From b7aec3eac4d6b7ae0d9451682d4e9dac88343a12 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Thu, 8 Aug 2024 11:08:43 -0400 Subject: [PATCH 01/33] add profiling notebook, hotfix a few classes --- docs/examples/profiling.ipynb | 5905 +++++++++++++++++ docs/getting_started/install.md | 2 +- project/configs/algorithm/example.yaml | 2 + .../algorithm/example_from_config.yaml | 17 - .../algorithm/optimizer/custom_adam.yaml | 6 + project/configs/config.yaml | 2 +- project/configs/datamodule/imagenet.yaml | 3 + .../configs/trainer/callbacks/default.yaml | 3 + project/experiment.py | 9 +- project/main.py | 2 +- pyproject.toml | 2 + 11 files changed, 5925 insertions(+), 28 deletions(-) create mode 100644 docs/examples/profiling.ipynb create mode 100644 project/configs/algorithm/example.yaml delete mode 100644 project/configs/algorithm/example_from_config.yaml create mode 100644 project/configs/algorithm/optimizer/custom_adam.yaml diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb new file mode 100644 index 00000000..2ac62445 --- /dev/null +++ b/docs/examples/profiling.ipynb @@ -0,0 +1,5905 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Benchmarking" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Setup" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The Mila Research Template leverages built-in PyTorch and Lightning functionality to make benchmarking accesible and flexible. " + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "/home/mila/c/cesar.valdez/idt/ResearchTemplate\n", + "/home/mila/c/cesar.valdez/idt/ResearchTemplate\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/IPython/core/magics/osm.py:417: UserWarning: This is now an optional IPython functionality, setting dhist requires you to install the `pickleshare` library.\n", + " self.shell.db['dhist'] = compress_dhist(dhist)[-100:]\n" + ] + } + ], + "source": [ + "import os\n", + "import rootutils\n", + "\n", + "home_dir = rootutils.find_root(search_from=\"profiling.ipynb\", indicator=\".git\")\n", + "%cd $home_dir\n", + "print(os.getcwd())" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Finding bottlenecks: Dataloading vs Training" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "A potential use of .." + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2mCONFIG\u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2malgorithm\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mhydra_zen.funcs.zen_processing \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_zen_target\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.no_op.NoOp \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_zen_partial\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_zen_wrappers\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mhydra_zen.third_party.pydantic.pydantic_parser \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mnetwork\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtorchvision.models.resnet.resnet18 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mweights\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mprogress\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_classes\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${instance_attr:datamodule.num_classes} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mdatamodule\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.datamodules.ImageNetDataModule \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdata_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:DATA_DIR} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_workers\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:NUM_WORKERS} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mval_split\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnormalize\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mshuffle\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mpin_memory\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mseed\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m42 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mbatch_size\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m64 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mtrainer\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.Trainer \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlogger\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40maccelerator\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mstrategy\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdevices\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmin_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmax_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdefault_root_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdetect_anomaly\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mcallbacks\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_checkpoint\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.ModelCheckpoint \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdirpath\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir}/checkpoints \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mfilename\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mepoch_{epoch:03d} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_last\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_top_k\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mauto_insert_metric_name\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_weights_only\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_train_steps\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mtrain_time_interval\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mearly_stopping\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.EarlyStopping \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmin_delta\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.0 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mpatience\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m5 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstrict\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_finite\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstopping_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdivergence_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_summary\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichModelSummary \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmax_depth\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m2 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mrich_progress_bar\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichProgressBar \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mthroughput\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.callbacks.samples_per_second.MeasureSam\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_train_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_val_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mlog_level\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40minfo \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mseed\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40m87282 \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mname\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mNoOp-resnet18-imagenet \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mdebug\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", + "\u001b[2m└── \u001b[0m\u001b[2mverbose\u001b[0m\n", + "\u001b[2m \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", + "\u001b[2;36m[08/07/24 16:28:58]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=131611;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=866100;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m,\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", + "seed manually set to 87282\n", + "Seed set to 87282\n", + "/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=NoOp trainer.max_epochs=1 ...\n", + "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", + "GPU available: True (cuda), used: True\n", + "TPU available: False, using: 0 TPU cores\n", + "HPU available: False, using: 0 HPUs\n", + "\u001b[2;36m[16:28:58]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=867639;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=12083;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#163\u001b\\\u001b[2m163\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7fb2a6f64f80\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", + "Extracting train archive: 100%|████| 1000/1000 [10:10<00:00, 1.64Directories/s]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", + "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", + "┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━┩\n", + "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNorm2d │ 128 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequential │ 147 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequential │ 525 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequential │ 2.1 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequential │ 8.4 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ AdaptiveAvgPool2d │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\n", + "└────┴─────────────────┴───────────────────┴────────┴───────┘\n", + "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", + "\u001b[1mNon-trainable params\u001b[0m: 0 \n", + "\u001b[1mTotal params\u001b[0m: 11.7 M \n", + "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/180\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0mt/s\u001b[0m \n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m100.703 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/180\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m100.703 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/180\u001b[0m \u001b[37m0:00:02 • 0:01:48\u001b[0m \u001b[37m1.64it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:35\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:35\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:31\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:31\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:26\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:26\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:22\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:22\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:05 • 0:01:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:05 • 0:01:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:07 • 0:01:19\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:07 • 0:01:19\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:07 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:07 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:08 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:08 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:08 • 0:01:19\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:08 • 0:01:19\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:09 • 0:01:17\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:09 • 0:01:17\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:09 • 0:01:16\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:09 • 0:01:16\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:11 • 0:01:13\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:11 • 0:01:13\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:13 • 0:01:11\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:13 • 0:01:11\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:13 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:13 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:14 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:14 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:14 • 0:01:09\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:14 • 0:01:09\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:14 • 0:01:08\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:14 • 0:01:08\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:16 • 0:01:07\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:16 • 0:01:07\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:16 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:16 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:17 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:17 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:17 • 0:01:05\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:17 • 0:01:05\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:17 • 0:01:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:17 • 0:01:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:25 • 0:00:56\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:25 • 0:00:56\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:26 • 0:00:56\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:26 • 0:00:56\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:26 • 0:00:55\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:26 • 0:00:55\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:27 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:27 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:28 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:28 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:28 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:28 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:29 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:29 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:32 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:32 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:33 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:33 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:33 • 0:00:46\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:33 • 0:00:46\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:34 • 0:00:46\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:34 • 0:00:46\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:34 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:34 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:35 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:35 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:37 • 0:00:41\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:37 • 0:00:41\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:38 • 0:00:41\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:38 • 0:00:41\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:38 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:38 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:39 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:39 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:39 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:39 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:40 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:40 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:42 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:42 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:44 • 0:00:33\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:44 • 0:00:33\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:45 • 0:00:33\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:45 • 0:00:33\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:46 • 0:00:33\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:46 • 0:00:33\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:47 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:47 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:47 • 0:00:31\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:47 • 0:00:31\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:48 • 0:00:31\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:48 • 0:00:31\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:48 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:48 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:49 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:49 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:49 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:49 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:50 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:50 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:51 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:51 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:51 • 0:00:27\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:51 • 0:00:27\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:52 • 0:00:27\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:52 • 0:00:27\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:52 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:52 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:54 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:54 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:55 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:55 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:55 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:55 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:00:56 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:00:56 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:00:57 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:00:57 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:00:57 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:00:57 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:00:58 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:00:58 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:00:59 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:00:59 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:00:59 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:00:59 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:00 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:00 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:05 • 0:00:14\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:05 • 0:00:14\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:07 • 0:00:12\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:07 • 0:00:12\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:07 • 0:00:11\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:07 • 0:00:11\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:08 • 0:00:11\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:08 • 0:00:11\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:08 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:08 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:10 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:10 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:11 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:11 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:11 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:11 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:12 • 0:00:06\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:12 • 0:00:06\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:13 • 0:00:06\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:13 • 0:00:06\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:13 • 0:00:05\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:13 • 0:00:05\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:14 • 0:00:05\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:14 • 0:00:05\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:14 • 0:00:04\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:14 • 0:00:04\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:16 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:16 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:17 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:17 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:17 • 0:00:01\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:17 • 0:00:01\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:18 • 0:00:01\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:18 • 0:00:01\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m151.611 \u001b[0m\n", + " \u001b[37mval/samples_per_…\u001b[0m\n", + " \u001b[37m120.811 \u001b[0m\n", + " \u001b[37mval/samples_per_…\u001b[0m\n", + " \u001b[37m128.323 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m150.129 \u001b[0m\n", + "\u001b[?25h/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=NoOp trainer.max_epochs=1 ...\n", + "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", + "\u001b[2;36m[16:41:35]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=972670;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=502992;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#381\u001b\\\u001b[2m381\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m[16:41:36]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=825781;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=661210;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37m2.10it/s\u001b[0m \n", + "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", + "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", + "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4964688718318939 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 140.0721893310547 \u001b[0m\u001b[35m \u001b[0m│\n", + "└──────────────────────────────┴──────────────────────────────┘\n", + "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.11it/s\u001b[0m \n", + "\u001b[?25hval val/samples_per_second_epoch: \u001b[1;36m140.0721893310547\u001b[0m\n" + ] + } + ], + "source": [ + "!python project/main.py \\\n", + " algorithm=NoOp \\\n", + " trainer.max_epochs=1 \\\n", + " +trainer.limit_train_batches=0.01\\\n", + " +trainer.limit_val_batches=0.01\\\n", + " datamodule=imagenet" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2mCONFIG\u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2malgorithm\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.example.ExampleAlgorithm \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_partial_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mnetwork\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtorchvision.models.resnet.resnet18 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mweights\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mprogress\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_classes\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${instance_attr:datamodule.num_classes} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mdatamodule\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.datamodules.ImageNetDataModule \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdata_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:DATA_DIR} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_workers\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:NUM_WORKERS} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mval_split\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnormalize\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mshuffle\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mpin_memory\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mseed\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m42 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mbatch_size\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m64 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mtrainer\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.Trainer \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlogger\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40maccelerator\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mstrategy\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdevices\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmin_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmax_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdefault_root_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdetect_anomaly\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mcallbacks\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_checkpoint\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.ModelCheckpoint \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdirpath\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir}/checkpoints \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mfilename\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mepoch_{epoch:03d} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_last\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_top_k\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mauto_insert_metric_name\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_weights_only\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_train_steps\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mtrain_time_interval\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mearly_stopping\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.EarlyStopping \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmin_delta\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.0 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mpatience\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m5 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstrict\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_finite\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstopping_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdivergence_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_summary\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichModelSummary \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmax_depth\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m2 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mrich_progress_bar\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichProgressBar \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mthroughput\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.callbacks.samples_per_second.MeasureSam\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_train_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_val_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mlog_level\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40minfo \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mseed\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40m81570 \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mname\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mexample-resnet18-imagenet \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mdebug\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", + "\u001b[2m└── \u001b[0m\u001b[2mverbose\u001b[0m\n", + "\u001b[2m \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", + "\u001b[2;36m[08/07/24 17:07:38]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=50545;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=71944;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m,\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", + "seed manually set to 81570\n", + "Seed set to 81570\n", + "/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=example trainer.max_epochs ...\n", + "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", + "GPU available: True (cuda), used: True\n", + "TPU available: False, using: 0 TPU cores\n", + "HPU available: False, using: 0 HPUs\n", + "\u001b[2;36m[17:07:38]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=285735;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=912000;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#156\u001b\\\u001b[2m156\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7eff7e751f10\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m[17:07:43]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=702216;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=293228;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m train_accuracy. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=476337;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=48315;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m val_accuracy. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=513180;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=101826;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m test_accuracy. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=370748;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=635161;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m train_top5_accuracy. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=727524;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=265133;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m val_top5_accuracy. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=753886;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=808003;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m test_top5_accuracy. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=781296;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=524717;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#381\u001b\\\u001b[2m381\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m[17:07:44]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=248066;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=550199;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m[17:08:02]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Not adding metrics to the pl \u001b]8;id=656135;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=914060;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#64\u001b\\\u001b[2m64\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module because they are already \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m present. \u001b[2m \u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "┏━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━┓\n", + "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mIn sizes\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mOut sizes\u001b[0m\u001b[1;35m \u001b[0m┃\n", + "┡━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━┩\n", + "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\u001b[37m \u001b[0m\u001b[37m [64, 3,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1000]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\u001b[37m \u001b[0m\u001b[37m [64, 3,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNor… │ 128 │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequenti… │ 147 K │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequenti… │ 525 K │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 128,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 28, 28]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequenti… │ 2.1 M │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 256,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m128, 28,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 14, 14]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 28]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequenti… │ 8.4 M │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 512,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m256, 14,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 7, 7]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 14]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ Adaptive… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 512,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 512, 7,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1, 1]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 7]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 512]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1000]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m11\u001b[0m\u001b[2m \u001b[0m│ train_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m12\u001b[0m\u001b[2m \u001b[0m│ val_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m13\u001b[0m\u001b[2m \u001b[0m│ test_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m14\u001b[0m\u001b[2m \u001b[0m│ train_top5_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m15\u001b[0m\u001b[2m \u001b[0m│ val_top5_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m16\u001b[0m\u001b[2m \u001b[0m│ test_top5_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", + "└────┴─────────────────────┴───────────┴────────┴───────┴──────────┴───────────┘\n", + "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", + "\u001b[1mNon-trainable params\u001b[0m: 0 \n", + "\u001b[1mTotal params\u001b[0m: 11.7 M \n", + "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/180\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0mt/s\u001b[0m \n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m51.620 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/180\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m51.620 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/180\u001b[0m \u001b[37m0:00:01 • 0:00:21\u001b[0m \u001b[37m8.90it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.059 train/loss: \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/180\u001b[0m \u001b[37m0:00:02 • 0:01:01\u001b[0m \u001b[37m2.92it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.059 train/loss: \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/180\u001b[0m \u001b[37m0:00:02 • 0:01:01\u001b[0m \u001b[37m2.92it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.015 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m6.940 train/loss: \u001b[0m\n", + " \u001b[37m6.940 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:04\u001b[0m \u001b[37m2.76it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.015 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m6.940 train/loss: \u001b[0m\n", + " \u001b[37m6.940 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:04\u001b[0m \u001b[37m2.76it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.146 train/loss: \u001b[0m\n", + " \u001b[37m7.146 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:08\u001b[0m \u001b[37m2.58it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.146 train/loss: \u001b[0m\n", + " \u001b[37m7.146 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:08\u001b[0m \u001b[37m2.58it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.059 train/loss: \u001b[0m\n", + " \u001b[37m7.059 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.44it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.059 train/loss: \u001b[0m\n", + " \u001b[37m7.059 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.44it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.076 train/loss: \u001b[0m\n", + " \u001b[37m7.076 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.43it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.076 train/loss: \u001b[0m\n", + " \u001b[37m7.076 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.43it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.096 train/loss: \u001b[0m\n", + " \u001b[37m7.096 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.096 train/loss: \u001b[0m\n", + " \u001b[37m7.096 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.142 train/loss: \u001b[0m\n", + " \u001b[37m7.142 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.142 train/loss: \u001b[0m\n", + " \u001b[37m7.142 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.030 train/loss: \u001b[0m\n", + " \u001b[37m7.030 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:19\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.030 train/loss:\u001b[0m\n", + " \u001b[37m7.030 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:19\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.100 train/loss:\u001b[0m\n", + " \u001b[37m7.100 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:06 • 0:01:31\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.100 train/loss:\u001b[0m\n", + " \u001b[37m7.100 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:06 • 0:01:31\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.008 train/loss:\u001b[0m\n", + " \u001b[37m7.008 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:07 • 0:01:28\u001b[0m \u001b[37m1.91it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.008 train/loss:\u001b[0m\n", + " \u001b[37m7.008 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:07 • 0:01:28\u001b[0m \u001b[37m1.91it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.302 train/loss:\u001b[0m\n", + " \u001b[37m7.302 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:07 • 0:01:26\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.302 train/loss:\u001b[0m\n", + " \u001b[37m7.302 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:07 • 0:01:26\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.344 train/loss:\u001b[0m\n", + " \u001b[37m7.344 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:08 • 0:01:30\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.344 train/loss:\u001b[0m\n", + " \u001b[37m7.344 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:08 • 0:01:30\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.050 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.361 train/loss:\u001b[0m\n", + " \u001b[37m7.361 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:08 • 0:01:28\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.050 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.361 train/loss:\u001b[0m\n", + " \u001b[37m7.361 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:08 • 0:01:28\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.182 train/loss:\u001b[0m\n", + " \u001b[37m7.182 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:10 • 0:01:35\u001b[0m \u001b[37m1.73it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.182 train/loss:\u001b[0m\n", + " \u001b[37m7.182 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:10 • 0:01:35\u001b[0m \u001b[37m1.73it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.168 train/loss:\u001b[0m\n", + " \u001b[37m7.168 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:10 • 0:01:37\u001b[0m \u001b[37m1.69it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.168 train/loss:\u001b[0m\n", + " \u001b[37m7.168 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:10 • 0:01:37\u001b[0m \u001b[37m1.69it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.324 train/loss:\u001b[0m\n", + " \u001b[37m7.324 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:11 • 0:01:35\u001b[0m \u001b[37m1.72it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.324 train/loss:\u001b[0m\n", + " \u001b[37m7.324 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:11 • 0:01:35\u001b[0m \u001b[37m1.72it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.436 train/loss:\u001b[0m\n", + " \u001b[37m7.436 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:11 • 0:01:32\u001b[0m \u001b[37m1.75it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.436 train/loss:\u001b[0m\n", + " \u001b[37m7.436 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:11 • 0:01:32\u001b[0m \u001b[37m1.75it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.111 train/loss:\u001b[0m\n", + " \u001b[37m7.111 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:12 • 0:01:31\u001b[0m \u001b[37m1.78it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.111 train/loss:\u001b[0m\n", + " \u001b[37m7.111 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:12 • 0:01:31\u001b[0m \u001b[37m1.78it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.521 train/loss:\u001b[0m\n", + " \u001b[37m7.521 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:12 • 0:01:29\u001b[0m \u001b[37m1.79it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.521 train/loss:\u001b[0m\n", + " \u001b[37m7.521 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:12 • 0:01:29\u001b[0m \u001b[37m1.79it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.065 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.163 train/loss:\u001b[0m\n", + " \u001b[37m7.163 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:13 • 0:01:28\u001b[0m \u001b[37m1.81it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.065 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.163 train/loss:\u001b[0m\n", + " \u001b[37m7.163 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:13 • 0:01:28\u001b[0m \u001b[37m1.81it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.038 train/loss:\u001b[0m\n", + " \u001b[37m7.038 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:13 • 0:01:26\u001b[0m \u001b[37m1.83it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.038 train/loss:\u001b[0m\n", + " \u001b[37m7.038 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:13 • 0:01:26\u001b[0m \u001b[37m1.83it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.171 train/loss:\u001b[0m\n", + " \u001b[37m7.171 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:13 • 0:01:25\u001b[0m \u001b[37m1.85it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.171 train/loss:\u001b[0m\n", + " \u001b[37m7.171 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:13 • 0:01:25\u001b[0m \u001b[37m1.85it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.327 train/loss:\u001b[0m\n", + " \u001b[37m7.327 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:14 • 0:01:23\u001b[0m \u001b[37m1.87it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.327 train/loss:\u001b[0m\n", + " \u001b[37m7.327 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:14 • 0:01:23\u001b[0m \u001b[37m1.87it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.207 train/loss:\u001b[0m\n", + " \u001b[37m7.207 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:14 • 0:01:22\u001b[0m \u001b[37m1.90it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.207 train/loss:\u001b[0m\n", + " \u001b[37m7.207 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:14 • 0:01:22\u001b[0m \u001b[37m1.90it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.271 train/loss:\u001b[0m\n", + " \u001b[37m7.271 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:15 • 0:01:20\u001b[0m \u001b[37m1.92it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.271 train/loss:\u001b[0m\n", + " \u001b[37m7.271 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:15 • 0:01:20\u001b[0m \u001b[37m1.92it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.566 train/loss:\u001b[0m\n", + " \u001b[37m7.566 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:15 • 0:01:19\u001b[0m \u001b[37m1.93it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.566 train/loss:\u001b[0m\n", + " \u001b[37m7.566 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:15 • 0:01:19\u001b[0m \u001b[37m1.93it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.203 train/loss:\u001b[0m\n", + " \u001b[37m7.203 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:15 • 0:01:18\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.203 train/loss:\u001b[0m\n", + " \u001b[37m7.203 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:15 • 0:01:18\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.262 train/loss:\u001b[0m\n", + " \u001b[37m7.262 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:16 • 0:01:17\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.262 train/loss:\u001b[0m\n", + " \u001b[37m7.262 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:16 • 0:01:17\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.109 train/loss:\u001b[0m\n", + " \u001b[37m7.109 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:16 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.109 train/loss:\u001b[0m\n", + " \u001b[37m7.109 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:16 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.343 train/loss:\u001b[0m\n", + " \u001b[37m7.343 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:17 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.343 train/loss:\u001b[0m\n", + " \u001b[37m7.343 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:17 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.190 train/loss:\u001b[0m\n", + " \u001b[37m7.190 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:17 • 0:01:15\u001b[0m \u001b[37m1.98it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.190 train/loss:\u001b[0m\n", + " \u001b[37m7.190 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:17 • 0:01:15\u001b[0m \u001b[37m1.98it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.051 train/loss:\u001b[0m\n", + " \u001b[37m7.051 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:18 • 0:01:14\u001b[0m \u001b[37m1.99it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.051 train/loss:\u001b[0m\n", + " \u001b[37m7.051 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:18 • 0:01:14\u001b[0m \u001b[37m1.99it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.174 train/loss:\u001b[0m\n", + " \u001b[37m7.174 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:18 • 0:01:13\u001b[0m \u001b[37m2.00it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.174 train/loss:\u001b[0m\n", + " \u001b[37m7.174 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:18 • 0:01:13\u001b[0m \u001b[37m2.00it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.390 train/loss:\u001b[0m\n", + " \u001b[37m7.390 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:18 • 0:01:12\u001b[0m \u001b[37m2.01it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.390 train/loss:\u001b[0m\n", + " \u001b[37m7.390 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:18 • 0:01:12\u001b[0m \u001b[37m2.01it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.300 train/loss:\u001b[0m\n", + " \u001b[37m7.300 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:19 • 0:01:11\u001b[0m \u001b[37m2.02it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.300 train/loss:\u001b[0m\n", + " \u001b[37m7.300 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:19 • 0:01:11\u001b[0m \u001b[37m2.02it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.237 train/loss:\u001b[0m\n", + " \u001b[37m7.237 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:19 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.237 train/loss:\u001b[0m\n", + " \u001b[37m7.237 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:19 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.232 train/loss:\u001b[0m\n", + " \u001b[37m7.232 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:20 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.232 train/loss:\u001b[0m\n", + " \u001b[37m7.232 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:20 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.971 train/loss:\u001b[0m\n", + " \u001b[37m6.971 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:20 • 0:01:09\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.971 train/loss:\u001b[0m\n", + " \u001b[37m6.971 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:20 • 0:01:09\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.262 train/loss:\u001b[0m\n", + " \u001b[37m7.262 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.262 train/loss:\u001b[0m\n", + " \u001b[37m7.262 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.142 train/loss:\u001b[0m\n", + " \u001b[37m7.142 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.142 train/loss:\u001b[0m\n", + " \u001b[37m7.142 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.984 train/loss:\u001b[0m\n", + " \u001b[37m6.984 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.984 train/loss:\u001b[0m\n", + " \u001b[37m6.984 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.149 train/loss:\u001b[0m\n", + " \u001b[37m7.149 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.149 train/loss:\u001b[0m\n", + " \u001b[37m7.149 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.986 train/loss:\u001b[0m\n", + " \u001b[37m6.986 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:22 • 0:01:06\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.986 train/loss:\u001b[0m\n", + " \u001b[37m6.986 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:22 • 0:01:06\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.067 train/loss:\u001b[0m\n", + " \u001b[37m7.067 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:23 • 0:01:06\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.067 train/loss:\u001b[0m\n", + " \u001b[37m7.067 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:23 • 0:01:06\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.224 train/loss:\u001b[0m\n", + " \u001b[37m7.224 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:23 • 0:01:05\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.224 train/loss:\u001b[0m\n", + " \u001b[37m7.224 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:23 • 0:01:05\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.053 train/loss:\u001b[0m\n", + " \u001b[37m7.053 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:24 • 0:01:05\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.053 train/loss:\u001b[0m\n", + " \u001b[37m7.053 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:24 • 0:01:05\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.016 train/loss:\u001b[0m\n", + " \u001b[37m7.016 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:24 • 0:01:04\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.016 train/loss:\u001b[0m\n", + " \u001b[37m7.016 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:24 • 0:01:04\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.050 train/loss:\u001b[0m\n", + " \u001b[37m7.050 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.050 train/loss:\u001b[0m\n", + " \u001b[37m7.050 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.025 train/loss:\u001b[0m\n", + " \u001b[37m7.025 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.025 train/loss:\u001b[0m\n", + " \u001b[37m7.025 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.886 train/loss:\u001b[0m\n", + " \u001b[37m6.886 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:25 • 0:01:02\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.886 train/loss:\u001b[0m\n", + " \u001b[37m6.886 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:25 • 0:01:02\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.926 train/loss:\u001b[0m\n", + " \u001b[37m6.926 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.926 train/loss:\u001b[0m\n", + " \u001b[37m6.926 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.090 train/loss:\u001b[0m\n", + " \u001b[37m7.090 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.090 train/loss:\u001b[0m\n", + " \u001b[37m7.090 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.913 train/loss:\u001b[0m\n", + " \u001b[37m6.913 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.913 train/loss:\u001b[0m\n", + " \u001b[37m6.913 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.100 train/loss:\u001b[0m\n", + " \u001b[37m7.100 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.100 train/loss:\u001b[0m\n", + " \u001b[37m7.100 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.028 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.985 train/loss:\u001b[0m\n", + " \u001b[37m6.985 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:28 • 0:00:59\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.028 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.985 train/loss:\u001b[0m\n", + " \u001b[37m6.985 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:28 • 0:00:59\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.191 train/loss:\u001b[0m\n", + " \u001b[37m7.191 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.191 train/loss:\u001b[0m\n", + " \u001b[37m7.191 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.939 train/loss:\u001b[0m\n", + " \u001b[37m6.939 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.939 train/loss:\u001b[0m\n", + " \u001b[37m6.939 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.859 train/loss:\u001b[0m\n", + " \u001b[37m6.859 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.859 train/loss:\u001b[0m\n", + " \u001b[37m6.859 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.076 train/loss:\u001b[0m\n", + " \u001b[37m7.076 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.076 train/loss:\u001b[0m\n", + " \u001b[37m7.076 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.115 train/loss:\u001b[0m\n", + " \u001b[37m7.115 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:30 • 0:00:56\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.115 train/loss:\u001b[0m\n", + " \u001b[37m7.115 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:30 • 0:00:56\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.936 train/loss:\u001b[0m\n", + " \u001b[37m6.936 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.936 train/loss:\u001b[0m\n", + " \u001b[37m6.936 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.901 train/loss:\u001b[0m\n", + " \u001b[37m6.901 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.901 train/loss:\u001b[0m\n", + " \u001b[37m6.901 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.068 train/loss:\u001b[0m\n", + " \u001b[37m7.068 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.068 train/loss:\u001b[0m\n", + " \u001b[37m7.068 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.932 train/loss:\u001b[0m\n", + " \u001b[37m6.932 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.932 train/loss:\u001b[0m\n", + " \u001b[37m6.932 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.934 train/loss:\u001b[0m\n", + " \u001b[37m6.934 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.934 train/loss:\u001b[0m\n", + " \u001b[37m6.934 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.022 train/loss:\u001b[0m\n", + " \u001b[37m7.022 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.022 train/loss:\u001b[0m\n", + " \u001b[37m7.022 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.905 train/loss:\u001b[0m\n", + " \u001b[37m6.905 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.905 train/loss:\u001b[0m\n", + " \u001b[37m6.905 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.929 train/loss:\u001b[0m\n", + " \u001b[37m6.929 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.929 train/loss:\u001b[0m\n", + " \u001b[37m6.929 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.060 train/loss:\u001b[0m\n", + " \u001b[37m7.060 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.060 train/loss:\u001b[0m\n", + " \u001b[37m7.060 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.089 train/loss:\u001b[0m\n", + " \u001b[37m7.089 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:34 • 0:00:51\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.089 train/loss:\u001b[0m\n", + " \u001b[37m7.089 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:34 • 0:00:51\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.011 train/loss:\u001b[0m\n", + " \u001b[37m7.011 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:34 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.011 train/loss:\u001b[0m\n", + " \u001b[37m7.011 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:34 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.953 train/loss:\u001b[0m\n", + " \u001b[37m6.953 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:35 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.953 train/loss:\u001b[0m\n", + " \u001b[37m6.953 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:35 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.890 train/loss:\u001b[0m\n", + " \u001b[37m6.890 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:35 • 0:00:48\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.890 train/loss:\u001b[0m\n", + " \u001b[37m6.890 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:35 • 0:00:48\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.829 train/loss:\u001b[0m\n", + " \u001b[37m6.829 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.829 train/loss:\u001b[0m\n", + " \u001b[37m6.829 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.009 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.902 train/loss:\u001b[0m\n", + " \u001b[37m6.902 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.009 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.902 train/loss:\u001b[0m\n", + " \u001b[37m6.902 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.996 train/loss:\u001b[0m\n", + " \u001b[37m6.996 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:37 • 0:00:47\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.996 train/loss:\u001b[0m\n", + " \u001b[37m6.996 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:37 • 0:00:47\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.074 train/loss:\u001b[0m\n", + " \u001b[37m7.074 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:37 • 0:00:46\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.074 train/loss:\u001b[0m\n", + " \u001b[37m7.074 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:37 • 0:00:46\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.826 train/loss:\u001b[0m\n", + " \u001b[37m6.826 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.826 train/loss:\u001b[0m\n", + " \u001b[37m6.826 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.075 train/loss:\u001b[0m\n", + " \u001b[37m7.075 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.075 train/loss:\u001b[0m\n", + " \u001b[37m7.075 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.881 train/loss:\u001b[0m\n", + " \u001b[37m6.881 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.28it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.881 train/loss:\u001b[0m\n", + " \u001b[37m6.881 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.28it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.860 train/loss:\u001b[0m\n", + " \u001b[37m6.860 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.860 train/loss:\u001b[0m\n", + " \u001b[37m6.860 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.005 train/loss:\u001b[0m\n", + " \u001b[37m7.005 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:39 • 0:00:42\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.005 train/loss:\u001b[0m\n", + " \u001b[37m7.005 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:39 • 0:00:42\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.864 train/loss:\u001b[0m\n", + " \u001b[37m6.864 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:40 • 0:00:42\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.864 train/loss:\u001b[0m\n", + " \u001b[37m6.864 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:40 • 0:00:42\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.953 train/loss:\u001b[0m\n", + " \u001b[37m6.953 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:40 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.953 train/loss:\u001b[0m\n", + " \u001b[37m6.953 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:40 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.968 train/loss:\u001b[0m\n", + " \u001b[37m6.968 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:41 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.968 train/loss:\u001b[0m\n", + " \u001b[37m6.968 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:41 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.973 train/loss:\u001b[0m\n", + " \u001b[37m6.973 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:41 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.973 train/loss:\u001b[0m\n", + " \u001b[37m6.973 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:41 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.986 train/loss:\u001b[0m\n", + " \u001b[37m6.986 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:42 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.986 train/loss:\u001b[0m\n", + " \u001b[37m6.986 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:42 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.007 train/loss:\u001b[0m\n", + " \u001b[37m7.007 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.007 train/loss:\u001b[0m\n", + " \u001b[37m7.007 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.029 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.940 train/loss:\u001b[0m\n", + " \u001b[37m6.940 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.029 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.940 train/loss:\u001b[0m\n", + " \u001b[37m6.940 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.932 train/loss:\u001b[0m\n", + " \u001b[37m6.932 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:43 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.932 train/loss:\u001b[0m\n", + " \u001b[37m6.932 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:43 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.997 train/loss:\u001b[0m\n", + " \u001b[37m6.997 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:43 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.997 train/loss:\u001b[0m\n", + " \u001b[37m6.997 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:43 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.936 train/loss:\u001b[0m\n", + " \u001b[37m6.936 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:44 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.936 train/loss:\u001b[0m\n", + " \u001b[37m6.936 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:44 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.014 train/loss:\u001b[0m\n", + " \u001b[37m7.014 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:44 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.014 train/loss:\u001b[0m\n", + " \u001b[37m7.014 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:44 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.937 train/loss:\u001b[0m\n", + " \u001b[37m6.937 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.937 train/loss:\u001b[0m\n", + " \u001b[37m6.937 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.915 train/loss:\u001b[0m\n", + " \u001b[37m6.915 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.915 train/loss:\u001b[0m\n", + " \u001b[37m6.915 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.923 train/loss:\u001b[0m\n", + " \u001b[37m6.923 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:45 • 0:00:36\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.923 train/loss:\u001b[0m\n", + " \u001b[37m6.923 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:45 • 0:00:36\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.014 train/loss:\u001b[0m\n", + " \u001b[37m7.014 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:46 • 0:00:36\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.014 train/loss:\u001b[0m\n", + " \u001b[37m7.014 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:46 • 0:00:36\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.963 train/loss:\u001b[0m\n", + " \u001b[37m6.963 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:46 • 0:00:35\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.963 train/loss:\u001b[0m\n", + " \u001b[37m6.963 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:46 • 0:00:35\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.873 train/loss:\u001b[0m\n", + " \u001b[37m6.873 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:47 • 0:00:35\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.873 train/loss:\u001b[0m\n", + " \u001b[37m6.873 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:47 • 0:00:35\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.033 train/loss:\u001b[0m\n", + " \u001b[37m7.033 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:47 • 0:00:34\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.033 train/loss:\u001b[0m\n", + " \u001b[37m7.033 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:47 • 0:00:34\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.034 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.787 train/loss:\u001b[0m\n", + " \u001b[37m6.787 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:48 • 0:00:35\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.034 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.787 train/loss:\u001b[0m\n", + " \u001b[37m6.787 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:48 • 0:00:35\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.946 train/loss:\u001b[0m\n", + " \u001b[37m6.946 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.946 train/loss:\u001b[0m\n", + " \u001b[37m6.946 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.966 train/loss:\u001b[0m\n", + " \u001b[37m6.966 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.966 train/loss:\u001b[0m\n", + " \u001b[37m6.966 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.899 train/loss:\u001b[0m\n", + " \u001b[37m6.899 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:49 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.899 train/loss:\u001b[0m\n", + " \u001b[37m6.899 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:49 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.031 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.886 train/loss:\u001b[0m\n", + " \u001b[37m6.886 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:50 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.031 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.886 train/loss:\u001b[0m\n", + " \u001b[37m6.886 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:50 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.892 train/loss:\u001b[0m\n", + " \u001b[37m6.892 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:51 • 0:00:33\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.892 train/loss:\u001b[0m\n", + " \u001b[37m6.892 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:51 • 0:00:33\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.880 train/loss:\u001b[0m\n", + " \u001b[37m6.880 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:51 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.880 train/loss:\u001b[0m\n", + " \u001b[37m6.880 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:51 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.897 train/loss:\u001b[0m\n", + " \u001b[37m6.897 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.897 train/loss:\u001b[0m\n", + " \u001b[37m6.897 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.891 train/loss:\u001b[0m\n", + " \u001b[37m6.891 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.891 train/loss:\u001b[0m\n", + " \u001b[37m6.891 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.979 train/loss:\u001b[0m\n", + " \u001b[37m6.979 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:53 • 0:00:31\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.979 train/loss:\u001b[0m\n", + " \u001b[37m6.979 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:53 • 0:00:31\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.860 train/loss:\u001b[0m\n", + " \u001b[37m6.860 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.860 train/loss:\u001b[0m\n", + " \u001b[37m6.860 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.014 train/loss:\u001b[0m\n", + " \u001b[37m7.014 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.014 train/loss:\u001b[0m\n", + " \u001b[37m7.014 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.908 train/loss:\u001b[0m\n", + " \u001b[37m6.908 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.908 train/loss:\u001b[0m\n", + " \u001b[37m6.908 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.802 train/loss:\u001b[0m\n", + " \u001b[37m6.802 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.802 train/loss:\u001b[0m\n", + " \u001b[37m6.802 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.839 train/loss:\u001b[0m\n", + " \u001b[37m6.839 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:56 • 0:00:30\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.839 train/loss:\u001b[0m\n", + " \u001b[37m6.839 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:56 • 0:00:30\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.880 train/loss:\u001b[0m\n", + " \u001b[37m6.880 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.880 train/loss:\u001b[0m\n", + " \u001b[37m6.880 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.812 train/loss:\u001b[0m\n", + " \u001b[37m6.812 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.812 train/loss:\u001b[0m\n", + " \u001b[37m6.812 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.848 train/loss:\u001b[0m\n", + " \u001b[37m6.848 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.848 train/loss:\u001b[0m\n", + " \u001b[37m6.848 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.946 train/loss:\u001b[0m\n", + " \u001b[37m6.946 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.946 train/loss:\u001b[0m\n", + " \u001b[37m6.946 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.939 train/loss:\u001b[0m\n", + " \u001b[37m6.939 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.939 train/loss:\u001b[0m\n", + " \u001b[37m6.939 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.839 train/loss:\u001b[0m\n", + " \u001b[37m6.839 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.839 train/loss:\u001b[0m\n", + " \u001b[37m6.839 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.827 train/loss:\u001b[0m\n", + " \u001b[37m6.827 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.827 train/loss:\u001b[0m\n", + " \u001b[37m6.827 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.862 train/loss:\u001b[0m\n", + " \u001b[37m6.862 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.862 train/loss:\u001b[0m\n", + " \u001b[37m6.862 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.034 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.838 train/loss:\u001b[0m\n", + " \u001b[37m6.838 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.034 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.838 train/loss:\u001b[0m\n", + " \u001b[37m6.838 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.004 train/loss:\u001b[0m\n", + " \u001b[37m7.004 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.004 train/loss:\u001b[0m\n", + " \u001b[37m7.004 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.028 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.801 train/loss:\u001b[0m\n", + " \u001b[37m6.801 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.028 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.801 train/loss:\u001b[0m\n", + " \u001b[37m6.801 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.042 train/loss:\u001b[0m\n", + " \u001b[37m7.042 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.042 train/loss:\u001b[0m\n", + " \u001b[37m7.042 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.851 train/loss:\u001b[0m\n", + " \u001b[37m6.851 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.851 train/loss:\u001b[0m\n", + " \u001b[37m6.851 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.904 train/loss:\u001b[0m\n", + " \u001b[37m6.904 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:01:01 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.904 train/loss:\u001b[0m\n", + " \u001b[37m6.904 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:01:02 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.850 train/loss:\u001b[0m\n", + " \u001b[37m6.850 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:01:02 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.850 train/loss:\u001b[0m\n", + " \u001b[37m6.850 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:01:02 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.885 train/loss:\u001b[0m\n", + " \u001b[37m6.885 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:01:02 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.885 train/loss:\u001b[0m\n", + " \u001b[37m6.885 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:01:02 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.815 train/loss:\u001b[0m\n", + " \u001b[37m6.815 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:01:03 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.815 train/loss:\u001b[0m\n", + " \u001b[37m6.815 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:01:03 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.871 train/loss:\u001b[0m\n", + " \u001b[37m6.871 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:01:03 • 0:00:21\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.871 train/loss:\u001b[0m\n", + " \u001b[37m6.871 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:01:03 • 0:00:21\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.864 train/loss:\u001b[0m\n", + " \u001b[37m6.864 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:01:04 • 0:00:21\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.864 train/loss:\u001b[0m\n", + " \u001b[37m6.864 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:01:04 • 0:00:21\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.936 train/loss:\u001b[0m\n", + " \u001b[37m6.936 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.936 train/loss:\u001b[0m\n", + " \u001b[37m6.936 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.031 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.878 train/loss:\u001b[0m\n", + " \u001b[37m6.878 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.031 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.878 train/loss:\u001b[0m\n", + " \u001b[37m6.878 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.026 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.902 train/loss:\u001b[0m\n", + " \u001b[37m6.902 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.026 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.902 train/loss:\u001b[0m\n", + " \u001b[37m6.902 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.955 train/loss:\u001b[0m\n", + " \u001b[37m6.955 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.955 train/loss:\u001b[0m\n", + " \u001b[37m6.955 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.852 train/loss:\u001b[0m\n", + " \u001b[37m6.852 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.852 train/loss:\u001b[0m\n", + " \u001b[37m6.852 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.869 train/loss:\u001b[0m\n", + " \u001b[37m6.869 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.869 train/loss:\u001b[0m\n", + " \u001b[37m6.869 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.902 train/loss:\u001b[0m\n", + " \u001b[37m6.902 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.902 train/loss:\u001b[0m\n", + " \u001b[37m6.902 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.872 train/loss:\u001b[0m\n", + " \u001b[37m6.872 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.872 train/loss:\u001b[0m\n", + " \u001b[37m6.872 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.959 train/loss:\u001b[0m\n", + " \u001b[37m6.959 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:07 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.959 train/loss:\u001b[0m\n", + " \u001b[37m6.959 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:07 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.851 train/loss:\u001b[0m\n", + " \u001b[37m6.851 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.851 train/loss:\u001b[0m\n", + " \u001b[37m6.851 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.966 train/loss:\u001b[0m\n", + " \u001b[37m6.966 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.966 train/loss:\u001b[0m\n", + " \u001b[37m6.966 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.838 train/loss:\u001b[0m\n", + " \u001b[37m6.838 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.838 train/loss:\u001b[0m\n", + " \u001b[37m6.838 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.921 train/loss:\u001b[0m\n", + " \u001b[37m6.921 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.921 train/loss:\u001b[0m\n", + " \u001b[37m6.921 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.838 train/loss:\u001b[0m\n", + " \u001b[37m6.838 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.838 train/loss:\u001b[0m\n", + " \u001b[37m6.838 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.012 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.876 train/loss:\u001b[0m\n", + " \u001b[37m6.876 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.012 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.876 train/loss:\u001b[0m\n", + " \u001b[37m6.876 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.829 train/loss:\u001b[0m\n", + " \u001b[37m6.829 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:10 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.829 train/loss:\u001b[0m\n", + " \u001b[37m6.829 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:10 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.862 train/loss:\u001b[0m\n", + " \u001b[37m6.862 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:11 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.862 train/loss:\u001b[0m\n", + " \u001b[37m6.862 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:11 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.914 train/loss:\u001b[0m\n", + " \u001b[37m6.914 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:11 • 0:00:12\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.914 train/loss:\u001b[0m\n", + " \u001b[37m6.914 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:11 • 0:00:12\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.050 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.764 train/loss:\u001b[0m\n", + " \u001b[37m6.764 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.050 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.764 train/loss:\u001b[0m\n", + " \u001b[37m6.764 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.895 train/loss:\u001b[0m\n", + " \u001b[37m6.895 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.895 train/loss:\u001b[0m\n", + " \u001b[37m6.895 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.911 train/loss:\u001b[0m\n", + " \u001b[37m6.911 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:13 • 0:00:11\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.911 train/loss:\u001b[0m\n", + " \u001b[37m6.911 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:13 • 0:00:11\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.845 train/loss:\u001b[0m\n", + " \u001b[37m6.845 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:14 • 0:00:11\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.845 train/loss:\u001b[0m\n", + " \u001b[37m6.845 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:14 • 0:00:11\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.828 train/loss:\u001b[0m\n", + " \u001b[37m6.828 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:15 • 0:00:11\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.828 train/loss:\u001b[0m\n", + " \u001b[37m6.828 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:15 • 0:00:11\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.887 train/loss:\u001b[0m\n", + " \u001b[37m6.887 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.887 train/loss:\u001b[0m\n", + " \u001b[37m6.887 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.836 train/loss:\u001b[0m\n", + " \u001b[37m6.836 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.836 train/loss:\u001b[0m\n", + " \u001b[37m6.836 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.854 train/loss:\u001b[0m\n", + " \u001b[37m6.854 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.854 train/loss:\u001b[0m\n", + " \u001b[37m6.854 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.854 train/loss:\u001b[0m\n", + " \u001b[37m6.854 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.854 train/loss:\u001b[0m\n", + " \u001b[37m6.854 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.915 train/loss:\u001b[0m\n", + " \u001b[37m6.915 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.915 train/loss:\u001b[0m\n", + " \u001b[37m6.915 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.931 train/loss:\u001b[0m\n", + " \u001b[37m6.931 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.931 train/loss:\u001b[0m\n", + " \u001b[37m6.931 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.966 train/loss:\u001b[0m\n", + " \u001b[37m6.966 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.966 train/loss:\u001b[0m\n", + " \u001b[37m6.966 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.800 train/loss:\u001b[0m\n", + " \u001b[37m6.800 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.800 train/loss:\u001b[0m\n", + " \u001b[37m6.800 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.989 train/loss:\u001b[0m\n", + " \u001b[37m6.989 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.989 train/loss:\u001b[0m\n", + " \u001b[37m6.989 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.854 train/loss:\u001b[0m\n", + " \u001b[37m6.854 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.854 train/loss:\u001b[0m\n", + " \u001b[37m6.854 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.731 train/loss:\u001b[0m\n", + " \u001b[37m6.731 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:19 • 0:00:05\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.731 train/loss:\u001b[0m\n", + " \u001b[37m6.731 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:19 • 0:00:05\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.039 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.065 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.726 train/loss:\u001b[0m\n", + " \u001b[37m6.726 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:20 • 0:00:05\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.039 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.065 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.726 train/loss:\u001b[0m\n", + " \u001b[37m6.726 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:20 • 0:00:05\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.851 train/loss:\u001b[0m\n", + " \u001b[37m6.851 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:20 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.851 train/loss:\u001b[0m\n", + " \u001b[37m6.851 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:20 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.945 train/loss:\u001b[0m\n", + " \u001b[37m6.945 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:21 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.945 train/loss:\u001b[0m\n", + " \u001b[37m6.945 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:21 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.852 train/loss:\u001b[0m\n", + " \u001b[37m6.852 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:21 • 0:00:03\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.852 train/loss:\u001b[0m\n", + " \u001b[37m6.852 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:21 • 0:00:03\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.031 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.797 train/loss:\u001b[0m\n", + " \u001b[37m6.797 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:22 • 0:00:03\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.031 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.797 train/loss:\u001b[0m\n", + " \u001b[37m6.797 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:22 • 0:00:03\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.819 train/loss:\u001b[0m\n", + " \u001b[37m6.819 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.819 train/loss:\u001b[0m\n", + " \u001b[37m6.819 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.871 train/loss:\u001b[0m\n", + " \u001b[37m6.871 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.871 train/loss:\u001b[0m\n", + " \u001b[37m6.871 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.764 train/loss:\u001b[0m\n", + " \u001b[37m6.764 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:23 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.764 train/loss:\u001b[0m\n", + " \u001b[37m6.764 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:23 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.791 train/loss:\u001b[0m\n", + " \u001b[37m6.791 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:24 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.791 train/loss:\u001b[0m\n", + " \u001b[37m6.791 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:24 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.066 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.775 train/loss:\u001b[0m\n", + " \u001b[37m6.775 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.066 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.775 train/loss:\u001b[0m\n", + " \u001b[37m6.775 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.766 train/loss:\u001b[0m\n", + " \u001b[37m6.766 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m140.174 \u001b[0m\n", + " \u001b[37mval/samples_per_…\u001b[0m\n", + " \u001b[37m138.743 \u001b[0m\n", + " \u001b[37mval/samples_per_…\u001b[0m\n", + " \u001b[37m136.694 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m146.335 \u001b[0m\n", + "\u001b[?25h/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=example trainer.max_epochs ...\n", + "\u001b[2;36m[17:09:43]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Not adding metrics to the pl \u001b]8;id=733785;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=345137;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#64\u001b\\\u001b[2m64\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module because they are already \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m present. \u001b[2m \u001b[0m\n", + "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: ClassificationMetricsCallback, LearningRateMonitor\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=371304;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=737977;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#381\u001b\\\u001b[2m381\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=648908;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=28924;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m[17:10:01]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Not adding metrics to the pl \u001b]8;id=33733;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=344699;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#64\u001b\\\u001b[2m64\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module because they are already \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m present. \u001b[2m \u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37m2.17it/s\u001b[0m \n", + "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", + "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", + "│\u001b[36m \u001b[0m\u001b[36m val/accuracy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.003010033629834652 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36m val/cross_entropy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 6.422720432281494 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 6.422720432281494 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 140.8275146484375 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36m val/top5_accuracy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.029230769723653793 \u001b[0m\u001b[35m \u001b[0m│\n", + "└──────────────────────────────┴──────────────────────────────┘\n", + "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.17it/s\u001b[0m \n", + "\u001b[?25hval val/accuracy: \u001b[1;36m0.003010033629834652\u001b[0m\n", + "val val/top5_accuracy: \u001b[1;36m0.029230769723653793\u001b[0m\n", + "val val/cross_entropy: \u001b[1;36m6.422720432281494\u001b[0m\n", + "val val/samples_per_second_epoch: \u001b[1;36m140.8275146484375\u001b[0m\n" + ] + } + ], + "source": [ + "!HYDRA_FULL_ERROR=1 python project/main.py \\\n", + " algorithm=example \\\n", + " trainer.max_epochs=1 \\\n", + " +trainer.limit_train_batches=0.01\\\n", + " +trainer.limit_val_batches=0.01\\\n", + " datamodule=imagenet" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Testing for throughput across GPUs" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Using the Mila Research template, it is possible to sweep over different parameters for testing purposes. \n", + "For example, suppose we wanted to figure out how different GPUs perform relative to each other. \n", + "\n", + "[Mila's official documentation](https://docs.mila.quebec/Information.html) shows which GPUs are installed on the cluster. Typing ```savail``` on the command line shows their current availability. \n", + "Testing their capacity can yield insights into their suitability for different training cases." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "GPU Avail / Total \n", + "===============================\n", + "2g.20gb 31 / 48 \n", + "3g.40gb 9 / 48 \n", + "4g.40gb 7 / 24 \n", + "a100 8 / 16 \n", + "a100l 0 / 72 \n", + "a6000 0 / 8 \n", + "rtx8000 11 / 400 \n", + "v100 2 / 40 \n" + ] + } + ], + "source": [ + "!savail" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We can observe the following prominent GPU classes: a100, a100l, a6000, rtx8000, v100. \n", + "We will now proceed to specify different GPUs over training runs and compare their throughput." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Add an example of a sweep over some parameters, \n", + "# with the training throughput as the metric, \n", + "# :: callbacks/samples_per_second, ### or add a devicestatsmonitor in\n", + "# and using different kinds of GPUs. " + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Making sense of the former: if a GPU with lower maximum capacity is readily available, training on it may be more time and resource effective than waiting for higher capacity GPUs to become available.\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Logging with Weights & Biases (wandb)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The Mila Research template integrates wandb functionality as a logger specification. \n", + "This has the advantage of being able to track additional metrics and create accompanying visualizations. \n", + "We will now create a wandb report comparing throughput between GPUs. \n" + ] + }, + { + "cell_type": "code", + "execution_count": 86, + "metadata": {}, + "outputs": [], + "source": [ + "# Create a wandb report with the throughput comparison \n", + "# between the different GPU types.\n", + "# i.e. specify wandb as the logger and log the throughput" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We would like to maximize our throughput given GPU choice" + ] + }, + { + "cell_type": "code", + "execution_count": 74, + "metadata": {}, + "outputs": [], + "source": [ + "## Find the best datamodule parameters to maximize the throughput \n", + "## (batches per second) without training (NoOP algo)" + ] + }, + { + "cell_type": "code", + "execution_count": 75, + "metadata": {}, + "outputs": [], + "source": [ + "### Measure the performance on different GPUS using the optimal datamodule \n", + "### params from before (and keeping other parameters the same)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We will now sweep over model hyper-parameters to maximize the utilization of our selected GPU." + ] + }, + { + "cell_type": "code", + "execution_count": 76, + "metadata": {}, + "outputs": [], + "source": [ + "#### Using the results from before, do a simple sweep over model hyper-parameters \n", + "#### to maximize the utilization of the selected GPU (which was selected as a tradeoff \n", + "#### between performance and difficulty to get an allocation). For example if the \n", + "#### RTX8000's are 20% slower than A100s but 5x easier to get an allocation on, use those instead." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Additional resources\n", + "\n", + "[GPU Training (Basic) - LightningAI](https://lightning.ai/docs/pytorch/stable/accelerators/gpu_basic.html) \n", + "[DeviceStatsMonitor class - LightningAI](https://lightning.ai/docs/pytorch/stable/api/lightning.pytorch.callbacks.DeviceStatsMonitor.html) \n", + "[PyTorch Profiler + W&B integration - Weights & Biases](https://wandb.ai/wandb/trace/reports/Using-the-PyTorch-Profiler-with-W-B--Vmlldzo5MDE3NjU)" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "researchtemplate", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.12.4" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/docs/getting_started/install.md b/docs/getting_started/install.md index 2384c9ff..75cbc5fe 100644 --- a/docs/getting_started/install.md +++ b/docs/getting_started/install.md @@ -29,7 +29,7 @@ You need to have [Conda](https://docs.conda.io/en/latest/) installed on your mac Notes: - - If you don't Conda installed, you can download it from [here](https://docs.conda.io/en/latest/miniconda.html). + - If you don't have Conda installed, you can download it from [here](https://docs.conda.io/en/latest/miniconda.html). - If you'd rather use a virtual environment instead of Conda, you can totally do so, as long as you have a version of Python >= 3.12. diff --git a/project/configs/algorithm/example.yaml b/project/configs/algorithm/example.yaml new file mode 100644 index 00000000..ab974cf4 --- /dev/null +++ b/project/configs/algorithm/example.yaml @@ -0,0 +1,2 @@ +_target_: project.algorithms.example.ExampleAlgorithm +_partial_: true diff --git a/project/configs/algorithm/example_from_config.yaml b/project/configs/algorithm/example_from_config.yaml deleted file mode 100644 index f79db6c6..00000000 --- a/project/configs/algorithm/example_from_config.yaml +++ /dev/null @@ -1,17 +0,0 @@ -defaults: - # Use the example as a schema for this config, and inherit its default values. - # BUG: This doesn't work when the lr scheduler or optimizer types change from their defaults, - # because OmegaConf seems to use the type of the value (not of the field?) when merging configs. - # - ExampleAlgorithm - - # Use a custom config for the Adam optimizer (optimizer/custom_adam.yaml) at hp.optimizer - - optimizer/custom_adam@hp.optimizer - # Apply the config for the StepLR learning rate scheduler at `hp.lr_scheduler` in this config. - - lr_scheduler/StepLR@hp.lr_scheduler - -_target_: project.algorithms.example.ExampleAlgorithm -_partial_: true -hp: - _target_: project.algorithms.example.ExampleAlgorithm.HParams - lr_scheduler: - step_size: 5 # Required argument for the StepLR scheduler. diff --git a/project/configs/algorithm/optimizer/custom_adam.yaml b/project/configs/algorithm/optimizer/custom_adam.yaml new file mode 100644 index 00000000..b1d27a41 --- /dev/null +++ b/project/configs/algorithm/optimizer/custom_adam.yaml @@ -0,0 +1,6 @@ +_target_: torch.optim.Adam +lr: 0.001 +betas: [0.9, 0.999] +eps: 1e-08 +weight_decay: 0 +amsgrad: false diff --git a/project/configs/config.yaml b/project/configs/config.yaml index 0d180a55..72a76392 100644 --- a/project/configs/config.yaml +++ b/project/configs/config.yaml @@ -2,7 +2,7 @@ defaults: - base_config - _self_ - datamodule: cifar10 - - algorithm: ExampleAlgorithm + - algorithm: example - network: resnet18 - trainer: default.yaml - trainer/callbacks: default.yaml diff --git a/project/configs/datamodule/imagenet.yaml b/project/configs/datamodule/imagenet.yaml index 3e82c78b..826d1f19 100644 --- a/project/configs/datamodule/imagenet.yaml +++ b/project/configs/datamodule/imagenet.yaml @@ -1,4 +1,7 @@ defaults: - vision _target_: project.datamodules.ImageNetDataModule +batch_size: 128 +train_transforms: + _target_: project.datamodules.image_classification.imagenet.imagenet_train_transforms # todo: add good configuration options here. diff --git a/project/configs/trainer/callbacks/default.yaml b/project/configs/trainer/callbacks/default.yaml index 519005d7..e7bcf79a 100644 --- a/project/configs/trainer/callbacks/default.yaml +++ b/project/configs/trainer/callbacks/default.yaml @@ -24,3 +24,6 @@ early_stopping: model_summary: max_depth: 2 + +throughput: + _target_: project.algorithms.callbacks.samples_per_second.MeasureSamplesPerSecondCallback \ No newline at end of file diff --git a/project/experiment.py b/project/experiment.py index af8c45ae..35616c2c 100644 --- a/project/experiment.py +++ b/project/experiment.py @@ -5,7 +5,6 @@ import logging import os import random -from collections.abc import Callable from dataclasses import dataclass, is_dataclass from logging import getLogger as get_logger from typing import Any @@ -15,7 +14,6 @@ import rich.logging import rich.traceback import torch -from hydra_zen.third_party.pydantic import pydantic_parser from lightning import Callback, LightningModule, Trainer, seed_everything from omegaconf import DictConfig from torch import nn @@ -32,12 +30,7 @@ logger = get_logger(__name__) -# todo: fix this. -def _use_pydantic[C: Callable](fn: C) -> C: - return functools.partial(hydra_zen.instantiate, _target_wrapper_=pydantic_parser) # type: ignore - - -instantiate = _use_pydantic(hydra_zen.instantiate) +instantiate = hydra_zen.instantiate @dataclass diff --git a/project/main.py b/project/main.py index 8910098e..50d001c8 100644 --- a/project/main.py +++ b/project/main.py @@ -94,7 +94,7 @@ def evaluation(experiment: Experiment) -> tuple[str, float | None, dict]: # We want to report the training error. metrics = { **experiment.trainer.logged_metrics, - **experiment.trainer.callback_metrics, + **experiment.trainer._metrics, **experiment.trainer.progress_bar_metrics, } rich.print(metrics) diff --git a/pyproject.toml b/pyproject.toml index 62f97b2b..a76bf0c0 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -31,6 +31,8 @@ dependencies = [ "simple-parsing>=0.1.5", "pydantic==2.7.4", "milatools>=0.0.18", + "rootutils>=0.0.1", + "ipykernel>=6.28.0" ] requires-python = ">=3.12" readme = "README.md" From 8e0dad2d314f758b7263df0c5c74180ab68325d3 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Thu, 8 Aug 2024 11:08:43 -0400 Subject: [PATCH 02/33] add profiling notebook, hotfix a few classes --- docs/examples/profiling.ipynb | 5905 +++++++++++++++++ docs/install.md | 6 +- project/configs/algorithm/example.yaml | 10 - .../algorithm/optimizer/custom_adam.yaml | 12 +- project/configs/datamodule/imagenet.yaml | 3 + .../configs/trainer/callbacks/default.yaml | 7 +- project/experiment.py | 10 +- project/main.py | 2 +- pyproject.toml | 4 + 9 files changed, 5927 insertions(+), 32 deletions(-) create mode 100644 docs/examples/profiling.ipynb diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb new file mode 100644 index 00000000..2ac62445 --- /dev/null +++ b/docs/examples/profiling.ipynb @@ -0,0 +1,5905 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Benchmarking" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Setup" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The Mila Research Template leverages built-in PyTorch and Lightning functionality to make benchmarking accesible and flexible. " + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "/home/mila/c/cesar.valdez/idt/ResearchTemplate\n", + "/home/mila/c/cesar.valdez/idt/ResearchTemplate\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/IPython/core/magics/osm.py:417: UserWarning: This is now an optional IPython functionality, setting dhist requires you to install the `pickleshare` library.\n", + " self.shell.db['dhist'] = compress_dhist(dhist)[-100:]\n" + ] + } + ], + "source": [ + "import os\n", + "import rootutils\n", + "\n", + "home_dir = rootutils.find_root(search_from=\"profiling.ipynb\", indicator=\".git\")\n", + "%cd $home_dir\n", + "print(os.getcwd())" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Finding bottlenecks: Dataloading vs Training" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "A potential use of .." + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2mCONFIG\u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2malgorithm\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mhydra_zen.funcs.zen_processing \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_zen_target\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.no_op.NoOp \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_zen_partial\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_zen_wrappers\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mhydra_zen.third_party.pydantic.pydantic_parser \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mnetwork\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtorchvision.models.resnet.resnet18 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mweights\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mprogress\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_classes\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${instance_attr:datamodule.num_classes} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mdatamodule\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.datamodules.ImageNetDataModule \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdata_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:DATA_DIR} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_workers\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:NUM_WORKERS} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mval_split\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnormalize\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mshuffle\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mpin_memory\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mseed\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m42 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mbatch_size\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m64 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mtrainer\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.Trainer \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlogger\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40maccelerator\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mstrategy\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdevices\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmin_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmax_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdefault_root_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdetect_anomaly\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mcallbacks\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_checkpoint\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.ModelCheckpoint \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdirpath\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir}/checkpoints \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mfilename\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mepoch_{epoch:03d} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_last\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_top_k\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mauto_insert_metric_name\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_weights_only\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_train_steps\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mtrain_time_interval\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mearly_stopping\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.EarlyStopping \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmin_delta\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.0 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mpatience\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m5 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstrict\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_finite\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstopping_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdivergence_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_summary\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichModelSummary \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmax_depth\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m2 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mrich_progress_bar\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichProgressBar \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mthroughput\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.callbacks.samples_per_second.MeasureSam\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_train_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_val_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mlog_level\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40minfo \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mseed\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40m87282 \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mname\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mNoOp-resnet18-imagenet \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mdebug\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", + "\u001b[2m└── \u001b[0m\u001b[2mverbose\u001b[0m\n", + "\u001b[2m \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", + "\u001b[2;36m[08/07/24 16:28:58]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=131611;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=866100;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m,\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", + "seed manually set to 87282\n", + "Seed set to 87282\n", + "/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=NoOp trainer.max_epochs=1 ...\n", + "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", + "GPU available: True (cuda), used: True\n", + "TPU available: False, using: 0 TPU cores\n", + "HPU available: False, using: 0 HPUs\n", + "\u001b[2;36m[16:28:58]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=867639;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=12083;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#163\u001b\\\u001b[2m163\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7fb2a6f64f80\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", + "Extracting train archive: 100%|████| 1000/1000 [10:10<00:00, 1.64Directories/s]\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", + "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", + "┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━┩\n", + "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNorm2d │ 128 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequential │ 147 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequential │ 525 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequential │ 2.1 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequential │ 8.4 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ AdaptiveAvgPool2d │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\n", + "└────┴─────────────────┴───────────────────┴────────┴───────┘\n", + "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", + "\u001b[1mNon-trainable params\u001b[0m: 0 \n", + "\u001b[1mTotal params\u001b[0m: 11.7 M \n", + "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/180\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0mt/s\u001b[0m \n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m100.703 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/180\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m100.703 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/180\u001b[0m \u001b[37m0:00:02 • 0:01:48\u001b[0m \u001b[37m1.64it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:35\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:35\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:31\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:31\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:26\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:26\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:22\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:22\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:05 • 0:01:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:05 • 0:01:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:07 • 0:01:19\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:07 • 0:01:19\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:07 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:07 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:08 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:08 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:08 • 0:01:19\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:08 • 0:01:19\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:09 • 0:01:17\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:09 • 0:01:17\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:09 • 0:01:16\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:09 • 0:01:16\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:11 • 0:01:13\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:11 • 0:01:13\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:13 • 0:01:11\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:13 • 0:01:11\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:13 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:13 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:14 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:14 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:14 • 0:01:09\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:14 • 0:01:09\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:14 • 0:01:08\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:14 • 0:01:08\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:16 • 0:01:07\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:16 • 0:01:07\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:16 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:16 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:17 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:17 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:17 • 0:01:05\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:17 • 0:01:05\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:17 • 0:01:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:17 • 0:01:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:25 • 0:00:56\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:25 • 0:00:56\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:26 • 0:00:56\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:26 • 0:00:56\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:26 • 0:00:55\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:26 • 0:00:55\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:27 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:27 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:28 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:28 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:28 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:28 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:29 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:29 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:32 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:32 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:33 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:33 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:33 • 0:00:46\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:33 • 0:00:46\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:34 • 0:00:46\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:34 • 0:00:46\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:34 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:34 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:35 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:35 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:37 • 0:00:41\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:37 • 0:00:41\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:38 • 0:00:41\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:38 • 0:00:41\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:38 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:38 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:39 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:39 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:39 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:39 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:40 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:40 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:42 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:42 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:44 • 0:00:33\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:44 • 0:00:33\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:45 • 0:00:33\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:45 • 0:00:33\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:46 • 0:00:33\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:46 • 0:00:33\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:47 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:47 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:47 • 0:00:31\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:47 • 0:00:31\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:48 • 0:00:31\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:48 • 0:00:31\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:48 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:48 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:49 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:49 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:49 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:49 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:50 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:50 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:51 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:51 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:51 • 0:00:27\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:51 • 0:00:27\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:52 • 0:00:27\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:52 • 0:00:27\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:52 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:52 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:54 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:54 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:55 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:55 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:55 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:55 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:00:56 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:00:56 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:00:57 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:00:57 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:00:57 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:00:57 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:00:58 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:00:58 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:00:59 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:00:59 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:00:59 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:00:59 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:00 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:00 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:05 • 0:00:14\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:05 • 0:00:14\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:07 • 0:00:12\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:07 • 0:00:12\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:07 • 0:00:11\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:07 • 0:00:11\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:08 • 0:00:11\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:08 • 0:00:11\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:08 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:08 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:10 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:10 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:11 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:11 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:11 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:11 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:12 • 0:00:06\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:12 • 0:00:06\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:13 • 0:00:06\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:13 • 0:00:06\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:13 • 0:00:05\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:13 • 0:00:05\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:14 • 0:00:05\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:14 • 0:00:05\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:14 • 0:00:04\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:14 • 0:00:04\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:16 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:16 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:17 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:17 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:17 • 0:00:01\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:17 • 0:00:01\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:18 • 0:00:01\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:18 • 0:00:01\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m149.625 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m151.611 \u001b[0m\n", + " \u001b[37mval/samples_per_…\u001b[0m\n", + " \u001b[37m120.811 \u001b[0m\n", + " \u001b[37mval/samples_per_…\u001b[0m\n", + " \u001b[37m128.323 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m150.129 \u001b[0m\n", + "\u001b[?25h/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=NoOp trainer.max_epochs=1 ...