AlexNet/train_alexnet.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's train AlexNet from scratch!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/a75wu/miniconda3/envs/dev/lib/python3.10/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "import datasets\n",
    "import torch\n",
    "\n",
    "from datasets import load_dataset\n",
    "from torchvision import transforms\n",
    "from typing import Tuple\n",
    "\n",
    "from trainer import Trainer\n",
    "from utils import get_device, imshow"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using CUDA...\n"
     ]
    }
   ],
   "source": [
    "# setup device\n",
    "device = get_device()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training configurations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "NUM_EPOCHS = 30 # roughly ~90 cycles specified in alexnet paper\n",
    "BATCH_SIZE = 128 # original paper uses 128\n",
    "MOMENTUM = 0.9\n",
    "WEIGHT_DECAY = 0.0005\n",
    "LEARNING_RATE = 0.01\n",
    "NUM_CLASSES = 200 # tiny imagenet"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Dataset\n",
    "\n",
    "We're going to use hugging face to load our dummy dataset (Tiny Imagenet) for training."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train: 100000, test: 10000\n"
     ]
    }
   ],
   "source": [
    "train_data = load_dataset('zh-plus/tiny-imagenet', split='train')\n",
    "test_data = load_dataset('zh-plus/tiny-imagenet', split='valid')\n",
    "print(f'train: {len(train_data)}, test: {len(test_data)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "class TinyImageNetDataset(torch.utils.data.Dataset):\n",
    "\n",
    "    def __init__(self, hf_dataset: datasets.Dataset, transforms: transforms.Compose = None) -> None:\n",
    "        self.dataset = hf_dataset\n",
    "        self.transforms = transforms\n",
    "\n",
    "    def __len__(self) -> int:\n",
    "        return len(self.dataset)\n",
    "    \n",
    "    def __getitem__(self, idx: int) -> Tuple[torch.Tensor, int]:\n",
    "        sample = self.dataset[idx]\n",
    "\n",
    "        if self.transforms:\n",
    "            # some images are apparently greyscale so we need to convert to RGB\n",
    "            sample[\"image\"] = self.transforms(sample[\"image\"].convert(\"RGB\"))\n",
    "\n",
    "        image, label = sample[\"image\"], sample[\"label\"]\n",
    "\n",
    "        return image, label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# data preprocessing\n",
    "\n",
    "train_transform = transforms.Compose([\n",
    "    transforms.Resize(227),\n",
    "    transforms.RandomHorizontalFlip(),\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),\n",
    "])\n",
    "\n",
    "test_transform = transforms.Compose([\n",
    "    transforms.Resize(227),\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),\n",
    "])\n",
    "\n",
    "train_set = TinyImageNetDataset(train_data, train_transform)\n",
    "test_set = TinyImageNetDataset(test_data, test_transform)\n",
    "\n",
    "# split into 90-10 train and validation dataset\n",
    "train_set, val_set = torch.utils.data.random_split(train_set, [0.9, 0.1])\n",
    "\n",
    "# NOTE: apparently adding num_workers is very important for improving training speed lol\n",
    "# https://stackoverflow.com/questions/53998282/how-does-the-number-of-workers-parameter-in-pytorch-dataloader-actually-work\n",
    "train_dataloader = torch.utils.data.DataLoader(train_set, batch_size=BATCH_SIZE, shuffle=True, num_workers=2)\n",
    "val_dataloader = torch.utils.data.DataLoader(val_set, batch_size=BATCH_SIZE, shuffle=True, num_workers=2)\n",
    "test_dataloader = torch.utils.data.DataLoader(test_set, batch_size=BATCH_SIZE, shuffle=False, num_workers=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train: 704\n",
      "validation: 79\n",
      "test: 79\n"
     ]
    }
   ],
   "source": [
    "print(f\"train: {len(train_dataloader)}\\nvalidation: {len(val_dataloader)}\\ntest: {len(test_dataloader)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([128, 3, 227, 227])"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "image, label = next(iter(train_dataloader))\n",
    "image.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Observe Pretrained AlexNet\n",
    "Let's see what the pytorch AlexNet looks like"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using cache found in /home/a75wu/.cache/torch/hub/pytorch_vision_v0.10.0\n",
      "/home/a75wu/miniconda3/envs/dev/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n",
      "  warnings.warn(\n",
