notebooks/preprocessing/significant_perts_edist.ipynb

{
 "nbformat": 4,
 "nbformat_minor": 0,
 "metadata": {
  "colab": {
   "provenance": []
  },
  "kernelspec": {
   "name": "python3",
   "display_name": "Python 3 (ipykernel)",
   "language": "python"
  },
  "language_info": {
   "name": "python"
  }
 },
 "cells": [
  {
   "metadata": {},
   "cell_type": "code",
   "outputs": [],
   "execution_count": null,
   "source": "# Author: Will Connell, A. Wenteler"
  },
  {
   "cell_type": "code",
   "source": [
    "!pip install scanpy\n",
    "!pip install scperturb\n",
    "!pip install scikit-misc"
   ],
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "HJ8YJlya5MFa",
    "outputId": "4e4ba291-3014-4cca-b609-e22678e2eedc",
    "ExecuteTime": {
     "end_time": "2024-07-18T10:16:20.093442Z",
     "start_time": "2024-07-18T10:16:14.668469Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: scanpy in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (1.10.0)\r\n",
      "Requirement already satisfied: anndata>=0.8 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (0.10.6)\r\n",
      "Requirement already satisfied: h5py>=3.1 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (3.10.0)\r\n",
      "Requirement already satisfied: joblib in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (1.3.2)\r\n",
      "Requirement already satisfied: legacy-api-wrap>=1.4 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (1.4)\r\n",
      "Requirement already satisfied: matplotlib>=3.6 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (3.8.3)\r\n",
      "Requirement already satisfied: natsort in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (8.4.0)\r\n",
      "Requirement already satisfied: networkx>=2.7 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (3.2.1)\r\n",
      "Requirement already satisfied: numba>=0.56 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (0.59.1)\r\n",
      "Requirement already satisfied: numpy>=1.23 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (1.26.4)\r\n",
      "Requirement already satisfied: packaging>=21.3 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (24.0)\r\n",
      "Requirement already satisfied: pandas>=1.5 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (2.2.1)\r\n",
      "Requirement already satisfied: patsy in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (0.5.6)\r\n",
      "Requirement already satisfied: pynndescent>=0.5 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (0.5.12)\r\n",
      "Requirement already satisfied: scikit-learn>=0.24 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (1.4.1.post1)\r\n",
      "Requirement already satisfied: scipy>=1.8 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (1.12.0)\r\n",
      "Requirement already satisfied: seaborn>=0.13 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (0.13.2)\r\n",
      "Requirement already satisfied: session-info in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (1.0.0)\r\n",
      "Requirement already satisfied: statsmodels>=0.13 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (0.14.1)\r\n",
      "Requirement already satisfied: tqdm in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (4.66.2)\r\n",
      "Requirement already satisfied: umap-learn!=0.5.0,>=0.5 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy) (0.5.5)\r\n",
      "Requirement already satisfied: array-api-compat!=1.5,>1.4 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from anndata>=0.8->scanpy) (1.6)\r\n",
      "Requirement already satisfied: exceptiongroup in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from anndata>=0.8->scanpy) (1.2.0)\r\n",
      "Requirement already satisfied: contourpy>=1.0.1 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from matplotlib>=3.6->scanpy) (1.2.0)\r\n",
      "Requirement already satisfied: cycler>=0.10 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from matplotlib>=3.6->scanpy) (0.12.1)\r\n",
      "Requirement already satisfied: fonttools>=4.22.0 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from matplotlib>=3.6->scanpy) (4.50.0)\r\n",
      "Requirement already satisfied: kiwisolver>=1.3.1 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from matplotlib>=3.6->scanpy) (1.4.5)\r\n",
      "Requirement already satisfied: pillow>=8 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from matplotlib>=3.6->scanpy) (10.2.0)\r\n",
      "Requirement already satisfied: pyparsing>=2.3.1 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from matplotlib>=3.6->scanpy) (3.1.2)\r\n",
      "Requirement already satisfied: python-dateutil>=2.7 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from matplotlib>=3.6->scanpy) (2.9.0.post0)\r\n",
      "Requirement already satisfied: llvmlite<0.43,>=0.42.0dev0 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from numba>=0.56->scanpy) (0.42.0)\r\n",
      "Requirement already satisfied: pytz>=2020.1 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from pandas>=1.5->scanpy) (2024.1)\r\n",
      "Requirement already satisfied: tzdata>=2022.7 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from pandas>=1.5->scanpy) (2024.1)\r\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scikit-learn>=0.24->scanpy) (3.4.0)\r\n",
      "Requirement already satisfied: six in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from patsy->scanpy) (1.16.0)\r\n",
      "Requirement already satisfied: stdlib-list in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from session-info->scanpy) (0.10.0)\r\n",
      "\r\n",
      "\u001B[1m[\u001B[0m\u001B[34;49mnotice\u001B[0m\u001B[1;39;49m]\u001B[0m\u001B[39;49m A new release of pip is available: \u001B[0m\u001B[31;49m23.2.1\u001B[0m\u001B[39;49m -> \u001B[0m\u001B[32;49m24.1.2\u001B[0m\r\n",
      "\u001B[1m[\u001B[0m\u001B[34;49mnotice\u001B[0m\u001B[1;39;49m]\u001B[0m\u001B[39;49m To update, run: \u001B[0m\u001B[32;49mpip install --upgrade pip\u001B[0m\r\n",
      "Collecting scperturb\r\n",
      "  Obtaining dependency information for scperturb from https://files.pythonhosted.org/packages/0b/96/1f1cdce4b007c24cb7171cf368ced8f1e6723f7132bf0e3afb43cee5858d/scperturb-0.1.0-py3-none-any.whl.metadata\r\n",
      "  Downloading scperturb-0.1.0-py3-none-any.whl.metadata (1.5 kB)\r\n",
      "Requirement already satisfied: pandas>=1.1 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scperturb) (2.2.1)\r\n",
      "Requirement already satisfied: scanpy>=1.7 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scperturb) (1.10.0)\r\n",
      "Requirement already satisfied: tqdm>=4.62 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scperturb) (4.66.2)\r\n",
      "Requirement already satisfied: numpy in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scperturb) (1.26.4)\r\n",
      "Requirement already satisfied: scipy in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scperturb) (1.12.0)\r\n",
      "Requirement already satisfied: scikit-learn in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scperturb) (1.4.1.post1)\r\n",
      "Requirement already satisfied: statsmodels in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scperturb) (0.14.1)\r\n",
      "Requirement already satisfied: python-dateutil>=2.8.2 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from pandas>=1.1->scperturb) (2.9.0.post0)\r\n",
      "Requirement already satisfied: pytz>=2020.1 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from pandas>=1.1->scperturb) (2024.1)\r\n",
      "Requirement already satisfied: tzdata>=2022.7 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from pandas>=1.1->scperturb) (2024.1)\r\n",
      "Requirement already satisfied: anndata>=0.8 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (0.10.6)\r\n",
      "Requirement already satisfied: h5py>=3.1 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (3.10.0)\r\n",