\n", + "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", + "\u001b[2;36m[16:41:35]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=972670;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=502992;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#381\u001b\\\u001b[2m381\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m[16:41:36]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=825781;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=661210;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37m2.10it/s\u001b[0m \n", + "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", + "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", + "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4964688718318939 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 140.0721893310547 \u001b[0m\u001b[35m \u001b[0m│\n", + "└──────────────────────────────┴──────────────────────────────┘\n", + "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.11it/s\u001b[0m \n", + "\u001b[?25hval val/samples_per_second_epoch: \u001b[1;36m140.0721893310547\u001b[0m\n" + ] + } + ], + "source": [ + "!python project/main.py \\\n", + " algorithm=NoOp \\\n", + " trainer.max_epochs=1 \\\n", + " +trainer.limit_train_batches=0.01\\\n", + " +trainer.limit_val_batches=0.01\\\n", + " datamodule=imagenet" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2mCONFIG\u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2malgorithm\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.example.ExampleAlgorithm \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_partial_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mnetwork\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtorchvision.models.resnet.resnet18 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mweights\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mprogress\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_classes\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${instance_attr:datamodule.num_classes} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mdatamodule\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.datamodules.ImageNetDataModule \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdata_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:DATA_DIR} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_workers\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:NUM_WORKERS} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mval_split\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnormalize\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mshuffle\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mpin_memory\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mseed\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m42 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mbatch_size\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m64 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mtrainer\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.Trainer \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlogger\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40maccelerator\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mstrategy\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdevices\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmin_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmax_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdefault_root_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdetect_anomaly\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mcallbacks\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_checkpoint\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.ModelCheckpoint \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdirpath\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir}/checkpoints \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mfilename\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mepoch_{epoch:03d} \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_last\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_top_k\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mauto_insert_metric_name\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_weights_only\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_train_steps\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mtrain_time_interval\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mearly_stopping\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.EarlyStopping \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmin_delta\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.0 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mpatience\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m5 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstrict\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_finite\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstopping_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdivergence_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_summary\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichModelSummary \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmax_depth\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m2 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mrich_progress_bar\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichProgressBar \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mthroughput\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.callbacks.samples_per_second.MeasureSam\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_train_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_val_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mlog_level\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40minfo \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mseed\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40m81570 \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mname\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mexample-resnet18-imagenet \u001b[0m\n", + "\u001b[2m├── \u001b[0m\u001b[2mdebug\u001b[0m\n", + "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", + "\u001b[2m└── \u001b[0m\u001b[2mverbose\u001b[0m\n", + "\u001b[2m \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", + "\u001b[2;36m[08/07/24 17:07:38]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=50545;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=71944;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m,\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", + "seed manually set to 81570\n", + "Seed set to 81570\n", + "/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=example trainer.max_epochs ...\n", + "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", + "GPU available: True (cuda), used: True\n", + "TPU available: False, using: 0 TPU cores\n", + "HPU available: False, using: 0 HPUs\n", + "\u001b[2;36m[17:07:38]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=285735;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=912000;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#156\u001b\\\u001b[2m156\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7eff7e751f10\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m[17:07:43]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=702216;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=293228;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m train_accuracy. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=476337;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=48315;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m val_accuracy. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=513180;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=101826;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m test_accuracy. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=370748;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=635161;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m train_top5_accuracy. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=727524;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=265133;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m val_top5_accuracy. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=753886;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=808003;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m test_top5_accuracy. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=781296;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=524717;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#381\u001b\\\u001b[2m381\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m[17:07:44]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=248066;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=550199;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m[17:08:02]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Not adding metrics to the pl \u001b]8;id=656135;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=914060;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#64\u001b\\\u001b[2m64\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module because they are already \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m present. \u001b[2m \u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "┏━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━┓\n", + "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mIn sizes\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mOut sizes\u001b[0m\u001b[1;35m \u001b[0m┃\n", + "┡━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━┩\n", + "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\u001b[37m \u001b[0m\u001b[37m [64, 3,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1000]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\u001b[37m \u001b[0m\u001b[37m [64, 3,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNor… │ 128 │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequenti… │ 147 K │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequenti… │ 525 K │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 128,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 28, 28]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequenti… │ 2.1 M │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 256,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m128, 28,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 14, 14]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 28]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequenti… │ 8.4 M │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 512,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m256, 14,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 7, 7]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 14]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ Adaptive… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 512,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 512, 7,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1, 1]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 7]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 512]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1000]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m11\u001b[0m\u001b[2m \u001b[0m│ train_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m12\u001b[0m\u001b[2m \u001b[0m│ val_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m13\u001b[0m\u001b[2m \u001b[0m│ test_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m14\u001b[0m\u001b[2m \u001b[0m│ train_top5_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m15\u001b[0m\u001b[2m \u001b[0m│ val_top5_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m16\u001b[0m\u001b[2m \u001b[0m│ test_top5_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", + "└────┴─────────────────────┴───────────┴────────┴───────┴──────────┴───────────┘\n", + "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", + "\u001b[1mNon-trainable params\u001b[0m: 0 \n", + "\u001b[1mTotal params\u001b[0m: 11.7 M \n", + "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/180\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0mt/s\u001b[0m \n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m51.620 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/180\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m51.620 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/180\u001b[0m \u001b[37m0:00:01 • 0:00:21\u001b[0m \u001b[37m8.90it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.059 train/loss: \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/180\u001b[0m \u001b[37m0:00:02 • 0:01:01\u001b[0m \u001b[37m2.92it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.059 train/loss: \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/180\u001b[0m \u001b[37m0:00:02 • 0:01:01\u001b[0m \u001b[37m2.92it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.015 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m6.940 train/loss: \u001b[0m\n", + " \u001b[37m6.940 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:04\u001b[0m \u001b[37m2.76it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.015 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m6.940 train/loss: \u001b[0m\n", + " \u001b[37m6.940 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:04\u001b[0m \u001b[37m2.76it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.146 train/loss: \u001b[0m\n", + " \u001b[37m7.146 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:08\u001b[0m \u001b[37m2.58it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.146 train/loss: \u001b[0m\n", + " \u001b[37m7.146 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:08\u001b[0m \u001b[37m2.58it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.059 train/loss: \u001b[0m\n", + " \u001b[37m7.059 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.44it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.059 train/loss: \u001b[0m\n", + " \u001b[37m7.059 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.44it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.076 train/loss: \u001b[0m\n", + " \u001b[37m7.076 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.43it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.076 train/loss: \u001b[0m\n", + " \u001b[37m7.076 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.43it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.096 train/loss: \u001b[0m\n", + " \u001b[37m7.096 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.096 train/loss: \u001b[0m\n", + " \u001b[37m7.096 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.142 train/loss: \u001b[0m\n", + " \u001b[37m7.142 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.142 train/loss: \u001b[0m\n", + " \u001b[37m7.142 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accura…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entro…\u001b[0m\n", + " \u001b[37m7.030 train/loss: \u001b[0m\n", + " \u001b[37m7.030 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:19\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.030 train/loss:\u001b[0m\n", + " \u001b[37m7.030 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:19\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.100 train/loss:\u001b[0m\n", + " \u001b[37m7.100 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:06 • 0:01:31\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.100 train/loss:\u001b[0m\n", + " \u001b[37m7.100 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:06 • 0:01:31\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.008 train/loss:\u001b[0m\n", + " \u001b[37m7.008 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:07 • 0:01:28\u001b[0m \u001b[37m1.91it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.008 train/loss:\u001b[0m\n", + " \u001b[37m7.008 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:07 • 0:01:28\u001b[0m \u001b[37m1.91it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.302 train/loss:\u001b[0m\n", + " \u001b[37m7.302 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:07 • 0:01:26\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.302 train/loss:\u001b[0m\n", + " \u001b[37m7.302 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:07 • 0:01:26\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.344 train/loss:\u001b[0m\n", + " \u001b[37m7.344 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:08 • 0:01:30\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.344 train/loss:\u001b[0m\n", + " \u001b[37m7.344 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:08 • 0:01:30\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.050 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.361 train/loss:\u001b[0m\n", + " \u001b[37m7.361 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:08 • 0:01:28\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.050 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.361 train/loss:\u001b[0m\n", + " \u001b[37m7.361 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:08 • 0:01:28\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.182 train/loss:\u001b[0m\n", + " \u001b[37m7.182 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:10 • 0:01:35\u001b[0m \u001b[37m1.73it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.182 train/loss:\u001b[0m\n", + " \u001b[37m7.182 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:10 • 0:01:35\u001b[0m \u001b[37m1.73it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.168 train/loss:\u001b[0m\n", + " \u001b[37m7.168 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:10 • 0:01:37\u001b[0m \u001b[37m1.69it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.168 train/loss:\u001b[0m\n", + " \u001b[37m7.168 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:10 • 0:01:37\u001b[0m \u001b[37m1.69it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.324 train/loss:\u001b[0m\n", + " \u001b[37m7.324 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:11 • 0:01:35\u001b[0m \u001b[37m1.72it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.324 train/loss:\u001b[0m\n", + " \u001b[37m7.324 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:11 • 0:01:35\u001b[0m \u001b[37m1.72it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.436 train/loss:\u001b[0m\n", + " \u001b[37m7.436 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:11 • 0:01:32\u001b[0m \u001b[37m1.75it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.436 train/loss:\u001b[0m\n", + " \u001b[37m7.436 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:11 • 0:01:32\u001b[0m \u001b[37m1.75it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.111 train/loss:\u001b[0m\n", + " \u001b[37m7.111 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:12 • 0:01:31\u001b[0m \u001b[37m1.78it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.111 train/loss:\u001b[0m\n", + " \u001b[37m7.111 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:12 • 0:01:31\u001b[0m \u001b[37m1.78it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.521 train/loss:\u001b[0m\n", + " \u001b[37m7.521 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:12 • 0:01:29\u001b[0m \u001b[37m1.79it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.521 train/loss:\u001b[0m\n", + " \u001b[37m7.521 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:12 • 0:01:29\u001b[0m \u001b[37m1.79it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.065 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.163 train/loss:\u001b[0m\n", + " \u001b[37m7.163 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:13 • 0:01:28\u001b[0m \u001b[37m1.81it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.065 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.163 train/loss:\u001b[0m\n", + " \u001b[37m7.163 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:13 • 0:01:28\u001b[0m \u001b[37m1.81it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.038 train/loss:\u001b[0m\n", + " \u001b[37m7.038 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:13 • 0:01:26\u001b[0m \u001b[37m1.83it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.038 train/loss:\u001b[0m\n", + " \u001b[37m7.038 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:13 • 0:01:26\u001b[0m \u001b[37m1.83it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.171 train/loss:\u001b[0m\n", + " \u001b[37m7.171 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:13 • 0:01:25\u001b[0m \u001b[37m1.85it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.171 train/loss:\u001b[0m\n", + " \u001b[37m7.171 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:13 • 0:01:25\u001b[0m \u001b[37m1.85it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.327 train/loss:\u001b[0m\n", + " \u001b[37m7.327 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:14 • 0:01:23\u001b[0m \u001b[37m1.87it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.327 train/loss:\u001b[0m\n", + " \u001b[37m7.327 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:14 • 0:01:23\u001b[0m \u001b[37m1.87it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.207 train/loss:\u001b[0m\n", + " \u001b[37m7.207 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:14 • 0:01:22\u001b[0m \u001b[37m1.90it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.207 train/loss:\u001b[0m\n", + " \u001b[37m7.207 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:14 • 0:01:22\u001b[0m \u001b[37m1.90it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.271 train/loss:\u001b[0m\n", + " \u001b[37m7.271 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:15 • 0:01:20\u001b[0m \u001b[37m1.92it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.271 train/loss:\u001b[0m\n", + " \u001b[37m7.271 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:15 • 0:01:20\u001b[0m \u001b[37m1.92it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.566 train/loss:\u001b[0m\n", + " \u001b[37m7.566 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:15 • 0:01:19\u001b[0m \u001b[37m1.93it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.566 train/loss:\u001b[0m\n", + " \u001b[37m7.566 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:15 • 0:01:19\u001b[0m \u001b[37m1.93it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.203 train/loss:\u001b[0m\n", + " \u001b[37m7.203 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:15 • 0:01:18\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.203 train/loss:\u001b[0m\n", + " \u001b[37m7.203 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:15 • 0:01:18\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.262 train/loss:\u001b[0m\n", + " \u001b[37m7.262 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:16 • 0:01:17\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.262 train/loss:\u001b[0m\n", + " \u001b[37m7.262 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:16 • 0:01:17\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.109 train/loss:\u001b[0m\n", + " \u001b[37m7.109 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:16 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.109 train/loss:\u001b[0m\n", + " \u001b[37m7.109 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:16 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.343 train/loss:\u001b[0m\n", + " \u001b[37m7.343 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:17 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.343 train/loss:\u001b[0m\n", + " \u001b[37m7.343 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:17 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.190 train/loss:\u001b[0m\n", + " \u001b[37m7.190 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:17 • 0:01:15\u001b[0m \u001b[37m1.98it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.190 train/loss:\u001b[0m\n", + " \u001b[37m7.190 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:17 • 0:01:15\u001b[0m \u001b[37m1.98it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.051 train/loss:\u001b[0m\n", + " \u001b[37m7.051 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:18 • 0:01:14\u001b[0m \u001b[37m1.99it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.051 train/loss:\u001b[0m\n", + " \u001b[37m7.051 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:18 • 0:01:14\u001b[0m \u001b[37m1.99it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.174 train/loss:\u001b[0m\n", + " \u001b[37m7.174 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:18 • 0:01:13\u001b[0m \u001b[37m2.00it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.174 train/loss:\u001b[0m\n", + " \u001b[37m7.174 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:18 • 0:01:13\u001b[0m \u001b[37m2.00it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.390 train/loss:\u001b[0m\n", + " \u001b[37m7.390 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:18 • 0:01:12\u001b[0m \u001b[37m2.01it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.390 train/loss:\u001b[0m\n", + " \u001b[37m7.390 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:18 • 0:01:12\u001b[0m \u001b[37m2.01it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.300 train/loss:\u001b[0m\n", + " \u001b[37m7.300 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:19 • 0:01:11\u001b[0m \u001b[37m2.02it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.300 train/loss:\u001b[0m\n", + " \u001b[37m7.300 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:19 • 0:01:11\u001b[0m \u001b[37m2.02it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.237 train/loss:\u001b[0m\n", + " \u001b[37m7.237 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:19 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.237 train/loss:\u001b[0m\n", + " \u001b[37m7.237 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:19 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.232 train/loss:\u001b[0m\n", + " \u001b[37m7.232 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:20 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.232 train/loss:\u001b[0m\n", + " \u001b[37m7.232 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:20 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.971 train/loss:\u001b[0m\n", + " \u001b[37m6.971 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:20 • 0:01:09\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.971 train/loss:\u001b[0m\n", + " \u001b[37m6.971 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:20 • 0:01:09\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.262 train/loss:\u001b[0m\n", + " \u001b[37m7.262 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.262 train/loss:\u001b[0m\n", + " \u001b[37m7.262 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.142 train/loss:\u001b[0m\n", + " \u001b[37m7.142 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.142 train/loss:\u001b[0m\n", + " \u001b[37m7.142 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.984 train/loss:\u001b[0m\n", + " \u001b[37m6.984 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.984 train/loss:\u001b[0m\n", + " \u001b[37m6.984 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.149 train/loss:\u001b[0m\n", + " \u001b[37m7.149 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.149 train/loss:\u001b[0m\n", + " \u001b[37m7.149 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.986 train/loss:\u001b[0m\n", + " \u001b[37m6.986 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:22 • 0:01:06\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.986 train/loss:\u001b[0m\n", + " \u001b[37m6.986 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:22 • 0:01:06\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.067 train/loss:\u001b[0m\n", + " \u001b[37m7.067 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:23 • 0:01:06\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.067 train/loss:\u001b[0m\n", + " \u001b[37m7.067 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:23 • 0:01:06\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.224 train/loss:\u001b[0m\n", + " \u001b[37m7.224 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:23 • 0:01:05\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.224 train/loss:\u001b[0m\n", + " \u001b[37m7.224 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:23 • 0:01:05\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.053 train/loss:\u001b[0m\n", + " \u001b[37m7.053 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:24 • 0:01:05\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.053 train/loss:\u001b[0m\n", + " \u001b[37m7.053 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:24 • 0:01:05\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.016 train/loss:\u001b[0m\n", + " \u001b[37m7.016 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:24 • 0:01:04\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.016 train/loss:\u001b[0m\n", + " \u001b[37m7.016 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:24 • 0:01:04\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.050 train/loss:\u001b[0m\n", + " \u001b[37m7.050 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.050 train/loss:\u001b[0m\n", + " \u001b[37m7.050 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.025 train/loss:\u001b[0m\n", + " \u001b[37m7.025 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.025 train/loss:\u001b[0m\n", + " \u001b[37m7.025 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.886 train/loss:\u001b[0m\n", + " \u001b[37m6.886 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:25 • 0:01:02\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.886 train/loss:\u001b[0m\n", + " \u001b[37m6.886 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:25 • 0:01:02\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.926 train/loss:\u001b[0m\n", + " \u001b[37m6.926 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.926 train/loss:\u001b[0m\n", + " \u001b[37m6.926 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.090 train/loss:\u001b[0m\n", + " \u001b[37m7.090 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.090 train/loss:\u001b[0m\n", + " \u001b[37m7.090 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.913 train/loss:\u001b[0m\n", + " \u001b[37m6.913 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.913 train/loss:\u001b[0m\n", + " \u001b[37m6.913 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.100 train/loss:\u001b[0m\n", + " \u001b[37m7.100 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.100 train/loss:\u001b[0m\n", + " \u001b[37m7.100 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.028 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.985 train/loss:\u001b[0m\n", + " \u001b[37m6.985 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:28 • 0:00:59\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.028 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.985 train/loss:\u001b[0m\n", + " \u001b[37m6.985 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:28 • 0:00:59\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.191 train/loss:\u001b[0m\n", + " \u001b[37m7.191 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.191 train/loss:\u001b[0m\n", + " \u001b[37m7.191 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.939 train/loss:\u001b[0m\n", + " \u001b[37m6.939 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.939 train/loss:\u001b[0m\n", + " \u001b[37m6.939 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.859 train/loss:\u001b[0m\n", + " \u001b[37m6.859 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.859 train/loss:\u001b[0m\n", + " \u001b[37m6.859 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.076 train/loss:\u001b[0m\n", + " \u001b[37m7.076 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.076 train/loss:\u001b[0m\n", + " \u001b[37m7.076 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.115 train/loss:\u001b[0m\n", + " \u001b[37m7.115 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:30 • 0:00:56\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.115 train/loss:\u001b[0m\n", + " \u001b[37m7.115 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:30 • 0:00:56\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.936 train/loss:\u001b[0m\n", + " \u001b[37m6.936 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.936 train/loss:\u001b[0m\n", + " \u001b[37m6.936 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.901 train/loss:\u001b[0m\n", + " \u001b[37m6.901 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.901 train/loss:\u001b[0m\n", + " \u001b[37m6.901 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.068 train/loss:\u001b[0m\n", + " \u001b[37m7.068 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.068 train/loss:\u001b[0m\n", + " \u001b[37m7.068 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.932 train/loss:\u001b[0m\n", + " \u001b[37m6.932 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.932 train/loss:\u001b[0m\n", + " \u001b[37m6.932 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.934 train/loss:\u001b[0m\n", + " \u001b[37m6.934 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.934 train/loss:\u001b[0m\n", + " \u001b[37m6.934 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.022 train/loss:\u001b[0m\n", + " \u001b[37m7.022 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.022 train/loss:\u001b[0m\n", + " \u001b[37m7.022 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.905 train/loss:\u001b[0m\n", + " \u001b[37m6.905 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.905 train/loss:\u001b[0m\n", + " \u001b[37m6.905 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.929 train/loss:\u001b[0m\n", + " \u001b[37m6.929 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.929 train/loss:\u001b[0m\n", + " \u001b[37m6.929 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.060 train/loss:\u001b[0m\n", + " \u001b[37m7.060 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.060 train/loss:\u001b[0m\n", + " \u001b[37m7.060 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.089 train/loss:\u001b[0m\n", + " \u001b[37m7.089 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:34 • 0:00:51\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.089 train/loss:\u001b[0m\n", + " \u001b[37m7.089 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:34 • 0:00:51\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.011 train/loss:\u001b[0m\n", + " \u001b[37m7.011 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:34 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.011 train/loss:\u001b[0m\n", + " \u001b[37m7.011 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:34 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.953 train/loss:\u001b[0m\n", + " \u001b[37m6.953 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:35 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.953 train/loss:\u001b[0m\n", + " \u001b[37m6.953 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:35 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.890 train/loss:\u001b[0m\n", + " \u001b[37m6.890 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:35 • 0:00:48\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.890 train/loss:\u001b[0m\n", + " \u001b[37m6.890 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:35 • 0:00:48\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.829 train/loss:\u001b[0m\n", + " \u001b[37m6.829 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.829 train/loss:\u001b[0m\n", + " \u001b[37m6.829 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.009 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.902 train/loss:\u001b[0m\n", + " \u001b[37m6.902 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.009 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.902 train/loss:\u001b[0m\n", + " \u001b[37m6.902 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.996 train/loss:\u001b[0m\n", + " \u001b[37m6.996 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:37 • 0:00:47\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.996 train/loss:\u001b[0m\n", + " \u001b[37m6.996 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:37 • 0:00:47\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.074 train/loss:\u001b[0m\n", + " \u001b[37m7.074 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:37 • 0:00:46\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.074 train/loss:\u001b[0m\n", + " \u001b[37m7.074 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:37 • 0:00:46\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.826 train/loss:\u001b[0m\n", + " \u001b[37m6.826 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.826 train/loss:\u001b[0m\n", + " \u001b[37m6.826 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.075 train/loss:\u001b[0m\n", + " \u001b[37m7.075 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.075 train/loss:\u001b[0m\n", + " \u001b[37m7.075 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.881 train/loss:\u001b[0m\n", + " \u001b[37m6.881 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.28it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.881 train/loss:\u001b[0m\n", + " \u001b[37m6.881 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.28it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.860 train/loss:\u001b[0m\n", + " \u001b[37m6.860 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.860 train/loss:\u001b[0m\n", + " \u001b[37m6.860 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.005 train/loss:\u001b[0m\n", + " \u001b[37m7.005 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:39 • 0:00:42\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.005 train/loss:\u001b[0m\n", + " \u001b[37m7.005 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:39 • 0:00:42\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.864 train/loss:\u001b[0m\n", + " \u001b[37m6.864 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:40 • 0:00:42\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.864 train/loss:\u001b[0m\n", + " \u001b[37m6.864 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:40 • 0:00:42\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.953 train/loss:\u001b[0m\n", + " \u001b[37m6.953 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:40 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.953 train/loss:\u001b[0m\n", + " \u001b[37m6.953 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:40 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.968 train/loss:\u001b[0m\n", + " \u001b[37m6.968 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:41 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.968 train/loss:\u001b[0m\n", + " \u001b[37m6.968 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:41 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.973 train/loss:\u001b[0m\n", + " \u001b[37m6.973 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:41 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.973 train/loss:\u001b[0m\n", + " \u001b[37m6.973 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:41 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.986 train/loss:\u001b[0m\n", + " \u001b[37m6.986 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:42 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.986 train/loss:\u001b[0m\n", + " \u001b[37m6.986 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:42 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.007 train/loss:\u001b[0m\n", + " \u001b[37m7.007 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.007 train/loss:\u001b[0m\n", + " \u001b[37m7.007 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.029 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.940 train/loss:\u001b[0m\n", + " \u001b[37m6.940 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.029 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.940 train/loss:\u001b[0m\n", + " \u001b[37m6.940 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.932 train/loss:\u001b[0m\n", + " \u001b[37m6.932 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:43 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.932 train/loss:\u001b[0m\n", + " \u001b[37m6.932 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:43 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.997 train/loss:\u001b[0m\n", + " \u001b[37m6.997 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:43 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.997 train/loss:\u001b[0m\n", + " \u001b[37m6.997 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:43 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.936 train/loss:\u001b[0m\n", + " \u001b[37m6.936 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:44 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.936 train/loss:\u001b[0m\n", + " \u001b[37m6.936 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:44 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.014 train/loss:\u001b[0m\n", + " \u001b[37m7.014 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:44 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.014 train/loss:\u001b[0m\n", + " \u001b[37m7.014 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:44 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.937 train/loss:\u001b[0m\n", + " \u001b[37m6.937 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.937 train/loss:\u001b[0m\n", + " \u001b[37m6.937 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.915 train/loss:\u001b[0m\n", + " \u001b[37m6.915 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.915 train/loss:\u001b[0m\n", + " \u001b[37m6.915 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.923 train/loss:\u001b[0m\n", + " \u001b[37m6.923 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:45 • 0:00:36\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.923 train/loss:\u001b[0m\n", + " \u001b[37m6.923 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:45 • 0:00:36\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.014 train/loss:\u001b[0m\n", + " \u001b[37m7.014 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:46 • 0:00:36\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.014 train/loss:\u001b[0m\n", + " \u001b[37m7.014 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:46 • 0:00:36\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.963 train/loss:\u001b[0m\n", + " \u001b[37m6.963 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:46 • 0:00:35\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.963 train/loss:\u001b[0m\n", + " \u001b[37m6.963 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:46 • 0:00:35\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.873 train/loss:\u001b[0m\n", + " \u001b[37m6.873 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:47 • 0:00:35\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.873 train/loss:\u001b[0m\n", + " \u001b[37m6.873 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:47 • 0:00:35\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.033 train/loss:\u001b[0m\n", + " \u001b[37m7.033 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:47 • 0:00:34\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.033 train/loss:\u001b[0m\n", + " \u001b[37m7.033 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:47 • 0:00:34\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.034 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.787 train/loss:\u001b[0m\n", + " \u001b[37m6.787 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:48 • 0:00:35\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.034 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.787 train/loss:\u001b[0m\n", + " \u001b[37m6.787 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:48 • 0:00:35\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.946 train/loss:\u001b[0m\n", + " \u001b[37m6.946 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.946 train/loss:\u001b[0m\n", + " \u001b[37m6.946 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.966 train/loss:\u001b[0m\n", + " \u001b[37m6.966 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.966 train/loss:\u001b[0m\n", + " \u001b[37m6.966 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.899 train/loss:\u001b[0m\n", + " \u001b[37m6.899 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:49 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.899 train/loss:\u001b[0m\n", + " \u001b[37m6.899 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:49 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.031 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.886 train/loss:\u001b[0m\n", + " \u001b[37m6.886 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:50 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.031 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.886 train/loss:\u001b[0m\n", + " \u001b[37m6.886 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:50 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.892 train/loss:\u001b[0m\n", + " \u001b[37m6.892 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:51 • 0:00:33\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.892 train/loss:\u001b[0m\n", + " \u001b[37m6.892 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:51 • 0:00:33\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.880 train/loss:\u001b[0m\n", + " \u001b[37m6.880 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:51 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.880 train/loss:\u001b[0m\n", + " \u001b[37m6.880 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:51 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.897 train/loss:\u001b[0m\n", + " \u001b[37m6.897 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.897 train/loss:\u001b[0m\n", + " \u001b[37m6.897 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.891 train/loss:\u001b[0m\n", + " \u001b[37m6.891 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.891 train/loss:\u001b[0m\n", + " \u001b[37m6.891 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.979 train/loss:\u001b[0m\n", + " \u001b[37m6.979 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:53 • 0:00:31\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.979 train/loss:\u001b[0m\n", + " \u001b[37m6.979 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:53 • 0:00:31\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.860 train/loss:\u001b[0m\n", + " \u001b[37m6.860 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.860 train/loss:\u001b[0m\n", + " \u001b[37m6.860 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.014 train/loss:\u001b[0m\n", + " \u001b[37m7.014 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.014 train/loss:\u001b[0m\n", + " \u001b[37m7.014 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.908 train/loss:\u001b[0m\n", + " \u001b[37m6.908 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.908 train/loss:\u001b[0m\n", + " \u001b[37m6.908 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.802 train/loss:\u001b[0m\n", + " \u001b[37m6.802 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.802 train/loss:\u001b[0m\n", + " \u001b[37m6.802 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.839 train/loss:\u001b[0m\n", + " \u001b[37m6.839 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:56 • 0:00:30\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.839 train/loss:\u001b[0m\n", + " \u001b[37m6.839 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:56 • 0:00:30\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.880 train/loss:\u001b[0m\n", + " \u001b[37m6.880 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.880 train/loss:\u001b[0m\n", + " \u001b[37m6.880 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.812 train/loss:\u001b[0m\n", + " \u001b[37m6.812 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.812 train/loss:\u001b[0m\n", + " \u001b[37m6.812 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.848 train/loss:\u001b[0m\n", + " \u001b[37m6.848 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.848 train/loss:\u001b[0m\n", + " \u001b[37m6.848 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.946 train/loss:\u001b[0m\n", + " \u001b[37m6.946 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.946 train/loss:\u001b[0m\n", + " \u001b[37m6.946 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.939 train/loss:\u001b[0m\n", + " \u001b[37m6.939 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.939 train/loss:\u001b[0m\n", + " \u001b[37m6.939 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.839 train/loss:\u001b[0m\n", + " \u001b[37m6.839 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.839 train/loss:\u001b[0m\n", + " \u001b[37m6.839 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.827 train/loss:\u001b[0m\n", + " \u001b[37m6.827 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.827 train/loss:\u001b[0m\n", + " \u001b[37m6.827 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.862 train/loss:\u001b[0m\n", + " \u001b[37m6.862 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.862 train/loss:\u001b[0m\n", + " \u001b[37m6.862 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.034 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.838 train/loss:\u001b[0m\n", + " \u001b[37m6.838 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.034 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.838 train/loss:\u001b[0m\n", + " \u001b[37m6.838 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.004 train/loss:\u001b[0m\n", + " \u001b[37m7.004 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.004 train/loss:\u001b[0m\n", + " \u001b[37m7.004 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.028 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.801 train/loss:\u001b[0m\n", + " \u001b[37m6.801 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.028 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.801 train/loss:\u001b[0m\n", + " \u001b[37m6.801 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.042 train/loss:\u001b[0m\n", + " \u001b[37m7.042 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m7.042 train/loss:\u001b[0m\n", + " \u001b[37m7.042 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.851 train/loss:\u001b[0m\n", + " \u001b[37m6.851 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.851 train/loss:\u001b[0m\n", + " \u001b[37m6.851 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.904 train/loss:\u001b[0m\n", + " \u001b[37m6.904 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:01:01 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.904 train/loss:\u001b[0m\n", + " \u001b[37m6.904 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:01:02 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.850 train/loss:\u001b[0m\n", + " \u001b[37m6.850 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:01:02 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.850 train/loss:\u001b[0m\n", + " \u001b[37m6.850 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:01:02 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.885 train/loss:\u001b[0m\n", + " \u001b[37m6.885 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:01:02 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.885 train/loss:\u001b[0m\n", + " \u001b[37m6.885 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:01:02 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.815 train/loss:\u001b[0m\n", + " \u001b[37m6.815 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:01:03 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.815 train/loss:\u001b[0m\n", + " \u001b[37m6.815 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:01:03 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.871 train/loss:\u001b[0m\n", + " \u001b[37m6.871 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:01:03 • 0:00:21\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.871 train/loss:\u001b[0m\n", + " \u001b[37m6.871 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:01:03 • 0:00:21\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.864 train/loss:\u001b[0m\n", + " \u001b[37m6.864 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:01:04 • 0:00:21\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.864 train/loss:\u001b[0m\n", + " \u001b[37m6.864 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:01:04 • 0:00:21\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.936 train/loss:\u001b[0m\n", + " \u001b[37m6.936 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.936 train/loss:\u001b[0m\n", + " \u001b[37m6.936 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.031 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.878 train/loss:\u001b[0m\n", + " \u001b[37m6.878 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.031 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.878 train/loss:\u001b[0m\n", + " \u001b[37m6.878 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.026 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.902 train/loss:\u001b[0m\n", + " \u001b[37m6.902 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.026 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.902 train/loss:\u001b[0m\n", + " \u001b[37m6.902 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.955 train/loss:\u001b[0m\n", + " \u001b[37m6.955 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.955 train/loss:\u001b[0m\n", + " \u001b[37m6.955 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.852 train/loss:\u001b[0m\n", + " \u001b[37m6.852 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.852 train/loss:\u001b[0m\n", + " \u001b[37m6.852 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.869 train/loss:\u001b[0m\n", + " \u001b[37m6.869 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.048 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.869 train/loss:\u001b[0m\n", + " \u001b[37m6.869 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.902 train/loss:\u001b[0m\n", + " \u001b[37m6.902 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.902 train/loss:\u001b[0m\n", + " \u001b[37m6.902 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.872 train/loss:\u001b[0m\n", + " \u001b[37m6.872 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.872 train/loss:\u001b[0m\n", + " \u001b[37m6.872 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.959 train/loss:\u001b[0m\n", + " \u001b[37m6.959 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:07 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.959 train/loss:\u001b[0m\n", + " \u001b[37m6.959 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:07 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.851 train/loss:\u001b[0m\n", + " \u001b[37m6.851 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.851 train/loss:\u001b[0m\n", + " \u001b[37m6.851 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.966 train/loss:\u001b[0m\n", + " \u001b[37m6.966 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.966 train/loss:\u001b[0m\n", + " \u001b[37m6.966 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.838 train/loss:\u001b[0m\n", + " \u001b[37m6.838 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.032 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.838 train/loss:\u001b[0m\n", + " \u001b[37m6.838 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.921 train/loss:\u001b[0m\n", + " \u001b[37m6.921 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.921 train/loss:\u001b[0m\n", + " \u001b[37m6.921 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.838 train/loss:\u001b[0m\n", + " \u001b[37m6.838 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.838 train/loss:\u001b[0m\n", + " \u001b[37m6.838 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.012 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.876 train/loss:\u001b[0m\n", + " \u001b[37m6.876 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.012 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.876 train/loss:\u001b[0m\n", + " \u001b[37m6.876 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.829 train/loss:\u001b[0m\n", + " \u001b[37m6.829 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:10 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.008 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.829 train/loss:\u001b[0m\n", + " \u001b[37m6.829 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:10 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.862 train/loss:\u001b[0m\n", + " \u001b[37m6.862 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:11 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.017 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.862 train/loss:\u001b[0m\n", + " \u001b[37m6.862 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:11 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.914 train/loss:\u001b[0m\n", + " \u001b[37m6.914 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:11 • 0:00:12\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.914 train/loss:\u001b[0m\n", + " \u001b[37m6.914 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:11 • 0:00:12\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.050 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.764 train/loss:\u001b[0m\n", + " \u001b[37m6.764 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.050 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.764 train/loss:\u001b[0m\n", + " \u001b[37m6.764 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.895 train/loss:\u001b[0m\n", + " \u001b[37m6.895 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.895 train/loss:\u001b[0m\n", + " \u001b[37m6.895 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.911 train/loss:\u001b[0m\n", + " \u001b[37m6.911 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:13 • 0:00:11\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.911 train/loss:\u001b[0m\n", + " \u001b[37m6.911 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:13 • 0:00:11\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.845 train/loss:\u001b[0m\n", + " \u001b[37m6.845 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:14 • 0:00:11\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.845 train/loss:\u001b[0m\n", + " \u001b[37m6.845 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:14 • 0:00:11\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.828 train/loss:\u001b[0m\n", + " \u001b[37m6.828 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:15 • 0:00:11\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.828 train/loss:\u001b[0m\n", + " \u001b[37m6.828 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:15 • 0:00:11\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.887 train/loss:\u001b[0m\n", + " \u001b[37m6.887 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.887 train/loss:\u001b[0m\n", + " \u001b[37m6.887 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.836 train/loss:\u001b[0m\n", + " \u001b[37m6.836 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.836 train/loss:\u001b[0m\n", + " \u001b[37m6.836 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.854 train/loss:\u001b[0m\n", + " \u001b[37m6.854 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.854 train/loss:\u001b[0m\n", + " \u001b[37m6.854 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.854 train/loss:\u001b[0m\n", + " \u001b[37m6.854 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.854 train/loss:\u001b[0m\n", + " \u001b[37m6.854 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.915 train/loss:\u001b[0m\n", + " \u001b[37m6.915 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.915 train/loss:\u001b[0m\n", + " \u001b[37m6.915 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.931 train/loss:\u001b[0m\n", + " \u001b[37m6.931 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.931 train/loss:\u001b[0m\n", + " \u001b[37m6.931 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.966 train/loss:\u001b[0m\n", + " \u001b[37m6.966 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.966 train/loss:\u001b[0m\n", + " \u001b[37m6.966 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.800 train/loss:\u001b[0m\n", + " \u001b[37m6.800 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.800 train/loss:\u001b[0m\n", + " \u001b[37m6.800 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.989 train/loss:\u001b[0m\n", + " \u001b[37m6.989 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.989 train/loss:\u001b[0m\n", + " \u001b[37m6.989 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.854 train/loss:\u001b[0m\n", + " \u001b[37m6.854 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.854 train/loss:\u001b[0m\n", + " \u001b[37m6.854 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.731 train/loss:\u001b[0m\n", + " \u001b[37m6.731 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:19 • 0:00:05\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.731 train/loss:\u001b[0m\n", + " \u001b[37m6.731 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:19 • 0:00:05\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.039 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.065 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.726 train/loss:\u001b[0m\n", + " \u001b[37m6.726 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:20 • 0:00:05\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.039 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.065 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.726 train/loss:\u001b[0m\n", + " \u001b[37m6.726 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:20 • 0:00:05\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.851 train/loss:\u001b[0m\n", + " \u001b[37m6.851 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:20 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.851 train/loss:\u001b[0m\n", + " \u001b[37m6.851 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:20 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.945 train/loss:\u001b[0m\n", + " \u001b[37m6.945 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:21 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.945 train/loss:\u001b[0m\n", + " \u001b[37m6.945 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:21 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.852 train/loss:\u001b[0m\n", + " \u001b[37m6.852 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:21 • 0:00:03\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.852 train/loss:\u001b[0m\n", + " \u001b[37m6.852 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:21 • 0:00:03\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.031 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.797 train/loss:\u001b[0m\n", + " \u001b[37m6.797 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:22 • 0:00:03\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.031 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.797 train/loss:\u001b[0m\n", + " \u001b[37m6.797 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:22 • 0:00:03\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.819 train/loss:\u001b[0m\n", + " \u001b[37m6.819 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.819 train/loss:\u001b[0m\n", + " \u001b[37m6.819 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.871 train/loss:\u001b[0m\n", + " \u001b[37m6.871 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.871 train/loss:\u001b[0m\n", + " \u001b[37m6.871 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.764 train/loss:\u001b[0m\n", + " \u001b[37m6.764 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:23 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.033 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.764 train/loss:\u001b[0m\n", + " \u001b[37m6.764 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:23 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.791 train/loss:\u001b[0m\n", + " \u001b[37m6.791 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:24 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.016 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.791 train/loss:\u001b[0m\n", + " \u001b[37m6.791 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:24 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.066 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.775 train/loss:\u001b[0m\n", + " \u001b[37m6.775 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.013 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.066 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.775 train/loss:\u001b[0m\n", + " \u001b[37m6.775 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.014 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.049 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.855 train/loss:\u001b[0m\n", + " \u001b[37m6.855 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m147.992 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", + " \u001b[37m5156802.000 \u001b[0m\n", + " \u001b[37mtrain/accuracy: \u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/top5_accur…\u001b[0m\n", + " \u001b[37m0.000 \u001b[0m\n", + " \u001b[37mtrain/cross_entr…\u001b[0m\n", + " \u001b[37m6.766 train/loss:\u001b[0m\n", + " \u001b[37m6.766 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m140.174 \u001b[0m\n", + " \u001b[37mval/samples_per_…\u001b[0m\n", + " \u001b[37m138.743 \u001b[0m\n", + " \u001b[37mval/samples_per_…\u001b[0m\n", + " \u001b[37m136.694 \u001b[0m\n", + " \u001b[37mtrain/samples_pe…\u001b[0m\n", + " \u001b[37m146.335 \u001b[0m\n", + "\u001b[?25h/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=example trainer.max_epochs ...\n", + "\u001b[2;36m[17:09:43]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Not adding metrics to the pl \u001b]8;id=733785;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=345137;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#64\u001b\\\u001b[2m64\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module because they are already \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m present. \u001b[2m \u001b[0m\n", + "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: ClassificationMetricsCallback, LearningRateMonitor\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=371304;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=737977;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#381\u001b\\\u001b[2m381\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=648908;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=28924;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m[17:10:01]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Not adding metrics to the pl \u001b]8;id=33733;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=344699;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#64\u001b\\\u001b[2m64\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m module because they are already \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m present. \u001b[2m \u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37m2.17it/s\u001b[0m \n", + "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", + "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", + "│\u001b[36m \u001b[0m\u001b[36m val/accuracy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.003010033629834652 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36m val/cross_entropy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 6.422720432281494 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 6.422720432281494 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 140.8275146484375 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36m val/top5_accuracy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.029230769723653793 \u001b[0m\u001b[35m \u001b[0m│\n", + "└──────────────────────────────┴──────────────────────────────┘\n", + "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.17it/s\u001b[0m \n", + "\u001b[?25hval val/accuracy: \u001b[1;36m0.003010033629834652\u001b[0m\n", + "val val/top5_accuracy: \u001b[1;36m0.029230769723653793\u001b[0m\n", + "val val/cross_entropy: \u001b[1;36m6.422720432281494\u001b[0m\n", + "val val/samples_per_second_epoch: \u001b[1;36m140.8275146484375\u001b[0m\n" + ] + } + ], + "source": [ + "!HYDRA_FULL_ERROR=1 python project/main.py \\\n", + " algorithm=example \\\n", + " trainer.max_epochs=1 \\\n", + " +trainer.limit_train_batches=0.01\\\n", + " +trainer.limit_val_batches=0.01\\\n", + " datamodule=imagenet" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Testing for throughput across GPUs" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Using the Mila Research template, it is possible to sweep over different parameters for testing purposes. \n", + "For example, suppose we wanted to figure out how different GPUs perform relative to each other. \n", + "\n", + "[Mila's official documentation](https://docs.mila.quebec/Information.html) shows which GPUs are installed on the cluster. Typing ```savail``` on the command line shows their current availability. \n", + "Testing their capacity can yield insights into their suitability for different training cases." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "GPU Avail / Total \n", + "===============================\n", + "2g.20gb 31 / 48 \n", + "3g.40gb 9 / 48 \n", + "4g.40gb 7 / 24 \n", + "a100 8 / 16 \n", + "a100l 0 / 72 \n", + "a6000 0 / 8 \n", + "rtx8000 11 / 400 \n", + "v100 2 / 40 \n" + ] + } + ], + "source": [ + "!savail" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We can observe the following prominent GPU classes: a100, a100l, a6000, rtx8000, v100. \n", + "We will now proceed to specify different GPUs over training runs and compare their throughput." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Add an example of a sweep over some parameters, \n", + "# with the training throughput as the metric, \n", + "# :: callbacks/samples_per_second, ### or add a devicestatsmonitor in\n", + "# and using different kinds of GPUs. " + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Making sense of the former: if a GPU with lower maximum capacity is readily available, training on it may be more time and resource effective than waiting for higher capacity GPUs to become available.\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Logging with Weights & Biases (wandb)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The Mila Research template integrates wandb functionality as a logger specification. \n", + "This has the advantage of being able to track additional metrics and create accompanying visualizations. \n", + "We will now create a wandb report comparing throughput between GPUs. \n" + ] + }, + { + "cell_type": "code", + "execution_count": 86, + "metadata": {}, + "outputs": [], + "source": [ + "# Create a wandb report with the throughput comparison \n", + "# between the different GPU types.\n", + "# i.e. specify wandb as the logger and log the throughput" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We would like to maximize our throughput given GPU choice" + ] + }, + { + "cell_type": "code", + "execution_count": 74, + "metadata": {}, + "outputs": [], + "source": [ + "## Find the best datamodule parameters to maximize the throughput \n", + "## (batches per second) without training (NoOP algo)" + ] + }, + { + "cell_type": "code", + "execution_count": 75, + "metadata": {}, + "outputs": [], + "source": [ + "### Measure the performance on different GPUS using the optimal datamodule \n", + "### params from before (and keeping other parameters the same)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We will now sweep over model hyper-parameters to maximize the utilization of our selected GPU." + ] + }, + { + "cell_type": "code", + "execution_count": 76, + "metadata": {}, + "outputs": [], + "source": [ + "#### Using the results from before, do a simple sweep over model hyper-parameters \n", + "#### to maximize the utilization of the selected GPU (which was selected as a tradeoff \n", + "#### between performance and difficulty to get an allocation). For example if the \n", + "#### RTX8000's are 20% slower than A100s but 5x easier to get an allocation on, use those instead." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Additional resources\n", + "\n", + "[GPU Training (Basic) - LightningAI](https://lightning.ai/docs/pytorch/stable/accelerators/gpu_basic.html) \n", + "[DeviceStatsMonitor class - LightningAI](https://lightning.ai/docs/pytorch/stable/api/lightning.pytorch.callbacks.DeviceStatsMonitor.html) \n", + "[PyTorch Profiler + W&B integration - Weights & Biases](https://wandb.ai/wandb/trace/reports/Using-the-PyTorch-Profiler-with-W-B--Vmlldzo5MDE3NjU)" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "researchtemplate", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.12.4" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/docs/install.md b/docs/install.md index 7d5a3558..9e7d3605 100644 --- a/docs/install.md +++ b/docs/install.md @@ -21,10 +21,8 @@ There are two ways to install this project 1. On your machine: - ```console - curl -sSf https://rye.astral.sh/get | bash - rye sync # Creates a virtual environment and installs dependencies in it. - ``` + - If you don't have Conda installed, you can download it from [here](https://docs.conda.io/en/latest/miniconda.html). + - If you'd rather use a virtual environment instead of Conda, you can totally do so, as long as you have a version of Python >= 3.12. 2. On the Mila cluster: diff --git a/project/configs/algorithm/example.yaml b/project/configs/algorithm/example.yaml index 83d7da68..ab974cf4 100644 --- a/project/configs/algorithm/example.yaml +++ b/project/configs/algorithm/example.yaml @@ -1,12 +1,2 @@ -# This is an example of how you can use a config file to configure a LightningModule. -# In this case we configure the example algorithm. - _target_: project.algorithms.example.ExampleAlgorithm -# NOTE: Why _partial_? Because the config doesn't create the algo directly, it creates a function -# that will accept the datamodule and network and create the algo. _partial_: true -_recursive_: false -optimizer_config: - _target_: torch.optim.AdamW - _partial_: true - lr: 0.001 diff --git a/project/configs/algorithm/optimizer/custom_adam.yaml b/project/configs/algorithm/optimizer/custom_adam.yaml index a48f8215..b1d27a41 100644 --- a/project/configs/algorithm/optimizer/custom_adam.yaml +++ b/project/configs/algorithm/optimizer/custom_adam.yaml @@ -1,6 +1,6 @@ -defaults: - - Adam -# Learning rate of the optimizer. -lr: 4e-3 -# Weight decay coefficient. -weight_decay: 1e-5 +_target_: torch.optim.Adam +lr: 0.001 +betas: [0.9, 0.999] +eps: 1e-08 +weight_decay: 0 +amsgrad: false diff --git a/project/configs/datamodule/imagenet.yaml b/project/configs/datamodule/imagenet.yaml index 23804087..45b68c6e 100644 --- a/project/configs/datamodule/imagenet.yaml +++ b/project/configs/datamodule/imagenet.yaml @@ -2,4 +2,7 @@ defaults: - vision - _self_ _target_: project.datamodules.ImageNetDataModule +batch_size: 128 +train_transforms: + _target_: project.datamodules.image_classification.imagenet.imagenet_train_transforms # todo: add good configuration options here. diff --git a/project/configs/trainer/callbacks/default.yaml b/project/configs/trainer/callbacks/default.yaml index 8c645e12..b31d7b58 100644 --- a/project/configs/trainer/callbacks/default.yaml +++ b/project/configs/trainer/callbacks/default.yaml @@ -25,8 +25,5 @@ early_stopping: model_summary: max_depth: 2 -lr_monitor: - _target_: lightning.pytorch.callbacks.LearningRateMonitor - -device_utilisation: - _target_: lightning.pytorch.callbacks.DeviceStatsMonitor +throughput: + _target_: project.algorithms.callbacks.samples_per_second.MeasureSamplesPerSecondCallback diff --git a/project/experiment.py b/project/experiment.py index ee0f4f61..e5222279 100644 --- a/project/experiment.py +++ b/project/experiment.py @@ -16,7 +16,11 @@ import logging import os import random +<<<<<<< HEAD from dataclasses import dataclass +======= +from dataclasses import dataclass, is_dataclass +>>>>>>> b7aec3e (add profiling notebook, hotfix a few classes) from logging import getLogger as get_logger from typing import Any @@ -41,12 +45,6 @@ logger = get_logger(__name__) -# BUG: Always using the pydantic parser when instantiating things would be nice, but it currently -# causes issues related to pickling: https://github.com/mit-ll-responsible-ai/hydra-zen/issues/717 -# def _use_pydantic[C: Callable](fn: C) -> C: -# return functools.partial(hydra_zen.instantiate, _target_wrapper_=pydantic_parser) # type: ignore -# instantiate = _use_pydantic(hydra_zen.instantiate) - instantiate = hydra_zen.instantiate diff --git a/project/main.py b/project/main.py index b9d15498..cb49cfd4 100644 --- a/project/main.py +++ b/project/main.py @@ -109,7 +109,7 @@ def evaluation(experiment: Experiment) -> tuple[str, float | None, dict]: # We want to report the training error. metrics = { **experiment.trainer.logged_metrics, - **experiment.trainer.callback_metrics, + **experiment.trainer._metrics, **experiment.trainer.progress_bar_metrics, } rich.print(metrics) diff --git a/pyproject.toml b/pyproject.toml index 05129f8b..04debe18 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -23,6 +23,10 @@ dependencies = [ "torch-jax-interop>=0.0.7", "pydantic>=2.8.2", "simple-parsing>=0.1.5", + "pydantic==2.7.4", + "milatools>=0.0.18", + "rootutils>=0.0.1", + "ipykernel>=6.28.0" ] readme = "README.md" requires-python = ">= 3.10" From 7f2032f206512e645a9cc2c660c94aacf2ed7610 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Fri, 30 Aug 2024 18:19:34 -0400 Subject: [PATCH 03/33] removed pyrootutils, fixed typos, nbstripout check --- docs/examples/profiling.ipynb | 5729 +-------------------------------- 1 file changed, 25 insertions(+), 5704 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 2ac62445..9e6ae30a 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -18,38 +18,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "The Mila Research Template leverages built-in PyTorch and Lightning functionality to make benchmarking accesible and flexible. " - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "/home/mila/c/cesar.valdez/idt/ResearchTemplate\n", - "/home/mila/c/cesar.valdez/idt/ResearchTemplate\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/IPython/core/magics/osm.py:417: UserWarning: This is now an optional IPython functionality, setting dhist requires you to install the `pickleshare` library.\n", - " self.shell.db['dhist'] = compress_dhist(dhist)[-100:]\n" - ] - } - ], - "source": [ - "import os\n", - "import rootutils\n", - "\n", - "home_dir = rootutils.find_root(search_from=\"profiling.ipynb\", indicator=\".git\")\n", - "%cd $home_dir\n", - "print(os.getcwd())" + "The Mila Research Template leverages built-in PyTorch and Lightning functionality to make benchmarking accessible and flexible. " ] }, { @@ -68,5661 +37,30 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2mCONFIG\u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2malgorithm\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mhydra_zen.funcs.zen_processing \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_zen_target\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.no_op.NoOp \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_zen_partial\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_zen_wrappers\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mhydra_zen.third_party.pydantic.pydantic_parser \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mnetwork\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtorchvision.models.resnet.resnet18 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mweights\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mprogress\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_classes\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${instance_attr:datamodule.num_classes} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mdatamodule\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.datamodules.ImageNetDataModule \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdata_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:DATA_DIR} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_workers\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:NUM_WORKERS} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mval_split\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnormalize\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mshuffle\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mpin_memory\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mseed\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m42 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mbatch_size\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m64 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mtrainer\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.Trainer \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlogger\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40maccelerator\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mstrategy\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdevices\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmin_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmax_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdefault_root_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdetect_anomaly\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mcallbacks\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_checkpoint\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.ModelCheckpoint \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdirpath\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir}/checkpoints \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mfilename\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mepoch_{epoch:03d} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_last\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_top_k\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mauto_insert_metric_name\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_weights_only\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_train_steps\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mtrain_time_interval\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mearly_stopping\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.EarlyStopping \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmin_delta\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.0 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mpatience\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m5 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstrict\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_finite\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstopping_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdivergence_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_summary\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichModelSummary \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmax_depth\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m2 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mrich_progress_bar\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichProgressBar \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mthroughput\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.callbacks.samples_per_second.MeasureSam\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_train_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_val_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mlog_level\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40minfo \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mseed\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40m87282 \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mname\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mNoOp-resnet18-imagenet \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mdebug\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", - "\u001b[2m└── \u001b[0m\u001b[2mverbose\u001b[0m\n", - "\u001b[2m \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", - "\u001b[2;36m[08/07/24 16:28:58]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=131611;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=866100;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m,\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 87282\n", - "Seed set to 87282\n", - "/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=NoOp trainer.max_epochs=1 ...\n", - "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", - "GPU available: True (cuda), used: True\n", - "TPU available: False, using: 0 TPU cores\n", - "HPU available: False, using: 0 HPUs\n", - "\u001b[2;36m[16:28:58]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=867639;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=12083;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#163\u001b\\\u001b[2m163\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7fb2a6f64f80\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", - "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "Extracting train archive: 100%|████| 1000/1000 [10:10<00:00, 1.64Directories/s]\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", - "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", - "┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━┩\n", - "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNorm2d │ 128 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequential │ 147 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequential │ 525 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequential │ 2.1 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequential │ 8.4 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ AdaptiveAvgPool2d │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\n", - "└────┴─────────────────┴───────────────────┴────────┴───────┘\n", - "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", - "\u001b[1mNon-trainable params\u001b[0m: 0 \n", - "\u001b[1mTotal params\u001b[0m: 11.7 M \n", - "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/180\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0mt/s\u001b[0m \n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m100.703 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/180\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m100.703 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/180\u001b[0m \u001b[37m0:00:02 • 0:01:48\u001b[0m \u001b[37m1.64it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:35\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:35\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:31\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:31\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:26\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:26\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:22\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:22\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:05 • 0:01:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:05 • 0:01:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:07 • 0:01:19\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:07 • 0:01:19\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:07 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:07 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:08 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:08 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:08 • 0:01:19\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:08 • 0:01:19\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:09 • 0:01:17\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:09 • 0:01:17\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:09 • 0:01:16\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:09 • 0:01:16\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:11 • 0:01:13\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:11 • 0:01:13\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:13 • 0:01:11\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:13 • 0:01:11\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:13 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:13 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:14 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:14 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:14 • 0:01:09\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:14 • 0:01:09\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:14 • 0:01:08\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:14 • 0:01:08\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:16 • 0:01:07\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:16 • 0:01:07\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:16 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:16 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:17 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:17 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:17 • 0:01:05\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:17 • 0:01:05\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:17 • 0:01:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:17 • 0:01:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:25 • 0:00:56\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:25 • 0:00:56\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:26 • 0:00:56\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:26 • 0:00:56\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:26 • 0:00:55\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:26 • 0:00:55\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:27 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:27 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:28 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:28 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:28 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:28 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:29 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:29 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:32 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:32 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:33 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:33 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:33 • 0:00:46\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:33 • 0:00:46\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:34 • 0:00:46\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:34 • 0:00:46\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:34 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:34 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:35 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:35 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:37 • 0:00:41\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:37 • 0:00:41\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:38 • 0:00:41\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:38 • 0:00:41\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:38 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:38 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:39 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:39 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:39 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:39 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:40 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:40 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:42 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:42 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:44 • 0:00:33\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:44 • 0:00:33\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:45 • 0:00:33\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:45 • 0:00:33\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:46 • 0:00:33\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:46 • 0:00:33\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:47 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:47 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:47 • 0:00:31\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:47 • 0:00:31\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:48 • 0:00:31\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:48 • 0:00:31\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:48 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:48 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:49 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:49 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:49 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:49 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:50 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:50 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:51 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:51 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:51 • 0:00:27\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:51 • 0:00:27\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:52 • 0:00:27\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:52 • 0:00:27\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:52 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:52 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:54 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:54 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:55 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:55 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:55 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:55 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:00:56 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:00:56 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:00:57 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:00:57 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:00:57 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:00:57 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:00:58 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:00:58 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:00:59 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:00:59 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:00:59 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:00:59 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:00 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:00 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:05 • 0:00:14\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:05 • 0:00:14\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:07 • 0:00:12\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:07 • 0:00:12\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:07 • 0:00:11\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:07 • 0:00:11\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:08 • 0:00:11\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:08 • 0:00:11\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:08 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:08 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:10 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:10 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:11 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:11 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:11 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:11 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:12 • 0:00:06\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:12 • 0:00:06\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:13 • 0:00:06\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:13 • 0:00:06\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:13 • 0:00:05\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:13 • 0:00:05\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:14 • 0:00:05\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:14 • 0:00:05\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:14 • 0:00:04\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:14 • 0:00:04\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:16 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:16 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:17 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:17 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:17 • 0:00:01\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:17 • 0:00:01\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:18 • 0:00:01\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:18 • 0:00:01\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m151.611 \u001b[0m\n", - " \u001b[37mval/samples_per_…\u001b[0m\n", - " \u001b[37m120.811 \u001b[0m\n", - " \u001b[37mval/samples_per_…\u001b[0m\n", - " \u001b[37m128.323 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m150.129 \u001b[0m\n", - "\u001b[?25h/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=NoOp trainer.max_epochs=1 ...\n", - "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "\u001b[2;36m[16:41:35]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=972670;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=502992;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#381\u001b\\\u001b[2m381\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m[16:41:36]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=825781;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=661210;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37m2.10it/s\u001b[0m \n", - "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", - "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", - "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4964688718318939 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 140.0721893310547 \u001b[0m\u001b[35m \u001b[0m│\n", - "└──────────────────────────────┴──────────────────────────────┘\n", - "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.11it/s\u001b[0m \n", - "\u001b[?25hval val/samples_per_second_epoch: \u001b[1;36m140.0721893310547\u001b[0m\n" - ] - } - ], + "outputs": [], "source": [ - "!python project/main.py \\\n", - " algorithm=NoOp \\\n", - " trainer.max_epochs=1 \\\n", - " +trainer.limit_train_batches=0.01\\\n", - " +trainer.limit_val_batches=0.01\\\n", - " datamodule=imagenet" + "#!python project/main.py \\\n", + "# algorithm=NoOp \\\n", + "# trainer.max_epochs=1 \\\n", + "# +trainer.limit_train_batches=0.01\\\n", + "# +trainer.limit_val_batches=0.01\\\n", + "# datamodule=imagenet" ] }, { "cell_type": "code", - "execution_count": 10, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2mCONFIG\u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2malgorithm\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.example.ExampleAlgorithm \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_partial_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mnetwork\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtorchvision.models.resnet.resnet18 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mweights\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mprogress\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_classes\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${instance_attr:datamodule.num_classes} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mdatamodule\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.datamodules.ImageNetDataModule \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdata_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:DATA_DIR} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_workers\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:NUM_WORKERS} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mval_split\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnormalize\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mshuffle\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mpin_memory\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mseed\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m42 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mbatch_size\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m64 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mtrainer\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.Trainer \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlogger\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40maccelerator\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mstrategy\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdevices\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmin_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmax_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdefault_root_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdetect_anomaly\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mcallbacks\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_checkpoint\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.ModelCheckpoint \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdirpath\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir}/checkpoints \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mfilename\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mepoch_{epoch:03d} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_last\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_top_k\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mauto_insert_metric_name\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_weights_only\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_train_steps\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mtrain_time_interval\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mearly_stopping\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.EarlyStopping \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmin_delta\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.0 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mpatience\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m5 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstrict\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_finite\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstopping_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdivergence_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_summary\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichModelSummary \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmax_depth\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m2 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mrich_progress_bar\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichProgressBar \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mthroughput\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.callbacks.samples_per_second.MeasureSam\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_train_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_val_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mlog_level\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40minfo \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mseed\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40m81570 \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mname\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mexample-resnet18-imagenet \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mdebug\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", - "\u001b[2m└── \u001b[0m\u001b[2mverbose\u001b[0m\n", - "\u001b[2m \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", - "\u001b[2;36m[08/07/24 17:07:38]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=50545;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=71944;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m,\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 81570\n", - "Seed set to 81570\n", - "/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=example trainer.max_epochs ...\n", - "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", - "GPU available: True (cuda), used: True\n", - "TPU available: False, using: 0 TPU cores\n", - "HPU available: False, using: 0 HPUs\n", - "\u001b[2;36m[17:07:38]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=285735;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=912000;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#156\u001b\\\u001b[2m156\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7eff7e751f10\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m[17:07:43]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=702216;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=293228;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m train_accuracy. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=476337;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=48315;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m val_accuracy. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=513180;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=101826;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m test_accuracy. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=370748;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=635161;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m train_top5_accuracy. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=727524;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=265133;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m val_top5_accuracy. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=753886;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=808003;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m test_top5_accuracy. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=781296;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=524717;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#381\u001b\\\u001b[2m381\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m[17:07:44]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=248066;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=550199;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m[17:08:02]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Not adding metrics to the pl \u001b]8;id=656135;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=914060;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#64\u001b\\\u001b[2m64\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module because they are already \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m present. \u001b[2m \u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "┏━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━┓\n", - "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mIn sizes\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mOut sizes\u001b[0m\u001b[1;35m \u001b[0m┃\n", - "┡━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━┩\n", - "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\u001b[37m \u001b[0m\u001b[37m [64, 3,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1000]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\u001b[37m \u001b[0m\u001b[37m [64, 3,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNor… │ 128 │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequenti… │ 147 K │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequenti… │ 525 K │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 128,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 28, 28]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequenti… │ 2.1 M │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 256,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m128, 28,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 14, 14]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 28]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequenti… │ 8.4 M │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 512,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m256, 14,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 7, 7]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 14]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ Adaptive… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 512,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 512, 7,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1, 1]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 7]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 512]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1000]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m11\u001b[0m\u001b[2m \u001b[0m│ train_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m12\u001b[0m\u001b[2m \u001b[0m│ val_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m13\u001b[0m\u001b[2m \u001b[0m│ test_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m14\u001b[0m\u001b[2m \u001b[0m│ train_top5_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m15\u001b[0m\u001b[2m \u001b[0m│ val_top5_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m16\u001b[0m\u001b[2m \u001b[0m│ test_top5_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", - "└────┴─────────────────────┴───────────┴────────┴───────┴──────────┴───────────┘\n", - "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", - "\u001b[1mNon-trainable params\u001b[0m: 0 \n", - "\u001b[1mTotal params\u001b[0m: 11.7 M \n", - "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/180\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0mt/s\u001b[0m \n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m51.620 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/180\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m51.620 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/180\u001b[0m \u001b[37m0:00:01 • 0:00:21\u001b[0m \u001b[37m8.90it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.059 train/loss: \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/180\u001b[0m \u001b[37m0:00:02 • 0:01:01\u001b[0m \u001b[37m2.92it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.059 train/loss: \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/180\u001b[0m \u001b[37m0:00:02 • 0:01:01\u001b[0m \u001b[37m2.92it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.015 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m6.940 train/loss: \u001b[0m\n", - " \u001b[37m6.940 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:04\u001b[0m \u001b[37m2.76it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.015 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m6.940 train/loss: \u001b[0m\n", - " \u001b[37m6.940 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:04\u001b[0m \u001b[37m2.76it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.146 train/loss: \u001b[0m\n", - " \u001b[37m7.146 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:08\u001b[0m \u001b[37m2.58it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.146 train/loss: \u001b[0m\n", - " \u001b[37m7.146 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:08\u001b[0m \u001b[37m2.58it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.059 train/loss: \u001b[0m\n", - " \u001b[37m7.059 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.44it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.059 train/loss: \u001b[0m\n", - " \u001b[37m7.059 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.44it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.076 train/loss: \u001b[0m\n", - " \u001b[37m7.076 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.43it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.076 train/loss: \u001b[0m\n", - " \u001b[37m7.076 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.43it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.096 train/loss: \u001b[0m\n", - " \u001b[37m7.096 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.096 train/loss: \u001b[0m\n", - " \u001b[37m7.096 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.142 train/loss: \u001b[0m\n", - " \u001b[37m7.142 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.142 train/loss: \u001b[0m\n", - " \u001b[37m7.142 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.030 train/loss: \u001b[0m\n", - " \u001b[37m7.030 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:19\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.030 train/loss:\u001b[0m\n", - " \u001b[37m7.030 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:19\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.100 train/loss:\u001b[0m\n", - " \u001b[37m7.100 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:06 • 0:01:31\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.100 train/loss:\u001b[0m\n", - " \u001b[37m7.100 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:06 • 0:01:31\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.008 train/loss:\u001b[0m\n", - " \u001b[37m7.008 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:07 • 0:01:28\u001b[0m \u001b[37m1.91it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.008 train/loss:\u001b[0m\n", - " \u001b[37m7.008 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:07 • 0:01:28\u001b[0m \u001b[37m1.91it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.302 train/loss:\u001b[0m\n", - " \u001b[37m7.302 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:07 • 0:01:26\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.302 train/loss:\u001b[0m\n", - " \u001b[37m7.302 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:07 • 0:01:26\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.344 train/loss:\u001b[0m\n", - " \u001b[37m7.344 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:08 • 0:01:30\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.344 train/loss:\u001b[0m\n", - " \u001b[37m7.344 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:08 • 0:01:30\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.050 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.361 train/loss:\u001b[0m\n", - " \u001b[37m7.361 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:08 • 0:01:28\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.050 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.361 train/loss:\u001b[0m\n", - " \u001b[37m7.361 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:08 • 0:01:28\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.182 train/loss:\u001b[0m\n", - " \u001b[37m7.182 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:10 • 0:01:35\u001b[0m \u001b[37m1.73it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.182 train/loss:\u001b[0m\n", - " \u001b[37m7.182 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:10 • 0:01:35\u001b[0m \u001b[37m1.73it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.168 train/loss:\u001b[0m\n", - " \u001b[37m7.168 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:10 • 0:01:37\u001b[0m \u001b[37m1.69it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.168 train/loss:\u001b[0m\n", - " \u001b[37m7.168 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:10 • 0:01:37\u001b[0m \u001b[37m1.69it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.324 train/loss:\u001b[0m\n", - " \u001b[37m7.324 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:11 • 0:01:35\u001b[0m \u001b[37m1.72it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.324 train/loss:\u001b[0m\n", - " \u001b[37m7.324 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:11 • 0:01:35\u001b[0m \u001b[37m1.72it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.436 train/loss:\u001b[0m\n", - " \u001b[37m7.436 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:11 • 0:01:32\u001b[0m \u001b[37m1.75it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.436 train/loss:\u001b[0m\n", - " \u001b[37m7.436 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:11 • 0:01:32\u001b[0m \u001b[37m1.75it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.111 train/loss:\u001b[0m\n", - " \u001b[37m7.111 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:12 • 0:01:31\u001b[0m \u001b[37m1.78it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.111 train/loss:\u001b[0m\n", - " \u001b[37m7.111 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:12 • 0:01:31\u001b[0m \u001b[37m1.78it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.521 train/loss:\u001b[0m\n", - " \u001b[37m7.521 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:12 • 0:01:29\u001b[0m \u001b[37m1.79it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.521 train/loss:\u001b[0m\n", - " \u001b[37m7.521 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:12 • 0:01:29\u001b[0m \u001b[37m1.79it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.065 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.163 train/loss:\u001b[0m\n", - " \u001b[37m7.163 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:13 • 0:01:28\u001b[0m \u001b[37m1.81it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.065 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.163 train/loss:\u001b[0m\n", - " \u001b[37m7.163 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:13 • 0:01:28\u001b[0m \u001b[37m1.81it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.038 train/loss:\u001b[0m\n", - " \u001b[37m7.038 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:13 • 0:01:26\u001b[0m \u001b[37m1.83it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.038 train/loss:\u001b[0m\n", - " \u001b[37m7.038 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:13 • 0:01:26\u001b[0m \u001b[37m1.83it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.171 train/loss:\u001b[0m\n", - " \u001b[37m7.171 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:13 • 0:01:25\u001b[0m \u001b[37m1.85it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.171 train/loss:\u001b[0m\n", - " \u001b[37m7.171 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:13 • 0:01:25\u001b[0m \u001b[37m1.85it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.327 train/loss:\u001b[0m\n", - " \u001b[37m7.327 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:14 • 0:01:23\u001b[0m \u001b[37m1.87it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.327 train/loss:\u001b[0m\n", - " \u001b[37m7.327 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:14 • 0:01:23\u001b[0m \u001b[37m1.87it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.207 train/loss:\u001b[0m\n", - " \u001b[37m7.207 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:14 • 0:01:22\u001b[0m \u001b[37m1.90it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.207 train/loss:\u001b[0m\n", - " \u001b[37m7.207 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:14 • 0:01:22\u001b[0m \u001b[37m1.90it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.271 train/loss:\u001b[0m\n", - " \u001b[37m7.271 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:15 • 0:01:20\u001b[0m \u001b[37m1.92it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.271 train/loss:\u001b[0m\n", - " \u001b[37m7.271 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:15 • 0:01:20\u001b[0m \u001b[37m1.92it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.566 train/loss:\u001b[0m\n", - " \u001b[37m7.566 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:15 • 0:01:19\u001b[0m \u001b[37m1.93it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.566 train/loss:\u001b[0m\n", - " \u001b[37m7.566 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:15 • 0:01:19\u001b[0m \u001b[37m1.93it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.203 train/loss:\u001b[0m\n", - " \u001b[37m7.203 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:15 • 0:01:18\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.203 train/loss:\u001b[0m\n", - " \u001b[37m7.203 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:15 • 0:01:18\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.262 train/loss:\u001b[0m\n", - " \u001b[37m7.262 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:16 • 0:01:17\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.262 train/loss:\u001b[0m\n", - " \u001b[37m7.262 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:16 • 0:01:17\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.109 train/loss:\u001b[0m\n", - " \u001b[37m7.109 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:16 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.109 train/loss:\u001b[0m\n", - " \u001b[37m7.109 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:16 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.343 train/loss:\u001b[0m\n", - " \u001b[37m7.343 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:17 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.343 train/loss:\u001b[0m\n", - " \u001b[37m7.343 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:17 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.190 train/loss:\u001b[0m\n", - " \u001b[37m7.190 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:17 • 0:01:15\u001b[0m \u001b[37m1.98it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.190 train/loss:\u001b[0m\n", - " \u001b[37m7.190 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:17 • 0:01:15\u001b[0m \u001b[37m1.98it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.051 train/loss:\u001b[0m\n", - " \u001b[37m7.051 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:18 • 0:01:14\u001b[0m \u001b[37m1.99it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.051 train/loss:\u001b[0m\n", - " \u001b[37m7.051 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:18 • 0:01:14\u001b[0m \u001b[37m1.99it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.174 train/loss:\u001b[0m\n", - " \u001b[37m7.174 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:18 • 0:01:13\u001b[0m \u001b[37m2.00it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.174 train/loss:\u001b[0m\n", - " \u001b[37m7.174 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:18 • 0:01:13\u001b[0m \u001b[37m2.00it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.390 train/loss:\u001b[0m\n", - " \u001b[37m7.390 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:18 • 0:01:12\u001b[0m \u001b[37m2.01it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.390 train/loss:\u001b[0m\n", - " \u001b[37m7.390 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:18 • 0:01:12\u001b[0m \u001b[37m2.01it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.300 train/loss:\u001b[0m\n", - " \u001b[37m7.300 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:19 • 0:01:11\u001b[0m \u001b[37m2.02it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.300 train/loss:\u001b[0m\n", - " \u001b[37m7.300 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:19 • 0:01:11\u001b[0m \u001b[37m2.02it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.237 train/loss:\u001b[0m\n", - " \u001b[37m7.237 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:19 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.237 train/loss:\u001b[0m\n", - " \u001b[37m7.237 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:19 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.232 train/loss:\u001b[0m\n", - " \u001b[37m7.232 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:20 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.232 train/loss:\u001b[0m\n", - " \u001b[37m7.232 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:20 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.971 train/loss:\u001b[0m\n", - " \u001b[37m6.971 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:20 • 0:01:09\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.971 train/loss:\u001b[0m\n", - " \u001b[37m6.971 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:20 • 0:01:09\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.262 train/loss:\u001b[0m\n", - " \u001b[37m7.262 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.262 train/loss:\u001b[0m\n", - " \u001b[37m7.262 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.142 train/loss:\u001b[0m\n", - " \u001b[37m7.142 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.142 train/loss:\u001b[0m\n", - " \u001b[37m7.142 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.984 train/loss:\u001b[0m\n", - " \u001b[37m6.984 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.984 train/loss:\u001b[0m\n", - " \u001b[37m6.984 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.149 train/loss:\u001b[0m\n", - " \u001b[37m7.149 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.149 train/loss:\u001b[0m\n", - " \u001b[37m7.149 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.986 train/loss:\u001b[0m\n", - " \u001b[37m6.986 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:22 • 0:01:06\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.986 train/loss:\u001b[0m\n", - " \u001b[37m6.986 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:22 • 0:01:06\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.067 train/loss:\u001b[0m\n", - " \u001b[37m7.067 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:23 • 0:01:06\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.067 train/loss:\u001b[0m\n", - " \u001b[37m7.067 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:23 • 0:01:06\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.224 train/loss:\u001b[0m\n", - " \u001b[37m7.224 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:23 • 0:01:05\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.224 train/loss:\u001b[0m\n", - " \u001b[37m7.224 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:23 • 0:01:05\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.053 train/loss:\u001b[0m\n", - " \u001b[37m7.053 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:24 • 0:01:05\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.053 train/loss:\u001b[0m\n", - " \u001b[37m7.053 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:24 • 0:01:05\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.016 train/loss:\u001b[0m\n", - " \u001b[37m7.016 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:24 • 0:01:04\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.016 train/loss:\u001b[0m\n", - " \u001b[37m7.016 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:24 • 0:01:04\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.050 train/loss:\u001b[0m\n", - " \u001b[37m7.050 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.050 train/loss:\u001b[0m\n", - " \u001b[37m7.050 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.025 train/loss:\u001b[0m\n", - " \u001b[37m7.025 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.025 train/loss:\u001b[0m\n", - " \u001b[37m7.025 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.886 train/loss:\u001b[0m\n", - " \u001b[37m6.886 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:25 • 0:01:02\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.886 train/loss:\u001b[0m\n", - " \u001b[37m6.886 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:25 • 0:01:02\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.926 train/loss:\u001b[0m\n", - " \u001b[37m6.926 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.926 train/loss:\u001b[0m\n", - " \u001b[37m6.926 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.090 train/loss:\u001b[0m\n", - " \u001b[37m7.090 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.090 train/loss:\u001b[0m\n", - " \u001b[37m7.090 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.913 train/loss:\u001b[0m\n", - " \u001b[37m6.913 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.913 train/loss:\u001b[0m\n", - " \u001b[37m6.913 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.100 train/loss:\u001b[0m\n", - " \u001b[37m7.100 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.100 train/loss:\u001b[0m\n", - " \u001b[37m7.100 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.028 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.985 train/loss:\u001b[0m\n", - " \u001b[37m6.985 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:28 • 0:00:59\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.028 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.985 train/loss:\u001b[0m\n", - " \u001b[37m6.985 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:28 • 0:00:59\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.191 train/loss:\u001b[0m\n", - " \u001b[37m7.191 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.191 train/loss:\u001b[0m\n", - " \u001b[37m7.191 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.939 train/loss:\u001b[0m\n", - " \u001b[37m6.939 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.939 train/loss:\u001b[0m\n", - " \u001b[37m6.939 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.859 train/loss:\u001b[0m\n", - " \u001b[37m6.859 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.859 train/loss:\u001b[0m\n", - " \u001b[37m6.859 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.076 train/loss:\u001b[0m\n", - " \u001b[37m7.076 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.076 train/loss:\u001b[0m\n", - " \u001b[37m7.076 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.115 train/loss:\u001b[0m\n", - " \u001b[37m7.115 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:30 • 0:00:56\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.115 train/loss:\u001b[0m\n", - " \u001b[37m7.115 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:30 • 0:00:56\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.936 train/loss:\u001b[0m\n", - " \u001b[37m6.936 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.936 train/loss:\u001b[0m\n", - " \u001b[37m6.936 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.901 train/loss:\u001b[0m\n", - " \u001b[37m6.901 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.901 train/loss:\u001b[0m\n", - " \u001b[37m6.901 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.068 train/loss:\u001b[0m\n", - " \u001b[37m7.068 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.068 train/loss:\u001b[0m\n", - " \u001b[37m7.068 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.932 train/loss:\u001b[0m\n", - " \u001b[37m6.932 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.932 train/loss:\u001b[0m\n", - " \u001b[37m6.932 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.934 train/loss:\u001b[0m\n", - " \u001b[37m6.934 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.934 train/loss:\u001b[0m\n", - " \u001b[37m6.934 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.022 train/loss:\u001b[0m\n", - " \u001b[37m7.022 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.022 train/loss:\u001b[0m\n", - " \u001b[37m7.022 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.905 train/loss:\u001b[0m\n", - " \u001b[37m6.905 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.905 train/loss:\u001b[0m\n", - " \u001b[37m6.905 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.929 train/loss:\u001b[0m\n", - " \u001b[37m6.929 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.929 train/loss:\u001b[0m\n", - " \u001b[37m6.929 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.060 train/loss:\u001b[0m\n", - " \u001b[37m7.060 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.060 train/loss:\u001b[0m\n", - " \u001b[37m7.060 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.089 train/loss:\u001b[0m\n", - " \u001b[37m7.089 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:34 • 0:00:51\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.089 train/loss:\u001b[0m\n", - " \u001b[37m7.089 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:34 • 0:00:51\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.011 train/loss:\u001b[0m\n", - " \u001b[37m7.011 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:34 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.011 train/loss:\u001b[0m\n", - " \u001b[37m7.011 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:34 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.953 train/loss:\u001b[0m\n", - " \u001b[37m6.953 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:35 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.953 train/loss:\u001b[0m\n", - " \u001b[37m6.953 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:35 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.890 train/loss:\u001b[0m\n", - " \u001b[37m6.890 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:35 • 0:00:48\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.890 train/loss:\u001b[0m\n", - " \u001b[37m6.890 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:35 • 0:00:48\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.829 train/loss:\u001b[0m\n", - " \u001b[37m6.829 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.829 train/loss:\u001b[0m\n", - " \u001b[37m6.829 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.009 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.902 train/loss:\u001b[0m\n", - " \u001b[37m6.902 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.009 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.902 train/loss:\u001b[0m\n", - " \u001b[37m6.902 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.996 train/loss:\u001b[0m\n", - " \u001b[37m6.996 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:37 • 0:00:47\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.996 train/loss:\u001b[0m\n", - " \u001b[37m6.996 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:37 • 0:00:47\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.074 train/loss:\u001b[0m\n", - " \u001b[37m7.074 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:37 • 0:00:46\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.074 train/loss:\u001b[0m\n", - " \u001b[37m7.074 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:37 • 0:00:46\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.826 train/loss:\u001b[0m\n", - " \u001b[37m6.826 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.826 train/loss:\u001b[0m\n", - " \u001b[37m6.826 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.075 train/loss:\u001b[0m\n", - " \u001b[37m7.075 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.075 train/loss:\u001b[0m\n", - " \u001b[37m7.075 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.881 train/loss:\u001b[0m\n", - " \u001b[37m6.881 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.28it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.881 train/loss:\u001b[0m\n", - " \u001b[37m6.881 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.28it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.860 train/loss:\u001b[0m\n", - " \u001b[37m6.860 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.860 train/loss:\u001b[0m\n", - " \u001b[37m6.860 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.005 train/loss:\u001b[0m\n", - " \u001b[37m7.005 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:39 • 0:00:42\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.005 train/loss:\u001b[0m\n", - " \u001b[37m7.005 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:39 • 0:00:42\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.864 train/loss:\u001b[0m\n", - " \u001b[37m6.864 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:40 • 0:00:42\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.864 train/loss:\u001b[0m\n", - " \u001b[37m6.864 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:40 • 0:00:42\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.953 train/loss:\u001b[0m\n", - " \u001b[37m6.953 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:40 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.953 train/loss:\u001b[0m\n", - " \u001b[37m6.953 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:40 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.968 train/loss:\u001b[0m\n", - " \u001b[37m6.968 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:41 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.968 train/loss:\u001b[0m\n", - " \u001b[37m6.968 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:41 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.973 train/loss:\u001b[0m\n", - " \u001b[37m6.973 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:41 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.973 train/loss:\u001b[0m\n", - " \u001b[37m6.973 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:41 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.986 train/loss:\u001b[0m\n", - " \u001b[37m6.986 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:42 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.986 train/loss:\u001b[0m\n", - " \u001b[37m6.986 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:42 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.007 train/loss:\u001b[0m\n", - " \u001b[37m7.007 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.007 train/loss:\u001b[0m\n", - " \u001b[37m7.007 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.029 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.940 train/loss:\u001b[0m\n", - " \u001b[37m6.940 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.029 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.940 train/loss:\u001b[0m\n", - " \u001b[37m6.940 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.932 train/loss:\u001b[0m\n", - " \u001b[37m6.932 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:43 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.932 train/loss:\u001b[0m\n", - " \u001b[37m6.932 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:43 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.997 train/loss:\u001b[0m\n", - " \u001b[37m6.997 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:43 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.997 train/loss:\u001b[0m\n", - " \u001b[37m6.997 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:43 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.936 train/loss:\u001b[0m\n", - " \u001b[37m6.936 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:44 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.936 train/loss:\u001b[0m\n", - " \u001b[37m6.936 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:44 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.014 train/loss:\u001b[0m\n", - " \u001b[37m7.014 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:44 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.014 train/loss:\u001b[0m\n", - " \u001b[37m7.014 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:44 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.937 train/loss:\u001b[0m\n", - " \u001b[37m6.937 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.937 train/loss:\u001b[0m\n", - " \u001b[37m6.937 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.915 train/loss:\u001b[0m\n", - " \u001b[37m6.915 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.915 train/loss:\u001b[0m\n", - " \u001b[37m6.915 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.923 train/loss:\u001b[0m\n", - " \u001b[37m6.923 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:45 • 0:00:36\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.923 train/loss:\u001b[0m\n", - " \u001b[37m6.923 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:45 • 0:00:36\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.014 train/loss:\u001b[0m\n", - " \u001b[37m7.014 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:46 • 0:00:36\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.014 train/loss:\u001b[0m\n", - " \u001b[37m7.014 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:46 • 0:00:36\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.963 train/loss:\u001b[0m\n", - " \u001b[37m6.963 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:46 • 0:00:35\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.963 train/loss:\u001b[0m\n", - " \u001b[37m6.963 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:46 • 0:00:35\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.873 train/loss:\u001b[0m\n", - " \u001b[37m6.873 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:47 • 0:00:35\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.873 train/loss:\u001b[0m\n", - " \u001b[37m6.873 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:47 • 0:00:35\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.033 train/loss:\u001b[0m\n", - " \u001b[37m7.033 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:47 • 0:00:34\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.033 train/loss:\u001b[0m\n", - " \u001b[37m7.033 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:47 • 0:00:34\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.034 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.787 train/loss:\u001b[0m\n", - " \u001b[37m6.787 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:48 • 0:00:35\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.034 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.787 train/loss:\u001b[0m\n", - " \u001b[37m6.787 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:48 • 0:00:35\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.946 train/loss:\u001b[0m\n", - " \u001b[37m6.946 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.946 train/loss:\u001b[0m\n", - " \u001b[37m6.946 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.966 train/loss:\u001b[0m\n", - " \u001b[37m6.966 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.966 train/loss:\u001b[0m\n", - " \u001b[37m6.966 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.899 train/loss:\u001b[0m\n", - " \u001b[37m6.899 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:49 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.899 train/loss:\u001b[0m\n", - " \u001b[37m6.899 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:49 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.031 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.886 train/loss:\u001b[0m\n", - " \u001b[37m6.886 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:50 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.031 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.886 train/loss:\u001b[0m\n", - " \u001b[37m6.886 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:50 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.892 train/loss:\u001b[0m\n", - " \u001b[37m6.892 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:51 • 0:00:33\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.892 train/loss:\u001b[0m\n", - " \u001b[37m6.892 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:51 • 0:00:33\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.880 train/loss:\u001b[0m\n", - " \u001b[37m6.880 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:51 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.880 train/loss:\u001b[0m\n", - " \u001b[37m6.880 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:51 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.897 train/loss:\u001b[0m\n", - " \u001b[37m6.897 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.897 train/loss:\u001b[0m\n", - " \u001b[37m6.897 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.891 train/loss:\u001b[0m\n", - " \u001b[37m6.891 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.891 train/loss:\u001b[0m\n", - " \u001b[37m6.891 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.979 train/loss:\u001b[0m\n", - " \u001b[37m6.979 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:53 • 0:00:31\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.979 train/loss:\u001b[0m\n", - " \u001b[37m6.979 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:53 • 0:00:31\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.860 train/loss:\u001b[0m\n", - " \u001b[37m6.860 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.860 train/loss:\u001b[0m\n", - " \u001b[37m6.860 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.014 train/loss:\u001b[0m\n", - " \u001b[37m7.014 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.014 train/loss:\u001b[0m\n", - " \u001b[37m7.014 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.908 train/loss:\u001b[0m\n", - " \u001b[37m6.908 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.908 train/loss:\u001b[0m\n", - " \u001b[37m6.908 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.802 train/loss:\u001b[0m\n", - " \u001b[37m6.802 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.802 train/loss:\u001b[0m\n", - " \u001b[37m6.802 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.839 train/loss:\u001b[0m\n", - " \u001b[37m6.839 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:56 • 0:00:30\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.839 train/loss:\u001b[0m\n", - " \u001b[37m6.839 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:56 • 0:00:30\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.880 train/loss:\u001b[0m\n", - " \u001b[37m6.880 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.880 train/loss:\u001b[0m\n", - " \u001b[37m6.880 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.812 train/loss:\u001b[0m\n", - " \u001b[37m6.812 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.812 train/loss:\u001b[0m\n", - " \u001b[37m6.812 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.848 train/loss:\u001b[0m\n", - " \u001b[37m6.848 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.848 train/loss:\u001b[0m\n", - " \u001b[37m6.848 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.946 train/loss:\u001b[0m\n", - " \u001b[37m6.946 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.946 train/loss:\u001b[0m\n", - " \u001b[37m6.946 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.939 train/loss:\u001b[0m\n", - " \u001b[37m6.939 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.939 train/loss:\u001b[0m\n", - " \u001b[37m6.939 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.839 train/loss:\u001b[0m\n", - " \u001b[37m6.839 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.839 train/loss:\u001b[0m\n", - " \u001b[37m6.839 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.827 train/loss:\u001b[0m\n", - " \u001b[37m6.827 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.827 train/loss:\u001b[0m\n", - " \u001b[37m6.827 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.862 train/loss:\u001b[0m\n", - " \u001b[37m6.862 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.862 train/loss:\u001b[0m\n", - " \u001b[37m6.862 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.034 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.838 train/loss:\u001b[0m\n", - " \u001b[37m6.838 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.034 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.838 train/loss:\u001b[0m\n", - " \u001b[37m6.838 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.004 train/loss:\u001b[0m\n", - " \u001b[37m7.004 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.004 train/loss:\u001b[0m\n", - " \u001b[37m7.004 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.028 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.801 train/loss:\u001b[0m\n", - " \u001b[37m6.801 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.028 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.801 train/loss:\u001b[0m\n", - " \u001b[37m6.801 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.042 train/loss:\u001b[0m\n", - " \u001b[37m7.042 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.042 train/loss:\u001b[0m\n", - " \u001b[37m7.042 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.851 train/loss:\u001b[0m\n", - " \u001b[37m6.851 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.851 train/loss:\u001b[0m\n", - " \u001b[37m6.851 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.904 train/loss:\u001b[0m\n", - " \u001b[37m6.904 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:01:01 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.904 train/loss:\u001b[0m\n", - " \u001b[37m6.904 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:01:02 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.850 train/loss:\u001b[0m\n", - " \u001b[37m6.850 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:01:02 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.850 train/loss:\u001b[0m\n", - " \u001b[37m6.850 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:01:02 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.885 train/loss:\u001b[0m\n", - " \u001b[37m6.885 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:01:02 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.885 train/loss:\u001b[0m\n", - " \u001b[37m6.885 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:01:02 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.815 train/loss:\u001b[0m\n", - " \u001b[37m6.815 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:01:03 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.815 train/loss:\u001b[0m\n", - " \u001b[37m6.815 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:01:03 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.871 train/loss:\u001b[0m\n", - " \u001b[37m6.871 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:01:03 • 0:00:21\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.871 train/loss:\u001b[0m\n", - " \u001b[37m6.871 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:01:03 • 0:00:21\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.864 train/loss:\u001b[0m\n", - " \u001b[37m6.864 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:01:04 • 0:00:21\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.864 train/loss:\u001b[0m\n", - " \u001b[37m6.864 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:01:04 • 0:00:21\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.936 train/loss:\u001b[0m\n", - " \u001b[37m6.936 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.936 train/loss:\u001b[0m\n", - " \u001b[37m6.936 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.031 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.878 train/loss:\u001b[0m\n", - " \u001b[37m6.878 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.031 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.878 train/loss:\u001b[0m\n", - " \u001b[37m6.878 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.026 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.902 train/loss:\u001b[0m\n", - " \u001b[37m6.902 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.026 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.902 train/loss:\u001b[0m\n", - " \u001b[37m6.902 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.955 train/loss:\u001b[0m\n", - " \u001b[37m6.955 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.955 train/loss:\u001b[0m\n", - " \u001b[37m6.955 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.852 train/loss:\u001b[0m\n", - " \u001b[37m6.852 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.852 train/loss:\u001b[0m\n", - " \u001b[37m6.852 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.869 train/loss:\u001b[0m\n", - " \u001b[37m6.869 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.869 train/loss:\u001b[0m\n", - " \u001b[37m6.869 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.902 train/loss:\u001b[0m\n", - " \u001b[37m6.902 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.902 train/loss:\u001b[0m\n", - " \u001b[37m6.902 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.872 train/loss:\u001b[0m\n", - " \u001b[37m6.872 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.872 train/loss:\u001b[0m\n", - " \u001b[37m6.872 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.959 train/loss:\u001b[0m\n", - " \u001b[37m6.959 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:07 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.959 train/loss:\u001b[0m\n", - " \u001b[37m6.959 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:07 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.851 train/loss:\u001b[0m\n", - " \u001b[37m6.851 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.851 train/loss:\u001b[0m\n", - " \u001b[37m6.851 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.966 train/loss:\u001b[0m\n", - " \u001b[37m6.966 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.966 train/loss:\u001b[0m\n", - " \u001b[37m6.966 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.838 train/loss:\u001b[0m\n", - " \u001b[37m6.838 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.838 train/loss:\u001b[0m\n", - " \u001b[37m6.838 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.921 train/loss:\u001b[0m\n", - " \u001b[37m6.921 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.921 train/loss:\u001b[0m\n", - " \u001b[37m6.921 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.838 train/loss:\u001b[0m\n", - " \u001b[37m6.838 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.838 train/loss:\u001b[0m\n", - " \u001b[37m6.838 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.012 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.876 train/loss:\u001b[0m\n", - " \u001b[37m6.876 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.012 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.876 train/loss:\u001b[0m\n", - " \u001b[37m6.876 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.829 train/loss:\u001b[0m\n", - " \u001b[37m6.829 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:10 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.829 train/loss:\u001b[0m\n", - " \u001b[37m6.829 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:10 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.862 train/loss:\u001b[0m\n", - " \u001b[37m6.862 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:11 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.862 train/loss:\u001b[0m\n", - " \u001b[37m6.862 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:11 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.914 train/loss:\u001b[0m\n", - " \u001b[37m6.914 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:11 • 0:00:12\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.914 train/loss:\u001b[0m\n", - " \u001b[37m6.914 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:11 • 0:00:12\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.050 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.764 train/loss:\u001b[0m\n", - " \u001b[37m6.764 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.050 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.764 train/loss:\u001b[0m\n", - " \u001b[37m6.764 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.895 train/loss:\u001b[0m\n", - " \u001b[37m6.895 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.895 train/loss:\u001b[0m\n", - " \u001b[37m6.895 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.911 train/loss:\u001b[0m\n", - " \u001b[37m6.911 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:13 • 0:00:11\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.911 train/loss:\u001b[0m\n", - " \u001b[37m6.911 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:13 • 0:00:11\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.845 train/loss:\u001b[0m\n", - " \u001b[37m6.845 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:14 • 0:00:11\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.845 train/loss:\u001b[0m\n", - " \u001b[37m6.845 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:14 • 0:00:11\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.828 train/loss:\u001b[0m\n", - " \u001b[37m6.828 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:15 • 0:00:11\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.828 train/loss:\u001b[0m\n", - " \u001b[37m6.828 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:15 • 0:00:11\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.887 train/loss:\u001b[0m\n", - " \u001b[37m6.887 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.887 train/loss:\u001b[0m\n", - " \u001b[37m6.887 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.836 train/loss:\u001b[0m\n", - " \u001b[37m6.836 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.836 train/loss:\u001b[0m\n", - " \u001b[37m6.836 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.854 train/loss:\u001b[0m\n", - " \u001b[37m6.854 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.854 train/loss:\u001b[0m\n", - " \u001b[37m6.854 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.854 train/loss:\u001b[0m\n", - " \u001b[37m6.854 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.854 train/loss:\u001b[0m\n", - " \u001b[37m6.854 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.915 train/loss:\u001b[0m\n", - " \u001b[37m6.915 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.915 train/loss:\u001b[0m\n", - " \u001b[37m6.915 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.931 train/loss:\u001b[0m\n", - " \u001b[37m6.931 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.931 train/loss:\u001b[0m\n", - " \u001b[37m6.931 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.966 train/loss:\u001b[0m\n", - " \u001b[37m6.966 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.966 train/loss:\u001b[0m\n", - " \u001b[37m6.966 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.800 train/loss:\u001b[0m\n", - " \u001b[37m6.800 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.800 train/loss:\u001b[0m\n", - " \u001b[37m6.800 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.989 train/loss:\u001b[0m\n", - " \u001b[37m6.989 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.989 train/loss:\u001b[0m\n", - " \u001b[37m6.989 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.854 train/loss:\u001b[0m\n", - " \u001b[37m6.854 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.854 train/loss:\u001b[0m\n", - " \u001b[37m6.854 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.731 train/loss:\u001b[0m\n", - " \u001b[37m6.731 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:19 • 0:00:05\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.731 train/loss:\u001b[0m\n", - " \u001b[37m6.731 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:19 • 0:00:05\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.039 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.065 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.726 train/loss:\u001b[0m\n", - " \u001b[37m6.726 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:20 • 0:00:05\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.039 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.065 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.726 train/loss:\u001b[0m\n", - " \u001b[37m6.726 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:20 • 0:00:05\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.851 train/loss:\u001b[0m\n", - " \u001b[37m6.851 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:20 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.851 train/loss:\u001b[0m\n", - " \u001b[37m6.851 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:20 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.945 train/loss:\u001b[0m\n", - " \u001b[37m6.945 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:21 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.945 train/loss:\u001b[0m\n", - " \u001b[37m6.945 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:21 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.852 train/loss:\u001b[0m\n", - " \u001b[37m6.852 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:21 • 0:00:03\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.852 train/loss:\u001b[0m\n", - " \u001b[37m6.852 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:21 • 0:00:03\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.031 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.797 train/loss:\u001b[0m\n", - " \u001b[37m6.797 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:22 • 0:00:03\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.031 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.797 train/loss:\u001b[0m\n", - " \u001b[37m6.797 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:22 • 0:00:03\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.819 train/loss:\u001b[0m\n", - " \u001b[37m6.819 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.819 train/loss:\u001b[0m\n", - " \u001b[37m6.819 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.871 train/loss:\u001b[0m\n", - " \u001b[37m6.871 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.871 train/loss:\u001b[0m\n", - " \u001b[37m6.871 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.764 train/loss:\u001b[0m\n", - " \u001b[37m6.764 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:23 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.764 train/loss:\u001b[0m\n", - " \u001b[37m6.764 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:23 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.791 train/loss:\u001b[0m\n", - " \u001b[37m6.791 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:24 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.791 train/loss:\u001b[0m\n", - " \u001b[37m6.791 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:24 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.066 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.775 train/loss:\u001b[0m\n", - " \u001b[37m6.775 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.066 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.775 train/loss:\u001b[0m\n", - " \u001b[37m6.775 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.766 train/loss:\u001b[0m\n", - " \u001b[37m6.766 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m140.174 \u001b[0m\n", - " \u001b[37mval/samples_per_…\u001b[0m\n", - " \u001b[37m138.743 \u001b[0m\n", - " \u001b[37mval/samples_per_…\u001b[0m\n", - " \u001b[37m136.694 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m146.335 \u001b[0m\n", - "\u001b[?25h/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=example trainer.max_epochs ...\n", - "\u001b[2;36m[17:09:43]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Not adding metrics to the pl \u001b]8;id=733785;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=345137;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#64\u001b\\\u001b[2m64\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module because they are already \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m present. \u001b[2m \u001b[0m\n", - "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: ClassificationMetricsCallback, LearningRateMonitor\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=371304;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=737977;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#381\u001b\\\u001b[2m381\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=648908;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=28924;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m[17:10:01]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Not adding metrics to the pl \u001b]8;id=33733;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=344699;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#64\u001b\\\u001b[2m64\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module because they are already \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m present. \u001b[2m \u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37m2.17it/s\u001b[0m \n", - "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", - "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", - "│\u001b[36m \u001b[0m\u001b[36m val/accuracy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.003010033629834652 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36m val/cross_entropy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 6.422720432281494 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 6.422720432281494 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 140.8275146484375 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36m val/top5_accuracy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.029230769723653793 \u001b[0m\u001b[35m \u001b[0m│\n", - "└──────────────────────────────┴──────────────────────────────┘\n", - "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.17it/s\u001b[0m \n", - "\u001b[?25hval val/accuracy: \u001b[1;36m0.003010033629834652\u001b[0m\n", - "val val/top5_accuracy: \u001b[1;36m0.029230769723653793\u001b[0m\n", - "val val/cross_entropy: \u001b[1;36m6.422720432281494\u001b[0m\n", - "val val/samples_per_second_epoch: \u001b[1;36m140.8275146484375\u001b[0m\n" - ] - } - ], + "outputs": [], "source": [ - "!HYDRA_FULL_ERROR=1 python project/main.py \\\n", - " algorithm=example \\\n", - " trainer.max_epochs=1 \\\n", - " +trainer.limit_train_batches=0.01\\\n", - " +trainer.limit_val_batches=0.01\\\n", - " datamodule=imagenet" + "#!HYDRA_FULL_ERROR=1 python project/main.py \\\n", + "# algorithm=example \\\n", + "# trainer.max_epochs=1 \\\n", + "# +trainer.limit_train_batches=0.01\\\n", + "# +trainer.limit_val_batches=0.01\\\n", + "# datamodule=imagenet" ] }, { @@ -5745,28 +83,11 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "GPU Avail / Total \n", - "===============================\n", - "2g.20gb 31 / 48 \n", - "3g.40gb 9 / 48 \n", - "4g.40gb 7 / 24 \n", - "a100 8 / 16 \n", - "a100l 0 / 72 \n", - "a6000 0 / 8 \n", - "rtx8000 11 / 400 \n", - "v100 2 / 40 \n" - ] - } - ], + "outputs": [], "source": [ - "!savail" + "#!savail" ] }, { @@ -5814,7 +135,7 @@ }, { "cell_type": "code", - "execution_count": 86, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ @@ -5832,7 +153,7 @@ }, { "cell_type": "code", - "execution_count": 74, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ @@ -5842,7 +163,7 @@ }, { "cell_type": "code", - "execution_count": 75, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ @@ -5859,7 +180,7 @@ }, { "cell_type": "code", - "execution_count": 76, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ @@ -5897,7 +218,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.12.4" + "version": "3.10.14" } }, "nbformat": 4, From dae2e14e2c10b7624cc3381a86f380305eb9c2fb Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Tue, 3 Sep 2024 15:45:27 -0400 Subject: [PATCH 04/33] nbstripout compliance --- docs/examples/profiling.ipynb | 30 +- mkdocs.yml | 5 +- pyproject.toml | 6 +- requirements-dev.lock | 507 ++++++++++++++++++++++------------ requirements.lock | 456 ++++++++++++++++++++---------- 5 files changed, 678 insertions(+), 326 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 9e6ae30a..84d99e02 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -4,7 +4,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "# Benchmarking" + "# Profiling your code" ] }, { @@ -41,12 +41,28 @@ "metadata": {}, "outputs": [], "source": [ - "#!python project/main.py \\\n", - "# algorithm=NoOp \\\n", - "# trainer.max_epochs=1 \\\n", - "# +trainer.limit_train_batches=0.01\\\n", - "# +trainer.limit_val_batches=0.01\\\n", - "# datamodule=imagenet" + "import os\n", + "from pathlib import Path\n", + "\n", + "# Set the working directory to the project root\n", + "notebook_path = Path().resolve() \n", + "project_root = notebook_path.parent.parent\n", + "os.chdir(str(project_root))\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "!python project/main.py \\\n", + " algorithm=example \\\n", + " datamodule=imagenet \\\n", + " ++trainer.max_epochs=0 \\\n", + " ++trainer.limit_train_batches=0\\\n", + "\n", + "#!python ../../project/main.py throws an error about relative paths" ] }, { diff --git a/mkdocs.yml b/mkdocs.yml index 3e3e813a..9f7e4387 100644 --- a/mkdocs.yml +++ b/mkdocs.yml @@ -3,7 +3,9 @@ site_description: A project template and directory structure for Python data sci site_url: https://mila-iqia.github.io/ResearchTemplate/ repo_url: https://www.github.com/mila-iqia/ResearchTemplate # edit_uri: edit/master/docs - +nav: + - Home: index.md + - Profiling your code: docs/examples/profiling.ipynb theme: name: material features: @@ -92,6 +94,7 @@ plugins: video_controls: True css_style: width: "100%" + - mkdocs-jupyter # todo: take a look at https://github.com/drivendataorg/cookiecutter-data-science/blob/master/docs/mkdocs.yml # - admonition # - pymdownx.details diff --git a/pyproject.toml b/pyproject.toml index 04debe18..dc991099 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -23,10 +23,11 @@ dependencies = [ "torch-jax-interop>=0.0.7", "pydantic>=2.8.2", "simple-parsing>=0.1.5", - "pydantic==2.7.4", + "pydantic==2.8.2", "milatools>=0.0.18", "rootutils>=0.0.1", - "ipykernel>=6.28.0" + "ipykernel>=6.28.0", + "dill>=0.3.8", ] readme = "README.md" requires-python = ">= 3.10" @@ -42,6 +43,7 @@ docs = [ "mkdocs>=1.6.0", "mkdocs-video>=1.5.0", "mkdocs-section-index>=0.3.9", + "mkdocs-jupyter>=0.24.8", ] gpu = ["jax[cuda12]>=0.4.31"] diff --git a/requirements-dev.lock b/requirements-dev.lock index 93aa7a90..b873c192 100644 --- a/requirements-dev.lock +++ b/requirements-dev.lock @@ -10,126 +10,166 @@ # universal: true -e file:. -absl-py==2.1.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via file:///- +absl-py==2.1.0 # via chex # via optax # via orbax-checkpoint -aiohappyeyeballs==2.3.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +aiohappyeyeballs==2.3.5 # via aiohttp -aiohttp==3.10.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +aiohttp==3.10.3 # via fsspec -aiosignal==1.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +aiosignal==1.3.1 # via aiohttp -annotated-types==0.7.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +annotated-types==0.7.0 # via pydantic -antlr4-python3-runtime==4.9.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +antlr4-python3-runtime==4.9.3 # via hydra-core # via omegaconf -async-timeout==4.0.3 ; python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +asttokens==2.4.1 + # via stack-data +async-timeout==4.0.3 # via aiohttp -attrs==24.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +attrs==24.2.0 # via aiohttp -babel==2.16.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via jsonschema + # via referencing +babel==2.16.0 # via mkdocs-material -beautifulsoup4==4.12.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +bcrypt==4.2.0 + # via paramiko +beautifulsoup4==4.12.3 # via gdown -black==24.8.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via nbconvert +black==24.8.0 # via research-project-template -bracex==2.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +bleach==6.1.0 + # via nbconvert +blessed==1.20.0 + # via milatools +bracex==2.5 # via wcmatch -certifi==2024.7.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +certifi==2024.7.4 # via requests # via sentry-sdk -cfgv==3.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +cffi==1.17.0 + # via cryptography + # via pynacl +cfgv==3.4.0 # via pre-commit -charset-normalizer==3.3.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +charset-normalizer==3.3.2 # via requests -chex==0.1.86 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +chex==0.1.86 # via optax -click==8.1.7 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +click==8.1.7 # via black # via mkdocs # via mkdocstrings # via wandb -cloudpickle==3.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +cloudpickle==3.0.0 # via submitit -colorama==0.4.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_system == 'Windows' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (sys_platform == 'win32' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) - # via click - # via colorlog +colorama==0.4.6 # via griffe - # via mkdocs # via mkdocs-material - # via pytest - # via tqdm -colorlog==6.8.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +colorlog==6.8.2 # via hydra-colorlog -contourpy==1.2.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +comm==0.2.2 + # via ipykernel +contourpy==1.2.1 # via matplotlib -coverage==7.6.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +coverage==7.6.1 + # via coverage # via pytest-cov # via pytest-testmon -cycler==0.12.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +cryptography==43.0.0 + # via paramiko +cycler==0.12.1 # via matplotlib -distlib==0.3.8 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +debugpy==1.8.5 + # via ipykernel +decorator==5.1.1 + # via fabric + # via ipython +defusedxml==0.7.1 + # via nbconvert +deprecated==1.2.14 + # via fabric +dill==0.3.8 + # via research-project-template +distlib==0.3.8 # via virtualenv -docker-pycreds==0.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +docker-pycreds==0.4.0 # via wandb -docstring-parser==0.16 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +docstring-parser==0.16 # via simple-parsing -etils==1.7.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +etils==1.7.0 # via optax # via orbax-checkpoint -exceptiongroup==1.2.2 ; python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +exceptiongroup==1.2.2 + # via ipython # via pytest -execnet==2.1.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +execnet==2.1.1 # via pytest-xdist -filelock==3.15.4 ; (python_version < '3.13' and platform_machine == 'x86_64' and platform_system == 'Linux') or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +executing==2.0.1 + # via stack-data +fabric==3.2.2 + # via milatools +fastjsonschema==2.20.0 + # via nbformat +filelock==3.15.4 # via gdown # via torch # via triton # via virtualenv -flax==0.8.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +flax==0.8.5 # via torch-jax-interop -fonttools==4.53.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +fonttools==4.53.1 # via matplotlib -frozenlist==1.4.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +frozenlist==1.4.1 # via aiohttp # via aiosignal -fsspec==2024.6.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +fsspec==2024.6.1 # via etils # via lightning # via pytorch-lightning # via torch -gdown==5.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +gdown==5.2.0 # via research-project-template -ghp-import==2.1.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +ghp-import==2.1.0 # via mkdocs -gitdb==4.0.11 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +gitdb==4.0.11 # via gitpython -gitpython==3.1.43 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +gitpython==3.1.43 # via wandb -griffe==0.49.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +griffe==0.49.0 # via mkdocstrings-python -hydra-colorlog==1.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-colorlog==1.2.0 # via research-project-template -hydra-core==1.3.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-core==1.3.2 # via hydra-colorlog # via hydra-submitit-launcher # via hydra-zen # via research-project-template -hydra-submitit-launcher==1.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-submitit-launcher==1.2.0 # via research-project-template -hydra-zen==0.13.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-zen==0.13.0 # via research-project-template -identify==2.6.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +identify==2.6.0 # via pre-commit -idna==3.7 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +idna==3.7 # via requests # via yarl -importlib-resources==6.4.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +importlib-resources==6.4.2 # via etils -iniconfig==2.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +iniconfig==2.0.0 # via pytest +invoke==2.2.0 + # via fabric +ipykernel==6.29.5 + # via mkdocs-jupyter + # via research-project-template +ipython==8.27.0 + # via ipykernel jax==0.4.31 # via chex # via flax @@ -138,108 +178,155 @@ jax==0.4.31 # via pytorch2jax # via research-project-template # via torch-jax-interop -jax-cuda12-pjrt==0.4.31 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +jax-cuda12-pjrt==0.4.31 # via jax-cuda12-plugin jax-cuda12-plugin==0.4.31 # via jax -jaxlib==0.4.31 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via jax-cuda12-plugin +jaxlib==0.4.31 # via chex # via jax # via optax # via orbax-checkpoint # via pytorch2jax -jinja2==3.1.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +jedi==0.19.1 + # via ipython +jinja2==3.1.4 # via mkdocs # via mkdocs-material # via mkdocstrings + # via nbconvert # via torch -kiwisolver==1.4.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +jsonschema==4.23.0 + # via nbformat +jsonschema-specifications==2023.12.1 + # via jsonschema +jupyter-client==8.6.2 + # via ipykernel + # via nbclient +jupyter-core==5.7.2 + # via ipykernel + # via jupyter-client + # via nbclient + # via nbconvert + # via nbformat +jupyterlab-pygments==0.3.0 + # via nbconvert +jupytext==1.16.4 + # via mkdocs-jupyter +kiwisolver==1.4.5 # via matplotlib -lightning==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +lightning==2.4.0 # via research-project-template -lightning-utilities==0.11.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +lightning-utilities==0.11.6 # via lightning # via pytorch-lightning # via torchmetrics -lxml==5.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +lxml==5.3.0 # via mkdocs-video -markdown==3.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +markdown==3.6 # via mkdocs # via mkdocs-autorefs # via mkdocs-material # via mkdocstrings # via pymdown-extensions -markdown-it-py==3.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +markdown-it-py==3.0.0 + # via jupytext + # via mdit-py-plugins # via rich -markupsafe==2.1.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +markupsafe==2.1.5 # via jinja2 # via mkdocs # via mkdocs-autorefs # via mkdocstrings -matplotlib==3.9.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via nbconvert +matplotlib==3.9.2 # via research-project-template -mdurl==0.1.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +matplotlib-inline==0.1.7 + # via ipykernel + # via ipython +mdit-py-plugins==0.4.1 + # via jupytext +mdurl==0.1.2 # via markdown-it-py -mergedeep==1.3.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mergedeep==1.3.4 # via mkdocs # via mkdocs-get-deps -mkdocs==1.6.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +milatools==0.1.5 + # via research-project-template +mistune==3.0.2 + # via nbconvert +mkdocs==1.6.0 # via mkdocs-autorefs # via mkdocs-awesome-pages-plugin # via mkdocs-gen-files + # via mkdocs-jupyter # via mkdocs-literate-nav # via mkdocs-material # via mkdocs-section-index # via mkdocs-video # via mkdocstrings # via research-project-template -mkdocs-autorefs==1.0.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-autorefs==1.0.1 # via mkdocstrings -mkdocs-awesome-pages-plugin==2.9.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-awesome-pages-plugin==2.9.3 # via research-project-template -mkdocs-gen-files==0.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-gen-files==0.5.0 # via research-project-template -mkdocs-get-deps==0.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-get-deps==0.2.0 # via mkdocs -mkdocs-literate-nav==0.6.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-jupyter==0.24.8 + # via research-project-template +mkdocs-literate-nav==0.6.1 # via research-project-template -mkdocs-material==9.5.31 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-material==9.5.31 + # via mkdocs-jupyter # via research-project-template -mkdocs-material-extensions==1.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-material-extensions==1.3.1 # via mkdocs-material -mkdocs-section-index==0.3.9 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-section-index==0.3.9 # via research-project-template -mkdocs-video==1.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-video==1.5.0 # via research-project-template -mkdocstrings==0.25.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocstrings==0.25.2 + # via mkdocstrings # via mkdocstrings-python # via research-project-template -mkdocstrings-python==1.10.8 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocstrings-python==1.10.8 # via mkdocstrings -mktestdocs==0.2.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -ml-dtypes==0.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mktestdocs==0.2.2 +ml-dtypes==0.4.0 # via jax # via jaxlib # via tensorstore -mpmath==1.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mpmath==1.3.0 # via sympy -msgpack==1.0.8 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +msgpack==1.0.8 # via flax # via orbax-checkpoint -multidict==6.0.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +multidict==6.0.5 # via aiohttp # via yarl -mypy-extensions==1.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mypy-extensions==1.0.0 # via black -natsort==8.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +natsort==8.4.0 # via mkdocs-awesome-pages-plugin -nest-asyncio==1.6.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +nbclient==0.10.0 + # via nbconvert +nbconvert==7.16.4 + # via mkdocs-jupyter +nbformat==5.10.4 + # via jupytext + # via nbclient + # via nbconvert +nest-asyncio==1.6.0 + # via ipykernel # via orbax-checkpoint -networkx==3.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +networkx==3.3 # via torch -nodeenv==1.9.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +nodeenv==1.9.1 # via pre-commit -numpy==1.26.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or ((python_version >= '3.11' or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or ((python_version >= '3.12' or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +numpy==1.26.4 # via chex # via contourpy # via flax @@ -255,103 +342,130 @@ numpy==1.26.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python # via tensorstore # via torchmetrics # via torchvision -nvidia-cublas-cu12==12.1.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))) +nvidia-cublas-cu12==12.1.3.1 # via jax-cuda12-plugin # via nvidia-cudnn-cu12 # via nvidia-cusolver-cu12 # via torch -nvidia-cuda-cupti-cu12==12.1.105 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cuda-cupti-cu12==12.1.105 # via jax-cuda12-plugin # via torch -nvidia-cuda-nvcc-cu12==12.6.20 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +nvidia-cuda-nvcc-cu12==12.6.20 # via jax-cuda12-plugin -nvidia-cuda-nvrtc-cu12==12.1.105 ; platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +nvidia-cuda-nvrtc-cu12==12.1.105 # via torch -nvidia-cuda-runtime-cu12==12.1.105 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cuda-runtime-cu12==12.1.105 # via jax-cuda12-plugin # via torch -nvidia-cudnn-cu12==9.1.0.70 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cudnn-cu12==9.1.0.70 # via jax-cuda12-plugin # via torch -nvidia-cufft-cu12==11.0.2.54 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cufft-cu12==11.0.2.54 # via jax-cuda12-plugin # via torch -nvidia-curand-cu12==10.3.2.106 ; platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +nvidia-curand-cu12==10.3.2.106 # via torch -nvidia-cusolver-cu12==11.4.5.107 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cusolver-cu12==11.4.5.107 # via jax-cuda12-plugin # via torch -nvidia-cusparse-cu12==12.1.0.106 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))) +nvidia-cusparse-cu12==12.1.0.106 # via jax-cuda12-plugin # via nvidia-cusolver-cu12 # via torch -nvidia-nccl-cu12==2.20.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-nccl-cu12==2.20.5 # via jax-cuda12-plugin # via torch -nvidia-nvjitlink-cu12==12.6.20 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))))) +nvidia-nvjitlink-cu12==12.6.20 # via jax-cuda12-plugin # via nvidia-cusolver-cu12 # via nvidia-cusparse-cu12 -nvidia-nvtx-cu12==12.1.105 ; platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +nvidia-nvtx-cu12==12.1.105 # via torch -omegaconf==2.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +omegaconf==2.3.0 # via hydra-core # via hydra-zen -opt-einsum==3.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +opt-einsum==3.3.0 # via jax -optax==0.2.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +optax==0.2.3 # via flax -orbax-checkpoint==0.5.23 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +orbax-checkpoint==0.5.23 # via flax -packaging==24.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +packaging==24.1 # via black # via hydra-core + # via ipykernel + # via jupytext # via lightning # via lightning-utilities # via matplotlib # via mkdocs + # via nbconvert # via pytest # via pytorch-lightning # via torchmetrics -paginate==0.5.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +paginate==0.5.6 # via mkdocs-material -pathspec==0.12.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pandocfilters==1.5.1 + # via nbconvert +paramiko==3.4.1 + # via fabric +parso==0.8.4 + # via jedi +pathspec==0.12.1 # via black # via mkdocs -pillow==10.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pexpect==4.9.0 + # via ipython +pillow==10.4.0 # via matplotlib # via torchvision -platformdirs==4.2.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +platformdirs==4.2.2 # via black + # via jupyter-core # via mkdocs-get-deps # via mkdocstrings # via virtualenv # via wandb -pluggy==1.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pluggy==1.5.0 # via pytest -pre-commit==3.8.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -protobuf==5.27.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pre-commit==3.8.0 +prompt-toolkit==3.0.47 + # via ipython + # via questionary +protobuf==5.27.3 # via orbax-checkpoint # via wandb -psutil==6.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +psutil==6.0.0 + # via ipykernel # via wandb -py-cpuinfo==9.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +ptyprocess==0.7.0 + # via pexpect +pure-eval==0.2.3 + # via stack-data +py-cpuinfo==9.0.0 # via pytest-benchmark -pydantic==2.8.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pycparser==2.22 + # via cffi +pydantic==2.8.2 # via research-project-template -pydantic-core==2.20.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pydantic-core==2.20.1 # via pydantic -pygments==2.18.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pygments==2.18.0 + # via ipython + # via mkdocs-jupyter # via mkdocs-material + # via nbconvert # via rich -pymdown-extensions==10.9 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pymdown-extensions==10.9 # via mkdocs-material # via mkdocstrings -pyparsing==3.1.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pynacl==1.5.0 + # via paramiko +pyparsing==3.1.2 # via matplotlib -pysocks==1.7.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pysocks==1.7.1 # via requests -pytest==8.3.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pytest==8.3.2 # via pytest-benchmark # via pytest-cov # via pytest-datadir @@ -361,26 +475,30 @@ pytest==8.3.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python # via pytest-testmon # via pytest-timeout # via pytest-xdist -pytest-benchmark==4.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -pytest-cov==5.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -pytest-datadir==1.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pytest-benchmark==4.0.0 +pytest-cov==5.0.0 +pytest-datadir==1.5.0 # via pytest-regressions -pytest-env==1.1.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -pytest-regressions==2.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pytest-env==1.1.3 +pytest-regressions==2.5.0 # via tensor-regression -pytest-skip-slow==0.0.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -pytest-testmon==2.1.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -pytest-timeout==2.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -pytest-xdist==3.6.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -python-dateutil==2.9.0.post0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pytest-skip-slow==0.0.5 +pytest-testmon==2.1.1 +pytest-timeout==2.3.1 +pytest-xdist==3.6.1 +python-dateutil==2.9.0.post0 # via ghp-import + # via jupyter-client # via matplotlib -pytorch-lightning==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +python-dotenv==1.0.1 + # via rootutils +pytorch-lightning==2.4.0 # via lightning -pytorch2jax==0.1.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pytorch2jax==0.1.0 # via torch-jax-interop -pyyaml==6.0.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pyyaml==6.0.2 # via flax + # via jupytext # via lightning # via mkdocs # via mkdocs-get-deps @@ -392,54 +510,74 @@ pyyaml==6.0.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python # via pytorch-lightning # via pyyaml-env-tag # via wandb -pyyaml-env-tag==0.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pyyaml-env-tag==0.1 # via mkdocs -regex==2024.7.24 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pyzmq==26.2.0 + # via ipykernel + # via jupyter-client +questionary==1.10.0 + # via milatools +referencing==0.35.1 + # via jsonschema + # via jsonschema-specifications +regex==2024.7.24 # via mkdocs-material -requests==2.32.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +requests==2.32.3 # via gdown # via mkdocs-material # via wandb -rich==13.7.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +rich==13.7.1 # via flax + # via milatools + # via research-project-template +rootutils==1.0.7 # via research-project-template -ruff==0.6.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -scipy==1.14.0 ; (python_version >= '3.12' or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +rpds-py==0.20.0 + # via jsonschema + # via referencing +ruff==0.6.0 +scipy==1.14.0 # via jax # via jaxlib -sentry-sdk==2.13.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +sentry-sdk==2.13.0 # via wandb -setproctitle==1.3.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +setproctitle==1.3.3 # via wandb -setuptools==72.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.12' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') - # via chex - # via lightning-utilities - # via wandb -simple-parsing==0.1.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +simple-parsing==0.1.5 # via research-project-template -six==1.16.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +six==1.16.0 + # via asttokens + # via bleach + # via blessed # via docker-pycreds # via python-dateutil -smmap==5.0.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +smmap==5.0.1 # via gitdb -soupsieve==2.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +soupsieve==2.6 # via beautifulsoup4 -submitit==1.5.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +sshconf==0.2.7 + # via milatools +stack-data==0.6.3 + # via ipython +submitit==1.5.1 # via hydra-submitit-launcher -sympy==1.13.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +sympy==1.13.2 # via torch -tensor-regression==0.0.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -tensorstore==0.1.64 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +tensor-regression==0.0.6 +tensorstore==0.1.64 # via flax # via orbax-checkpoint -tomli==2.0.1 ; (python_full_version <= '3.11.0a6' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +tinycss2==1.3.0 + # via nbconvert +tomli==2.0.1 # via black # via coverage + # via jupytext # via pytest # via pytest-env -toolz==0.12.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +toolz==0.12.1 # via chex -torch==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torch==2.4.0 # via lightning # via pytorch-lightning # via pytorch2jax @@ -448,28 +586,44 @@ torch==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_ # via torch-jax-interop # via torchmetrics # via torchvision -torch-jax-interop==0.0.7 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torch-jax-interop==0.0.7 # via research-project-template -torchmetrics==1.4.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torchmetrics==1.4.1 # via lightning # via pytorch-lightning -torchvision==0.19.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torchvision==0.19.0 # via research-project-template -tqdm==4.66.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +tornado==6.4.1 + # via ipykernel + # via jupyter-client +tqdm==4.66.5 # via gdown # via lightning + # via milatools # via pytorch-lightning # via research-project-template -triton==3.0.0 ; python_version < '3.13' and platform_machine == 'x86_64' and platform_system == 'Linux' +traitlets==5.14.3 + # via comm + # via ipykernel + # via ipython + # via jupyter-client + # via jupyter-core + # via matplotlib-inline + # via nbclient + # via nbconvert + # via nbformat +triton==3.0.0 # via torch -typing-extensions==4.12.2 ; (python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +typing-extensions==4.12.2 # via black # via chex # via etils # via flax # via hydra-zen + # via ipython # via lightning # via lightning-utilities + # via milatools # via orbax-checkpoint # via pydantic # via pydantic-core @@ -477,18 +631,29 @@ typing-extensions==4.12.2 ; (python_version < '3.11' and ((python_version < '3.1 # via simple-parsing # via submitit # via torch -urllib3==2.2.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +urllib3==2.2.2 # via requests # via sentry-sdk -virtualenv==20.26.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +virtualenv==20.26.3 # via pre-commit -wandb==0.17.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +wandb==0.17.6 # via research-project-template -watchdog==4.0.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +watchdog==4.0.2 # via mkdocs -wcmatch==9.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +wcmatch==9.0 # via mkdocs-awesome-pages-plugin -yarl==1.9.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +wcwidth==0.2.13 + # via blessed + # via prompt-toolkit +webencodings==0.5.1 + # via bleach + # via tinycss2 +wrapt==1.16.0 + # via deprecated +yarl==1.9.4 # via aiohttp -zipp==3.20.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +zipp==3.20.0 # via etils +setuptools==72.2.0 + # via lightning-utilities + # via wandb diff --git a/requirements.lock b/requirements.lock index a1e9e7ec..b253afa2 100644 --- a/requirements.lock +++ b/requirements.lock @@ -10,109 +10,150 @@ # universal: true -e file:. -absl-py==2.1.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via file:///- +absl-py==2.1.0 # via chex # via optax # via orbax-checkpoint -aiohappyeyeballs==2.3.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +aiohappyeyeballs==2.3.5 # via aiohttp -aiohttp==3.10.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +aiohttp==3.10.3 # via fsspec -aiosignal==1.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +aiosignal==1.3.1 # via aiohttp -annotated-types==0.7.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +annotated-types==0.7.0 # via pydantic -antlr4-python3-runtime==4.9.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +antlr4-python3-runtime==4.9.3 # via hydra-core # via omegaconf -async-timeout==4.0.3 ; python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +asttokens==2.4.1 + # via stack-data +async-timeout==4.0.3 # via aiohttp -attrs==24.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +attrs==24.2.0 # via aiohttp -babel==2.16.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via jsonschema + # via referencing +babel==2.16.0 # via mkdocs-material -beautifulsoup4==4.12.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +bcrypt==4.2.0 + # via paramiko +beautifulsoup4==4.12.3 # via gdown -black==24.8.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via nbconvert +black==24.8.0 # via research-project-template -bracex==2.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +bleach==6.1.0 + # via nbconvert +blessed==1.20.0 + # via milatools +bracex==2.5 # via wcmatch -certifi==2024.7.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +certifi==2024.7.4 # via requests # via sentry-sdk -charset-normalizer==3.3.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +cffi==1.17.0 + # via cryptography + # via pynacl +charset-normalizer==3.3.2 # via requests -chex==0.1.86 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +chex==0.1.86 # via optax -click==8.1.7 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +click==8.1.7 # via black # via mkdocs # via mkdocstrings # via wandb -cloudpickle==3.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +cloudpickle==3.0.0 # via submitit -colorama==0.4.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_system == 'Windows' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (sys_platform == 'win32' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) - # via click - # via colorlog +colorama==0.4.6 # via griffe - # via mkdocs # via mkdocs-material - # via tqdm -colorlog==6.8.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +colorlog==6.8.2 # via hydra-colorlog -contourpy==1.2.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +comm==0.2.2 + # via ipykernel +contourpy==1.2.1 # via matplotlib -cycler==0.12.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +cryptography==43.0.0 + # via paramiko +cycler==0.12.1 # via matplotlib -docker-pycreds==0.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +debugpy==1.8.5 + # via ipykernel +decorator==5.1.1 + # via fabric + # via ipython +defusedxml==0.7.1 + # via nbconvert +deprecated==1.2.14 + # via fabric +dill==0.3.8 + # via research-project-template +docker-pycreds==0.4.0 # via wandb -docstring-parser==0.16 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +docstring-parser==0.16 # via simple-parsing -etils==1.7.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +etils==1.7.0 # via optax # via orbax-checkpoint -filelock==3.15.4 ; (python_version < '3.13' and platform_machine == 'x86_64' and platform_system == 'Linux') or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +exceptiongroup==1.2.2 + # via ipython +executing==2.0.1 + # via stack-data +fabric==3.2.2 + # via milatools +fastjsonschema==2.20.0 + # via nbformat +filelock==3.15.4 # via gdown # via torch # via triton -flax==0.8.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +flax==0.8.5 # via torch-jax-interop -fonttools==4.53.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +fonttools==4.53.1 # via matplotlib -frozenlist==1.4.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +frozenlist==1.4.1 # via aiohttp # via aiosignal -fsspec==2024.6.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +fsspec==2024.6.1 # via etils # via lightning # via pytorch-lightning # via torch -gdown==5.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +gdown==5.2.0 # via research-project-template -ghp-import==2.1.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +ghp-import==2.1.0 # via mkdocs -gitdb==4.0.11 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +gitdb==4.0.11 # via gitpython -gitpython==3.1.43 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +gitpython==3.1.43 # via wandb -griffe==0.49.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +griffe==0.49.0 # via mkdocstrings-python -hydra-colorlog==1.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-colorlog==1.2.0 # via research-project-template -hydra-core==1.3.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-core==1.3.2 # via hydra-colorlog # via hydra-submitit-launcher # via hydra-zen # via research-project-template -hydra-submitit-launcher==1.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-submitit-launcher==1.2.0 # via research-project-template -hydra-zen==0.13.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-zen==0.13.0 # via research-project-template -idna==3.7 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +idna==3.7 # via requests # via yarl -importlib-resources==6.4.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +importlib-resources==6.4.2 # via etils +invoke==2.2.0 + # via fabric +ipykernel==6.29.5 + # via mkdocs-jupyter + # via research-project-template +ipython==8.27.0 + # via ipykernel jax==0.4.31 # via chex # via flax @@ -121,105 +162,152 @@ jax==0.4.31 # via pytorch2jax # via research-project-template # via torch-jax-interop -jax-cuda12-pjrt==0.4.31 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +jax-cuda12-pjrt==0.4.31 # via jax-cuda12-plugin jax-cuda12-plugin==0.4.31 # via jax -jaxlib==0.4.31 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via jax-cuda12-plugin +jaxlib==0.4.31 # via chex # via jax # via optax # via orbax-checkpoint # via pytorch2jax -jinja2==3.1.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +jedi==0.19.1 + # via ipython +jinja2==3.1.4 # via mkdocs # via mkdocs-material # via mkdocstrings + # via nbconvert # via torch -kiwisolver==1.4.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +jsonschema==4.23.0 + # via nbformat +jsonschema-specifications==2023.12.1 + # via jsonschema +jupyter-client==8.6.2 + # via ipykernel + # via nbclient +jupyter-core==5.7.2 + # via ipykernel + # via jupyter-client + # via nbclient + # via nbconvert + # via nbformat +jupyterlab-pygments==0.3.0 + # via nbconvert +jupytext==1.16.4 + # via mkdocs-jupyter +kiwisolver==1.4.5 # via matplotlib -lightning==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +lightning==2.4.0 # via research-project-template -lightning-utilities==0.11.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +lightning-utilities==0.11.6 # via lightning # via pytorch-lightning # via torchmetrics -lxml==5.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +lxml==5.3.0 # via mkdocs-video -markdown==3.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +markdown==3.6 # via mkdocs # via mkdocs-autorefs # via mkdocs-material # via mkdocstrings # via pymdown-extensions -markdown-it-py==3.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +markdown-it-py==3.0.0 + # via jupytext + # via mdit-py-plugins # via rich -markupsafe==2.1.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +markupsafe==2.1.5 # via jinja2 # via mkdocs # via mkdocs-autorefs # via mkdocstrings -matplotlib==3.9.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via nbconvert +matplotlib==3.9.2 # via research-project-template -mdurl==0.1.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +matplotlib-inline==0.1.7 + # via ipykernel + # via ipython +mdit-py-plugins==0.4.1 + # via jupytext +mdurl==0.1.2 # via markdown-it-py -mergedeep==1.3.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mergedeep==1.3.4 # via mkdocs # via mkdocs-get-deps -mkdocs==1.6.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +milatools==0.1.5 + # via research-project-template +mistune==3.0.2 + # via nbconvert +mkdocs==1.6.0 # via mkdocs-autorefs # via mkdocs-awesome-pages-plugin # via mkdocs-gen-files + # via mkdocs-jupyter # via mkdocs-literate-nav # via mkdocs-material # via mkdocs-section-index # via mkdocs-video # via mkdocstrings # via research-project-template -mkdocs-autorefs==1.0.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-autorefs==1.0.1 # via mkdocstrings -mkdocs-awesome-pages-plugin==2.9.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-awesome-pages-plugin==2.9.3 # via research-project-template -mkdocs-gen-files==0.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-gen-files==0.5.0 # via research-project-template -mkdocs-get-deps==0.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-get-deps==0.2.0 # via mkdocs -mkdocs-literate-nav==0.6.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-jupyter==0.24.8 + # via research-project-template +mkdocs-literate-nav==0.6.1 # via research-project-template -mkdocs-material==9.5.31 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-material==9.5.31 + # via mkdocs-jupyter # via research-project-template -mkdocs-material-extensions==1.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-material-extensions==1.3.1 # via mkdocs-material -mkdocs-section-index==0.3.9 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-section-index==0.3.9 # via research-project-template -mkdocs-video==1.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-video==1.5.0 # via research-project-template -mkdocstrings==0.25.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocstrings==0.25.2 + # via mkdocstrings # via mkdocstrings-python # via research-project-template -mkdocstrings-python==1.10.8 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocstrings-python==1.10.8 # via mkdocstrings -ml-dtypes==0.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +ml-dtypes==0.4.0 # via jax # via jaxlib # via tensorstore -mpmath==1.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mpmath==1.3.0 # via sympy -msgpack==1.0.8 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +msgpack==1.0.8 # via flax # via orbax-checkpoint -multidict==6.0.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +multidict==6.0.5 # via aiohttp # via yarl -mypy-extensions==1.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mypy-extensions==1.0.0 # via black -natsort==8.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +natsort==8.4.0 # via mkdocs-awesome-pages-plugin -nest-asyncio==1.6.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +nbclient==0.10.0 + # via nbconvert +nbconvert==7.16.4 + # via mkdocs-jupyter +nbformat==5.10.4 + # via jupytext + # via nbclient + # via nbconvert +nest-asyncio==1.6.0 + # via ipykernel # via orbax-checkpoint -networkx==3.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +networkx==3.3 # via torch -numpy==1.26.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or ((python_version >= '3.11' or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or ((python_version >= '3.12' or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +numpy==1.26.4 # via chex # via contourpy # via flax @@ -234,104 +322,135 @@ numpy==1.26.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python # via tensorstore # via torchmetrics # via torchvision -nvidia-cublas-cu12==12.1.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))) +nvidia-cublas-cu12==12.1.3.1 # via jax-cuda12-plugin # via nvidia-cudnn-cu12 # via nvidia-cusolver-cu12 # via torch -nvidia-cuda-cupti-cu12==12.1.105 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cuda-cupti-cu12==12.1.105 # via jax-cuda12-plugin # via torch -nvidia-cuda-nvcc-cu12==12.6.20 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +nvidia-cuda-nvcc-cu12==12.6.20 # via jax-cuda12-plugin -nvidia-cuda-nvrtc-cu12==12.1.105 ; platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +nvidia-cuda-nvrtc-cu12==12.1.105 # via torch -nvidia-cuda-runtime-cu12==12.1.105 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cuda-runtime-cu12==12.1.105 # via jax-cuda12-plugin # via torch -nvidia-cudnn-cu12==9.1.0.70 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cudnn-cu12==9.1.0.70 # via jax-cuda12-plugin # via torch -nvidia-cufft-cu12==11.0.2.54 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cufft-cu12==11.0.2.54 # via jax-cuda12-plugin # via torch -nvidia-curand-cu12==10.3.2.106 ; platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +nvidia-curand-cu12==10.3.2.106 # via torch -nvidia-cusolver-cu12==11.4.5.107 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cusolver-cu12==11.4.5.107 # via jax-cuda12-plugin # via torch -nvidia-cusparse-cu12==12.1.0.106 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))) +nvidia-cusparse-cu12==12.1.0.106 # via jax-cuda12-plugin # via nvidia-cusolver-cu12 # via torch -nvidia-nccl-cu12==2.20.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-nccl-cu12==2.20.5 # via jax-cuda12-plugin # via torch -nvidia-nvjitlink-cu12==12.6.20 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))))) +nvidia-nvjitlink-cu12==12.6.20 # via jax-cuda12-plugin # via nvidia-cusolver-cu12 # via nvidia-cusparse-cu12 -nvidia-nvtx-cu12==12.1.105 ; platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +nvidia-nvtx-cu12==12.1.105 # via torch -omegaconf==2.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +omegaconf==2.3.0 # via hydra-core # via hydra-zen -opt-einsum==3.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +opt-einsum==3.3.0 # via jax -optax==0.2.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +optax==0.2.3 # via flax -orbax-checkpoint==0.5.23 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +orbax-checkpoint==0.5.23 # via flax -packaging==24.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +packaging==24.1 # via black # via hydra-core + # via ipykernel + # via jupytext # via lightning # via lightning-utilities # via matplotlib # via mkdocs + # via nbconvert # via pytorch-lightning # via torchmetrics -paginate==0.5.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +paginate==0.5.6 # via mkdocs-material -pathspec==0.12.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pandocfilters==1.5.1 + # via nbconvert +paramiko==3.4.1 + # via fabric +parso==0.8.4 + # via jedi +pathspec==0.12.1 # via black # via mkdocs -pillow==10.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pexpect==4.9.0 + # via ipython +pillow==10.4.0 # via matplotlib # via torchvision -platformdirs==4.2.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +platformdirs==4.2.2 # via black + # via jupyter-core # via mkdocs-get-deps # via mkdocstrings # via wandb -protobuf==5.27.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +prompt-toolkit==3.0.47 + # via ipython + # via questionary +protobuf==5.27.3 # via orbax-checkpoint # via wandb -psutil==6.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +psutil==6.0.0 + # via ipykernel # via wandb -pydantic==2.8.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +ptyprocess==0.7.0 + # via pexpect +pure-eval==0.2.3 + # via stack-data +pycparser==2.22 + # via cffi +pydantic==2.8.2 # via research-project-template -pydantic-core==2.20.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pydantic-core==2.20.1 # via pydantic -pygments==2.18.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pygments==2.18.0 + # via ipython + # via mkdocs-jupyter # via mkdocs-material + # via nbconvert # via rich -pymdown-extensions==10.9 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pymdown-extensions==10.9 # via mkdocs-material # via mkdocstrings -pyparsing==3.1.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pynacl==1.5.0 + # via paramiko +pyparsing==3.1.2 # via matplotlib -pysocks==1.7.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pysocks==1.7.1 # via requests -python-dateutil==2.9.0.post0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +python-dateutil==2.9.0.post0 # via ghp-import + # via jupyter-client # via matplotlib -pytorch-lightning==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +python-dotenv==1.0.1 + # via rootutils +pytorch-lightning==2.4.0 # via lightning -pytorch2jax==0.1.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pytorch2jax==0.1.0 # via torch-jax-interop -pyyaml==6.0.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pyyaml==6.0.2 # via flax + # via jupytext # via lightning # via mkdocs # via mkdocs-get-deps @@ -341,49 +460,69 @@ pyyaml==6.0.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python # via pytorch-lightning # via pyyaml-env-tag # via wandb -pyyaml-env-tag==0.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pyyaml-env-tag==0.1 # via mkdocs -regex==2024.7.24 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pyzmq==26.2.0 + # via ipykernel + # via jupyter-client +questionary==1.10.0 + # via milatools +referencing==0.35.1 + # via jsonschema + # via jsonschema-specifications +regex==2024.7.24 # via mkdocs-material -requests==2.32.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +requests==2.32.3 # via gdown # via mkdocs-material # via wandb -rich==13.7.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +rich==13.7.1 # via flax + # via milatools # via research-project-template -scipy==1.14.0 ; (python_version >= '3.12' or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +rootutils==1.0.7 + # via research-project-template +rpds-py==0.20.0 + # via jsonschema + # via referencing +scipy==1.14.0 # via jax # via jaxlib -sentry-sdk==2.13.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') - # via wandb -setproctitle==1.3.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +sentry-sdk==2.13.0 # via wandb -setuptools==72.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.12' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') - # via chex - # via lightning-utilities +setproctitle==1.3.3 # via wandb -simple-parsing==0.1.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +simple-parsing==0.1.5 # via research-project-template -six==1.16.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +six==1.16.0 + # via asttokens + # via bleach + # via blessed # via docker-pycreds # via python-dateutil -smmap==5.0.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +smmap==5.0.1 # via gitdb -soupsieve==2.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +soupsieve==2.6 # via beautifulsoup4 -submitit==1.5.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +sshconf==0.2.7 + # via milatools +stack-data==0.6.3 + # via ipython +submitit==1.5.1 # via hydra-submitit-launcher -sympy==1.13.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +sympy==1.13.2 # via torch -tensorstore==0.1.64 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +tensorstore==0.1.64 # via flax # via orbax-checkpoint -tomli==2.0.1 ; python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +tinycss2==1.3.0 + # via nbconvert +tomli==2.0.1 # via black -toolz==0.12.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via jupytext +toolz==0.12.1 # via chex -torch==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torch==2.4.0 # via lightning # via pytorch-lightning # via pytorch2jax @@ -391,28 +530,44 @@ torch==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_ # via torch-jax-interop # via torchmetrics # via torchvision -torch-jax-interop==0.0.7 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torch-jax-interop==0.0.7 # via research-project-template -torchmetrics==1.4.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torchmetrics==1.4.1 # via lightning # via pytorch-lightning -torchvision==0.19.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torchvision==0.19.0 # via research-project-template -tqdm==4.66.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +tornado==6.4.1 + # via ipykernel + # via jupyter-client +tqdm==4.66.5 # via gdown # via lightning + # via milatools # via pytorch-lightning # via research-project-template -triton==3.0.0 ; python_version < '3.13' and platform_machine == 'x86_64' and platform_system == 'Linux' +traitlets==5.14.3 + # via comm + # via ipykernel + # via ipython + # via jupyter-client + # via jupyter-core + # via matplotlib-inline + # via nbclient + # via nbconvert + # via nbformat +triton==3.0.0 # via torch -typing-extensions==4.12.2 ; (python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +typing-extensions==4.12.2 # via black # via chex # via etils # via flax # via hydra-zen + # via ipython # via lightning # via lightning-utilities + # via milatools # via orbax-checkpoint # via pydantic # via pydantic-core @@ -420,16 +575,27 @@ typing-extensions==4.12.2 ; (python_version < '3.11' and ((python_version < '3.1 # via simple-parsing # via submitit # via torch -urllib3==2.2.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +urllib3==2.2.2 # via requests # via sentry-sdk -wandb==0.17.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +wandb==0.17.6 # via research-project-template -watchdog==4.0.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +watchdog==4.0.2 # via mkdocs -wcmatch==9.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +wcmatch==9.0 # via mkdocs-awesome-pages-plugin -yarl==1.9.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +wcwidth==0.2.13 + # via blessed + # via prompt-toolkit +webencodings==0.5.1 + # via bleach + # via tinycss2 +wrapt==1.16.0 + # via deprecated +yarl==1.9.4 # via aiohttp -zipp==3.20.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +zipp==3.20.0 # via etils +setuptools==72.2.0 + # via lightning-utilities + # via wandb From 24c84067e9a6209c46947acb32c2dacaefb3fd09 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Thu, 8 Aug 2024 11:08:43 -0400 Subject: [PATCH 05/33] add profiling notebook, hotfix a few classes --- project/experiment.py | 4 ---- 1 file changed, 4 deletions(-) diff --git a/project/experiment.py b/project/experiment.py index e5222279..683eb397 100644 --- a/project/experiment.py +++ b/project/experiment.py @@ -16,11 +16,7 @@ import logging import os import random -<<<<<<< HEAD from dataclasses import dataclass -======= -from dataclasses import dataclass, is_dataclass ->>>>>>> b7aec3e (add profiling notebook, hotfix a few classes) from logging import getLogger as get_logger from typing import Any From 56b84b77dc314ab28a7309082dd1a72482179e58 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Fri, 30 Aug 2024 18:19:34 -0400 Subject: [PATCH 06/33] removed pyrootutils, fixed typos, nbstripout check --- docs/examples/profiling.ipynb | 5729 +-------------------------------- 1 file changed, 25 insertions(+), 5704 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 2ac62445..9e6ae30a 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -18,38 +18,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "The Mila Research Template leverages built-in PyTorch and Lightning functionality to make benchmarking accesible and flexible. " - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "/home/mila/c/cesar.valdez/idt/ResearchTemplate\n", - "/home/mila/c/cesar.valdez/idt/ResearchTemplate\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/IPython/core/magics/osm.py:417: UserWarning: This is now an optional IPython functionality, setting dhist requires you to install the `pickleshare` library.