      "/home/a75wu/miniconda3/envs/dev/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=None`.\n",
      "  warnings.warn(msg)\n"
     ]
    }
   ],
   "source": [
    "model = torch.hub.load('pytorch/vision:v0.10.0', 'alexnet', pretrained=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AlexNet(\n",
       "  (features): Sequential(\n",
       "    (0): Conv2d(3, 64, kernel_size=(11, 11), stride=(4, 4), padding=(2, 2))\n",
       "    (1): ReLU(inplace=True)\n",
       "    (2): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (3): Conv2d(64, 192, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
       "    (4): ReLU(inplace=True)\n",
       "    (5): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (6): Conv2d(192, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (7): ReLU(inplace=True)\n",
       "    (8): Conv2d(384, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (9): ReLU(inplace=True)\n",
       "    (10): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (11): ReLU(inplace=True)\n",
       "    (12): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "  )\n",
       "  (avgpool): AdaptiveAvgPool2d(output_size=(6, 6))\n",
       "  (classifier): Sequential(\n",
       "    (0): Dropout(p=0.5, inplace=False)\n",
       "    (1): Linear(in_features=9216, out_features=4096, bias=True)\n",
       "    (2): ReLU(inplace=True)\n",
       "    (3): Dropout(p=0.5, inplace=False)\n",
       "    (4): Linear(in_features=4096, out_features=4096, bias=True)\n",
       "    (5): ReLU(inplace=True)\n",
       "    (6): Linear(in_features=4096, out_features=1000, bias=True)\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Some of the implementation details of the current pytorch model is a different than the original AlexNet paper but this github issue goes into some detail about it:\n",
    "\n",
    "https://github.com/pytorch/vision/issues/549\n",
    "\n",
    "NOTE: \n",
    "> \"we added the avg_pool to make the model support images of different sizes. For images of size 224x224, the avg_pool is a no-op.\"\n",
    "\n",
    "**We will try our best to implement the EXACT model mentioned in the paper instead of the PyTorch implementation**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Our AlexNet Implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AlexNet(\n",
       "  (features): Sequential(\n",
       "    (0): Conv2d(3, 96, kernel_size=(11, 11), stride=(4, 4))\n",
       "    (1): ReLU()\n",
       "    (2): LocalResponseNorm(5, alpha=0.001, beta=0.75, k=2)\n",
       "    (3): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (4): Conv2d(96, 256, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
       "    (5): ReLU()\n",
       "    (6): LocalResponseNorm(5, alpha=0.001, beta=0.75, k=2)\n",
       "    (7): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (8): Conv2d(256, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (9): ReLU()\n",
       "    (10): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (11): ReLU()\n",
       "    (12): Conv2d(384, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (13): ReLU()\n",
       "    (14): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "  )\n",
       "  (classifier): Sequential(\n",
       "    (0): Dropout(p=0.5, inplace=False)\n",
       "    (1): Linear(in_features=9216, out_features=4096, bias=True)\n",
       "    (2): ReLU()\n",
       "    (3): Dropout(p=0.5, inplace=False)\n",
       "    (4): Linear(in_features=4096, out_features=4096, bias=True)\n",
       "    (5): ReLU()\n",
       "    (6): Linear(in_features=4096, out_features=1000, bias=True)\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from alexnet import AlexNet\n",
    "model = AlexNet()\n",
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of parameters: 62378344\n",
      "Number of trainable parameters: 62378344\n"
     ]
    }
   ],
   "source": [
    "n_params = sum(p.numel() for p in model.parameters())\n",
    "n_trainable_params = sum(p.numel() for p in model.parameters() if p.requires_grad)\n",
    "\n",
    "print(f\"Number of parameters: {n_params}\")\n",