      "Requirement already satisfied: joblib in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (1.3.2)\r\n",
      "Requirement already satisfied: legacy-api-wrap>=1.4 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (1.4)\r\n",
      "Requirement already satisfied: matplotlib>=3.6 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (3.8.3)\r\n",
      "Requirement already satisfied: natsort in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (8.4.0)\r\n",
      "Requirement already satisfied: networkx>=2.7 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (3.2.1)\r\n",
      "Requirement already satisfied: numba>=0.56 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (0.59.1)\r\n",
      "Requirement already satisfied: packaging>=21.3 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (24.0)\r\n",
      "Requirement already satisfied: patsy in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (0.5.6)\r\n",
      "Requirement already satisfied: pynndescent>=0.5 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (0.5.12)\r\n",
      "Requirement already satisfied: seaborn>=0.13 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (0.13.2)\r\n",
      "Requirement already satisfied: session-info in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (1.0.0)\r\n",
      "Requirement already satisfied: umap-learn!=0.5.0,>=0.5 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scanpy>=1.7->scperturb) (0.5.5)\r\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scikit-learn->scperturb) (3.4.0)\r\n",
      "Requirement already satisfied: array-api-compat!=1.5,>1.4 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from anndata>=0.8->scanpy>=1.7->scperturb) (1.6)\r\n",
      "Requirement already satisfied: exceptiongroup in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from anndata>=0.8->scanpy>=1.7->scperturb) (1.2.0)\r\n",
      "Requirement already satisfied: contourpy>=1.0.1 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from matplotlib>=3.6->scanpy>=1.7->scperturb) (1.2.0)\r\n",
      "Requirement already satisfied: cycler>=0.10 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from matplotlib>=3.6->scanpy>=1.7->scperturb) (0.12.1)\r\n",
      "Requirement already satisfied: fonttools>=4.22.0 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from matplotlib>=3.6->scanpy>=1.7->scperturb) (4.50.0)\r\n",
      "Requirement already satisfied: kiwisolver>=1.3.1 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from matplotlib>=3.6->scanpy>=1.7->scperturb) (1.4.5)\r\n",
      "Requirement already satisfied: pillow>=8 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from matplotlib>=3.6->scanpy>=1.7->scperturb) (10.2.0)\r\n",
      "Requirement already satisfied: pyparsing>=2.3.1 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from matplotlib>=3.6->scanpy>=1.7->scperturb) (3.1.2)\r\n",
      "Requirement already satisfied: llvmlite<0.43,>=0.42.0dev0 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from numba>=0.56->scanpy>=1.7->scperturb) (0.42.0)\r\n",
      "Requirement already satisfied: six in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from patsy->scanpy>=1.7->scperturb) (1.16.0)\r\n",
      "Requirement already satisfied: stdlib-list in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from session-info->scanpy>=1.7->scperturb) (0.10.0)\r\n",
      "Downloading scperturb-0.1.0-py3-none-any.whl (8.5 kB)\r\n",
      "Installing collected packages: scperturb\r\n",
      "Successfully installed scperturb-0.1.0\r\n",
      "\r\n",
      "\u001B[1m[\u001B[0m\u001B[34;49mnotice\u001B[0m\u001B[1;39;49m]\u001B[0m\u001B[39;49m A new release of pip is available: \u001B[0m\u001B[31;49m23.2.1\u001B[0m\u001B[39;49m -> \u001B[0m\u001B[32;49m24.1.2\u001B[0m\r\n",
      "\u001B[1m[\u001B[0m\u001B[34;49mnotice\u001B[0m\u001B[1;39;49m]\u001B[0m\u001B[39;49m To update, run: \u001B[0m\u001B[32;49mpip install --upgrade pip\u001B[0m\r\n",
      "Requirement already satisfied: scikit-misc in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (0.3.1)\r\n",
      "Requirement already satisfied: numpy>=1.22.3 in /Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages (from scikit-misc) (1.26.4)\r\n",
      "\r\n",
      "\u001B[1m[\u001B[0m\u001B[34;49mnotice\u001B[0m\u001B[1;39;49m]\u001B[0m\u001B[39;49m A new release of pip is available: \u001B[0m\u001B[31;49m23.2.1\u001B[0m\u001B[39;49m -> \u001B[0m\u001B[32;49m24.1.2\u001B[0m\r\n",
      "\u001B[1m[\u001B[0m\u001B[34;49mnotice\u001B[0m\u001B[1;39;49m]\u001B[0m\u001B[39;49m To update, run: \u001B[0m\u001B[32;49mpip install --upgrade pip\u001B[0m\r\n"
     ]
    }
   ],
   "execution_count": 1
  },
  {
   "cell_type": "code",
   "source": [
    "# datastes here: http://projects.sanderlab.org/scperturb/datavzrd/scPerturb_vzrd_v1/dataset_info/index_1.html\n",
    "\n",
    "!wget https://zenodo.org/record/7041849/files/NormanWeissman2019_filtered.h5ad"
   ],
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cjOp3K2h5XzM",
    "outputId": "a89b6756-59c9-4a33-8dba-adb57c3812ec",
    "ExecuteTime": {
     "end_time": "2024-07-18T10:20:42.901716Z",
     "start_time": "2024-07-18T10:20:42.784662Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "zsh:1: command not found: wget\r\n"
     ]
    }
   ],
   "execution_count": 1
  },
  {
   "cell_type": "code",
   "metadata": {
    "id": "V73VgLBL48ZT",
    "ExecuteTime": {
     "end_time": "2024-09-23T23:20:57.052793Z",
     "start_time": "2024-09-23T23:20:57.033957Z"
    }
   },
   "source": [
    "import scanpy as sc\n",
    "import anndata as ad\n",
    "from scipy.sparse import csr_matrix\n",
    "\n",
    "from scperturb import *\n",
    "\n",
    "import scanpy as sc\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import mpl_sizes\n",
    "\n",
    "formatter = mpl_sizes.get_format(\"NeurIPS\") # options: ICLR, ICML, NeurIPS, InfThesis"
   ],
   "outputs": [],
   "execution_count": 39
  },
  {
   "cell_type": "code",
   "source": [
    "'''\n",
    "See here: https://github.com/sanderlab/scPerturb/blob/f64821e5e621b3f6147b47f318ae1775d4eff324/revision/snakemake/Snakefile#L177-L215\n",
    "'''\n",
    "\n",
    "def equal_subsampling(adata, obs_key, N_min=None):\n",
    "    '''\n",
    "    Subsample to same class sizes. Classes given by obs_key pointing to categorical in adata.obs.\n",
    "    If N_min is given, downsamples to at least this number instead of the number of cells in the smallest class\n",
    "    and throws out classes with less than N_min cells.\n",
    "    '''\n",
    "    counts = adata.obs[obs_key].value_counts()\n",
    "    groups = counts.index[counts >= N_min]  # ignore groups with less than N_min cells to begin with\n",
    "\n",
    "    if len(groups) == 0:\n",
    "        raise ValueError(f\"No groups have at least {N_min} cells.\")\n",
    "\n",
    "    # We select downsampling target counts by min-max, i.e.\n",
    "    # the largest N such that every group has at least N cells before downsampling.\n",
    "    N = np.min(counts[groups])\n",
    "    N = max(N, N_min) if N_min else N\n",
    "\n",
    "    # subsample indices per group\n",
    "    indices = [\n",
    "        np.random.choice(\n",
    "            adata.obs_names[adata.obs[obs_key] == group], size=N, replace=False\n",
    "        )\n",
    "        for group in groups\n",
    "    ]\n",
    "    selection = np.hstack(np.array(indices))\n",
    "    return adata[selection].copy()"
   ],
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Kec7xKA65IFK",
    "outputId": "993bb731-2dc1-4f72-d850-dfe488905909",
    "ExecuteTime": {
     "end_time": "2024-07-19T19:33:21.149139Z",
     "start_time": "2024-07-19T19:33:21.142209Z"
    }
   },
   "outputs": [],
   "execution_count": 21
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-23T23:03:40.654576Z",
     "start_time": "2024-09-23T23:03:18.498984Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# Load your anndata object\n",