\n", - " self.shell.db['dhist'] = compress_dhist(dhist)[-100:]\n" - ] - } - ], - "source": [ - "import os\n", - "import rootutils\n", - "\n", - "home_dir = rootutils.find_root(search_from=\"profiling.ipynb\", indicator=\".git\")\n", - "%cd $home_dir\n", - "print(os.getcwd())" + "The Mila Research Template leverages built-in PyTorch and Lightning functionality to make benchmarking accessible and flexible. " ] }, { @@ -68,5661 +37,30 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2mCONFIG\u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2malgorithm\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mhydra_zen.funcs.zen_processing \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_zen_target\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.no_op.NoOp \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_zen_partial\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_zen_wrappers\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mhydra_zen.third_party.pydantic.pydantic_parser \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mnetwork\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtorchvision.models.resnet.resnet18 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mweights\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mprogress\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_classes\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${instance_attr:datamodule.num_classes} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mdatamodule\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.datamodules.ImageNetDataModule \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdata_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:DATA_DIR} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_workers\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:NUM_WORKERS} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mval_split\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnormalize\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mshuffle\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mpin_memory\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mseed\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m42 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mbatch_size\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m64 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mtrainer\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.Trainer \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlogger\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40maccelerator\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mstrategy\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdevices\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmin_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmax_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdefault_root_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdetect_anomaly\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mcallbacks\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_checkpoint\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.ModelCheckpoint \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdirpath\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir}/checkpoints \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mfilename\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mepoch_{epoch:03d} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_last\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_top_k\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mauto_insert_metric_name\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_weights_only\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_train_steps\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mtrain_time_interval\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mearly_stopping\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.EarlyStopping \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmin_delta\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.0 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mpatience\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m5 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstrict\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_finite\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstopping_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdivergence_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_summary\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichModelSummary \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmax_depth\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m2 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mrich_progress_bar\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichProgressBar \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mthroughput\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.callbacks.samples_per_second.MeasureSam\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_train_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_val_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mlog_level\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40minfo \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mseed\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40m87282 \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mname\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mNoOp-resnet18-imagenet \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mdebug\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", - "\u001b[2m└── \u001b[0m\u001b[2mverbose\u001b[0m\n", - "\u001b[2m \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", - "\u001b[2;36m[08/07/24 16:28:58]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=131611;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=866100;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m,\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 87282\n", - "Seed set to 87282\n", - "/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=NoOp trainer.max_epochs=1 ...\n", - "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", - "GPU available: True (cuda), used: True\n", - "TPU available: False, using: 0 TPU cores\n", - "HPU available: False, using: 0 HPUs\n", - "\u001b[2;36m[16:28:58]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=867639;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=12083;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#163\u001b\\\u001b[2m163\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7fb2a6f64f80\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", - "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "Extracting train archive: 100%|████| 1000/1000 [10:10<00:00, 1.64Directories/s]\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", - "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", - "┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━┩\n", - "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNorm2d │ 128 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequential │ 147 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequential │ 525 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequential │ 2.1 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequential │ 8.4 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ AdaptiveAvgPool2d │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\n", - "└────┴─────────────────┴───────────────────┴────────┴───────┘\n", - "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", - "\u001b[1mNon-trainable params\u001b[0m: 0 \n", - "\u001b[1mTotal params\u001b[0m: 11.7 M \n", - "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/180\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0mt/s\u001b[0m \n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m100.703 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/180\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m100.703 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/180\u001b[0m \u001b[37m0:00:02 • 0:01:48\u001b[0m \u001b[37m1.64it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:35\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:35\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:31\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:31\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:26\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:26\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:24\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:22\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:22\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:05 • 0:01:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:05 • 0:01:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:06 • 0:01:17\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:07 • 0:01:19\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:07 • 0:01:19\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:07 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:07 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:08 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:08 • 0:01:18\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:08 • 0:01:19\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:08 • 0:01:19\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:09 • 0:01:17\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:09 • 0:01:17\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:09 • 0:01:16\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:09 • 0:01:16\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:10 • 0:01:15\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:11 • 0:01:14\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:11 • 0:01:13\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:11 • 0:01:13\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:12 • 0:01:12\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:13 • 0:01:11\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:13 • 0:01:11\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:13 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:13 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:14 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:14 • 0:01:10\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:14 • 0:01:09\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:14 • 0:01:09\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:14 • 0:01:08\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:14 • 0:01:08\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:15 • 0:01:08\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:16 • 0:01:07\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:16 • 0:01:07\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:16 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:16 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:17 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:17 • 0:01:06\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:17 • 0:01:05\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:17 • 0:01:05\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:17 • 0:01:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:17 • 0:01:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:18 • 0:01:04\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:19 • 0:01:03\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:20 • 0:01:02\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:21 • 0:01:01\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:22 • 0:01:00\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:23 • 0:00:59\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:24 • 0:00:58\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:25 • 0:00:57\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:25 • 0:00:56\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:25 • 0:00:56\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:26 • 0:00:56\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:26 • 0:00:56\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:26 • 0:00:55\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:26 • 0:00:55\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:27 • 0:00:54\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:27 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:27 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:28 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:28 • 0:00:53\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:28 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:28 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:29 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:29 • 0:00:52\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:29 • 0:00:51\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:30 • 0:00:50\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.27it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:31 • 0:00:49\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:32 • 0:00:48\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:32 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:32 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:33 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:33 • 0:00:47\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:33 • 0:00:46\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:33 • 0:00:46\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:34 • 0:00:46\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:34 • 0:00:46\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:34 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:34 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:35 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:35 • 0:00:45\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:35 • 0:00:44\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:36 • 0:00:43\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:37 • 0:00:42\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:37 • 0:00:41\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:37 • 0:00:41\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:38 • 0:00:41\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:38 • 0:00:41\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:38 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:38 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:39 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:39 • 0:00:40\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:39 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:39 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:40 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:40 • 0:00:39\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:40 • 0:00:38\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:41 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:42 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:42 • 0:00:37\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:42 • 0:00:36\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:43 • 0:00:35\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:44 • 0:00:34\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:44 • 0:00:33\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:44 • 0:00:33\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:45 • 0:00:33\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:45 • 0:00:33\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:46 • 0:00:33\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:46 • 0:00:33\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:46 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:47 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:47 • 0:00:32\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:47 • 0:00:31\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:47 • 0:00:31\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:48 • 0:00:31\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:48 • 0:00:31\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:48 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:48 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:49 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:49 • 0:00:30\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:49 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:49 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:50 • 0:00:29\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:50 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:50 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:51 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:51 • 0:00:28\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:51 • 0:00:27\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:51 • 0:00:27\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:52 • 0:00:27\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:52 • 0:00:27\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:52 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:52 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:53 • 0:00:26\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:54 • 0:00:25\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:54 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:54 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:55 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:55 • 0:00:24\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:55 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:55 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:00:56 • 0:00:23\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:00:56 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:00:56 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:00:57 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:00:57 • 0:00:22\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:00:57 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:00:57 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:00:58 • 0:00:21\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:00:58 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:00:58 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:00:59 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:00:59 • 0:00:20\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:00:59 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:00:59 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:00 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:00 • 0:00:19\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:00 • 0:00:18\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:01 • 0:00:17\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:02 • 0:00:16\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:03 • 0:00:15\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:04 • 0:00:14\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:05 • 0:00:14\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:05 • 0:00:14\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:05 • 0:00:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:06 • 0:00:12\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:07 • 0:00:12\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:07 • 0:00:12\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:07 • 0:00:11\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:07 • 0:00:11\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:08 • 0:00:11\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:08 • 0:00:11\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:08 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:08 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:09 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:10 • 0:00:09\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:10 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:10 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:11 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:11 • 0:00:08\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:11 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:11 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:12 • 0:00:07\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:12 • 0:00:06\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:12 • 0:00:06\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:13 • 0:00:06\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:13 • 0:00:06\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:13 • 0:00:05\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:13 • 0:00:05\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:14 • 0:00:05\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:14 • 0:00:05\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:14 • 0:00:04\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:14 • 0:00:04\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:15 • 0:00:04\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:16 • 0:00:03\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:16 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:16 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:17 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:17 • 0:00:02\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:17 • 0:00:01\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:17 • 0:00:01\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:18 • 0:00:01\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:18 • 0:00:01\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m149.625 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:18 • 0:00:00\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m151.611 \u001b[0m\n", - " \u001b[37mval/samples_per_…\u001b[0m\n", - " \u001b[37m120.811 \u001b[0m\n", - " \u001b[37mval/samples_per_…\u001b[0m\n", - " \u001b[37m128.323 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m150.129 \u001b[0m\n", - "\u001b[?25h/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=NoOp trainer.max_epochs=1 ...\n", - "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "\u001b[2;36m[16:41:35]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=972670;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=502992;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#381\u001b\\\u001b[2m381\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m[16:41:36]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=825781;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=661210;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37m2.10it/s\u001b[0m \n", - "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", - "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", - "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4964688718318939 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 140.0721893310547 \u001b[0m\u001b[35m \u001b[0m│\n", - "└──────────────────────────────┴──────────────────────────────┘\n", - "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.11it/s\u001b[0m \n", - "\u001b[?25hval val/samples_per_second_epoch: \u001b[1;36m140.0721893310547\u001b[0m\n" - ] - } - ], + "outputs": [], "source": [ - "!python project/main.py \\\n", - " algorithm=NoOp \\\n", - " trainer.max_epochs=1 \\\n", - " +trainer.limit_train_batches=0.01\\\n", - " +trainer.limit_val_batches=0.01\\\n", - " datamodule=imagenet" + "#!python project/main.py \\\n", + "# algorithm=NoOp \\\n", + "# trainer.max_epochs=1 \\\n", + "# +trainer.limit_train_batches=0.01\\\n", + "# +trainer.limit_val_batches=0.01\\\n", + "# datamodule=imagenet" ] }, { "cell_type": "code", - "execution_count": 10, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2mCONFIG\u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2malgorithm\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.example.ExampleAlgorithm \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40m_partial_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mnetwork\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtorchvision.models.resnet.resnet18 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mweights\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mprogress\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_classes\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${instance_attr:datamodule.num_classes} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mdatamodule\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.datamodules.ImageNetDataModule \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdata_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:DATA_DIR} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnum_workers\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${constant:NUM_WORKERS} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mval_split\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mnormalize\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mshuffle\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mpin_memory\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mseed\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m42 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mbatch_size\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m64 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mtrainer\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.Trainer \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlogger\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40maccelerator\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mstrategy\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mauto \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdevices\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmin_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mmax_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdefault_root_dir\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mdetect_anomaly\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mcallbacks\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_checkpoint\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.ModelCheckpoint \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdirpath\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m${hydra:runtime.output_dir}/checkpoints \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mfilename\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mepoch_{epoch:03d} \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_last\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_top_k\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m1 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mauto_insert_metric_name\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_weights_only\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_train_steps\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mtrain_time_interval\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mevery_n_epochs\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40msave_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mearly_stopping\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.EarlyStopping \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmonitor\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mval/loss \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmin_delta\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.0 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mpatience\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m5 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mverbose\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mfalse \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmode\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mmin \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstrict\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_finite\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mtrue \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mstopping_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mdivergence_threshold\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mcheck_on_train_epoch_end\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mnull \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmodel_summary\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichModelSummary \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mmax_depth\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m2 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mrich_progress_bar\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mlightning.pytorch.callbacks.RichProgressBar \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40mthroughput\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;91;40m_target_\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40mproject.algorithms.callbacks.samples_per_second.MeasureSam\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_train_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;91;40mlimit_val_batches\u001b[0m\u001b[2;97;40m:\u001b[0m\u001b[2;97;40m \u001b[0m\u001b[2;40m0.01 \u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m \u001b[0m\u001b[2;40m \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mlog_level\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40minfo \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mseed\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40m81570 \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mname\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mexample-resnet18-imagenet \u001b[0m\n", - "\u001b[2m├── \u001b[0m\u001b[2mdebug\u001b[0m\n", - "\u001b[2m│ \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", - "\u001b[2m└── \u001b[0m\u001b[2mverbose\u001b[0m\n", - "\u001b[2m \u001b[0m\u001b[2m└── \u001b[0m\u001b[2;40mFalse \u001b[0m\n", - "\u001b[2;36m[08/07/24 17:07:38]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=50545;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=71944;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m,\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 81570\n", - "Seed set to 81570\n", - "/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=example trainer.max_epochs ...\n", - "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", - "GPU available: True (cuda), used: True\n", - "TPU available: False, using: 0 TPU cores\n", - "HPU available: False, using: 0 HPUs\n", - "\u001b[2;36m[17:07:38]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=285735;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=912000;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#156\u001b\\\u001b[2m156\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7eff7e751f10\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m[17:07:43]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=702216;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=293228;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m train_accuracy. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=476337;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=48315;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m val_accuracy. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=513180;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=101826;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m test_accuracy. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=370748;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=635161;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m train_top5_accuracy. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=727524;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=265133;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m val_top5_accuracy. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Setting a new metric on the pl \u001b]8;id=753886;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=808003;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#75\u001b\\\u001b[2m75\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module at attribute \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m test_top5_accuracy. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=781296;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=524717;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#381\u001b\\\u001b[2m381\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m[17:07:44]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=248066;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=550199;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m[17:08:02]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Not adding metrics to the pl \u001b]8;id=656135;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=914060;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#64\u001b\\\u001b[2m64\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module because they are already \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m present. \u001b[2m \u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "┏━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━┓\n", - "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mIn sizes\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mOut sizes\u001b[0m\u001b[1;35m \u001b[0m┃\n", - "┡━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━┩\n", - "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\u001b[37m \u001b[0m\u001b[37m [64, 3,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1000]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\u001b[37m \u001b[0m\u001b[37m [64, 3,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNor… │ 128 │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequenti… │ 147 K │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequenti… │ 525 K │ train │\u001b[37m \u001b[0m\u001b[37m[64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 128,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 28, 28]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequenti… │ 2.1 M │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 256,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m128, 28,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 14, 14]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 28]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequenti… │ 8.4 M │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 512,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m256, 14,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 7, 7]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 14]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ Adaptive… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 512,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 512, 7,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1, 1]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 7]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 512]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1000]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m11\u001b[0m\u001b[2m \u001b[0m│ train_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m12\u001b[0m\u001b[2m \u001b[0m│ val_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m13\u001b[0m\u001b[2m \u001b[0m│ test_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m14\u001b[0m\u001b[2m \u001b[0m│ train_top5_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m15\u001b[0m\u001b[2m \u001b[0m│ val_top5_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m16\u001b[0m\u001b[2m \u001b[0m│ test_top5_accuracy │ Multicla… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m ?\u001b[0m\u001b[37m \u001b[0m│\n", - "└────┴─────────────────────┴───────────┴────────┴───────┴──────────┴───────────┘\n", - "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", - "\u001b[1mNon-trainable params\u001b[0m: 0 \n", - "\u001b[1mTotal params\u001b[0m: 11.7 M \n", - "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/180\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0mt/s\u001b[0m \n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m51.620 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/180\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m51.620 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/180\u001b[0m \u001b[37m0:00:01 • 0:00:21\u001b[0m \u001b[37m8.90it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.059 train/loss: \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/180\u001b[0m \u001b[37m0:00:02 • 0:01:01\u001b[0m \u001b[37m2.92it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.059 train/loss: \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/180\u001b[0m \u001b[37m0:00:02 • 0:01:01\u001b[0m \u001b[37m2.92it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.015 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m6.940 train/loss: \u001b[0m\n", - " \u001b[37m6.940 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:04\u001b[0m \u001b[37m2.76it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.015 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m6.940 train/loss: \u001b[0m\n", - " \u001b[37m6.940 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/180\u001b[0m \u001b[37m0:00:02 • 0:01:04\u001b[0m \u001b[37m2.76it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.146 train/loss: \u001b[0m\n", - " \u001b[37m7.146 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:08\u001b[0m \u001b[37m2.58it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.146 train/loss: \u001b[0m\n", - " \u001b[37m7.146 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/180\u001b[0m \u001b[37m0:00:03 • 0:01:08\u001b[0m \u001b[37m2.58it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.059 train/loss: \u001b[0m\n", - " \u001b[37m7.059 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.44it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.059 train/loss: \u001b[0m\n", - " \u001b[37m7.059 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.44it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.076 train/loss: \u001b[0m\n", - " \u001b[37m7.076 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.43it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.076 train/loss: \u001b[0m\n", - " \u001b[37m7.076 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m7/180\u001b[0m \u001b[37m0:00:03 • 0:01:12\u001b[0m \u001b[37m2.43it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.096 train/loss: \u001b[0m\n", - " \u001b[37m7.096 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.096 train/loss: \u001b[0m\n", - " \u001b[37m7.096 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m8/180\u001b[0m \u001b[37m0:00:04 • 0:01:13\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.142 train/loss: \u001b[0m\n", - " \u001b[37m7.142 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.142 train/loss: \u001b[0m\n", - " \u001b[37m7.142 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m9/180\u001b[0m \u001b[37m0:00:05 • 0:01:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: 5156802.000\u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accura…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entro…\u001b[0m\n", - " \u001b[37m7.030 train/loss: \u001b[0m\n", - " \u001b[37m7.030 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:19\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.030 train/loss:\u001b[0m\n", - " \u001b[37m7.030 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m10/180\u001b[0m \u001b[37m0:00:05 • 0:01:19\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.100 train/loss:\u001b[0m\n", - " \u001b[37m7.100 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:06 • 0:01:31\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.100 train/loss:\u001b[0m\n", - " \u001b[37m7.100 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m11/180\u001b[0m \u001b[37m0:00:06 • 0:01:31\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.008 train/loss:\u001b[0m\n", - " \u001b[37m7.008 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:07 • 0:01:28\u001b[0m \u001b[37m1.91it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.008 train/loss:\u001b[0m\n", - " \u001b[37m7.008 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m12/180\u001b[0m \u001b[37m0:00:07 • 0:01:28\u001b[0m \u001b[37m1.91it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.302 train/loss:\u001b[0m\n", - " \u001b[37m7.302 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:07 • 0:01:26\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.302 train/loss:\u001b[0m\n", - " \u001b[37m7.302 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m13/180\u001b[0m \u001b[37m0:00:07 • 0:01:26\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.344 train/loss:\u001b[0m\n", - " \u001b[37m7.344 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:08 • 0:01:30\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.344 train/loss:\u001b[0m\n", - " \u001b[37m7.344 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m14/180\u001b[0m \u001b[37m0:00:08 • 0:01:30\u001b[0m \u001b[37m1.86it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.050 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.361 train/loss:\u001b[0m\n", - " \u001b[37m7.361 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:08 • 0:01:28\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.050 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.361 train/loss:\u001b[0m\n", - " \u001b[37m7.361 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m15/180\u001b[0m \u001b[37m0:00:08 • 0:01:28\u001b[0m \u001b[37m1.88it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.182 train/loss:\u001b[0m\n", - " \u001b[37m7.182 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:10 • 0:01:35\u001b[0m \u001b[37m1.73it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.182 train/loss:\u001b[0m\n", - " \u001b[37m7.182 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m16/180\u001b[0m \u001b[37m0:00:10 • 0:01:35\u001b[0m \u001b[37m1.73it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.168 train/loss:\u001b[0m\n", - " \u001b[37m7.168 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:10 • 0:01:37\u001b[0m \u001b[37m1.69it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.168 train/loss:\u001b[0m\n", - " \u001b[37m7.168 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m17/180\u001b[0m \u001b[37m0:00:10 • 0:01:37\u001b[0m \u001b[37m1.69it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.324 train/loss:\u001b[0m\n", - " \u001b[37m7.324 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:11 • 0:01:35\u001b[0m \u001b[37m1.72it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.324 train/loss:\u001b[0m\n", - " \u001b[37m7.324 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m18/180\u001b[0m \u001b[37m0:00:11 • 0:01:35\u001b[0m \u001b[37m1.72it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.436 train/loss:\u001b[0m\n", - " \u001b[37m7.436 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:11 • 0:01:32\u001b[0m \u001b[37m1.75it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.436 train/loss:\u001b[0m\n", - " \u001b[37m7.436 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m19/180\u001b[0m \u001b[37m0:00:11 • 0:01:32\u001b[0m \u001b[37m1.75it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.111 train/loss:\u001b[0m\n", - " \u001b[37m7.111 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:12 • 0:01:31\u001b[0m \u001b[37m1.78it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.111 train/loss:\u001b[0m\n", - " \u001b[37m7.111 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/180\u001b[0m \u001b[37m0:00:12 • 0:01:31\u001b[0m \u001b[37m1.78it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.521 train/loss:\u001b[0m\n", - " \u001b[37m7.521 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:12 • 0:01:29\u001b[0m \u001b[37m1.79it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.521 train/loss:\u001b[0m\n", - " \u001b[37m7.521 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m21/180\u001b[0m \u001b[37m0:00:12 • 0:01:29\u001b[0m \u001b[37m1.79it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.065 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.163 train/loss:\u001b[0m\n", - " \u001b[37m7.163 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:13 • 0:01:28\u001b[0m \u001b[37m1.81it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.065 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.163 train/loss:\u001b[0m\n", - " \u001b[37m7.163 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m22/180\u001b[0m \u001b[37m0:00:13 • 0:01:28\u001b[0m \u001b[37m1.81it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.038 train/loss:\u001b[0m\n", - " \u001b[37m7.038 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:13 • 0:01:26\u001b[0m \u001b[37m1.83it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.038 train/loss:\u001b[0m\n", - " \u001b[37m7.038 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m23/180\u001b[0m \u001b[37m0:00:13 • 0:01:26\u001b[0m \u001b[37m1.83it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.171 train/loss:\u001b[0m\n", - " \u001b[37m7.171 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:13 • 0:01:25\u001b[0m \u001b[37m1.85it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.171 train/loss:\u001b[0m\n", - " \u001b[37m7.171 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m24/180\u001b[0m \u001b[37m0:00:13 • 0:01:25\u001b[0m \u001b[37m1.85it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.327 train/loss:\u001b[0m\n", - " \u001b[37m7.327 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:14 • 0:01:23\u001b[0m \u001b[37m1.87it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.327 train/loss:\u001b[0m\n", - " \u001b[37m7.327 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m25/180\u001b[0m \u001b[37m0:00:14 • 0:01:23\u001b[0m \u001b[37m1.87it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.207 train/loss:\u001b[0m\n", - " \u001b[37m7.207 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:14 • 0:01:22\u001b[0m \u001b[37m1.90it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.207 train/loss:\u001b[0m\n", - " \u001b[37m7.207 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m26/180\u001b[0m \u001b[37m0:00:14 • 0:01:22\u001b[0m \u001b[37m1.90it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.271 train/loss:\u001b[0m\n", - " \u001b[37m7.271 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:15 • 0:01:20\u001b[0m \u001b[37m1.92it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.271 train/loss:\u001b[0m\n", - " \u001b[37m7.271 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m27/180\u001b[0m \u001b[37m0:00:15 • 0:01:20\u001b[0m \u001b[37m1.92it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.566 train/loss:\u001b[0m\n", - " \u001b[37m7.566 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:15 • 0:01:19\u001b[0m \u001b[37m1.93it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.566 train/loss:\u001b[0m\n", - " \u001b[37m7.566 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m28/180\u001b[0m \u001b[37m0:00:15 • 0:01:19\u001b[0m \u001b[37m1.93it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.203 train/loss:\u001b[0m\n", - " \u001b[37m7.203 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:15 • 0:01:18\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.203 train/loss:\u001b[0m\n", - " \u001b[37m7.203 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m29/180\u001b[0m \u001b[37m0:00:15 • 0:01:18\u001b[0m \u001b[37m1.94it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.262 train/loss:\u001b[0m\n", - " \u001b[37m7.262 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:16 • 0:01:17\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.262 train/loss:\u001b[0m\n", - " \u001b[37m7.262 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/180\u001b[0m \u001b[37m0:00:16 • 0:01:17\u001b[0m \u001b[37m1.95it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.109 train/loss:\u001b[0m\n", - " \u001b[37m7.109 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:16 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.109 train/loss:\u001b[0m\n", - " \u001b[37m7.109 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m31/180\u001b[0m \u001b[37m0:00:16 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.343 train/loss:\u001b[0m\n", - " \u001b[37m7.343 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:17 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.343 train/loss:\u001b[0m\n", - " \u001b[37m7.343 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m32/180\u001b[0m \u001b[37m0:00:17 • 0:01:16\u001b[0m \u001b[37m1.97it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.190 train/loss:\u001b[0m\n", - " \u001b[37m7.190 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:17 • 0:01:15\u001b[0m \u001b[37m1.98it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.190 train/loss:\u001b[0m\n", - " \u001b[37m7.190 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m33/180\u001b[0m \u001b[37m0:00:17 • 0:01:15\u001b[0m \u001b[37m1.98it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.051 train/loss:\u001b[0m\n", - " \u001b[37m7.051 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:18 • 0:01:14\u001b[0m \u001b[37m1.99it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.051 train/loss:\u001b[0m\n", - " \u001b[37m7.051 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m34/180\u001b[0m \u001b[37m0:00:18 • 0:01:14\u001b[0m \u001b[37m1.99it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.174 train/loss:\u001b[0m\n", - " \u001b[37m7.174 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:18 • 0:01:13\u001b[0m \u001b[37m2.00it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.174 train/loss:\u001b[0m\n", - " \u001b[37m7.174 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m35/180\u001b[0m \u001b[37m0:00:18 • 0:01:13\u001b[0m \u001b[37m2.00it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.390 train/loss:\u001b[0m\n", - " \u001b[37m7.390 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:18 • 0:01:12\u001b[0m \u001b[37m2.01it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.390 train/loss:\u001b[0m\n", - " \u001b[37m7.390 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m36/180\u001b[0m \u001b[37m0:00:18 • 0:01:12\u001b[0m \u001b[37m2.01it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.300 train/loss:\u001b[0m\n", - " \u001b[37m7.300 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:19 • 0:01:11\u001b[0m \u001b[37m2.02it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.300 train/loss:\u001b[0m\n", - " \u001b[37m7.300 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m37/180\u001b[0m \u001b[37m0:00:19 • 0:01:11\u001b[0m \u001b[37m2.02it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.237 train/loss:\u001b[0m\n", - " \u001b[37m7.237 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:19 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.237 train/loss:\u001b[0m\n", - " \u001b[37m7.237 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m38/180\u001b[0m \u001b[37m0:00:19 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.232 train/loss:\u001b[0m\n", - " \u001b[37m7.232 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:20 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.232 train/loss:\u001b[0m\n", - " \u001b[37m7.232 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m39/180\u001b[0m \u001b[37m0:00:20 • 0:01:10\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.971 train/loss:\u001b[0m\n", - " \u001b[37m6.971 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:20 • 0:01:09\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.971 train/loss:\u001b[0m\n", - " \u001b[37m6.971 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m40/180\u001b[0m \u001b[37m0:00:20 • 0:01:09\u001b[0m \u001b[37m2.03it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.262 train/loss:\u001b[0m\n", - " \u001b[37m7.262 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.262 train/loss:\u001b[0m\n", - " \u001b[37m7.262 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m41/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.142 train/loss:\u001b[0m\n", - " \u001b[37m7.142 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.142 train/loss:\u001b[0m\n", - " \u001b[37m7.142 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m42/180\u001b[0m \u001b[37m0:00:21 • 0:01:08\u001b[0m \u001b[37m2.04it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.984 train/loss:\u001b[0m\n", - " \u001b[37m6.984 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.984 train/loss:\u001b[0m\n", - " \u001b[37m6.984 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m43/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.149 train/loss:\u001b[0m\n", - " \u001b[37m7.149 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.149 train/loss:\u001b[0m\n", - " \u001b[37m7.149 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m44/180\u001b[0m \u001b[37m0:00:22 • 0:01:07\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.986 train/loss:\u001b[0m\n", - " \u001b[37m6.986 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:22 • 0:01:06\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.986 train/loss:\u001b[0m\n", - " \u001b[37m6.986 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m45/180\u001b[0m \u001b[37m0:00:22 • 0:01:06\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.067 train/loss:\u001b[0m\n", - " \u001b[37m7.067 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:23 • 0:01:06\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.067 train/loss:\u001b[0m\n", - " \u001b[37m7.067 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m46/180\u001b[0m \u001b[37m0:00:23 • 0:01:06\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.224 train/loss:\u001b[0m\n", - " \u001b[37m7.224 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:23 • 0:01:05\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.224 train/loss:\u001b[0m\n", - " \u001b[37m7.224 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m47/180\u001b[0m \u001b[37m0:00:23 • 0:01:05\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.053 train/loss:\u001b[0m\n", - " \u001b[37m7.053 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:24 • 0:01:05\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.053 train/loss:\u001b[0m\n", - " \u001b[37m7.053 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m48/180\u001b[0m \u001b[37m0:00:24 • 0:01:05\u001b[0m \u001b[37m2.05it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.016 train/loss:\u001b[0m\n", - " \u001b[37m7.016 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:24 • 0:01:04\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.016 train/loss:\u001b[0m\n", - " \u001b[37m7.016 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m49/180\u001b[0m \u001b[37m0:00:24 • 0:01:04\u001b[0m \u001b[37m2.06it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.050 train/loss:\u001b[0m\n", - " \u001b[37m7.050 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.050 train/loss:\u001b[0m\n", - " \u001b[37m7.050 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m50/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.07it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.025 train/loss:\u001b[0m\n", - " \u001b[37m7.025 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.025 train/loss:\u001b[0m\n", - " \u001b[37m7.025 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m51/180\u001b[0m \u001b[37m0:00:25 • 0:01:03\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.886 train/loss:\u001b[0m\n", - " \u001b[37m6.886 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:25 • 0:01:02\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.886 train/loss:\u001b[0m\n", - " \u001b[37m6.886 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m52/180\u001b[0m \u001b[37m0:00:25 • 0:01:02\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.926 train/loss:\u001b[0m\n", - " \u001b[37m6.926 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.926 train/loss:\u001b[0m\n", - " \u001b[37m6.926 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m53/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.090 train/loss:\u001b[0m\n", - " \u001b[37m7.090 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.090 train/loss:\u001b[0m\n", - " \u001b[37m7.090 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m54/180\u001b[0m \u001b[37m0:00:26 • 0:01:01\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.913 train/loss:\u001b[0m\n", - " \u001b[37m6.913 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.913 train/loss:\u001b[0m\n", - " \u001b[37m6.913 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m55/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.100 train/loss:\u001b[0m\n", - " \u001b[37m7.100 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.100 train/loss:\u001b[0m\n", - " \u001b[37m7.100 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m56/180\u001b[0m \u001b[37m0:00:27 • 0:01:00\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.028 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.985 train/loss:\u001b[0m\n", - " \u001b[37m6.985 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:28 • 0:00:59\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.028 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.985 train/loss:\u001b[0m\n", - " \u001b[37m6.985 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m57/180\u001b[0m \u001b[37m0:00:28 • 0:00:59\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.191 train/loss:\u001b[0m\n", - " \u001b[37m7.191 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.191 train/loss:\u001b[0m\n", - " \u001b[37m7.191 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m58/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.939 train/loss:\u001b[0m\n", - " \u001b[37m6.939 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.939 train/loss:\u001b[0m\n", - " \u001b[37m6.939 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m59/180\u001b[0m \u001b[37m0:00:28 • 0:00:58\u001b[0m \u001b[37m2.11it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.859 train/loss:\u001b[0m\n", - " \u001b[37m6.859 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.859 train/loss:\u001b[0m\n", - " \u001b[37m6.859 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m60/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.076 train/loss:\u001b[0m\n", - " \u001b[37m7.076 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.076 train/loss:\u001b[0m\n", - " \u001b[37m7.076 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m61/180\u001b[0m \u001b[37m0:00:29 • 0:00:57\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.115 train/loss:\u001b[0m\n", - " \u001b[37m7.115 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:30 • 0:00:56\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.115 train/loss:\u001b[0m\n", - " \u001b[37m7.115 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m62/180\u001b[0m \u001b[37m0:00:30 • 0:00:56\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.936 train/loss:\u001b[0m\n", - " \u001b[37m6.936 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.936 train/loss:\u001b[0m\n", - " \u001b[37m6.936 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m63/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.901 train/loss:\u001b[0m\n", - " \u001b[37m6.901 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.901 train/loss:\u001b[0m\n", - " \u001b[37m6.901 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m64/180\u001b[0m \u001b[37m0:00:30 • 0:00:55\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.068 train/loss:\u001b[0m\n", - " \u001b[37m7.068 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.068 train/loss:\u001b[0m\n", - " \u001b[37m7.068 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m65/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.932 train/loss:\u001b[0m\n", - " \u001b[37m6.932 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.932 train/loss:\u001b[0m\n", - " \u001b[37m6.932 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m66/180\u001b[0m \u001b[37m0:00:31 • 0:00:54\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.934 train/loss:\u001b[0m\n", - " \u001b[37m6.934 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.934 train/loss:\u001b[0m\n", - " \u001b[37m6.934 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m67/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.022 train/loss:\u001b[0m\n", - " \u001b[37m7.022 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.022 train/loss:\u001b[0m\n", - " \u001b[37m7.022 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m68/180\u001b[0m \u001b[37m0:00:32 • 0:00:53\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.905 train/loss:\u001b[0m\n", - " \u001b[37m6.905 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.905 train/loss:\u001b[0m\n", - " \u001b[37m6.905 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m69/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.929 train/loss:\u001b[0m\n", - " \u001b[37m6.929 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.929 train/loss:\u001b[0m\n", - " \u001b[37m6.929 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m70/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.060 train/loss:\u001b[0m\n", - " \u001b[37m7.060 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.060 train/loss:\u001b[0m\n", - " \u001b[37m7.060 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m71/180\u001b[0m \u001b[37m0:00:33 • 0:00:52\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.089 train/loss:\u001b[0m\n", - " \u001b[37m7.089 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:34 • 0:00:51\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.089 train/loss:\u001b[0m\n", - " \u001b[37m7.089 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m72/180\u001b[0m \u001b[37m0:00:34 • 0:00:51\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.011 train/loss:\u001b[0m\n", - " \u001b[37m7.011 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:34 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.011 train/loss:\u001b[0m\n", - " \u001b[37m7.011 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m73/180\u001b[0m \u001b[37m0:00:34 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.953 train/loss:\u001b[0m\n", - " \u001b[37m6.953 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:35 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.953 train/loss:\u001b[0m\n", - " \u001b[37m6.953 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m74/180\u001b[0m \u001b[37m0:00:35 • 0:00:50\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.890 train/loss:\u001b[0m\n", - " \u001b[37m6.890 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:35 • 0:00:48\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.890 train/loss:\u001b[0m\n", - " \u001b[37m6.890 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m75/180\u001b[0m \u001b[37m0:00:35 • 0:00:48\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.829 train/loss:\u001b[0m\n", - " \u001b[37m6.829 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.829 train/loss:\u001b[0m\n", - " \u001b[37m6.829 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m76/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.009 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.902 train/loss:\u001b[0m\n", - " \u001b[37m6.902 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.009 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.902 train/loss:\u001b[0m\n", - " \u001b[37m6.902 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m77/180\u001b[0m \u001b[37m0:00:36 • 0:00:47\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.996 train/loss:\u001b[0m\n", - " \u001b[37m6.996 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:37 • 0:00:47\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.996 train/loss:\u001b[0m\n", - " \u001b[37m6.996 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m78/180\u001b[0m \u001b[37m0:00:37 • 0:00:47\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.074 train/loss:\u001b[0m\n", - " \u001b[37m7.074 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:37 • 0:00:46\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.074 train/loss:\u001b[0m\n", - " \u001b[37m7.074 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m79/180\u001b[0m \u001b[37m0:00:37 • 0:00:46\u001b[0m \u001b[37m2.21it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.826 train/loss:\u001b[0m\n", - " \u001b[37m6.826 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.826 train/loss:\u001b[0m\n", - " \u001b[37m6.826 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m80/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.075 train/loss:\u001b[0m\n", - " \u001b[37m7.075 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.075 train/loss:\u001b[0m\n", - " \u001b[37m7.075 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m81/180\u001b[0m \u001b[37m0:00:38 • 0:00:45\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.881 train/loss:\u001b[0m\n", - " \u001b[37m6.881 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.28it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.881 train/loss:\u001b[0m\n", - " \u001b[37m6.881 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m82/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.28it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.860 train/loss:\u001b[0m\n", - " \u001b[37m6.860 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.860 train/loss:\u001b[0m\n", - " \u001b[37m6.860 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m83/180\u001b[0m \u001b[37m0:00:39 • 0:00:43\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.005 train/loss:\u001b[0m\n", - " \u001b[37m7.005 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:39 • 0:00:42\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.005 train/loss:\u001b[0m\n", - " \u001b[37m7.005 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m84/180\u001b[0m \u001b[37m0:00:39 • 0:00:42\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.864 train/loss:\u001b[0m\n", - " \u001b[37m6.864 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:40 • 0:00:42\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.864 train/loss:\u001b[0m\n", - " \u001b[37m6.864 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m85/180\u001b[0m \u001b[37m0:00:40 • 0:00:42\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.953 train/loss:\u001b[0m\n", - " \u001b[37m6.953 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:40 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.953 train/loss:\u001b[0m\n", - " \u001b[37m6.953 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m86/180\u001b[0m \u001b[37m0:00:40 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.968 train/loss:\u001b[0m\n", - " \u001b[37m6.968 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:41 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.968 train/loss:\u001b[0m\n", - " \u001b[37m6.968 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m87/180\u001b[0m \u001b[37m0:00:41 • 0:00:41\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.973 train/loss:\u001b[0m\n", - " \u001b[37m6.973 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:41 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.973 train/loss:\u001b[0m\n", - " \u001b[37m6.973 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m88/180\u001b[0m \u001b[37m0:00:41 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.986 train/loss:\u001b[0m\n", - " \u001b[37m6.986 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:42 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.986 train/loss:\u001b[0m\n", - " \u001b[37m6.986 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m89/180\u001b[0m \u001b[37m0:00:42 • 0:00:40\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.007 train/loss:\u001b[0m\n", - " \u001b[37m7.007 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.007 train/loss:\u001b[0m\n", - " \u001b[37m7.007 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m90/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.029 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.940 train/loss:\u001b[0m\n", - " \u001b[37m6.940 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.029 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.940 train/loss:\u001b[0m\n", - " \u001b[37m6.940 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m91/180\u001b[0m \u001b[37m0:00:42 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.932 train/loss:\u001b[0m\n", - " \u001b[37m6.932 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:43 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.932 train/loss:\u001b[0m\n", - " \u001b[37m6.932 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m92/180\u001b[0m \u001b[37m0:00:43 • 0:00:39\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.997 train/loss:\u001b[0m\n", - " \u001b[37m6.997 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:43 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.997 train/loss:\u001b[0m\n", - " \u001b[37m6.997 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m93/180\u001b[0m \u001b[37m0:00:43 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.936 train/loss:\u001b[0m\n", - " \u001b[37m6.936 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:44 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.936 train/loss:\u001b[0m\n", - " \u001b[37m6.936 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m94/180\u001b[0m \u001b[37m0:00:44 • 0:00:38\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.014 train/loss:\u001b[0m\n", - " \u001b[37m7.014 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:44 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.014 train/loss:\u001b[0m\n", - " \u001b[37m7.014 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m95/180\u001b[0m \u001b[37m0:00:44 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.937 train/loss:\u001b[0m\n", - " \u001b[37m6.937 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.937 train/loss:\u001b[0m\n", - " \u001b[37m6.937 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m96/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.915 train/loss:\u001b[0m\n", - " \u001b[37m6.915 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.915 train/loss:\u001b[0m\n", - " \u001b[37m6.915 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m97/180\u001b[0m \u001b[37m0:00:45 • 0:00:37\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.923 train/loss:\u001b[0m\n", - " \u001b[37m6.923 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:45 • 0:00:36\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.923 train/loss:\u001b[0m\n", - " \u001b[37m6.923 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m98/180\u001b[0m \u001b[37m0:00:45 • 0:00:36\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.014 train/loss:\u001b[0m\n", - " \u001b[37m7.014 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:46 • 0:00:36\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.014 train/loss:\u001b[0m\n", - " \u001b[37m7.014 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m99/180\u001b[0m \u001b[37m0:00:46 • 0:00:36\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.963 train/loss:\u001b[0m\n", - " \u001b[37m6.963 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:46 • 0:00:35\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.963 train/loss:\u001b[0m\n", - " \u001b[37m6.963 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m100/180\u001b[0m \u001b[37m0:00:46 • 0:00:35\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.873 train/loss:\u001b[0m\n", - " \u001b[37m6.873 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:47 • 0:00:35\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.873 train/loss:\u001b[0m\n", - " \u001b[37m6.873 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m101/180\u001b[0m \u001b[37m0:00:47 • 0:00:35\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.033 train/loss:\u001b[0m\n", - " \u001b[37m7.033 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:47 • 0:00:34\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.033 train/loss:\u001b[0m\n", - " \u001b[37m7.033 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m102/180\u001b[0m \u001b[37m0:00:47 • 0:00:34\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.034 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.787 train/loss:\u001b[0m\n", - " \u001b[37m6.787 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:48 • 0:00:35\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.034 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.787 train/loss:\u001b[0m\n", - " \u001b[37m6.787 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m103/180\u001b[0m \u001b[37m0:00:48 • 0:00:35\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.946 train/loss:\u001b[0m\n", - " \u001b[37m6.946 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.946 train/loss:\u001b[0m\n", - " \u001b[37m6.946 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m104/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.966 train/loss:\u001b[0m\n", - " \u001b[37m6.966 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.966 train/loss:\u001b[0m\n", - " \u001b[37m6.966 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m105/180\u001b[0m \u001b[37m0:00:49 • 0:00:34\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.899 train/loss:\u001b[0m\n", - " \u001b[37m6.899 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:49 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.899 train/loss:\u001b[0m\n", - " \u001b[37m6.899 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m106/180\u001b[0m \u001b[37m0:00:49 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.031 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.886 train/loss:\u001b[0m\n", - " \u001b[37m6.886 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:50 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.031 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.886 train/loss:\u001b[0m\n", - " \u001b[37m6.886 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m107/180\u001b[0m \u001b[37m0:00:50 • 0:00:33\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.892 train/loss:\u001b[0m\n", - " \u001b[37m6.892 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:51 • 0:00:33\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.892 train/loss:\u001b[0m\n", - " \u001b[37m6.892 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m108/180\u001b[0m \u001b[37m0:00:51 • 0:00:33\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.880 train/loss:\u001b[0m\n", - " \u001b[37m6.880 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:51 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.880 train/loss:\u001b[0m\n", - " \u001b[37m6.880 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m109/180\u001b[0m \u001b[37m0:00:51 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.897 train/loss:\u001b[0m\n", - " \u001b[37m6.897 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.897 train/loss:\u001b[0m\n", - " \u001b[37m6.897 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m110/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.891 train/loss:\u001b[0m\n", - " \u001b[37m6.891 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.891 train/loss:\u001b[0m\n", - " \u001b[37m6.891 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m111/180\u001b[0m \u001b[37m0:00:52 • 0:00:32\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.979 train/loss:\u001b[0m\n", - " \u001b[37m6.979 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:53 • 0:00:31\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.979 train/loss:\u001b[0m\n", - " \u001b[37m6.979 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m112/180\u001b[0m \u001b[37m0:00:53 • 0:00:31\u001b[0m \u001b[37m2.22it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.860 train/loss:\u001b[0m\n", - " \u001b[37m6.860 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.860 train/loss:\u001b[0m\n", - " \u001b[37m6.860 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m113/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.014 train/loss:\u001b[0m\n", - " \u001b[37m7.014 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.014 train/loss:\u001b[0m\n", - " \u001b[37m7.014 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m114/180\u001b[0m \u001b[37m0:00:54 • 0:00:31\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.908 train/loss:\u001b[0m\n", - " \u001b[37m6.908 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.908 train/loss:\u001b[0m\n", - " \u001b[37m6.908 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m115/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.802 train/loss:\u001b[0m\n", - " \u001b[37m6.802 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.802 train/loss:\u001b[0m\n", - " \u001b[37m6.802 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m116/180\u001b[0m \u001b[37m0:00:55 • 0:00:30\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.839 train/loss:\u001b[0m\n", - " \u001b[37m6.839 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:56 • 0:00:30\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.839 train/loss:\u001b[0m\n", - " \u001b[37m6.839 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m117/180\u001b[0m \u001b[37m0:00:56 • 0:00:30\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.880 train/loss:\u001b[0m\n", - " \u001b[37m6.880 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.880 train/loss:\u001b[0m\n", - " \u001b[37m6.880 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m118/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.812 train/loss:\u001b[0m\n", - " \u001b[37m6.812 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.812 train/loss:\u001b[0m\n", - " \u001b[37m6.812 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m119/180\u001b[0m \u001b[37m0:00:56 • 0:00:29\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.848 train/loss:\u001b[0m\n", - " \u001b[37m6.848 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.848 train/loss:\u001b[0m\n", - " \u001b[37m6.848 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m120/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.946 train/loss:\u001b[0m\n", - " \u001b[37m6.946 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.946 train/loss:\u001b[0m\n", - " \u001b[37m6.946 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m121/180\u001b[0m \u001b[37m0:00:57 • 0:00:28\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.939 train/loss:\u001b[0m\n", - " \u001b[37m6.939 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.939 train/loss:\u001b[0m\n", - " \u001b[37m6.939 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m122/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.839 train/loss:\u001b[0m\n", - " \u001b[37m6.839 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.839 train/loss:\u001b[0m\n", - " \u001b[37m6.839 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m123/180\u001b[0m \u001b[37m0:00:58 • 0:00:27\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.827 train/loss:\u001b[0m\n", - " \u001b[37m6.827 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.827 train/loss:\u001b[0m\n", - " \u001b[37m6.827 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m124/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.862 train/loss:\u001b[0m\n", - " \u001b[37m6.862 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.862 train/loss:\u001b[0m\n", - " \u001b[37m6.862 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m125/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.034 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.838 train/loss:\u001b[0m\n", - " \u001b[37m6.838 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.034 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.838 train/loss:\u001b[0m\n", - " \u001b[37m6.838 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m126/180\u001b[0m \u001b[37m0:00:59 • 0:00:26\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.004 train/loss:\u001b[0m\n", - " \u001b[37m7.004 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.004 train/loss:\u001b[0m\n", - " \u001b[37m7.004 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m127/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.028 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.801 train/loss:\u001b[0m\n", - " \u001b[37m6.801 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.028 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.801 train/loss:\u001b[0m\n", - " \u001b[37m6.801 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m128/180\u001b[0m \u001b[37m0:01:00 • 0:00:25\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.042 train/loss:\u001b[0m\n", - " \u001b[37m7.042 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m7.042 train/loss:\u001b[0m\n", - " \u001b[37m7.042 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m129/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.851 train/loss:\u001b[0m\n", - " \u001b[37m6.851 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.851 train/loss:\u001b[0m\n", - " \u001b[37m6.851 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m130/180\u001b[0m \u001b[37m0:01:01 • 0:00:24\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.904 train/loss:\u001b[0m\n", - " \u001b[37m6.904 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:01:01 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.904 train/loss:\u001b[0m\n", - " \u001b[37m6.904 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m131/180\u001b[0m \u001b[37m0:01:02 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.850 train/loss:\u001b[0m\n", - " \u001b[37m6.850 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:01:02 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.850 train/loss:\u001b[0m\n", - " \u001b[37m6.850 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m132/180\u001b[0m \u001b[37m0:01:02 • 0:00:23\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.885 train/loss:\u001b[0m\n", - " \u001b[37m6.885 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:01:02 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.885 train/loss:\u001b[0m\n", - " \u001b[37m6.885 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m133/180\u001b[0m \u001b[37m0:01:02 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.815 train/loss:\u001b[0m\n", - " \u001b[37m6.815 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:01:03 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.815 train/loss:\u001b[0m\n", - " \u001b[37m6.815 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m134/180\u001b[0m \u001b[37m0:01:03 • 0:00:22\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.871 train/loss:\u001b[0m\n", - " \u001b[37m6.871 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:01:03 • 0:00:21\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.871 train/loss:\u001b[0m\n", - " \u001b[37m6.871 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m135/180\u001b[0m \u001b[37m0:01:03 • 0:00:21\u001b[0m \u001b[37m2.15it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.864 train/loss:\u001b[0m\n", - " \u001b[37m6.864 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:01:04 • 0:00:21\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.864 train/loss:\u001b[0m\n", - " \u001b[37m6.864 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m136/180\u001b[0m \u001b[37m0:01:04 • 0:00:21\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.936 train/loss:\u001b[0m\n", - " \u001b[37m6.936 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.936 train/loss:\u001b[0m\n", - " \u001b[37m6.936 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m137/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.031 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.878 train/loss:\u001b[0m\n", - " \u001b[37m6.878 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.031 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.878 train/loss:\u001b[0m\n", - " \u001b[37m6.878 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m138/180\u001b[0m \u001b[37m0:01:04 • 0:00:20\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.026 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.902 train/loss:\u001b[0m\n", - " \u001b[37m6.902 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.026 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.902 train/loss:\u001b[0m\n", - " \u001b[37m6.902 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m139/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.955 train/loss:\u001b[0m\n", - " \u001b[37m6.955 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.955 train/loss:\u001b[0m\n", - " \u001b[37m6.955 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m140/180\u001b[0m \u001b[37m0:01:05 • 0:00:19\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.852 train/loss:\u001b[0m\n", - " \u001b[37m6.852 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.852 train/loss:\u001b[0m\n", - " \u001b[37m6.852 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m141/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.869 train/loss:\u001b[0m\n", - " \u001b[37m6.869 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.048 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.869 train/loss:\u001b[0m\n", - " \u001b[37m6.869 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m142/180\u001b[0m \u001b[37m0:01:06 • 0:00:18\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.902 train/loss:\u001b[0m\n", - " \u001b[37m6.902 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.902 train/loss:\u001b[0m\n", - " \u001b[37m6.902 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m143/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.872 train/loss:\u001b[0m\n", - " \u001b[37m6.872 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.872 train/loss:\u001b[0m\n", - " \u001b[37m6.872 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m144/180\u001b[0m \u001b[37m0:01:07 • 0:00:17\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.959 train/loss:\u001b[0m\n", - " \u001b[37m6.959 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:07 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.959 train/loss:\u001b[0m\n", - " \u001b[37m6.959 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m145/180\u001b[0m \u001b[37m0:01:07 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.851 train/loss:\u001b[0m\n", - " \u001b[37m6.851 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.851 train/loss:\u001b[0m\n", - " \u001b[37m6.851 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m146/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.966 train/loss:\u001b[0m\n", - " \u001b[37m6.966 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.966 train/loss:\u001b[0m\n", - " \u001b[37m6.966 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m147/180\u001b[0m \u001b[37m0:01:08 • 0:00:16\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.838 train/loss:\u001b[0m\n", - " \u001b[37m6.838 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.032 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.838 train/loss:\u001b[0m\n", - " \u001b[37m6.838 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m148/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.921 train/loss:\u001b[0m\n", - " \u001b[37m6.921 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.921 train/loss:\u001b[0m\n", - " \u001b[37m6.921 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m149/180\u001b[0m \u001b[37m0:01:09 • 0:00:15\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.838 train/loss:\u001b[0m\n", - " \u001b[37m6.838 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.838 train/loss:\u001b[0m\n", - " \u001b[37m6.838 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m150/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.012 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.876 train/loss:\u001b[0m\n", - " \u001b[37m6.876 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.012 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.876 train/loss:\u001b[0m\n", - " \u001b[37m6.876 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m151/180\u001b[0m \u001b[37m0:01:10 • 0:00:14\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.829 train/loss:\u001b[0m\n", - " \u001b[37m6.829 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:10 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.008 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.829 train/loss:\u001b[0m\n", - " \u001b[37m6.829 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m152/180\u001b[0m \u001b[37m0:01:10 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.862 train/loss:\u001b[0m\n", - " \u001b[37m6.862 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:11 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.017 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.862 train/loss:\u001b[0m\n", - " \u001b[37m6.862 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m153/180\u001b[0m \u001b[37m0:01:11 • 0:00:13\u001b[0m \u001b[37m2.20it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.914 train/loss:\u001b[0m\n", - " \u001b[37m6.914 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:11 • 0:00:12\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.914 train/loss:\u001b[0m\n", - " \u001b[37m6.914 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m154/180\u001b[0m \u001b[37m0:01:11 • 0:00:12\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.050 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.764 train/loss:\u001b[0m\n", - " \u001b[37m6.764 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.050 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.764 train/loss:\u001b[0m\n", - " \u001b[37m6.764 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m155/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.895 train/loss:\u001b[0m\n", - " \u001b[37m6.895 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.895 train/loss:\u001b[0m\n", - " \u001b[37m6.895 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m156/180\u001b[0m \u001b[37m0:01:12 • 0:00:12\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.911 train/loss:\u001b[0m\n", - " \u001b[37m6.911 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:13 • 0:00:11\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.911 train/loss:\u001b[0m\n", - " \u001b[37m6.911 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m157/180\u001b[0m \u001b[37m0:01:13 • 0:00:11\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.845 train/loss:\u001b[0m\n", - " \u001b[37m6.845 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:14 • 0:00:11\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.845 train/loss:\u001b[0m\n", - " \u001b[37m6.845 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m158/180\u001b[0m \u001b[37m0:01:14 • 0:00:11\u001b[0m \u001b[37m2.12it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.828 train/loss:\u001b[0m\n", - " \u001b[37m6.828 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:15 • 0:00:11\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.828 train/loss:\u001b[0m\n", - " \u001b[37m6.828 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m159/180\u001b[0m \u001b[37m0:01:15 • 0:00:11\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.887 train/loss:\u001b[0m\n", - " \u001b[37m6.887 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.887 train/loss:\u001b[0m\n", - " \u001b[37m6.887 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m160/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.836 train/loss:\u001b[0m\n", - " \u001b[37m6.836 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.836 train/loss:\u001b[0m\n", - " \u001b[37m6.836 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m161/180\u001b[0m \u001b[37m0:01:15 • 0:00:10\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.854 train/loss:\u001b[0m\n", - " \u001b[37m6.854 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.854 train/loss:\u001b[0m\n", - " \u001b[37m6.854 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m162/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.854 train/loss:\u001b[0m\n", - " \u001b[37m6.854 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.854 train/loss:\u001b[0m\n", - " \u001b[37m6.854 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m163/180\u001b[0m \u001b[37m0:01:16 • 0:00:09\u001b[0m \u001b[37m2.09it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.915 train/loss:\u001b[0m\n", - " \u001b[37m6.915 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.915 train/loss:\u001b[0m\n", - " \u001b[37m6.915 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m164/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.08it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.931 train/loss:\u001b[0m\n", - " \u001b[37m6.931 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.931 train/loss:\u001b[0m\n", - " \u001b[37m6.931 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m165/180\u001b[0m \u001b[37m0:01:17 • 0:00:08\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.966 train/loss:\u001b[0m\n", - " \u001b[37m6.966 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.966 train/loss:\u001b[0m\n", - " \u001b[37m6.966 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m166/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.800 train/loss:\u001b[0m\n", - " \u001b[37m6.800 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.800 train/loss:\u001b[0m\n", - " \u001b[37m6.800 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m167/180\u001b[0m \u001b[37m0:01:18 • 0:00:07\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.989 train/loss:\u001b[0m\n", - " \u001b[37m6.989 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.989 train/loss:\u001b[0m\n", - " \u001b[37m6.989 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m168/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.854 train/loss:\u001b[0m\n", - " \u001b[37m6.854 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.854 train/loss:\u001b[0m\n", - " \u001b[37m6.854 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m169/180\u001b[0m \u001b[37m0:01:19 • 0:00:06\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.731 train/loss:\u001b[0m\n", - " \u001b[37m6.731 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:19 • 0:00:05\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.731 train/loss:\u001b[0m\n", - " \u001b[37m6.731 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m170/180\u001b[0m \u001b[37m0:01:19 • 0:00:05\u001b[0m \u001b[37m2.13it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.039 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.065 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.726 train/loss:\u001b[0m\n", - " \u001b[37m6.726 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:20 • 0:00:05\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.039 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.065 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.726 train/loss:\u001b[0m\n", - " \u001b[37m6.726 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m171/180\u001b[0m \u001b[37m0:01:20 • 0:00:05\u001b[0m \u001b[37m2.16it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.851 train/loss:\u001b[0m\n", - " \u001b[37m6.851 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:20 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.851 train/loss:\u001b[0m\n", - " \u001b[37m6.851 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m172/180\u001b[0m \u001b[37m0:01:20 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.945 train/loss:\u001b[0m\n", - " \u001b[37m6.945 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:21 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.945 train/loss:\u001b[0m\n", - " \u001b[37m6.945 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m173/180\u001b[0m \u001b[37m0:01:21 • 0:00:04\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.852 train/loss:\u001b[0m\n", - " \u001b[37m6.852 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:21 • 0:00:03\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.852 train/loss:\u001b[0m\n", - " \u001b[37m6.852 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m174/180\u001b[0m \u001b[37m0:01:21 • 0:00:03\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.031 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.797 train/loss:\u001b[0m\n", - " \u001b[37m6.797 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:22 • 0:00:03\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.031 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.797 train/loss:\u001b[0m\n", - " \u001b[37m6.797 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m175/180\u001b[0m \u001b[37m0:01:22 • 0:00:03\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.819 train/loss:\u001b[0m\n", - " \u001b[37m6.819 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.819 train/loss:\u001b[0m\n", - " \u001b[37m6.819 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m176/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.14it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.871 train/loss:\u001b[0m\n", - " \u001b[37m6.871 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.871 train/loss:\u001b[0m\n", - " \u001b[37m6.871 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m177/180\u001b[0m \u001b[37m0:01:23 • 0:00:02\u001b[0m \u001b[37m2.18it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.764 train/loss:\u001b[0m\n", - " \u001b[37m6.764 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:23 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.033 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.764 train/loss:\u001b[0m\n", - " \u001b[37m6.764 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m178/180\u001b[0m \u001b[37m0:01:23 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.791 train/loss:\u001b[0m\n", - " \u001b[37m6.791 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:24 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.016 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.791 train/loss:\u001b[0m\n", - " \u001b[37m6.791 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m \u001b[37m179/180\u001b[0m \u001b[37m0:01:24 • 0:00:01\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.066 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.775 train/loss:\u001b[0m\n", - " \u001b[37m6.775 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.013 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.066 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.775 train/loss:\u001b[0m\n", - " \u001b[37m6.775 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.014 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.049 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.855 train/loss:\u001b[0m\n", - " \u001b[37m6.855 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m147.992 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m180/180\u001b[0m \u001b[37m0:01:24 • 0:00:00\u001b[0m \u001b[37m2.19it/s\u001b[0m \u001b[37mv_num: \u001b[0m\n", - " \u001b[37m5156802.000 \u001b[0m\n", - " \u001b[37mtrain/accuracy: \u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/top5_accur…\u001b[0m\n", - " \u001b[37m0.000 \u001b[0m\n", - " \u001b[37mtrain/cross_entr…\u001b[0m\n", - " \u001b[37m6.766 train/loss:\u001b[0m\n", - " \u001b[37m6.766 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m140.174 \u001b[0m\n", - " \u001b[37mval/samples_per_…\u001b[0m\n", - " \u001b[37m138.743 \u001b[0m\n", - " \u001b[37mval/samples_per_…\u001b[0m\n", - " \u001b[37m136.694 \u001b[0m\n", - " \u001b[37mtrain/samples_pe…\u001b[0m\n", - " \u001b[37m146.335 \u001b[0m\n", - "\u001b[?25h/home/mila/c/cesar.valdez/.conda/envs/research_template/lib/python3.12/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=example trainer.max_epochs ...\n", - "\u001b[2;36m[17:09:43]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Not adding metrics to the pl \u001b]8;id=733785;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=345137;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#64\u001b\\\u001b[2m64\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module because they are already \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m present. \u001b[2m \u001b[0m\n", - "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: ClassificationMetricsCallback, LearningRateMonitor\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=371304;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=737977;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#381\u001b\\\u001b[2m381\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=648908;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=28924;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#338\u001b\\\u001b[2m338\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m[17:10:01]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Not adding metrics to the pl \u001b]8;id=33733;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py\u001b\\\u001b[2mclassification_metrics.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=344699;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/algorithms/callbacks/classification_metrics.py#64\u001b\\\u001b[2m64\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m module because they are already \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m present. \u001b[2m \u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.17it/s\u001b[0m \u001b[37m2.17it/s\u001b[0m \n", - "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", - "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", - "│\u001b[36m \u001b[0m\u001b[36m val/accuracy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.003010033629834652 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36m val/cross_entropy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 6.422720432281494 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 6.422720432281494 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 140.8275146484375 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36m val/top5_accuracy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.029230769723653793 \u001b[0m\u001b[35m \u001b[0m│\n", - "└──────────────────────────────┴──────────────────────────────┘\n", - "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m20/20\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m2.17it/s\u001b[0m \n", - "\u001b[?25hval val/accuracy: \u001b[1;36m0.003010033629834652\u001b[0m\n", - "val val/top5_accuracy: \u001b[1;36m0.029230769723653793\u001b[0m\n", - "val val/cross_entropy: \u001b[1;36m6.422720432281494\u001b[0m\n", - "val val/samples_per_second_epoch: \u001b[1;36m140.8275146484375\u001b[0m\n" - ] - } - ], + "outputs": [], "source": [ - "!HYDRA_FULL_ERROR=1 python project/main.py \\\n", - " algorithm=example \\\n", - " trainer.max_epochs=1 \\\n", - " +trainer.limit_train_batches=0.01\\\n", - " +trainer.limit_val_batches=0.01\\\n", - " datamodule=imagenet" + "#!HYDRA_FULL_ERROR=1 python project/main.py \\\n", + "# algorithm=example \\\n", + "# trainer.max_epochs=1 \\\n", + "# +trainer.limit_train_batches=0.01\\\n", + "# +trainer.limit_val_batches=0.01\\\n", + "# datamodule=imagenet" ] }, { @@ -5745,28 +83,11 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "GPU Avail / Total \n", - "===============================\n", - "2g.20gb 31 / 48 \n", - "3g.40gb 9 / 48 \n", - "4g.40gb 7 / 24 \n", - "a100 8 / 16 \n", - "a100l 0 / 72 \n", - "a6000 0 / 8 \n", - "rtx8000 11 / 400 \n", - "v100 2 / 40 \n" - ] - } - ], + "outputs": [], "source": [ - "!savail" + "#!savail" ] }, { @@ -5814,7 +135,7 @@ }, { "cell_type": "code", - "execution_count": 86, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ @@ -5832,7 +153,7 @@ }, { "cell_type": "code", - "execution_count": 74, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ @@ -5842,7 +163,7 @@ }, { "cell_type": "code", - "execution_count": 75, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ @@ -5859,7 +180,7 @@ }, { "cell_type": "code", - "execution_count": 76, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ @@ -5897,7 +218,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.12.4" + "version": "3.10.14" } }, "nbformat": 4, From bf514943072cf4ae9a765be235a66fa5afcd2aad Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Tue, 3 Sep 2024 15:45:27 -0400 Subject: [PATCH 07/33] nbstripout compliance --- docs/examples/profiling.ipynb | 30 +- mkdocs.yml | 5 +- pyproject.toml | 6 +- requirements-dev.lock | 507 ++++++++++++++++++++++------------ requirements.lock | 456 ++++++++++++++++++++---------- 5 files changed, 678 insertions(+), 326 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 9e6ae30a..84d99e02 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -4,7 +4,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "# Benchmarking" + "# Profiling your code" ] }, { @@ -41,12 +41,28 @@ "metadata": {}, "outputs": [], "source": [ - "#!python project/main.py \\\n", - "# algorithm=NoOp \\\n", - "# trainer.max_epochs=1 \\\n", - "# +trainer.limit_train_batches=0.01\\\n", - "# +trainer.limit_val_batches=0.01\\\n", - "# datamodule=imagenet" + "import os\n", + "from pathlib import Path\n", + "\n", + "# Set the working directory to the project root\n", + "notebook_path = Path().resolve() \n", + "project_root = notebook_path.parent.parent\n", + "os.chdir(str(project_root))\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "!python project/main.py \\\n", + " algorithm=example \\\n", + " datamodule=imagenet \\\n", + " ++trainer.max_epochs=0 \\\n", + " ++trainer.limit_train_batches=0\\\n", + "\n", + "#!python ../../project/main.py throws an error about relative paths" ] }, { diff --git a/mkdocs.yml b/mkdocs.yml index 3e3e813a..9f7e4387 100644 --- a/mkdocs.yml +++ b/mkdocs.yml @@ -3,7 +3,9 @@ site_description: A project template and directory structure for Python data sci site_url: https://mila-iqia.github.io/ResearchTemplate/ repo_url: https://www.github.com/mila-iqia/ResearchTemplate # edit_uri: edit/master/docs - +nav: + - Home: index.md + - Profiling your code: docs/examples/profiling.ipynb theme: name: material features: @@ -92,6 +94,7 @@ plugins: video_controls: True css_style: width: "100%" + - mkdocs-jupyter # todo: take a look at https://github.com/drivendataorg/cookiecutter-data-science/blob/master/docs/mkdocs.yml # - admonition # - pymdownx.details diff --git a/pyproject.toml b/pyproject.toml index 04debe18..dc991099 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -23,10 +23,11 @@ dependencies = [ "torch-jax-interop>=0.0.7", "pydantic>=2.8.2", "simple-parsing>=0.1.5", - "pydantic==2.7.4", + "pydantic==2.8.2", "milatools>=0.0.18", "rootutils>=0.0.1", - "ipykernel>=6.28.0" + "ipykernel>=6.28.0", + "dill>=0.3.8", ] readme = "README.md" requires-python = ">= 3.10" @@ -42,6 +43,7 @@ docs = [ "mkdocs>=1.6.0", "mkdocs-video>=1.5.0", "mkdocs-section-index>=0.3.9", + "mkdocs-jupyter>=0.24.8", ] gpu = ["jax[cuda12]>=0.4.31"] diff --git a/requirements-dev.lock b/requirements-dev.lock index 93aa7a90..b873c192 100644 --- a/requirements-dev.lock +++ b/requirements-dev.lock @@ -10,126 +10,166 @@ # universal: true -e file:. -absl-py==2.1.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via file:///- +absl-py==2.1.0 # via chex # via optax # via orbax-checkpoint -aiohappyeyeballs==2.3.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +aiohappyeyeballs==2.3.5 # via aiohttp -aiohttp==3.10.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +aiohttp==3.10.3 # via fsspec -aiosignal==1.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +aiosignal==1.3.1 # via aiohttp -annotated-types==0.7.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +annotated-types==0.7.0 # via pydantic -antlr4-python3-runtime==4.9.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +antlr4-python3-runtime==4.9.3 # via hydra-core # via omegaconf -async-timeout==4.0.3 ; python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +asttokens==2.4.1 + # via stack-data +async-timeout==4.0.3 # via aiohttp -attrs==24.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +attrs==24.2.0 # via aiohttp -babel==2.16.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via jsonschema + # via referencing +babel==2.16.0 # via mkdocs-material -beautifulsoup4==4.12.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +bcrypt==4.2.0 + # via paramiko +beautifulsoup4==4.12.3 # via gdown -black==24.8.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via nbconvert +black==24.8.0 # via research-project-template -bracex==2.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +bleach==6.1.0 + # via nbconvert +blessed==1.20.0 + # via milatools +bracex==2.5 # via wcmatch -certifi==2024.7.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +certifi==2024.7.4 # via requests # via sentry-sdk -cfgv==3.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +cffi==1.17.0 + # via cryptography + # via pynacl +cfgv==3.4.0 # via pre-commit -charset-normalizer==3.3.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +charset-normalizer==3.3.2 # via requests -chex==0.1.86 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +chex==0.1.86 # via optax -click==8.1.7 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +click==8.1.7 # via black # via mkdocs # via mkdocstrings # via wandb -cloudpickle==3.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +cloudpickle==3.0.0 # via submitit -colorama==0.4.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_system == 'Windows' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (sys_platform == 'win32' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) - # via click - # via colorlog +colorama==0.4.6 # via griffe - # via mkdocs # via mkdocs-material - # via pytest - # via tqdm -colorlog==6.8.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +colorlog==6.8.2 # via hydra-colorlog -contourpy==1.2.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +comm==0.2.2 + # via ipykernel +contourpy==1.2.1 # via matplotlib -coverage==7.6.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +coverage==7.6.1 + # via coverage # via pytest-cov # via pytest-testmon -cycler==0.12.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +cryptography==43.0.0 + # via paramiko +cycler==0.12.1 # via matplotlib -distlib==0.3.8 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +debugpy==1.8.5 + # via ipykernel +decorator==5.1.1 + # via fabric + # via ipython +defusedxml==0.7.1 + # via nbconvert +deprecated==1.2.14 + # via fabric +dill==0.3.8 + # via research-project-template +distlib==0.3.8 # via virtualenv -docker-pycreds==0.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +docker-pycreds==0.4.0 # via wandb -docstring-parser==0.16 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +docstring-parser==0.16 # via simple-parsing -etils==1.7.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +etils==1.7.0 # via optax # via orbax-checkpoint -exceptiongroup==1.2.2 ; python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +exceptiongroup==1.2.2 + # via ipython # via pytest -execnet==2.1.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +execnet==2.1.1 # via pytest-xdist -filelock==3.15.4 ; (python_version < '3.13' and platform_machine == 'x86_64' and platform_system == 'Linux') or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +executing==2.0.1 + # via stack-data +fabric==3.2.2 + # via milatools +fastjsonschema==2.20.0 + # via nbformat +filelock==3.15.4 # via gdown # via torch # via triton # via virtualenv -flax==0.8.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +flax==0.8.5 # via torch-jax-interop -fonttools==4.53.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +fonttools==4.53.1 # via matplotlib -frozenlist==1.4.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +frozenlist==1.4.1 # via aiohttp # via aiosignal -fsspec==2024.6.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +fsspec==2024.6.1 # via etils # via lightning # via pytorch-lightning # via torch -gdown==5.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +gdown==5.2.0 # via research-project-template -ghp-import==2.1.