    "print(f\"Number of trainable parameters: {n_trainable_params}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:37<00:00,  7.19it/s, loss=5.37]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 1] LOSS : train=5.6569632718508895 val=5.361310868323604 | ACCURACY (Top-1) : train=0.00456099072471261 val=0.004647943191230297 | TOP-5 : train=0.024158824573863636 val=0.023635284810126583\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:39<00:00,  7.11it/s, loss=5.34]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 2] LOSS : train=5.356095892461863 val=5.3487432878228685 | ACCURACY (Top-1) : train=0.004827326163649559 val=0.005735759623348713 | TOP-5 : train=0.02434747869318182 val=0.025118670886075948\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:32<00:00,  7.62it/s, loss=5.13]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 3] LOSS : train=5.257141777737574 val=5.10165585747248 | ACCURACY (Top-1) : train=0.007956765592098236 val=0.01275712065398693 | TOP-5 : train=0.038507634943181816 val=0.05844541139240506\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:42<00:00,  6.84it/s, loss=4.92]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 4] LOSS : train=5.018083686178381 val=4.863555654694762 | ACCURACY (Top-1) : train=0.02147327922284603 val=0.0390625 | TOP-5 : train=0.08804598721590909 val=0.12895569620253164\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:36<00:00,  7.30it/s, loss=4.45]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 5] LOSS : train=4.66817624934695 val=4.410591222062895 | ACCURACY (Top-1) : train=0.049893468618392944 val=0.06784018874168396 | TOP-5 : train=0.1739501953125 val=0.2331882911392405\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:39<00:00,  7.09it/s, loss=4]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 6] LOSS : train=4.199221126057885 val=4.047299366963061 | ACCURACY (Top-1) : train=0.1011962890625 val=0.12776899337768555 | TOP-5 : train=0.29114879261363635 val=0.32743275316455694\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:36<00:00,  7.26it/s, loss=3.66]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 7] LOSS : train=3.85190109197389 val=3.7797603788255136 | ACCURACY (Top-1) : train=0.15103426575660706 val=0.1643591821193695 | TOP-5 : train=0.38041548295454547 val=0.39408623417721517\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:38<00:00,  7.13it/s, loss=3.36]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 8] LOSS : train=3.417269053784284 val=3.3564166177677204 | ACCURACY (Top-1) : train=0.22166860103607178 val=0.2350672483444214 | TOP-5 : train=0.48197798295454547 val=0.4933742088607595\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:41<00:00,  6.91it/s, loss=3.4]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 9] LOSS : train=3.325157389722087 val=3.313815829120105 | ACCURACY (Top-1) : train=0.23643910884857178 val=0.2431764304637909 | TOP-5 : train=0.5035844282670454 val=0.5069224683544303\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:32<00:00,  7.58it/s, loss=3.26]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 10] LOSS : train=3.265972266820344 val=3.271509484399723 | ACCURACY (Top-1) : train=0.24708141386508942 val=0.25019779801368713 | TOP-5 : train=0.5155140269886364 val=0.514932753164557\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:38<00:00,  7.12it/s, loss=3.22]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 11] LOSS : train=3.2150866091251373 val=3.2310529479497596 | ACCURACY (Top-1) : train=0.2565252184867859 val=0.25632911920547485 | TOP-5 : train=0.5268332741477273 val=0.5234375\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:45<00:00,  6.69it/s, loss=3.14]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 12] LOSS : train=3.1671993864530865 val=3.1724180843256695 | ACCURACY (Top-1) : train=0.26481491327285767 val=0.26908624172210693 | TOP-5 : train=0.5379416725852273 val=0.5352056962025317\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:37<00:00,  7.25it/s, loss=3.06]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 13] LOSS : train=3.121407210826874 val=3.122367267367206 | ACCURACY (Top-1) : train=0.27268290519714355 val=0.2761076092720032 | TOP-5 : train=0.5489723899147727 val=0.5446993670886076\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:41<00:00,  6.92it/s, loss=3.07]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 14] LOSS : train=3.0817639726129444 val=3.138366729398317 | ACCURACY (Top-1) : train=0.28103914856910706 val=0.2761076092720032 | TOP-5 : train=0.5564963600852273 val=0.5436115506329114\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:37<00:00,  7.20it/s, loss=3.02]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 15] LOSS : train=2.980340000241995 val=3.0474748309654527 | ACCURACY (Top-1) : train=0.2997491955757141 val=0.2939082384109497 | TOP-5 : train=0.5786354758522727 val=0.564873417721519\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:34<00:00,  7.42it/s, loss=3.05]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 16] LOSS : train=2.967280241914771 val=3.043632377552081 | ACCURACY (Top-1) : train=0.3021795153617859 val=0.2920292913913727 | TOP-5 : train=0.5815984552556818 val=0.5652689873417721\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:40<00:00,  7.01it/s, loss=2.92]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 17] LOSS : train=2.9573537788607855 val=3.025141785416422 | ACCURACY (Top-1) : train=0.3042657971382141 val=0.2967761158943176 | TOP-5 : train=0.5843505859375 val=0.5682357594936709\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:36<00:00,  7.30it/s, loss=2.95]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 18] LOSS : train=2.9505811773917894 val=3.0305239701572853 | ACCURACY (Top-1) : train=0.30465421080589294 val=0.29282042384147644 | TOP-5 : train=0.5847722833806818 val=0.5676424050632911\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:37<00:00,  7.19it/s, loss=2.9]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 19] LOSS : train=2.9441729692572896 val=3.028598344778713 | ACCURACY (Top-1) : train=0.30630770325660706 val=0.29568830132484436 | TOP-5 : train=0.5867919921875 val=0.5660601265822784\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:38<00:00,  7.17it/s, loss=2.93]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 20] LOSS : train=2.935957385057753 val=3.014673797390129 | ACCURACY (Top-1) : train=0.30728426575660706 val=0.2962816655635834 | TOP-5 : train=0.5887895063920454 val=0.5695213607594937\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:38<00:00,  7.15it/s, loss=2.88]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 21] LOSS : train=2.9312188113954933 val=3.024732378464711 | ACCURACY (Top-1) : train=0.3079279065132141 val=0.2977650463581085 | TOP-5 : train=0.5887451171875 val=0.5703125\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:39<00:00,  7.06it/s, loss=2.88]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 22] LOSS : train=2.9175763164054263 val=3.0081283352043053 | ACCURACY (Top-1) : train=0.31034713983535767 val=0.30251187086105347 | TOP-5 : train=0.5926513671875 val=0.572685917721519\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:39<00:00,  7.09it/s, loss=2.94]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 23] LOSS : train=2.915942318737507 val=3.0088868654226957 | ACCURACY (Top-1) : train=0.3112127184867859 val=0.2984572947025299 | TOP-5 : train=0.5915860262784091 val=0.5746637658227848\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:37<00:00,  7.20it/s, loss=2.89]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 24] LOSS : train=2.91580056399107 val=3.0050605792033522 | ACCURACY (Top-1) : train=0.31081321835517883 val=0.29934731125831604 | TOP-5 : train=0.5929953835227273 val=0.5714003164556962\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:35<00:00,  7.36it/s, loss=2.91]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 25] LOSS : train=2.913320042531599 val=2.999567249153234 | ACCURACY (Top-1) : train=0.31143465638160706 val=0.3040941655635834 | TOP-5 : train=0.5926180752840909 val=0.5738726265822784\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:40<00:00,  6.98it/s, loss=3]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 26] LOSS : train=2.912034350362691 val=3.000708848615236 | ACCURACY (Top-1) : train=0.3120006322860718 val=0.2997428774833679 | TOP-5 : train=0.5922296697443182 val=0.571004746835443\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:36<00:00,  7.32it/s, loss=2.83]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 27] LOSS : train=2.9125874746929514 val=3.015918722635583 | ACCURACY (Top-1) : train=0.3126331865787506 val=0.297468364238739 | TOP-5 : train=0.5923517400568182 val=0.5695213607594937\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:38<00:00,  7.13it/s, loss=2.9]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 28] LOSS : train=2.912517122585665 val=2.9981011318255075 | ACCURACY (Top-1) : train=0.3111572265625 val=0.30191850662231445 | TOP-5 : train=0.5936279296875 val=0.575751582278481\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 704/704 [01:39<00:00,  7.09it/s, loss=2.84]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | [EPOCH 29] LOSS : train=2.910572686317292 val=3.000654649130906 | ACCURACY (Top-1) : train=0.3120672106742859 val=0.30290743708610535 | TOP-5 : train=0.5935169566761364 val=0.5732792721518988\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 68%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▋                                                          | 477/704 [01:07<00:31,  7.11it/s, loss=2.83]\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[13], line 8\u001b[0m\n\u001b[1;32m      1\u001b[0m alexnet_trainer \u001b[38;5;241m=\u001b[39m Trainer(model,\n\u001b[1;32m      