    "adata = sc.read_h5ad(\"../data/norman_2019_raw.h5ad\")"
   ],
   "outputs": [],
   "execution_count": 26
  },
  {
   "metadata": {},
   "cell_type": "code",
   "source": "adata",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AnnData object with n_obs × n_vars = 111445 × 33694\n",
       "    obs: 'guide_identity', 'read_count', 'UMI_count', 'coverage', 'gemgroup', 'good_coverage', 'number_of_cells', 'guide_AHR', 'guide_ARID1A', 'guide_ARRDC3', 'guide_ATL1', 'guide_BAK1', 'guide_BCL2L11', 'guide_BCORL1', 'guide_BPGM', 'guide_C19orf26', 'guide_C3orf72', 'guide_CBFA2T3', 'guide_CBL', 'guide_CDKN1A', 'guide_CDKN1B', 'guide_CDKN1C', 'guide_CEBPA', 'guide_CEBPB', 'guide_CEBPE', 'guide_CELF2', 'guide_CITED1', 'guide_CKS1B', 'guide_CLDN6', 'guide_CNN1', 'guide_CNNM4', 'guide_COL1A1', 'guide_COL2A1', 'guide_CSRNP1', 'guide_DLX2', 'guide_DUSP9', 'guide_EGR1', 'guide_ELMSAN1', 'guide_ETS2', 'guide_FEV', 'guide_FOSB', 'guide_FOXA1', 'guide_FOXA3', 'guide_FOXF1', 'guide_FOXL2', 'guide_FOXO4', 'guide_GLB1L2', 'guide_HES7', 'guide_HK2', 'guide_HNF4A', 'guide_HOXA13', 'guide_HOXB9', 'guide_HOXC13', 'guide_IER5L', 'guide_IGDCC3', 'guide_IKZF3', 'guide_IRF1', 'guide_ISL2', 'guide_JUN', 'guide_KIAA1804', 'guide_KIF18B', 'guide_KIF2C', 'guide_KLF1', 'guide_KMT2A', 'guide_LHX1', 'guide_LYL1', 'guide_MAML2', 'guide_MAP2K3', 'guide_MAP2K6', 'guide_MAP4K3', 'guide_MAP4K5', 'guide_MAP7D1', 'guide_MAPK1', 'guide_MEIS1', 'guide_MIDN', 'guide_NCL', 'guide_NIT1', 'guide_OSR2', 'guide_PLK4', 'guide_POU3F2', 'guide_PRDM1', 'guide_PRTG', 'guide_PTPN1', 'guide_PTPN12', 'guide_PTPN13', 'guide_PTPN9', 'guide_RHOXF2', 'guide_RREB1', 'guide_RUNX1T1', 'guide_S1PR2', 'guide_SAMD1', 'guide_SET', 'guide_SGK1', 'guide_SLC38A2', 'guide_SLC4A1', 'guide_SLC6A9', 'guide_SNAI1', 'guide_SPI1', 'guide_STIL', 'guide_TBX2', 'guide_TBX3', 'guide_TGFBR2', 'guide_TMSB4X', 'guide_TP73', 'guide_TSC22D1', 'guide_UBASH3A', 'guide_UBASH3B', 'guide_ZBTB1', 'guide_ZBTB10', 'guide_ZBTB25', 'guide_ZC3HAV1', 'guide_ZNF318', 'guide_ids'\n",
       "    var: 'gene_symbols'"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 24
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-23T23:04:02.820690Z",
     "start_time": "2024-09-23T23:04:02.815094Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# rename 'guide_ids' to 'perturbation'\n",
    "adata.obs['perturbation'] = adata.obs['guide_ids']\n",
    "total_unique_perts = adata.obs['perturbation'].nunique()\n",
    "total_unique_perts"
   ],
   "outputs": [
    {
     "data": {
      "text/plain": [
       "237"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 28
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-23T22:33:07.183332Z",
     "start_time": "2024-09-23T22:33:07.107127Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# filter all the 'perturbation's that contain '_' from the a data object\n",
    "adata = adata[~adata.obs['perturbation'].str.contains('_')]\n",
    "total_single_gene_perts = adata.obs['perturbation'].nunique()\n",
    "print(\"Number of single gene perturbations: \", total_single_gene_perts - 1)"
   ],
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of single gene perturbations:  236\n"
     ]
    }
   ],
   "execution_count": 26
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-23T23:05:06.385860Z",
     "start_time": "2024-09-23T23:04:05.322246Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# Basic QC and preprocessing\n",
    "print('Shape before filtering: ', adata.shape)\n",
    "sc.pp.filter_cells(adata, min_counts=1000)\n",
    "print('Number of perturbations: ', adata.obs['perturbation'].nunique())\n",
    "\n",
    "# Apply equal subsampling\n",
    "# adata = equal_subsampling(adata, 'perturbation', N_min=300)\n",
    "print('Number of perturbations after subsampling: ', adata.obs['perturbation'].nunique())\n",
    "\n",
    "adata.layers['counts'] = adata.X.copy()\n",
    "print('Normalizing counts...')\n",
    "sc.pp.normalize_per_cell(adata)\n",
    "sc.pp.filter_genes(adata, min_cells=50)\n",
    "print('Applying log1p...')\n",
    "sc.pp.log1p(adata)\n",
    "print('Shape after filtering: ', adata.shape)\n",
    "\n",
    "# Select HVGs\n",
    "print('Selecting highly variable genes...')\n",
    "sc.pp.highly_variable_genes(adata, n_top_genes=2000,\n",
    "                            subset=False, flavor='seurat_v3', layer='counts')\n",
    "print('Number of perturbations after selecting HVGs: ', adata.obs['perturbation'].nunique())\n",
    "\n",
    "print('Running PCA...')\n",
    "sc.pp.pca(adata, use_highly_variable=True, n_comps=50)\n",
    "\n",
    "# Save the filtered and processed anndata object\n",
    "# adata.write_h5ad('filtered_anndata.h5ad')\n",
    "\n",
    "# Display the final number of cells per perturbation\n",
    "cell_counts = adata.obs['perturbation'].value_counts()\n",
    "print(cell_counts)"
   ],
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape before filtering:  (111445, 33694)\n",
      "Number of perturbations:  237\n",
      "Number of perturbations after subsampling:  237\n",
      "Normalizing counts...\n",
      "Applying log1p...\n",
      "Shape after filtering:  (111445, 17414)\n",
      "Selecting highly variable genes...\n",
      "Number of perturbations after selecting HVGs:  237\n",
      "Running PCA...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages/scanpy/preprocessing/_pca.py:374: FutureWarning: Argument `use_highly_variable` is deprecated, consider using the mask argument. Use_highly_variable=True can be called through mask_var=\"highly_variable\". Use_highly_variable=False can be called through mask_var=None\n",
      "  warn(msg, FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "perturbation\n",
      "                 11855\n",
      "KLF1              1960\n",
      "BAK1              1457\n",
      "CEBPE             1233\n",
      "CEBPE,RUNX1T1     1219\n",
      "                 ...  \n",
      "CBL,UBASH3A         64\n",
      "CEBPA,CEBPB         64\n",
      "CEBPB,JUN           59\n",
      "C3orf72,FOXL2       59\n",
      "CEBPA,JUN           54\n",
      "Name: count, Length: 237, dtype: int64\n"
     ]
    }
   ],
   "execution_count": 29
  },
  {
   "cell_type": "code",
   "source": [
    "df = etest(adata, runs=10000, n_jobs=-1)\n",
    "\n",
    "print(df.shape)\n",
    "df['significant_adj'].sum() / len(df)\n",
    "# save the df results\n",
    "df.to_csv('etest_results_single_perts.csv')"
   ],
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "mH4nnSMN6Ch_",
    "outputId": "99c6027d-475c-498f-886e-e1ae1d7c1dfa",
    "ExecuteTime": {
     "end_time": "2024-07-19T20:02:04.744104Z",
     "start_time": "2024-07-19T19:54:09.535191Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  0%|          | 0/10000 [00:00<?, ?it/s]\u001B[A\n",
      "  4%|▎         | 368/10000 [00:31<06:16, 25.60it/s]A\n",
      "  0%|          | 16/10000 [00:06<1:18:49,  2.11it/s]\u001B[A\n",
      "  0%|          | 24/10000 [00:06<45:19,  3.67it/s]  \u001B[A\n",
      "  0%|          | 32/10000 [00:07<30:47,  5.39it/s]\u001B[A\n",
      "  0%|          | 40/10000 [00:07<21:30,  7.72it/s]\u001B[A\n",
      "  0%|          | 48/10000 [00:07<16:37,  9.98it/s]\u001B[A\n",
      "  1%|          | 56/10000 [00:08<13:11, 12.56it/s]\u001B[A\n",
      "  1%|          | 64/10000 [00:08<10:56, 15.14it/s]\u001B[A\n",
      "  1%|          | 72/10000 [00:08<09:24, 17.59it/s]\u001B[A\n",
      "  1%|          | 80/10000 [00:08<08:17, 19.93it/s]\u001B[A\n",
      "  1%|          | 88/10000 [00:09<07:44, 21.34it/s]\u001B[A\n",
      "  1%|          | 96/10000 [00:09<07:16, 22.70it/s]\u001B[A\n",
      "  1%|          | 104/10000 [00:09<07:26, 22.16it/s]\u001B[A\n",
      "  1%|          | 112/10000 [00:10<06:57, 23.66it/s]\u001B[A\n",
      "  1%|          | 120/10000 [00:10<07:20, 22.45it/s]\u001B[A\n",
      "  1%|▏         | 128/10000 [00:10<07:11, 22.88it/s]\u001B[A\n",