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +ghp-import==2.1.0 # via mkdocs -gitdb==4.0.11 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +gitdb==4.0.11 # via gitpython -gitpython==3.1.43 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +gitpython==3.1.43 # via wandb -griffe==0.49.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +griffe==0.49.0 # via mkdocstrings-python -hydra-colorlog==1.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-colorlog==1.2.0 # via research-project-template -hydra-core==1.3.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-core==1.3.2 # via hydra-colorlog # via hydra-submitit-launcher # via hydra-zen # via research-project-template -hydra-submitit-launcher==1.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-submitit-launcher==1.2.0 # via research-project-template -hydra-zen==0.13.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-zen==0.13.0 # via research-project-template -identify==2.6.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +identify==2.6.0 # via pre-commit -idna==3.7 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +idna==3.7 # via requests # via yarl -importlib-resources==6.4.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +importlib-resources==6.4.2 # via etils -iniconfig==2.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +iniconfig==2.0.0 # via pytest +invoke==2.2.0 + # via fabric +ipykernel==6.29.5 + # via mkdocs-jupyter + # via research-project-template +ipython==8.27.0 + # via ipykernel jax==0.4.31 # via chex # via flax @@ -138,108 +178,155 @@ jax==0.4.31 # via pytorch2jax # via research-project-template # via torch-jax-interop -jax-cuda12-pjrt==0.4.31 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +jax-cuda12-pjrt==0.4.31 # via jax-cuda12-plugin jax-cuda12-plugin==0.4.31 # via jax -jaxlib==0.4.31 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via jax-cuda12-plugin +jaxlib==0.4.31 # via chex # via jax # via optax # via orbax-checkpoint # via pytorch2jax -jinja2==3.1.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +jedi==0.19.1 + # via ipython +jinja2==3.1.4 # via mkdocs # via mkdocs-material # via mkdocstrings + # via nbconvert # via torch -kiwisolver==1.4.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +jsonschema==4.23.0 + # via nbformat +jsonschema-specifications==2023.12.1 + # via jsonschema +jupyter-client==8.6.2 + # via ipykernel + # via nbclient +jupyter-core==5.7.2 + # via ipykernel + # via jupyter-client + # via nbclient + # via nbconvert + # via nbformat +jupyterlab-pygments==0.3.0 + # via nbconvert +jupytext==1.16.4 + # via mkdocs-jupyter +kiwisolver==1.4.5 # via matplotlib -lightning==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +lightning==2.4.0 # via research-project-template -lightning-utilities==0.11.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +lightning-utilities==0.11.6 # via lightning # via pytorch-lightning # via torchmetrics -lxml==5.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +lxml==5.3.0 # via mkdocs-video -markdown==3.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +markdown==3.6 # via mkdocs # via mkdocs-autorefs # via mkdocs-material # via mkdocstrings # via pymdown-extensions -markdown-it-py==3.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +markdown-it-py==3.0.0 + # via jupytext + # via mdit-py-plugins # via rich -markupsafe==2.1.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +markupsafe==2.1.5 # via jinja2 # via mkdocs # via mkdocs-autorefs # via mkdocstrings -matplotlib==3.9.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via nbconvert +matplotlib==3.9.2 # via research-project-template -mdurl==0.1.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +matplotlib-inline==0.1.7 + # via ipykernel + # via ipython +mdit-py-plugins==0.4.1 + # via jupytext +mdurl==0.1.2 # via markdown-it-py -mergedeep==1.3.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mergedeep==1.3.4 # via mkdocs # via mkdocs-get-deps -mkdocs==1.6.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +milatools==0.1.5 + # via research-project-template +mistune==3.0.2 + # via nbconvert +mkdocs==1.6.0 # via mkdocs-autorefs # via mkdocs-awesome-pages-plugin # via mkdocs-gen-files + # via mkdocs-jupyter # via mkdocs-literate-nav # via mkdocs-material # via mkdocs-section-index # via mkdocs-video # via mkdocstrings # via research-project-template -mkdocs-autorefs==1.0.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-autorefs==1.0.1 # via mkdocstrings -mkdocs-awesome-pages-plugin==2.9.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-awesome-pages-plugin==2.9.3 # via research-project-template -mkdocs-gen-files==0.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-gen-files==0.5.0 # via research-project-template -mkdocs-get-deps==0.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-get-deps==0.2.0 # via mkdocs -mkdocs-literate-nav==0.6.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-jupyter==0.24.8 + # via research-project-template +mkdocs-literate-nav==0.6.1 # via research-project-template -mkdocs-material==9.5.31 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-material==9.5.31 + # via mkdocs-jupyter # via research-project-template -mkdocs-material-extensions==1.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-material-extensions==1.3.1 # via mkdocs-material -mkdocs-section-index==0.3.9 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-section-index==0.3.9 # via research-project-template -mkdocs-video==1.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-video==1.5.0 # via research-project-template -mkdocstrings==0.25.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocstrings==0.25.2 + # via mkdocstrings # via mkdocstrings-python # via research-project-template -mkdocstrings-python==1.10.8 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocstrings-python==1.10.8 # via mkdocstrings -mktestdocs==0.2.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -ml-dtypes==0.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mktestdocs==0.2.2 +ml-dtypes==0.4.0 # via jax # via jaxlib # via tensorstore -mpmath==1.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mpmath==1.3.0 # via sympy -msgpack==1.0.8 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +msgpack==1.0.8 # via flax # via orbax-checkpoint -multidict==6.0.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +multidict==6.0.5 # via aiohttp # via yarl -mypy-extensions==1.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mypy-extensions==1.0.0 # via black -natsort==8.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +natsort==8.4.0 # via mkdocs-awesome-pages-plugin -nest-asyncio==1.6.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +nbclient==0.10.0 + # via nbconvert +nbconvert==7.16.4 + # via mkdocs-jupyter +nbformat==5.10.4 + # via jupytext + # via nbclient + # via nbconvert +nest-asyncio==1.6.0 + # via ipykernel # via orbax-checkpoint -networkx==3.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +networkx==3.3 # via torch -nodeenv==1.9.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +nodeenv==1.9.1 # via pre-commit -numpy==1.26.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or ((python_version >= '3.11' or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or ((python_version >= '3.12' or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +numpy==1.26.4 # via chex # via contourpy # via flax @@ -255,103 +342,130 @@ numpy==1.26.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python # via tensorstore # via torchmetrics # via torchvision -nvidia-cublas-cu12==12.1.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))) +nvidia-cublas-cu12==12.1.3.1 # via jax-cuda12-plugin # via nvidia-cudnn-cu12 # via nvidia-cusolver-cu12 # via torch -nvidia-cuda-cupti-cu12==12.1.105 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cuda-cupti-cu12==12.1.105 # via jax-cuda12-plugin # via torch -nvidia-cuda-nvcc-cu12==12.6.20 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +nvidia-cuda-nvcc-cu12==12.6.20 # via jax-cuda12-plugin -nvidia-cuda-nvrtc-cu12==12.1.105 ; platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +nvidia-cuda-nvrtc-cu12==12.1.105 # via torch -nvidia-cuda-runtime-cu12==12.1.105 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cuda-runtime-cu12==12.1.105 # via jax-cuda12-plugin # via torch -nvidia-cudnn-cu12==9.1.0.70 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cudnn-cu12==9.1.0.70 # via jax-cuda12-plugin # via torch -nvidia-cufft-cu12==11.0.2.54 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cufft-cu12==11.0.2.54 # via jax-cuda12-plugin # via torch -nvidia-curand-cu12==10.3.2.106 ; platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +nvidia-curand-cu12==10.3.2.106 # via torch -nvidia-cusolver-cu12==11.4.5.107 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cusolver-cu12==11.4.5.107 # via jax-cuda12-plugin # via torch -nvidia-cusparse-cu12==12.1.0.106 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))) +nvidia-cusparse-cu12==12.1.0.106 # via jax-cuda12-plugin # via nvidia-cusolver-cu12 # via torch -nvidia-nccl-cu12==2.20.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-nccl-cu12==2.20.5 # via jax-cuda12-plugin # via torch -nvidia-nvjitlink-cu12==12.6.20 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))))) +nvidia-nvjitlink-cu12==12.6.20 # via jax-cuda12-plugin # via nvidia-cusolver-cu12 # via nvidia-cusparse-cu12 -nvidia-nvtx-cu12==12.1.105 ; platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +nvidia-nvtx-cu12==12.1.105 # via torch -omegaconf==2.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +omegaconf==2.3.0 # via hydra-core # via hydra-zen -opt-einsum==3.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +opt-einsum==3.3.0 # via jax -optax==0.2.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +optax==0.2.3 # via flax -orbax-checkpoint==0.5.23 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +orbax-checkpoint==0.5.23 # via flax -packaging==24.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +packaging==24.1 # via black # via hydra-core + # via ipykernel + # via jupytext # via lightning # via lightning-utilities # via matplotlib # via mkdocs + # via nbconvert # via pytest # via pytorch-lightning # via torchmetrics -paginate==0.5.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +paginate==0.5.6 # via mkdocs-material -pathspec==0.12.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pandocfilters==1.5.1 + # via nbconvert +paramiko==3.4.1 + # via fabric +parso==0.8.4 + # via jedi +pathspec==0.12.1 # via black # via mkdocs -pillow==10.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pexpect==4.9.0 + # via ipython +pillow==10.4.0 # via matplotlib # via torchvision -platformdirs==4.2.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +platformdirs==4.2.2 # via black + # via jupyter-core # via mkdocs-get-deps # via mkdocstrings # via virtualenv # via wandb -pluggy==1.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pluggy==1.5.0 # via pytest -pre-commit==3.8.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -protobuf==5.27.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pre-commit==3.8.0 +prompt-toolkit==3.0.47 + # via ipython + # via questionary +protobuf==5.27.3 # via orbax-checkpoint # via wandb -psutil==6.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +psutil==6.0.0 + # via ipykernel # via wandb -py-cpuinfo==9.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +ptyprocess==0.7.0 + # via pexpect +pure-eval==0.2.3 + # via stack-data +py-cpuinfo==9.0.0 # via pytest-benchmark -pydantic==2.8.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pycparser==2.22 + # via cffi +pydantic==2.8.2 # via research-project-template -pydantic-core==2.20.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pydantic-core==2.20.1 # via pydantic -pygments==2.18.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pygments==2.18.0 + # via ipython + # via mkdocs-jupyter # via mkdocs-material + # via nbconvert # via rich -pymdown-extensions==10.9 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pymdown-extensions==10.9 # via mkdocs-material # via mkdocstrings -pyparsing==3.1.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pynacl==1.5.0 + # via paramiko +pyparsing==3.1.2 # via matplotlib -pysocks==1.7.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pysocks==1.7.1 # via requests -pytest==8.3.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pytest==8.3.2 # via pytest-benchmark # via pytest-cov # via pytest-datadir @@ -361,26 +475,30 @@ pytest==8.3.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python # via pytest-testmon # via pytest-timeout # via pytest-xdist -pytest-benchmark==4.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -pytest-cov==5.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -pytest-datadir==1.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pytest-benchmark==4.0.0 +pytest-cov==5.0.0 +pytest-datadir==1.5.0 # via pytest-regressions -pytest-env==1.1.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -pytest-regressions==2.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pytest-env==1.1.3 +pytest-regressions==2.5.0 # via tensor-regression -pytest-skip-slow==0.0.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -pytest-testmon==2.1.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -pytest-timeout==2.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -pytest-xdist==3.6.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -python-dateutil==2.9.0.post0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pytest-skip-slow==0.0.5 +pytest-testmon==2.1.1 +pytest-timeout==2.3.1 +pytest-xdist==3.6.1 +python-dateutil==2.9.0.post0 # via ghp-import + # via jupyter-client # via matplotlib -pytorch-lightning==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +python-dotenv==1.0.1 + # via rootutils +pytorch-lightning==2.4.0 # via lightning -pytorch2jax==0.1.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pytorch2jax==0.1.0 # via torch-jax-interop -pyyaml==6.0.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pyyaml==6.0.2 # via flax + # via jupytext # via lightning # via mkdocs # via mkdocs-get-deps @@ -392,54 +510,74 @@ pyyaml==6.0.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python # via pytorch-lightning # via pyyaml-env-tag # via wandb -pyyaml-env-tag==0.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pyyaml-env-tag==0.1 # via mkdocs -regex==2024.7.24 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pyzmq==26.2.0 + # via ipykernel + # via jupyter-client +questionary==1.10.0 + # via milatools +referencing==0.35.1 + # via jsonschema + # via jsonschema-specifications +regex==2024.7.24 # via mkdocs-material -requests==2.32.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +requests==2.32.3 # via gdown # via mkdocs-material # via wandb -rich==13.7.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +rich==13.7.1 # via flax + # via milatools + # via research-project-template +rootutils==1.0.7 # via research-project-template -ruff==0.6.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -scipy==1.14.0 ; (python_version >= '3.12' or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +rpds-py==0.20.0 + # via jsonschema + # via referencing +ruff==0.6.0 +scipy==1.14.0 # via jax # via jaxlib -sentry-sdk==2.13.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +sentry-sdk==2.13.0 # via wandb -setproctitle==1.3.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +setproctitle==1.3.3 # via wandb -setuptools==72.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.12' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') - # via chex - # via lightning-utilities - # via wandb -simple-parsing==0.1.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +simple-parsing==0.1.5 # via research-project-template -six==1.16.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +six==1.16.0 + # via asttokens + # via bleach + # via blessed # via docker-pycreds # via python-dateutil -smmap==5.0.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +smmap==5.0.1 # via gitdb -soupsieve==2.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +soupsieve==2.6 # via beautifulsoup4 -submitit==1.5.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +sshconf==0.2.7 + # via milatools +stack-data==0.6.3 + # via ipython +submitit==1.5.1 # via hydra-submitit-launcher -sympy==1.13.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +sympy==1.13.2 # via torch -tensor-regression==0.0.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') -tensorstore==0.1.64 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +tensor-regression==0.0.6 +tensorstore==0.1.64 # via flax # via orbax-checkpoint -tomli==2.0.1 ; (python_full_version <= '3.11.0a6' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +tinycss2==1.3.0 + # via nbconvert +tomli==2.0.1 # via black # via coverage + # via jupytext # via pytest # via pytest-env -toolz==0.12.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +toolz==0.12.1 # via chex -torch==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torch==2.4.0 # via lightning # via pytorch-lightning # via pytorch2jax @@ -448,28 +586,44 @@ torch==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_ # via torch-jax-interop # via torchmetrics # via torchvision -torch-jax-interop==0.0.7 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torch-jax-interop==0.0.7 # via research-project-template -torchmetrics==1.4.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torchmetrics==1.4.1 # via lightning # via pytorch-lightning -torchvision==0.19.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torchvision==0.19.0 # via research-project-template -tqdm==4.66.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +tornado==6.4.1 + # via ipykernel + # via jupyter-client +tqdm==4.66.5 # via gdown # via lightning + # via milatools # via pytorch-lightning # via research-project-template -triton==3.0.0 ; python_version < '3.13' and platform_machine == 'x86_64' and platform_system == 'Linux' +traitlets==5.14.3 + # via comm + # via ipykernel + # via ipython + # via jupyter-client + # via jupyter-core + # via matplotlib-inline + # via nbclient + # via nbconvert + # via nbformat +triton==3.0.0 # via torch -typing-extensions==4.12.2 ; (python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +typing-extensions==4.12.2 # via black # via chex # via etils # via flax # via hydra-zen + # via ipython # via lightning # via lightning-utilities + # via milatools # via orbax-checkpoint # via pydantic # via pydantic-core @@ -477,18 +631,29 @@ typing-extensions==4.12.2 ; (python_version < '3.11' and ((python_version < '3.1 # via simple-parsing # via submitit # via torch -urllib3==2.2.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +urllib3==2.2.2 # via requests # via sentry-sdk -virtualenv==20.26.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +virtualenv==20.26.3 # via pre-commit -wandb==0.17.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +wandb==0.17.6 # via research-project-template -watchdog==4.0.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +watchdog==4.0.2 # via mkdocs -wcmatch==9.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +wcmatch==9.0 # via mkdocs-awesome-pages-plugin -yarl==1.9.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +wcwidth==0.2.13 + # via blessed + # via prompt-toolkit +webencodings==0.5.1 + # via bleach + # via tinycss2 +wrapt==1.16.0 + # via deprecated +yarl==1.9.4 # via aiohttp -zipp==3.20.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +zipp==3.20.0 # via etils +setuptools==72.2.0 + # via lightning-utilities + # via wandb diff --git a/requirements.lock b/requirements.lock index a1e9e7ec..b253afa2 100644 --- a/requirements.lock +++ b/requirements.lock @@ -10,109 +10,150 @@ # universal: true -e file:. -absl-py==2.1.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via file:///- +absl-py==2.1.0 # via chex # via optax # via orbax-checkpoint -aiohappyeyeballs==2.3.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +aiohappyeyeballs==2.3.5 # via aiohttp -aiohttp==3.10.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +aiohttp==3.10.3 # via fsspec -aiosignal==1.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +aiosignal==1.3.1 # via aiohttp -annotated-types==0.7.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +annotated-types==0.7.0 # via pydantic -antlr4-python3-runtime==4.9.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +antlr4-python3-runtime==4.9.3 # via hydra-core # via omegaconf -async-timeout==4.0.3 ; python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +asttokens==2.4.1 + # via stack-data +async-timeout==4.0.3 # via aiohttp -attrs==24.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +attrs==24.2.0 # via aiohttp -babel==2.16.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via jsonschema + # via referencing +babel==2.16.0 # via mkdocs-material -beautifulsoup4==4.12.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +bcrypt==4.2.0 + # via paramiko +beautifulsoup4==4.12.3 # via gdown -black==24.8.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via nbconvert +black==24.8.0 # via research-project-template -bracex==2.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +bleach==6.1.0 + # via nbconvert +blessed==1.20.0 + # via milatools +bracex==2.5 # via wcmatch -certifi==2024.7.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +certifi==2024.7.4 # via requests # via sentry-sdk -charset-normalizer==3.3.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +cffi==1.17.0 + # via cryptography + # via pynacl +charset-normalizer==3.3.2 # via requests -chex==0.1.86 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +chex==0.1.86 # via optax -click==8.1.7 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +click==8.1.7 # via black # via mkdocs # via mkdocstrings # via wandb -cloudpickle==3.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +cloudpickle==3.0.0 # via submitit -colorama==0.4.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_system == 'Windows' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (sys_platform == 'win32' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) - # via click - # via colorlog +colorama==0.4.6 # via griffe - # via mkdocs # via mkdocs-material - # via tqdm -colorlog==6.8.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +colorlog==6.8.2 # via hydra-colorlog -contourpy==1.2.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +comm==0.2.2 + # via ipykernel +contourpy==1.2.1 # via matplotlib -cycler==0.12.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +cryptography==43.0.0 + # via paramiko +cycler==0.12.1 # via matplotlib -docker-pycreds==0.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +debugpy==1.8.5 + # via ipykernel +decorator==5.1.1 + # via fabric + # via ipython +defusedxml==0.7.1 + # via nbconvert +deprecated==1.2.14 + # via fabric +dill==0.3.8 + # via research-project-template +docker-pycreds==0.4.0 # via wandb -docstring-parser==0.16 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +docstring-parser==0.16 # via simple-parsing -etils==1.7.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +etils==1.7.0 # via optax # via orbax-checkpoint -filelock==3.15.4 ; (python_version < '3.13' and platform_machine == 'x86_64' and platform_system == 'Linux') or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +exceptiongroup==1.2.2 + # via ipython +executing==2.0.1 + # via stack-data +fabric==3.2.2 + # via milatools +fastjsonschema==2.20.0 + # via nbformat +filelock==3.15.4 # via gdown # via torch # via triton -flax==0.8.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +flax==0.8.5 # via torch-jax-interop -fonttools==4.53.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +fonttools==4.53.1 # via matplotlib -frozenlist==1.4.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +frozenlist==1.4.1 # via aiohttp # via aiosignal -fsspec==2024.6.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +fsspec==2024.6.1 # via etils # via lightning # via pytorch-lightning # via torch -gdown==5.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +gdown==5.2.0 # via research-project-template -ghp-import==2.1.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +ghp-import==2.1.0 # via mkdocs -gitdb==4.0.11 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +gitdb==4.0.11 # via gitpython -gitpython==3.1.43 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +gitpython==3.1.43 # via wandb -griffe==0.49.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +griffe==0.49.0 # via mkdocstrings-python -hydra-colorlog==1.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-colorlog==1.2.0 # via research-project-template -hydra-core==1.3.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-core==1.3.2 # via hydra-colorlog # via hydra-submitit-launcher # via hydra-zen # via research-project-template -hydra-submitit-launcher==1.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-submitit-launcher==1.2.0 # via research-project-template -hydra-zen==0.13.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +hydra-zen==0.13.0 # via research-project-template -idna==3.7 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +idna==3.7 # via requests # via yarl -importlib-resources==6.4.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +importlib-resources==6.4.2 # via etils +invoke==2.2.0 + # via fabric +ipykernel==6.29.5 + # via mkdocs-jupyter + # via research-project-template +ipython==8.27.0 + # via ipykernel jax==0.4.31 # via chex # via flax @@ -121,105 +162,152 @@ jax==0.4.31 # via pytorch2jax # via research-project-template # via torch-jax-interop -jax-cuda12-pjrt==0.4.31 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +jax-cuda12-pjrt==0.4.31 # via jax-cuda12-plugin jax-cuda12-plugin==0.4.31 # via jax -jaxlib==0.4.31 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via jax-cuda12-plugin +jaxlib==0.4.31 # via chex # via jax # via optax # via orbax-checkpoint # via pytorch2jax -jinja2==3.1.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +jedi==0.19.1 + # via ipython +jinja2==3.1.4 # via mkdocs # via mkdocs-material # via mkdocstrings + # via nbconvert # via torch -kiwisolver==1.4.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +jsonschema==4.23.0 + # via nbformat +jsonschema-specifications==2023.12.1 + # via jsonschema +jupyter-client==8.6.2 + # via ipykernel + # via nbclient +jupyter-core==5.7.2 + # via ipykernel + # via jupyter-client + # via nbclient + # via nbconvert + # via nbformat +jupyterlab-pygments==0.3.0 + # via nbconvert +jupytext==1.16.4 + # via mkdocs-jupyter +kiwisolver==1.4.5 # via matplotlib -lightning==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +lightning==2.4.0 # via research-project-template -lightning-utilities==0.11.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +lightning-utilities==0.11.6 # via lightning # via pytorch-lightning # via torchmetrics -lxml==5.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +lxml==5.3.0 # via mkdocs-video -markdown==3.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +markdown==3.6 # via mkdocs # via mkdocs-autorefs # via mkdocs-material # via mkdocstrings # via pymdown-extensions -markdown-it-py==3.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +markdown-it-py==3.0.0 + # via jupytext + # via mdit-py-plugins # via rich -markupsafe==2.1.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +markupsafe==2.1.5 # via jinja2 # via mkdocs # via mkdocs-autorefs # via mkdocstrings -matplotlib==3.9.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via nbconvert +matplotlib==3.9.2 # via research-project-template -mdurl==0.1.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +matplotlib-inline==0.1.7 + # via ipykernel + # via ipython +mdit-py-plugins==0.4.1 + # via jupytext +mdurl==0.1.2 # via markdown-it-py -mergedeep==1.3.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mergedeep==1.3.4 # via mkdocs # via mkdocs-get-deps -mkdocs==1.6.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +milatools==0.1.5 + # via research-project-template +mistune==3.0.2 + # via nbconvert +mkdocs==1.6.0 # via mkdocs-autorefs # via mkdocs-awesome-pages-plugin # via mkdocs-gen-files + # via mkdocs-jupyter # via mkdocs-literate-nav # via mkdocs-material # via mkdocs-section-index # via mkdocs-video # via mkdocstrings # via research-project-template -mkdocs-autorefs==1.0.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-autorefs==1.0.1 # via mkdocstrings -mkdocs-awesome-pages-plugin==2.9.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-awesome-pages-plugin==2.9.3 # via research-project-template -mkdocs-gen-files==0.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-gen-files==0.5.0 # via research-project-template -mkdocs-get-deps==0.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-get-deps==0.2.0 # via mkdocs -mkdocs-literate-nav==0.6.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-jupyter==0.24.8 + # via research-project-template +mkdocs-literate-nav==0.6.1 # via research-project-template -mkdocs-material==9.5.31 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-material==9.5.31 + # via mkdocs-jupyter # via research-project-template -mkdocs-material-extensions==1.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-material-extensions==1.3.1 # via mkdocs-material -mkdocs-section-index==0.3.9 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-section-index==0.3.9 # via research-project-template -mkdocs-video==1.5.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocs-video==1.5.0 # via research-project-template -mkdocstrings==0.25.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocstrings==0.25.2 + # via mkdocstrings # via mkdocstrings-python # via research-project-template -mkdocstrings-python==1.10.8 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mkdocstrings-python==1.10.8 # via mkdocstrings -ml-dtypes==0.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +ml-dtypes==0.4.0 # via jax # via jaxlib # via tensorstore -mpmath==1.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mpmath==1.3.0 # via sympy -msgpack==1.0.8 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +msgpack==1.0.8 # via flax # via orbax-checkpoint -multidict==6.0.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +multidict==6.0.5 # via aiohttp # via yarl -mypy-extensions==1.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +mypy-extensions==1.0.0 # via black -natsort==8.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +natsort==8.4.0 # via mkdocs-awesome-pages-plugin -nest-asyncio==1.6.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +nbclient==0.10.0 + # via nbconvert +nbconvert==7.16.4 + # via mkdocs-jupyter +nbformat==5.10.4 + # via jupytext + # via nbclient + # via nbconvert +nest-asyncio==1.6.0 + # via ipykernel # via orbax-checkpoint -networkx==3.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +networkx==3.3 # via torch -numpy==1.26.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or ((python_version >= '3.11' or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or ((python_version >= '3.12' or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +numpy==1.26.4 # via chex # via contourpy # via flax @@ -234,104 +322,135 @@ numpy==1.26.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python # via tensorstore # via torchmetrics # via torchvision -nvidia-cublas-cu12==12.1.3.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))) +nvidia-cublas-cu12==12.1.3.1 # via jax-cuda12-plugin # via nvidia-cudnn-cu12 # via nvidia-cusolver-cu12 # via torch -nvidia-cuda-cupti-cu12==12.1.105 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cuda-cupti-cu12==12.1.105 # via jax-cuda12-plugin # via torch -nvidia-cuda-nvcc-cu12==12.6.20 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +nvidia-cuda-nvcc-cu12==12.6.20 # via jax-cuda12-plugin -nvidia-cuda-nvrtc-cu12==12.1.105 ; platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +nvidia-cuda-nvrtc-cu12==12.1.105 # via torch -nvidia-cuda-runtime-cu12==12.1.105 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cuda-runtime-cu12==12.1.105 # via jax-cuda12-plugin # via torch -nvidia-cudnn-cu12==9.1.0.70 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cudnn-cu12==9.1.0.70 # via jax-cuda12-plugin # via torch -nvidia-cufft-cu12==11.0.2.54 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cufft-cu12==11.0.2.54 # via jax-cuda12-plugin # via torch -nvidia-curand-cu12==10.3.2.106 ; platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +nvidia-curand-cu12==10.3.2.106 # via torch -nvidia-cusolver-cu12==11.4.5.107 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-cusolver-cu12==11.4.5.107 # via jax-cuda12-plugin # via torch -nvidia-cusparse-cu12==12.1.0.106 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))) +nvidia-cusparse-cu12==12.1.0.106 # via jax-cuda12-plugin # via nvidia-cusolver-cu12 # via torch -nvidia-nccl-cu12==2.20.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) +nvidia-nccl-cu12==2.20.5 # via jax-cuda12-plugin # via torch -nvidia-nvjitlink-cu12==12.6.20 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') or (platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))))))) +nvidia-nvjitlink-cu12==12.6.20 # via jax-cuda12-plugin # via nvidia-cusolver-cu12 # via nvidia-cusparse-cu12 -nvidia-nvtx-cu12==12.1.105 ; platform_machine == 'x86_64' and platform_system == 'Linux' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +nvidia-nvtx-cu12==12.1.105 # via torch -omegaconf==2.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +omegaconf==2.3.0 # via hydra-core # via hydra-zen -opt-einsum==3.3.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +opt-einsum==3.3.0 # via jax -optax==0.2.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +optax==0.2.3 # via flax -orbax-checkpoint==0.5.23 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +orbax-checkpoint==0.5.23 # via flax -packaging==24.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +packaging==24.1 # via black # via hydra-core + # via ipykernel + # via jupytext # via lightning # via lightning-utilities # via matplotlib # via mkdocs + # via nbconvert # via pytorch-lightning # via torchmetrics -paginate==0.5.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +paginate==0.5.6 # via mkdocs-material -pathspec==0.12.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pandocfilters==1.5.1 + # via nbconvert +paramiko==3.4.1 + # via fabric +parso==0.8.4 + # via jedi +pathspec==0.12.1 # via black # via mkdocs -pillow==10.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pexpect==4.9.0 + # via ipython +pillow==10.4.0 # via matplotlib # via torchvision -platformdirs==4.2.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +platformdirs==4.2.2 # via black + # via jupyter-core # via mkdocs-get-deps # via mkdocstrings # via wandb -protobuf==5.27.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +prompt-toolkit==3.0.47 + # via ipython + # via questionary +protobuf==5.27.3 # via orbax-checkpoint # via wandb -psutil==6.0.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +psutil==6.0.0 + # via ipykernel # via wandb -pydantic==2.8.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +ptyprocess==0.7.0 + # via pexpect +pure-eval==0.2.3 + # via stack-data +pycparser==2.22 + # via cffi +pydantic==2.8.2 # via research-project-template -pydantic-core==2.20.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pydantic-core==2.20.1 # via pydantic -pygments==2.18.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pygments==2.18.0 + # via ipython + # via mkdocs-jupyter # via mkdocs-material + # via nbconvert # via rich -pymdown-extensions==10.9 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pymdown-extensions==10.9 # via mkdocs-material # via mkdocstrings -pyparsing==3.1.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pynacl==1.5.0 + # via paramiko +pyparsing==3.1.2 # via matplotlib -pysocks==1.7.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pysocks==1.7.1 # via requests -python-dateutil==2.9.0.post0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +python-dateutil==2.9.0.post0 # via ghp-import + # via jupyter-client # via matplotlib -pytorch-lightning==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +python-dotenv==1.0.1 + # via rootutils +pytorch-lightning==2.4.0 # via lightning -pytorch2jax==0.1.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pytorch2jax==0.1.0 # via torch-jax-interop -pyyaml==6.0.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pyyaml==6.0.2 # via flax + # via jupytext # via lightning # via mkdocs # via mkdocs-get-deps @@ -341,49 +460,69 @@ pyyaml==6.0.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python # via pytorch-lightning # via pyyaml-env-tag # via wandb -pyyaml-env-tag==0.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pyyaml-env-tag==0.1 # via mkdocs -regex==2024.7.24 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +pyzmq==26.2.0 + # via ipykernel + # via jupyter-client +questionary==1.10.0 + # via milatools +referencing==0.35.1 + # via jsonschema + # via jsonschema-specifications +regex==2024.7.24 # via mkdocs-material -requests==2.32.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +requests==2.32.3 # via gdown # via mkdocs-material # via wandb -rich==13.7.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +rich==13.7.1 # via flax + # via milatools # via research-project-template -scipy==1.14.0 ; (python_version >= '3.12' or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +rootutils==1.0.7 + # via research-project-template +rpds-py==0.20.0 + # via jsonschema + # via referencing +scipy==1.14.0 # via jax # via jaxlib -sentry-sdk==2.13.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') - # via wandb -setproctitle==1.3.3 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +sentry-sdk==2.13.0 # via wandb -setuptools==72.2.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.12' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') - # via chex - # via lightning-utilities +setproctitle==1.3.3 # via wandb -simple-parsing==0.1.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +simple-parsing==0.1.5 # via research-project-template -six==1.16.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +six==1.16.0 + # via asttokens + # via bleach + # via blessed # via docker-pycreds # via python-dateutil -smmap==5.0.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +smmap==5.0.1 # via gitdb -soupsieve==2.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +soupsieve==2.6 # via beautifulsoup4 -submitit==1.5.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +sshconf==0.2.7 + # via milatools +stack-data==0.6.3 + # via ipython +submitit==1.5.1 # via hydra-submitit-launcher -sympy==1.13.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +sympy==1.13.2 # via torch -tensorstore==0.1.64 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +tensorstore==0.1.64 # via flax # via orbax-checkpoint -tomli==2.0.1 ; python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux')) +tinycss2==1.3.0 + # via nbconvert +tomli==2.0.1 # via black -toolz==0.12.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') + # via jupytext +toolz==0.12.1 # via chex -torch==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torch==2.4.0 # via lightning # via pytorch-lightning # via pytorch2jax @@ -391,28 +530,44 @@ torch==2.4.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_ # via torch-jax-interop # via torchmetrics # via torchvision -torch-jax-interop==0.0.7 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torch-jax-interop==0.0.7 # via research-project-template -torchmetrics==1.4.1 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torchmetrics==1.4.1 # via lightning # via pytorch-lightning -torchvision==0.19.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +torchvision==0.19.0 # via research-project-template -tqdm==4.66.5 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +tornado==6.4.1 + # via ipykernel + # via jupyter-client +tqdm==4.66.5 # via gdown # via lightning + # via milatools # via pytorch-lightning # via research-project-template -triton==3.0.0 ; python_version < '3.13' and platform_machine == 'x86_64' and platform_system == 'Linux' +traitlets==5.14.3 + # via comm + # via ipykernel + # via ipython + # via jupyter-client + # via jupyter-core + # via matplotlib-inline + # via nbclient + # via nbconvert + # via nbformat +triton==3.0.0 # via torch -typing-extensions==4.12.2 ; (python_version < '3.11' and ((python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux'))) or (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +typing-extensions==4.12.2 # via black # via chex # via etils # via flax # via hydra-zen + # via ipython # via lightning # via lightning-utilities + # via milatools # via orbax-checkpoint # via pydantic # via pydantic-core @@ -420,16 +575,27 @@ typing-extensions==4.12.2 ; (python_version < '3.11' and ((python_version < '3.1 # via simple-parsing # via submitit # via torch -urllib3==2.2.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +urllib3==2.2.2 # via requests # via sentry-sdk -wandb==0.17.6 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +wandb==0.17.6 # via research-project-template -watchdog==4.0.2 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +watchdog==4.0.2 # via mkdocs -wcmatch==9.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +wcmatch==9.0 # via mkdocs-awesome-pages-plugin -yarl==1.9.4 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +wcwidth==0.2.13 + # via blessed + # via prompt-toolkit +webencodings==0.5.1 + # via bleach + # via tinycss2 +wrapt==1.16.0 + # via deprecated +yarl==1.9.4 # via aiohttp -zipp==3.20.0 ; (python_version < '3.13' and sys_platform == 'linux') or (python_version < '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version < '3.13' and sys_platform != 'linux') or (python_version >= '3.13' and sys_platform == 'linux') or (python_version >= '3.13' and sys_platform == 'linux' and (python_version <= '3.9' or sys_platform != 'linux')) or (python_version >= '3.13' and sys_platform != 'linux') +zipp==3.20.0 # via etils +setuptools==72.2.0 + # via lightning-utilities + # via wandb From dde4112452c9bad8c471e91554c5f87338189ccd Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Wed, 4 Sep 2024 09:12:51 -0400 Subject: [PATCH 08/33] pre-commit check --- docs/install.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/install.md b/docs/install.md index 9e7d3605..ffa5e01f 100644 --- a/docs/install.md +++ b/docs/install.md @@ -24,7 +24,7 @@ There are two ways to install this project - If you don't have Conda installed, you can download it from [here](https://docs.conda.io/en/latest/miniconda.html). - If you'd rather use a virtual environment instead of Conda, you can totally do so, as long as you have a version of Python >= 3.12. - 2. On the Mila cluster: + 4. On the Mila cluster: If you're on the `mila` cluster, you can run this setup script (on a *compute* node): From 912bda06a9cb0dbd87efa9b0414cb17792ee2b80 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Wed, 4 Sep 2024 12:24:26 -0400 Subject: [PATCH 09/33] lockfile regen, config update, misc changes to make profiling notebook work --- docs/examples/profiling.ipynb | 47 +++-- project/algorithms/example.py | 5 - project/configs/algorithm/example.yaml | 5 + project/configs/datamodule/imagenet.yaml | 3 - project/configs/experiment/example.yaml | 2 +- project/configs/network/resnet18.yaml | 2 +- .../image_classification/imagenet.py | 1 + requirements-dev.lock | 185 ++++++++++++++++- requirements.lock | 186 +++++++++++++++++- 9 files changed, 404 insertions(+), 32 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 84d99e02..c8f2a3ad 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -18,7 +18,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "The Mila Research Template leverages built-in PyTorch and Lightning functionality to make benchmarking accessible and flexible. " + "The Mila Research Template leverages built-in PyTorch and Lightning functionality to make benchmarking accessible and flexible. \n", + "Make sure to read the Mila Docs page on profiling before going through this example. \n", + "[PLACEHOLDER - Profiling](https://docs.mila.quebec/) " ] }, { @@ -32,7 +34,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "A potential use of .." + "WIP TEXT \n", + "Extending the examples in the documentation, the research template can be utilized in a similar fashion to the examples in the official documentation, albeit with other nifty tools, such as: native WandB integration, iteration using the GPUs available on the official Mila cluster, among other tools. See below." ] }, { @@ -47,7 +50,7 @@ "# Set the working directory to the project root\n", "notebook_path = Path().resolve() \n", "project_root = notebook_path.parent.parent\n", - "os.chdir(str(project_root))\n" + "os.chdir(str(project_root))" ] }, { @@ -56,13 +59,13 @@ "metadata": {}, "outputs": [], "source": [ - "!python project/main.py \\\n", - " algorithm=example \\\n", + "!HYDRA_FULL_ERROR=1 python project/main.py \\\n", + " algorithm=no_op \\\n", " datamodule=imagenet \\\n", - " ++trainer.max_epochs=0 \\\n", - " ++trainer.limit_train_batches=0\\\n", - "\n", - "#!python ../../project/main.py throws an error about relative paths" + " ++logger=wandb \\\n", + " ++trainer.max_epochs=1 \\\n", + " ++trainer.limit_train_batches=30 \\\n", + " ++trainer.limit_val_batches=30" ] }, { @@ -71,12 +74,13 @@ "metadata": {}, "outputs": [], "source": [ - "#!HYDRA_FULL_ERROR=1 python project/main.py \\\n", - "# algorithm=example \\\n", - "# trainer.max_epochs=1 \\\n", - "# +trainer.limit_train_batches=0.01\\\n", - "# +trainer.limit_val_batches=0.01\\\n", - "# datamodule=imagenet" + "!python project/main.py \\\n", + " algorithm=example \\\n", + " datamodule=imagenet \\\n", + " ++logger=wandb \\\n", + " ++trainer.max_epochs=1 \\\n", + " ++trainer.limit_train_batches=30 \\\n", + " ++trainer.limit_val_batches=30" ] }, { @@ -93,7 +97,7 @@ "Using the Mila Research template, it is possible to sweep over different parameters for testing purposes. \n", "For example, suppose we wanted to figure out how different GPUs perform relative to each other. \n", "\n", - "[Mila's official documentation](https://docs.mila.quebec/Information.html) shows which GPUs are installed on the cluster. Typing ```savail``` on the command line shows their current availability. \n", + "[Mila's official documentation](https://docs.mila.quebec/Information.html) has a comprehensive rundown of the GPUs that are installed on the cluster. Typing ```savail``` on the command line, when logged into the cluster, shows their current availability. \n", "Testing their capacity can yield insights into their suitability for different training cases." ] }, @@ -103,14 +107,14 @@ "metadata": {}, "outputs": [], "source": [ - "#!savail" + "!savail" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "We can observe the following prominent GPU classes: a100, a100l, a6000, rtx8000, v100. \n", + "We can observe the following prominent GPU classes: a100, a100l, a6000, rtx8000, v100 and MiG partitions with sizes 2g.20gb, 3g.40gb, 4g.40gb. \n", "We will now proceed to specify different GPUs over training runs and compare their throughput." ] }, @@ -122,8 +126,11 @@ "source": [ "# Add an example of a sweep over some parameters, \n", "# with the training throughput as the metric, \n", - "# :: callbacks/samples_per_second, ### or add a devicestatsmonitor in\n", - "# and using different kinds of GPUs. " + "# :: callbacks/samples_per_second, \n", + "# or add a devicestatsmonitor in\n", + "# and using different kinds of GPUs. \n", + "\n", + "## salloc --gres=gpu:a100:1 -c 6 --mem=32G -t 48:00:00 --partition=unkillable" ] }, { diff --git a/project/algorithms/example.py b/project/algorithms/example.py index 5643ad07..f9a96817 100644 --- a/project/algorithms/example.py +++ b/project/algorithms/example.py @@ -7,14 +7,12 @@ ``` """ -import dataclasses import functools from logging import getLogger from typing import Any, Literal import torch from lightning import LightningModule -from omegaconf import DictConfig from torch import Tensor from torch.nn import functional as F from torch.optim.optimizer import Optimizer @@ -48,9 +46,6 @@ def __init__( self.datamodule = datamodule self.network = network self.optimizer_config = optimizer_config - assert dataclasses.is_dataclass(optimizer_config) or isinstance( - optimizer_config, dict | DictConfig - ), optimizer_config # Used by Pytorch-Lightning to compute the input/output shapes of the network. self.example_input_array = torch.zeros( diff --git a/project/configs/algorithm/example.yaml b/project/configs/algorithm/example.yaml index ab974cf4..e947cabd 100644 --- a/project/configs/algorithm/example.yaml +++ b/project/configs/algorithm/example.yaml @@ -1,2 +1,7 @@ _target_: project.algorithms.example.ExampleAlgorithm _partial_: true + +optimizer_config: + _target_: torch.optim.adam.Adam + _partial_: true + lr: 0.0003 diff --git a/project/configs/datamodule/imagenet.yaml b/project/configs/datamodule/imagenet.yaml index 45b68c6e..23804087 100644 --- a/project/configs/datamodule/imagenet.yaml +++ b/project/configs/datamodule/imagenet.yaml @@ -2,7 +2,4 @@ defaults: - vision - _self_ _target_: project.datamodules.ImageNetDataModule -batch_size: 128 -train_transforms: - _target_: project.datamodules.image_classification.imagenet.imagenet_train_transforms # todo: add good configuration options here. diff --git a/project/configs/experiment/example.yaml b/project/configs/experiment/example.yaml index 73779651..2bd64461 100644 --- a/project/configs/experiment/example.yaml +++ b/project/configs/experiment/example.yaml @@ -9,7 +9,6 @@ defaults: - override /network: resnet18 - override /trainer: default - override /trainer/callbacks: default - - override /trainer/logger: wandb # all parameters below will be merged with parameters from default configurations set above # this allows you to overwrite only specified parameters @@ -30,6 +29,7 @@ trainer: gradient_clip_val: 0.5 logger: wandb: + _target_: lightning.pytorch.loggers.wandb.WandbLogger project: "ResearchTemplate" name: ${oc.env:SLURM_JOB_ID}_${oc.env:SLURM_PROCID} save_dir: "${hydra:runtime.output_dir}" diff --git a/project/configs/network/resnet18.yaml b/project/configs/network/resnet18.yaml index fe4bfc24..1a923ae8 100644 --- a/project/configs/network/resnet18.yaml +++ b/project/configs/network/resnet18.yaml @@ -1,2 +1,2 @@ _target_: torchvision.models.resnet18 -num_classes: ${instance_attr:datamodule.num_classes} +num_classes: ${instance_attr:datamodule.num_classes,1000} diff --git a/project/datamodules/image_classification/imagenet.py b/project/datamodules/image_classification/imagenet.py index b32debec..f00e7278 100644 --- a/project/datamodules/image_classification/imagenet.py +++ b/project/datamodules/image_classification/imagenet.py @@ -115,6 +115,7 @@ def __init__( self.train_kwargs = self.train_kwargs | {"split": "train"} self.valid_kwargs = self.valid_kwargs | {"split": "train"} self.test_kwargs = self.test_kwargs | {"split": "val"} + self.num_classes = type(self).num_classes # self.test_dataset_cls = UnlabeledImagenet def prepare_data(self) -> None: diff --git a/requirements-dev.lock b/requirements-dev.lock index f60258bf..32d7b591 100644 --- a/requirements-dev.lock +++ b/requirements-dev.lock @@ -4,12 +4,13 @@ # last locked with the following flags: # pre: false # features: [] -# all-features: false +# all-features: true # with-sources: false # generate-hashes: false # universal: false -e file:. + # via file:///- absl-py==2.1.0 # via chex # via optax @@ -32,12 +33,23 @@ async-timeout==4.0.3 # via aiohttp attrs==24.2.0 # via aiohttp + # via jsonschema + # via referencing +babel==2.16.0 + # via mkdocs-material bcrypt==4.2.0 # via paramiko beautifulsoup4==4.12.3 # via gdown + # via nbconvert +black==24.8.0 + # via research-project-template +bleach==6.1.0 + # via nbconvert blessed==1.20.0 # via milatools +bracex==2.5 + # via wcmatch certifi==2024.8.30 # via requests # via sentry-sdk @@ -51,9 +63,15 @@ charset-normalizer==3.3.2 chex==0.1.86 # via optax click==8.1.7 + # via black + # via mkdocs + # via mkdocstrings # via wandb cloudpickle==3.0.0 # via submitit +colorama==0.4.6 + # via griffe + # via mkdocs-material colorlog==6.8.2 # via hydra-colorlog comm==0.2.2 @@ -76,6 +94,8 @@ debugpy==1.8.5 decorator==5.1.1 # via fabric # via ipython +defusedxml==0.7.1 + # via nbconvert deprecated==1.2.14 # via fabric dill==0.3.8 @@ -103,6 +123,8 @@ executing==2.1.0 # via stack-data fabric==3.2.2 # via milatools +fastjsonschema==2.20.0 + # via nbformat filelock==3.15.4 # via datasets # via gdown @@ -128,10 +150,14 @@ fsspec==2024.6.1 # via torch gdown==5.2.0 # via research-project-template +ghp-import==2.1.0 + # via mkdocs gitdb==4.0.11 # via gitpython gitpython==3.1.43 # via wandb +griffe==1.2.0 + # via mkdocstrings-python huggingface-hub==0.24.6 # via datasets # via evaluate @@ -162,6 +188,7 @@ iniconfig==2.0.0 invoke==2.2.0 # via fabric ipykernel==6.29.5 + # via mkdocs-jupyter # via research-project-template ipython==8.27.0 # via ipykernel @@ -173,6 +200,11 @@ jax==0.4.31 # via pytorch2jax # via research-project-template # via torch-jax-interop +jax-cuda12-pjrt==0.4.31 + # via jax-cuda12-plugin +jax-cuda12-plugin==0.4.31 + # via jax + # via jax-cuda12-plugin jaxlib==0.4.31 # via chex # via jax @@ -182,14 +214,30 @@ jaxlib==0.4.31 jedi==0.19.1 # via ipython jinja2==3.1.4 + # via mkdocs + # via mkdocs-material + # via mkdocstrings + # via nbconvert # via torch joblib==1.4.2 # via scikit-learn +jsonschema==4.23.0 + # via nbformat +jsonschema-specifications==2023.12.1 + # via jsonschema jupyter-client==8.6.2 # via ipykernel + # via nbclient jupyter-core==5.7.2 # via ipykernel # via jupyter-client + # via nbclient + # via nbconvert + # via nbformat +jupyterlab-pygments==0.3.0 + # via nbconvert +jupytext==1.16.4 + # via mkdocs-jupyter kiwisolver==1.4.6 # via matplotlib lightning==2.4.0 @@ -198,19 +246,79 @@ lightning-utilities==0.11.7 # via lightning # via pytorch-lightning # via torchmetrics +lxml==5.3.0 + # via mkdocs-video +markdown==3.7 + # via mkdocs + # via mkdocs-autorefs + # via mkdocs-material + # via mkdocstrings + # via pymdown-extensions markdown-it-py==3.0.0 + # via jupytext + # via mdit-py-plugins # via rich markupsafe==2.1.5 # via jinja2 + # via mkdocs + # via mkdocs-autorefs + # via mkdocstrings + # via nbconvert matplotlib==3.9.2 # via research-project-template matplotlib-inline==0.1.7 # via ipykernel # via ipython +mdit-py-plugins==0.4.1 + # via jupytext mdurl==0.1.2 # via markdown-it-py +mergedeep==1.3.4 + # via mkdocs + # via mkdocs-get-deps milatools==0.1.5 # via research-project-template +mistune==3.0.2 + # via nbconvert +mkdocs==1.6.1 + # via mkdocs-autorefs + # via mkdocs-awesome-pages-plugin + # via mkdocs-gen-files + # via mkdocs-jupyter + # via mkdocs-literate-nav + # via mkdocs-material + # via mkdocs-section-index + # via mkdocs-video + # via mkdocstrings + # via research-project-template +mkdocs-autorefs==1.2.0 + # via mkdocstrings + # via mkdocstrings-python +mkdocs-awesome-pages-plugin==2.9.3 + # via research-project-template +mkdocs-gen-files==0.5.0 + # via research-project-template +mkdocs-get-deps==0.2.0 + # via mkdocs +mkdocs-jupyter==0.25.0 + # via research-project-template +mkdocs-literate-nav==0.6.1 + # via research-project-template +mkdocs-material==9.5.34 + # via mkdocs-jupyter + # via research-project-template +mkdocs-material-extensions==1.3.1 + # via mkdocs-material +mkdocs-section-index==0.3.9 + # via research-project-template +mkdocs-video==1.5.0 + # via research-project-template +mkdocstrings==0.26.0 + # via mkdocstrings + # via mkdocstrings-python + # via research-project-template +mkdocstrings-python==1.11.1 + # via mkdocstrings mktestdocs==0.2.2 ml-dtypes==0.4.0 # via jax @@ -227,6 +335,18 @@ multidict==6.0.5 multiprocess==0.70.16 # via datasets # via evaluate +mypy-extensions==1.0.0 + # via black +natsort==8.4.0 + # via mkdocs-awesome-pages-plugin +nbclient==0.10.0 + # via nbconvert +nbconvert==7.16.4 + # via mkdocs-jupyter +nbformat==5.10.4 + # via jupytext + # via nbclient + # via nbconvert nest-asyncio==1.6.0 # via ipykernel # via orbax-checkpoint @@ -257,29 +377,40 @@ numpy==1.26.4 # via torchvision # via transformers nvidia-cublas-cu12==12.1.3.1 + # via jax-cuda12-plugin # via nvidia-cudnn-cu12 # via nvidia-cusolver-cu12 # via torch nvidia-cuda-cupti-cu12==12.1.105 + # via jax-cuda12-plugin # via torch +nvidia-cuda-nvcc-cu12==12.6.68 + # via jax-cuda12-plugin nvidia-cuda-nvrtc-cu12==12.1.105 # via torch nvidia-cuda-runtime-cu12==12.1.105 + # via jax-cuda12-plugin # via torch nvidia-cudnn-cu12==9.1.0.70 + # via jax-cuda12-plugin # via torch nvidia-cufft-cu12==11.0.2.54 + # via jax-cuda12-plugin # via torch nvidia-curand-cu12==10.3.2.106 # via torch nvidia-cusolver-cu12==11.4.5.107 + # via jax-cuda12-plugin # via torch nvidia-cusparse-cu12==12.1.0.106 + # via jax-cuda12-plugin # via nvidia-cusolver-cu12 # via torch nvidia-nccl-cu12==2.20.5 + # via jax-cuda12-plugin # via torch nvidia-nvjitlink-cu12==12.6.68 + # via jax-cuda12-plugin # via nvidia-cusolver-cu12 # via nvidia-cusparse-cu12 nvidia-nvtx-cu12==12.1.105 @@ -294,32 +425,46 @@ optax==0.2.3 orbax-checkpoint==0.6.1 # via flax packaging==24.1 + # via black # via datasets # via evaluate # via huggingface-hub # via hydra-core # via ipykernel + # via jupytext # via lightning # via lightning-utilities # via matplotlib + # via mkdocs + # via nbconvert # via pytest # via pytorch-lightning # via torchmetrics # via transformers +paginate==0.5.7 + # via mkdocs-material pandas==2.2.2 # via datasets # via evaluate +pandocfilters==1.5.1 + # via nbconvert paramiko==3.4.1 # via fabric parso==0.8.4 # via jedi +pathspec==0.12.1 + # via black + # via mkdocs pexpect==4.9.0 # via ipython pillow==10.4.0 # via matplotlib # via torchvision platformdirs==4.2.2 + # via black # via jupyter-core + # via mkdocs-get-deps + # via mkdocstrings # via virtualenv # via wandb pluggy==1.5.0 @@ -350,7 +495,13 @@ pydantic-core==2.20.1 # via pydantic pygments==2.18.0 # via ipython + # via mkdocs-jupyter + # via mkdocs-material + # via nbconvert # via rich +pymdown-extensions==10.9 + # via mkdocs-material + # via mkdocstrings pynacl==1.5.0 # via paramiko pyparsing==3.1.4 @@ -379,6 +530,7 @@ pytest-testmon==2.1.1 pytest-timeout==2.3.1 pytest-xdist==3.6.1 python-dateutil==2.9.0.post0 + # via ghp-import # via jupyter-client # via matplotlib # via pandas @@ -392,32 +544,47 @@ pyyaml==6.0.2 # via datasets # via flax # via huggingface-hub + # via jupytext # via lightning + # via mkdocs + # via mkdocs-get-deps # via omegaconf # via orbax-checkpoint # via pre-commit + # via pymdown-extensions # via pytest-regressions # via pytorch-lightning + # via pyyaml-env-tag # via transformers # via wandb +pyyaml-env-tag==0.1 + # via mkdocs pyzmq==26.2.0 # via ipykernel # via jupyter-client questionary==1.10.0 # via milatools +referencing==0.35.1 + # via jsonschema + # via jsonschema-specifications regex==2024.7.24 + # via mkdocs-material # via transformers requests==2.32.3 # via datasets # via evaluate # via gdown # via huggingface-hub + # via mkdocs-material # via transformers # via wandb rich==13.8.0 # via flax # via milatools # via research-project-template +rpds-py==0.20.0 + # via jsonschema + # via referencing ruff==0.6.3 safetensors==0.4.4 # via transformers @@ -435,6 +602,7 @@ simple-parsing==0.1.5 # via research-project-template six==1.16.0 # via asttokens + # via bleach # via blessed # via docker-pycreds # via python-dateutil @@ -456,10 +624,14 @@ tensorstore==0.1.64 # via orbax-checkpoint threadpoolctl==3.5.0 # via scikit-learn +tinycss2==1.3.0 + # via nbconvert tokenizers==0.19.1 # via transformers tomli==2.0.1 + # via black # via coverage + # via jupytext # via pytest # via pytest-env toolz==0.12.1 @@ -500,11 +672,15 @@ traitlets==5.14.3 # via jupyter-client # via jupyter-core # via matplotlib-inline + # via nbclient + # via nbconvert + # via nbformat transformers==4.44.2 # via research-project-template triton==3.0.0 # via torch typing-extensions==4.12.2 + # via black # via chex # via etils # via flax @@ -530,9 +706,16 @@ virtualenv==20.26.3 # via pre-commit wandb==0.17.8 # via research-project-template +watchdog==5.0.2 + # via mkdocs +wcmatch==9.0 + # via mkdocs-awesome-pages-plugin wcwidth==0.2.13 # via blessed # via prompt-toolkit +webencodings==0.5.1 + # via bleach + # via tinycss2 wrapt==1.16.0 # via deprecated xxhash==3.5.0 diff --git a/requirements.lock b/requirements.lock index ca777ea9..59733b51 100644 --- a/requirements.lock +++ b/requirements.lock @@ -4,12 +4,13 @@ # last locked with the following flags: # pre: false # features: [] -# all-features: false +# all-features: true # with-sources: false # generate-hashes: false # universal: false -e file:. + # via file:///- absl-py==2.1.0 # via chex # via optax @@ -32,12 +33,23 @@ async-timeout==4.0.3 # via aiohttp attrs==24.2.0 # via aiohttp + # via jsonschema + # via referencing +babel==2.16.0 + # via mkdocs-material bcrypt==4.2.0 # via paramiko beautifulsoup4==4.12.3 # via gdown + # via nbconvert +black==24.8.0 + # via research-project-template +bleach==6.1.0 + # via nbconvert blessed==1.20.0 # via milatools +bracex==2.5 + # via wcmatch certifi==2024.8.30 # via requests # via sentry-sdk @@ -49,9 +61,15 @@ charset-normalizer==3.3.2 chex==0.1.86 # via optax click==8.1.7 + # via black + # via mkdocs + # via mkdocstrings # via wandb cloudpickle==3.0.0 # via submitit +colorama==0.4.6 + # via griffe + # via mkdocs-material colorlog==6.8.2 # via hydra-colorlog comm==0.2.2 @@ -70,6 +88,8 @@ debugpy==1.8.5 decorator==5.1.1 # via fabric # via ipython +defusedxml==0.7.1 + # via nbconvert deprecated==1.2.14 # via fabric dill==0.3.8 @@ -92,6 +112,8 @@ executing==2.1.0 # via stack-data fabric==3.2.2 # via milatools +fastjsonschema==2.20.0 + # via nbformat filelock==3.15.4 # via datasets # via gdown @@ -116,10 +138,14 @@ fsspec==2024.6.1 # via torch gdown==5.2.0 # via research-project-template +ghp-import==2.1.0 + # via mkdocs gitdb==4.0.11 # via gitpython gitpython==3.1.43 # via wandb +griffe==1.2.0 + # via mkdocstrings-python huggingface-hub==0.24.6 # via datasets # via evaluate @@ -146,6 +172,7 @@ importlib-resources==6.4.4 invoke==2.2.0 # via fabric ipykernel==6.29.5 + # via mkdocs-jupyter # via research-project-template ipython==8.27.0 # via ipykernel @@ -157,6 +184,11 @@ jax==0.4.31 # via pytorch2jax # via research-project-template # via torch-jax-interop +jax-cuda12-pjrt==0.4.31 + # via jax-cuda12-plugin +jax-cuda12-plugin==0.4.31 + # via jax + # via jax-cuda12-plugin jaxlib==0.4.31 # via chex # via jax @@ -166,14 +198,30 @@ jaxlib==0.4.31 jedi==0.19.1 # via ipython jinja2==3.1.4 + # via mkdocs + # via mkdocs-material + # via mkdocstrings + # via nbconvert # via torch joblib==1.4.2 # via scikit-learn +jsonschema==4.23.0 + # via nbformat +jsonschema-specifications==2023.12.1 + # via jsonschema jupyter-client==8.6.2 # via ipykernel + # via nbclient jupyter-core==5.7.2 # via ipykernel # via jupyter-client + # via nbclient + # via nbconvert + # via nbformat +jupyterlab-pygments==0.3.0 + # via nbconvert +jupytext==1.16.4 + # via mkdocs-jupyter kiwisolver==1.4.6 # via matplotlib lightning==2.4.0 @@ -182,19 +230,79 @@ lightning-utilities==0.11.7 # via lightning # via pytorch-lightning # via torchmetrics +lxml==5.3.0 + # via mkdocs-video +markdown==3.7 + # via mkdocs + # via mkdocs-autorefs + # via mkdocs-material + # via mkdocstrings + # via pymdown-extensions markdown-it-py==3.0.0 + # via jupytext + # via mdit-py-plugins # via rich markupsafe==2.1.5 # via jinja2 + # via mkdocs + # via mkdocs-autorefs + # via mkdocstrings + # via nbconvert matplotlib==3.9.2 # via research-project-template matplotlib-inline==0.1.7 # via ipykernel # via ipython +mdit-py-plugins==0.4.1 + # via jupytext mdurl==0.1.2 # via markdown-it-py +mergedeep==1.3.4 + # via mkdocs + # via mkdocs-get-deps milatools==0.1.5 # via research-project-template +mistune==3.0.2 + # via nbconvert +mkdocs==1.6.1 + # via mkdocs-autorefs + # via mkdocs-awesome-pages-plugin + # via mkdocs-gen-files + # via mkdocs-jupyter + # via mkdocs-literate-nav + # via mkdocs-material + # via mkdocs-section-index + # via mkdocs-video + # via mkdocstrings + # via research-project-template +mkdocs-autorefs==1.2.0 + # via mkdocstrings + # via mkdocstrings-python +mkdocs-awesome-pages-plugin==2.9.3 + # via research-project-template +mkdocs-gen-files==0.5.0 + # via research-project-template +mkdocs-get-deps==0.2.0 + # via mkdocs +mkdocs-jupyter==0.25.0 + # via research-project-template +mkdocs-literate-nav==0.6.1 + # via research-project-template +mkdocs-material==9.5.34 + # via mkdocs-jupyter + # via research-project-template +mkdocs-material-extensions==1.3.1 + # via mkdocs-material +mkdocs-section-index==0.3.9 + # via research-project-template +mkdocs-video==1.5.0 + # via research-project-template +mkdocstrings==0.26.0 + # via mkdocstrings + # via mkdocstrings-python + # via research-project-template +mkdocstrings-python==1.11.1 + # via mkdocstrings ml-dtypes==0.4.0 # via jax # via jaxlib @@ -210,6 +318,18 @@ multidict==6.0.5 multiprocess==0.70.16 # via datasets # via evaluate +mypy-extensions==1.0.0 + # via black +natsort==8.4.0 + # via mkdocs-awesome-pages-plugin +nbclient==0.10.0 + # via nbconvert +nbconvert==7.16.4 + # via mkdocs-jupyter +nbformat==5.10.4 + # via jupytext + # via nbclient + # via nbconvert nest-asyncio==1.6.0 # via ipykernel # via orbax-checkpoint @@ -237,29 +357,40 @@ numpy==2.1.1 # via torchvision # via transformers nvidia-cublas-cu12==12.1.3.1 + # via jax-cuda12-plugin # via nvidia-cudnn-cu12 # via nvidia-cusolver-cu12 # via torch nvidia-cuda-cupti-cu12==12.1.105 + # via jax-cuda12-plugin # via torch +nvidia-cuda-nvcc-cu12==12.6.68 + # via jax-cuda12-plugin nvidia-cuda-nvrtc-cu12==12.1.105 # via torch nvidia-cuda-runtime-cu12==12.1.105 + # via jax-cuda12-plugin # via torch nvidia-cudnn-cu12==9.1.0.70 + # via jax-cuda12-plugin # via torch nvidia-cufft-cu12==11.0.2.54 + # via jax-cuda12-plugin # via torch nvidia-curand-cu12==10.3.2.106 # via torch nvidia-cusolver-cu12==11.4.5.107 + # via jax-cuda12-plugin # via torch nvidia-cusparse-cu12==12.1.0.106 + # via jax-cuda12-plugin # via nvidia-cusolver-cu12 # via torch nvidia-nccl-cu12==2.20.5 + # via jax-cuda12-plugin # via torch nvidia-nvjitlink-cu12==12.6.68 + # via jax-cuda12-plugin # via nvidia-cusolver-cu12 # via nvidia-cusparse-cu12 nvidia-nvtx-cu12==12.1.105 @@ -274,31 +405,45 @@ optax==0.2.3 orbax-checkpoint==0.6.1 # via flax packaging==24.1 + # via black # via datasets # via evaluate # via huggingface-hub # via hydra-core # via ipykernel + # via jupytext # via lightning # via lightning-utilities # via matplotlib + # via mkdocs + # via nbconvert # via pytorch-lightning # via torchmetrics # via transformers +paginate==0.5.7 + # via mkdocs-material pandas==2.2.2 # via datasets # via evaluate +pandocfilters==1.5.1 + # via nbconvert paramiko==3.4.1 # via fabric parso==0.8.4 # via jedi +pathspec==0.12.1 + # via black + # via mkdocs pexpect==4.9.0 # via ipython pillow==10.4.0 # via matplotlib # via torchvision platformdirs==4.2.2 + # via black # via jupyter-core + # via mkdocs-get-deps + # via mkdocstrings # via wandb prompt-toolkit==3.0.47 # via ipython @@ -323,7 +468,13 @@ pydantic-core==2.20.1 # via pydantic pygments==2.18.0 # via ipython + # via mkdocs-jupyter + # via mkdocs-material + # via nbconvert # via rich +pymdown-extensions==10.9 + # via mkdocs-material + # via mkdocstrings pynacl==1.5.0 # via paramiko pyparsing==3.1.4 @@ -331,6 +482,7 @@ pyparsing==3.1.4 pysocks==1.7.1 # via requests python-dateutil==2.9.0.post0 + # via ghp-import # via jupyter-client # via matplotlib # via pandas @@ -344,30 +496,45 @@ pyyaml==6.0.2 # via datasets # via flax # via huggingface-hub + # via jupytext # via lightning + # via mkdocs + # via mkdocs-get-deps # via omegaconf # via orbax-checkpoint + # via pymdown-extensions # via pytorch-lightning + # via pyyaml-env-tag # via transformers # via wandb +pyyaml-env-tag==0.1 + # via mkdocs pyzmq==26.2.0 # via ipykernel # via jupyter-client questionary==1.10.0 # via milatools +referencing==0.35.1 + # via jsonschema + # via jsonschema-specifications regex==2024.7.24 + # via mkdocs-material # via transformers requests==2.32.3 # via datasets # via evaluate # via gdown # via huggingface-hub + # via mkdocs-material # via transformers # via wandb rich==13.8.0 # via flax # via milatools # via research-project-template +rpds-py==0.20.0 + # via jsonschema + # via referencing safetensors==0.4.4 # via transformers scikit-learn==1.5.1 @@ -384,6 +551,7 @@ simple-parsing==0.1.5 # via research-project-template six==1.16.0 # via asttokens + # via bleach # via blessed # via docker-pycreds # via python-dateutil @@ -404,8 +572,13 @@ tensorstore==0.1.64 # via orbax-checkpoint threadpoolctl==3.5.0 # via scikit-learn +tinycss2==1.3.0 + # via nbconvert tokenizers==0.19.1 # via transformers +tomli==2.0.1 + # via black + # via jupytext toolz==0.12.1 # via chex torch==2.4.0 @@ -443,11 +616,15 @@ traitlets==5.14.3 # via jupyter-client # via jupyter-core # via matplotlib-inline + # via nbclient + # via nbconvert + # via nbformat transformers==4.44.2 # via research-project-template triton==3.0.0 # via torch typing-extensions==4.12.2 + # via black # via chex # via etils # via flax @@ -471,9 +648,16 @@ urllib3==2.2.2 # via sentry-sdk wandb==0.17.8 # via research-project-template +watchdog==5.0.2 + # via mkdocs +wcmatch==9.0 + # via mkdocs-awesome-pages-plugin wcwidth==0.2.13 # via blessed # via prompt-toolkit +webencodings==0.5.1 + # via bleach + # via tinycss2 wrapt==1.16.0 # via deprecated xxhash==3.5.0 From 303e6d9af4b8d72e1264b1c7d8bd4df87455c66a Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Tue, 10 Sep 2024 13:25:44 -0400 Subject: [PATCH 10/33] precommit exclusions, more WIP text --- .pre-commit-config.yaml | 1 + docs/examples/profiling.ipynb | 165 ++++++++++++++++++++++++++++------ 2 files changed, 140 insertions(+), 26 deletions(-) diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml index 0cf78069..aeac45c8 100644 --- a/.pre-commit-config.yaml +++ b/.pre-commit-config.yaml @@ -60,6 +60,7 @@ repos: rev: 0.7.1 hooks: - id: nbstripout + exclude: profiling.ipynb require_serial: true # md formatting diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index c8f2a3ad..379d3cfa 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -7,6 +7,17 @@ "# Profiling your code" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The Mila Research Template leverages built-in PyTorch and Lightning functionality to make model profiling and benchmarking accessible and flexible. \n", + "Make sure to read the Mila Docs page on profiling before going through this example. \n", + "[PLACEHOLDER - Profiling](https://docs.mila.quebec/) . \n", + "\n", + "The research template profiling notebook extends the examples in the official documentation with additional tools: notably, native WandB integration to monitor performance and using hydra multiruns to compare the available GPUs on the official Mila cluster. See below. The goal of this notebook is to introduce profiling, present tools useful for doing so and to provide general concepts and guidelines for optimizing your code, within the Mila cluster ecosystem.\n" + ] + }, { "cell_type": "markdown", "metadata": {}, @@ -14,28 +25,55 @@ "### Setup" ] }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "from pathlib import Path\n", + "# Set the working directory to the project root\n", + "notebook_path = Path().resolve() \n", + "project_root = notebook_path.parent.parent\n", + "os.chdir(str(project_root))" + ] + }, { "cell_type": "markdown", "metadata": {}, "source": [ - "The Mila Research Template leverages built-in PyTorch and Lightning functionality to make benchmarking accessible and flexible. \n", - "Make sure to read the Mila Docs page on profiling before going through this example. \n", - "[PLACEHOLDER - Profiling](https://docs.mila.quebec/) " + "## Introduction" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "As a deep learning researcher, training comparatively slow models as opposed to faster, optimized ones can greatly influence your research output. In addition, being a user of a shared cluster, being efficient about the use of institutional resources is a benefit to all the users in the ecosystem. Given the ample variety of available resources and training schemes to achieve the same modeling objective, optimizing your code isn't necessarily a straightforward task. \n", + "\n", + "While there's many costs involved in getting a model to train, some are more relevant than others when it comes to making your code more efficient. Setting a performance baseline, by observing said costs and identifying underperforming components in the code while properly contextualizing them within a broader training scheme is the very first step to optimizing your code. Once a baseline performance expectation is set, we can modify and observe our code's performance in a comparative manner to then determine if the performed optimizations are better. A profiler can help us in this endeavor." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## What is a profiler and what is it good for?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## Finding bottlenecks: Dataloading vs Training" + "A profiler is a tool that allows you to measure the time and memory consumption of the model’s operators. Specifically, the PyTorch profiler output provides clues about operations relevant to model training. Examples include the total amount of time spent doing low-level mathematical operations in the GPU, and whether these are unexpectedly slow or take a disproportionate amount of time, indicating they should be avoided or optimized. Identifying problematic operations can greatly help us validate or rethink our baseline model performance expectations.\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "WIP TEXT \n", - "Extending the examples in the documentation, the research template can be utilized in a similar fashion to the examples in the official documentation, albeit with other nifty tools, such as: native WandB integration, iteration using the GPUs available on the official Mila cluster, among other tools. See below." + "## Setting baseline model performance expectations " ] }, { @@ -44,28 +82,73 @@ "metadata": {}, "outputs": [], "source": [ - "import os\n", - "from pathlib import Path\n", + "# model size\n", + "# https://discuss.pytorch.org/t/finding-model-size/130275/2\n", + "model = ???\n", + "param_size = 0\n", + "for param in model.parameters():\n", + " param_size += param.nelement() * param.element_size()\n", + "buffer_size = 0\n", + "for buffer in model.buffers():\n", + " buffer_size += buffer.nelement() * buffer.element_size()\n", "\n", - "# Set the working directory to the project root\n", - "notebook_path = Path().resolve() \n", - "project_root = notebook_path.parent.parent\n", - "os.chdir(str(project_root))" + "size_all_mb = (param_size + buffer_size) / 1024**2\n", + "print('model size: {:.3f}MB'.format(size_all_mb))\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "MBU = \n", + "\\frac{\\# \\text{ Params} \\cdot \\text{bytes per param} \\cdot \\text{tokens per second}}{\\text{Memory Bandwidth}}\n", + "$$\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Identifying potential bottleneck sources " + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Finding a bottleneck is not necessarily straightforward or clear from the start. A sensible first step is to determine whether a potential slowdown originates from data loading or model computation. Querying both ends of the process can be done to determine whether the master process has a significant stall when fetching the next batch, or not. \n", + "If it's close to 0, then data loading outpaces compute, and compute is the bottleneck. \n", + "If it's much greater than 0, then compute outpaces data loading, and data loading is the bottleneck. \n", + "You might not care about CPU usage by the master process and data loaders, so long as the GPU remains fully utilized. \n", + "Nonetheless, a profiler may record that anyways. How do you look out for relevant stuff? Here are a few ideas.\n" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 6, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "LexerNoViableAltException: \\\n", + " ^\n", + "See https://hydra.cc/docs/1.2/advanced/override_grammar/basic for details\n", + "\n", + "Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.\n" + ] + } + ], "source": [ - "!HYDRA_FULL_ERROR=1 python project/main.py \\\n", + "!python project/main.py \\\n", " algorithm=no_op \\\n", " datamodule=imagenet \\\n", - " ++logger=wandb \\\n", + " ++hydra=profiling_multirun \\\n", " ++trainer.max_epochs=1 \\\n", " ++trainer.limit_train_batches=30 \\\n", - " ++trainer.limit_val_batches=30" + " ++trainer.limit_val_batches=30 \\\n" ] }, { @@ -94,22 +177,43 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Using the Mila Research template, it is possible to sweep over different parameters for testing purposes. \n", - "For example, suppose we wanted to figure out how different GPUs perform relative to each other. \n", - "\n", - "[Mila's official documentation](https://docs.mila.quebec/Information.html) has a comprehensive rundown of the GPUs that are installed on the cluster. Typing ```savail``` on the command line, when logged into the cluster, shows their current availability. \n", - "Testing their capacity can yield insights into their suitability for different training cases." + "As the Mila Research template is built with hydra as a configuration manager, it integrates [Multi-runs](https://hydra.cc/docs/tutorials/basic/running_your_app/multi-run/) by default. This makes it possible to sweep over different parameters for profiling or throughput testing purposes or both. For example, suppose we wanted to figure out how different GPUs perform relative to each other. \n", + "[Mila's official documentation](https://docs.mila.quebec/Information.html) has a comprehensive rundown of the GPUs that are installed on the cluster. Typing ```savail``` on the command line, when logged into the cluster, shows their current availability. Testing their capacity can yield insights into their suitability for different training cases." ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 1, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "GPU Avail / Total \n", + "===============================\n", + "2g.20gb 12 / 48 \n", + "3g.40gb 0 / 48 \n", + "4g.40gb 0 / 24 \n", + "a100 0 / 32 \n", + "a100l 0 / 88 \n", + "a6000 0 / 8 \n", + "rtx8000 8 / 408 \n", + "v100 2 / 56 \n" + ] + } + ], "source": [ "!savail" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "As these jobs are part of the cluster, [Submitit](https://hydra.cc/docs/plugins/submitit_launcher/)" + ] + }, { "cell_type": "markdown", "metadata": {}, @@ -164,7 +268,15 @@ "source": [ "# Create a wandb report with the throughput comparison \n", "# between the different GPU types.\n", - "# i.e. specify wandb as the logger and log the throughput" + "# i.e. specify wandb as the logger and log the throughput\n", + "!python project/main.py \\\n", + " algorithm=no_op \\\n", + " datamodule=imagenet \\\n", + " ++logger=wandb \\\n", + " ++trainer.max_epochs=1 \\\n", + " ++trainer.limit_train_batches=30 \\\n", + " ++trainer.limit_val_batches=30 \\\n", + " hydra=profiling_multirun" ] }, { @@ -221,7 +333,8 @@ "\n", "[GPU Training (Basic) - LightningAI](https://lightning.ai/docs/pytorch/stable/accelerators/gpu_basic.html) \n", "[DeviceStatsMonitor class - LightningAI](https://lightning.ai/docs/pytorch/stable/api/lightning.pytorch.callbacks.DeviceStatsMonitor.html) \n", - "[PyTorch Profiler + W&B integration - Weights & Biases](https://wandb.ai/wandb/trace/reports/Using-the-PyTorch-Profiler-with-W-B--Vmlldzo5MDE3NjU)" + "[PyTorch Profiler + W&B integration - Weights & Biases](https://wandb.ai/wandb/trace/reports/Using-the-PyTorch-Profiler-with-W-B--Vmlldzo5MDE3NjU) \n", + "[Advanced profiling for model optimization - Accelerating Generative AI with PyTorch: Segment Anything, Fast](https://pytorch.org/blog/accelerating-generative-ai/)" ] } ], From 787ddf0a533c54346250f534a81bbcddbf492da7 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Wed, 11 Sep 2024 14:55:03 -0400 Subject: [PATCH 11/33] profiling nb progress, new WIP configs for pending throughput multirun --- docs/examples/profiling.ipynb | 2096 ++++++++++++++++- project/configs/algorithm/no_op.yaml | 2 + .../configs/hydra/profiling_cpu_vs_gpu.yaml | 23 + .../configs/hydra/profiling_gpu_multirun.yaml | 23 + project/configs/resources/cpu.yaml | 13 + project/configs/trainer/cpu.yaml | 6 - project/configs/trainer/profiling.yaml | 4 + pyproject.toml | 2 + requirements-dev.lock | 116 + requirements.lock | 116 + 10 files changed, 2321 insertions(+), 80 deletions(-) create mode 100644 project/configs/algorithm/no_op.yaml create mode 100644 project/configs/hydra/profiling_cpu_vs_gpu.yaml create mode 100644 project/configs/hydra/profiling_gpu_multirun.yaml create mode 100644 project/configs/resources/cpu.yaml delete mode 100644 project/configs/trainer/cpu.yaml create mode 100644 project/configs/trainer/profiling.yaml diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 379d3cfa..99a3321a 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -12,8 +12,7 @@ "metadata": {}, "source": [ "The Mila Research Template leverages built-in PyTorch and Lightning functionality to make model profiling and benchmarking accessible and flexible. \n", - "Make sure to read the Mila Docs page on profiling before going through this example. \n", - "[PLACEHOLDER - Profiling](https://docs.mila.quebec/) . \n", + "Make sure to read the Mila Docs page on [PLACEHOLDER - profiling](https://docs.mila.quebec/) before going through this example. \n", "\n", "The research template profiling notebook extends the examples in the official documentation with additional tools: notably, native WandB integration to monitor performance and using hydra multiruns to compare the available GPUs on the official Mila cluster. See below. The goal of this notebook is to introduce profiling, present tools useful for doing so and to provide general concepts and guidelines for optimizing your code, within the Mila cluster ecosystem.\n" ] @@ -27,7 +26,7 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 1, "metadata": {}, "outputs": [], "source": [ @@ -50,60 +49,1033 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "As a deep learning researcher, training comparatively slow models as opposed to faster, optimized ones can greatly influence your research output. In addition, being a user of a shared cluster, being efficient about the use of institutional resources is a benefit to all the users in the ecosystem. Given the ample variety of available resources and training schemes to achieve the same modeling objective, optimizing your code isn't necessarily a straightforward task. \n", + "As a deep learning researcher, training comparatively slow models as opposed to faster, optimized ones can greatly impact your research output. In addition, as a user of a shared cluster, being efficient about the use of institutional resources is a benefit to all the users in the ecosystem. Given the ample variety of available resources and training schemes to achieve the same modeling objective, optimizing your code isn't necessarily a straightforward task. \n", "\n", - "While there's many costs involved in getting a model to train, some are more relevant than others when it comes to making your code more efficient. Setting a performance baseline, by observing said costs and identifying underperforming components in the code while properly contextualizing them within a broader training scheme is the very first step to optimizing your code. Once a baseline performance expectation is set, we can modify and observe our code's performance in a comparative manner to then determine if the performed optimizations are better. A profiler can help us in this endeavor." + "While there's many costs involved in getting a model to train, some are more relevant than others when it comes to making your code more efficient. Setting a performance baseline, by observing said costs and identifying underperforming components in the code while properly contextualizing them within a broader training scheme is the very first step to optimizing your code. Once a baseline performance expectation is set, we can modify and observe our code's performance in a comparative manner to then determine if the performed optimizations are better." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## What is a profiler and what is it good for?" + "## Instrumenting your code" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "A profiler is a tool that allows you to measure the time and memory consumption of the model’s operators. Specifically, the PyTorch profiler output provides clues about operations relevant to model training. Examples include the total amount of time spent doing low-level mathematical operations in the GPU, and whether these are unexpectedly slow or take a disproportionate amount of time, indicating they should be avoided or optimized. Identifying problematic operations can greatly help us validate or rethink our baseline model performance expectations.\n" + "Setting up artifacts within your code to monitor metrics of interest can help set a cost baseline and evidence potential areas for improvement. Common metrics to watch for include but are not limited to:\n", + " \n", + "- Training speed (samples/s)\n", + "- CPU/GPU utilization \n", + "- RAM/VRAM utilization\n", + "\n", + "In the Mila ResearchTemplate, this can be done by passing a callback to the trainer. Supported configs are found within the project template at `configs/trainer/callbacks`. Here, we will use the default callback, which in turn implements early stopping and tracks the learning rate, device utilisation and throughput, each through a specific callback instance." + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/11/24 14:43:09]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=107891;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=976209;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m .yaml last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/11/24 14:43:10]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=289962;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=183361;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m not find \u001b[32m'hydra/sweeper/orion'\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Available options in \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'hydra/sweeper'\u001b[0m: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m basic \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Config search path: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra.conf\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mmain\u001b[0m, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mproject.configs\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m-colorlog, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra_plugins.hydra_c\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=482159;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=10099;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for resources/cpu.yaml \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=628064;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=886454;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m config resources/cpu.yaml: In \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'resources/cpu'\u001b[0m: Could not \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m override \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'resources/hydra/launcher'\u001b[0m. No \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m match in the defaults list. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=832035;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=182042;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m hydra/profiling_gpu_multirun.yam \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m l last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=843502;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=121431;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m hydra/profiling_gpu_multirun.yam \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m l: In \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'hydra/profiling_gpu_multirun'\u001b[0m: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Could not override \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'hydra/hydra/launcher'\u001b[0m. No match \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m in the defaults list. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=110377;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=270279;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m hydra/profiling_cpu_vs_gpu.yaml \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=710479;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=254129;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m hydra/profiling_cpu_vs_gpu.yaml: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m In \u001b[32m'hydra/profiling_cpu_vs_gpu'\u001b[0m: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Could not override \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'hydra/hydra/launcher'\u001b[0m. No match \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m in the defaults list. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m46/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m58 \u001b[0m ]\n", + " \u001b[31mit/s\u001b[0m \n", + "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/11/24 14:43:11]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=331996;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=767802;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=271388;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=490691;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", + "seed manually set to 8214\n", + "Seed set to 8214\n", + "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", + "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", + "GPU available: True (cuda), used: True\n", + "TPU available: False, using: 0 TPU cores\n", + "HPU available: False, using: 0 HPUs\n", + "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/pytorch/trainer/connectors/logger_connector/logger_connector.py:75: Starting from v1.9.0, `tensorboardX` has been removed as a dependency of the `lightning.pytorch` package, due to potential conflicts with other packages in the ML ecosystem. For this reason, `logger=True` will use `CSVLogger` as the default logger, unless the `tensorboard` or `tensorboardX` packages are found. Please `pip install lightning[extra]` or one of them to enable TensorBoard support by default\n", + "\u001b[2;36m[14:43:12]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=669543;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=940579;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7f456e514340\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", + "You are using a CUDA device ('NVIDIA A100-SXM4-80GB MIG 3g.40gb') that has Tensor Cores. To properly utilize them, you should set `torch.set_float32_matmul_precision('medium' | 'high')` which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=633055;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=868539;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m[14:43:13]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=809000;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=571941;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", + "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", + "┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━┩\n", + "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNorm2d │ 128 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequential │ 147 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequential │ 525 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequential │ 2.1 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequential │ 8.4 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ AdaptiveAvgPool2d │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\n", + "└────┴─────────────────┴───────────────────┴────────┴───────┘\n", + "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", + "\u001b[1mNon-trainable params\u001b[0m: 0 \n", + "\u001b[1mTotal params\u001b[0m: 11.7 M \n", + "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", + "\u001b[1mModules in train mode\u001b[0m: 68 \n", + "\u001b[1mModules in eval mode\u001b[0m: 0 \n", + "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m3.39it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", + "s/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (30) \n", + "is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower \n", + "value for log_every_n_steps if you want to see logs for the training epoch.\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m217.574 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m217.574 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:10\u001b[0m \u001b[37m2.98it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m \u001b[37mv_num: 20.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:08\u001b[0m \u001b[37m3.29it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:08\u001b[0m \u001b[37m3.29it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:01 • 0:00:08\u001b[0m \u001b[37m3.41it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:01 • 0:00:08\u001b[0m \u001b[37m3.41it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.52it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.52it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.59it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.59it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:02 • 0:00:06\u001b[0m \u001b[37m3.60it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:02 • 0:00:06\u001b[0m \u001b[37m3.60it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.71it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.71it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.76it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.76it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.77it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.77it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.05it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.05it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.10it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.10it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.12it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.12it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.14it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.14it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.16it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.16it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.18it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.18it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.26it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.26it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.29it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.29it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.33it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.33it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m239.891 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m266.122 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m222.725 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m208.827 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.348 \u001b[0m\n", + "\u001b[?25hThe following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", + "\u001b[2;36m[14:43:46]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=931799;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=729770;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=52378;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=369808;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.37it/s\u001b[0m \u001b[37m3.37it/s\u001b[0m \n", + "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", + "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", + "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.48725733160972595 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 216.92962646484375 \u001b[0m\u001b[35m \u001b[0m│\n", + "└──────────────────────────────┴──────────────────────────────┘\n", + "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.37it/s\u001b[0m \n", + "\u001b[?25h\u001b[2;36m[14:44:06]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=535481;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=180588;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#165\u001b\\\u001b[2m165\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m loss metric. \u001b[2m \u001b[0m\n", + "val val/samples_per_second_epoch: \u001b[1;36m216.92962646484375\u001b[0m\n" + ] + } + ], + "source": [ + "!python project/main.py \\\n", + " algorithm=no_op \\\n", + " datamodule=imagenet \\\n", + " trainer=profiling \\\n", + " trainer/callbacks=default" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Optional: log metrics on wandb" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## Setting baseline model performance expectations " + "In addition to callback specification, the Mila Research template integrates wandb as a logger specification, which has enables the tracking of additional metrics through visualizations and dashboard creation." ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 15, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/11/24 14:45:32]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=182872;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=964435;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m .yaml last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=108986;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=875791;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m not find \u001b[32m'hydra/sweeper/orion'\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Available options in \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'hydra/sweeper'\u001b[0m: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m basic \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Config search path: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra.conf\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mmain\u001b[0m, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mproject.configs\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m-colorlog, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra_plugins.hydra_c\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=346622;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=662566;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for resources/cpu.yaml \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=272758;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=311553;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m config resources/cpu.yaml: In \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'resources/cpu'\u001b[0m: Could not \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m override \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'resources/hydra/launcher'\u001b[0m. No \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m match in the defaults list. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=493766;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=805763;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m hydra/profiling_gpu_multirun.yam \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m l last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=610702;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=982945;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m hydra/profiling_gpu_multirun.yam \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m l: In \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'hydra/profiling_gpu_multirun'\u001b[0m: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Could not override \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'hydra/hydra/launcher'\u001b[0m. No match \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m in the defaults list. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=534702;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=374156;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m hydra/profiling_cpu_vs_gpu.yaml \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=260404;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=549776;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m hydra/profiling_cpu_vs_gpu.yaml: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m In \u001b[32m'hydra/profiling_cpu_vs_gpu'\u001b[0m: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Could not override \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'hydra/hydra/launcher'\u001b[0m. No match \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m in the defaults list. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m46/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m87 \u001b[0m ]\n", + " \u001b[31mit/s\u001b[0m \n", + "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=880941;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=76069;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=979499;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=716702;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", + "seed manually set to 73525\n", + "Seed set to 73525\n", + "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", + "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", + "GPU available: True (cuda), used: True\n", + "TPU available: False, using: 0 TPU cores\n", + "HPU available: False, using: 0 HPUs\n", + "\u001b[2;36m[14:45:33]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=615378;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=155209;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7f4c95712470\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", + "You are using a CUDA device ('NVIDIA A100-SXM4-80GB MIG 3g.40gb') that has Tensor Cores. To properly utilize them, you should set `torch.set_float32_matmul_precision('medium' | 'high')` which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=978903;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=1269;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m[14:45:34]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=5270;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=603572;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mcesar-valdez\u001b[0m (\u001b[33mcesar-valdez-mcgill-university\u001b[0m). Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.17.8\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in \u001b[35m\u001b[1m./wandb/run-20240911_144537-cv9r056m\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run \u001b[1m`wandb offline`\u001b[0m to turn off syncing.\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mdefault\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/cv9r056m\u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", + "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", + "┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━┩\n", + "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNorm2d │ 128 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequential │ 147 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequential │ 525 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequential │ 2.1 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequential │ 8.4 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ AdaptiveAvgPool2d │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\n", + "└────┴─────────────────┴───────────────────┴────────┴───────┘\n", + "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", + "\u001b[1mNon-trainable params\u001b[0m: 0 \n", + "\u001b[1mTotal params\u001b[0m: 11.7 M \n", + "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", + "\u001b[1mModules in train mode\u001b[0m: 68 \n", + "\u001b[1mModules in eval mode\u001b[0m: 0 \n", + "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m3.03it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", + "s/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (30) \n", + "is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower \n", + "value for log_every_n_steps if you want to see logs for the training epoch.\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m194.480 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m194.480 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:12\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m0m \u001b[37mv_num: 056m\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:02 • 0:00:12\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:02 • 0:00:12\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:10\u001b[0m \u001b[37m2.53it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:10\u001b[0m \u001b[37m2.53it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:09\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:09\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:09\u001b[0m \u001b[37m2.87it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:09\u001b[0m \u001b[37m2.87it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:03 • 0:00:08\u001b[0m \u001b[37m2.94it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:03 • 0:00:08\u001b[0m \u001b[37m2.94it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.03it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.03it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.12it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.12it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.27it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.27it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.19it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.19it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.26it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.26it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.27it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.27it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.32it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.32it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.33it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.33it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.36it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.36it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.37it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.37it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.42it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.42it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.46it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.46it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m230.148 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m256.431 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m178.465 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m201.937 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m231.156 \u001b[0m\n", + "\u001b[?25h/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", + "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", + "\u001b[2;36m[14:46:09]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=96605;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=535638;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=206179;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=390492;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.33it/s\u001b[0m \u001b[37m3.32it/s\u001b[0m \n", + "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", + "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", + "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4357786178588867 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 214.4339141845703 \u001b[0m\u001b[35m \u001b[0m│\n", + "└──────────────────────────────┴──────────────────────────────┘\n", + "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.33it/s\u001b[0m \n", + "\u001b[?25h\u001b[2;36m[14:46:30]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=47061;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=655270;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#165\u001b\\\u001b[2m165\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m loss metric. \u001b[2m \u001b[0m\n", + "val val/samples_per_second_epoch: \u001b[1;36m214.4339141845703\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run history:\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: epoch ▁▁█\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: lr-SGD ▁\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch ▁\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step ▁▁▂▂▂▃▃▃▄▄▅▅▅▆▆▆▇▇███▁▂▂▂▃▃▄▄▄▅▅▅▆▆▆▇▇██\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss █▁\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_epoch ▁█\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_step ▁▇██▇█▅▄▆▆▄▇▄█▇▇▇▆▇▄▇▆▇▇██▆▅▆▆▇▇▅▇█▇▇▇▇▇\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run summary:\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: epoch 1\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: lr-SGD 0.123\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch 231.15637\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step 30\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss 0.43578\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_epoch 214.43391\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_step 226.7135\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run \u001b[33mdefault\u001b[0m at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/cv9r056m\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Synced 5 W&B file(s), 0 media file(s), 2 artifact file(s) and 0 other file(s)\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Find logs at: \u001b[35m\u001b[1m./wandb/run-20240911_144537-cv9r056m/logs\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m wandb version 0.17.9 is available! To upgrade, please run:\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m $ pip install wandb --upgrade\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m The new W&B backend becomes opt-out in version 0.18.0; try it out with `wandb.require(\"core\")`! See https://wandb.me/wandb-core for more information.\n" + ] + } + ], "source": [ - "# model size\n", - "# https://discuss.pytorch.org/t/finding-model-size/130275/2\n", - "model = ???