2\u001b[0m                           batch_size\u001b[38;5;241m=\u001b[39mBATCH_SIZE,\n\u001b[1;32m      3\u001b[0m                           learning_rate\u001b[38;5;241m=\u001b[39mLEARNING_RATE,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m      6\u001b[0m                           num_epochs\u001b[38;5;241m=\u001b[39mNUM_EPOCHS,\n\u001b[1;32m      7\u001b[0m                           device\u001b[38;5;241m=\u001b[39mdevice)\n\u001b[0;32m----> 8\u001b[0m \u001b[43malexnet_trainer\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtrain\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtrain_dataloader\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mval_dataloader\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m/fsys1/home/a75wu/StuffFromScratch/AlexNet/trainer.py:86\u001b[0m, in \u001b[0;36mTrainer.train\u001b[0;34m(self, train_dataloader, val_dataloader)\u001b[0m\n\u001b[1;32m     82\u001b[0m running_top_k_acc \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0.0\u001b[39m\n\u001b[1;32m     84\u001b[0m pbar \u001b[38;5;241m=\u001b[39m tqdm(\u001b[38;5;28menumerate\u001b[39m(train_dataloader), total\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mlen\u001b[39m(train_dataloader))\n\u001b[0;32m---> 86\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m i, (inputs, labels) \u001b[38;5;129;01min\u001b[39;00m pbar:\n\u001b[1;32m     87\u001b[0m     inputs, labels \u001b[38;5;241m=\u001b[39m inputs\u001b[38;5;241m.\u001b[39mto(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdevice), labels\u001b[38;5;241m.\u001b[39mto(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdevice)\n\u001b[1;32m     89\u001b[0m     \u001b[38;5;66;03m# zero gradients for every batch\u001b[39;00m\n",
      "File \u001b[0;32m~/miniconda3/envs/dev/lib/python3.10/site-packages/tqdm/std.py:1181\u001b[0m, in \u001b[0;36mtqdm.__iter__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m   1178\u001b[0m time \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_time\n\u001b[1;32m   1180\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m-> 1181\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m obj \u001b[38;5;129;01min\u001b[39;00m iterable:\n\u001b[1;32m   1182\u001b[0m         \u001b[38;5;28;01myield\u001b[39;00m obj\n\u001b[1;32m   1183\u001b[0m         \u001b[38;5;66;03m# Update and possibly print the progressbar.\u001b[39;00m\n\u001b[1;32m   1184\u001b[0m         \u001b[38;5;66;03m# Note: does not call self.update(1) for speed optimisation.\u001b[39;00m\n",
      "File \u001b[0;32m~/miniconda3/envs/dev/lib/python3.10/site-packages/torch/utils/data/dataloader.py:630\u001b[0m, in \u001b[0;36m_BaseDataLoaderIter.__next__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    627\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_sampler_iter \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m    628\u001b[0m     \u001b[38;5;66;03m# TODO(https://github.com/pytorch/pytorch/issues/76750)\u001b[39;00m\n\u001b[1;32m    629\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_reset()  \u001b[38;5;66;03m# type: ignore[call-arg]\u001b[39;00m\n\u001b[0;32m--> 630\u001b[0m data \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_next_data\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    631\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_num_yielded \u001b[38;5;241m+\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1\u001b[39m\n\u001b[1;32m    632\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_dataset_kind \u001b[38;5;241m==\u001b[39m _DatasetKind\u001b[38;5;241m.\u001b[39mIterable \u001b[38;5;129;01mand\u001b[39;00m \\\n\u001b[1;32m    633\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_IterableDataset_len_called \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m \\\n\u001b[1;32m    634\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_num_yielded \u001b[38;5;241m>\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_IterableDataset_len_called:\n",
      "File \u001b[0;32m~/miniconda3/envs/dev/lib/python3.10/site-packages/torch/utils/data/dataloader.py:1327\u001b[0m, in \u001b[0;36m_MultiProcessingDataLoaderIter._next_data\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m   1324\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_process_data(data)\n\u001b[1;32m   1326\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_shutdown \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_tasks_outstanding \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m0\u001b[39m\n\u001b[0;32m-> 1327\u001b[0m idx, data \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_get_data\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1328\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_tasks_outstanding \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1\u001b[39m\n\u001b[1;32m   1329\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_dataset_kind \u001b[38;5;241m==\u001b[39m _DatasetKind\u001b[38;5;241m.\u001b[39mIterable:\n\u001b[1;32m   1330\u001b[0m     \u001b[38;5;66;03m# Check for _IterableDatasetStopIteration\u001b[39;00m\n",