      "  1%|▏         | 136/10000 [00:11<07:35, 21.67it/s]\u001B[A\n",
      "  1%|▏         | 144/10000 [00:11<08:14, 19.95it/s]\u001B[A\n",
      "  2%|▏         | 152/10000 [00:12<08:04, 20.35it/s]\u001B[A\n",
      "  2%|▏         | 160/10000 [00:12<07:50, 20.90it/s]\u001B[A\n",
      "  2%|▏         | 168/10000 [00:13<08:12, 19.96it/s]\u001B[A\n",
      "  2%|▏         | 176/10000 [00:13<07:53, 20.74it/s]\u001B[A\n",
      "  2%|▏         | 184/10000 [00:13<07:59, 20.45it/s]\u001B[A\n",
      "  2%|▏         | 192/10000 [00:14<08:17, 19.72it/s]\u001B[A\n",
      "  2%|▏         | 200/10000 [00:14<07:53, 20.70it/s]\u001B[A\n",
      "  2%|▏         | 208/10000 [00:15<08:25, 19.36it/s]\u001B[A\n",
      "  2%|▏         | 216/10000 [00:15<07:50, 20.78it/s]\u001B[A\n",
      "  2%|▏         | 224/10000 [00:15<08:38, 18.86it/s]\u001B[A\n",
      "  2%|▏         | 232/10000 [00:16<08:16, 19.65it/s]\u001B[A\n",
      "  2%|▏         | 240/10000 [00:16<08:02, 20.21it/s]\u001B[A\n",
      "  2%|▏         | 248/10000 [00:16<07:59, 20.35it/s]\u001B[A\n",
      "  3%|▎         | 256/10000 [00:17<07:53, 20.57it/s]\u001B[A\n",
      "  3%|▎         | 264/10000 [00:17<07:37, 21.30it/s]\u001B[A\n",
      "  3%|▎         | 272/10000 [00:18<07:10, 22.59it/s]\u001B[A\n",
      "  3%|▎         | 280/10000 [00:18<06:51, 23.62it/s]\u001B[A\n",
      "  3%|▎         | 288/10000 [00:18<07:00, 23.11it/s]\u001B[A\n",
      "  3%|▎         | 296/10000 [00:19<07:05, 22.78it/s]\u001B[A\n",
      "  3%|▎         | 304/10000 [00:19<07:58, 20.28it/s]\u001B[A\n",
      "  3%|▎         | 312/10000 [00:20<10:00, 16.14it/s]\u001B[A\n",
      "  3%|▎         | 320/10000 [00:20<09:17, 17.36it/s]\u001B[A\n",
      "  3%|▎         | 328/10000 [00:21<08:54, 18.11it/s]\u001B[A\n",
      "  3%|▎         | 336/10000 [00:21<08:15, 19.51it/s]\u001B[A\n",
      "  3%|▎         | 344/10000 [00:21<07:36, 21.16it/s]\u001B[A\n",
      "  4%|▎         | 352/10000 [00:22<07:28, 21.51it/s]\u001B[A\n",
      "  4%|▎         | 360/10000 [00:22<07:18, 22.00it/s]\u001B[A\n",
      "  4%|▎         | 368/10000 [00:22<07:20, 21.85it/s]\u001B[A\n",
      "  4%|▍         | 376/10000 [00:23<07:30, 21.35it/s]\u001B[A\n",
      "  4%|▍         | 384/10000 [00:23<07:26, 21.52it/s]\u001B[A\n",
      "  4%|▍         | 392/10000 [00:23<07:27, 21.48it/s]\u001B[A\n",
      "  4%|▍         | 400/10000 [00:24<07:17, 21.95it/s]\u001B[A\n",
      "  4%|▍         | 408/10000 [00:24<07:11, 22.23it/s]\u001B[A\n",
      "  4%|▍         | 416/10000 [00:24<07:09, 22.33it/s]\u001B[A\n",
      "  4%|▍         | 424/10000 [00:25<06:56, 23.00it/s]\u001B[A\n",
      "  4%|▍         | 432/10000 [00:25<08:38, 18.44it/s]\u001B[A\n",
      "  4%|▍         | 440/10000 [00:26<09:16, 17.19it/s]\u001B[A\n",
      "  4%|▍         | 448/10000 [00:26<09:00, 17.69it/s]\u001B[A\n",
      "  5%|▍         | 456/10000 [00:27<08:10, 19.47it/s]\u001B[A\n",
      "  5%|▍         | 464/10000 [00:27<07:54, 20.11it/s]\u001B[A\n",
      "  5%|▍         | 472/10000 [00:27<07:51, 20.21it/s]\u001B[A\n",
      "  5%|▍         | 480/10000 [00:28<07:00, 22.62it/s]\u001B[A\n",
      "  5%|▍         | 488/10000 [00:28<07:05, 22.34it/s]\u001B[A\n",
      "  5%|▍         | 496/10000 [00:28<07:10, 22.07it/s]\u001B[A\n",
      "  5%|▌         | 504/10000 [00:29<07:24, 21.35it/s]\u001B[A\n",
      "  5%|▌         | 512/10000 [00:29<07:15, 21.80it/s]\u001B[A\n",
      "  5%|▌         | 520/10000 [00:29<06:30, 24.30it/s]\u001B[A\n",
      "  5%|▌         | 528/10000 [00:30<06:44, 23.44it/s]\u001B[A\n",
      "  5%|▌         | 536/10000 [00:30<07:03, 22.36it/s]\u001B[A\n",
      "  5%|▌         | 544/10000 [00:31<07:28, 21.09it/s]\u001B[A\n",
      "  6%|▌         | 552/10000 [00:31<07:13, 21.81it/s]\u001B[A\n",
      "  6%|▌         | 560/10000 [00:31<07:07, 22.06it/s]\u001B[A\n",
      "  6%|▌         | 568/10000 [00:32<08:01, 19.59it/s]\u001B[A\n",
      "  6%|▌         | 576/10000 [00:32<07:56, 19.78it/s]\u001B[A\n",
      "  6%|▌         | 584/10000 [00:33<07:51, 19.97it/s]\u001B[A\n",
      "  6%|▌         | 592/10000 [00:33<07:56, 19.73it/s]\u001B[A\n",
      "  6%|▌         | 600/10000 [00:33<07:21, 21.30it/s]\u001B[A\n",
      "  6%|▌         | 608/10000 [00:34<07:12, 21.69it/s]\u001B[A\n",
      "  6%|▌         | 616/10000 [00:34<07:32, 20.74it/s]\u001B[A\n",
      "  6%|▌         | 624/10000 [00:34<07:03, 22.12it/s]\u001B[A\n",
      "  6%|▋         | 632/10000 [00:35<07:04, 22.04it/s]\u001B[A\n",
      "  6%|▋         | 640/10000 [00:35<07:00, 22.26it/s]\u001B[A\n",
      "  6%|▋         | 648/10000 [00:35<06:40, 23.34it/s]\u001B[A\n",
      "  7%|▋         | 656/10000 [00:36<06:57, 22.40it/s]\u001B[A\n",
      "  7%|▋         | 664/10000 [00:36<06:56, 22.40it/s]\u001B[A\n",
      "  7%|▋         | 672/10000 [00:37<07:18, 21.30it/s]\u001B[A\n",
      "  7%|▋         | 680/10000 [00:37<06:56, 22.37it/s]\u001B[A\n",
      "  7%|▋         | 688/10000 [00:37<06:45, 22.98it/s]\u001B[A\n",
      "  7%|▋         | 696/10000 [00:38<06:41, 23.17it/s]\u001B[A\n",
      "  7%|▋         | 704/10000 [00:38<06:44, 23.00it/s]\u001B[A\n",
      "  7%|▋         | 712/10000 [00:38<06:39, 23.26it/s]\u001B[A\n",
      "  7%|▋         | 720/10000 [00:39<06:29, 23.84it/s]\u001B[A\n",
      "  7%|▋         | 728/10000 [00:39<06:40, 23.15it/s]\u001B[A\n",
      "  7%|▋         | 736/10000 [00:39<06:43, 22.96it/s]\u001B[A\n",
      "  7%|▋         | 744/10000 [00:40<06:32, 23.57it/s]\u001B[A\n",
      "  8%|▊         | 752/10000 [00:40<06:39, 23.17it/s]\u001B[A\n",
      "  8%|▊         | 760/10000 [00:40<06:41, 23.03it/s]\u001B[A\n",
      "  8%|▊         | 768/10000 [00:41<06:30, 23.66it/s]\u001B[A\n",
      "  8%|▊         | 776/10000 [00:41<06:56, 22.13it/s]\u001B[A\n",
      "  8%|▊         | 784/10000 [00:41<06:42, 22.88it/s]\u001B[A\n",
      "  8%|▊         | 792/10000 [00:42<06:43, 22.84it/s]\u001B[A\n",
      "  8%|▊         | 800/10000 [00:42<06:36, 23.21it/s]\u001B[A\n",
      "  8%|▊         | 808/10000 [00:42<06:30, 23.54it/s]\u001B[A\n",
      "  8%|▊         | 816/10000 [00:43<06:57, 22.02it/s]\u001B[A\n",
      "  8%|▊         | 824/10000 [00:43<06:33, 23.33it/s]\u001B[A\n",
      "  8%|▊         | 832/10000 [00:44<06:54, 22.10it/s]\u001B[A\n",
      "  8%|▊         | 840/10000 [00:44<06:40, 22.88it/s]\u001B[A\n",
      "  8%|▊         | 848/10000 [00:44<06:44, 22.65it/s]\u001B[A\n",
      "  9%|▊         | 856/10000 [00:45<06:56, 21.96it/s]\u001B[A\n",
      "  9%|▊         | 864/10000 [00:45<06:51, 22.19it/s]\u001B[A\n",
      "  9%|▊         | 872/10000 [00:45<06:32, 23.24it/s]\u001B[A\n",
      "  9%|▉         | 880/10000 [00:46<06:51, 22.17it/s]\u001B[A\n",
      "  9%|▉         | 888/10000 [00:46<06:22, 23.84it/s]\u001B[A\n",
      "  9%|▉         | 896/10000 [00:46<06:29, 23.39it/s]\u001B[A\n",
      "  9%|▉         | 904/10000 [00:47<06:31, 23.24it/s]\u001B[A\n",
      "  9%|▉         | 912/10000 [00:47<06:19, 23.97it/s]\u001B[A\n",
      "  9%|▉         | 920/10000 [00:47<06:31, 23.20it/s]\u001B[A\n",
      "  9%|▉         | 928/10000 [00:48<06:23, 23.63it/s]\u001B[A\n",
      "  9%|▉         | 936/10000 [00:48<06:25, 23.51it/s]\u001B[A\n",
      "  9%|▉         | 944/10000 [00:48<06:39, 22.66it/s]\u001B[A\n",
      " 10%|▉         | 952/10000 [00:49<06:33, 22.99it/s]\u001B[A\n",
      " 10%|▉         | 960/10000 [00:49<06:34, 22.94it/s]\u001B[A\n",
      " 10%|▉         | 968/10000 [00:49<06:14, 24.09it/s]\u001B[A\n",
      " 10%|▉         | 976/10000 [00:50<06:20, 23.71it/s]\u001B[A\n",
      " 10%|▉         | 984/10000 [00:50<06:16, 23.92it/s]\u001B[A\n",
      " 10%|▉         | 992/10000 [00:50<06:18, 23.77it/s]\u001B[A\n",
      " 10%|█         | 1000/10000 [00:51<06:23, 23.45it/s]\u001B[A\n",
      " 10%|█         | 1008/10000 [00:51<06:17, 23.80it/s]\u001B[A\n",
      " 10%|█         | 1016/10000 [00:51<06:27, 23.19it/s]\u001B[A\n",
      " 10%|█         | 1024/10000 [00:52<07:09, 20.87it/s]\u001B[A\n",