\n", - "param_size = 0\n", - "for param in model.parameters():\n", - " param_size += param.nelement() * param.element_size()\n", - "buffer_size = 0\n", - "for buffer in model.buffers():\n", - " buffer_size += buffer.nelement() * buffer.element_size()\n", - "\n", - "size_all_mb = (param_size + buffer_size) / 1024**2\n", - "print('model size: {:.3f}MB'.format(size_all_mb))\n" + "!python project/main.py \\\n", + " algorithm=no_op \\\n", + " datamodule=imagenet \\\n", + " trainer=profiling \\\n", + " trainer/logger=wandb" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "$$\n", - "MBU = \n", - "\\frac{\\# \\text{ Params} \\cdot \\text{bytes per param} \\cdot \\text{tokens per second}}{\\text{Memory Bandwidth}}\n", - "$$\n" + "We can now visualize the results of our run at `wandb_url`" ] }, { @@ -117,27 +1089,470 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Finding a bottleneck is not necessarily straightforward or clear from the start. A sensible first step is to determine whether a potential slowdown originates from data loading or model computation. Querying both ends of the process can be done to determine whether the master process has a significant stall when fetching the next batch, or not. \n", + "Finding a bottleneck is not necessarily clear or straightforward from the start. A sensible first step is to determine whether a potential slowdown originates from data loading or model computation. Querying both ends of the process can be done to determine whether the master process has a significant stall when fetching the next batch, or not. \n", "If it's close to 0, then data loading outpaces compute, and compute is the bottleneck. \n", "If it's much greater than 0, then compute outpaces data loading, and data loading is the bottleneck. \n", - "You might not care about CPU usage by the master process and data loaders, so long as the GPU remains fully utilized. \n", - "Nonetheless, a profiler may record that anyways. How do you look out for relevant stuff? Here are a few ideas.\n" + "You might not care about CPU usage by the master process and data loaders, so long as the GPU remains fully utilized." ] }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "LexerNoViableAltException: \\\n", - " ^\n", - "See https://hydra.cc/docs/1.2/advanced/override_grammar/basic for details\n", - "\n", - "Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.\n" + "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/11/24 12:09:08]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=626004;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=19589;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m .yaml last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=693194;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=60895;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m not find \u001b[32m'hydra/sweeper/orion'\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Available options in \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'hydra/sweeper'\u001b[0m: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m basic \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Config search path: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra.conf\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mmain\u001b[0m, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mproject.configs\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m-colorlog, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra_plugins.hydra_c\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=346486;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=788581;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for resources/cpu.yaml \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=322509;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=244602;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m config resources/cpu.yaml: In \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'resources/cpu'\u001b[0m: Could not \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m override \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'resources/hydra/launcher'\u001b[0m. No \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m match in the defaults list. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=971168;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=544750;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m hydra/profiling_multirun.yaml \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=100610;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=775759;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m hydra/profiling_multirun.yaml: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m In \u001b[32m'hydra/profiling_multirun'\u001b[0m: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Could not override \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'hydra/hydra/launcher'\u001b[0m. No match \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m in the defaults list. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=286899;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=677441;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for hydra/cpu_vs_gpu.yaml \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=454750;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=684550;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m config hydra/cpu_vs_gpu.yaml: In \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'hydra/cpu_vs_gpu'\u001b[0m: Could not \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m override \u001b[32m'hydra/hydra/launcher'\u001b[0m. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m No match in the defaults list. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m46/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m104 \u001b[0m ]\n", + " \u001b[31mit/s\u001b[0m \n", + "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/11/24 12:09:09]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=766136;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=160792;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=475361;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=860693;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", + "seed manually set to 1633\n", + "Seed set to 1633\n", + "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", + "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", + "GPU available: True (cuda), used: True\n", + "TPU available: False, using: 0 TPU cores\n", + "HPU available: False, using: 0 HPUs\n", + "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/pytorch/trainer/connectors/logger_connector/logger_connector.py:75: Starting from v1.9.0, `tensorboardX` has been removed as a dependency of the `lightning.pytorch` package, due to potential conflicts with other packages in the ML ecosystem. For this reason, `logger=True` will use `CSVLogger` as the default logger, unless the `tensorboard` or `tensorboardX` packages are found. Please `pip install lightning[extra]` or one of them to enable TensorBoard support by default\n", + "\u001b[2;36m[12:09:09]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=640269;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=664528;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7feedd34ffa0\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", + "You are using a CUDA device ('NVIDIA A100-SXM4-80GB MIG 3g.40gb') that has Tensor Cores. To properly utilize them, you should set `torch.set_float32_matmul_precision('medium' | 'high')` which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision\n", + "\u001b[2;36m[12:09:10]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=465463;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=673508;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=62465;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=50264;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", + "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", + "┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━┩\n", + "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNorm2d │ 128 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequential │ 147 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequential │ 525 K │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequential │ 2.1 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequential │ 8.4 M │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ AdaptiveAvgPool2d │ 0 │ train │\n", + "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\n", + "└────┴─────────────────┴───────────────────┴────────┴───────┘\n", + "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", + "\u001b[1mNon-trainable params\u001b[0m: 0 \n", + "\u001b[1mTotal params\u001b[0m: 11.7 M \n", + "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", + "\u001b[1mModules in train mode\u001b[0m: 68 \n", + "\u001b[1mModules in eval mode\u001b[0m: 0 \n", + "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", + "s/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (30) \n", + "is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower \n", + "value for log_every_n_steps if you want to see logs for the training epoch.\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m217.110 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m217.110 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:10\u001b[0m \u001b[37m2.88it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m \u001b[37mv_num: 19.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:01 • 0:00:08\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:01 • 0:00:08\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.59it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.59it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.55it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.55it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:02 • 0:00:06\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:02 • 0:00:06\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.73it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.73it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.76it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.76it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.72it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.72it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:04 • 0:00:04\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:04 • 0:00:04\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.65it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.65it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:05 • 0:00:03\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:05 • 0:00:03\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:05 • 0:00:03\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:05 • 0:00:03\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.71it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.71it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:06 • 0:00:02\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:06 • 0:00:02\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:06 • 0:00:02\u001b[0m \u001b[37m3.72it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:06 • 0:00:02\u001b[0m \u001b[37m3.72it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.73it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.73it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.75it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.75it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:07 • 0:00:01\u001b[0m \u001b[37m3.77it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:07 • 0:00:01\u001b[0m \u001b[37m3.77it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:07 • 0:00:01\u001b[0m \u001b[37m3.78it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:07 • 0:00:01\u001b[0m \u001b[37m3.78it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.77it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.77it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m233.653 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m280.538 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m212.582 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m215.113 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m246.492 \u001b[0m\n", + "\u001b[?25hThe following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", + "\u001b[2;36m[12:09:41]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=848753;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=339553;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=232176;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=670495;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37m3.40it/s\u001b[0m \n", + "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", + "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", + "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4903430640697479 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 219.41116333007812 \u001b[0m\u001b[35m \u001b[0m│\n", + "└──────────────────────────────┴──────────────────────────────┘\n", + "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.40it/s\u001b[0m \n", + "\u001b[?25h\u001b[2;36m[12:10:01]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=977916;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=724301;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#165\u001b\\\u001b[2m165\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m loss metric. \u001b[2m \u001b[0m\n", + "val val/samples_per_second_epoch: \u001b[1;36m219.41116333007812\u001b[0m\n" ] } ], @@ -145,10 +1560,518 @@ "!python project/main.py \\\n", " algorithm=no_op \\\n", " datamodule=imagenet \\\n", - " ++hydra=profiling_multirun \\\n", - " ++trainer.max_epochs=1 \\\n", - " ++trainer.limit_train_batches=30 \\\n", - " ++trainer.limit_val_batches=30 \\\n" + " trainer=profiling" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/10/24 16:16:47]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=791467;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=848502;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m .yaml last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/10/24 16:16:48]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=992568;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=88644;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m not find \u001b[32m'hydra/sweeper/orion'\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Available options in \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[32m'hydra/sweeper'\u001b[0m: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m basic \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Config search path: \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra.conf\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mmain\u001b[0m, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mproject.configs\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m-colorlog, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra_plugins.hydra_c\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Config file hydra/default.yaml \u001b]8;id=434209;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=149573;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#362\u001b\\\u001b[2m362\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m was modified, regenerating the \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m schema. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 76%\u001b[0m \u001b[91m━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m34/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 76%\u001b[0m \u001b[91m━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m34/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m45/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m50 \u001b[0m ]\n", + " \u001b[31mit/s\u001b[0m \n", + "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=499898;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=831070;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=697638;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=810789;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", + "seed manually set to 56210\n", + "Seed set to 56210\n", + "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=example datamodule=imagene ...\n", + "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", + "GPU available: True (cuda), used: True\n", + "TPU available: False, using: 0 TPU cores\n", + "HPU available: False, using: 0 HPUs\n", + "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/pytorch/trainer/connectors/logger_connector/logger_connector.py:75: Starting from v1.9.0, `tensorboardX` has been removed as a dependency of the `lightning.pytorch` package, due to potential conflicts with other packages in the ML ecosystem. For this reason, `logger=True` will use `CSVLogger` as the default logger, unless the `tensorboard` or `tensorboardX` packages are found. Please `pip install lightning[extra]` or one of them to enable TensorBoard support by default\n", + "\u001b[2;36m[16:16:49]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=2388;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=591001;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7fb4dc9fca60\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m[16:16:54]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=961539;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=3680;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=34379;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=596652;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━┓\n", + "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35m In sizes\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35m Out sizes\u001b[0m\u001b[1;35m \u001b[0m┃\n", + "┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━┩\n", + "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\u001b[37m \u001b[0m\u001b[37m [64, 3,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 1000]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224, 224]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\u001b[37m \u001b[0m\u001b[37m [64, 3,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224, 224]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNorm… │ 128 │ train │\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112, 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112, 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112, 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequential │ 147 K │ train │\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequential │ 525 K │ train │\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 128,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 28, 28]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequential │ 2.1 M │ train │\u001b[37m \u001b[0m\u001b[37m [64, 128,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 256,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 28, 28]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 14, 14]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequential │ 8.4 M │ train │\u001b[37m \u001b[0m\u001b[37m [64, 256,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 512,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 14, 14]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 7, 7]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ AdaptiveA… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m [64, 512,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 512,\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 7, 7]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1, 1]\u001b[0m\u001b[37m \u001b[0m│\n", + "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\u001b[37m \u001b[0m\u001b[37m [64, 512]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 1000]\u001b[0m\u001b[37m \u001b[0m│\n", + "└────┴─────────────────┴────────────┴────────┴───────┴────────────┴────────────┘\n", + "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", + "\u001b[1mNon-trainable params\u001b[0m: 0 \n", + "\u001b[1mTotal params\u001b[0m: 11.7 M \n", + "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", + "\u001b[1mModules in train mode\u001b[0m: 68 \n", + "\u001b[1mModules in eval mode\u001b[0m: 0 \n", + "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m2.28it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", + "s/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (30) \n", + "is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower \n", + "value for log_every_n_steps if you want to see logs for the training epoch.\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m146.211 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m146.211 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:02 • 0:00:12\u001b[0m \u001b[37m2.28it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m \u001b[37mv_num: 16.000\u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:02 • 0:00:11\u001b[0m \u001b[37m2.39it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:02 • 0:00:11\u001b[0m \u001b[37m2.39it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:03 • 0:00:11\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:03 • 0:00:11\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:03 • 0:00:11\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:03 • 0:00:11\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:03 • 0:00:10\u001b[0m \u001b[37m2.40it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:03 • 0:00:10\u001b[0m \u001b[37m2.40it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:04 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:04 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:04 • 0:00:10\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:04 • 0:00:10\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:05 • 0:00:09\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:05 • 0:00:09\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:05 • 0:00:09\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:05 • 0:00:09\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:06 • 0:00:08\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:06 • 0:00:08\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:06 • 0:00:08\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:06 • 0:00:08\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:07 • 0:00:07\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:07 • 0:00:07\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:07 • 0:00:07\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:07 • 0:00:07\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:07 • 0:00:07\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:07 • 0:00:07\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:08 • 0:00:06\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:08 • 0:00:06\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:09 • 0:00:06\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:09 • 0:00:06\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:09 • 0:00:05\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:09 • 0:00:05\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:09 • 0:00:05\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:09 • 0:00:05\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:10 • 0:00:05\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:10 • 0:00:05\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:10 • 0:00:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:10 • 0:00:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:11 • 0:00:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:11 • 0:00:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:11 • 0:00:03\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:11 • 0:00:03\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:12 • 0:00:03\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:12 • 0:00:03\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:12 • 0:00:02\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:12 • 0:00:02\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:12 • 0:00:02\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:12 • 0:00:02\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:13 • 0:00:01\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:13 • 0:00:01\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:13 • 0:00:01\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:13 • 0:00:01\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m138.248 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m158.500 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m147.835 \u001b[0m\n", + " \u001b[37mval/samples_per_s…\u001b[0m\n", + " \u001b[37m143.576 \u001b[0m\n", + " \u001b[37mtrain/samples_per…\u001b[0m\n", + " \u001b[37m146.852 \u001b[0m\n", + "\u001b[?25h\u001b[2;36m[16:17:48]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=302525;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=670131;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=923056;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=385077;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:13 • 0:00:00\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37m2.11it/s\u001b[0m \n", + "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", + "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", + "│\u001b[36m \u001b[0m\u001b[36m val/accuracy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.0234375 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 8.902314186096191 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 137.38766479492188 \u001b[0m\u001b[35m \u001b[0m│\n", + "└──────────────────────────────┴──────────────────────────────┘\n", + "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:13 • 0:00:00\u001b[0m \u001b[37m2.10it/s\u001b[0m \n", + "\u001b[?25hval accuracy: \u001b[1;36m2.3\u001b[0m%\n", + "val val/accuracy: \u001b[1;36m0.0234375\u001b[0m\n", + "val val/samples_per_second_epoch: \u001b[1;36m137.38766479492188\u001b[0m\n" + ] + } + ], + "source": [ + "!python project/main.py \\\n", + " algorithm=example \\\n", + " datamodule=imagenet \\\n", + " trainer=profiling" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## GPU vs CPU" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Do we even need a GPU? Compare speed using CPU only vs the slowest GPU available, for a low number of steps\n", + "If the CPU performance loosely comparable (for instance, only 1.5-2x slower) than with a GPU, then it might be worth considering! (LMK if this happens, one thing could be to try to increase the # of CPUs and measure performance scaling, then ship this kind of job to a DRAC cluster)\n", + "In most workflows, using a GPU actually helps a lot.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Traceback (most recent call last):\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\", line 177, in \n", + " main()\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/main.py\", line 94, in decorated_main\n", + " _run_hydra(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 394, in _run_hydra\n", + " _run_app(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 457, in _run_app\n", + " run_and_report(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 223, in run_and_report\n", + " raise ex\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 220, in run_and_report\n", + " return func()\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 458, in \n", + " lambda: hydra.run(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/hydra.py\", line 105, in run\n", + " cfg = self.compose_config(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/hydra.py\", line 594, in compose_config\n", + " cfg = self.config_loader.load_configuration(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/config_loader_impl.py\", line 142, in load_configuration\n", + " return self._load_configuration_impl(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/config_loader_impl.py\", line 253, in _load_configuration_impl\n", + " defaults_list = create_defaults_list(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/defaults_list.py\", line 752, in create_defaults_list\n", + " overrides.ensure_overrides_used()\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/defaults_list.py\", line 168, in ensure_overrides_used\n", + " raise ConfigCompositionException(msg)\n", + "hydra.errors.ConfigCompositionException: In 'resources/cpu': Could not override 'resources/hydra/launcher@hydra.launcher'. No match in the defaults list.\n" + ] + } + ], + "source": [ + "!HYDRA_FULL_ERROR=1 python project/main.py \\\n", + " algorithm=example \\\n", + " datamodule=imagenet \\\n", + " trainer=profiling \\\n", + " resources=cpu" ] }, { @@ -157,13 +2080,11 @@ "metadata": {}, "outputs": [], "source": [ - "!python project/main.py \\\n", + "!HYDRA_FULL_ERROR=1 python project/main.py \\\n", " algorithm=example \\\n", " datamodule=imagenet \\\n", - " ++logger=wandb \\\n", - " ++trainer.max_epochs=1 \\\n", - " ++trainer.limit_train_batches=30 \\\n", - " ++trainer.limit_val_batches=30" + " trainer=profiling \\\n", + " resources=one_gpu" ] }, { @@ -228,6 +2149,10 @@ "metadata": {}, "outputs": [], "source": [ + "## What performance do you get with each type of GPU? \n", + "# (Based on the VRAM requirements of the job (step 1), \n", + "# try all the GPU types on the Cluster that can accommodate this kind of job)\n", + "\n", "# Add an example of a sweep over some parameters, \n", "# with the training throughput as the metric, \n", "# :: callbacks/samples_per_second, \n", @@ -248,16 +2173,25 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### Logging with Weights & Biases (wandb)" + "## GPU utilization" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "How well are we using the GPU?\n", + "Once we've selected the target GPU that we want to use, measure the GPU utilization. Is the GPU utilization high? (>80%?)\n", + "If it's high (>80%), then we can either stop here, or we can keep going a bit further\n", + "If it's low, then we can use the PyTorch profiler (or any other tool) to try to figure out what the bottleneck i\n", + "## maybe look at submitit's array_parallelism" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "The Mila Research template integrates wandb functionality as a logger specification. \n", - "This has the advantage of being able to track additional metrics and create accompanying visualizations. \n", - "We will now create a wandb report comparing throughput between GPUs. \n" + "We would like to maximize our throughput given GPU choice" ] }, { @@ -266,24 +2200,15 @@ "metadata": {}, "outputs": [], "source": [ - "# Create a wandb report with the throughput comparison \n", - "# between the different GPU types.\n", - "# i.e. specify wandb as the logger and log the throughput\n", - "!python project/main.py \\\n", - " algorithm=no_op \\\n", - " datamodule=imagenet \\\n", - " ++logger=wandb \\\n", - " ++trainer.max_epochs=1 \\\n", - " ++trainer.limit_train_batches=30 \\\n", - " ++trainer.limit_val_batches=30 \\\n", - " hydra=profiling_multirun" + "### Measure the performance on different GPUS using the optimal datamodule \n", + "### params from before (and keeping other parameters the same)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "We would like to maximize our throughput given GPU choice" + "We will now sweep over model hyper-parameters to maximize the utilization of our selected GPU." ] }, { @@ -292,37 +2217,60 @@ "metadata": {}, "outputs": [], "source": [ - "## Find the best datamodule parameters to maximize the throughput \n", - "## (batches per second) without training (NoOP algo)" + "#### Using the results from before, do a simple sweep over model hyper-parameters \n", + "#### to maximize the utilization of the selected GPU (which was selected as a tradeoff \n", + "#### between performance and difficulty to get an allocation). For example if the \n", + "#### RTX8000's are 20% slower than A100s but 5x easier to get an allocation on, use those instead." ] }, { - "cell_type": "code", - "execution_count": null, + "cell_type": "markdown", "metadata": {}, - "outputs": [], "source": [ - "### Measure the performance on different GPUS using the optimal datamodule \n", - "### params from before (and keeping other parameters the same)" + "## What is a profiler and what is it good for?" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "We will now sweep over model hyper-parameters to maximize the utilization of our selected GPU." + "The former process was a bit contrived - We can zero down specifically on subprocesses... \n", + "A profiler is a tool that allows you to measure the time and memory consumption of the model’s operators. Specifically, the PyTorch profiler output provides clues about operations relevant to model training. Examples include the total amount of time spent doing low-level mathematical operations in the GPU, and whether these are unexpectedly slow or take a disproportionate amount of time, indicating they should be avoided or optimized. Identifying problematic operations can greatly help us validate or rethink our baseline model performance expectations.\n", + "\n", + "[Multiple](https://developer.nvidia.com/blog/profiling-and-optimizing-deep-neural-networks-with-dlprof-and-pyprof/) [profilers](https://github.com/plasma-umass/scalene) [exist](https://docs.python.org/3/library/profile.html). For the purposes of this example we'll use the default [PyTorch Profiler](https://pytorch.org/tutorials/recipes/recipes/profiler_recipe.html). " ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 18, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "------------------------- ------------ ------------ ------------ ------------ ------------ ------------ ------------ ------------ \n", + " Name Self CPU % Self CPU CPU total % CPU total CPU time avg CPU Mem Self CPU Mem # of Calls \n", + "------------------------- ------------ ------------ ------------ ------------ ------------ ------------ ------------ ------------ \n", + " cudaDeviceSynchronize 100.00% 14.444us 100.00% 14.444us 14.444us 0 b 0 b 1 \n", + "------------------------- ------------ ------------ ------------ ------------ ------------ ------------ ------------ ------------ \n", + "Self CPU time total: 14.444us\n", + "\n" + ] + } + ], "source": [ - "#### Using the results from before, do a simple sweep over model hyper-parameters \n", - "#### to maximize the utilization of the selected GPU (which was selected as a tradeoff \n", - "#### between performance and difficulty to get an allocation). For example if the \n", - "#### RTX8000's are 20% slower than A100s but 5x easier to get an allocation on, use those instead." + "from torch.profiler import ProfilerActivity, profile \n", + "\n", + "profiler = profile(\n", + " activities=[ProfilerActivity.CPU, ProfilerActivity.CUDA],\n", + " record_shapes=True,\n", + " profile_memory=True,\n", + " with_stack=True,\n", + ")\n", + "profiler.start()\n", + "profiler.stop()\n", + "print(profiler.key_averages().table(sort_by=\"cpu_time_total\", row_limit=10))\n" ] }, { diff --git a/project/configs/algorithm/no_op.yaml b/project/configs/algorithm/no_op.yaml new file mode 100644 index 00000000..f7b78c63 --- /dev/null +++ b/project/configs/algorithm/no_op.yaml @@ -0,0 +1,2 @@ +_target_: project.algorithms.no_op.NoOp +_partial_: true diff --git a/project/configs/hydra/profiling_cpu_vs_gpu.yaml b/project/configs/hydra/profiling_cpu_vs_gpu.yaml new file mode 100644 index 00000000..e50348ef --- /dev/null +++ b/project/configs/hydra/profiling_cpu_vs_gpu.yaml @@ -0,0 +1,23 @@ +# enable color logging +defaults: + - _self_ + - override hydra_logging: disabled + - override job_logging: disabled + - override hydra/launcher: submitit_slurm + +run: + # output directory, generated dynamically on each run + dir: logs/${name}/runs/${now:%Y-%m-%d}/${now:%H-%M-%S} +sweep: + dir: logs/${name}/multiruns/${now:%Y-%m-%d}/${now:%H-%M-%S} + subdir: ${hydra.job.num} + +verbose: False + +#submitit: + #salloc --gres=gpu:a100:1 -c 6 --mem=32G -t 48:00:00 --partition=unkillable + #gres: gpu:a100:1 + #cpus_per_gpu: 6 + #mem_per_cpu: 32 + #timeout_min: 180 + #partition: unkillable diff --git a/project/configs/hydra/profiling_gpu_multirun.yaml b/project/configs/hydra/profiling_gpu_multirun.yaml new file mode 100644 index 00000000..e50348ef --- /dev/null +++ b/project/configs/hydra/profiling_gpu_multirun.yaml @@ -0,0 +1,23 @@ +# enable color logging +defaults: + - _self_ + - override hydra_logging: disabled + - override job_logging: disabled + - override hydra/launcher: submitit_slurm + +run: + # output directory, generated dynamically on each run + dir: logs/${name}/runs/${now:%Y-%m-%d}/${now:%H-%M-%S} +sweep: + dir: logs/${name}/multiruns/${now:%Y-%m-%d}/${now:%H-%M-%S} + subdir: ${hydra.job.num} + +verbose: False + +#submitit: + #salloc --gres=gpu:a100:1 -c 6 --mem=32G -t 48:00:00 --partition=unkillable + #gres: gpu:a100:1 + #cpus_per_gpu: 6 + #mem_per_cpu: 32 + #timeout_min: 180 + #partition: unkillable diff --git a/project/configs/resources/cpu.yaml b/project/configs/resources/cpu.yaml new file mode 100644 index 00000000..e411b889 --- /dev/null +++ b/project/configs/resources/cpu.yaml @@ -0,0 +1,13 @@ +defaults: + - override hydra/launcher: submitit_slurm + +trainer: + accelerator: cpu + strategy: null + devices: 1 + +hydra: + mode: MULTIRUN + launcher: + gpus_per_node: null + cpus_per_task: 1 diff --git a/project/configs/trainer/cpu.yaml b/project/configs/trainer/cpu.yaml deleted file mode 100644 index cfc6bea8..00000000 --- a/project/configs/trainer/cpu.yaml +++ /dev/null @@ -1,6 +0,0 @@ -defaults: - - default.yaml - -accelerator: cpu -gpus: 0 -strategy: null diff --git a/project/configs/trainer/profiling.yaml b/project/configs/trainer/profiling.yaml new file mode 100644 index 00000000..690c446c --- /dev/null +++ b/project/configs/trainer/profiling.yaml @@ -0,0 +1,4 @@ +_target_: lightning.Trainer +max_epochs: 1 +limit_train_batches: 30 +limit_val_batches: 30 diff --git a/pyproject.toml b/pyproject.toml index 63888026..b1a9e84c 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -30,6 +30,8 @@ dependencies = [ "transformers>=4.44.0", "scikit-learn>=1.5.1", "evaluate>=0.4.2", + "jupyterlab>=4.2.5", + "notebook>=7.2.2", ] readme = "README.md" requires-python = ">= 3.10" diff --git a/requirements-dev.lock b/requirements-dev.lock index 32d7b591..e96a56fc 100644 --- a/requirements-dev.lock +++ b/requirements-dev.lock @@ -27,8 +27,19 @@ annotated-types==0.7.0 antlr4-python3-runtime==4.9.3 # via hydra-core # via omegaconf +anyio==4.4.0 + # via httpx + # via jupyter-server +argon2-cffi==23.1.0 + # via jupyter-server +argon2-cffi-bindings==21.2.0 + # via argon2-cffi +arrow==1.3.0 + # via isoduration asttokens==2.4.1 # via stack-data +async-lru==2.0.4 + # via jupyterlab async-timeout==4.0.3 # via aiohttp attrs==24.2.0 @@ -36,6 +47,7 @@ attrs==24.2.0 # via jsonschema # via referencing babel==2.16.0 + # via jupyterlab-server # via mkdocs-material bcrypt==4.2.0 # via paramiko @@ -51,9 +63,12 @@ blessed==1.20.0 bracex==2.5 # via wcmatch certifi==2024.8.30 + # via httpcore + # via httpx # via requests # via sentry-sdk cffi==1.17.0 + # via argon2-cffi-bindings # via cryptography # via pynacl cfgv==3.4.0 @@ -115,6 +130,7 @@ etils==1.9.4 evaluate==0.4.2 # via research-project-template exceptiongroup==1.2.2 + # via anyio # via ipython # via pytest execnet==2.1.1 @@ -137,6 +153,8 @@ flax==0.8.5 # via torch-jax-interop fonttools==4.53.1 # via matplotlib +fqdn==1.5.1 + # via jsonschema frozenlist==1.4.1 # via aiohttp # via aiosignal @@ -158,6 +176,12 @@ gitpython==3.1.43 # via wandb griffe==1.2.0 # via mkdocstrings-python +h11==0.14.0 + # via httpcore +httpcore==1.0.5 + # via httpx +httpx==0.27.2 + # via jupyterlab huggingface-hub==0.24.6 # via datasets # via evaluate @@ -179,6 +203,9 @@ hydra-zen==0.13.0 identify==2.6.0 # via pre-commit idna==3.8 + # via anyio + # via httpx + # via jsonschema # via requests # via yarl importlib-resources==6.4.4 @@ -188,10 +215,13 @@ iniconfig==2.0.0 invoke==2.2.0 # via fabric ipykernel==6.29.5 + # via jupyterlab # via mkdocs-jupyter # via research-project-template ipython==8.27.0 # via ipykernel +isoduration==20.11.0 + # via jsonschema jax==0.4.31 # via chex # via flax @@ -214,6 +244,9 @@ jaxlib==0.4.31 jedi==0.19.1 # via ipython jinja2==3.1.4 + # via jupyter-server + # via jupyterlab + # via jupyterlab-server # via mkdocs # via mkdocs-material # via mkdocstrings @@ -221,21 +254,48 @@ jinja2==3.1.4 # via torch joblib==1.4.2 # via scikit-learn +json5==0.9.25 + # via jupyterlab-server +jsonpointer==3.0.0 + # via jsonschema jsonschema==4.23.0 + # via jupyter-events + # via jupyterlab-server # via nbformat jsonschema-specifications==2023.12.1 # via jsonschema jupyter-client==8.6.2 # via ipykernel + # via jupyter-server # via nbclient jupyter-core==5.7.2 # via ipykernel # via jupyter-client + # via jupyter-server + # via jupyterlab # via nbclient # via nbconvert # via nbformat +jupyter-events==0.10.0 + # via jupyter-server +jupyter-lsp==2.2.5 + # via jupyterlab +jupyter-server==2.14.2 + # via jupyter-lsp + # via jupyterlab + # via jupyterlab-server + # via notebook + # via notebook-shim +jupyter-server-terminals==0.5.3 + # via jupyter-server +jupyterlab==4.2.5 + # via notebook + # via research-project-template jupyterlab-pygments==0.3.0 # via nbconvert +jupyterlab-server==2.27.3 + # via jupyterlab + # via notebook jupytext==1.16.4 # via mkdocs-jupyter kiwisolver==1.4.6 @@ -342,8 +402,10 @@ natsort==8.4.0 nbclient==0.10.0 # via nbconvert nbconvert==7.16.4 + # via jupyter-server # via mkdocs-jupyter nbformat==5.10.4 + # via jupyter-server # via jupytext # via nbclient # via nbconvert @@ -354,6 +416,11 @@ networkx==3.3 # via torch nodeenv==1.9.1 # via pre-commit +notebook==7.2.2 + # via research-project-template +notebook-shim==0.2.4 + # via jupyterlab + # via notebook numpy==1.26.4 # via chex # via contourpy @@ -424,6 +491,8 @@ optax==0.2.3 # via flax orbax-checkpoint==0.6.1 # via flax +overrides==7.7.0 + # via jupyter-server packaging==24.1 # via black # via datasets @@ -431,6 +500,9 @@ packaging==24.1 # via huggingface-hub # via hydra-core # via ipykernel + # via jupyter-server + # via jupyterlab + # via jupyterlab-server # via jupytext # via lightning # via lightning-utilities @@ -470,6 +542,8 @@ platformdirs==4.2.2 pluggy==1.5.0 # via pytest pre-commit==3.8.0 +prometheus-client==0.20.0 + # via jupyter-server prompt-toolkit==3.0.47 # via ipython # via questionary @@ -481,6 +555,7 @@ psutil==6.0.0 # via wandb ptyprocess==0.7.0 # via pexpect + # via terminado pure-eval==0.2.3 # via stack-data py-cpuinfo==9.0.0 @@ -530,10 +605,13 @@ pytest-testmon==2.1.1 pytest-timeout==2.3.1 pytest-xdist==3.6.1 python-dateutil==2.9.0.post0 + # via arrow # via ghp-import # via jupyter-client # via matplotlib # via pandas +python-json-logger==2.0.7 + # via jupyter-events pytorch-lightning==2.4.0 # via lightning pytorch2jax==0.1.0 @@ -544,6 +622,7 @@ pyyaml==6.0.2 # via datasets # via flax # via huggingface-hub + # via jupyter-events # via jupytext # via lightning # via mkdocs @@ -562,11 +641,13 @@ pyyaml-env-tag==0.1 pyzmq==26.2.0 # via ipykernel # via jupyter-client + # via jupyter-server questionary==1.10.0 # via milatools referencing==0.35.1 # via jsonschema # via jsonschema-specifications + # via jupyter-events regex==2024.7.24 # via mkdocs-material # via transformers @@ -575,9 +656,16 @@ requests==2.32.3 # via evaluate # via gdown # via huggingface-hub + # via jupyterlab-server # via mkdocs-material # via transformers # via wandb +rfc3339-validator==0.1.4 + # via jsonschema + # via jupyter-events +rfc3986-validator==0.1.1 + # via jsonschema + # via jupyter-events rich==13.8.0 # via flax # via milatools @@ -594,6 +682,8 @@ scipy==1.14.1 # via jax # via jaxlib # via scikit-learn +send2trash==1.8.3 + # via jupyter-server sentry-sdk==2.13.0 # via wandb setproctitle==1.3.3 @@ -606,8 +696,12 @@ six==1.16.0 # via blessed # via docker-pycreds # via python-dateutil + # via rfc3339-validator smmap==5.0.1 # via gitdb +sniffio==1.3.1 + # via anyio + # via httpx soupsieve==2.6 # via beautifulsoup4 sshconf==0.2.7 @@ -622,6 +716,9 @@ tensor-regression==0.0.6 tensorstore==0.1.64 # via flax # via orbax-checkpoint +terminado==0.18.1 + # via jupyter-server + # via jupyter-server-terminals threadpoolctl==3.5.0 # via scikit-learn tinycss2==1.3.0 @@ -631,6 +728,7 @@ tokenizers==0.19.1 tomli==2.0.1 # via black # via coverage + # via jupyterlab # via jupytext # via pytest # via pytest-env @@ -655,6 +753,10 @@ torchvision==0.19.0 tornado==6.4.1 # via ipykernel # via jupyter-client + # via jupyter-server + # via jupyterlab + # via notebook + # via terminado tqdm==4.66.5 # via datasets # via evaluate @@ -671,6 +773,9 @@ traitlets==5.14.3 # via ipython # via jupyter-client # via jupyter-core + # via jupyter-events + # via jupyter-server + # via jupyterlab # via matplotlib-inline # via nbclient # via nbconvert @@ -679,7 +784,11 @@ transformers==4.44.2 # via research-project-template triton==3.0.0 # via torch +types-python-dateutil==2.9.0.20240906 + # via arrow typing-extensions==4.12.2 + # via anyio + # via async-lru # via black # via chex # via etils @@ -699,6 +808,8 @@ typing-extensions==4.12.2 # via torch tzdata==2024.1 # via pandas +uri-template==1.3.0 + # via jsonschema urllib3==2.2.2 # via requests # via sentry-sdk @@ -713,9 +824,13 @@ wcmatch==9.0 wcwidth==0.2.13 # via blessed # via prompt-toolkit +webcolors==24.8.0 + # via jsonschema webencodings==0.5.1 # via bleach # via tinycss2 +websocket-client==1.8.0 + # via jupyter-server wrapt==1.16.0 # via deprecated xxhash==3.5.0 @@ -726,5 +841,6 @@ yarl==1.9.7 zipp==3.20.1 # via etils setuptools==74.1.1 + # via jupyterlab # via lightning-utilities # via wandb diff --git a/requirements.lock b/requirements.lock index 59733b51..3491c8ca 100644 --- a/requirements.lock +++ b/requirements.lock @@ -27,8 +27,19 @@ annotated-types==0.7.0 antlr4-python3-runtime==4.9.3 # via hydra-core # via omegaconf +anyio==4.4.0 + # via httpx + # via jupyter-server +argon2-cffi==23.1.0 + # via jupyter-server +argon2-cffi-bindings==21.2.0 + # via argon2-cffi +arrow==1.3.0 + # via isoduration asttokens==2.4.1 # via stack-data +async-lru==2.0.4 + # via jupyterlab async-timeout==4.0.3 # via aiohttp attrs==24.2.0 @@ -36,6 +47,7 @@ attrs==24.2.0 # via jsonschema # via referencing babel==2.16.0 + # via jupyterlab-server # via mkdocs-material bcrypt==4.2.0 # via paramiko @@ -51,9 +63,12 @@ blessed==1.20.0 bracex==2.5 # via wcmatch certifi==2024.8.30 + # via httpcore + # via httpx # via requests # via sentry-sdk cffi==1.17.0 + # via argon2-cffi-bindings # via cryptography # via pynacl charset-normalizer==3.3.2 @@ -107,6 +122,7 @@ etils==1.9.4 evaluate==0.4.2 # via research-project-template exceptiongroup==1.2.2 + # via anyio # via ipython executing==2.1.0 # via stack-data @@ -125,6 +141,8 @@ flax==0.8.5 # via torch-jax-interop fonttools==4.53.1 # via matplotlib +fqdn==1.5.1 + # via jsonschema frozenlist==1.4.1 # via aiohttp # via aiosignal @@ -146,6 +164,12 @@ gitpython==3.1.43 # via wandb griffe==1.2.0 # via mkdocstrings-python +h11==0.14.0 + # via httpcore +httpcore==1.0.5 + # via httpx +httpx==0.27.2 + # via jupyterlab huggingface-hub==0.24.6 # via datasets # via evaluate @@ -165,6 +189,9 @@ hydra-submitit-launcher==1.2.0 hydra-zen==0.13.0 # via research-project-template idna==3.8 + # via anyio + # via httpx + # via jsonschema # via requests # via yarl importlib-resources==6.4.4 @@ -172,10 +199,13 @@ importlib-resources==6.4.4 invoke==2.2.0 # via fabric ipykernel==6.29.5 + # via jupyterlab # via mkdocs-jupyter # via research-project-template ipython==8.27.0 # via ipykernel +isoduration==20.11.0 + # via jsonschema jax==0.4.31 # via chex # via flax @@ -198,6 +228,9 @@ jaxlib==0.4.31 jedi==0.19.1 # via ipython jinja2==3.1.4 + # via jupyter-server + # via jupyterlab + # via jupyterlab-server # via mkdocs # via mkdocs-material # via mkdocstrings @@ -205,21 +238,48 @@ jinja2==3.1.4 # via torch joblib==1.4.2 # via scikit-learn +json5==0.9.25 + # via jupyterlab-server +jsonpointer==3.0.0 + # via jsonschema jsonschema==4.23.0 + # via jupyter-events + # via jupyterlab-server # via nbformat jsonschema-specifications==2023.12.1 # via jsonschema jupyter-client==8.6.2 # via ipykernel + # via jupyter-server # via nbclient jupyter-core==5.7.2 # via ipykernel # via jupyter-client + # via jupyter-server + # via jupyterlab # via nbclient # via nbconvert # via nbformat +jupyter-events==0.10.0 + # via jupyter-server +jupyter-lsp==2.2.5 + # via jupyterlab +jupyter-server==2.14.2 + # via jupyter-lsp + # via jupyterlab + # via jupyterlab-server + # via notebook + # via notebook-shim +jupyter-server-terminals==0.5.3 + # via jupyter-server +jupyterlab==4.2.5 + # via notebook + # via research-project-template jupyterlab-pygments==0.3.0 # via nbconvert +jupyterlab-server==2.27.3 + # via jupyterlab + # via notebook jupytext==1.16.4 # via mkdocs-jupyter kiwisolver==1.4.6 @@ -325,8 +385,10 @@ natsort==8.4.0 nbclient==0.10.0 # via nbconvert nbconvert==7.16.4 + # via jupyter-server # via mkdocs-jupyter nbformat==5.10.4 + # via jupyter-server # via jupytext # via nbclient # via nbconvert @@ -335,6 +397,11 @@ nest-asyncio==1.6.0 # via orbax-checkpoint networkx==3.3 # via torch +notebook==7.2.2 + # via research-project-template +notebook-shim==0.2.4 + # via jupyterlab + # via notebook numpy==2.1.1 # via chex # via contourpy @@ -404,6 +471,8 @@ optax==0.2.3 # via flax orbax-checkpoint==0.6.1 # via flax +overrides==7.7.0 + # via jupyter-server packaging==24.1 # via black # via datasets @@ -411,6 +480,9 @@ packaging==24.1 # via huggingface-hub # via hydra-core # via ipykernel + # via jupyter-server + # via jupyterlab + # via jupyterlab-server # via jupytext # via lightning # via lightning-utilities @@ -445,6 +517,8 @@ platformdirs==4.2.2 # via mkdocs-get-deps # via mkdocstrings # via wandb +prometheus-client==0.20.0 + # via jupyter-server prompt-toolkit==3.0.47 # via ipython # via questionary @@ -456,6 +530,7 @@ psutil==6.0.0 # via wandb ptyprocess==0.7.0 # via pexpect + # via terminado pure-eval==0.2.3 # via stack-data pyarrow==17.0.0 @@ -482,10 +557,13 @@ pyparsing==3.1.4 pysocks==1.7.1 # via requests python-dateutil==2.9.0.post0 + # via arrow # via ghp-import # via jupyter-client # via matplotlib # via pandas +python-json-logger==2.0.7 + # via jupyter-events pytorch-lightning==2.4.0 # via lightning pytorch2jax==0.1.0 @@ -496,6 +574,7 @@ pyyaml==6.0.2 # via datasets # via flax # via huggingface-hub + # via jupyter-events # via jupytext # via lightning # via mkdocs @@ -512,11 +591,13 @@ pyyaml-env-tag==0.1 pyzmq==26.2.0 # via ipykernel # via jupyter-client + # via jupyter-server questionary==1.10.0 # via milatools referencing==0.35.1 # via jsonschema # via jsonschema-specifications + # via jupyter-events regex==2024.7.24 # via mkdocs-material # via transformers @@ -525,9 +606,16 @@ requests==2.32.3 # via evaluate # via gdown # via huggingface-hub + # via jupyterlab-server # via mkdocs-material # via transformers # via wandb +rfc3339-validator==0.1.4 + # via jsonschema + # via jupyter-events +rfc3986-validator==0.1.1 + # via jsonschema + # via jupyter-events rich==13.8.0 # via flax # via milatools @@ -543,6 +631,8 @@ scipy==1.14.1 # via jax # via jaxlib # via scikit-learn +send2trash==1.8.3 + # via jupyter-server sentry-sdk==2.13.0 # via wandb setproctitle==1.3.3 @@ -555,8 +645,12 @@ six==1.16.0 # via blessed # via docker-pycreds # via python-dateutil + # via rfc3339-validator smmap==5.0.1 # via gitdb +sniffio==1.3.1 + # via anyio + # via httpx soupsieve==2.6 # via beautifulsoup4 sshconf==0.2.7 @@ -570,6 +664,9 @@ sympy==1.13.2 tensorstore==0.1.64 # via flax # via orbax-checkpoint +terminado==0.18.1 + # via jupyter-server + # via jupyter-server-terminals threadpoolctl==3.5.0 # via scikit-learn tinycss2==1.3.0 @@ -578,6 +675,7 @@ tokenizers==0.19.1 # via transformers tomli==2.0.1 # via black + # via jupyterlab # via jupytext toolz==0.12.1 # via chex @@ -599,6 +697,10 @@ torchvision==0.19.0 tornado==6.4.1 # via ipykernel # via jupyter-client + # via jupyter-server + # via jupyterlab + # via notebook + # via terminado tqdm==4.66.5 # via datasets # via evaluate @@ -615,6 +717,9 @@ traitlets==5.14.3 # via ipython # via jupyter-client # via jupyter-core + # via jupyter-events + # via jupyter-server + # via jupyterlab # via matplotlib-inline # via nbclient # via nbconvert @@ -623,7 +728,11 @@ transformers==4.44.2 # via research-project-template triton==3.0.0 # via torch +types-python-dateutil==2.9.0.20240906 + # via arrow typing-extensions==4.12.2 + # via anyio + # via async-lru # via black # via chex # via etils @@ -643,6 +752,8 @@ typing-extensions==4.12.2 # via torch tzdata==2024.1 # via pandas +uri-template==1.3.0 + # via jsonschema urllib3==2.2.2 # via requests # via sentry-sdk @@ -655,9 +766,13 @@ wcmatch==9.0 wcwidth==0.2.13 # via blessed # via prompt-toolkit +webcolors==24.8.0 + # via jsonschema webencodings==0.5.1 # via bleach # via tinycss2 +websocket-client==1.8.0 + # via jupyter-server wrapt==1.16.0 # via deprecated xxhash==3.5.0 @@ -668,5 +783,6 @@ yarl==1.9.7 zipp==3.20.1 # via etils setuptools==74.1.1 + # via jupyterlab # via lightning-utilities # via wandb From f189cf42577ffa29728430d6b4158c010693c45b Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Thu, 12 Sep 2024 13:13:43 -0400 Subject: [PATCH 12/33] wandb logging working, notebook progress, CPU GPU throughput comparisons --- docs/examples/profiling.ipynb | 2196 +++++++++-------- project/configs/config.yaml | 3 +- .../experiment/cluster_sweep_example.yaml | 1 + .../configs/hydra/profiling_cpu_vs_gpu.yaml | 23 - .../configs/hydra/profiling_gpu_multirun.yaml | 23 - project/configs/resources/cpu.yaml | 9 +- project/configs/resources/one_gpu.yaml | 2 +- project/configs/trainer/default.yaml | 3 +- project/configs/trainer/profiling.yaml | 4 + project/main.py | 6 +- 10 files changed, 1140 insertions(+), 1130 deletions(-) delete mode 100644 project/configs/hydra/profiling_cpu_vs_gpu.yaml delete mode 100644 project/configs/hydra/profiling_gpu_multirun.yaml diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 99a3321a..79e18c74 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -76,22 +76,42 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/11/24 14:43:09]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=107891;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=976209;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", + "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:00:24]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Config file config.yaml was \u001b]8;id=986330;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=4653;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#362\u001b\\\u001b[2m362\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m modified, regenerating the \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m schema. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:00:25]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Config file \u001b]8;id=694205;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=36089;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#362\u001b\\\u001b[2m362\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m .yaml last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/11/24 14:43:10]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=289962;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=183361;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m .yaml was modified, regenerating \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m the schema. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 2%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m1/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=623339;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=998599;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", @@ -110,77 +130,43 @@ "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=482159;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=10099;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for resources/cpu.yaml \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=628064;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=886454;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m config resources/cpu.yaml: In \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'resources/cpu'\u001b[0m: Could not \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m override \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'resources/hydra/launcher'\u001b[0m. No \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m match in the defaults list. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=832035;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=182042;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m hydra/profiling_gpu_multirun.yam \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m l last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=843502;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=121431;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m hydra/profiling_gpu_multirun.yam \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m l: In \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'hydra/profiling_gpu_multirun'\u001b[0m: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Could not override \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'hydra/hydra/launcher'\u001b[0m. No match \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m in the defaults list. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=110377;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=270279;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m hydra/profiling_cpu_vs_gpu.yaml \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=710479;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=254129;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m hydra/profiling_cpu_vs_gpu.yaml: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m In \u001b[32m'hydra/profiling_cpu_vs_gpu'\u001b[0m: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Could not override \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'hydra/hydra/launcher'\u001b[0m. No match \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m in the defaults list. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m46/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m58 \u001b[0m ]\n", + "Creating schemas for Hydra config files...\u001b[35m 2%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m1/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Config file resources/cpu.yaml \u001b]8;id=63512;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=552908;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#362\u001b\\\u001b[2m362\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m was modified, regenerating the \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m schema. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 2%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m1/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 75%\u001b[0m \u001b[91m━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m249 \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 75%\u001b[0m \u001b[91m━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m249 \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m44/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m133 \u001b[0m ]\n", " \u001b[31mit/s\u001b[0m \n", - "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/11/24 14:43:11]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=331996;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=767802;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:00:26]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=546460;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=812556;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=271388;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=490691;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=483169;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=300052;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 8214\n", - "Seed set to 8214\n", + "seed manually set to 59953\n", + "Seed set to 59953\n", "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", "GPU available: True (cuda), used: True\n", "TPU available: False, using: 0 TPU cores\n", "HPU available: False, using: 0 HPUs\n", "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/pytorch/trainer/connectors/logger_connector/logger_connector.py:75: Starting from v1.9.0, `tensorboardX` has been removed as a dependency of the `lightning.pytorch` package, due to potential conflicts with other packages in the ML ecosystem. For this reason, `logger=True` will use `CSVLogger` as the default logger, unless the `tensorboard` or `tensorboardX` packages are found. Please `pip install lightning[extra]` or one of them to enable TensorBoard support by default\n", - "\u001b[2;36m[14:43:12]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=669543;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=940579;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[11:00:26]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=170869;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=964538;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7f456e514340\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7ff3e73e0bb0\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", "You are using a CUDA device ('NVIDIA A100-SXM4-80GB MIG 3g.40gb') that has Tensor Cores. To properly utilize them, you should set `torch.set_float32_matmul_precision('medium' | 'high')` which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=633055;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=868539;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[11:00:27]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=930890;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=368284;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m[14:43:13]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=809000;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=571941;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=532901;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=167949;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", @@ -204,340 +190,340 @@ "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", "\u001b[1mModules in train mode\u001b[0m: 68 \n", "\u001b[1mModules in eval mode\u001b[0m: 0 \n", - "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m3.39it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", + "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m3.66it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", "s/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (30) \n", "is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower \n", "value for log_every_n_steps if you want to see logs for the training epoch.\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m217.574 \u001b[0m\n", + " \u001b[37m234.865 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m217.574 \u001b[0m\n", + " \u001b[37m234.865 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:10\u001b[0m \u001b[37m2.98it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m \u001b[37mv_num: 20.000\u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:10\u001b[0m \u001b[37m2.78it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m \u001b[37mv_num: 23.000\u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:08\u001b[0m \u001b[37m3.29it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.15it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:08\u001b[0m \u001b[37m3.29it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.15it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:01 • 0:00:08\u001b[0m \u001b[37m3.41it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:01 • 0:00:08\u001b[0m \u001b[37m3.41it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.52it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.52it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.59it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.59it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:02 • 0:00:06\u001b[0m \u001b[37m3.60it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.50it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:02 • 0:00:06\u001b[0m \u001b[37m3.60it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.50it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.71it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.63it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.71it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.63it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.76it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.67it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.76it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.67it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.77it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.66it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.77it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.66it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:04 • 0:00:04\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:04 • 0:00:04\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.57it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.57it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.05it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.59it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.05it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.59it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.10it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.10it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.12it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.12it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.14it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.14it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.16it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.63it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.16it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.63it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.18it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:06 • 0:00:02\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.18it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:06 • 0:00:02\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.26it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.67it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.26it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.67it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.29it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:07 • 0:00:01\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.29it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:07 • 0:00:01\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.33it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.33it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m239.891 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: 20.000 \u001b[0m\n", + " \u001b[37m229.398 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m266.122 \u001b[0m\n", + " \u001b[37m225.043 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m222.725 \u001b[0m\n", + " \u001b[37m209.168 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m208.827 \u001b[0m\n", + " \u001b[37m206.239 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.348 \u001b[0m\n", + " \u001b[37m240.032 \u001b[0m\n", "\u001b[?25hThe following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "\u001b[2;36m[14:43:46]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=931799;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=729770;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[11:00:59]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=723830;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=838534;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=52378;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=369808;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[11:01:00]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=439812;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=420927;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.37it/s\u001b[0m \u001b[37m3.37it/s\u001b[0m \n", + "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37m3.24it/s\u001b[0m \n", "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", - "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.48725733160972595 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 216.92962646484375 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.46772629022598267 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 210.00535583496094 \u001b[0m\u001b[35m \u001b[0m│\n", "└──────────────────────────────┴──────────────────────────────┘\n", - "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.37it/s\u001b[0m \n", - "\u001b[?25h\u001b[2;36m[14:44:06]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=535481;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=180588;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#165\u001b\\\u001b[2m165\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.25it/s\u001b[0m \n", + "\u001b[?25h\u001b[2;36m[11:01:20]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=488021;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=564713;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#165\u001b\\\u001b[2m165\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m loss metric. \u001b[2m \u001b[0m\n", - "val val/samples_per_second_epoch: \u001b[1;36m216.92962646484375\u001b[0m\n" + "val val/samples_per_second_epoch: \u001b[1;36m210.00535583496094\u001b[0m\n" ] } ], @@ -560,26 +546,26 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "In addition to callback specification, the Mila Research template integrates wandb as a logger specification, which has enables the tracking of additional metrics through visualizations and dashboard creation." + "In addition to callback specification, the Mila Research template integrates wandb as a logger specification, which enables the tracking of additional metrics through visualizations and dashboard creation." ] }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 7, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/11/24 14:45:32]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=182872;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=964435;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", + "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:20:50]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Config file \u001b]8;id=977657;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=927078;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#362\u001b\\\u001b[2m362\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m .yaml last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=108986;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=875791;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m .yaml was modified, regenerating \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m the schema. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=277302;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=189170;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", @@ -598,84 +584,51 @@ "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=346622;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=662566;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for resources/cpu.yaml \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=272758;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=311553;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m config resources/cpu.yaml: In \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'resources/cpu'\u001b[0m: Could not \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m override \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'resources/hydra/launcher'\u001b[0m. No \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m match in the defaults list. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=493766;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=805763;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m hydra/profiling_gpu_multirun.yam \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m l last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=610702;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=982945;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m hydra/profiling_gpu_multirun.yam \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m l: In \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'hydra/profiling_gpu_multirun'\u001b[0m: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Could not override \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'hydra/hydra/launcher'\u001b[0m. No match \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m in the defaults list. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=534702;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=374156;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m hydra/profiling_cpu_vs_gpu.yaml \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=260404;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=549776;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m hydra/profiling_cpu_vs_gpu.yaml: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m In \u001b[32m'hydra/profiling_cpu_vs_gpu'\u001b[0m: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Could not override \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'hydra/hydra/launcher'\u001b[0m. No match \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m in the defaults list. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m46/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m87 \u001b[0m ]\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m44/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m9,1…\u001b[0m ]\n", " \u001b[31mit/s\u001b[0m \n", - "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=880941;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=76069;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[1A\u001b[2K\u001b[1A\u001b[2K" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2;36m[09/12/24 11:22:43]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=487389;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=365722;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=979499;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=716702;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=827521;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=598585;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 73525\n", - "Seed set to 73525\n", + "seed manually set to 68143\n", + "Seed set to 68143\n", "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", "GPU available: True (cuda), used: True\n", "TPU available: False, using: 0 TPU cores\n", "HPU available: False, using: 0 HPUs\n", - "\u001b[2;36m[14:45:33]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=615378;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=155209;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[11:22:44]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=460901;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=404017;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7f4c95712470\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7f7b377eccd0\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", "You are using a CUDA device ('NVIDIA A100-SXM4-80GB MIG 3g.40gb') that has Tensor Cores. To properly utilize them, you should set `torch.set_float32_matmul_precision('medium' | 'high')` which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=978903;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=1269;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=114934;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=518424;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m[14:45:34]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=5270;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=603572;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=190722;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=233764;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mcesar-valdez\u001b[0m (\u001b[33mcesar-valdez-mcgill-university\u001b[0m). Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n", "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.17.8\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in \u001b[35m\u001b[1m./wandb/run-20240911_144537-cv9r056m\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in \u001b[35m\u001b[1m./wandb/run-20240912_112248-rkoyrnjk\u001b[0m\n", "\u001b[34m\u001b[1mwandb\u001b[0m: Run \u001b[1m`wandb offline`\u001b[0m to turn off syncing.\n", "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mdefault\u001b[0m\n", "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/cv9r056m\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/rkoyrnjk\u001b[0m\n", "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", @@ -698,368 +651,389 @@ "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", "\u001b[1mModules in train mode\u001b[0m: 68 \n", "\u001b[1mModules in eval mode\u001b[0m: 0 \n", - "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m3.03it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", + "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m3.38it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", "s/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (30) \n", "is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower \n", "value for log_every_n_steps if you want to see logs for the training epoch.\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m194.480 \u001b[0m\n", + " \u001b[37m227.765 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m194.480 \u001b[0m\n", + " \u001b[37m227.765 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:12\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m0m \u001b[37mv_num: 056m\u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:10\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m0m \u001b[37mv_num: rnjk\u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:02 • 0:00:12\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.06it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:02 • 0:00:12\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.06it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:10\u001b[0m \u001b[37m2.53it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.14it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:10\u001b[0m \u001b[37m2.53it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.14it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:09\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.28it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:09\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.28it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:09\u001b[0m \u001b[37m2.87it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:09\u001b[0m \u001b[37m2.87it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:03 • 0:00:08\u001b[0m \u001b[37m2.94it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:03 • 0:00:08\u001b[0m \u001b[37m2.94it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.03it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.37it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.03it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.37it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.12it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.12it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.27it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.53it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.27it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.53it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.52it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.52it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.19it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.19it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.26it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.46it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.26it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.46it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.27it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.27it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.32it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.32it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.33it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.33it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.36it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.36it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.37it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.37it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.50it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.50it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.42it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.51it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.42it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.51it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.54it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.54it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.55it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.55it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.46it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.57it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.46it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.57it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.57it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.57it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m230.148 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: 056m \u001b[0m\n", + " \u001b[37m227.828 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m256.431 \u001b[0m\n", + " \u001b[37m263.316 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m178.465 \u001b[0m\n", + " \u001b[37m217.226 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m201.937 \u001b[0m\n", + " \u001b[37m200.034 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m231.156 \u001b[0m\n", + " \u001b[37m233.353 \u001b[0m\n", "\u001b[?25h/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "\u001b[2;36m[14:46:09]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=96605;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=535638;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[11:23:19]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=909552;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=694999;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=206179;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=390492;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=730783;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=629164;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.33it/s\u001b[0m \u001b[37m3.32it/s\u001b[0m \n", + "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.19it/s\u001b[0m \u001b[37m3.19it/s\u001b[0m \n", "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", - "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4357786178588867 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 214.4339141845703 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.490227609872818 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 206.32473754882812 \u001b[0m\u001b[35m \u001b[0m│\n", "└──────────────────────────────┴──────────────────────────────┘\n", - "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.33it/s\u001b[0m \n", - "\u001b[?25h\u001b[2;36m[14:46:30]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=47061;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=655270;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#165\u001b\\\u001b[2m165\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.19it/s\u001b[0m \n", + "\u001b[?25h\u001b[2;36m[11:23:41]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=357854;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=808910;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#165\u001b\\\u001b[2m165\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m loss metric. \u001b[2m \u001b[0m\n", - "val val/samples_per_second_epoch: \u001b[1;36m214.4339141845703\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run history:\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: epoch ▁▁█\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: lr-SGD ▁\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch ▁\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step ▁▁▂▂▂▃▃▃▄▄▅▅▅▆▆▆▇▇███▁▂▂▂▃▃▄▄▄▅▅▅▆▆▆▇▇██\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss █▁\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_epoch ▁█\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_step ▁▇██▇█▅▄▆▆▄▇▄█▇▇▇▆▇▄▇▆▇▇██▆▅▆▆▇▇▅▇█▇▇▇▇▇\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run summary:\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: epoch 1\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: lr-SGD 0.123\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch 231.15637\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step 30\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss 0.43578\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_epoch 214.43391\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_step 226.7135\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run \u001b[33mdefault\u001b[0m at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/cv9r056m\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Synced 5 W&B file(s), 0 media file(s), 2 artifact file(s) and 0 other file(s)\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Find logs at: \u001b[35m\u001b[1m./wandb/run-20240911_144537-cv9r056m/logs\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m wandb version 0.17.9 is available! To upgrade, please run:\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m $ pip install wandb --upgrade\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m The new W&B backend becomes opt-out in version 0.18.0; try it out with `wandb.require(\"core\")`! See https://wandb.me/wandb-core for more information.\n" + "val val/samples_per_second_epoch: \u001b[1;36m206.32473754882812\u001b[0m\n", + "^C\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[32m\u001b[41mERROR\u001b[0m Control-C detected -- Run data was not synced\n", + "Traceback (most recent call last):\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\", line 177, in \n", + " main()\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/main.py\", line 94, in decorated_main\n", + " _run_hydra(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 394, in _run_hydra\n", + " _run_app(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 457, in _run_app\n", + " run_and_report(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 220, in run_and_report\n", + " return func()\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 458, in \n", + " lambda: hydra.run(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/hydra.py\", line 119, in run\n", + " ret = run_job(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/core/utils.py\", line 186, in run_job\n", + " ret.return_value = task_function(task_cfg)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\", line 71, in main\n", + " metric_name, objective, _metrics = run(experiment)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\", line 92, in run\n", + " wandb.finish()\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 4351, in finish\n", + " wandb.run.finish(exit_code=exit_code, quiet=quiet)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 452, in wrapper\n", + " return func(self, *args, **kwargs)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 393, in wrapper\n", + " return func(self, *args, **kwargs)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 2153, in finish\n", + " return self._finish(exit_code, quiet)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 2187, in _finish\n", + " self._atexit_cleanup(exit_code=exit_code)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 2436, in _atexit_cleanup\n", + " self._on_finish()\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 2699, in _on_finish\n", + " _ = exit_handle.wait(timeout=-1, on_progress=self._on_progress_exit)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/lib/mailbox.py\", line 283, in wait\n", + " found, abandoned = self._slot._get_and_clear(timeout=wait_timeout)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/lib/mailbox.py\", line 130, in _get_and_clear\n", + " if self._wait(timeout=timeout):\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/lib/mailbox.py\", line 126, in _wait\n", + " return self._event.wait(timeout=timeout)\n", + " File \"/home/mila/c/cesar.valdez/.rye/py/cpython@3.10.14/lib/python3.10/threading.py\", line 607, in wait\n", + " signaled = self._cond.wait(timeout)\n", + " File \"/home/mila/c/cesar.valdez/.rye/py/cpython@3.10.14/lib/python3.10/threading.py\", line 324, in wait\n", + " gotit = waiter.acquire(True, timeout)\n", + "KeyboardInterrupt\n" ] } ], @@ -1068,7 +1042,8 @@ " algorithm=no_op \\\n", " datamodule=imagenet \\\n", " trainer=profiling \\\n", - " trainer/logger=wandb" + " trainer/logger=wandb \\\n", + " trainer.logger.wandb.name=\"WandB logging test\" " ] }, { @@ -1089,29 +1064,31 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Finding a bottleneck is not necessarily clear or straightforward from the start. A sensible first step is to determine whether a potential slowdown originates from data loading or model computation. Querying both ends of the process can be done to determine whether the master process has a significant stall when fetching the next batch, or not. \n", - "If it's close to 0, then data loading outpaces compute, and compute is the bottleneck. \n", - "If it's much greater than 0, then compute outpaces data loading, and data loading is the bottleneck. \n", - "You might not care about CPU usage by the master process and data loaders, so long as the GPU remains fully utilized." + "Finding bottlenecks in your code is not necessarily clear or straightforward from the start. A sensible first step is to determine whether potential slowdowns originate from data loading or model computation. Running a model with and without training and contrasting the obtained outputs can help us determine whether the master process has a significant stall when fetching the next batch for training or not. Analyzing the difference between outputs can tell us the following about our model: \n", + "\n", + "- If the difference between data loading and training is close to 0, then the data loading procedure outpaces model computation, and computation is the bottleneck. \n", + "- If the difference between data loading and training is much greater than 0, then model computation outpaces data loading, and data loading is the bottleneck. \n", + "\n", + "To showcase the former, we will proceed to run two separate model loops on imagenet: the first one doing data loading without any training, followed by one with." ] }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/11/24 12:09:08]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=626004;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=19589;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", + "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:33:46]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Config file \u001b]8;id=248360;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=640404;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#362\u001b\\\u001b[2m362\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m .yaml last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=693194;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=60895;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m .yaml was modified, regenerating \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m the schema. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=157116;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=812221;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", @@ -1130,73 +1107,44 @@ "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=346486;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=788581;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for resources/cpu.yaml \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=322509;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=244602;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m config resources/cpu.yaml: In \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'resources/cpu'\u001b[0m: Could not \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m override \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'resources/hydra/launcher'\u001b[0m. No \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m match in the defaults list. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=971168;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=544750;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m hydra/profiling_multirun.yaml \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=100610;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=775759;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m hydra/profiling_multirun.yaml: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m In \u001b[32m'hydra/profiling_multirun'\u001b[0m: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Could not override \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'hydra/hydra/launcher'\u001b[0m. No match \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m in the defaults list. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=286899;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=677441;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for hydra/cpu_vs_gpu.yaml \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=454750;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=684550;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m config hydra/cpu_vs_gpu.yaml: In \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'hydra/cpu_vs_gpu'\u001b[0m: Could not \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m override \u001b[32m'hydra/hydra/launcher'\u001b[0m. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m No match in the defaults list. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 72%\u001b[0m \u001b[91m━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m46/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m104 \u001b[0m ]\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m44/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m8,5…\u001b[0m ]\n", " \u001b[31mit/s\u001b[0m \n", - "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/11/24 12:09:09]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=766136;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=160792;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:33:47]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=59367;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=581334;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=475361;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=860693;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=335963;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=390514;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 1633\n", - "Seed set to 1633\n", + "seed manually set to 4698\n", + "Seed set to 4698\n", "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", "GPU available: True (cuda), used: True\n", "TPU available: False, using: 0 TPU cores\n", "HPU available: False, using: 0 HPUs\n", - "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/pytorch/trainer/connectors/logger_connector/logger_connector.py:75: Starting from v1.9.0, `tensorboardX` has been removed as a dependency of the `lightning.pytorch` package, due to potential conflicts with other packages in the ML ecosystem. For this reason, `logger=True` will use `CSVLogger` as the default logger, unless the `tensorboard` or `tensorboardX` packages are found. Please `pip install lightning[extra]` or one of them to enable TensorBoard support by default\n", - "\u001b[2;36m[12:09:09]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=640269;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=664528;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[11:33:47]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=874387;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=504683;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7feedd34ffa0\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7f54a5cbf760\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", "You are using a CUDA device ('NVIDIA A100-SXM4-80GB MIG 3g.40gb') that has Tensor Cores. To properly utilize them, you should set `torch.set_float32_matmul_precision('medium' | 'high')` which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision\n", - "\u001b[2;36m[12:09:10]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=465463;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=673508;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[11:33:48]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=447340;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=420839;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=62465;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=50264;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=348156;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=474339;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mcesar-valdez\u001b[0m (\u001b[33mcesar-valdez-mcgill-university\u001b[0m). Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.17.8\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in \u001b[35m\u001b[1m./wandb/run-20240912_113352-7zvrvdg7\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run \u001b[1m`wandb offline`\u001b[0m to turn off syncing.\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mDataloading only\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/7zvrvdg7\u001b[0m\n", "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", @@ -1219,340 +1167,368 @@ "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", "\u001b[1mModules in train mode\u001b[0m: 68 \n", "\u001b[1mModules in eval mode\u001b[0m: 0 \n", - "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", + "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", "s/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (30) \n", "is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower \n", "value for log_every_n_steps if you want to see logs for the training epoch.\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m217.110 \u001b[0m\n", + " \u001b[37m228.319 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m217.110 \u001b[0m\n", + " \u001b[37m228.319 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:10\u001b[0m \u001b[37m2.88it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m \u001b[37mv_num: 19.000\u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:10\u001b[0m \u001b[37m2.80it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m0m \u001b[37mv_num: vdg7\u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.07it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.07it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:01 • 0:00:08\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:01 • 0:00:08\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.59it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.32it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.59it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.32it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.55it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.29it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.55it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.29it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:02 • 0:00:06\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:02 • 0:00:06\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.73it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.73it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.76it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.76it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.72it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.36it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.72it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.36it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:04 • 0:00:04\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:04 • 0:00:04\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.65it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.32it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.65it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.32it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:05 • 0:00:03\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:05 • 0:00:03\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:05 • 0:00:03\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:05 • 0:00:03\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.71it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.71it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:06 • 0:00:02\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:06 • 0:00:02\u001b[0m \u001b[37m3.70it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:06 • 0:00:02\u001b[0m \u001b[37m3.72it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:06 • 0:00:02\u001b[0m \u001b[37m3.72it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.73it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.73it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.75it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.18it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.75it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.18it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:07 • 0:00:01\u001b[0m \u001b[37m3.77it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:07 • 0:00:01\u001b[0m \u001b[37m3.77it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:07 • 0:00:01\u001b[0m \u001b[37m3.78it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:07 • 0:00:01\u001b[0m \u001b[37m3.78it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.77it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.77it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.653 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.79it/s\u001b[0m \u001b[37mv_num: 19.000 \u001b[0m\n", + " \u001b[37m200.945 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m280.538 \u001b[0m\n", + " \u001b[37m232.539 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m212.582 \u001b[0m\n", + " \u001b[37m218.772 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m215.113 \u001b[0m\n", + " \u001b[37m186.436 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m246.492 \u001b[0m\n", - "\u001b[?25hThe following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "\u001b[2;36m[12:09:41]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=848753;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=339553;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + " \u001b[37m214.588 \u001b[0m\n", + "\u001b[?25h/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", + "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", + "\u001b[2;36m[11:34:24]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=162286;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=523976;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=232176;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=670495;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[11:34:25]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=401060;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=345444;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37m3.40it/s\u001b[0m \n", + "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.13it/s\u001b[0m \u001b[37m3.12it/s\u001b[0m \n", "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", - "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4903430640697479 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 219.41116333007812 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.5581756830215454 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 201.9569091796875 \u001b[0m\u001b[35m \u001b[0m│\n", "└──────────────────────────────┴──────────────────────────────┘\n", - "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.40it/s\u001b[0m \n", - "\u001b[?25h\u001b[2;36m[12:10:01]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=977916;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=724301;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#165\u001b\\\u001b[2m165\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.13it/s\u001b[0m \n", + "\u001b[?25h\u001b[2;36m[11:34:46]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=109688;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=212673;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#165\u001b\\\u001b[2m165\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m loss metric. \u001b[2m \u001b[0m\n", - "val val/samples_per_second_epoch: \u001b[1;36m219.41116333007812\u001b[0m\n" + "val val/samples_per_second_epoch: \u001b[1;36m201.9569091796875\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run history:\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: epoch ▁▁█\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: lr-SGD ▁\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch ▁\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step ▁▁▂▂▂▃▃▃▄▄▅▅▅▆▆▆▇▇███▁▂▂▂▃▃▄▄▄▅▅▅▆▆▆▇▇██\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss ▁█\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_epoch ▁█\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_step ▁▆█▇▇▇▆▃▅▇▅▅▄▅▇▇▅▄▇▇▆▇█▄▇▇▅▄▆▆▇▆▆█▅▇▇▇▆▇\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run summary:\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: epoch 1\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: lr-SGD 0.123\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch 214.58818\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step 30\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss 0.55818\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_epoch 201.95691\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_step 213.31876\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run \u001b[33mDataloading only\u001b[0m at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/7zvrvdg7\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Synced 5 W&B file(s), 0 media file(s), 0 artifact file(s) and 0 other file(s)\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Find logs at: \u001b[35m\u001b[1m./wandb/run-20240912_113352-7zvrvdg7/logs\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m wandb version 0.18.0 is available! To upgrade, please run:\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m $ pip install wandb --upgrade\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m The new W&B backend becomes opt-out in version 0.18.0; try it out with `wandb.require(\"core\")`! See https://wandb.me/wandb-core for more information.\n" ] } ], @@ -1560,26 +1536,28 @@ "!python project/main.py \\\n", " algorithm=no_op \\\n", " datamodule=imagenet \\\n", - " trainer=profiling" + " trainer=profiling \\\n", + " trainer/logger=wandb \\\n", + " trainer.logger.wandb.name=\"Dataloading only\"" ] }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/10/24 16:16:47]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=791467;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=848502;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:35:07]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=207449;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=656180;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m .yaml last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/10/24 16:16:48]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=992568;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=88644;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:35:08]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=968768;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=228614;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", @@ -1598,42 +1576,43 @@ "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Config file hydra/default.yaml \u001b]8;id=434209;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=149573;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#362\u001b\\\u001b[2m362\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m was modified, regenerating the \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m schema. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 76%\u001b[0m \u001b[91m━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m34/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 76%\u001b[0m \u001b[91m━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m34/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m45/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m50 \u001b[0m ]\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m44/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m9,3…\u001b[0m ]\n", " \u001b[31mit/s\u001b[0m \n", - "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=499898;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=831070;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=203320;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=916240;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=697638;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=810789;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=919392;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=127866;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 56210\n", - "Seed set to 56210\n", + "seed manually set to 3755\n", + "Seed set to 3755\n", "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=example datamodule=imagene ...\n", "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", "GPU available: True (cuda), used: True\n", "TPU available: False, using: 0 TPU cores\n", "HPU available: False, using: 0 HPUs\n", - "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/pytorch/trainer/connectors/logger_connector/logger_connector.py:75: Starting from v1.9.0, `tensorboardX` has been removed as a dependency of the `lightning.pytorch` package, due to potential conflicts with other packages in the ML ecosystem. For this reason, `logger=True` will use `CSVLogger` as the default logger, unless the `tensorboard` or `tensorboardX` packages are found. Please `pip install lightning[extra]` or one of them to enable TensorBoard support by default\n", - "\u001b[2;36m[16:16:49]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=2388;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=591001;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[11:35:09]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=202501;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=864263;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7fb4dc9fca60\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m[16:16:54]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=961539;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=3680;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7f1d791c7f40\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "You are using a CUDA device ('NVIDIA A100-SXM4-80GB MIG 3g.40gb') that has Tensor Cores. To properly utilize them, you should set `torch.set_float32_matmul_precision('medium' | 'high')` which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision\n", + "\u001b[2;36m[11:35:13]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=647810;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=482525;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=34379;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=596652;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=658441;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=714596;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mcesar-valdez\u001b[0m (\u001b[33mcesar-valdez-mcgill-university\u001b[0m). Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.17.8\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in \u001b[35m\u001b[1m./wandb/run-20240912_113516-5szkhfvp\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run \u001b[1m`wandb offline`\u001b[0m to turn off syncing.