      "File \u001b[0;32m~/miniconda3/envs/dev/lib/python3.10/site-packages/torch/utils/data/dataloader.py:1293\u001b[0m, in \u001b[0;36m_MultiProcessingDataLoaderIter._get_data\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m   1289\u001b[0m     \u001b[38;5;66;03m# In this case, `self._data_queue` is a `queue.Queue`,. But we don't\u001b[39;00m\n\u001b[1;32m   1290\u001b[0m     \u001b[38;5;66;03m# need to call `.task_done()` because we don't use `.join()`.\u001b[39;00m\n\u001b[1;32m   1291\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m   1292\u001b[0m     \u001b[38;5;28;01mwhile\u001b[39;00m \u001b[38;5;28;01mTrue\u001b[39;00m:\n\u001b[0;32m-> 1293\u001b[0m         success, data \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_try_get_data\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1294\u001b[0m         \u001b[38;5;28;01mif\u001b[39;00m success:\n\u001b[1;32m   1295\u001b[0m             \u001b[38;5;28;01mreturn\u001b[39;00m data\n",
      "File \u001b[0;32m~/miniconda3/envs/dev/lib/python3.10/site-packages/torch/utils/data/dataloader.py:1131\u001b[0m, in \u001b[0;36m_MultiProcessingDataLoaderIter._try_get_data\u001b[0;34m(self, timeout)\u001b[0m\n\u001b[1;32m   1118\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_try_get_data\u001b[39m(\u001b[38;5;28mself\u001b[39m, timeout\u001b[38;5;241m=\u001b[39m_utils\u001b[38;5;241m.\u001b[39mMP_STATUS_CHECK_INTERVAL):\n\u001b[1;32m   1119\u001b[0m     \u001b[38;5;66;03m# Tries to fetch data from `self._data_queue` once for a given timeout.\u001b[39;00m\n\u001b[1;32m   1120\u001b[0m     \u001b[38;5;66;03m# This can also be used as inner loop of fetching without timeout, with\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   1128\u001b[0m     \u001b[38;5;66;03m# Returns a 2-tuple:\u001b[39;00m\n\u001b[1;32m   1129\u001b[0m     \u001b[38;5;66;03m#   (bool: whether successfully get data, any: data if successful else None)\u001b[39;00m\n\u001b[1;32m   1130\u001b[0m     \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m-> 1131\u001b[0m         data \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_data_queue\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtimeout\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtimeout\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1132\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m (\u001b[38;5;28;01mTrue\u001b[39;00m, data)\n\u001b[1;32m   1133\u001b[0m     \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m   1134\u001b[0m         \u001b[38;5;66;03m# At timeout and error, we manually check whether any worker has\u001b[39;00m\n\u001b[1;32m   1135\u001b[0m         \u001b[38;5;66;03m# failed. Note that this is the only mechanism for Windows to detect\u001b[39;00m\n\u001b[1;32m   1136\u001b[0m         \u001b[38;5;66;03m# worker failures.\u001b[39;00m\n",
      "File \u001b[0;32m~/miniconda3/envs/dev/lib/python3.10/multiprocessing/queues.py:113\u001b[0m, in \u001b[0;36mQueue.get\u001b[0;34m(self, block, timeout)\u001b[0m\n\u001b[1;32m    111\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m block:\n\u001b[1;32m    112\u001b[0m     timeout \u001b[38;5;241m=\u001b[39m deadline \u001b[38;5;241m-\u001b[39m time\u001b[38;5;241m.\u001b[39mmonotonic()\n\u001b[0;32m--> 113\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_poll\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtimeout\u001b[49m\u001b[43m)\u001b[49m:\n\u001b[1;32m    114\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m Empty\n\u001b[1;32m    115\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_poll():\n",
      "File \u001b[0;32m~/miniconda3/envs/dev/lib/python3.10/multiprocessing/connection.py:257\u001b[0m, in \u001b[0;36m_ConnectionBase.poll\u001b[0;34m(self, timeout)\u001b[0m\n\u001b[1;32m    255\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_check_closed()\n\u001b[1;32m    256\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_check_readable()\n\u001b[0;32m--> 257\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_poll\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtimeout\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/envs/dev/lib/python3.10/multiprocessing/connection.py:424\u001b[0m, in \u001b[0;36mConnection._poll\u001b[0;34m(self, timeout)\u001b[0m\n\u001b[1;32m    423\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_poll\u001b[39m(\u001b[38;5;28mself\u001b[39m, timeout):\n\u001b[0;32m--> 424\u001b[0m     r \u001b[38;5;241m=\u001b[39m \u001b[43mwait\u001b[49m\u001b[43m(\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtimeout\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    425\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mbool\u001b[39m(r)\n",