      " 10%|█         | 1032/10000 [00:52<07:21, 20.32it/s]\u001B[A\n",
      " 10%|█         | 1040/10000 [00:53<06:50, 21.85it/s]\u001B[A\n",
      " 10%|█         | 1048/10000 [00:53<07:01, 21.23it/s]\u001B[A\n",
      " 11%|█         | 1056/10000 [00:53<06:27, 23.09it/s]\u001B[A\n",
      " 11%|█         | 1064/10000 [00:54<06:37, 22.45it/s]\u001B[A\n",
      " 11%|█         | 1072/10000 [00:54<06:39, 22.36it/s]\u001B[A\n",
      " 11%|█         | 1080/10000 [00:54<06:24, 23.18it/s]\u001B[A\n",
      " 11%|█         | 1088/10000 [00:55<06:16, 23.65it/s]\u001B[A\n",
      " 11%|█         | 1096/10000 [00:55<06:39, 22.28it/s]\u001B[A\n",
      " 11%|█         | 1104/10000 [00:55<06:37, 22.36it/s]\u001B[A\n",
      " 11%|█         | 1112/10000 [00:56<06:42, 22.10it/s]\u001B[A\n",
      " 11%|█         | 1120/10000 [00:56<06:25, 23.05it/s]\u001B[A\n",
      " 11%|█▏        | 1128/10000 [00:56<06:28, 22.84it/s]\u001B[A\n",
      " 11%|█▏        | 1136/10000 [00:57<06:44, 21.91it/s]\u001B[A\n",
      " 11%|█▏        | 1144/10000 [00:57<06:17, 23.44it/s]\u001B[A\n",
      " 12%|█▏        | 1152/10000 [00:58<06:22, 23.14it/s]\u001B[A\n",
      " 12%|█▏        | 1160/10000 [00:58<06:09, 23.90it/s]\u001B[A\n",
      " 12%|█▏        | 1168/10000 [00:58<06:20, 23.18it/s]\u001B[A\n",
      " 12%|█▏        | 1176/10000 [00:59<06:10, 23.79it/s]\u001B[A\n",
      " 12%|█▏        | 1184/10000 [00:59<06:15, 23.51it/s]\u001B[A\n",
      " 12%|█▏        | 1192/10000 [00:59<06:13, 23.60it/s]\u001B[A\n",
      " 12%|█▏        | 1200/10000 [01:00<06:25, 22.85it/s]\u001B[A\n",
      " 12%|█▏        | 1208/10000 [01:00<06:06, 24.00it/s]\u001B[A\n",
      " 12%|█▏        | 1216/10000 [01:00<06:16, 23.34it/s]\u001B[A\n",
      " 12%|█▏        | 1224/10000 [01:01<06:15, 23.36it/s]\u001B[A\n",
      " 12%|█▏        | 1232/10000 [01:01<06:19, 23.09it/s]\u001B[A\n",
      " 12%|█▏        | 1240/10000 [01:01<06:15, 23.34it/s]\u001B[A\n",
      " 12%|█▏        | 1248/10000 [01:02<06:28, 22.54it/s]\u001B[A\n",
      " 13%|█▎        | 1256/10000 [01:02<06:16, 23.23it/s]\u001B[A\n",
      " 13%|█▎        | 1264/10000 [01:02<06:05, 23.93it/s]\u001B[A\n",
      " 13%|█▎        | 1272/10000 [01:03<05:59, 24.27it/s]\u001B[A\n",
      " 13%|█▎        | 1280/10000 [01:03<06:00, 24.22it/s]\u001B[A\n",
      " 13%|█▎        | 1288/10000 [01:03<06:09, 23.59it/s]\u001B[A\n",
      " 13%|█▎        | 1296/10000 [01:04<06:04, 23.89it/s]\u001B[A\n",
      " 13%|█▎        | 1304/10000 [01:04<05:51, 24.72it/s]\u001B[A\n",
      " 13%|█▎        | 1312/10000 [01:04<06:05, 23.75it/s]\u001B[A\n",
      " 13%|█▎        | 1320/10000 [01:05<06:07, 23.62it/s]\u001B[A\n",
      " 13%|█▎        | 1328/10000 [01:06<09:31, 15.18it/s]\u001B[A\n",
      " 13%|█▎        | 1336/10000 [01:06<10:16, 14.05it/s]\u001B[A\n",
      " 13%|█▎        | 1344/10000 [01:07<09:53, 14.59it/s]\u001B[A\n",
      " 14%|█▎        | 1352/10000 [01:07<10:18, 13.98it/s]\u001B[A\n",
      " 14%|█▎        | 1360/10000 [01:09<13:40, 10.53it/s]\u001B[A\n",
      " 14%|█▎        | 1368/10000 [01:09<12:23, 11.62it/s]\u001B[A\n",
      " 14%|█▍        | 1376/10000 [01:10<13:45, 10.45it/s]\u001B[A\n",
      " 14%|█▍        | 1384/10000 [01:11<13:24, 10.71it/s]\u001B[A\n",
      " 14%|█▍        | 1392/10000 [01:11<12:30, 11.47it/s]\u001B[A\n",
      " 14%|█▍        | 1400/10000 [01:12<10:40, 13.43it/s]\u001B[A\n",
      " 14%|█▍        | 1408/10000 [01:12<09:56, 14.40it/s]\u001B[A\n",
      " 14%|█▍        | 1416/10000 [01:13<09:19, 15.34it/s]\u001B[A\n",
      " 14%|█▍        | 1424/10000 [01:13<08:13, 17.39it/s]\u001B[A\n",
      " 14%|█▍        | 1432/10000 [01:14<10:13, 13.97it/s]\u001B[A\n",
      " 14%|█▍        | 1440/10000 [01:14<08:33, 16.67it/s]\u001B[A\n",
      " 14%|█▍        | 1448/10000 [01:15<10:26, 13.65it/s]\u001B[A\n",
      " 15%|█▍        | 1456/10000 [01:15<08:36, 16.53it/s]\u001B[A\n",
      " 15%|█▍        | 1464/10000 [01:16<09:35, 14.83it/s]\u001B[A\n",
      " 15%|█▍        | 1472/10000 [01:16<09:09, 15.51it/s]\u001B[A\n",
      " 15%|█▍        | 1480/10000 [01:17<08:25, 16.87it/s]\u001B[A\n",
      " 15%|█▍        | 1488/10000 [01:17<07:37, 18.59it/s]\u001B[A\n",
      " 15%|█▍        | 1496/10000 [01:17<07:08, 19.86it/s]\u001B[A\n",
      " 15%|█▌        | 1504/10000 [01:18<06:48, 20.79it/s]\u001B[A\n",
      " 15%|█▌        | 1512/10000 [01:18<06:52, 20.59it/s]\u001B[A\n",
      " 15%|█▌        | 1520/10000 [01:18<06:44, 20.94it/s]\u001B[A\n",
      " 15%|█▌        | 1528/10000 [01:19<06:15, 22.54it/s]\u001B[A\n",
      " 15%|█▌        | 1536/10000 [01:19<06:05, 23.15it/s]\u001B[A\n",
      " 15%|█▌        | 1544/10000 [01:19<06:14, 22.60it/s]\u001B[A\n",
      " 16%|█▌        | 1552/10000 [01:20<05:59, 23.51it/s]\u001B[A\n",
      " 16%|█▌        | 1560/10000 [01:20<05:50, 24.09it/s]\u001B[A\n",
      " 16%|█▌        | 1568/10000 [01:20<05:37, 25.01it/s]\u001B[A\n",
      " 16%|█▌        | 1576/10000 [01:21<05:53, 23.85it/s]\u001B[A\n",
      " 16%|█▌        | 1584/10000 [01:21<05:32, 25.35it/s]\u001B[A\n",
      " 16%|█▌        | 1592/10000 [01:21<05:31, 25.39it/s]\u001B[A\n",
      " 16%|█▌        | 1600/10000 [01:22<05:39, 24.78it/s]\u001B[A\n",
      " 16%|█▌        | 1608/10000 [01:22<05:40, 24.64it/s]\u001B[A\n",
      " 16%|█▌        | 1616/10000 [01:22<06:25, 21.73it/s]\u001B[A\n",
      " 16%|█▌        | 1624/10000 [01:23<06:19, 22.08it/s]\u001B[A\n",
      " 16%|█▋        | 1632/10000 [01:23<06:20, 22.01it/s]\u001B[A\n",
      " 16%|█▋        | 1640/10000 [01:23<06:04, 22.96it/s]\u001B[A\n",
      " 16%|█▋        | 1648/10000 [01:24<05:50, 23.81it/s]\u001B[A\n",
      " 17%|█▋        | 1656/10000 [01:24<07:58, 17.45it/s]\u001B[A\n",
      " 17%|█▋        | 1664/10000 [01:25<08:08, 17.05it/s]\u001B[A\n",
      " 17%|█▋        | 1672/10000 [01:25<07:57, 17.44it/s]\u001B[A\n",
      " 17%|█▋        | 1680/10000 [01:26<07:43, 17.94it/s]\u001B[A\n",
      " 17%|█▋        | 1688/10000 [01:26<07:23, 18.74it/s]\u001B[A\n",
      " 17%|█▋        | 1696/10000 [01:26<06:50, 20.22it/s]\u001B[A\n",
      " 17%|█▋        | 1704/10000 [01:27<06:37, 20.87it/s]\u001B[A\n",
      " 17%|█▋        | 1712/10000 [01:27<06:17, 21.96it/s]\u001B[A\n",
      " 17%|█▋        | 1720/10000 [01:27<06:06, 22.57it/s]\u001B[A\n",
      " 17%|█▋        | 1728/10000 [01:28<06:22, 21.60it/s]\u001B[A\n",
      " 17%|█▋        | 1736/10000 [01:28<05:49, 23.68it/s]\u001B[A\n",
      " 17%|█▋        | 1744/10000 [01:29<05:49, 23.65it/s]\u001B[A\n",
      " 18%|█▊        | 1752/10000 [01:29<05:41, 24.19it/s]\u001B[A\n",
      " 18%|█▊        | 1760/10000 [01:29<05:42, 24.06it/s]\u001B[A\n",
      " 18%|█▊        | 1768/10000 [01:29<05:38, 24.29it/s]\u001B[A\n",
      " 18%|█▊        | 1776/10000 [01:30<05:40, 24.19it/s]\u001B[A\n",
      " 18%|█▊        | 1784/10000 [01:30<05:56, 23.03it/s]\u001B[A\n",
      " 18%|█▊        | 1792/10000 [01:31<05:55, 23.10it/s]\u001B[A\n",
      " 18%|█▊        | 1800/10000 [01:31<05:39, 24.17it/s]\u001B[A\n",
      " 18%|█▊        | 1808/10000 [01:31<05:40, 24.05it/s]\u001B[A\n",
      " 18%|█▊        | 1816/10000 [01:32<06:06, 22.31it/s]\u001B[A\n",
      " 18%|█▊        | 1824/10000 [01:32<06:00, 22.67it/s]\u001B[A\n",
      " 18%|█▊        | 1832/10000 [01:32<05:59, 22.71it/s]\u001B[A\n",
      " 18%|█▊        | 1840/10000 [01:33<05:36, 24.22it/s]\u001B[A\n",
      " 18%|█▊        | 1848/10000 [01:33<05:28, 24.80it/s]\u001B[A\n",
      " 19%|█▊        | 1856/10000 [01:33<05:40, 23.90it/s]\u001B[A\n",
      " 19%|█▊        | 1864/10000 [01:34<05:49, 23.29it/s]\u001B[A\n",
      " 19%|█▊        | 1872/10000 [01:34<05:39, 23.93it/s]\u001B[A\n",
      " 19%|█▉        | 1880/10000 [01:34<05:36, 24.12it/s]\u001B[A\n",
      " 19%|█▉        | 1888/10000 [01:35<05:37, 24.02it/s]\u001B[A\n",
      " 19%|█▉        | 1896/10000 [01:35<05:38, 23.94it/s]\u001B[A\n",
      " 19%|█▉        | 1904/10000 [01:35<05:25, 24.90it/s]\u001B[A\n",