\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mDataloading + Training\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/5szkhfvp\u001b[0m\n", "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━┓\n", "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35m In sizes\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35m Out sizes\u001b[0m\u001b[1;35m \u001b[0m┃\n", @@ -1666,340 +1645,371 @@ "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", "\u001b[1mModules in train mode\u001b[0m: 68 \n", "\u001b[1mModules in eval mode\u001b[0m: 0 \n", - "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m2.28it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", + "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m4.21it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", "s/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (30) \n", "is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower \n", "value for log_every_n_steps if you want to see logs for the training epoch.\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m146.211 \u001b[0m\n", + " \u001b[37m271.139 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m146.211 \u001b[0m\n", + " \u001b[37m271.139 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:02 • 0:00:12\u001b[0m \u001b[37m2.28it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m \u001b[37mv_num: 16.000\u001b[0m\n", + "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:07\u001b[0m \u001b[37m4.17it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m0m \u001b[37mv_num: hfvp\u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:02 • 0:00:11\u001b[0m \u001b[37m2.39it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.97it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:02 • 0:00:11\u001b[0m \u001b[37m2.39it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.97it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:03 • 0:00:11\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:03 • 0:00:11\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:03 • 0:00:11\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.55it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:03 • 0:00:11\u001b[0m \u001b[37m2.38it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.55it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:03 • 0:00:10\u001b[0m \u001b[37m2.40it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.53it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:03 • 0:00:10\u001b[0m \u001b[37m2.40it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.53it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:04 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:04 • 0:00:10\u001b[0m \u001b[37m2.33it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:04 • 0:00:10\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.41it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:04 • 0:00:10\u001b[0m \u001b[37m2.32it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.41it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:05 • 0:00:09\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.39it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:05 • 0:00:09\u001b[0m \u001b[37m2.34it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.39it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:05 • 0:00:09\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.38it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:05 • 0:00:09\u001b[0m \u001b[37m2.35it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.38it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:06 • 0:00:08\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.38it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:06 • 0:00:08\u001b[0m \u001b[37m2.37it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.38it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:06 • 0:00:08\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:05 • 0:00:06\u001b[0m \u001b[37m3.39it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:06 • 0:00:08\u001b[0m \u001b[37m2.36it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:05 • 0:00:06\u001b[0m \u001b[37m3.39it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:07 • 0:00:07\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:07 • 0:00:07\u001b[0m \u001b[37m2.31it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:07 • 0:00:07\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:07 • 0:00:07\u001b[0m \u001b[37m2.30it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:07 • 0:00:07\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:06 • 0:00:05\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:07 • 0:00:07\u001b[0m \u001b[37m2.29it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:06 • 0:00:05\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:08 • 0:00:06\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:06 • 0:00:05\u001b[0m \u001b[37m3.18it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:08 • 0:00:06\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:06 • 0:00:05\u001b[0m \u001b[37m3.18it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:09 • 0:00:06\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.19it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:09 • 0:00:06\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.19it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:09 • 0:00:05\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:07 • 0:00:04\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:09 • 0:00:05\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:07 • 0:00:04\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:09 • 0:00:05\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:07 • 0:00:04\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:09 • 0:00:05\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:07 • 0:00:04\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:10 • 0:00:05\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:10 • 0:00:05\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:10 • 0:00:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:10 • 0:00:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:08 • 0:00:03\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:11 • 0:00:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:08 • 0:00:03\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:11 • 0:00:04\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:08 • 0:00:03\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:11 • 0:00:03\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:11 • 0:00:03\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:12 • 0:00:03\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:12 • 0:00:03\u001b[0m \u001b[37m2.24it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:12 • 0:00:02\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:09 • 0:00:02\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:12 • 0:00:02\u001b[0m \u001b[37m2.23it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:09 • 0:00:02\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:12 • 0:00:02\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:12 • 0:00:02\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:13 • 0:00:01\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:13 • 0:00:01\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:13 • 0:00:01\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:10 • 0:00:01\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:13 • 0:00:01\u001b[0m \u001b[37m2.25it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:10 • 0:00:01\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m138.248 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:14 • 0:00:00\u001b[0m \u001b[37m2.26it/s\u001b[0m \u001b[37mv_num: 16.000 \u001b[0m\n", + " \u001b[37m177.496 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m158.500 \u001b[0m\n", + " \u001b[37m178.472 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m147.835 \u001b[0m\n", + " \u001b[37m189.712 \u001b[0m\n", " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m143.576 \u001b[0m\n", + " \u001b[37m189.159 \u001b[0m\n", " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m146.852 \u001b[0m\n", - "\u001b[?25h\u001b[2;36m[16:17:48]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=302525;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=670131;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + " \u001b[37m235.514 \u001b[0m\n", + "\u001b[?25h/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=example datamodule=imagene ...\n", + "\u001b[2;36m[11:35:53]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=306215;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=838732;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=923056;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=385077;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=346133;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=387093;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:13 • 0:00:00\u001b[0m \u001b[37m2.10it/s\u001b[0m \u001b[37m2.11it/s\u001b[0m \n", + "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.16it/s\u001b[0m \u001b[37m3.16it/s\u001b[0m \n", "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", - "│\u001b[36m \u001b[0m\u001b[36m val/accuracy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.0234375 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 8.902314186096191 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 137.38766479492188 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36m val/accuracy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.0 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 7.498830318450928 \u001b[0m\u001b[35m \u001b[0m│\n", + "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 203.5886688232422 \u001b[0m\u001b[35m \u001b[0m│\n", "└──────────────────────────────┴──────────────────────────────┘\n", - "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:13 • 0:00:00\u001b[0m \u001b[37m2.10it/s\u001b[0m \n", - "\u001b[?25hval accuracy: \u001b[1;36m2.3\u001b[0m%\n", - "val val/accuracy: \u001b[1;36m0.0234375\u001b[0m\n", - "val val/samples_per_second_epoch: \u001b[1;36m137.38766479492188\u001b[0m\n" + "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.16it/s\u001b[0m \n", + "\u001b[?25hval accuracy: \u001b[1;36m0.0\u001b[0m%\n", + "val val/accuracy: \u001b[1;36m0.0\u001b[0m\n", + "val val/samples_per_second_epoch: \u001b[1;36m203.5886688232422\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: W&B sync reduced upload amount by 18.3%\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run history:\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: epoch ▁▁█\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: lr-Adam ▁\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch ▁\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step ▁▁▂▂▂▃▃▃▄▄▅▅▅▆▆▆▇▇███▁▂▂▂▃▃▄▄▄▅▅▅▆▆▆▇▇██\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/accuracy ▁▁\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss ▁▁\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_epoch ▁█\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_step ▁▆█▆▇▆▅▄▅▃▄▅▄▃▇▅▆▆▆▆▇▆▇▇▇▇▅▄▅▆▇▆▄▇▇▆▇▆▇▆\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run summary:\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: epoch 1\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: lr-Adam 0.0003\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch 235.5143\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step 30\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/accuracy 0.0\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss 7.49883\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_epoch 203.58867\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_step 200.58806\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run \u001b[33mDataloading + Training\u001b[0m at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/5szkhfvp\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Synced 5 W&B file(s), 0 media file(s), 2 artifact file(s) and 0 other file(s)\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Find logs at: \u001b[35m\u001b[1m./wandb/run-20240912_113516-5szkhfvp/logs\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m wandb version 0.18.0 is available! To upgrade, please run:\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m $ pip install wandb --upgrade\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m The new W&B backend becomes opt-out in version 0.18.0; try it out with `wandb.require(\"core\")`! See https://wandb.me/wandb-core for more information.\n" ] } ], @@ -2007,28 +2017,71 @@ "!python project/main.py \\\n", " algorithm=example \\\n", " datamodule=imagenet \\\n", - " trainer=profiling" + " trainer=profiling \\\n", + " trainer/logger=wandb \\\n", + " trainer.logger.wandb.name=\"Dataloading + Training\"" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## GPU vs CPU" + "As evidenced in the former, adding training to our run results in a difference in the ballpark of 100 samples/s. This would indicate that we have a computation bottleneck." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "Do we even need a GPU? Compare speed using CPU only vs the slowest GPU available, for a low number of steps\n", - "If the CPU performance loosely comparable (for instance, only 1.5-2x slower) than with a GPU, then it might be worth considering! (LMK if this happens, one thing could be to try to increase the # of CPUs and measure performance scaling, then ship this kind of job to a DRAC cluster)\n", - "In most workflows, using a GPU actually helps a lot.\n" + "## Comparing throughput: GPU vs CPU model training" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Advancements in Graphical Processing Units (GPUs) are widely known to have enabled the deep learning revolution, particularly through faster computation, relative to CPUs. Given that we have the option to run both GPU and CPU workloads, let's compare their throughput. In most workflows, the speedup provided by a GPU is dramatic. For a few select workloads, particularly those with a low number of steps or lighter computation requirements, if a 1.5-2x slower performance is observed when using a CPU, as opposed to a GPU, the former may be worth considering, as they're a far less contested resource on the cluster and pose far fewer availability issues." + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2;36m[09/12/24 12:18:52]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=941276;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=775917;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m12\u001b[0m/\u001b[1;36m12\u001b[0m-\u001b[1;36m18\u001b[0m-\u001b[1;36m52\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=38093;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=703144;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[33malgorithm\u001b[0m=\u001b[35mexample\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mimagenet\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mcpu\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mtrainer\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer/\u001b[33mlogger\u001b[0m=\u001b[35mwandb\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mCPU\u001b[0m \u001b[2m \u001b[0m\n" + ] + } + ], + "source": [ + "#Compare speed using CPU only vs the slowest GPU available, for a low number of steps\n", + "!python project/main.py \\\n", + " algorithm=example \\\n", + " datamodule=imagenet \\\n", + " resources=cpu \\\n", + " trainer=profiling \\\n", + " trainer/logger=wandb \\\n", + " trainer.logger.wandb.name=\"CPU\" \\\n", + " trainer.logger.wandb.group=\"GPU vs CPU\"" ] }, { "cell_type": "code", - "execution_count": 23, + "execution_count": 15, "metadata": {}, "outputs": [ { @@ -2056,13 +2109,14 @@ " cfg = self.config_loader.load_configuration(\n", " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/config_loader_impl.py\", line 142, in load_configuration\n", " return self._load_configuration_impl(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/config_loader_impl.py\", line 253, in _load_configuration_impl\n", - " defaults_list = create_defaults_list(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/defaults_list.py\", line 752, in create_defaults_list\n", - " overrides.ensure_overrides_used()\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/defaults_list.py\", line 168, in ensure_overrides_used\n", - " raise ConfigCompositionException(msg)\n", - "hydra.errors.ConfigCompositionException: In 'resources/cpu': Could not override 'resources/hydra/launcher@hydra.launcher'. No match in the defaults list.\n" + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/config_loader_impl.py\", line 244, in _load_configuration_impl\n", + " parsed_overrides, caching_repo = self._parse_overrides_and_create_caching_repo(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/config_loader_impl.py\", line 228, in _parse_overrides_and_create_caching_repo\n", + " parsed_overrides = parser.parse_overrides(overrides=overrides)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/core/override_parser/overrides_parser.py\", line 96, in parse_overrides\n", + " raise OverrideParseException(\n", + "hydra.errors.OverrideParseException: mismatched input 'logger.wandb.name' expecting ID\n", + "See https://hydra.cc/docs/1.2/advanced/override_grammar/basic for details\n" ] } ], @@ -2071,40 +2125,23 @@ " algorithm=example \\\n", " datamodule=imagenet \\\n", " trainer=profiling \\\n", - " resources=cpu" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "!HYDRA_FULL_ERROR=1 python project/main.py \\\n", - " algorithm=example \\\n", - " datamodule=imagenet \\\n", - " trainer=profiling \\\n", - " resources=one_gpu" + " resources=one_gpu \\\n", + " trainer/logger=wandb \\\n", + " trainer.logger.wandb.name=\"GPU\" \\\n", + " trainer.logger.wandb.group=\"GPU vs CPU\"" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## Testing for throughput across GPUs" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "As the Mila Research template is built with hydra as a configuration manager, it integrates [Multi-runs](https://hydra.cc/docs/tutorials/basic/running_your_app/multi-run/) by default. This makes it possible to sweep over different parameters for profiling or throughput testing purposes or both. For example, suppose we wanted to figure out how different GPUs perform relative to each other. \n", - "[Mila's official documentation](https://docs.mila.quebec/Information.html) has a comprehensive rundown of the GPUs that are installed on the cluster. Typing ```savail``` on the command line, when logged into the cluster, shows their current availability. Testing their capacity can yield insights into their suitability for different training cases." + "[Mila's official documentation](https://docs.mila.quebec/Information.html) has a comprehensive rundown of the GPUs that are installed on the cluster. Typing ```savail``` on the command line when logged into the cluster, shows their current availability. Testing their capacity can yield insights into their suitability for different training workloads.\n", + "As the Mila Research template is built with hydra as a configuration manager, it integrates [Multi-runs](https://hydra.cc/docs/tutorials/basic/running_your_app/multi-run/) by default. This makes it possible to sweep over different parameters for profiling or throughput testing purposes or both. For example, suppose we wanted to figure out how different GPUs perform relative to each other. " ] }, { "cell_type": "code", - "execution_count": 1, + "execution_count": 8, "metadata": {}, "outputs": [ { @@ -2113,14 +2150,14 @@ "text": [ "GPU Avail / Total \n", "===============================\n", - "2g.20gb 12 / 48 \n", - "3g.40gb 0 / 48 \n", - "4g.40gb 0 / 24 \n", + "2g.20gb 3 / 48 \n", + "3g.40gb 1 / 48 \n", + "4g.40gb 1 / 24 \n", "a100 0 / 32 \n", "a100l 0 / 88 \n", "a6000 0 / 8 \n", - "rtx8000 8 / 408 \n", - "v100 2 / 56 \n" + "rtx8000 13 / 360 \n", + "v100 3 / 56 \n" ] } ], @@ -2132,14 +2169,12 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "As these jobs are part of the cluster, [Submitit](https://hydra.cc/docs/plugins/submitit_launcher/)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "We can observe the following prominent GPU classes: a100, a100l, a6000, rtx8000, v100 and MiG partitions with sizes 2g.20gb, 3g.40gb, 4g.40gb. \n", + "We can observe the following prominent GPU classes:\n", + "\n", + "- NVIDIA Tensor Core GPUs: A100, A100L, V100 (previous gen)\n", + "- NVIDIA RTX GPUs: A6000, RTX8000\n", + "- Multi-Instance GPU (MiG) partitions: 2g.20gb, 3g.40gb, 4g.40gb \n", + "\n", "We will now proceed to specify different GPUs over training runs and compare their throughput." ] }, @@ -2159,7 +2194,14 @@ "# or add a devicestatsmonitor in\n", "# and using different kinds of GPUs. \n", "\n", - "## salloc --gres=gpu:a100:1 -c 6 --mem=32G -t 48:00:00 --partition=unkillable" + "## salloc --gres=gpu:a100:1 -c 6 --mem=32G -t 48:00:00 --partition=unkillable\n", + "\n", + "!HYDRA_FULL_ERROR=1 python project/main.py \\\n", + " algorithm=example \\\n", + " datamodule=imagenet \\\n", + " trainer=profiling \\\n", + " resources=one_gpu \\\n", + " trainer/logger=wandb" ] }, { @@ -2273,6 +2315,16 @@ "print(profiler.key_averages().table(sort_by=\"cpu_time_total\", row_limit=10))\n" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [] + }, { "cell_type": "markdown", "metadata": {}, diff --git a/project/configs/config.yaml b/project/configs/config.yaml index 19150416..02edc596 100644 --- a/project/configs/config.yaml +++ b/project/configs/config.yaml @@ -10,7 +10,8 @@ defaults: - hydra: default.yaml # Allows launching LOTS of runs in parallel on a cluster thanks to the submitit launcher. - - resources: null + - optional resources: null + # experiment configs allow for version control of specific hyperparameters # e.g. best hyperparameters for given model and datamodule diff --git a/project/configs/experiment/cluster_sweep_example.yaml b/project/configs/experiment/cluster_sweep_example.yaml index 96992b8d..b9bae1c6 100644 --- a/project/configs/experiment/cluster_sweep_example.yaml +++ b/project/configs/experiment/cluster_sweep_example.yaml @@ -1,4 +1,5 @@ # @package _global_ +# yaml-language-server: $schema=../../../.schemas/experiment_cluster_sweep_example_schema.json defaults: - example.yaml - override /resources: one_gpu.yaml diff --git a/project/configs/hydra/profiling_cpu_vs_gpu.yaml b/project/configs/hydra/profiling_cpu_vs_gpu.yaml deleted file mode 100644 index e50348ef..00000000 --- a/project/configs/hydra/profiling_cpu_vs_gpu.yaml +++ /dev/null @@ -1,23 +0,0 @@ -# enable color logging -defaults: - - _self_ - - override hydra_logging: disabled - - override job_logging: disabled - - override hydra/launcher: submitit_slurm - -run: - # output directory, generated dynamically on each run - dir: logs/${name}/runs/${now:%Y-%m-%d}/${now:%H-%M-%S} -sweep: - dir: logs/${name}/multiruns/${now:%Y-%m-%d}/${now:%H-%M-%S} - subdir: ${hydra.job.num} - -verbose: False - -#submitit: - #salloc --gres=gpu:a100:1 -c 6 --mem=32G -t 48:00:00 --partition=unkillable - #gres: gpu:a100:1 - #cpus_per_gpu: 6 - #mem_per_cpu: 32 - #timeout_min: 180 - #partition: unkillable diff --git a/project/configs/hydra/profiling_gpu_multirun.yaml b/project/configs/hydra/profiling_gpu_multirun.yaml deleted file mode 100644 index e50348ef..00000000 --- a/project/configs/hydra/profiling_gpu_multirun.yaml +++ /dev/null @@ -1,23 +0,0 @@ -# enable color logging -defaults: - - _self_ - - override hydra_logging: disabled - - override job_logging: disabled - - override hydra/launcher: submitit_slurm - -run: - # output directory, generated dynamically on each run - dir: logs/${name}/runs/${now:%Y-%m-%d}/${now:%H-%M-%S} -sweep: - dir: logs/${name}/multiruns/${now:%Y-%m-%d}/${now:%H-%M-%S} - subdir: ${hydra.job.num} - -verbose: False - -#submitit: - #salloc --gres=gpu:a100:1 -c 6 --mem=32G -t 48:00:00 --partition=unkillable - #gres: gpu:a100:1 - #cpus_per_gpu: 6 - #mem_per_cpu: 32 - #timeout_min: 180 - #partition: unkillable diff --git a/project/configs/resources/cpu.yaml b/project/configs/resources/cpu.yaml index e411b889..0c781658 100644 --- a/project/configs/resources/cpu.yaml +++ b/project/configs/resources/cpu.yaml @@ -1,13 +1,10 @@ +# @package _global_ defaults: - - override hydra/launcher: submitit_slurm - -trainer: - accelerator: cpu - strategy: null - devices: 1 + - override /hydra/launcher: submitit_slurm hydra: mode: MULTIRUN launcher: gpus_per_node: null cpus_per_task: 1 + mem_gb: 16 diff --git a/project/configs/resources/one_gpu.yaml b/project/configs/resources/one_gpu.yaml index 9edcadc8..2a7215fd 100644 --- a/project/configs/resources/one_gpu.yaml +++ b/project/configs/resources/one_gpu.yaml @@ -1,8 +1,8 @@ # @package _global_ +# yaml-language-server: $schema=../../../.schemas/resources_one_gpu_schema.json defaults: - override /hydra/launcher: submitit_slurm trainer: - accelerator: gpu devices: 1 hydra: mode: MULTIRUN diff --git a/project/configs/trainer/default.yaml b/project/configs/trainer/default.yaml index f897d183..7558af3b 100644 --- a/project/configs/trainer/default.yaml +++ b/project/configs/trainer/default.yaml @@ -1,8 +1,9 @@ +# yaml-language-server: $schema=../../../.schemas/trainer_default_schema.json _target_: lightning.Trainer logger: null accelerator: auto strategy: auto -devices: 1 +devices: auto deterministic: true diff --git a/project/configs/trainer/profiling.yaml b/project/configs/trainer/profiling.yaml index 690c446c..eed1ff3e 100644 --- a/project/configs/trainer/profiling.yaml +++ b/project/configs/trainer/profiling.yaml @@ -1,4 +1,8 @@ +# yaml-language-server: $schema=../../../.schemas/trainer_profiling_schema.json _target_: lightning.Trainer max_epochs: 1 limit_train_batches: 30 limit_val_batches: 30 + +wandb: + tags: ["profiling"] diff --git a/project/main.py b/project/main.py index 42ca9338..6388c0af 100644 --- a/project/main.py +++ b/project/main.py @@ -11,16 +11,15 @@ import hydra import omegaconf import rich -import wandb from lightning import LightningDataModule from omegaconf import DictConfig +import wandb from project.configs import add_configs_to_hydra_store from project.configs.config import Config from project.experiment import Experiment, setup_experiment from project.utils.env_vars import REPO_ROOTDIR from project.utils.hydra_utils import resolve_dictconfig -from project.utils.utils import print_config if os.environ.get("CUDA_VISIBLE_DEVICES", "").startswith("MIG-"): # NOTE: Perhaps unsetting it would also work, but this works atm. @@ -39,7 +38,8 @@ ) def main(dict_config: DictConfig) -> dict: """Main entry point for training a model.""" - print_config(dict_config, resolve=False) + # print_config(dict_config, resolve=False) + ## add if statement contingent on global? env variables from project.utils.auto_schema import add_schemas_to_all_hydra_configs From e577f9586d82c09262e7534190e241c027c080af Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Mon, 16 Sep 2024 16:57:50 -0400 Subject: [PATCH 13/33] Cleaned notebook up, support for config parameters in wandb overview --- docs/examples/profiling.ipynb | 2313 ++++------------- project/configs/experiment/profiling_cpu.yaml | 15 + project/configs/experiment/profiling_gpu.yaml | 19 + project/configs/trainer/profiling.yaml | 8 - project/datamodules/vision.py | 2 +- project/main.py | 27 +- 6 files changed, 486 insertions(+), 1898 deletions(-) create mode 100644 project/configs/experiment/profiling_cpu.yaml create mode 100644 project/configs/experiment/profiling_gpu.yaml delete mode 100644 project/configs/trainer/profiling.yaml diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 79e18c74..ec180172 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -76,42 +76,21 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 15, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:00:24]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Config file config.yaml was \u001b]8;id=986330;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=4653;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#362\u001b\\\u001b[2m362\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m modified, regenerating the \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m schema. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:00:25]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Config file \u001b]8;id=694205;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=36089;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#362\u001b\\\u001b[2m362\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/16/24 16:09:10]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=800276;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=394219;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m .yaml was modified, regenerating \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m the schema. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 2%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m1/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=623339;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=998599;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m .yaml last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=497869;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=538994;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", @@ -130,404 +109,45 @@ "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 2%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m1/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Config file resources/cpu.yaml \u001b]8;id=63512;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=552908;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#362\u001b\\\u001b[2m362\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m was modified, regenerating the \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m schema. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 2%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m1/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 75%\u001b[0m \u001b[91m━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m249 \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 75%\u001b[0m \u001b[91m━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[32m33/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m249 \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m44/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m133 \u001b[0m ]\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m46/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m7,5…\u001b[0m ]\n", " \u001b[31mit/s\u001b[0m \n", - "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:00:26]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=546460;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=812556;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/16/24 16:09:11]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=35654;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=593628;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=483169;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=300052;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=907199;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=326931;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 59953\n", - "Seed set to 59953\n", + "seed manually set to 30897\n", + "Seed set to 30897\n", "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", - "GPU available: True (cuda), used: True\n", + "GPU available: False, used: False\n", "TPU available: False, using: 0 TPU cores\n", "HPU available: False, using: 0 HPUs\n", "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/pytorch/trainer/connectors/logger_connector/logger_connector.py:75: Starting from v1.9.0, `tensorboardX` has been removed as a dependency of the `lightning.pytorch` package, due to potential conflicts with other packages in the ML ecosystem. For this reason, `logger=True` will use `CSVLogger` as the default logger, unless the `tensorboard` or `tensorboardX` packages are found. Please `pip install lightning[extra]` or one of them to enable TensorBoard support by default\n", - "\u001b[2;36m[11:00:26]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=170869;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=964538;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[16:09:11]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=422359;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=276611;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7ff3e73e0bb0\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7fc41bf0de40\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "You are using a CUDA device ('NVIDIA A100-SXM4-80GB MIG 3g.40gb') that has Tensor Cores. To properly utilize them, you should set `torch.set_float32_matmul_precision('medium' | 'high')` which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision\n", - "\u001b[2;36m[11:00:27]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=930890;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=368284;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[16:09:12]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=765323;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=527622;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=532901;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=167949;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=300968;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=508269;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", - "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", - "┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━┩\n", - "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNorm2d │ 128 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequential │ 147 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequential │ 525 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequential │ 2.1 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequential │ 8.4 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ AdaptiveAvgPool2d │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\n", - "└────┴─────────────────┴───────────────────┴────────┴───────┘\n", - "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", - "\u001b[1mNon-trainable params\u001b[0m: 0 \n", - "\u001b[1mTotal params\u001b[0m: 11.7 M \n", - "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", - "\u001b[1mModules in train mode\u001b[0m: 68 \n", - "\u001b[1mModules in eval mode\u001b[0m: 0 \n", - "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m3.66it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", - "s/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (30) \n", - "is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower \n", - "value for log_every_n_steps if you want to see logs for the training epoch.\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m234.865 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m234.865 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:10\u001b[0m \u001b[37m2.78it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m \u001b[37mv_num: 23.000\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.15it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.15it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.50it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.50it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.63it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.63it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.67it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:05\u001b[0m \u001b[37m3.67it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.66it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.66it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:04 • 0:00:04\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:04 • 0:00:04\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.57it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.57it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.59it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.59it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.62it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.63it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.63it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:06 • 0:00:02\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:06 • 0:00:02\u001b[0m \u001b[37m3.61it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.64it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.67it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.67it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:07 • 0:00:01\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:07 • 0:00:01\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m229.398 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.68it/s\u001b[0m \u001b[37mv_num: 23.000 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m225.043 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m209.168 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m206.239 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m240.032 \u001b[0m\n", - "\u001b[?25hThe following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "\u001b[2;36m[11:00:59]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=723830;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=838534;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m[11:01:00]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=439812;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=420927;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37m3.24it/s\u001b[0m \n", - "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", - "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", - "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.46772629022598267 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 210.00535583496094 \u001b[0m\u001b[35m \u001b[0m│\n", - "└──────────────────────────────┴──────────────────────────────┘\n", - "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.25it/s\u001b[0m \n", - "\u001b[?25h\u001b[2;36m[11:01:20]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=488021;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=564713;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#165\u001b\\\u001b[2m165\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m loss metric. \u001b[2m \u001b[0m\n", - "val val/samples_per_second_epoch: \u001b[1;36m210.00535583496094\u001b[0m\n" + "^C\n", + "\n", + "Detected KeyboardInterrupt, attempting graceful shutdown ...\n" ] } ], "source": [ + "#%%capture\n", "!python project/main.py \\\n", " algorithm=no_op \\\n", " datamodule=imagenet \\\n", @@ -539,33 +159,33 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### Optional: log metrics on wandb" + "### Logging metrics on WandB" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "In addition to callback specification, the Mila Research template integrates wandb as a logger specification, which enables the tracking of additional metrics through visualizations and dashboard creation." + "In addition to callback specification, the Mila Research template integrates wandb as a logger specification, which enables the tracking of additional metrics through visualizations and dashboard creation. Given the flexibility and widespread adoption of using WandB as a logger, we'll be using it for the remainder of this tutorial." ] }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 18, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:20:50]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Config file \u001b]8;id=977657;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=927078;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#362\u001b\\\u001b[2m362\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/16/24 16:31:23]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=965865;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=902936;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m .yaml was modified, regenerating \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m the schema. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=277302;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=189170;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m .yaml last time. Trying again. \u001b[2m \u001b[0m\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=381213;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=757802;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", @@ -584,52 +204,44 @@ "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m44/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m9,1…\u001b[0m ]\n", + "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", + "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m45/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m8,5…\u001b[0m ]\n", " \u001b[31mit/s\u001b[0m \n", - "\u001b[1A\u001b[2K\u001b[1A\u001b[2K" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2;36m[09/12/24 11:22:43]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=487389;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=365722;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/16/24 16:31:24]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=106742;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=184714;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=827521;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=598585;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=771749;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=511000;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 68143\n", - "Seed set to 68143\n", - "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", + "seed manually set to 77069\n", + "Seed set to 77069\n", + "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py experiment=profiling_cpu trainer/log ...\n", "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", - "GPU available: True (cuda), used: True\n", + "GPU available: False, used: False\n", "TPU available: False, using: 0 TPU cores\n", "HPU available: False, using: 0 HPUs\n", - "\u001b[2;36m[11:22:44]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=460901;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=404017;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", + "`Trainer(limit_val_batches=1)` was configured so 1 batch will be used.\n", + "\u001b[2;36m[16:31:25]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=903950;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=805041;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7f7b377eccd0\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7f40d1f33ac0\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "You are using a CUDA device ('NVIDIA A100-SXM4-80GB MIG 3g.40gb') that has Tensor Cores. To properly utilize them, you should set `torch.set_float32_matmul_precision('medium' | 'high')` which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=114934;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=518424;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=618681;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=810172;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=190722;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=233764;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=485875;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=801119;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mcesar-valdez\u001b[0m (\u001b[33mcesar-valdez-mcgill-university\u001b[0m). Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n", "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.17.8\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in \u001b[35m\u001b[1m./wandb/run-20240912_112248-rkoyrnjk\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in \u001b[35m\u001b[1m./wandb/run-20240916_163130-bd8pa1aw\u001b[0m\n", "\u001b[34m\u001b[1mwandb\u001b[0m: Run \u001b[1m`wandb offline`\u001b[0m to turn off syncing.\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mdefault\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mWandB logging test\u001b[0m\n", "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/rkoyrnjk\u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/bd8pa1aw\u001b[0m\n", "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", "┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━┩\n", @@ -651,399 +263,108 @@ "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", "\u001b[1mModules in train mode\u001b[0m: 68 \n", "\u001b[1mModules in eval mode\u001b[0m: 0 \n", - "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m3.38it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", - "s/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (30) \n", - "is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower \n", - "value for log_every_n_steps if you want to see logs for the training epoch.\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m227.765 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m227.765 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:10\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m0m \u001b[37mv_num: rnjk\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.06it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.06it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.14it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.14it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.28it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.28it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.35it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.37it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.37it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.53it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.53it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.52it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.52it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.46it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.46it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.47it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.48it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.50it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.50it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.51it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.51it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.54it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.54it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.55it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.55it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.57it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.57it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.57it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.57it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m227.828 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:08 • 0:00:00\u001b[0m \u001b[37m3.58it/s\u001b[0m \u001b[37mv_num: rnjk \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m263.316 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m217.226 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m200.034 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m233.353 \u001b[0m\n", - "\u001b[?25h/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", + "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (2) is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower value for log_every_n_steps if you want to see logs for the training epoch.\n", + "\u001b[2K\u001b[37mEpoch 0/1 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m4.45it/s\u001b[0m \u001b[37mv_num: a1aw\u001b[0m37mv_num: a1aw\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/1 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m4.45it/s\u001b[0m \u001b[37mv_num: a1aw\u001b[0m\n", + "\u001b[2KEpoch 1/1 \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/2\u001b[0m \u001b[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m7mv_num: a1aw\u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/2\u001b[0m \u001b[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m1/2\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m1/2\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m286.046 \u001b[0m`Trainer.fit` stopped: `max_epochs=2` reached.\n", + "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m175.405 \u001b[0m\n", + " \u001b[37mtrain/samples_per_…\u001b[0m\n", + " \u001b[37m175.405 \u001b[0m\n", + "\u001b[?25h/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py experiment=profiling_cpu trainer/log ...\n", "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "\u001b[2;36m[11:23:19]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=909552;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=694999;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[16:31:49]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=817434;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=930994;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=730783;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=629164;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[16:31:50]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=584010;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=413818;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.19it/s\u001b[0m \u001b[37m3.19it/s\u001b[0m \n", - "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", - "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", - "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.490227609872818 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 206.32473754882812 \u001b[0m\u001b[35m \u001b[0m│\n", - "└──────────────────────────────┴──────────────────────────────┘\n", - "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.19it/s\u001b[0m \n", - "\u001b[?25h\u001b[2;36m[11:23:41]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=357854;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=808910;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#165\u001b\\\u001b[2m165\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓━━━━\u001b[0m \u001b[37m1/1\u001b[0m \u001b[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", + "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", + "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", + "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.3719692826271057 \u001b[0m\u001b[35m \u001b[0m│\n", + "└───────────────────────────┴───────────────────────────┘\n", + "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/1\u001b[0m \u001b[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", + "\u001b[?25h\u001b[2;36m[16:32:08]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=785781;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=487120;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#160\u001b\\\u001b[2m160\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m loss metric. \u001b[2m \u001b[0m\n", - "val val/samples_per_second_epoch: \u001b[1;36m206.32473754882812\u001b[0m\n", - "^C\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[32m\u001b[41mERROR\u001b[0m Control-C detected -- Run data was not synced\n", - "Traceback (most recent call last):\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\", line 177, in \n", - " main()\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/main.py\", line 94, in decorated_main\n", - " _run_hydra(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 394, in _run_hydra\n", - " _run_app(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 457, in _run_app\n", - " run_and_report(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 220, in run_and_report\n", - " return func()\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 458, in \n", - " lambda: hydra.run(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/hydra.py\", line 119, in run\n", - " ret = run_job(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/core/utils.py\", line 186, in run_job\n", - " ret.return_value = task_function(task_cfg)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\", line 71, in main\n", - " metric_name, objective, _metrics = run(experiment)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\", line 92, in run\n", - " wandb.finish()\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 4351, in finish\n", - " wandb.run.finish(exit_code=exit_code, quiet=quiet)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 452, in wrapper\n", - " return func(self, *args, **kwargs)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 393, in wrapper\n", - " return func(self, *args, **kwargs)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 2153, in finish\n", - " return self._finish(exit_code, quiet)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 2187, in _finish\n", - " self._atexit_cleanup(exit_code=exit_code)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 2436, in _atexit_cleanup\n", - " self._on_finish()\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/wandb_run.py\", line 2699, in _on_finish\n", - " _ = exit_handle.wait(timeout=-1, on_progress=self._on_progress_exit)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/lib/mailbox.py\", line 283, in wait\n", - " found, abandoned = self._slot._get_and_clear(timeout=wait_timeout)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/lib/mailbox.py\", line 130, in _get_and_clear\n", - " if self._wait(timeout=timeout):\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/wandb/sdk/lib/mailbox.py\", line 126, in _wait\n", - " return self._event.wait(timeout=timeout)\n", - " File \"/home/mila/c/cesar.valdez/.rye/py/cpython@3.10.14/lib/python3.10/threading.py\", line 607, in wait\n", - " signaled = self._cond.wait(timeout)\n", - " File \"/home/mila/c/cesar.valdez/.rye/py/cpython@3.10.14/lib/python3.10/threading.py\", line 324, in wait\n", - " gotit = waiter.acquire(True, timeout)\n", - "KeyboardInterrupt\n" + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: W&B sync reduced upload amount by 21.7%\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run history:\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: epoch ▁▁▅▅█\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: lr-SGD ▁▁\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch █▁\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step ▁▃▃▅▆▆█\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss █▁▄\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Run summary:\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: epoch 2\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: lr-SGD 0.123\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch 175.40546\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step 4\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss 0.37197\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \n", + "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run \u001b[33mWandB logging test\u001b[0m at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/bd8pa1aw\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Synced 5 W&B file(s), 0 media file(s), 2 artifact file(s) and 0 other file(s)\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Find logs at: \u001b[35m\u001b[1m./wandb/run-20240916_163130-bd8pa1aw/logs\u001b[0m\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m wandb version 0.18.0 is available! To upgrade, please run:\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m $ pip install wandb --upgrade\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m The new W&B backend becomes opt-out in version 0.18.0; try it out with `wandb.require(\"core\")`! See https://wandb.me/wandb-core for more information.\n" ] } ], "source": [ + "%%capture\n", "!python project/main.py \\\n", - " algorithm=no_op \\\n", - " datamodule=imagenet \\\n", - " trainer=profiling \\\n", + " experiment=profiling_cpu \\\n", " trainer/logger=wandb \\\n", - " trainer.logger.wandb.name=\"WandB logging test\" " + " trainer.logger.wandb.name=\"WandB logging test\"" ] }, { @@ -1074,952 +395,81 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 21, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:33:46]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Config file \u001b]8;id=248360;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=640404;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#362\u001b\\\u001b[2m362\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m .yaml was modified, regenerating \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m the schema. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=157116;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=812221;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m not find \u001b[32m'hydra/sweeper/orion'\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Available options in \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'hydra/sweeper'\u001b[0m: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m basic \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Config search path: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra.conf\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mmain\u001b[0m, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mproject.configs\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m-colorlog, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra_plugins.hydra_c\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m44/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m8,5…\u001b[0m ]\n", - " \u001b[31mit/s\u001b[0m \n", - "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:33:47]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=59367;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=581334;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=335963;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=390514;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 4698\n", - "Seed set to 4698\n", - "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", - "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", - "GPU available: True (cuda), used: True\n", - "TPU available: False, using: 0 TPU cores\n", - "HPU available: False, using: 0 HPUs\n", - "\u001b[2;36m[11:33:47]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=874387;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=504683;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7f54a5cbf760\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", - "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "You are using a CUDA device ('NVIDIA A100-SXM4-80GB MIG 3g.40gb') that has Tensor Cores. To properly utilize them, you should set `torch.set_float32_matmul_precision('medium' | 'high')` which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision\n", - "\u001b[2;36m[11:33:48]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=447340;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=420839;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=348156;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=474339;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mcesar-valdez\u001b[0m (\u001b[33mcesar-valdez-mcgill-university\u001b[0m). Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.17.8\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in \u001b[35m\u001b[1m./wandb/run-20240912_113352-7zvrvdg7\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run \u001b[1m`wandb offline`\u001b[0m to turn off syncing.\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mDataloading only\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/7zvrvdg7\u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", - "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", - "┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━┩\n", - "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNorm2d │ 128 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequential │ 147 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequential │ 525 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequential │ 2.1 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequential │ 8.4 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ AdaptiveAvgPool2d │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\n", - "└────┴─────────────────┴───────────────────┴────────┴───────┘\n", - "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", - "\u001b[1mNon-trainable params\u001b[0m: 0 \n", - "\u001b[1mTotal params\u001b[0m: 11.7 M \n", - "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", - "\u001b[1mModules in train mode\u001b[0m: 68 \n", - "\u001b[1mModules in eval mode\u001b[0m: 0 \n", - "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", - "s/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (30) \n", - "is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower \n", - "value for log_every_n_steps if you want to see logs for the training epoch.\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m228.319 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m228.319 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:10\u001b[0m \u001b[37m2.80it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m0m \u001b[37mv_num: vdg7\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.07it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:01 • 0:00:09\u001b[0m \u001b[37m3.07it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:08\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.32it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.32it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.29it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.29it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:03 • 0:00:06\u001b[0m \u001b[37m3.45it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.49it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.40it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.36it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:04 • 0:00:05\u001b[0m \u001b[37m3.36it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.32it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.32it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:05 • 0:00:04\u001b[0m \u001b[37m3.31it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.34it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:06 • 0:00:03\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:07 • 0:00:02\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.18it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.18it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:08 • 0:00:01\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m200.945 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: vdg7 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m232.539 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m218.772 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m186.436 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m214.588 \u001b[0m\n", - "\u001b[?25h/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", - "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "\u001b[2;36m[11:34:24]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=162286;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=523976;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m[11:34:25]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=401060;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=345444;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.13it/s\u001b[0m \u001b[37m3.12it/s\u001b[0m \n", - "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", - "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", - "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.5581756830215454 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 201.9569091796875 \u001b[0m\u001b[35m \u001b[0m│\n", - "└──────────────────────────────┴──────────────────────────────┘\n", - "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.13it/s\u001b[0m \n", - "\u001b[?25h\u001b[2;36m[11:34:46]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=109688;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=212673;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#165\u001b\\\u001b[2m165\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m loss metric. \u001b[2m \u001b[0m\n", - "val val/samples_per_second_epoch: \u001b[1;36m201.9569091796875\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run history:\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: epoch ▁▁█\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: lr-SGD ▁\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch ▁\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step ▁▁▂▂▂▃▃▃▄▄▅▅▅▆▆▆▇▇███▁▂▂▂▃▃▄▄▄▅▅▅▆▆▆▇▇██\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss ▁█\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_epoch ▁█\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_step ▁▆█▇▇▇▆▃▅▇▅▅▄▅▇▇▅▄▇▇▆▇█▄▇▇▅▄▆▆▇▆▆█▅▇▇▇▆▇\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run summary:\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: epoch 1\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: lr-SGD 0.123\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch 214.58818\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step 30\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss 0.55818\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_epoch 201.95691\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_step 213.31876\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run \u001b[33mDataloading only\u001b[0m at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/7zvrvdg7\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Synced 5 W&B file(s), 0 media file(s), 0 artifact file(s) and 0 other file(s)\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Find logs at: \u001b[35m\u001b[1m./wandb/run-20240912_113352-7zvrvdg7/logs\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m wandb version 0.18.0 is available! To upgrade, please run:\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m $ pip install wandb --upgrade\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m The new W&B backend becomes opt-out in version 0.18.0; try it out with `wandb.require(\"core\")`! See https://wandb.me/wandb-core for more information.\n" + "\u001b[2;36m[09/16/24 16:44:50]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=44774;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=174608;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m16\u001b[0m-\u001b[1;36m44\u001b[0m-\u001b[1;36m50\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=55055;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=351128;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[33mexperiment\u001b[0m=\u001b[35mprofiling_cpu\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mDataloading\u001b[0m\\ only \u001b[2m \u001b[0m\n" ] } ], "source": [ + "%%capture\n", "!python project/main.py \\\n", - " algorithm=no_op \\\n", - " datamodule=imagenet \\\n", - " trainer=profiling \\\n", - " trainer/logger=wandb \\\n", + " experiment=profiling_cpu \\\n", " trainer.logger.wandb.name=\"Dataloading only\"" ] }, { "cell_type": "code", - "execution_count": 10, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:35:07]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=207449;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=656180;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m .yaml last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/12/24 11:35:08]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=968768;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=228614;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m not find \u001b[32m'hydra/sweeper/orion'\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Available options in \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'hydra/sweeper'\u001b[0m: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m basic \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Config search path: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra.conf\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mmain\u001b[0m, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mproject.configs\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m-colorlog, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra_plugins.hydra_c\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/44\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m44/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m9,3…\u001b[0m ]\n", - " \u001b[31mit/s\u001b[0m \n", - "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=203320;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=916240;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=919392;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=127866;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 3755\n", - "Seed set to 3755\n", - "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=example datamodule=imagene ...\n", - "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", - "GPU available: True (cuda), used: True\n", - "TPU available: False, using: 0 TPU cores\n", - "HPU available: False, using: 0 HPUs\n", - "\u001b[2;36m[11:35:09]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=202501;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=864263;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7f1d791c7f40\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", - "You are using a CUDA device ('NVIDIA A100-SXM4-80GB MIG 3g.40gb') that has Tensor Cores. To properly utilize them, you should set `torch.set_float32_matmul_precision('medium' | 'high')` which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision\n", - "\u001b[2;36m[11:35:13]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=647810;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=482525;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=658441;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=714596;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mcesar-valdez\u001b[0m (\u001b[33mcesar-valdez-mcgill-university\u001b[0m). Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.17.8\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in \u001b[35m\u001b[1m./wandb/run-20240912_113516-5szkhfvp\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run \u001b[1m`wandb offline`\u001b[0m to turn off syncing.\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mDataloading + Training\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/5szkhfvp\u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━┓\n", - "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35m In sizes\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35m Out sizes\u001b[0m\u001b[1;35m \u001b[0m┃\n", - "┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━┩\n", - "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\u001b[37m \u001b[0m\u001b[37m [64, 3,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 1000]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224, 224]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\u001b[37m \u001b[0m\u001b[37m [64, 3,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 224, 224]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNorm… │ 128 │ train │\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112, 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112, 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 112, 112]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 112, 112]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequential │ 147 K │ train │\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequential │ 525 K │ train │\u001b[37m \u001b[0m\u001b[37m [64, 64,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 128,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 56, 56]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 28, 28]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequential │ 2.1 M │ train │\u001b[37m \u001b[0m\u001b[37m [64, 128,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 256,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 28, 28]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 14, 14]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequential │ 8.4 M │ train │\u001b[37m \u001b[0m\u001b[37m [64, 256,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 512,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 14, 14]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 7, 7]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ AdaptiveA… │ 0 │ train │\u001b[37m \u001b[0m\u001b[37m [64, 512,\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m [64, 512,\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m│ │ │ │ │\u001b[37m \u001b[0m\u001b[37m 7, 7]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m 1, 1]\u001b[0m\u001b[37m \u001b[0m│\n", - "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\u001b[37m \u001b[0m\u001b[37m [64, 512]\u001b[0m\u001b[37m \u001b[0m│\u001b[37m \u001b[0m\u001b[37m[64, 1000]\u001b[0m\u001b[37m \u001b[0m│\n", - "└────┴─────────────────┴────────────┴────────┴───────┴────────────┴────────────┘\n", - "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", - "\u001b[1mNon-trainable params\u001b[0m: 0 \n", - "\u001b[1mTotal params\u001b[0m: 11.7 M \n", - "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", - "\u001b[1mModules in train mode\u001b[0m: 68 \n", - "\u001b[1mModules in eval mode\u001b[0m: 0 \n", - "\u001b[2K/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-package[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m4.21it/s\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", - "s/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (30) \n", - "is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower \n", - "value for log_every_n_steps if you want to see logs for the training epoch.\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/30\u001b[0m \u001b[37m0:00:00 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m271.139 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/30\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m271.139 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - "\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m3/30\u001b[0m \u001b[37m0:00:01 • 0:00:07\u001b[0m \u001b[37m4.17it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m0m \u001b[37mv_num: hfvp\u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.97it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━━\u001b[0m \u001b[37m4/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.97it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m5/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.69it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.55it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━━\u001b[0m \u001b[37m6/30\u001b[0m \u001b[37m0:00:02 • 0:00:07\u001b[0m \u001b[37m3.55it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.53it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m7/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.53it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━━━\u001b[0m \u001b[37m8/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.43it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.41it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━━━\u001b[0m \u001b[37m9/30\u001b[0m \u001b[37m0:00:03 • 0:00:07\u001b[0m \u001b[37m3.41it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.39it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━━\u001b[0m \u001b[37m10/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.39it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.38it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m11/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.38it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.38it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━━\u001b[0m \u001b[37m12/30\u001b[0m \u001b[37m0:00:04 • 0:00:06\u001b[0m \u001b[37m3.38it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:05 • 0:00:06\u001b[0m \u001b[37m3.39it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m13/30\u001b[0m \u001b[37m0:00:05 • 0:00:06\u001b[0m \u001b[37m3.39it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━━\u001b[0m \u001b[37m14/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m15/30\u001b[0m \u001b[37m0:00:05 • 0:00:05\u001b[0m \u001b[37m3.24it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:06 • 0:00:05\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m16/30\u001b[0m \u001b[37m0:00:06 • 0:00:05\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:06 • 0:00:05\u001b[0m \u001b[37m3.18it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━━\u001b[0m \u001b[37m17/30\u001b[0m \u001b[37m0:00:06 • 0:00:05\u001b[0m \u001b[37m3.18it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.19it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m18/30\u001b[0m \u001b[37m0:00:06 • 0:00:04\u001b[0m \u001b[37m3.19it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:07 • 0:00:04\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━━\u001b[0m \u001b[37m19/30\u001b[0m \u001b[37m0:00:07 • 0:00:04\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:07 • 0:00:04\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m20/30\u001b[0m \u001b[37m0:00:07 • 0:00:04\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[37m21/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:07 • 0:00:03\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━\u001b[0m \u001b[37m22/30\u001b[0m \u001b[37m0:00:08 • 0:00:03\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:08 • 0:00:03\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m23/30\u001b[0m \u001b[37m0:00:08 • 0:00:03\u001b[0m \u001b[37m3.21it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━━\u001b[0m \u001b[37m24/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m25/30\u001b[0m \u001b[37m0:00:08 • 0:00:02\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:09 • 0:00:02\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━━\u001b[0m \u001b[37m26/30\u001b[0m \u001b[37m0:00:09 • 0:00:02\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━\u001b[0m \u001b[37m27/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.23it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━\u001b[0m\u001b[35m╸\u001b[0m\u001b[90m━\u001b[0m \u001b[37m28/30\u001b[0m \u001b[37m0:00:09 • 0:00:01\u001b[0m \u001b[37m3.25it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:10 • 0:00:01\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m \u001b[37m29/30\u001b[0m \u001b[37m0:00:10 • 0:00:01\u001b[0m \u001b[37m3.22it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m177.496 \u001b[0m`Trainer.fit` stopped: `max_epochs=1` reached.\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/0 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:10 • 0:00:00\u001b[0m \u001b[37m3.20it/s\u001b[0m \u001b[37mv_num: hfvp \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m178.472 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m189.712 \u001b[0m\n", - " \u001b[37mval/samples_per_s…\u001b[0m\n", - " \u001b[37m189.159 \u001b[0m\n", - " \u001b[37mtrain/samples_per…\u001b[0m\n", - " \u001b[37m235.514 \u001b[0m\n", - "\u001b[?25h/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=example datamodule=imagene ...\n", - "\u001b[2;36m[11:35:53]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=306215;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=838732;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=346133;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=387093;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", - "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓ \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.16it/s\u001b[0m \u001b[37m3.16it/s\u001b[0m \n", - "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", - "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", - "│\u001b[36m \u001b[0m\u001b[36m val/accuracy \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.0 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 7.498830318450928 \u001b[0m\u001b[35m \u001b[0m│\n", - "│\u001b[36m \u001b[0m\u001b[36mval/samples_per_second_epoch\u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 203.5886688232422 \u001b[0m\u001b[35m \u001b[0m│\n", - "└──────────────────────────────┴──────────────────────────────┘\n", - "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m30/30\u001b[0m \u001b[37m0:00:09 • 0:00:00\u001b[0m \u001b[37m3.16it/s\u001b[0m \n", - "\u001b[?25hval accuracy: \u001b[1;36m0.0\u001b[0m%\n", - "val val/accuracy: \u001b[1;36m0.0\u001b[0m\n", - "val val/samples_per_second_epoch: \u001b[1;36m203.5886688232422\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: W&B sync reduced upload amount by 18.3%\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run history:\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: epoch ▁▁█\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: lr-Adam ▁\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch ▁\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step ▁▁▂▂▂▃▃▃▄▄▅▅▅▆▆▆▇▇███▁▂▂▂▃▃▄▄▄▅▅▅▆▆▆▇▇██\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/accuracy ▁▁\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss ▁▁\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_epoch ▁█\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_step ▁▆█▆▇▆▅▄▅▃▄▅▄▃▇▅▆▆▆▆▇▆▇▇▇▇▅▄▅▆▇▆▄▇▇▆▇▆▇▆\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run summary:\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: epoch 1\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: lr-Adam 0.0003\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch 235.5143\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step 30\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/accuracy 0.0\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss 7.49883\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_epoch 203.58867\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/samples_per_second_step 200.58806\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run \u001b[33mDataloading + Training\u001b[0m at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/5szkhfvp\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Synced 5 W&B file(s), 0 media file(s), 2 artifact file(s) and 0 other file(s)\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Find logs at: \u001b[35m\u001b[1m./wandb/run-20240912_113516-5szkhfvp/logs\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m wandb version 0.18.0 is available! To upgrade, please run:\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m $ pip install wandb --upgrade\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m The new W&B backend becomes opt-out in version 0.18.0; try it out with `wandb.require(\"core\")`! See https://wandb.me/wandb-core for more information.\n" - ] - } - ], + "outputs": [], "source": [ + "%%capture\n", "!python project/main.py \\\n", - " algorithm=example \\\n", - " datamodule=imagenet \\\n", - " trainer=profiling \\\n", - " trainer/logger=wandb \\\n", - " trainer.logger.wandb.name=\"Dataloading + Training\"" + " experiment=profiling_cpu \\\n", + " datamodule.num_workers = 4 \\\n", + " trainer.logger.wandb.name=\"Dataloading only\" " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling_cpu \\\n", + " datamodule.num_workers = 8 \\\n", + " trainer.logger.wandb.name=\"Dataloading only\" " + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling_cpu \\\n", + " datamodule.num_workers = 16 \\\n", + " trainer.logger.wandb.name=\"Dataloading only\" " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling_cpu \\\n", + " datamodule.num_workers = 32 \\\n", + " trainer.logger.wandb.name=\"Dataloading only\" " ] }, { @@ -2033,7 +483,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Comparing throughput: GPU vs CPU model training" + "## Training models with GPUs" ] }, { @@ -2045,90 +495,28 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2;36m[09/12/24 12:18:52]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=941276;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=775917;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m12\u001b[0m/\u001b[1;36m12\u001b[0m-\u001b[1;36m18\u001b[0m-\u001b[1;36m52\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=38093;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=703144;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[33malgorithm\u001b[0m=\u001b[35mexample\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mimagenet\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mcpu\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mtrainer\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer/\u001b[33mlogger\u001b[0m=\u001b[35mwandb\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35mCPU\u001b[0m \u001b[2m \u001b[0m\n" - ] - } - ], + "outputs": [], "source": [ - "#Compare speed using CPU only vs the slowest GPU available, for a low number of steps\n", + "%%capture\n", "!python project/main.py \\\n", + " experiment=profiling_cpu \\\n", " algorithm=example \\\n", - " datamodule=imagenet \\\n", - " resources=cpu \\\n", - " trainer=profiling \\\n", - " trainer/logger=wandb \\\n", - " trainer.logger.wandb.name=\"CPU\" \\\n", - " trainer.logger.wandb.group=\"GPU vs CPU\"" + " trainer.logger.wandb.name=\"Dataloading only\"" ] }, { "cell_type": "code", - "execution_count": 15, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Traceback (most recent call last):\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\", line 177, in \n", - " main()\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/main.py\", line 94, in decorated_main\n", - " _run_hydra(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 394, in _run_hydra\n", - " _run_app(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 457, in _run_app\n", - " run_and_report(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 223, in run_and_report\n", - " raise ex\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 220, in run_and_report\n", - " return func()\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 458, in \n", - " lambda: hydra.run(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/hydra.py\", line 105, in run\n", - " cfg = self.compose_config(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/hydra.py\", line 594, in compose_config\n", - " cfg = self.config_loader.load_configuration(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/config_loader_impl.py\", line 142, in load_configuration\n", - " return self._load_configuration_impl(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/config_loader_impl.py\", line 244, in _load_configuration_impl\n", - " parsed_overrides, caching_repo = self._parse_overrides_and_create_caching_repo(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/config_loader_impl.py\", line 228, in _parse_overrides_and_create_caching_repo\n", - " parsed_overrides = parser.parse_overrides(overrides=overrides)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/core/override_parser/overrides_parser.py\", line 96, in parse_overrides\n", - " raise OverrideParseException(\n", - "hydra.errors.OverrideParseException: mismatched input 'logger.wandb.name' expecting ID\n", - "See https://hydra.cc/docs/1.2/advanced/override_grammar/basic for details\n" - ] - } - ], + "outputs": [], "source": [ - "!HYDRA_FULL_ERROR=1 python project/main.py \\\n", - " algorithm=example \\\n", - " datamodule=imagenet \\\n", - " trainer=profiling \\\n", - " resources=one_gpu \\\n", - " trainer/logger=wandb \\\n", - " trainer.logger.wandb.name=\"GPU\" \\\n", - " trainer.logger.wandb.group=\"GPU vs CPU\"" + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling_gpu \\\n", + " hydra.launcher.gres=gpu:rtx8000:1 \\\n", + " trainer.logger.wandb.name=\"Dataloading + Training\"" ] }, { @@ -2150,14 +538,14 @@ "text": [ "GPU Avail / Total \n", "===============================\n", - "2g.20gb 3 / 48 \n", - "3g.40gb 1 / 48 \n", + "2g.20gb 6 / 48 \n", + "3g.40gb 3 / 48 \n", "4g.40gb 1 / 24 \n", "a100 0 / 32 \n", "a100l 0 / 88 \n", "a6000 0 / 8 \n", - "rtx8000 13 / 360 \n", - "v100 3 / 56 \n" + "rtx8000 14 / 408 \n", + "v100 0 / 56 \n" ] } ], @@ -2178,30 +566,53 @@ "We will now proceed to specify different GPUs over training runs and compare their throughput." ] }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2;36m[09/16/24 13:33:03]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=246287;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=666892;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m13\u001b[0m-\u001b[1;36m33\u001b[0m-\u001b[1;36m03\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=835344;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=966187;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[33malgorithm\u001b[0m=\u001b[35mexample\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mimagenet\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mtrainer\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mone_gpu\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer/\u001b[33mlogger\u001b[0m=\u001b[35mwandb\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mGPU\u001b[0m\\ -\\ A100 \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mgroup\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m =\u001b[35mGPU\u001b[0m\\ types \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mtags\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n" + ] + } + ], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling_gpu \\\n", + " hydra.launcher.gres=gpu:a100:1 \\\n", + " trainer.logger.wandb.name=\"A100\"" + ] + }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ - "## What performance do you get with each type of GPU? \n", - "# (Based on the VRAM requirements of the job (step 1), \n", - "# try all the GPU types on the Cluster that can accommodate this kind of job)\n", - "\n", - "# Add an example of a sweep over some parameters, \n", - "# with the training throughput as the metric, \n", - "# :: callbacks/samples_per_second, \n", - "# or add a devicestatsmonitor in\n", - "# and using different kinds of GPUs. \n", - "\n", - "## salloc --gres=gpu:a100:1 -c 6 --mem=32G -t 48:00:00 --partition=unkillable\n", - "\n", - "!HYDRA_FULL_ERROR=1 python project/main.py \\\n", - " algorithm=example \\\n", - " datamodule=imagenet \\\n", - " trainer=profiling \\\n", - " resources=one_gpu \\\n", - " trainer/logger=wandb" + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling_gpu \\\n", + " hydra.launcher.gres=gpu:v100:1 \\\n", + " trainer.logger.wandb.name=\"V100\"" ] }, { @@ -2222,47 +633,213 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "How well are we using the GPU?\n", - "Once we've selected the target GPU that we want to use, measure the GPU utilization. Is the GPU utilization high? (>80%?)\n", - "If it's high (>80%), then we can either stop here, or we can keep going a bit further\n", - "If it's low, then we can use the PyTorch profiler (or any other tool) to try to figure out what the bottleneck i\n", - "## maybe look at submitit's array_parallelism" + "How well are we using the GPU? Once we've done a few preliminary runs with candidate GPUs that we'd want to use, the GPU utilization can be measured and optimized. We generally aim for high GPU utilization. Is the GPU utilization high? (>80%?) \n", + "If it's low (<80%), then we can use the PyTorch profiler (or similar tools) to try to figure out where the bottleneck lies, and further tune our parameters to increase our utilization." ] }, { - "cell_type": "markdown", + "cell_type": "code", + "execution_count": 11, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2;36m[09/16/24 11:50:49]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=78076;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=898758;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m11\u001b[0m-\u001b[1;36m50\u001b[0m-\u001b[1;36m48\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=612036;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=652546;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[33malgorithm\u001b[0m=\u001b[35mexample\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mimagenet\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m datamodule.\u001b[33mbatch_size\u001b[0m=\u001b[1;36m1\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mtrainer\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mone_gpu\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer/\u001b[33mlogger\u001b[0m=\u001b[35mwandb\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mRTX8000\u001b[0m\\ -\\ batch\\ size\\ \u001b[1;36m1\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mgroup\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m =\u001b[35mBatch\u001b[0m\\ size\\ tests \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mtags\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n" + ] + } + ], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling_gpu \\\n", + " hydra.launcher.gres=gpu:rtx8000:1 \\\n", + " datamodule.batch_size=1" + ] + }, + { + "cell_type": "code", + "execution_count": 12, "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2;36m[09/16/24 12:06:22]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=731053;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=366557;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m12\u001b[0m-\u001b[1;36m06\u001b[0m-\u001b[1;36m21\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=462864;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=920530;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[33malgorithm\u001b[0m=\u001b[35mexample\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mimagenet\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m datamodule.\u001b[33mbatch_size\u001b[0m=\u001b[1;36m8\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mtrainer\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mone_gpu\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer/\u001b[33mlogger\u001b[0m=\u001b[35mwandb\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mRTX8000\u001b[0m\\ -\\ batch\\ size\\ \u001b[1;36m8\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mgroup\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m =\u001b[35mBatch\u001b[0m\\ size\\ tests \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mtags\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n" + ] + } + ], "source": [ - "We would like to maximize our throughput given GPU choice" + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling_gpu \\\n", + " hydra.launcher.gres=gpu:rtx8000:1 \\\n", + " datamodule.batch_size=8" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 13, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2;36m[09/16/24 12:20:27]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=271112;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=705864;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m12\u001b[0m-\u001b[1;36m20\u001b[0m-\u001b[1;36m26\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=897324;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=330527;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[33malgorithm\u001b[0m=\u001b[35mexample\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mimagenet\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m datamodule.\u001b[33mbatch_size\u001b[0m=\u001b[1;36m32\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mtrainer\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mone_gpu\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer/\u001b[33mlogger\u001b[0m=\u001b[35mwandb\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mRTX8000\u001b[0m\\ -\\ batch\\ size\\ \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m32\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mgroup\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m =\u001b[35mBatch\u001b[0m\\ size\\ tests \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mtags\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n" + ] + } + ], "source": [ - "### Measure the performance on different GPUS using the optimal datamodule \n", - "### params from before (and keeping other parameters the same)" + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling_gpu \\\n", + " hydra.launcher.gres=gpu:rtx8000:1 \\\n", + " datamodule.batch_size=32" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2;36m[09/16/24 12:35:51]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=831424;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=96207;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m12\u001b[0m-\u001b[1;36m35\u001b[0m-\u001b[1;36m50\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=663693;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=91463;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[33malgorithm\u001b[0m=\u001b[35mexample\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mimagenet\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m datamodule.\u001b[33mbatch_size\u001b[0m=\u001b[1;36m128\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mtrainer\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mone_gpu\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer/\u001b[33mlogger\u001b[0m=\u001b[35mwandb\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mRTX8000\u001b[0m\\ -\\ batch\\ size\\ \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m128\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mgroup\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m =\u001b[35mBatch\u001b[0m\\ size\\ tests \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mtags\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n" + ] + } + ], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling_gpu \\\n", + " hydra.launcher.gres=gpu:rtx8000:1 \\\n", + " datamodule.batch_size=128" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Additional optimization" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "We will now sweep over model hyper-parameters to maximize the utilization of our selected GPU." + "Once GPU selection and a reasonable batch size are chosen, more can be done to speed up a model's computation.\n", + "- a\n", + "- la" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 16, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2;36m[09/16/24 13:05:32]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=865887;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=436119;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m13\u001b[0m-\u001b[1;36m05\u001b[0m-\u001b[1;36m32\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=50432;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=381740;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[33malgorithm\u001b[0m=\u001b[35mexample\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mimagenet\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m datamodule.\u001b[33mbatch_size\u001b[0m=\u001b[1;36m32\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m datamodule.\u001b[33mpin_memory\u001b[0m=\u001b[3;91mFals\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[3;91me\u001b[0m \u001b[33mtrainer\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mone_gpu\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer/\u001b[33mlogger\u001b[0m=\u001b[35mwandb\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mGPU\u001b[0m\\ -\\ RTX8000\\ -\\ \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m pinned\\ memory \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mgroup\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m =\u001b[35mPin\u001b[0m\\ memory \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mtags\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n" + ] + } + ], "source": [ - "#### Using the results from before, do a simple sweep over model hyper-parameters \n", - "#### to maximize the utilization of the selected GPU (which was selected as a tradeoff \n", - "#### between performance and difficulty to get an allocation). For example if the \n", - "#### RTX8000's are 20% slower than A100s but 5x easier to get an allocation on, use those instead." + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling_gpu \\\n", + " datamodule.batch_size=32\n", + " datamodule.num_workers=4 ## optimal parameter from above tests, check" ] }, { @@ -2276,7 +853,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "The former process was a bit contrived - We can zero down specifically on subprocesses... \n", + "The former process, while straightforward, was a bit contrived - would having a bird's eye view of our models performance be of aid when trying to optimize its parameters? It certainly wouldn't hurt. Enter the profiler. \n", "A profiler is a tool that allows you to measure the time and memory consumption of the model’s operators. Specifically, the PyTorch profiler output provides clues about operations relevant to model training. Examples include the total amount of time spent doing low-level mathematical operations in the GPU, and whether these are unexpectedly slow or take a disproportionate amount of time, indicating they should be avoided or optimized. Identifying problematic operations can greatly help us validate or rethink our baseline model performance expectations.\n", "\n", "[Multiple](https://developer.nvidia.com/blog/profiling-and-optimizing-deep-neural-networks-with-dlprof-and-pyprof/) [profilers](https://github.com/plasma-umass/scalene) [exist](https://docs.python.org/3/library/profile.html). For the purposes of this example we'll use the default [PyTorch Profiler](https://pytorch.org/tutorials/recipes/recipes/profiler_recipe.html). " @@ -2315,16 +892,6 @@ "print(profiler.key_averages().table(sort_by=\"cpu_time_total\", row_limit=10))\n" ] }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [] - }, { "cell_type": "markdown", "metadata": {}, diff --git a/project/configs/experiment/profiling_cpu.yaml b/project/configs/experiment/profiling_cpu.yaml new file mode 100644 index 00000000..4c998d77 --- /dev/null +++ b/project/configs/experiment/profiling_cpu.yaml @@ -0,0 +1,15 @@ +# @package _global_ +# yaml-language-server: $schema=../../../.schemas/experiment_profiling_cpu_schema.json + +defaults: + - override /datamodule: imagenet + - override /algorithm: no_op + - override /trainer/logger: wandb + - override /resources: cpu + +trainer: + min_epochs: 1 + max_epochs: 2 + limit_train_batches: 2 + limit_val_batches: 2 + num_sanity_val_steps: 0 diff --git a/project/configs/experiment/profiling_gpu.yaml b/project/configs/experiment/profiling_gpu.yaml new file mode 100644 index 00000000..b31a8502 --- /dev/null +++ b/project/configs/experiment/profiling_gpu.yaml @@ -0,0 +1,19 @@ +# @package _global_ +# yaml-language-server: $schema=../../../.schemas/experiment_profiling_rtx8000_schema.json + +defaults: + - override /datamodule: imagenet + - override /algorithm: example + - override /trainer/logger: wandb + - override /resources: one_gpu + +hydra: + launcher: + partition: unkillable + +trainer: + min_epochs: 1 + max_epochs: 2 + limit_train_batches: 2 + limit_val_batches: 2 + num_sanity_val_steps: 0 diff --git a/project/configs/trainer/profiling.yaml b/project/configs/trainer/profiling.yaml deleted file mode 100644 index eed1ff3e..00000000 --- a/project/configs/trainer/profiling.yaml +++ /dev/null @@ -1,8 +0,0 @@ -# yaml-language-server: $schema=../../../.schemas/trainer_profiling_schema.json -_target_: lightning.Trainer -max_epochs: 1 -limit_train_batches: 30 -limit_val_batches: 30 - -wandb: - tags: ["profiling"] diff --git a/project/datamodules/vision.py b/project/datamodules/vision.py index 50d8dd12..df5fdee8 100644 --- a/project/datamodules/vision.py +++ b/project/datamodules/vision.py @@ -115,7 +115,7 @@ def __init__( self.test_kwargs["train"] = False self.batch_size_per_device: int = batch_size - self.save_hyperparameters(logger=False) + self.save_hyperparameters() def prepare_data(self) -> None: """Saves files to data_dir.""" diff --git a/project/main.py b/project/main.py index 6388c0af..4ba195a0 100644 --- a/project/main.py +++ b/project/main.py @@ -2,14 +2,12 @@ from __future__ import annotations -import dataclasses import os import warnings from logging import getLogger as get_logger from pathlib import Path import hydra -import omegaconf import rich from lightning import LightningDataModule from omegaconf import DictConfig @@ -53,23 +51,21 @@ def main(dict_config: DictConfig) -> dict: quiet=True, add_headers=False, ) - config: Config = resolve_dictconfig(dict_config) - + # assert False, OmegaConf.resolve(dict_config)["datamodule"] experiment: Experiment = setup_experiment(config) - if wandb.run: - wandb.config.update({k: v for k, v in os.environ.items() if k.startswith("SLURM")}) - wandb.config.update( - omegaconf.OmegaConf.to_container(dict_config, resolve=False, throw_on_missing=True) - ) - wandb.config.update( - dataclasses.asdict(config), - allow_val_change=True, - ) - metric_name, objective, _metrics = run(experiment) assert objective is not None + + if wandb.run: + wandb.run.config.update({k: v for k, v in os.environ.items() if k.startswith("SLURM")}) + # wandb.run.config.update( + # omegaconf.OmegaConf.to_container(dict_config, resolve=False, throw_on_missing=True) + # ) + # wandb.run.config.update(dataclasses.asdict(config), allow_val_change=True) + wandb.run.finish() + return dict(name=metric_name, type="objective", value=objective) # return {metric_name: objective} @@ -88,8 +84,7 @@ def run(experiment: Experiment) -> tuple[str, float | None, dict]: ) metric_name, error, metrics = evaluation(experiment) - if wandb.run: - wandb.finish() + return metric_name, error, metrics From 4dc7bbc7f20279dc692b9c5922d1fc4e9a2e540e Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Mon, 16 Sep 2024 17:10:05 -0400 Subject: [PATCH 14/33] Additional nb cleanup --- docs/examples/profiling.ipynb | 65 +++++++++++-------- project/configs/experiment/profiling_gpu.yaml | 2 +- 2 files changed, 39 insertions(+), 28 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index ec180172..5aa04ec7 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -417,20 +417,34 @@ "%%capture\n", "!python project/main.py \\\n", " experiment=profiling_cpu \\\n", - " trainer.logger.wandb.name=\"Dataloading only\"" + " trainer.logger.wandb.name=\"Dataloading only\" \\\n", + " datamodule.num_workers=1" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 24, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2;36m[09/16/24 17:02:15]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=318334;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=306596;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m17\u001b[0m-\u001b[1;36m02\u001b[0m-\u001b[1;36m14\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=259532;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=63312;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[33mexperiment\u001b[0m=\u001b[35mprofiling_cpu\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m datamodule.\u001b[33mnum_workers\u001b[0m=\u001b[1;36m4\u001b[0m \u001b[2m \u001b[0m\n" + ] + } + ], "source": [ "%%capture\n", "!python project/main.py \\\n", " experiment=profiling_cpu \\\n", - " datamodule.num_workers = 4 \\\n", - " trainer.logger.wandb.name=\"Dataloading only\" " + " datamodule.num_workers=4" ] }, { @@ -442,8 +456,7 @@ "%%capture\n", "!python project/main.py \\\n", " experiment=profiling_cpu \\\n", - " datamodule.num_workers = 8 \\\n", - " trainer.logger.wandb.name=\"Dataloading only\" " + " datamodule.num_workers=8" ] }, { @@ -455,8 +468,7 @@ "%%capture\n", "!python project/main.py \\\n", " experiment=profiling_cpu \\\n", - " datamodule.num_workers = 16 \\\n", - " trainer.logger.wandb.name=\"Dataloading only\" " + " datamodule.num_workers=16" ] }, { @@ -468,15 +480,7 @@ "%%capture\n", "!python project/main.py \\\n", " experiment=profiling_cpu \\\n", - " datamodule.num_workers = 32 \\\n", - " trainer.logger.wandb.name=\"Dataloading only\" " - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "As evidenced in the former, adding training to our run results in a difference in the ballpark of 100 samples/s. This would indicate that we have a computation bottleneck." + " datamodule.num_workers=32" ] }, { @@ -503,20 +507,14 @@ "!python project/main.py \\\n", " experiment=profiling_cpu \\\n", " algorithm=example \\\n", - " trainer.logger.wandb.name=\"Dataloading only\"" + " trainer.logger.wandb.name=\"Dataloading + Training\"" ] }, { - "cell_type": "code", - "execution_count": null, + "cell_type": "markdown", "metadata": {}, - "outputs": [], "source": [ - "%%capture\n", - "!python project/main.py \\\n", - " experiment=profiling_gpu \\\n", - " hydra.launcher.gres=gpu:rtx8000:1 \\\n", - " trainer.logger.wandb.name=\"Dataloading + Training\"" + "CHECK FOR ACCURACY ON DASHBOARD - As evidenced in the former, adding training to our run results in a difference in the ballpark of 100 samples/s. This would indicate that we have a computation bottleneck." ] }, { @@ -566,6 +564,19 @@ "We will now proceed to specify different GPUs over training runs and compare their throughput." ] }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling_gpu \\\n", + " hydra.launcher.gres=gpu:rtx8000:1 \\\n", + " trainer.logger.wandb.name=\"RTX8000\"" + ] + }, { "cell_type": "code", "execution_count": 18, diff --git a/project/configs/experiment/profiling_gpu.yaml b/project/configs/experiment/profiling_gpu.yaml index b31a8502..11ac2071 100644 --- a/project/configs/experiment/profiling_gpu.yaml +++ b/project/configs/experiment/profiling_gpu.yaml @@ -1,5 +1,5 @@ # @package _global_ -# yaml-language-server: $schema=../../../.schemas/experiment_profiling_rtx8000_schema.json +# yaml-language-server: $schema=../../../.schemas/experiment_profiling_gpu_schema.json defaults: - override /datamodule: imagenet From 064c11748c9b1e4959c5d112e6f11a85a4ad04a0 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Wed, 18 Sep 2024 11:29:37 -0400 Subject: [PATCH 15/33] latest nb changes --- docs/examples/profiling.ipynb | 45 +++++++---------------------------- 1 file changed, 8 insertions(+), 37 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 5aa04ec7..a940250a 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -395,24 +395,9 @@ }, { "cell_type": "code", - "execution_count": 21, + "execution_count": 2, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2;36m[09/16/24 16:44:50]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=44774;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=174608;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m16\u001b[0m-\u001b[1;36m44\u001b[0m-\u001b[1;36m50\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=55055;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=351128;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[33mexperiment\u001b[0m=\u001b[35mprofiling_cpu\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35mDataloading\u001b[0m\\ only \u001b[2m \u001b[0m\n" - ] - } - ], + "outputs": [], "source": [ "%%capture\n", "!python project/main.py \\\n", @@ -423,23 +408,9 @@ }, { "cell_type": "code", - "execution_count": 24, + "execution_count": 3, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2;36m[09/16/24 17:02:15]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=318334;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=306596;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m17\u001b[0m-\u001b[1;36m02\u001b[0m-\u001b[1;36m14\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=259532;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=63312;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[33mexperiment\u001b[0m=\u001b[35mprofiling_cpu\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m datamodule.