      "File \u001b[0;32m~/miniconda3/envs/dev/lib/python3.10/multiprocessing/connection.py:931\u001b[0m, in \u001b[0;36mwait\u001b[0;34m(object_list, timeout)\u001b[0m\n\u001b[1;32m    928\u001b[0m     deadline \u001b[38;5;241m=\u001b[39m time\u001b[38;5;241m.\u001b[39mmonotonic() \u001b[38;5;241m+\u001b[39m timeout\n\u001b[1;32m    930\u001b[0m \u001b[38;5;28;01mwhile\u001b[39;00m \u001b[38;5;28;01mTrue\u001b[39;00m:\n\u001b[0;32m--> 931\u001b[0m     ready \u001b[38;5;241m=\u001b[39m \u001b[43mselector\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mselect\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtimeout\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    932\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m ready:\n\u001b[1;32m    933\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m [key\u001b[38;5;241m.\u001b[39mfileobj \u001b[38;5;28;01mfor\u001b[39;00m (key, events) \u001b[38;5;129;01min\u001b[39;00m ready]\n",
      "File \u001b[0;32m~/miniconda3/envs/dev/lib/python3.10/selectors.py:416\u001b[0m, in \u001b[0;36m_PollLikeSelector.select\u001b[0;34m(self, timeout)\u001b[0m\n\u001b[1;32m    414\u001b[0m ready \u001b[38;5;241m=\u001b[39m []\n\u001b[1;32m    415\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 416\u001b[0m     fd_event_list \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_selector\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpoll\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtimeout\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    417\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mInterruptedError\u001b[39;00m:\n\u001b[1;32m    418\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m ready\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "alexnet_trainer = Trainer(model,\n",
    "                          batch_size=BATCH_SIZE,\n",
    "                          learning_rate=LEARNING_RATE,\n",
    "                          weight_decay=WEIGHT_DECAY,\n",
    "                          momentum=MOMENTUM,\n",
    "                          num_epochs=NUM_EPOCHS,\n",
    "                          device=device)\n",
    "alexnet_trainer.train(train_dataloader, val_dataloader)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Performance Metrics\n",
    "Let's implement Top-1 (accuracy) and Top-5 Error Rate Metrics\n",
    "\n",
    "https://pytorch.org/ignite/generated/ignite.metrics.TopKCategoricalAccuracy.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "top-k indices:\n",
      "tensor([[0, 1],\n",
      "        [2, 1],\n",
      "        [1, 0],\n",
      "        [0, 2]])\n",
      "\n",
      "ground truths:\n",
      "tensor([1, 0, 0, 2])\n",
      "tensor([[1, 1],\n",
      "        [0, 0],\n",
      "        [0, 0],\n",
      "        [2, 2]])\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.75"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# test top-k accuracy metric\n",
    "outputs = torch.tensor([\n",
    "    [0.7, 0.2, 0.05, 0.05],     # 1 is in the top 2\n",
    "    [0.2, 0.3, 0.4, 0.1],       # 0 is not in the top 2\n",
    "    [0.4, 0.4, 0.1, 0.1],       # 0 is in the top 2\n",
    "    [0.7, 0.05, 0.2, 0.05]      # 2 is in the top 2\n",
    "])\n",
    "labels = torch.tensor([1, 0, 0, 2])\n",
    "\n",
    "k = 2\n",
    "values, indices = torch.topk(outputs, k)\n",
    "print(f\"top-k indices:\\n{indices}\\n\")\n",
    "\n",
    "print(f\"ground truths:\\n{labels}\")\n",
    "print(labels.view(-1, 1).expand_as(indices))\n",
    "topk_correct = indices.eq(labels.view(-1, 1).expand_as(indices))\n",
    "accuracy = topk_correct.sum().item() / labels.size(0)\n",
    "\n",
    "accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | Test accuracy: 31.754350662231445 % | Top-5 accuracy: 58.7618670886076\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "alexnet_trainer.test(test_dataloader)\n",
    "alexnet_trainer.plot_metrics()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "dev",
   "language": "python",
   "name": "dev"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}