      " 19%|█▉        | 1912/10000 [01:36<05:38, 23.91it/s]\u001B[A\n",
      " 19%|█▉        | 1920/10000 [01:36<06:23, 21.04it/s]\u001B[A\n",
      " 19%|█▉        | 1928/10000 [01:36<06:12, 21.66it/s]\u001B[A\n",
      " 19%|█▉        | 1936/10000 [01:37<05:56, 22.59it/s]\u001B[A\n",
      " 19%|█▉        | 1944/10000 [01:37<06:04, 22.10it/s]\u001B[A\n",
      " 20%|█▉        | 1952/10000 [01:38<06:33, 20.45it/s]\u001B[A\n",
      " 20%|█▉        | 1960/10000 [01:38<06:05, 22.00it/s]\u001B[A\n",
      " 20%|█▉        | 1968/10000 [01:38<06:00, 22.29it/s]\u001B[A\n",
      " 20%|█▉        | 1976/10000 [01:39<06:15, 21.35it/s]\u001B[A\n",
      " 20%|█▉        | 1984/10000 [01:39<06:07, 21.79it/s]\u001B[A\n",
      " 20%|█▉        | 1992/10000 [01:39<05:35, 23.88it/s]\u001B[A\n",
      " 20%|██        | 2000/10000 [01:40<05:42, 23.35it/s]\u001B[A\n",
      " 20%|██        | 2008/10000 [01:40<05:34, 23.87it/s]\u001B[A\n",
      " 20%|██        | 2016/10000 [01:40<05:29, 24.22it/s]\u001B[A\n",
      " 20%|██        | 2024/10000 [01:41<05:29, 24.22it/s]\u001B[A\n",
      " 20%|██        | 2032/10000 [01:41<05:40, 23.38it/s]\u001B[A\n",
      " 20%|██        | 2040/10000 [01:41<05:25, 24.45it/s]\u001B[A\n",
      " 20%|██        | 2048/10000 [01:42<05:24, 24.53it/s]\u001B[A\n",
      " 21%|██        | 2056/10000 [01:42<05:29, 24.14it/s]\u001B[A\n",
      " 21%|██        | 2064/10000 [01:42<05:33, 23.81it/s]\u001B[A\n",
      " 21%|██        | 2072/10000 [01:43<05:45, 22.93it/s]\u001B[A\n",
      " 21%|██        | 2080/10000 [01:43<05:19, 24.80it/s]\u001B[A\n",
      " 21%|██        | 2088/10000 [01:43<05:36, 23.54it/s]\u001B[A\n",
      " 21%|██        | 2096/10000 [01:44<05:25, 24.28it/s]\u001B[A\n",
      " 21%|██        | 2104/10000 [01:44<05:35, 23.55it/s]\u001B[A\n",
      " 21%|██        | 2112/10000 [01:44<05:21, 24.53it/s]\u001B[A\n",
      " 21%|██        | 2120/10000 [01:45<05:25, 24.21it/s]\u001B[A\n",
      " 21%|██▏       | 2128/10000 [01:45<05:21, 24.46it/s]\u001B[A\n",
      " 21%|██▏       | 2136/10000 [01:45<05:23, 24.34it/s]\u001B[A\n",
      " 21%|██▏       | 2144/10000 [01:46<05:19, 24.59it/s]\u001B[A\n",
      " 22%|██▏       | 2152/10000 [01:46<05:06, 25.62it/s]\u001B[A\n",
      " 22%|██▏       | 2160/10000 [01:47<08:24, 15.55it/s]\u001B[A\n",
      " 22%|██▏       | 2168/10000 [01:47<08:05, 16.13it/s]\u001B[A\n",
      " 22%|██▏       | 2176/10000 [01:48<10:24, 12.53it/s]\u001B[A\n",
      " 22%|██▏       | 2184/10000 [01:49<11:39, 11.18it/s]\u001B[A\n",
      " 22%|██▏       | 2192/10000 [01:49<09:07, 14.25it/s]\u001B[A\n",
      " 22%|██▏       | 2200/10000 [01:50<07:45, 16.76it/s]\u001B[A\n",
      " 22%|██▏       | 2208/10000 [01:50<07:12, 18.02it/s]\u001B[A\n",
      " 22%|██▏       | 2216/10000 [01:50<06:50, 18.95it/s]\u001B[A\n",
      " 22%|██▏       | 2224/10000 [01:51<06:17, 20.62it/s]\u001B[A\n",
      " 22%|██▏       | 2232/10000 [01:51<05:56, 21.79it/s]\u001B[A\n",
      " 22%|██▏       | 2240/10000 [01:51<05:43, 22.56it/s]\u001B[A\n",
      " 22%|██▏       | 2248/10000 [01:52<05:45, 22.42it/s]\u001B[A\n",
      " 23%|██▎       | 2256/10000 [01:52<06:06, 21.15it/s]\u001B[A\n",
      " 23%|██▎       | 2264/10000 [01:52<05:54, 21.82it/s]\u001B[A\n",
      " 23%|██▎       | 2272/10000 [01:53<06:07, 21.04it/s]\u001B[A\n",
      " 23%|██▎       | 2280/10000 [01:53<06:15, 20.54it/s]\u001B[A\n",
      " 23%|██▎       | 2288/10000 [01:54<05:55, 21.67it/s]\u001B[A\n",
      " 23%|██▎       | 2296/10000 [01:54<05:57, 21.53it/s]\u001B[A\n",
      " 23%|██▎       | 2304/10000 [01:54<05:55, 21.65it/s]\u001B[A\n",
      " 23%|██▎       | 2312/10000 [01:55<05:44, 22.29it/s]\u001B[A\n",
      " 23%|██▎       | 2320/10000 [01:55<05:35, 22.87it/s]\u001B[A\n",
      " 23%|██▎       | 2328/10000 [01:55<05:45, 22.23it/s]\u001B[A\n",
      " 23%|██▎       | 2336/10000 [01:56<05:24, 23.64it/s]\u001B[A\n",
      " 23%|██▎       | 2344/10000 [01:56<05:24, 23.61it/s]\u001B[A\n",
      " 24%|██▎       | 2352/10000 [01:56<05:14, 24.33it/s]\u001B[A\n",
      " 24%|██▎       | 2360/10000 [01:57<05:09, 24.71it/s]\u001B[A\n",
      " 24%|██▎       | 2368/10000 [01:57<05:00, 25.38it/s]\u001B[A\n",
      " 24%|██▍       | 2376/10000 [01:57<05:12, 24.38it/s]\u001B[A\n",
      " 24%|██▍       | 2384/10000 [01:58<05:09, 24.63it/s]\u001B[A\n",
      " 24%|██▍       | 2392/10000 [01:58<05:05, 24.88it/s]\u001B[A\n",
      " 24%|██▍       | 2400/10000 [01:58<05:13, 24.26it/s]\u001B[A\n",
      " 24%|██▍       | 2408/10000 [01:59<05:01, 25.16it/s]\u001B[A\n",
      " 24%|██▍       | 2416/10000 [01:59<04:56, 25.57it/s]\u001B[A\n",
      " 24%|██▍       | 2424/10000 [01:59<05:07, 24.64it/s]\u001B[A\n",
      " 24%|██▍       | 2432/10000 [01:59<05:01, 25.09it/s]\u001B[A\n",
      " 24%|██▍       | 2440/10000 [02:00<05:13, 24.10it/s]\u001B[A\n",
      " 24%|██▍       | 2448/10000 [02:00<05:24, 23.29it/s]\u001B[A\n",
      " 25%|██▍       | 2456/10000 [02:01<05:52, 21.41it/s]\u001B[A\n",
      " 25%|██▍       | 2464/10000 [02:01<05:20, 23.51it/s]\u001B[A\n",
      " 25%|██▍       | 2472/10000 [02:01<05:22, 23.31it/s]\u001B[A\n",
      " 25%|██▍       | 2480/10000 [02:02<05:17, 23.67it/s]\u001B[A\n",
      " 25%|██▍       | 2488/10000 [02:02<05:14, 23.87it/s]\u001B[A\n",
      " 25%|██▍       | 2496/10000 [02:02<05:19, 23.47it/s]\u001B[A\n",
      " 25%|██▌       | 2504/10000 [02:03<05:14, 23.80it/s]\u001B[A\n",
      " 25%|██▌       | 2512/10000 [02:03<05:15, 23.76it/s]\u001B[A\n",
      " 25%|██▌       | 2520/10000 [02:03<05:19, 23.39it/s]\u001B[A\n",
      " 25%|██▌       | 2528/10000 [02:04<06:07, 20.33it/s]\u001B[A\n",
      " 25%|██▌       | 2536/10000 [02:04<07:07, 17.45it/s]\u001B[A\n",
      " 25%|██▌       | 2544/10000 [02:05<07:58, 15.60it/s]\u001B[A\n",
      " 26%|██▌       | 2552/10000 [02:06<07:53, 15.72it/s]\u001B[A\n",
      " 26%|██▌       | 2560/10000 [02:06<08:08, 15.24it/s]\u001B[A\n",
      " 26%|██▌       | 2568/10000 [02:07<07:44, 16.00it/s]\u001B[A\n",
      " 26%|██▌       | 2576/10000 [02:07<07:54, 15.64it/s]\u001B[A\n",
      " 26%|██▌       | 2584/10000 [02:08<07:42, 16.02it/s]\u001B[A\n",
      " 26%|██▌       | 2592/10000 [02:08<08:01, 15.37it/s]\u001B[A\n",
      " 26%|██▌       | 2600/10000 [02:09<07:29, 16.46it/s]\u001B[A\n",
      " 26%|██▌       | 2608/10000 [02:09<07:50, 15.71it/s]\u001B[A\n",
      " 26%|██▌       | 2616/10000 [02:10<08:08, 15.13it/s]\u001B[A\n",
      " 26%|██▌       | 2624/10000 [02:10<08:30, 14.44it/s]\u001B[A\n",
      " 26%|██▋       | 2632/10000 [02:11<07:30, 16.37it/s]\u001B[A\n",
      " 26%|██▋       | 2640/10000 [02:11<08:26, 14.53it/s]\u001B[A\n",
      " 26%|██▋       | 2648/10000 [02:12<07:53, 15.51it/s]\u001B[A\n",
      " 27%|██▋       | 2656/10000 [02:12<07:23, 16.57it/s]\u001B[A\n",
      " 27%|██▋       | 2664/10000 [02:12<06:33, 18.64it/s]\u001B[A\n",
      " 27%|██▋       | 2672/10000 [02:13<06:15, 19.54it/s]\u001B[A\n",
      " 27%|██▋       | 2680/10000 [02:13<06:08, 19.86it/s]\u001B[A\n",
      " 27%|██▋       | 2688/10000 [02:14<05:51, 20.81it/s]\u001B[A\n",
      " 27%|██▋       | 2696/10000 [02:14<05:32, 21.94it/s]\u001B[A\n",
      " 27%|██▋       | 2704/10000 [02:14<05:31, 21.98it/s]\u001B[A\n",
      " 27%|██▋       | 2712/10000 [02:15<05:22, 22.57it/s]\u001B[A\n",
      " 27%|██▋       | 2720/10000 [02:15<05:10, 23.44it/s]\u001B[A\n",
      " 27%|██▋       | 2728/10000 [02:15<05:38, 21.45it/s]\u001B[A\n",
      " 27%|██▋       | 2736/10000 [02:16<05:32, 21.85it/s]\u001B[A\n",
      " 27%|██▋       | 2744/10000 [02:16<05:18, 22.75it/s]\u001B[A\n",
      " 28%|██▊       | 2752/10000 [02:16<05:01, 24.04it/s]\u001B[A\n",
      " 28%|██▊       | 2760/10000 [02:17<05:08, 23.47it/s]\u001B[A\n",
      " 28%|██▊       | 2768/10000 [02:17<05:06, 23.60it/s]\u001B[A\n",
      " 28%|██▊       | 2776/10000 [02:17<05:06, 23.53it/s]\u001B[A\n",
      " 28%|██▊       | 2784/10000 [02:18<05:03, 23.77it/s]\u001B[A\n",
      " 28%|██▊       | 2792/10000 [02:18<04:58, 24.19it/s]\u001B[A\n",