\u001b[33mnum_workers\u001b[0m=\u001b[1;36m4\u001b[0m \u001b[2m \u001b[0m\n" - ] - } - ], + "outputs": [], "source": [ "%%capture\n", "!python project/main.py \\\n", @@ -449,7 +420,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 4, "metadata": {}, "outputs": [], "source": [ @@ -461,7 +432,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 5, "metadata": {}, "outputs": [], "source": [ @@ -473,7 +444,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 6, "metadata": {}, "outputs": [], "source": [ @@ -507,7 +478,7 @@ "!python project/main.py \\\n", " experiment=profiling_cpu \\\n", " algorithm=example \\\n", - " trainer.logger.wandb.name=\"Dataloading + Training\"" + " trainer.logger.wandb.name=\"CPU Dataloading + Training\"" ] }, { From 0370aa9da3f2a0dd02d1c2f965325958d4e6074d Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Thu, 19 Sep 2024 13:14:45 -0400 Subject: [PATCH 16/33] post feedback nb restructure --- docs/examples/profiling.ipynb | 647 ++++-------------- project/configs/experiment/profiling_cpu.yaml | 15 - project/configs/experiment/profiling_gpu.yaml | 19 - 3 files changed, 136 insertions(+), 545 deletions(-) delete mode 100644 project/configs/experiment/profiling_cpu.yaml delete mode 100644 project/configs/experiment/profiling_gpu.yaml diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index a940250a..4f577424 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -71,307 +71,148 @@ "- CPU/GPU utilization \n", "- RAM/VRAM utilization\n", "\n", - "In the Mila ResearchTemplate, this can be done by passing a callback to the trainer. Supported configs are found within the project template at `configs/trainer/callbacks`. Here, we will use the default callback, which in turn implements early stopping and tracks the learning rate, device utilisation and throughput, each through a specific callback instance." + "In the Mila ResearchTemplate, this can be done by passing a callback to the trainer. Supported configs are found within the project template at `configs/trainer/callbacks`. Throughout this tutorial, we will use the default callback, which in turn implements early stopping and tracks the learning rate, device utilisation and throughput, each through a specific callback instance." ] }, { - "cell_type": "code", - "execution_count": 15, + "cell_type": "markdown", "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/16/24 16:09:10]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=800276;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=394219;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m .yaml last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=497869;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=538994;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m not find \u001b[32m'hydra/sweeper/orion'\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Available options in \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'hydra/sweeper'\u001b[0m: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m basic \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Config search path: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra.conf\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mmain\u001b[0m, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mproject.configs\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m-colorlog, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra_plugins.hydra_c\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/46\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m46/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m7,5…\u001b[0m ]\n", - " \u001b[31mit/s\u001b[0m \n", - "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/16/24 16:09:11]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=35654;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=593628;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=907199;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=326931;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 30897\n", - "Seed set to 30897\n", - "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py algorithm=no_op datamodule=imagenet ...\n", - "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", - "GPU available: False, used: False\n", - "TPU available: False, using: 0 TPU cores\n", - "HPU available: False, using: 0 HPUs\n", - "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/pytorch/trainer/connectors/logger_connector/logger_connector.py:75: Starting from v1.9.0, `tensorboardX` has been removed as a dependency of the `lightning.pytorch` package, due to potential conflicts with other packages in the ML ecosystem. For this reason, `logger=True` will use `CSVLogger` as the default logger, unless the `tensorboard` or `tensorboardX` packages are found. Please `pip install lightning[extra]` or one of them to enable TensorBoard support by default\n", - "\u001b[2;36m[16:09:11]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=422359;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=276611;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7fc41bf0de40\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", - "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "\u001b[2;36m[16:09:12]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=765323;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=527622;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=300968;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=508269;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "^C\n", - "\n", - "Detected KeyboardInterrupt, attempting graceful shutdown ...\n" - ] - } - ], "source": [ - "#%%capture\n", - "!python project/main.py \\\n", - " algorithm=no_op \\\n", - " datamodule=imagenet \\\n", - " trainer=profiling \\\n", - " trainer/callbacks=default" + "### Logging metrics on WandB" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "### Logging metrics on WandB" + "In addition to callback specification, the Mila Research template integrates wandb as a logger specification, which enables the tracking of additional metrics through visualizations and dashboard creation. Given the flexibility and widespread adoption of using WandB as a logger, we'll be using it for the remainder of this tutorial, which will then be visualizable at `wandb_url` , as supporting information for the experiments contained herein." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "In addition to callback specification, the Mila Research template integrates wandb as a logger specification, which enables the tracking of additional metrics through visualizations and dashboard creation. Given the flexibility and widespread adoption of using WandB as a logger, we'll be using it for the remainder of this tutorial." + "## Training models with GPUs" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Advancements in Graphical Processing Units (GPUs) are widely known to have enabled the deep learning revolution, particularly through faster computation, relative to CPUs. Given that we have the option to run both GPU and CPU workloads, let's compare their throughput. In most workflows, the speedup provided by a GPU is dramatic. For a few select workloads, particularly those with a low number of steps or lighter computation requirements, if a 1.5-2x slower performance is observed when using a CPU, as opposed to a GPU, the former may be worth considering, as they're a far less contested resource on the cluster and pose far fewer availability issues." ] }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/16/24 16:31:23]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Unable to properly create the \u001b]8;id=965865;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=902936;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#368\u001b\\\u001b[2m368\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m schema for \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m .yaml last time. Trying again. \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Unable to create a schema for \u001b]8;id=381213;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=757802;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#396\u001b\\\u001b[2m396\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m config \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m experiment/cluster_sweep_example \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m .yaml: In \u001b[32m'hydra/config'\u001b[0m: Could \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m not find \u001b[32m'hydra/sweeper/orion'\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Available options in \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[32m'hydra/sweeper'\u001b[0m: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m basic \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Config search path: \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra.conf\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mmain\u001b[0m, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mproject.configs\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mhydra\u001b[0m-colorlog, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mpkg\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m\u001b[95mhydra_plugins.hydra_c\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[95molorlog.conf\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mprovider\u001b[0m=\u001b[35mschema\u001b[0m, \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mpath\u001b[0m=\u001b[35mstructured\u001b[0m:\u001b[35m/\u001b[0m\u001b[35m/\u001b[0m \u001b[2m \u001b[0m\n", - "Creating schemas for Hydra config files...\u001b[35m 0%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m0/45\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m-:-…\u001b[0m , \u001b[31m? \u001b[0m ]\n", - "\u001b[2K\u001b[1A\u001b[2KCreating schemas for Hydra config files...\u001b[35m 100%\u001b[0m \u001b[90m━━━━\u001b[0m \u001b[32m45/…\u001b[0m [ \u001b[33m0:0…\u001b[0m < \u001b[36m0:0…\u001b[0m , \u001b[31m8,5…\u001b[0m ]\n", - " \u001b[31mit/s\u001b[0m \n", - "\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[2;36m[09/16/24 16:31:24]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Updated the yaml schemas in the \u001b]8;id=106742;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py\u001b\\\u001b[2mauto_schema.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=184714;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/auto_schema.py#522\u001b\\\u001b[2m522\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m vscode settings file at \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35m/home/mila/c/cesar.valdez/idt/Re\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35msearchTemplate/.vscode/\u001b[0m\u001b[95msettings.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[95mjson.\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Instantiated the config at \u001b]8;id=771749;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py\u001b\\\u001b[2mhydra_utils.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=511000;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/utils/hydra_utils.py#370\u001b\\\u001b[2m370\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[32m'datamodule'\u001b[0m while trying to \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m find one of the \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0m\u001b[32m'datamodule.num_classes'\u001b[0m, \u001b[1;36m1000\u001b[0m\u001b[1m)\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m attributes. \u001b[2m \u001b[0m\n", - "seed manually set to 77069\n", - "Seed set to 77069\n", - "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py experiment=profiling_cpu trainer/log ...\n", - "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n", - "GPU available: False, used: False\n", - "TPU available: False, using: 0 TPU cores\n", - "HPU available: False, using: 0 HPUs\n", - "`Trainer(limit_val_batches=1)` was configured so 1 batch will be used.\n", - "\u001b[2;36m[16:31:25]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Datamodule was already instantiated \u001b]8;id=903950;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py\u001b\\\u001b[2mexperiment.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=805041;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/experiment.py#175\u001b\\\u001b[2m175\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[1m(\u001b[0mprobably to interpolate a field value\u001b[1m)\u001b[0m. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule_config\u001b[0m=\u001b[1m<\u001b[0m\u001b[1;95mproject.datamodules.ima\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mge_classification.imagenet.ImageNetDataMod\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;95mule\u001b[0m\u001b[39m object at \u001b[0m\u001b[1;36m0x7f40d1f33ac0\u001b[0m\u001b[1m>\u001b[0m \u001b[2m \u001b[0m\n", - "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=618681;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=810172;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=485875;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=801119;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mcesar-valdez\u001b[0m (\u001b[33mcesar-valdez-mcgill-university\u001b[0m). Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Tracking run with wandb version 0.17.8\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run data is saved locally in \u001b[35m\u001b[1m./wandb/run-20240916_163130-bd8pa1aw\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run \u001b[1m`wandb offline`\u001b[0m to turn off syncing.\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Syncing run \u001b[33mWandB logging test\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run at \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/bd8pa1aw\u001b[0m\n", - "┏━━━━┳━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┓\n", - "┃\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mName \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mType \u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mParams\u001b[0m\u001b[1;35m \u001b[0m┃\u001b[1;35m \u001b[0m\u001b[1;35mMode \u001b[0m\u001b[1;35m \u001b[0m┃\n", - "┡━━━━╇━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━┩\n", - "│\u001b[2m \u001b[0m\u001b[2m0 \u001b[0m\u001b[2m \u001b[0m│ network │ ResNet │ 11.7 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m1 \u001b[0m\u001b[2m \u001b[0m│ network.conv1 │ Conv2d │ 9.4 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m2 \u001b[0m\u001b[2m \u001b[0m│ network.bn1 │ BatchNorm2d │ 128 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m3 \u001b[0m\u001b[2m \u001b[0m│ network.relu │ ReLU │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m4 \u001b[0m\u001b[2m \u001b[0m│ network.maxpool │ MaxPool2d │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m5 \u001b[0m\u001b[2m \u001b[0m│ network.layer1 │ Sequential │ 147 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m6 \u001b[0m\u001b[2m \u001b[0m│ network.layer2 │ Sequential │ 525 K │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m7 \u001b[0m\u001b[2m \u001b[0m│ network.layer3 │ Sequential │ 2.1 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m8 \u001b[0m\u001b[2m \u001b[0m│ network.layer4 │ Sequential │ 8.4 M │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m9 \u001b[0m\u001b[2m \u001b[0m│ network.avgpool │ AdaptiveAvgPool2d │ 0 │ train │\n", - "│\u001b[2m \u001b[0m\u001b[2m10\u001b[0m\u001b[2m \u001b[0m│ network.fc │ Linear │ 513 K │ train │\n", - "└────┴─────────────────┴───────────────────┴────────┴───────┘\n", - "\u001b[1mTrainable params\u001b[0m: 11.7 M \n", - "\u001b[1mNon-trainable params\u001b[0m: 0 \n", - "\u001b[1mTotal params\u001b[0m: 11.7 M \n", - "\u001b[1mTotal estimated model params size (MB)\u001b[0m: 46 \n", - "\u001b[1mModules in train mode\u001b[0m: 68 \n", - "\u001b[1mModules in eval mode\u001b[0m: 0 \n", - "/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/pytorch/loops/fit_loop.py:298: The number of training batches (2) is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower value for log_every_n_steps if you want to see logs for the training epoch.\n", - "\u001b[2K\u001b[37mEpoch 0/1 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m4.45it/s\u001b[0m \u001b[37mv_num: a1aw\u001b[0m37mv_num: a1aw\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[37mEpoch 0/1 \u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m4.45it/s\u001b[0m \u001b[37mv_num: a1aw\u001b[0m\n", - "\u001b[2KEpoch 1/1 \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/2\u001b[0m \u001b[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m7mv_num: a1aw\u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[90m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m0/2\u001b[0m \u001b[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m1/2\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━\u001b[0m\u001b[90m╺\u001b[0m\u001b[90m━━━━━━━━\u001b[0m \u001b[37m1/2\u001b[0m \u001b[37m0:00:01 • -:--:--\u001b[0m \u001b[37m0.00it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m286.046 \u001b[0m`Trainer.fit` stopped: `max_epochs=2` reached.\n", - "\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2K\u001b[1A\u001b[2KEpoch 1/1 \u001b[35m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m2/2\u001b[0m \u001b[37m0:00:01 • 0:00:00\u001b[0m \u001b[37m2.72it/s\u001b[0m \u001b[37mv_num: a1aw \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m175.405 \u001b[0m\n", - " \u001b[37mtrain/samples_per_…\u001b[0m\n", - " \u001b[37m175.405 \u001b[0m\n", - "\u001b[?25h/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/lightning/fabric/plugins/environments/slurm.py:204: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python project/main.py experiment=profiling_cpu trainer/log ...\n", - "The following callbacks returned in `LightningModule.configure_callbacks` will override existing callbacks passed to Trainer: MeasureSamplesPerSecondCallback\n", - "\u001b[2;36m[16:31:49]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Train archive already fully extracted. \u001b]8;id=817434;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=930994;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#382\u001b\\\u001b[2m382\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2;36m[16:31:50]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Validation split already extracted. \u001b]8;id=584010;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py\u001b\\\u001b[2mimagenet.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=413818;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/datamodules/image_classification/imagenet.py#339\u001b\\\u001b[2m339\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m Skipping. \u001b[2m \u001b[0m\n", - "\u001b[2K┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓━━━━\u001b[0m \u001b[37m1/1\u001b[0m \u001b[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", - "┃\u001b[1m \u001b[0m\u001b[1m Validate metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n", - "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n", - "│\u001b[36m \u001b[0m\u001b[36m val/loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.3719692826271057 \u001b[0m\u001b[35m \u001b[0m│\n", - "└───────────────────────────┴───────────────────────────┘\n", - "\u001b[2K\u001b[37mValidation\u001b[0m \u001b[35m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[37m1/1\u001b[0m \u001b[37m0:00:00 • 0:00:00\u001b[0m \u001b[37m0.00it/s\u001b[0m \n", - "\u001b[?25h\u001b[2;36m[16:32:08]\u001b[0m\u001b[2;36m \u001b[0m\u001b[33mWARNING \u001b[0m Assuming that the objective to minimize is the \u001b]8;id=785781;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\u001b\\\u001b[2mmain.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=487120;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py#160\u001b\\\u001b[2m160\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m loss metric. \u001b[2m \u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: W&B sync reduced upload amount by 21.7%\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run history:\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: epoch ▁▁▅▅█\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: lr-SGD ▁▁\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch █▁\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step ▁▃▃▅▆▆█\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss █▁▄\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Run summary:\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: epoch 2\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: lr-SGD 0.123\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: train/samples_per_second_epoch 175.40546\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: trainer/global_step 4\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: val/loss 0.37197\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \n", - "\u001b[34m\u001b[1mwandb\u001b[0m: 🚀 View run \u001b[33mWandB logging test\u001b[0m at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/runs/bd8pa1aw\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: ⭐️ View project at: \u001b[34m\u001b[4mhttps://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Synced 5 W&B file(s), 0 media file(s), 2 artifact file(s) and 0 other file(s)\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: Find logs at: \u001b[35m\u001b[1m./wandb/run-20240916_163130-bd8pa1aw/logs\u001b[0m\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m wandb version 0.18.0 is available! To upgrade, please run:\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m $ pip install wandb --upgrade\n", - "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m The new W&B backend becomes opt-out in version 0.18.0; try it out with `wandb.require(\"core\")`! See https://wandb.me/wandb-core for more information.\n" + "\u001b[2;36m[09/19/24 13:05:55]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=143271;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=313435;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m19\u001b[0m/\u001b[1;36m13\u001b[0m-\u001b[1;36m05\u001b[0m-\u001b[1;36m54\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=857793;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=99902;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[33mexperiment\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m1\u001b[0m\\ RTX8000\\ GPU\\ \u001b[1;36m1\u001b[0m\\ CPU\\ \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Training \u001b[2m \u001b[0m\n", + "^C\n", + "Traceback (most recent call last):\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\", line 172, in \n", + " main()\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/main.py\", line 94, in decorated_main\n", + " _run_hydra(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 394, in _run_hydra\n", + " _run_app(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 465, in _run_app\n", + " run_and_report(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 220, in run_and_report\n", + " return func()\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 466, in \n", + " lambda: hydra.multirun(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/hydra.py\", line 162, in multirun\n", + " ret = sweeper.sweep(arguments=task_overrides)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/core_plugins/basic_sweeper.py\", line 177, in sweep\n", + " results = self.launcher.launch(batch, initial_job_idx=initial_job_idx)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\", line 146, in launch\n", + " return [j.results()[0] for j in jobs]\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\", line 146, in \n", + " return [j.results()[0] for j in jobs]\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/submitit/core/core.py\", line 284, in results\n", + " self.wait()\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/submitit/core/core.py\", line 400, in wait\n", + " _time.sleep(1)\n", + "KeyboardInterrupt\n" ] } ], "source": [ - "%%capture\n", + "#%%capture\n", "!python project/main.py \\\n", - " experiment=profiling_cpu \\\n", - " trainer/logger=wandb \\\n", - " trainer.logger.wandb.name=\"WandB logging test\"" + " experiment=profiling \\\n", + " hydra.launcher.gres='gpu:rtx8000:1' \\\n", + " hydra.launcher.cpus_per_task=1 \\\n", + " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training\"" ] }, { - "cell_type": "markdown", + "cell_type": "code", + "execution_count": 8, "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2;36m[09/19/24 13:07:09]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=701107;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=395625;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m19\u001b[0m/\u001b[1;36m13\u001b[0m-\u001b[1;36m07\u001b[0m-\u001b[1;36m08\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=233531;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=331458;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[33mexperiment\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mcpu\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m1\u001b[0m\\ CPU\\ Training \u001b[2m \u001b[0m\n", + "^C\n", + "Traceback (most recent call last):\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\", line 172, in \n", + " main()\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/main.py\", line 94, in decorated_main\n", + " _run_hydra(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 394, in _run_hydra\n", + " _run_app(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 465, in _run_app\n", + " run_and_report(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 220, in run_and_report\n", + " return func()\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 466, in \n", + " lambda: hydra.multirun(\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/hydra.py\", line 162, in multirun\n", + " ret = sweeper.sweep(arguments=task_overrides)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/core_plugins/basic_sweeper.py\", line 177, in sweep\n", + " results = self.launcher.launch(batch, initial_job_idx=initial_job_idx)\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\", line 146, in launch\n", + " return [j.results()[0] for j in jobs]\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\", line 146, in \n", + " return [j.results()[0] for j in jobs]\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/submitit/core/core.py\", line 284, in results\n", + " self.wait()\n", + " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/submitit/core/core.py\", line 400, in wait\n", + " _time.sleep(1)\n", + "KeyboardInterrupt\n" + ] + } + ], "source": [ - "We can now visualize the results of our run at `wandb_url`" + "#%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling \\\n", + " resources=cpu \\\n", + " trainer.logger.wandb.name=\"1 CPU Training\"" ] }, { @@ -390,7 +231,7 @@ "- If the difference between data loading and training is close to 0, then the data loading procedure outpaces model computation, and computation is the bottleneck. \n", "- If the difference between data loading and training is much greater than 0, then model computation outpaces data loading, and data loading is the bottleneck. \n", "\n", - "To showcase the former, we will proceed to run two separate model loops on imagenet: the first one doing data loading without any training, followed by one with." + "We will proceed to run a series of experiments to identify potential bottlenecks: changing the workers involved in the dataloading process and the numbers of cpu assigned per task when training on a GPU." ] }, { @@ -401,71 +242,38 @@ "source": [ "%%capture\n", "!python project/main.py \\\n", - " experiment=profiling_cpu \\\n", + " experiment=profiling \\\n", + " algorithm=no_op \\\n", " trainer.logger.wandb.name=\"Dataloading only\" \\\n", - " datamodule.num_workers=1" - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "metadata": {}, - "outputs": [], - "source": [ - "%%capture\n", - "!python project/main.py \\\n", - " experiment=profiling_cpu \\\n", - " datamodule.num_workers=4" + " datamodule.num_workers=1,4,8,16,32" ] }, { "cell_type": "code", - "execution_count": 4, - "metadata": {}, - "outputs": [], - "source": [ - "%%capture\n", - "!python project/main.py \\\n", - " experiment=profiling_cpu \\\n", - " datamodule.num_workers=8" - ] - }, - { - "cell_type": "code", - "execution_count": 5, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ "%%capture\n", - "!python project/main.py \\\n", - " experiment=profiling_cpu \\\n", - " datamodule.num_workers=16" + "!python project/main.py -m \\\n", + " experiment=profiling \\\n", + " hydra.launcher.gres='gpu:rtx8000:1' \\\n", + " hydra.launcher.cpus_per_task=2 \\\n", + " trainer.logger.wandb.name=\"1 RTX8000 GPU 2 CPU Training\"" ] }, { "cell_type": "code", - "execution_count": 6, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ "%%capture\n", - "!python project/main.py \\\n", - " experiment=profiling_cpu \\\n", - " datamodule.num_workers=32" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Training models with GPUs" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Advancements in Graphical Processing Units (GPUs) are widely known to have enabled the deep learning revolution, particularly through faster computation, relative to CPUs. Given that we have the option to run both GPU and CPU workloads, let's compare their throughput. In most workflows, the speedup provided by a GPU is dramatic. For a few select workloads, particularly those with a low number of steps or lighter computation requirements, if a 1.5-2x slower performance is observed when using a CPU, as opposed to a GPU, the former may be worth considering, as they're a far less contested resource on the cluster and pose far fewer availability issues." + "!python project/main.py -m \\\n", + " experiment=profiling \\\n", + " hydra.launcher.gres='gpu:rtx8000:1' \\\n", + " hydra.launcher.cpus_per_task=3 \\\n", + " trainer.logger.wandb.name=\"1 RTX8000 GPU 3 CPU Training\"" ] }, { @@ -475,17 +283,18 @@ "outputs": [], "source": [ "%%capture\n", - "!python project/main.py \\\n", - " experiment=profiling_cpu \\\n", - " algorithm=example \\\n", - " trainer.logger.wandb.name=\"CPU Dataloading + Training\"" + "!python project/main.py -m \\\n", + " experiment=profiling \\\n", + " hydra.launcher.gres='gpu:rtx8000:1' \\\n", + " hydra.launcher.cpus_per_task=4 \\\n", + " trainer.logger.wandb.name=\"1 RTX8000 GPU 4 CPU Training\"" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "CHECK FOR ACCURACY ON DASHBOARD - As evidenced in the former, adding training to our run results in a difference in the ballpark of 100 samples/s. This would indicate that we have a computation bottleneck." + "## Throughput across GPU types" ] }, { @@ -532,20 +341,7 @@ "- NVIDIA RTX GPUs: A6000, RTX8000\n", "- Multi-Instance GPU (MiG) partitions: 2g.20gb, 3g.40gb, 4g.40gb \n", "\n", - "We will now proceed to specify different GPUs over training runs and compare their throughput." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "%%capture\n", - "!python project/main.py \\\n", - " experiment=profiling_gpu \\\n", - " hydra.launcher.gres=gpu:rtx8000:1 \\\n", - " trainer.logger.wandb.name=\"RTX8000\"" + "We will now proceed to specify different GPUs over training runs and compare their throughput. If a GPU with lower maximum capacity is readily available, training on it may be more time and resource effective than waiting for higher capacity GPUs to become available.\n" ] }, { @@ -581,7 +377,9 @@ "!python project/main.py \\\n", " experiment=profiling_gpu \\\n", " hydra.launcher.gres=gpu:a100:1 \\\n", - " trainer.logger.wandb.name=\"A100\"" + " trainer.logger.wandb.name=\"A100\" \\\n", + " datamodule.num_workers=#optimal params as determined before\n", + " trainer.logger.wandb.name=\"A100 GPU X CPU X Num_workers\"" ] }, { @@ -594,14 +392,9 @@ "!python project/main.py \\\n", " experiment=profiling_gpu \\\n", " hydra.launcher.gres=gpu:v100:1 \\\n", - " trainer.logger.wandb.name=\"V100\"" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Making sense of the former: if a GPU with lower maximum capacity is readily available, training on it may be more time and resource effective than waiting for higher capacity GPUs to become available.\n" + " trainer.logger.wandb.name=\"V100\" \\\n", + " datamodule.num_workers=#optimal param as determined before \n", + " trainer.logger.wandb.name=\"V100 GPU X CPU X Num_workers\"" ] }, { @@ -615,7 +408,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "How well are we using the GPU? Once we've done a few preliminary runs with candidate GPUs that we'd want to use, the GPU utilization can be measured and optimized. We generally aim for high GPU utilization. Is the GPU utilization high? (>80%?) \n", + "How well are we using a given GPU? Once we've done a few preliminary runs with candidate GPUs that we'd want to use, the GPU utilization can be measured and optimized. We generally aim for high GPU utilization. Is the GPU utilization high? (>80%?) \n", "If it's low (<80%), then we can use the PyTorch profiler (or similar tools) to try to figure out where the bottleneck lies, and further tune our parameters to increase our utilization." ] }, @@ -653,175 +446,7 @@ "!python project/main.py \\\n", " experiment=profiling_gpu \\\n", " hydra.launcher.gres=gpu:rtx8000:1 \\\n", - " datamodule.batch_size=1" - ] - }, - { - "cell_type": "code", - "execution_count": 12, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2;36m[09/16/24 12:06:22]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=731053;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=366557;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m12\u001b[0m-\u001b[1;36m06\u001b[0m-\u001b[1;36m21\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=462864;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=920530;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[33malgorithm\u001b[0m=\u001b[35mexample\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mimagenet\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m datamodule.\u001b[33mbatch_size\u001b[0m=\u001b[1;36m8\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mtrainer\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mone_gpu\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer/\u001b[33mlogger\u001b[0m=\u001b[35mwandb\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35mRTX8000\u001b[0m\\ -\\ batch\\ size\\ \u001b[1;36m8\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mgroup\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m =\u001b[35mBatch\u001b[0m\\ size\\ tests \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mtags\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n" - ] - } - ], - "source": [ - "%%capture\n", - "!python project/main.py \\\n", - " experiment=profiling_gpu \\\n", - " hydra.launcher.gres=gpu:rtx8000:1 \\\n", - " datamodule.batch_size=8" - ] - }, - { - "cell_type": "code", - "execution_count": 13, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2;36m[09/16/24 12:20:27]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=271112;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=705864;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m12\u001b[0m-\u001b[1;36m20\u001b[0m-\u001b[1;36m26\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=897324;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=330527;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[33malgorithm\u001b[0m=\u001b[35mexample\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mimagenet\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m datamodule.\u001b[33mbatch_size\u001b[0m=\u001b[1;36m32\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mtrainer\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mone_gpu\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer/\u001b[33mlogger\u001b[0m=\u001b[35mwandb\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35mRTX8000\u001b[0m\\ -\\ batch\\ size\\ \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m32\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mgroup\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m =\u001b[35mBatch\u001b[0m\\ size\\ tests \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mtags\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n" - ] - } - ], - "source": [ - "%%capture\n", - "!python project/main.py \\\n", - " experiment=profiling_gpu \\\n", - " hydra.launcher.gres=gpu:rtx8000:1 \\\n", - " datamodule.batch_size=32" - ] - }, - { - "cell_type": "code", - "execution_count": 14, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2;36m[09/16/24 12:35:51]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=831424;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=96207;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m12\u001b[0m-\u001b[1;36m35\u001b[0m-\u001b[1;36m50\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=663693;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=91463;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[33malgorithm\u001b[0m=\u001b[35mexample\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mimagenet\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m datamodule.\u001b[33mbatch_size\u001b[0m=\u001b[1;36m128\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mtrainer\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mone_gpu\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer/\u001b[33mlogger\u001b[0m=\u001b[35mwandb\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35mRTX8000\u001b[0m\\ -\\ batch\\ size\\ \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m128\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mgroup\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m =\u001b[35mBatch\u001b[0m\\ size\\ tests \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mtags\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n" - ] - } - ], - "source": [ - "%%capture\n", - "!python project/main.py \\\n", - " experiment=profiling_gpu \\\n", - " hydra.launcher.gres=gpu:rtx8000:1 \\\n", - " datamodule.batch_size=128" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Additional optimization" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Once GPU selection and a reasonable batch size are chosen, more can be done to speed up a model's computation.\n", - "- a\n", - "- la" - ] - }, - { - "cell_type": "code", - "execution_count": 16, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2;36m[09/16/24 13:05:32]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=865887;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=436119;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m16\u001b[0m/\u001b[1;36m13\u001b[0m-\u001b[1;36m05\u001b[0m-\u001b[1;36m32\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=50432;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=381740;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[33malgorithm\u001b[0m=\u001b[35mexample\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mimagenet\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m datamodule.\u001b[33mbatch_size\u001b[0m=\u001b[1;36m32\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m datamodule.\u001b[33mpin_memory\u001b[0m=\u001b[3;91mFals\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[3;91me\u001b[0m \u001b[33mtrainer\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mone_gpu\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer/\u001b[33mlogger\u001b[0m=\u001b[35mwandb\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35mGPU\u001b[0m\\ -\\ RTX8000\\ -\\ \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m pinned\\ memory \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mgroup\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m =\u001b[35mPin\u001b[0m\\ memory \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mtags\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n" - ] - } - ], - "source": [ - "%%capture\n", - "!python project/main.py \\\n", - " experiment=profiling_gpu \\\n", - " datamodule.batch_size=32\n", - " datamodule.num_workers=4 ## optimal parameter from above tests, check" + " datamodule.batch_size=1,8,32,64,128" ] }, { diff --git a/project/configs/experiment/profiling_cpu.yaml b/project/configs/experiment/profiling_cpu.yaml deleted file mode 100644 index 4c998d77..00000000 --- a/project/configs/experiment/profiling_cpu.yaml +++ /dev/null @@ -1,15 +0,0 @@ -# @package _global_ -# yaml-language-server: $schema=../../../.schemas/experiment_profiling_cpu_schema.json - -defaults: - - override /datamodule: imagenet - - override /algorithm: no_op - - override /trainer/logger: wandb - - override /resources: cpu - -trainer: - min_epochs: 1 - max_epochs: 2 - limit_train_batches: 2 - limit_val_batches: 2 - num_sanity_val_steps: 0 diff --git a/project/configs/experiment/profiling_gpu.yaml b/project/configs/experiment/profiling_gpu.yaml deleted file mode 100644 index 11ac2071..00000000 --- a/project/configs/experiment/profiling_gpu.yaml +++ /dev/null @@ -1,19 +0,0 @@ -# @package _global_ -# yaml-language-server: $schema=../../../.schemas/experiment_profiling_gpu_schema.json - -defaults: - - override /datamodule: imagenet - - override /algorithm: example - - override /trainer/logger: wandb - - override /resources: one_gpu - -hydra: - launcher: - partition: unkillable - -trainer: - min_epochs: 1 - max_epochs: 2 - limit_train_batches: 2 - limit_val_batches: 2 - num_sanity_val_steps: 0 From 142b96335b31db14b7a57877104db0bd424182e9 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Thu, 19 Sep 2024 16:01:42 -0400 Subject: [PATCH 17/33] Added ImageNet training example --- docs/examples/profiling.ipynb | 190 ++++++++++++++++++++-------------- 1 file changed, 114 insertions(+), 76 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 4f577424..eab55ba6 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -26,7 +26,7 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": 2, "metadata": {}, "outputs": [], "source": [ @@ -85,7 +85,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "In addition to callback specification, the Mila Research template integrates wandb as a logger specification, which enables the tracking of additional metrics through visualizations and dashboard creation. Given the flexibility and widespread adoption of using WandB as a logger, we'll be using it for the remainder of this tutorial, which will then be visualizable at `wandb_url` , as supporting information for the experiments contained herein." + "In addition to specifying callbacks, the Mila Research template integrates using WandB as a logger, which enables the tracking of additional metrics through visualizations and dashboard creation. Given the flexibility and widespread adoption of the WandB logger, we'll be using it for the remainder of this tutorial, which will then be visualizable at `wandb_url` , as supporting information for the experiments contained herein." ] }, { @@ -104,115 +104,100 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "\u001b[2;36m[09/19/24 13:05:55]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=143271;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=313435;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[09/19/24 13:32:30]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=887253;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=551139;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m19\u001b[0m/\u001b[1;36m13\u001b[0m-\u001b[1;36m05\u001b[0m-\u001b[1;36m54\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=857793;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=99902;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m19\u001b[0m/\u001b[1;36m13\u001b[0m-\u001b[1;36m32\u001b[0m-\u001b[1;36m30\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=820907;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=581098;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m \u001b[33mexperiment\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[1;36m1\u001b[0m\\ RTX8000\\ GPU\\ \u001b[1;36m1\u001b[0m\\ CPU\\ \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Training \u001b[2m \u001b[0m\n", - "^C\n", - "Traceback (most recent call last):\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\", line 172, in \n", - " main()\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/main.py\", line 94, in decorated_main\n", - " _run_hydra(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 394, in _run_hydra\n", - " _run_app(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 465, in _run_app\n", - " run_and_report(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 220, in run_and_report\n", - " return func()\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 466, in \n", - " lambda: hydra.multirun(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/hydra.py\", line 162, in multirun\n", - " ret = sweeper.sweep(arguments=task_overrides)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/core_plugins/basic_sweeper.py\", line 177, in sweep\n", - " results = self.launcher.launch(batch, initial_job_idx=initial_job_idx)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\", line 146, in launch\n", - " return [j.results()[0] for j in jobs]\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\", line 146, in \n", - " return [j.results()[0] for j in jobs]\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/submitit/core/core.py\", line 284, in results\n", - " self.wait()\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/submitit/core/core.py\", line 400, in wait\n", - " _time.sleep(1)\n", - "KeyboardInterrupt\n" + "\u001b[2;36m \u001b[0m Training \u001b[2m \u001b[0m\n" ] } ], "source": [ - "#%%capture\n", + "%%capture\n", "!python project/main.py \\\n", " experiment=profiling \\\n", " hydra.launcher.gres='gpu:rtx8000:1' \\\n", " hydra.launcher.cpus_per_task=1 \\\n", - " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training\"" + " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - ImageNet\"" ] }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 5, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling \\\n", + " resources=cpu \\\n", + " trainer.logger.wandb.name=\"1 CPU Training - ImageNet\"" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Another example: the same comparison for training a small fcnetnetwork on MNIST. There, I suspect that the difference between GPU / CPU throughput shouldn't be that large." + ] + }, + { + "cell_type": "code", + "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "\u001b[2;36m[09/19/24 13:07:09]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=701107;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=395625;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m[09/19/24 15:49:07]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=854152;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=476561;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m19\u001b[0m/\u001b[1;36m13\u001b[0m-\u001b[1;36m07\u001b[0m-\u001b[1;36m08\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=233531;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=331458;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m19\u001b[0m/\u001b[1;36m15\u001b[0m-\u001b[1;36m49\u001b[0m-\u001b[1;36m06\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b[33mnetwork\u001b[0m=\u001b[35mfcnet\u001b[0m \u001b]8;id=577341;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=695335;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mmnist\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m \u001b[33mexperiment\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mresources\u001b[0m=\u001b[35mcpu\u001b[0m \u001b[2m \u001b[0m\n", "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m1\u001b[0m\\ CPU\\ Training \u001b[2m \u001b[0m\n", - "^C\n", - "Traceback (most recent call last):\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/project/main.py\", line 172, in \n", - " main()\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/main.py\", line 94, in decorated_main\n", - " _run_hydra(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 394, in _run_hydra\n", - " _run_app(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 465, in _run_app\n", - " run_and_report(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 220, in run_and_report\n", - " return func()\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/utils.py\", line 466, in \n", - " lambda: hydra.multirun(\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/hydra.py\", line 162, in multirun\n", - " ret = sweeper.sweep(arguments=task_overrides)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra/_internal/core_plugins/basic_sweeper.py\", line 177, in sweep\n", - " results = self.launcher.launch(batch, initial_job_idx=initial_job_idx)\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\", line 146, in launch\n", - " return [j.results()[0] for j in jobs]\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\", line 146, in \n", - " return [j.results()[0] for j in jobs]\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/submitit/core/core.py\", line 284, in results\n", - " self.wait()\n", - " File \"/home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/submitit/core/core.py\", line 400, in wait\n", - " _time.sleep(1)\n", - "KeyboardInterrupt\n" + "\u001b[2;36m \u001b[0m \u001b[1;36m1\u001b[0m\\ RTX8000\\ GPU\\ \u001b[1;36m1\u001b[0m\\ CPU\\ \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Training \u001b[2m \u001b[0m\n" ] } ], "source": [ - "#%%capture\n", + "%%capture\n", + "!python project/main.py \\\n", + " network=fcnet \\\n", + " datamodule=mnist \\\n", + " experiment=profiling \\\n", + " hydra.launcher.gres='gpu:rtx8000:1' \\\n", + " hydra.launcher.cpus_per_task=1 \\\n", + " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - MNIST\"" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", "!python project/main.py \\\n", + " network=fcnet \\\n", + " datamodule=mnist \\\n", " experiment=profiling \\\n", " resources=cpu \\\n", - " trainer.logger.wandb.name=\"1 CPU Training\"" + " trainer.logger.wandb.name=\"1 CPU Training - MNIST\"" ] }, { @@ -236,7 +221,52 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 6, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling \\\n", + " algorithm=no_op \\\n", + " trainer.logger.wandb.name=\"1 CPU Dataloading\" \\\n", + " hydra.launcher.cpus_per_task=1 \\\n", + " datamodule.num_workers=1,4,8,16,32" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling \\\n", + " algorithm=no_op \\\n", + " trainer.logger.wandb.name=\"2 CPU Dataloading\" \\\n", + " hydra.launcher.cpus_per_task=2 \\\n", + " datamodule.num_workers=1,4,8,16,32" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling \\\n", + " algorithm=no_op \\\n", + " trainer.logger.wandb.name=\"3 CPU Dataloading\" \\\n", + " hydra.launcher.cpus_per_task=3 \\\n", + " datamodule.num_workers=1,4,8,16,32" + ] + }, + { + "cell_type": "code", + "execution_count": 9, "metadata": {}, "outputs": [], "source": [ @@ -244,10 +274,18 @@ "!python project/main.py \\\n", " experiment=profiling \\\n", " algorithm=no_op \\\n", - " trainer.logger.wandb.name=\"Dataloading only\" \\\n", + " trainer.logger.wandb.name=\"4 CPU Dataloading\" \\\n", + " hydra.launcher.cpus_per_task=4 \\\n", " datamodule.num_workers=1,4,8,16,32" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Given the former dataloading configurations, we know that ... (DON'T RUN UNTIL POINTS 3.2-3.3 ARE VISIBLE)" + ] + }, { "cell_type": "code", "execution_count": null, @@ -255,7 +293,7 @@ "outputs": [], "source": [ "%%capture\n", - "!python project/main.py -m \\\n", + "!python project/main.py \\\n", " experiment=profiling \\\n", " hydra.launcher.gres='gpu:rtx8000:1' \\\n", " hydra.launcher.cpus_per_task=2 \\\n", @@ -269,7 +307,7 @@ "outputs": [], "source": [ "%%capture\n", - "!python project/main.py -m \\\n", + "!python project/main.py \\\n", " experiment=profiling \\\n", " hydra.launcher.gres='gpu:rtx8000:1' \\\n", " hydra.launcher.cpus_per_task=3 \\\n", @@ -283,7 +321,7 @@ "outputs": [], "source": [ "%%capture\n", - "!python project/main.py -m \\\n", + "!python project/main.py \\\n", " experiment=profiling \\\n", " hydra.launcher.gres='gpu:rtx8000:1' \\\n", " hydra.launcher.cpus_per_task=4 \\\n", From 38b42f049b5a08569e3fa9a5f6c93cd6860e8bd6 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Mon, 23 Sep 2024 10:40:48 -0400 Subject: [PATCH 18/33] nb restructuring --- docs/examples/profiling.ipynb | 202 ++++++++++++---------------------- 1 file changed, 70 insertions(+), 132 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index eab55ba6..89175861 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -26,7 +26,7 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 1, "metadata": {}, "outputs": [], "source": [ @@ -85,119 +85,19 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "In addition to specifying callbacks, the Mila Research template integrates using WandB as a logger, which enables the tracking of additional metrics through visualizations and dashboard creation. Given the flexibility and widespread adoption of the WandB logger, we'll be using it for the remainder of this tutorial, which will then be visualizable at `wandb_url` , as supporting information for the experiments contained herein." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Training models with GPUs" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Advancements in Graphical Processing Units (GPUs) are widely known to have enabled the deep learning revolution, particularly through faster computation, relative to CPUs. Given that we have the option to run both GPU and CPU workloads, let's compare their throughput. In most workflows, the speedup provided by a GPU is dramatic. For a few select workloads, particularly those with a low number of steps or lighter computation requirements, if a 1.5-2x slower performance is observed when using a CPU, as opposed to a GPU, the former may be worth considering, as they're a far less contested resource on the cluster and pose far fewer availability issues." - ] - }, - { - "cell_type": "code", - "execution_count": 4, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2;36m[09/19/24 13:32:30]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=887253;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=551139;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m19\u001b[0m/\u001b[1;36m13\u001b[0m-\u001b[1;36m32\u001b[0m-\u001b[1;36m30\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b]8;id=820907;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=581098;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[33mexperiment\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m1\u001b[0m\\ RTX8000\\ GPU\\ \u001b[1;36m1\u001b[0m\\ CPU\\ \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Training \u001b[2m \u001b[0m\n" - ] - } - ], - "source": [ - "%%capture\n", - "!python project/main.py \\\n", - " experiment=profiling \\\n", - " hydra.launcher.gres='gpu:rtx8000:1' \\\n", - " hydra.launcher.cpus_per_task=1 \\\n", - " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - ImageNet\"" - ] - }, - { - "cell_type": "code", - "execution_count": 5, - "metadata": {}, - "outputs": [], - "source": [ - "%%capture\n", - "!python project/main.py \\\n", - " experiment=profiling \\\n", - " resources=cpu \\\n", - " trainer.logger.wandb.name=\"1 CPU Training - ImageNet\"" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Another example: the same comparison for training a small fcnetnetwork on MNIST. There, I suspect that the difference between GPU / CPU throughput shouldn't be that large." - ] - }, - { - "cell_type": "code", - "execution_count": 10, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2;36m[09/19/24 15:49:07]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=854152;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=476561;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m19\u001b[0m/\u001b[1;36m15\u001b[0m-\u001b[1;36m49\u001b[0m-\u001b[1;36m06\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b[33mnetwork\u001b[0m=\u001b[35mfcnet\u001b[0m \u001b]8;id=577341;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=695335;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mmnist\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mexperiment\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m1\u001b[0m\\ RTX8000\\ GPU\\ \u001b[1;36m1\u001b[0m\\ CPU\\ \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Training \u001b[2m \u001b[0m\n" - ] - } - ], - "source": [ - "%%capture\n", - "!python project/main.py \\\n", - " network=fcnet \\\n", - " datamodule=mnist \\\n", - " experiment=profiling \\\n", - " hydra.launcher.gres='gpu:rtx8000:1' \\\n", - " hydra.launcher.cpus_per_task=1 \\\n", - " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - MNIST\"" + "In addition to specifying callbacks, the Mila Research template integrates using WandB as a logger, which enables the tracking of additional metrics through visualizations and dashboard creation. Given the flexibility and widespread adoption of the WandB logger, we'll be using it for the remainder of this tutorial, which will then be visualizable at `wandb_url` , as supporting information for the experiments contained herein. We will now proceed to establish a baseline to profile, diagnose and optimize throughout this example." ] }, { "cell_type": "code", - "execution_count": 11, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ "%%capture\n", "!python project/main.py \\\n", - " network=fcnet \\\n", - " datamodule=mnist \\\n", " experiment=profiling \\\n", - " resources=cpu \\\n", - " trainer.logger.wandb.name=\"1 CPU Training - MNIST\"" + " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - ResNet-18 - ImageNet\"" ] }, { @@ -229,7 +129,8 @@ "!python project/main.py \\\n", " experiment=profiling \\\n", " algorithm=no_op \\\n", - " trainer.logger.wandb.name=\"1 CPU Dataloading\" \\\n", + " resources=cpu \\\n", + " trainer.logger.wandb.tags=[\"1 CPU Dataloading\"] \\\n", " hydra.launcher.cpus_per_task=1 \\\n", " datamodule.num_workers=1,4,8,16,32" ] @@ -244,7 +145,8 @@ "!python project/main.py \\\n", " experiment=profiling \\\n", " algorithm=no_op \\\n", - " trainer.logger.wandb.name=\"2 CPU Dataloading\" \\\n", + " resources=cpu \\\n", + " trainer.logger.wandb.tags=[\"2 CPU Dataloading\"] \\\n", " hydra.launcher.cpus_per_task=2 \\\n", " datamodule.num_workers=1,4,8,16,32" ] @@ -259,7 +161,8 @@ "!python project/main.py \\\n", " experiment=profiling \\\n", " algorithm=no_op \\\n", - " trainer.logger.wandb.name=\"3 CPU Dataloading\" \\\n", + " resources=cpu \\\n", + " trainer.logger.wandb.tags=[\"3 CPU Dataloading\"] \\\n", " hydra.launcher.cpus_per_task=3 \\\n", " datamodule.num_workers=1,4,8,16,32" ] @@ -274,7 +177,8 @@ "!python project/main.py \\\n", " experiment=profiling \\\n", " algorithm=no_op \\\n", - " trainer.logger.wandb.name=\"4 CPU Dataloading\" \\\n", + " resources=cpu \\\n", + " trainer.logger.wandb.tags=[\"4 CPU Dataloading\"] \\\n", " hydra.launcher.cpus_per_task=4 \\\n", " datamodule.num_workers=1,4,8,16,32" ] @@ -283,49 +187,77 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Given the former dataloading configurations, we know that ... (DON'T RUN UNTIL POINTS 3.2-3.3 ARE VISIBLE)" + "## The advantages of training models with GPUs" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Advancements in Graphical Processing Units (GPUs) are widely known to have enabled the deep learning revolution, particularly through faster computation, relative to CPUs. Given that we have the option to run both GPU and CPU workloads, let's compare their throughput. In most workflows, the speedup provided by a GPU is dramatic. For a few select workloads, particularly those with a low number of steps or lighter computation requirements, if a 1.5-2x slower performance is observed when using a CPU, as opposed to a GPU, the former may be worth considering, as they're a far less contested resource on the cluster and pose far fewer availability issues. \n", + "\n", + "In this section, we'll train a model that's analogous to our ImageNet baseline - entirely on the CPU. We will also train two smaller fully connected networks on MNIST, a smaller dataset than ImageNet, to compare and contrast the differences in throughput when training with and without a GPU. " ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "%%capture\n", "!python project/main.py \\\n", " experiment=profiling \\\n", - " hydra.launcher.gres='gpu:rtx8000:1' \\\n", - " hydra.launcher.cpus_per_task=2 \\\n", - " trainer.logger.wandb.name=\"1 RTX8000 GPU 2 CPU Training\"" + " resources=cpu \\\n", + " trainer.logger.wandb.name=\"1 CPU Training - ResNet-18 - ImageNet\"" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 10, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2;36m[09/19/24 15:49:07]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=854152;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=476561;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m19\u001b[0m/\u001b[1;36m15\u001b[0m-\u001b[1;36m49\u001b[0m-\u001b[1;36m06\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b[33mnetwork\u001b[0m=\u001b[35mfcnet\u001b[0m \u001b]8;id=577341;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=695335;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", + "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mmnist\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[33mexperiment\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m \u001b[1;36m1\u001b[0m\\ RTX8000\\ GPU\\ \u001b[1;36m1\u001b[0m\\ CPU\\ \u001b[2m \u001b[0m\n", + "\u001b[2;36m \u001b[0m Training \u001b[2m \u001b[0m\n" + ] + } + ], "source": [ "%%capture\n", "!python project/main.py \\\n", + " network=fcnet \\\n", + " datamodule=mnist \\\n", " experiment=profiling \\\n", " hydra.launcher.gres='gpu:rtx8000:1' \\\n", - " hydra.launcher.cpus_per_task=3 \\\n", - " trainer.logger.wandb.name=\"1 RTX8000 GPU 3 CPU Training\"" + " hydra.launcher.cpus_per_task=1 \\\n", + " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - FcNet - MNIST\"" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "%%capture\n", "!python project/main.py \\\n", + " network=fcnet \\\n", + " datamodule=mnist \\\n", " experiment=profiling \\\n", - " hydra.launcher.gres='gpu:rtx8000:1' \\\n", - " hydra.launcher.cpus_per_task=4 \\\n", - " trainer.logger.wandb.name=\"1 RTX8000 GPU 4 CPU Training\"" + " resources=cpu \\\n", + " trainer.logger.wandb.name=\"1 CPU Training - FcNet- MNIST\"" ] }, { @@ -339,8 +271,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "[Mila's official documentation](https://docs.mila.quebec/Information.html) has a comprehensive rundown of the GPUs that are installed on the cluster. Typing ```savail``` on the command line when logged into the cluster, shows their current availability. Testing their capacity can yield insights into their suitability for different training workloads.\n", - "As the Mila Research template is built with hydra as a configuration manager, it integrates [Multi-runs](https://hydra.cc/docs/tutorials/basic/running_your_app/multi-run/) by default. This makes it possible to sweep over different parameters for profiling or throughput testing purposes or both. For example, suppose we wanted to figure out how different GPUs perform relative to each other. " + "Observing the former, we've made a solid case for utilizing GPUs for model training. Furthermore, when using GPUs, these vary in throughput; some are more powerful than others. [Mila's official documentation](https://docs.mila.quebec/Information.html) has a comprehensive rundown of the GPUs that are installed on the cluster. Typing ```savail``` on the command line when logged into the cluster, shows their current availability. Testing their capacity can yield insights into their suitability for different training workloads. Let's see what's available on the Mila cluster." ] }, { @@ -377,9 +308,16 @@ "\n", "- NVIDIA Tensor Core GPUs: A100, A100L, V100 (previous gen)\n", "- NVIDIA RTX GPUs: A6000, RTX8000\n", - "- Multi-Instance GPU (MiG) partitions: 2g.20gb, 3g.40gb, 4g.40gb \n", - "\n", - "We will now proceed to specify different GPUs over training runs and compare their throughput. If a GPU with lower maximum capacity is readily available, training on it may be more time and resource effective than waiting for higher capacity GPUs to become available.\n" + "- Multi-Instance GPU (MiG) partitions: 2g.20gb, 3g.40gb, 4g.40gb " + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "As the Mila Research template is built with hydra as a configuration manager, it integrates [Multi-runs](https://hydra.cc/docs/tutorials/basic/running_your_app/multi-run/) by default. This makes it possible to specify particular GPU resources for a given run, or sweeping over different parameters for profiling or throughput testing purposes or both. \n", + "For example, suppose we wanted to figure out how different GPUs perform relative to each other. \n", + "We are able to do this by specifying different GPUs over training runs and comparing their throughput. " ] }, { @@ -417,7 +355,7 @@ " hydra.launcher.gres=gpu:a100:1 \\\n", " trainer.logger.wandb.name=\"A100\" \\\n", " datamodule.num_workers=#optimal params as determined before\n", - " trainer.logger.wandb.name=\"A100 GPU X CPU X Num_workers\"" + " trainer.logger.wandb.name=\"A100 GPU X CPU X Num_workers - ResNet-18 - ImageNet\"" ] }, { @@ -432,22 +370,22 @@ " hydra.launcher.gres=gpu:v100:1 \\\n", " trainer.logger.wandb.name=\"V100\" \\\n", " datamodule.num_workers=#optimal param as determined before \n", - " trainer.logger.wandb.name=\"V100 GPU X CPU X Num_workers\"" + " trainer.logger.wandb.name=\"V100 GPU X CPU X Num_workers - ResNet-18 - ImageNet\"" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## GPU utilization" + "## Maximizing GPU efficiency and utilization" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "How well are we using a given GPU? Once we've done a few preliminary runs with candidate GPUs that we'd want to use, the GPU utilization can be measured and optimized. We generally aim for high GPU utilization. Is the GPU utilization high? (>80%?) \n", - "If it's low (<80%), then we can use the PyTorch profiler (or similar tools) to try to figure out where the bottleneck lies, and further tune our parameters to increase our utilization." + "While there is a clear difference in throughput between GPU types, if a GPU with lower maximum capacity is readily available, training on it may be more time and resource effective than waiting for higher capacity GPUs to become available. Optimizing a lower capacity GPU may be sufficient for your use case. How well are is a given GPU being utilized? Once we've done a few preliminary runs with candidate GPU configurations that we'd want to use, the GPU utilization can be measured and optimized. \n", + "We generally aim for high GPU utilization. Is the GPU utilization high? (>80%)? If it's low (<80%), then we can use the PyTorch profiler (or similar tools) to try to figure out where the bottleneck lies, and further tune our parameters to increase our utilization. " ] }, { @@ -480,7 +418,7 @@ } ], "source": [ - "%%capture\n", + "##%%capture ------ PLACEHOLDER: OPTIMAL PARAMETERS FROM SECTION 3 REQUIRED BEFORE RUNNING ------\n", "!python project/main.py \\\n", " experiment=profiling_gpu \\\n", " hydra.launcher.gres=gpu:rtx8000:1 \\\n", From 26c50bef342909493e1cad8230ce78e192dfdbc8 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Mon, 23 Sep 2024 11:07:28 -0400 Subject: [PATCH 19/33] pre-run placeholders --- docs/examples/profiling.ipynb | 39 ++++++++++++++++++++++++++--------- 1 file changed, 29 insertions(+), 10 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 89175861..c4a6d421 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -14,7 +14,7 @@ "The Mila Research Template leverages built-in PyTorch and Lightning functionality to make model profiling and benchmarking accessible and flexible. \n", "Make sure to read the Mila Docs page on [PLACEHOLDER - profiling](https://docs.mila.quebec/) before going through this example. \n", "\n", - "The research template profiling notebook extends the examples in the official documentation with additional tools: notably, native WandB integration to monitor performance and using hydra multiruns to compare the available GPUs on the official Mila cluster. See below. The goal of this notebook is to introduce profiling, present tools useful for doing so and to provide general concepts and guidelines for optimizing your code, within the Mila cluster ecosystem.\n" + "The Research Template's profiling notebook extends the examples in the official documentation with additional tools: notably, native WandB integration to monitor performance and using hydra multiruns to compare the available GPUs on the official Mila cluster. See below. The goal of this notebook is to introduce profiling, present tools useful for doing so and to provide general concepts and guidelines for optimizing your code, within the Mila cluster ecosystem.\n" ] }, { @@ -71,14 +71,14 @@ "- CPU/GPU utilization \n", "- RAM/VRAM utilization\n", "\n", - "In the Mila ResearchTemplate, this can be done by passing a callback to the trainer. Supported configs are found within the project template at `configs/trainer/callbacks`. Throughout this tutorial, we will use the default callback, which in turn implements early stopping and tracks the learning rate, device utilisation and throughput, each through a specific callback instance." + "In the Mila Research Template, this can be done by passing a callback to the trainer. Supported configs are found within the project template at `configs/trainer/callbacks`. Throughout this tutorial, we will use the default callback, which in turn implements early stopping and tracks the learning rate, device utilisation and throughput, each through a specific callback instance." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "### Logging metrics on WandB" + "### Running a baseline and logging metrics on WandB" ] }, { @@ -183,6 +183,25 @@ " datamodule.num_workers=1,4,8,16,32" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Once we've determined the optimal number of workers and CPUs in terms of dataloading throughput, we can train a model similar to our baseline, albeit with the newly obtained parameters, to then compare throughput and determine if there was a sizeable increase." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#%%capture ----- RUN WITH OPTIMAL PARAMETERS ONCE DETERMINED, LOCALLY -----\n", + "!python project/main.py \\\n", + " experiment=profiling \\\n", + " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - ResNet-18 - ImageNet\"" + ] + }, { "cell_type": "markdown", "metadata": {}, @@ -205,7 +224,7 @@ "metadata": {}, "outputs": [], "source": [ - "%%capture\n", + "%%capture ### USE OPTIMIZED NUM CPUS, WORKERS, BEFORE RUNNING\n", "!python project/main.py \\\n", " experiment=profiling \\\n", " resources=cpu \\\n", @@ -241,7 +260,7 @@ " datamodule=mnist \\\n", " experiment=profiling \\\n", " hydra.launcher.gres='gpu:rtx8000:1' \\\n", - " hydra.launcher.cpus_per_task=1 \\\n", + " ###hydra.launcher.cpus_per_task=1 \\ USE OPTIMIZED NUM CPUS, WORKERS\n", " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - FcNet - MNIST\"" ] }, @@ -251,13 +270,13 @@ "metadata": {}, "outputs": [], "source": [ - "%%capture\n", + "%%capture ### USE OPTIMIZED NUM CPUS, WORKERS, BEFORE RUNNING\n", "!python project/main.py \\\n", " network=fcnet \\\n", " datamodule=mnist \\\n", " experiment=profiling \\\n", " resources=cpu \\\n", - " trainer.logger.wandb.name=\"1 CPU Training - FcNet- MNIST\"" + " trainer.logger.wandb.name=\"1 CPU Training - FcNet - MNIST\"" ] }, { @@ -315,7 +334,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "As the Mila Research template is built with hydra as a configuration manager, it integrates [Multi-runs](https://hydra.cc/docs/tutorials/basic/running_your_app/multi-run/) by default. This makes it possible to specify particular GPU resources for a given run, or sweeping over different parameters for profiling or throughput testing purposes or both. \n", + "As the Mila Research Template is built with hydra as a configuration manager, it integrates [Multi-runs](https://hydra.cc/docs/tutorials/basic/running_your_app/multi-run/) by default. This makes it possible to specify particular GPU resources for a given run, or sweeping over different parameters for profiling or throughput testing purposes or both. \n", "For example, suppose we wanted to figure out how different GPUs perform relative to each other. \n", "We are able to do this by specifying different GPUs over training runs and comparing their throughput. " ] @@ -349,7 +368,7 @@ } ], "source": [ - "%%capture\n", + "%%capture ### USE OPTIMIZED NUM CPUS, WORKERS, BEFORE RUNNING\n", "!python project/main.py \\\n", " experiment=profiling_gpu \\\n", " hydra.launcher.gres=gpu:a100:1 \\\n", @@ -364,7 +383,7 @@ "metadata": {}, "outputs": [], "source": [ - "%%capture\n", + "%%capture ### USE OPTIMIZED NUM CPUS, WORKERS, BEFORE RUNNING\n", "!python project/main.py \\\n", " experiment=profiling_gpu \\\n", " hydra.launcher.gres=gpu:v100:1 \\\n", From d7f5931378dc9d8313c1c504928b1fe85bd3ce2f Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Mon, 23 Sep 2024 11:11:38 -0400 Subject: [PATCH 20/33] added placeholder for optimized training run, point 3.3 --- docs/examples/profiling.ipynb | 21 ++++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index c4a6d421..23c7fbd1 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -97,7 +97,8 @@ "%%capture\n", "!python project/main.py \\\n", " experiment=profiling \\\n", - " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - ResNet-18 - ImageNet\"" + " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - ResNet-18 - ImageNet\" \\\n", + " trainer.logger.wandb.tags=[\"Training baseline\"] " ] }, { @@ -199,7 +200,8 @@ "#%%capture ----- RUN WITH OPTIMAL PARAMETERS ONCE DETERMINED, LOCALLY -----\n", "!python project/main.py \\\n", " experiment=profiling \\\n", - " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - ResNet-18 - ImageNet\"" + " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - ResNet-18 - ImageNet\" \\\n", + " trainer.logger.wandb.tags=[\"Optimized\"] " ] }, { @@ -228,7 +230,8 @@ "!python project/main.py \\\n", " experiment=profiling \\\n", " resources=cpu \\\n", - " trainer.logger.wandb.name=\"1 CPU Training - ResNet-18 - ImageNet\"" + " trainer.logger.wandb.name=\"1 CPU Training - ResNet-18 - ImageNet\" \\\n", + " trainer.logger.wandb.tags=[\"CPU Training\"] " ] }, { @@ -261,7 +264,8 @@ " experiment=profiling \\\n", " hydra.launcher.gres='gpu:rtx8000:1' \\\n", " ###hydra.launcher.cpus_per_task=1 \\ USE OPTIMIZED NUM CPUS, WORKERS\n", - " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - FcNet - MNIST\"" + " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - FcNet - MNIST\"\n", + " trainer.logger.wandb.tags=[\"GPU Training\", \"MNIST\"] " ] }, { @@ -276,7 +280,8 @@ " datamodule=mnist \\\n", " experiment=profiling \\\n", " resources=cpu \\\n", - " trainer.logger.wandb.name=\"1 CPU Training - FcNet - MNIST\"" + " trainer.logger.wandb.name=\"1 CPU Training - FcNet - MNIST\" \\\n", + " trainer.logger.wandb.tags=[\"CPU Training\", \"MNIST\"] " ] }, { @@ -374,7 +379,8 @@ " hydra.launcher.gres=gpu:a100:1 \\\n", " trainer.logger.wandb.name=\"A100\" \\\n", " datamodule.num_workers=#optimal params as determined before\n", - " trainer.logger.wandb.name=\"A100 GPU X CPU X Num_workers - ResNet-18 - ImageNet\"" + " trainer.logger.wandb.name=\"A100 GPU X CPU X Num_workers - ResNet-18 - ImageNet\" \\\n", + " trainer.logger.wandb.tags=[\"GPU Training\", \"ImageNet\"]" ] }, { @@ -389,7 +395,8 @@ " hydra.launcher.gres=gpu:v100:1 \\\n", " trainer.logger.wandb.name=\"V100\" \\\n", " datamodule.num_workers=#optimal param as determined before \n", - " trainer.logger.wandb.name=\"V100 GPU X CPU X Num_workers - ResNet-18 - ImageNet\"" + " trainer.logger.wandb.name=\"V100 GPU X CPU X Num_workers - ResNet-18 - ImageNet\" \\\n", + " trainer.logger.wandb.tags=[\"GPU Training\", \"ImageNet\"]" ] }, { From 19b8de8386247a7b2e0807aa9719a2946deb3d20 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Mon, 23 Sep 2024 15:02:06 -0400 Subject: [PATCH 21/33] added cpu constraint --- docs/examples/profiling.ipynb | 103 +++++++++++++++++++--------------- 1 file changed, 58 insertions(+), 45 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 23c7fbd1..392ce6f6 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -90,22 +90,22 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "%%capture\n", "!python project/main.py \\\n", " experiment=profiling \\\n", - " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - ResNet-18 - ImageNet\" \\\n", - " trainer.logger.wandb.tags=[\"Training baseline\"] " + " trainer.logger.wandb.name=\"Baseline with training\" \\\n", + " trainer.logger.wandb.tags=[\"Training\",\"Baseline comparison\"] " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## Identifying potential bottleneck sources " + "## Identifying potential bottlenecks " ] }, { @@ -122,7 +122,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 18, "metadata": {}, "outputs": [], "source": [ @@ -130,15 +130,27 @@ "!python project/main.py \\\n", " experiment=profiling \\\n", " algorithm=no_op \\\n", - " resources=cpu \\\n", - " trainer.logger.wandb.tags=[\"1 CPU Dataloading\"] \\\n", - " hydra.launcher.cpus_per_task=1 \\\n", + " trainer.logger.wandb.name=\"Baseline without training\" \\\n", + " trainer.logger.wandb.tags=[\"No training\",\"Baseline comparison\"] " + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", + "!python project/main.py -m \\\n", + " experiment=profiling \\\n", + " algorithm=no_op \\\n", + " trainer.logger.wandb.tags=[\"1 CPU Dataloading\",\"Worker throughput\"] \\\n", " datamodule.num_workers=1,4,8,16,32" ] }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 26, "metadata": {}, "outputs": [], "source": [ @@ -147,14 +159,15 @@ " experiment=profiling \\\n", " algorithm=no_op \\\n", " resources=cpu \\\n", - " trainer.logger.wandb.tags=[\"2 CPU Dataloading\"] \\\n", + " trainer.logger.wandb.tags=[\"2 CPU Dataloading\",\"Worker throughput\"] \\\n", " hydra.launcher.cpus_per_task=2 \\\n", + " hydra.launcher.constraint=\"sapphire\" \\\n", " datamodule.num_workers=1,4,8,16,32" ] }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 27, "metadata": {}, "outputs": [], "source": [ @@ -163,14 +176,15 @@ " experiment=profiling \\\n", " algorithm=no_op \\\n", " resources=cpu \\\n", - " trainer.logger.wandb.tags=[\"3 CPU Dataloading\"] \\\n", + " trainer.logger.wandb.tags=[\"3 CPU Dataloading\",\"Worker throughput\"] \\\n", " hydra.launcher.cpus_per_task=3 \\\n", + " hydra.launcher.constraint=\"sapphire\" \\\n", " datamodule.num_workers=1,4,8,16,32" ] }, { "cell_type": "code", - "execution_count": 9, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ @@ -179,8 +193,9 @@ " experiment=profiling \\\n", " algorithm=no_op \\\n", " resources=cpu \\\n", - " trainer.logger.wandb.tags=[\"4 CPU Dataloading\"] \\\n", + " trainer.logger.wandb.tags=[\"4 CPU Dataloading\",\"Worker throughput\"] \\\n", " hydra.launcher.cpus_per_task=4 \\\n", + " hydra.launcher.constraint=\"sapphire\" \\\n", " datamodule.num_workers=1,4,8,16,32" ] }, @@ -192,32 +207,32 @@ ] }, { - "cell_type": "code", - "execution_count": null, + "cell_type": "markdown", "metadata": {}, - "outputs": [], "source": [ - "#%%capture ----- RUN WITH OPTIMAL PARAMETERS ONCE DETERMINED, LOCALLY -----\n", - "!python project/main.py \\\n", - " experiment=profiling \\\n", - " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - ResNet-18 - ImageNet\" \\\n", - " trainer.logger.wandb.tags=[\"Optimized\"] " + "## The advantages of training models with GPUs" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## The advantages of training models with GPUs" + "Advancements in Graphical Processing Units (GPUs) are widely known to have enabled the deep learning revolution, particularly through faster computation, relative to CPUs. Given that we have the option to run both GPU and CPU workloads, let's compare their throughput. In most workflows, the speedup provided by a GPU is dramatic. For a few select workloads, particularly those with a low number of steps or lighter computation requirements, if a 1.5-2x slower performance is observed when using a CPU, as opposed to a GPU, the former may be worth considering, as they're a far less contested resource on the cluster and pose far fewer availability issues. \n", + "\n", + "In this section, we'll train a model that's analogous to our ImageNet baseline - entirely on the CPU. We will use the optimal dataloading parameters, as determined by the previous runs. We will also train two smaller fully connected networks on MNIST, a smaller dataset than ImageNet, to compare and contrast the differences in throughput when training with and without a GPU. " ] }, { - "cell_type": "markdown", + "cell_type": "code", + "execution_count": null, "metadata": {}, + "outputs": [], "source": [ - "Advancements in Graphical Processing Units (GPUs) are widely known to have enabled the deep learning revolution, particularly through faster computation, relative to CPUs. Given that we have the option to run both GPU and CPU workloads, let's compare their throughput. In most workflows, the speedup provided by a GPU is dramatic. For a few select workloads, particularly those with a low number of steps or lighter computation requirements, if a 1.5-2x slower performance is observed when using a CPU, as opposed to a GPU, the former may be worth considering, as they're a far less contested resource on the cluster and pose far fewer availability issues. \n", - "\n", - "In this section, we'll train a model that's analogous to our ImageNet baseline - entirely on the CPU. We will also train two smaller fully connected networks on MNIST, a smaller dataset than ImageNet, to compare and contrast the differences in throughput when training with and without a GPU. " + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling \\\n", + " trainer.logger.wandb.name=\"Optimized training run\" \\\n", + " trainer.logger.wandb.tags=[\"Optimized\",\"CPU/GPU comparison\",\"GPU\",\"Baseline comparison\", \"GPU comparison\"] " ] }, { @@ -230,8 +245,8 @@ "!python project/main.py \\\n", " experiment=profiling \\\n", " resources=cpu \\\n", - " trainer.logger.wandb.name=\"1 CPU Training - ResNet-18 - ImageNet\" \\\n", - " trainer.logger.wandb.tags=[\"CPU Training\"] " + " trainer.logger.wandb.name=\"CPU training\" \\\n", + " trainer.logger.wandb.tags=[\"CPU/GPU comparison\",\"CPU\"] " ] }, { @@ -263,9 +278,9 @@ " datamodule=mnist \\\n", " experiment=profiling \\\n", " hydra.launcher.gres='gpu:rtx8000:1' \\\n", - " ###hydra.launcher.cpus_per_task=1 \\ USE OPTIMIZED NUM CPUS, WORKERS\n", - " trainer.logger.wandb.name=\"1 RTX8000 GPU 1 CPU Training - FcNet - MNIST\"\n", - " trainer.logger.wandb.tags=[\"GPU Training\", \"MNIST\"] " + " hydra.launcher.cpus_per_task=1 \\\n", + " trainer.logger.wandb.name=\"Optimized training run (FcNet/MNIST)\"\n", + " trainer.logger.wandb.tags=[\"CPU/GPU comparison\",\"GPU\",\"MNIST\"] " ] }, { @@ -280,8 +295,8 @@ " datamodule=mnist \\\n", " experiment=profiling \\\n", " resources=cpu \\\n", - " trainer.logger.wandb.name=\"1 CPU Training - FcNet - MNIST\" \\\n", - " trainer.logger.wandb.tags=[\"CPU Training\", \"MNIST\"] " + " trainer.logger.wandb.name=\"CPU training (FcNet/MNIST)\" \\\n", + " trainer.logger.wandb.tags=[\"CPU/GPU comparison\",\"CPU\",\"MNIST\"] " ] }, { @@ -373,14 +388,13 @@ } ], "source": [ - "%%capture ### USE OPTIMIZED NUM CPUS, WORKERS, BEFORE RUNNING\n", + "%%capture \n", "!python project/main.py \\\n", " experiment=profiling_gpu \\\n", - " hydra.launcher.gres=gpu:a100:1 \\\n", - " trainer.logger.wandb.name=\"A100\" \\\n", - " datamodule.num_workers=#optimal params as determined before\n", - " trainer.logger.wandb.name=\"A100 GPU X CPU X Num_workers - ResNet-18 - ImageNet\" \\\n", - " trainer.logger.wandb.tags=[\"GPU Training\", \"ImageNet\"]" + " hydra.launcher.gres='gpu:a100:1' \\\n", + " datamodule.num_workers= xxx \\\n", + " trainer.logger.wandb.name=\"A100 training\" \\\n", + " trainer.logger.wandb.tags=[\"GPU comparison\"]" ] }, { @@ -389,14 +403,13 @@ "metadata": {}, "outputs": [], "source": [ - "%%capture ### USE OPTIMIZED NUM CPUS, WORKERS, BEFORE RUNNING\n", + "%%capture\n", "!python project/main.py \\\n", " experiment=profiling_gpu \\\n", " hydra.launcher.gres=gpu:v100:1 \\\n", - " trainer.logger.wandb.name=\"V100\" \\\n", - " datamodule.num_workers=#optimal param as determined before \n", - " trainer.logger.wandb.name=\"V100 GPU X CPU X Num_workers - ResNet-18 - ImageNet\" \\\n", - " trainer.logger.wandb.tags=[\"GPU Training\", \"ImageNet\"]" + " datamodule.num_workers= xxx \\\n", + " trainer.logger.wandb.name=\"V100 training\" \\\n", + " trainer.logger.wandb.tags=[\"GPU comparison\"]" ] }, { From f5adb255ff063271b9c680ccac77104894972d46 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Tue, 24 Sep 2024 13:58:53 -0400 Subject: [PATCH 22/33] added all mnist runs --- docs/examples/profiling.ipynb | 158 ++++++++++++++++++++++++---------- 1 file changed, 114 insertions(+), 44 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 392ce6f6..937299b9 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -90,7 +90,7 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 2, "metadata": {}, "outputs": [], "source": [ @@ -98,7 +98,7 @@ "!python project/main.py \\\n", " experiment=profiling \\\n", " trainer.logger.wandb.name=\"Baseline with training\" \\\n", - " trainer.logger.wandb.tags=[\"Training\",\"Baseline comparison\"] " + " trainer.logger.wandb.tags=[\"Training\",\"Baseline comparison\", \"CPU/GPU comparison\"] " ] }, { @@ -122,7 +122,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 3, "metadata": {}, "outputs": [], "source": [ @@ -136,7 +136,7 @@ }, { "cell_type": "code", - "execution_count": 25, + "execution_count": 8, "metadata": {}, "outputs": [], "source": [ @@ -145,12 +145,12 @@ " experiment=profiling \\\n", " algorithm=no_op \\\n", " trainer.logger.wandb.tags=[\"1 CPU Dataloading\",\"Worker throughput\"] \\\n", - " datamodule.num_workers=1,4,8,16,32" + " datamodule.num_workers=1,4,8,16,32" ] }, { "cell_type": "code", - "execution_count": 26, + "execution_count": 5, "metadata": {}, "outputs": [], "source": [ @@ -160,6 +160,7 @@ " algorithm=no_op \\\n", " resources=cpu \\\n", " trainer.logger.wandb.tags=[\"2 CPU Dataloading\",\"Worker throughput\"] \\\n", + " hydra.launcher.timeout_min=60 \\\n", " hydra.launcher.cpus_per_task=2 \\\n", " hydra.launcher.constraint=\"sapphire\" \\\n", " datamodule.num_workers=1,4,8,16,32" @@ -167,7 +168,7 @@ }, { "cell_type": "code", - "execution_count": 27, + "execution_count": 6, "metadata": {}, "outputs": [], "source": [ @@ -177,6 +178,7 @@ " algorithm=no_op \\\n", " resources=cpu \\\n", " trainer.logger.wandb.tags=[\"3 CPU Dataloading\",\"Worker throughput\"] \\\n", + " hydra.launcher.timeout_min=60 \\\n", " hydra.launcher.cpus_per_task=3 \\\n", " hydra.launcher.constraint=\"sapphire\" \\\n", " datamodule.num_workers=1,4,8,16,32" @@ -184,7 +186,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 7, "metadata": {}, "outputs": [], "source": [ @@ -194,6 +196,7 @@ " algorithm=no_op \\\n", " resources=cpu \\\n", " trainer.logger.wandb.tags=[\"4 CPU Dataloading\",\"Worker throughput\"] \\\n", + " hydra.launcher.timeout_min=60 \\\n", " hydra.launcher.cpus_per_task=4 \\\n", " hydra.launcher.constraint=\"sapphire\" \\\n", " datamodule.num_workers=1,4,8,16,32" @@ -224,78 +227,141 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "%%capture\n", + "## make sure to run in a local session with the optimized number of CPU cores, num workers\n", "!python project/main.py \\\n", " experiment=profiling \\\n", + " algorithm=no_op \\\n", + " datamodule.num_workers=8 \\\n", + " trainer.logger.wandb.name=\"Optimized run without training\" \\\n", + " trainer.logger.wandb.tags=[\"Optimized\",\"CPU/GPU comparison\"] " + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling \\\n", + " resources=one_gpu \\\n", + " hydra.launcher.gres='gpu:rtx8000:1' \\\n", + " hydra.launcher.cpus_per_task=4 \\\n", + " datamodule.num_workers=8 \\\n", " trainer.logger.wandb.name=\"Optimized training run\" \\\n", - " trainer.logger.wandb.tags=[\"Optimized\",\"CPU/GPU comparison\",\"GPU\",\"Baseline comparison\", \"GPU comparison\"] " + " trainer.logger.wandb.tags=[\"Optimized\",\"CPU/GPU comparison\",\"GPU\",\"Baseline comparison\",\"GPU comparison\"] " ] }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 4, "metadata": {}, "outputs": [], "source": [ - "%%capture ### USE OPTIMIZED NUM CPUS, WORKERS, BEFORE RUNNING\n", + "%%capture \n", "!python project/main.py \\\n", " experiment=profiling \\\n", " resources=cpu \\\n", + " hydra.launcher.cpus_per_task=4 \\\n", + " datamodule.num_workers=8 \\\n", " trainer.logger.wandb.name=\"CPU training\" \\\n", " trainer.logger.wandb.tags=[\"CPU/GPU comparison\",\"CPU\"] " ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We will now proceed to run a similar comparison for the MNIST dataset, with a smaller, FcNet model." + ] + }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 2, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[2;36m[09/19/24 15:49:07]\u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m Submitit \u001b[32m'slurm'\u001b[0m sweep \u001b]8;id=854152;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=476561;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#120\u001b\\\u001b[2m120\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m output dir : \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m logs/default/multiruns/\u001b[1;36m202\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m4\u001b[0m-\u001b[1;36m09\u001b[0m-\u001b[1;36m19\u001b[0m/\u001b[1;36m15\u001b[0m-\u001b[1;36m49\u001b[0m-\u001b[1;36m06\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m\u001b[2;36m \u001b[0m\u001b[34mINFO \u001b[0m #\u001b[1;36m0\u001b[0m : \u001b[33mnetwork\u001b[0m=\u001b[35mfcnet\u001b[0m \u001b]8;id=577341;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py\u001b\\\u001b[2msubmitit_launcher.py\u001b[0m\u001b]8;;\u001b\\\u001b[2m:\u001b[0m\u001b]8;id=695335;file:///home/mila/c/cesar.valdez/idt/ResearchTemplate/.venv/lib/python3.10/site-packages/hydra_plugins/hydra_submitit_launcher/submitit_launcher.py#134\u001b\\\u001b[2m134\u001b[0m\u001b]8;;\u001b\\\n", - "\u001b[2;36m \u001b[0m \u001b[33mdatamodule\u001b[0m=\u001b[35mmnist\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[33mexperiment\u001b[0m=\u001b[35mprofiling\u001b[0m \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m trainer.logger.wandb.\u001b[33mname\u001b[0m= \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m \u001b[1;36m1\u001b[0m\\ RTX8000\\ GPU\\ \u001b[1;36m1\u001b[0m\\ CPU\\ \u001b[2m \u001b[0m\n", - "\u001b[2;36m \u001b[0m Training \u001b[2m \u001b[0m\n" - ] - } - ], + "outputs": [], "source": [ "%%capture\n", "!python project/main.py \\\n", + " experiment=profiling \\\n", " network=fcnet \\\n", " datamodule=mnist \\\n", - " experiment=profiling \\\n", - " hydra.launcher.gres='gpu:rtx8000:1' \\\n", - " hydra.launcher.cpus_per_task=1 \\\n", - " trainer.logger.wandb.name=\"Optimized training run (FcNet/MNIST)\"\n", + " trainer.logger.wandb.name=\"FcNet/MNIST baseline with training\" \\\n", " trainer.logger.wandb.tags=[\"CPU/GPU comparison\",\"GPU\",\"MNIST\"] " ] }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 3, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " experiment=profiling \\\n", + " algorithm=no_op \\\n", + " datamodule=mnist \\\n", + " trainer.logger.wandb.name=\"FcNet/MNIST baseline without training\" \\\n", + " trainer.logger.wandb.tags=[\"CPU/GPU comparison\",\"CPU\",\"MNIST\"] " + ] + }, + { + "cell_type": "code", + "execution_count": 13, "metadata": {}, "outputs": [], "source": [ - "%%capture ### USE OPTIMIZED NUM CPUS, WORKERS, BEFORE RUNNING\n", + "%%capture\n", "!python project/main.py \\\n", " network=fcnet \\\n", " datamodule=mnist \\\n", " experiment=profiling \\\n", " resources=cpu \\\n", - " trainer.logger.wandb.name=\"CPU training (FcNet/MNIST)\" \\\n", + " hydra.launcher.cpus_per_task=4 \\\n", + " datamodule.num_workers=8 \\\n", + " trainer.logger.wandb.name=\"FcNet/MNIST CPU training\" \\\n", + " trainer.logger.wandb.tags=[\"CPU/GPU comparison\",\"CPU\",\"MNIST\"] " + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", + "!python project/main.py \\\n", + " network=fcnet \\\n", + " datamodule=mnist \\\n", + " experiment=profiling \\\n", + " resources=one_gpu \\\n", + " hydra.launcher.gres='gpu:rtx8000:1' \\\n", + " hydra.launcher.cpus_per_task=4 \\\n", + " datamodule.num_workers=8 \\\n", + " trainer.logger.wandb.name=\"FcNet/MNIST optimized training run\" \\\n", + " trainer.logger.wandb.tags=[\"CPU/GPU comparison\",\"GPU\",\"MNIST\"] " + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [], + "source": [ + "%%capture\n", + "## make sure to run in a local session with the optimized number of CPU cores, num workers\n", + "!python project/main.py \\\n", + " experiment=profiling \\\n", + " algorithm=no_op \\\n", + " datamodule=mnist \\\n", + " datamodule.num_workers=8 \\\n", + " trainer.logger.wandb.name=\"FcNet/MNIST optimized run without training\" \\\n", " trainer.logger.wandb.tags=[\"CPU/GPU comparison\",\"CPU\",\"MNIST\"] " ] }, @@ -361,7 +427,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": null, "metadata": {}, "outputs": [ { @@ -390,9 +456,11 @@ "source": [ "%%capture \n", "!python project/main.py \\\n", - " experiment=profiling_gpu \\\n", + " experiment=profiling \\\n", + " resources=one_gpu \\\n", " hydra.launcher.gres='gpu:a100:1' \\\n", - " datamodule.num_workers= xxx \\\n", + " hydra.launcher.cpus_per_task=4 \\\n", + " datamodule.num_workers=8 \\\n", " trainer.logger.wandb.name=\"A100 training\" \\\n", " trainer.logger.wandb.tags=[\"GPU comparison\"]" ] @@ -405,9 +473,11 @@ "source": [ "%%capture\n", "!python project/main.py \\\n", - " experiment=profiling_gpu \\\n", - " hydra.launcher.gres=gpu:v100:1 \\\n", - " datamodule.num_workers= xxx \\\n", + " experiment=profiling\\\n", + " resources=one_gpu \\\n", + " hydra.launcher.gres='gpu:v100:1' \\\n", + " hydra.launcher.cpus_per_task=4 \\\n", + " datamodule.num_workers=8\\\n", " trainer.logger.wandb.name=\"V100 training\" \\\n", " trainer.logger.wandb.tags=[\"GPU comparison\"]" ] From 2140022de35ee2ee08bc57cd90b1c1a8b738006f Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Tue, 24 Sep 2024 14:25:30 -0400 Subject: [PATCH 23/33] fixed text on dataloading differences --- docs/examples/profiling.ipynb | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 937299b9..fdc322a8 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -85,7 +85,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "In addition to specifying callbacks, the Mila Research template integrates using WandB as a logger, which enables the tracking of additional metrics through visualizations and dashboard creation. Given the flexibility and widespread adoption of the WandB logger, we'll be using it for the remainder of this tutorial, which will then be visualizable at `wandb_url` , as supporting information for the experiments contained herein. We will now proceed to establish a baseline to profile, diagnose and optimize throughout this example." + "In addition to specifying callbacks, the Mila Research template integrates using WandB as a logger, which enables the tracking of additional metrics through visualizations and dashboard creation. Given the flexibility and widespread adoption of the WandB logger, we'll be using it for the remainder of this tutorial, which will then be visualizable in the following [WandB report](https://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/reports/Profiling--Vmlldzo5NDI1MjU0) , as supporting information for the experiments contained herein. We will now proceed to establish a baseline to profile, diagnose and optimize throughout this example." ] }, { @@ -114,8 +114,8 @@ "source": [ "Finding bottlenecks in your code is not necessarily clear or straightforward from the start. A sensible first step is to determine whether potential slowdowns originate from data loading or model computation. Running a model with and without training and contrasting the obtained outputs can help us determine whether the master process has a significant stall when fetching the next batch for training or not. Analyzing the difference between outputs can tell us the following about our model: \n", "\n", - "- If the difference between data loading and training is close to 0, then the data loading procedure outpaces model computation, and computation is the bottleneck. \n", - "- If the difference between data loading and training is much greater than 0, then model computation outpaces data loading, and data loading is the bottleneck. \n", + "- If the difference between data loading and training is close to 0, then the data loading procedure outpaces model computation, thus data loading is the bottleneck.\n", + "- If the difference between data loading and training is much greater than 0, then model computation outpaces data loading, thus computation is the bottleneck. \n", "\n", "We will proceed to run a series of experiments to identify potential bottlenecks: changing the workers involved in the dataloading process and the numbers of cpu assigned per task when training on a GPU." ] @@ -141,11 +141,12 @@ "outputs": [], "source": [ "%%capture\n", + "## make sure to have the one CPU in your local working session for the following run\n", "!python project/main.py -m \\\n", " experiment=profiling \\\n", " algorithm=no_op \\\n", " trainer.logger.wandb.tags=[\"1 CPU Dataloading\",\"Worker throughput\"] \\\n", - " datamodule.num_workers=1,4,8,16,32" + " datamodule.num_workers=1,4,8,16,32" ] }, { @@ -220,7 +221,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Advancements in Graphical Processing Units (GPUs) are widely known to have enabled the deep learning revolution, particularly through faster computation, relative to CPUs. Given that we have the option to run both GPU and CPU workloads, let's compare their throughput. In most workflows, the speedup provided by a GPU is dramatic. For a few select workloads, particularly those with a low number of steps or lighter computation requirements, if a 1.5-2x slower performance is observed when using a CPU, as opposed to a GPU, the former may be worth considering, as they're a far less contested resource on the cluster and pose far fewer availability issues. \n", + "Given that we have the option to run both GPU and CPU workloads, let's compare their throughput. In most workflows, the speedup provided by a GPU is dramatic. For a few select workloads, particularly those with a low number of steps or lighter computation requirements, if a 1.5-2x slower performance is observed when using a CPU, as opposed to a GPU, the former may be worth considering, as they're a far less contested resource on the cluster and pose far fewer availability issues. \n", "\n", "In this section, we'll train a model that's analogous to our ImageNet baseline - entirely on the CPU. We will use the optimal dataloading parameters, as determined by the previous runs. We will also train two smaller fully connected networks on MNIST, a smaller dataset than ImageNet, to compare and contrast the differences in throughput when training with and without a GPU. " ] From b1c227b8c75adb12c714cc65263114043485e6ef Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A9sar=20Miguel=20Valdez=20C=C3=B3rdova?= Date: Wed, 25 Sep 2024 11:59:02 -0400 Subject: [PATCH 24/33] Update docs/examples/profiling.ipynb Co-authored-by: Fabrice Normandin --- docs/examples/profiling.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index fdc322a8..ee270710 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -85,7 +85,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "In addition to specifying callbacks, the Mila Research template integrates using WandB as a logger, which enables the tracking of additional metrics through visualizations and dashboard creation. Given the flexibility and widespread adoption of the WandB logger, we'll be using it for the remainder of this tutorial, which will then be visualizable in the following [WandB report](https://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/reports/Profiling--Vmlldzo5NDI1MjU0) , as supporting information for the experiments contained herein. We will now proceed to establish a baseline to profile, diagnose and optimize throughout this example." + "In addition to callbacks, this template also uses the WandB logger to track metrics and visualize them in nice reports. We'll use it for the remainder of this tutorial. An accompanying wandb report can be viewed [here](https://wandb.ai/cesar-valdez-mcgill-university/ResearchTemplate/reports/Profiling--Vmlldzo5NDI1MjU0) . We will first measure the performance of our current code as a baseline, before any optimizations." ] }, { From 035a4b71373e9c887de3e991fbd4daa3d2006ecc Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A9sar=20Miguel=20Valdez=20C=C3=B3rdova?= Date: Wed, 25 Sep 2024 11:59:21 -0400 Subject: [PATCH 25/33] Update docs/examples/profiling.ipynb Co-authored-by: Fabrice Normandin --- docs/examples/profiling.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index ee270710..f5410296 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -97,7 +97,7 @@ "%%capture\n", "!python project/main.py \\\n", " experiment=profiling \\\n", - " trainer.logger.wandb.name=\"Baseline with training\" \\\n", + " trainer.logger.wandb.name=\"Baseline\" \\\n", " trainer.logger.wandb.tags=[\"Training\",\"Baseline comparison\", \"CPU/GPU comparison\"] " ] }, From c8c40acf37f2a5a42ebb0364ef2b7b14c870dd27 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A9sar=20Miguel=20Valdez=20C=C3=B3rdova?= Date: Wed, 25 Sep 2024 11:59:35 -0400 Subject: [PATCH 26/33] Update docs/examples/profiling.ipynb Co-authored-by: Fabrice Normandin --- docs/examples/profiling.ipynb | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index f5410296..8b3da3cc 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -112,12 +112,12 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Finding bottlenecks in your code is not necessarily clear or straightforward from the start. A sensible first step is to determine whether potential slowdowns originate from data loading or model computation. Running a model with and without training and contrasting the obtained outputs can help us determine whether the master process has a significant stall when fetching the next batch for training or not. Analyzing the difference between outputs can tell us the following about our model: \n", + "The first potential bottleneck to look out for is dataloading. An easy first step is to measure the throughput of your dataloading pipeline without any training. This can easily be done in this template with the `no_op` algorithm, which simply pulls batches from the dataloader, without doing any computation. By Comparing the throughput (in samples/sec) that we get with the no-op algorithm vs our algorithm, we can infer the following: \n", "\n", - "- If the difference between data loading and training is close to 0, then the data loading procedure outpaces model computation, thus data loading is the bottleneck.\n", - "- If the difference between data loading and training is much greater than 0, then model computation outpaces data loading, thus computation is the bottleneck. \n", + "- If the throughput is much higher without training (e.g. >3x faster), then the slowest part of our code is the model computation. This is good.\n", + "- If the difference in throughput (samples per second) between with and without training isn't significant, then data loading is the bottleneck, and we know where to focus our efforts to make the code more better. \n", "\n", - "We will proceed to run a series of experiments to identify potential bottlenecks: changing the workers involved in the dataloading process and the numbers of cpu assigned per task when training on a GPU." + "As can be seen in this figure (TODO: Add figure), we fall in the second case here: our dataloader is our bottleneck. Next, we will demonstrate how to improve the dataloading performance by changing the number of workers and the numbers of cpus." ] }, { From 2a95fddf5610f99887e18f9118b5f0b7c9b432ba Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A9sar=20Miguel=20Valdez=20C=C3=B3rdova?= Date: Wed, 25 Sep 2024 12:00:15 -0400 Subject: [PATCH 27/33] Update project/configs/resources/one_gpu.yaml Co-authored-by: Fabrice Normandin --- project/configs/resources/one_gpu.yaml | 1 - 1 file changed, 1 deletion(-) diff --git a/project/configs/resources/one_gpu.yaml b/project/configs/resources/one_gpu.yaml index 2a7215fd..0d00003a 100644 --- a/project/configs/resources/one_gpu.yaml +++ b/project/configs/resources/one_gpu.yaml @@ -1,5 +1,4 @@ # @package _global_ -# yaml-language-server: $schema=../../../.schemas/resources_one_gpu_schema.json defaults: - override /hydra/launcher: submitit_slurm trainer: From 4f4505a864899cdead5bf8e81bf7e96523f4b476 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A9sar=20Miguel=20Valdez=20C=C3=B3rdova?= Date: Wed, 25 Sep 2024 12:00:34 -0400 Subject: [PATCH 28/33] Update project/configs/trainer/default.yaml Co-authored-by: Fabrice Normandin --- project/configs/trainer/default.yaml | 1 - 1 file changed, 1 deletion(-) diff --git a/project/configs/trainer/default.yaml b/project/configs/trainer/default.yaml index 7558af3b..a9041d09 100644 --- a/project/configs/trainer/default.yaml +++ b/project/configs/trainer/default.yaml @@ -1,4 +1,3 @@ -# yaml-language-server: $schema=../../../.schemas/trainer_default_schema.json _target_: lightning.Trainer logger: null accelerator: auto From 768443a495aa665e793aa2f71a9ec8e284db34de Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A9sar=20Miguel=20Valdez=20C=C3=B3rdova?= Date: Wed, 25 Sep 2024 12:02:09 -0400 Subject: [PATCH 29/33] Update project/main.py Co-authored-by: Fabrice Normandin --- project/main.py | 1 - 1 file changed, 1 deletion(-) diff --git a/project/main.py b/project/main.py index 4ba195a0..567c4232 100644 --- a/project/main.py +++ b/project/main.py @@ -52,7 +52,6 @@ def main(dict_config: DictConfig) -> dict: add_headers=False, ) config: Config = resolve_dictconfig(dict_config) - # assert False, OmegaConf.resolve(dict_config)["datamodule"] experiment: Experiment = setup_experiment(config) metric_name, objective, _metrics = run(experiment) From 2b138f6c550346ac96d62ed316c2aa6c820fa251 Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Wed, 25 Sep 2024 13:19:13 -0400 Subject: [PATCH 30/33] fixed trailing callback_metrics error --- project/main.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/project/main.py b/project/main.py index 567c4232..61e65685 100644 --- a/project/main.py +++ b/project/main.py @@ -103,7 +103,7 @@ def evaluation(experiment: Experiment) -> tuple[str, float | None, dict]: # We want to report the training error. metrics = { **experiment.trainer.logged_metrics, - **experiment.trainer._metrics, + **experiment.trainer.callback_metrics, **experiment.trainer.progress_bar_metrics, } rich.print(metrics) From 7d5c1d2e1c437ed76f58c76a78804bbc8c1cdb8c Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Wed, 25 Sep 2024 13:50:41 -0400 Subject: [PATCH 31/33] changed profiling nb nav from mkdocs.yml to SUMMARY.md --- docs/SUMMARY.md | 1 + mkdocs.yml | 3 --- 2 files changed, 1 insertion(+), 3 deletions(-) diff --git a/docs/SUMMARY.md b/docs/SUMMARY.md index 37616f04..4fd3531d 100644 --- a/docs/SUMMARY.md +++ b/docs/SUMMARY.md @@ -8,6 +8,7 @@ * reference/* * Examples * examples/* + * [Profiling your code](examples/profiling.ipynb) * [Related projects](related.md) * [Getting Help](help.md) * [Contributing](contributing.md) diff --git a/mkdocs.yml b/mkdocs.yml index 9f7e4387..0765ac24 100644 --- a/mkdocs.yml +++ b/mkdocs.yml @@ -3,9 +3,6 @@ site_description: A project template and directory structure for Python data sci site_url: https://mila-iqia.github.io/ResearchTemplate/ repo_url: https://www.github.com/mila-iqia/ResearchTemplate # edit_uri: edit/master/docs -nav: - - Home: index.md - - Profiling your code: docs/examples/profiling.ipynb theme: name: material features: From 290b0d383e9089b45f5c499893adccaf7cf78e3f Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Wed, 25 Sep 2024 15:29:46 -0400 Subject: [PATCH 32/33] grammar --- docs/examples/profiling.ipynb | 34 +++++++++++++++++++++++++++++++++- 1 file changed, 33 insertions(+), 1 deletion(-) diff --git a/docs/examples/profiling.ipynb b/docs/examples/profiling.ipynb index 8b3da3cc..7ea82a68 100644 --- a/docs/examples/profiling.ipynb +++ b/docs/examples/profiling.ipynb @@ -38,6 +38,38 @@ "os.chdir(str(project_root))" ] }, + { + "cell_type": "code", + "execution_count": 17, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "\n", + " \n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from IPython.display import IFrame\n", + "IFrame(\"https://api.wandb.ai/links/cesar-valdez-mcgill-university/8tyi6ckt\",900,500))" + ] + }, { "cell_type": "markdown", "metadata": {}, @@ -115,7 +147,7 @@ "The first potential bottleneck to look out for is dataloading. An easy first step is to measure the throughput of your dataloading pipeline without any training. This can easily be done in this template with the `no_op` algorithm, which simply pulls batches from the dataloader, without doing any computation. By Comparing the throughput (in samples/sec) that we get with the no-op algorithm vs our algorithm, we can infer the following: \n", "\n", "- If the throughput is much higher without training (e.g. >3x faster), then the slowest part of our code is the model computation. This is good.\n", - "- If the difference in throughput (samples per second) between with and without training isn't significant, then data loading is the bottleneck, and we know where to focus our efforts to make the code more better. \n", + "- If the difference in throughput (samples per second) between runs with and without training isn't significant, then data loading is the bottleneck. We know then to focus our efforts on speeding up data loading to make the code run more efficiently. \n", "\n", "As can be seen in this figure (TODO: Add figure), we fall in the second case here: our dataloader is our bottleneck. Next, we will demonstrate how to improve the dataloading performance by changing the number of workers and the numbers of cpus." ] From 32f4e41f0444661648f8f2d2f0948a42c9431c4f Mon Sep 17 00:00:00 2001 From: cmvcordova Date: Thu, 3 Oct 2024 11:52:44 -0400 Subject: [PATCH 33/33] added profiling config --- docs/examples/profiling.md | 1 + project/configs/experiment/profiling.yaml | 13 +++++++++++++ 2 files changed, 14 insertions(+) create mode 100644 docs/examples/profiling.md create mode 100644 project/configs/experiment/profiling.yaml diff --git a/docs/examples/profiling.md b/docs/examples/profiling.md new file mode 100644 index 00000000..5922411c --- /dev/null +++ b/docs/examples/profiling.md @@ -0,0 +1 @@ + diff --git a/project/configs/experiment/profiling.yaml b/project/configs/experiment/profiling.yaml new file mode 100644 index 00000000..de7cbcf8 --- /dev/null +++ b/project/configs/experiment/profiling.yaml @@ -0,0 +1,13 @@ +# @package _global_ + +defaults: + - override /datamodule: imagenet + - override /algorithm: example + - override /trainer/logger: wandb + +trainer: + min_epochs: 1 + max_epochs: 2 + limit_train_batches: 30 + limit_val_batches: 2 + num_sanity_val_steps: 0