      " 28%|██▊       | 2800/10000 [02:18<04:40, 25.71it/s]\u001B[A\n",
      " 28%|██▊       | 2808/10000 [02:19<04:53, 24.51it/s]\u001B[A\n",
      " 28%|██▊       | 2816/10000 [02:19<04:51, 24.68it/s]\u001B[A\n",
      " 28%|██▊       | 2824/10000 [02:19<04:46, 25.08it/s]\u001B[A\n",
      " 28%|██▊       | 2832/10000 [02:20<04:52, 24.49it/s]\u001B[A\n",
      " 28%|██▊       | 2840/10000 [02:20<04:49, 24.76it/s]\u001B[A\n",
      " 28%|██▊       | 2848/10000 [02:20<04:53, 24.36it/s]\u001B[A\n",
      " 29%|██▊       | 2856/10000 [02:21<04:48, 24.77it/s]\u001B[A\n",
      " 29%|██▊       | 2864/10000 [02:21<04:46, 24.86it/s]\u001B[A\n",
      " 29%|██▊       | 2872/10000 [02:21<04:54, 24.19it/s]\u001B[A\n",
      " 29%|██▉       | 2880/10000 [02:21<04:36, 25.75it/s]\u001B[A\n",
      " 29%|██▉       | 2888/10000 [02:22<04:40, 25.36it/s]\u001B[A\n",
      " 29%|██▉       | 2896/10000 [02:22<04:51, 24.37it/s]\u001B[A\n",
      " 29%|██▉       | 2904/10000 [02:22<04:51, 24.38it/s]\u001B[A\n",
      " 29%|██▉       | 2912/10000 [02:23<04:50, 24.38it/s]\u001B[A\n",
      " 29%|██▉       | 2920/10000 [02:23<04:46, 24.67it/s]\u001B[A\n",
      " 29%|██▉       | 2928/10000 [02:23<04:42, 25.07it/s]\u001B[A\n",
      " 29%|██▉       | 2936/10000 [02:24<04:41, 25.11it/s]\u001B[A\n",
      " 29%|██▉       | 2944/10000 [02:24<04:55, 23.84it/s]\u001B[A\n",
      " 30%|██▉       | 2952/10000 [02:24<04:53, 23.98it/s]\u001B[A\n",
      " 30%|██▉       | 2960/10000 [02:25<04:44, 24.74it/s]\u001B[A\n",
      " 30%|██▉       | 2968/10000 [02:25<04:45, 24.65it/s]\u001B[A\n",
      " 30%|██▉       | 2976/10000 [02:25<04:40, 25.04it/s]\u001B[A\n",
      " 30%|██▉       | 2984/10000 [02:26<04:35, 25.47it/s]\u001B[A\n",
      " 30%|██▉       | 2992/10000 [02:26<04:47, 24.38it/s]\u001B[A\n",
      " 30%|███       | 3000/10000 [02:26<04:44, 24.64it/s]\u001B[A\n",
      " 30%|███       | 3008/10000 [02:27<04:38, 25.14it/s]\u001B[A\n",
      " 30%|███       | 3016/10000 [02:27<04:50, 24.05it/s]\u001B[A\n",
      " 30%|███       | 3024/10000 [02:27<04:48, 24.22it/s]\u001B[A\n",
      " 30%|███       | 3032/10000 [02:28<04:48, 24.18it/s]\u001B[A\n",
      " 30%|███       | 3040/10000 [02:28<04:35, 25.25it/s]\u001B[A\n",
      " 30%|███       | 3048/10000 [02:28<04:50, 23.89it/s]\u001B[A\n",
      " 31%|███       | 3056/10000 [02:29<04:47, 24.13it/s]\u001B[A\n",
      " 31%|███       | 3064/10000 [02:29<04:50, 23.86it/s]\u001B[A\n",
      " 31%|███       | 3072/10000 [02:29<04:37, 24.95it/s]\u001B[A\n",
      " 31%|███       | 3080/10000 [02:30<04:30, 25.56it/s]\u001B[A\n",
      " 31%|███       | 3088/10000 [02:30<04:34, 25.21it/s]\u001B[A\n",
      " 31%|███       | 3096/10000 [02:30<04:39, 24.72it/s]\u001B[A\n",
      " 31%|███       | 3104/10000 [02:31<05:22, 21.39it/s]\u001B[A\n",
      " 31%|███       | 3112/10000 [02:31<05:16, 21.74it/s]\u001B[A\n",
      " 31%|███       | 3120/10000 [02:31<04:46, 24.00it/s]\u001B[A\n",
      " 31%|███▏      | 3128/10000 [02:32<04:37, 24.74it/s]\u001B[A\n",
      " 31%|███▏      | 3136/10000 [02:32<06:06, 18.75it/s]\u001B[A\n",
      " 31%|███▏      | 3144/10000 [02:33<06:19, 18.09it/s]\u001B[A\n",
      " 32%|███▏      | 3152/10000 [02:33<05:37, 20.30it/s]\u001B[A\n",
      " 32%|███▏      | 3160/10000 [02:33<05:16, 21.63it/s]\u001B[A\n",
      " 32%|███▏      | 3168/10000 [02:34<05:10, 21.98it/s]\u001B[A\n",
      " 32%|███▏      | 3176/10000 [02:34<04:57, 22.95it/s]\u001B[A\n",
      " 32%|███▏      | 3184/10000 [02:34<04:51, 23.40it/s]\u001B[A\n",
      " 32%|███▏      | 3192/10000 [02:35<05:04, 22.36it/s]\u001B[A\n",
      " 32%|███▏      | 3200/10000 [02:35<04:58, 22.79it/s]\u001B[A\n",
      " 32%|███▏      | 3208/10000 [02:35<04:35, 24.61it/s]\u001B[A\n",
      " 32%|███▏      | 3216/10000 [02:36<04:39, 24.25it/s]\u001B[A\n",
      " 32%|███▏      | 3224/10000 [02:36<04:37, 24.41it/s]\u001B[A\n",
      " 32%|███▏      | 3232/10000 [02:36<04:34, 24.67it/s]\u001B[A\n",
      " 32%|███▏      | 3240/10000 [02:37<04:46, 23.59it/s]\u001B[A\n",
      " 32%|███▏      | 3248/10000 [02:37<04:52, 23.05it/s]\u001B[A\n",
      " 33%|███▎      | 3256/10000 [02:37<05:02, 22.28it/s]\u001B[A\n",
      " 33%|███▎      | 3264/10000 [02:38<04:46, 23.51it/s]\u001B[A\n",
      " 33%|███▎      | 3272/10000 [02:38<04:34, 24.55it/s]\u001B[A\n",
      " 33%|███▎      | 3280/10000 [02:38<04:31, 24.76it/s]\u001B[A\n",
      " 33%|███▎      | 3288/10000 [02:39<04:37, 24.21it/s]\u001B[A\n",
      " 33%|███▎      | 3296/10000 [02:39<04:42, 23.71it/s]\u001B[A\n",
      " 33%|███▎      | 3304/10000 [02:39<04:26, 25.12it/s]\u001B[A\n",
      " 33%|███▎      | 3312/10000 [02:40<04:24, 25.30it/s]\u001B[A\n",
      " 33%|███▎      | 3320/10000 [02:40<04:19, 25.72it/s]\u001B[A\n",
      " 33%|███▎      | 3328/10000 [02:40<04:22, 25.45it/s]\u001B[A\n",
      " 33%|███▎      | 3336/10000 [02:41<04:27, 24.91it/s]\u001B[A\n",
      " 33%|███▎      | 3344/10000 [02:41<04:11, 26.43it/s]\u001B[A\n",
      " 34%|███▎      | 3352/10000 [02:41<04:16, 25.96it/s]\u001B[A\n",
      " 34%|███▎      | 3360/10000 [02:42<04:18, 25.67it/s]\u001B[A\n",
      " 34%|███▎      | 3368/10000 [02:42<04:24, 25.05it/s]\u001B[A\n",
      " 34%|███▍      | 3376/10000 [02:42<04:39, 23.70it/s]\u001B[A\n",
      " 34%|███▍      | 3384/10000 [02:43<04:30, 24.45it/s]\u001B[A\n",
      " 34%|███▍      | 3392/10000 [02:43<04:24, 24.95it/s]\u001B[A\n",
      " 34%|███▍      | 3400/10000 [02:43<04:25, 24.89it/s]\u001B[A\n",
      " 34%|███▍      | 3408/10000 [02:43<04:16, 25.69it/s]\u001B[A\n",
      " 34%|███▍      | 3416/10000 [02:44<04:17, 25.55it/s]\u001B[A\n",
      " 34%|███▍      | 3424/10000 [02:44<04:20, 25.24it/s]\u001B[A\n",
      " 34%|███▍      | 3432/10000 [02:44<04:25, 24.78it/s]\u001B[A\n",
      " 34%|███▍      | 3440/10000 [02:45<04:28, 24.40it/s]\u001B[A\n",
      " 34%|███▍      | 3448/10000 [02:45<04:18, 25.31it/s]\u001B[A\n",
      " 35%|███▍      | 3456/10000 [02:46<04:44, 22.98it/s]\u001B[A\n",
      " 35%|███▍      | 3464/10000 [02:46<04:57, 22.00it/s]\u001B[A\n",
      " 35%|███▍      | 3472/10000 [02:46<04:38, 23.48it/s]\u001B[A\n",
      " 35%|███▍      | 3480/10000 [02:46<04:27, 24.38it/s]\u001B[A\n",
      " 35%|███▍      | 3488/10000 [02:47<04:26, 24.39it/s]\u001B[A\n",
      " 35%|███▍      | 3496/10000 [02:47<04:35, 23.63it/s]\u001B[A\n",
      " 35%|███▌      | 3504/10000 [02:47<04:27, 24.31it/s]\u001B[A\n",
      " 35%|███▌      | 3512/10000 [02:48<04:13, 25.54it/s]\u001B[A\n",
      " 35%|███▌      | 3520/10000 [02:48<04:15, 25.36it/s]\u001B[A\n",
      " 35%|███▌      | 3528/10000 [02:48<04:16, 25.27it/s]\u001B[A\n",
      " 35%|███▌      | 3536/10000 [02:49<04:16, 25.22it/s]\u001B[A\n",
      " 35%|███▌      | 3544/10000 [02:49<04:16, 25.14it/s]\u001B[A\n",
      " 36%|███▌      | 3552/10000 [02:49<04:07, 26.05it/s]\u001B[A\n",
      " 36%|███▌      | 3560/10000 [02:50<04:13, 25.40it/s]\u001B[A\n",
      " 36%|███▌      | 3568/10000 [02:50<04:13, 25.39it/s]\u001B[A\n",
      " 36%|███▌      | 3576/10000 [02:50<04:00, 26.67it/s]\u001B[A\n",
      " 36%|███▌      | 3584/10000 [02:51<05:13, 20.49it/s]\u001B[A\n",
      " 36%|███▌      | 3592/10000 [02:51<05:23, 19.79it/s]\u001B[A\n",
      " 36%|███▌      | 3600/10000 [02:52<04:47, 22.26it/s]\u001B[A\n",
      " 36%|███▌      | 3608/10000 [02:52<05:33, 19.16it/s]\u001B[A\n",
      " 36%|███▌      | 3616/10000 [02:52<05:15, 20.22it/s]\u001B[A\n",
      " 36%|███▌      | 3624/10000 [02:53<05:18, 20.03it/s]\u001B[A\n",
      " 36%|███▋      | 3632/10000 [02:53<05:07, 20.73it/s]\u001B[A\n",
      " 36%|███▋      | 3640/10000 [02:54<04:48, 22.04it/s]\u001B[A\n",
      " 36%|███▋      | 3648/10000 [02:54<04:59, 21.23it/s]\u001B[A\n",
      " 37%|███▋      | 3656/10000 [02:54<04:44, 22.30it/s]\u001B[A\n",
      " 37%|███▋      | 3664/10000 [02:55<04:53, 21.57it/s]\u001B[A\n",
      " 37%|███▋      | 3672/10000 [02:55<04:42, 22.42it/s]\u001B[A\n",
      " 37%|███▋      | 3680/10000 [02:55<04:39, 22.57it/s]\u001B[A\n",
      " 37%|███▋      | 3688/10000 [02:56<04:38, 22.68it/s]\u001B[A\n",
      " 37%|███▋      | 3696/10000 [02:56<04:31, 23.18it/s]\u001B[A\n",
      " 37%|███▋      | 3704/10000 [02:56<05:13, 20.11it/s]\u001B[A\n",
      " 37%|███▋      | 3712/10000 [02:57<04:58, 21.03it/s]\u001B[A\n",
      " 37%|███▋      | 3720/10000 [02:57<05:19, 19.64it/s]\u001B[A\n",
      " 37%|███▋      | 3728/10000 [02:58<04:48, 21.70it/s]\u001B[A\n",
      " 37%|███▋      | 3736/10000 [02:58<04:53, 21.31it/s]\u001B[A\n",
      " 37%|███▋      | 3744/10000 [02:58<04:34, 22.76it/s]\u001B[A\n",
      " 38%|███▊      | 3752/10000 [02:59<04:25, 23.51it/s]\u001B[A\n",
      " 38%|███▊      | 3760/10000 [02:59<04:14, 24.53it/s]\u001B[A\n",
      " 38%|███▊      | 3768/10000 [02:59<04:20, 23.97it/s]\u001B[A\n",
      " 38%|███▊      | 3776/10000 [03:00<04:15, 24.40it/s]\u001B[A\n",
      " 38%|███▊      | 3784/10000 [03:00<04:11, 24.72it/s]\u001B[A\n",
      " 38%|███▊      | 3792/10000 [03:00<04:01, 25.70it/s]\u001B[A\n",
      " 38%|███▊      | 3800/10000 [03:00<04:00, 25.80it/s]\u001B[A\n",
      " 38%|███▊      | 3808/10000 [03:01<04:08, 24.94it/s]\u001B[A\n",
      " 38%|███▊      | 3816/10000 [03:01<04:07, 24.96it/s]\u001B[A\n",
      " 38%|███▊      | 3824/10000 [03:01<04:20, 23.72it/s]\u001B[A\n",
      " 38%|███▊      | 3832/10000 [03:02<04:13, 24.36it/s]\u001B[A\n",
      " 38%|███▊      | 3840/10000 [03:02<04:20, 23.66it/s]\u001B[A\n",
      " 38%|███▊      | 3848/10000 [03:02<04:18, 23.77it/s]\u001B[A\n",
      " 39%|███▊      | 3856/10000 [03:03<04:13, 24.19it/s]\u001B[A\n",
      " 39%|███▊      | 3864/10000 [03:03<04:12, 24.26it/s]\u001B[A\n",
      " 39%|███▊      | 3872/10000 [03:03<04:10, 24.47it/s]\u001B[A\n",
      " 39%|███▉      | 3880/10000 [03:04<04:20, 23.45it/s]\u001B[A\n",
      " 39%|███▉      | 3888/10000 [03:04<04:10, 24.41it/s]\u001B[A\n",
      " 39%|███▉      | 3896/10000 [03:04<04:11, 24.31it/s]\u001B[A\n",
      " 39%|███▉      | 3904/10000 [03:05<04:07, 24.64it/s]\u001B[A\n",
      " 39%|███▉      | 3912/10000 [03:05<04:06, 24.74it/s]\u001B[A\n",
      " 39%|███▉      | 3920/10000 [03:05<04:11, 24.14it/s]\u001B[A\n",
      " 39%|███▉      | 3928/10000 [03:06<04:10, 24.28it/s]\u001B[A\n",
      " 39%|███▉      | 3936/10000 [03:06<04:05, 24.72it/s]\u001B[A\n",
      " 39%|███▉      | 3944/10000 [03:06<04:20, 23.24it/s]\u001B[A\n",
      " 40%|███▉      | 3952/10000 [03:07<04:11, 24.03it/s]\u001B[A\n",
      " 40%|███▉      | 3960/10000 [03:07<04:20, 23.17it/s]\u001B[A\n",
      " 40%|███▉      | 3968/10000 [03:07<04:14, 23.66it/s]\u001B[A\n",
      " 40%|███▉      | 3976/10000 [03:08<04:17, 23.43it/s]\u001B[A\n",
      " 40%|███▉      | 3984/10000 [03:08<04:08, 24.16it/s]\u001B[A\n",
      " 40%|███▉      | 3992/10000 [03:08<04:13, 23.67it/s]\u001B[A\n",
      " 40%|████      | 4000/10000 [03:09<04:16, 23.37it/s]\u001B[A\n",
      " 40%|████      | 4008/10000 [03:09<04:08, 24.08it/s]\u001B[A\n",
      " 40%|████      | 4016/10000 [03:10<04:12, 23.74it/s]\u001B[A\n",
      " 40%|████      | 4024/10000 [03:10<04:07, 24.13it/s]\u001B[A\n",
      " 40%|████      | 4032/10000 [03:10<04:02, 24.64it/s]\u001B[A\n",
      " 40%|████      | 4040/10000 [03:10<04:09, 23.87it/s]\u001B[A\n",
      " 40%|████      | 4048/10000 [03:11<04:09, 23.89it/s]\u001B[A\n",
      " 41%|████      | 4056/10000 [03:11<04:17, 23.13it/s]\u001B[A\n",
      " 41%|████      | 4064/10000 [03:12<04:10, 23.67it/s]\u001B[A\n",
      " 41%|████      | 4072/10000 [03:12<04:09, 23.78it/s]\u001B[A\n",
      " 41%|████      | 4080/10000 [03:12<04:21, 22.67it/s]\u001B[A\n",
      " 41%|████      | 4088/10000 [03:13<04:23, 22.45it/s]\u001B[A\n",
      " 41%|████      | 4096/10000 [03:13<04:08, 23.72it/s]\u001B[A\n",
      " 41%|████      | 4104/10000 [03:13<04:09, 23.62it/s]\u001B[A\n",
      " 41%|████      | 4112/10000 [03:14<04:11, 23.44it/s]\u001B[A\n",
      " 41%|████      | 4120/10000 [03:14<04:10, 23.43it/s]\u001B[A\n",
      " 41%|████▏     | 4128/10000 [03:14<03:58, 24.66it/s]\u001B[A\n",
      " 41%|████▏     | 4136/10000 [03:15<04:01, 24.29it/s]\u001B[A\n",
      " 41%|████▏     | 4144/10000 [03:15<03:54, 24.92it/s]\u001B[A\n",
      " 42%|████▏     | 4152/10000 [03:15<04:04, 23.88it/s]\u001B[A\n",
      " 42%|████▏     | 4160/10000 [03:16<04:06, 23.74it/s]\u001B[A\n",
      " 42%|████▏     | 4168/10000 [03:16<04:05, 23.79it/s]\u001B[A\n",
      " 42%|████▏     | 4176/10000 [03:16<04:18, 22.55it/s]\u001B[A\n",
      " 42%|████▏     | 4184/10000 [03:17<04:12, 23.01it/s]\u001B[A\n",
      " 42%|████▏     | 4192/10000 [03:17<04:15, 22.74it/s]\u001B[A\n",
      " 42%|████▏     | 4200/10000 [03:17<03:59, 24.21it/s]\u001B[A\n",
      " 42%|████▏     | 4208/10000 [03:18<04:06, 23.52it/s]\u001B[A\n",
      " 42%|████▏     | 4216/10000 [03:18<04:03, 23.72it/s]\u001B[A\n",
      " 42%|████▏     | 4224/10000 [03:18<04:11, 22.95it/s]\u001B[A\n",
      " 42%|████▏     | 4232/10000 [03:19<04:17, 22.40it/s]\u001B[A\n",
      " 42%|████▏     | 4240/10000 [03:19<04:53, 19.62it/s]\u001B[A\n",
      " 42%|████▏     | 4248/10000 [03:20<04:31, 21.16it/s]\u001B[A\n",
      " 43%|████▎     | 4256/10000 [03:20<04:18, 22.18it/s]\u001B[A\n",
      " 43%|████▎     | 4264/10000 [03:20<04:27, 21.45it/s]\u001B[A\n",
      " 43%|████▎     | 4272/10000 [03:21<04:14, 22.51it/s]\u001B[A\n",
      " 43%|████▎     | 4280/10000 [03:21<04:14, 22.51it/s]\u001B[A\n",
      " 43%|████▎     | 4288/10000 [03:21<04:30, 21.10it/s]\u001B[A\n",
      " 43%|████▎     | 4296/10000 [03:22<04:31, 20.99it/s]\u001B[A\n",
      " 43%|████▎     | 4304/10000 [03:22<04:29, 21.11it/s]\u001B[A\n",
      " 43%|████▎     | 4312/10000 [03:23<04:28, 21.20it/s]\u001B[A\n",
      " 43%|████▎     | 4320/10000 [03:23<04:20, 21.79it/s]\u001B[A\n",
      " 43%|████▎     | 4328/10000 [03:23<04:27, 21.20it/s]\u001B[A\n",
      " 43%|████▎     | 4336/10000 [03:24<04:32, 20.81it/s]\u001B[A\n",
      " 43%|████▎     | 4344/10000 [03:24<04:19, 21.76it/s]\u001B[A\n",
      " 44%|████▎     | 4352/10000 [03:24<04:17, 21.94it/s]\u001B[A\n",
      " 44%|████▎     | 4360/10000 [03:25<04:10, 22.56it/s]\u001B[A\n",
      " 44%|████▎     | 4368/10000 [03:25<04:22, 21.42it/s]\u001B[A\n",
      " 44%|████▍     | 4376/10000 [03:25<04:11, 22.37it/s]\u001B[A\n",
      " 44%|████▍     | 4384/10000 [03:26<04:14, 22.05it/s]\u001B[A\n",
      " 44%|████▍     | 4392/10000 [03:26<04:12, 22.24it/s]\u001B[A\n",
      " 44%|████▍     | 4400/10000 [03:27<04:15, 21.95it/s]\u001B[A\n",
      " 44%|████▍     | 4408/10000 [03:27<04:15, 21.87it/s]\u001B[A\n",
      " 44%|████▍     | 4416/10000 [03:27<04:06, 22.67it/s]\u001B[A\n",
      " 44%|████▍     | 4424/10000 [03:28<04:10, 22.26it/s]\u001B[A\n",
      " 44%|████▍     | 4432/10000 [03:28<04:03, 22.86it/s]\u001B[A\n",
      " 44%|████▍     | 4440/10000 [03:28<04:05, 22.69it/s]\u001B[A\n",
      " 44%|████▍     | 4448/10000 [03:29<04:08, 22.38it/s]\u001B[A\n",
      " 45%|████▍     | 4456/10000 [03:29<03:55, 23.58it/s]\u001B[A\n",
      " 45%|████▍     | 4464/10000 [03:29<04:06, 22.49it/s]\u001B[A\n",
      " 45%|████▍     | 4472/10000 [03:30<05:03, 18.22it/s]\u001B[A\n",
      " 45%|████▍     | 4480/10000 [03:30<04:21, 21.07it/s]\u001B[A\n",
      " 45%|████▍     | 4488/10000 [03:31<04:27, 20.58it/s]\u001B[A\n",
      " 45%|████▍     | 4496/10000 [03:31<04:13, 21.69it/s]\u001B[A\n",
      " 45%|████▌     | 4504/10000 [03:31<04:09, 22.01it/s]\u001B[A\n",
      " 45%|████▌     | 4512/10000 [03:32<04:18, 21.21it/s]\u001B[A\n",
      " 45%|████▌     | 4520/10000 [03:32<04:36, 19.83it/s]\u001B[A\n",
      " 45%|████▌     | 4528/10000 [03:33<06:28, 14.09it/s]\u001B[A\n",
      " 45%|████▌     | 4536/10000 [03:33<05:28, 16.65it/s]\u001B[A\n",
      " 45%|████▌     | 4544/10000 [03:34<05:10, 17.59it/s]\u001B[A\n",
      " 46%|████▌     | 4552/10000 [03:34<04:59, 18.16it/s]\u001B[A\n",
      " 46%|████▌     | 4560/10000 [03:35<04:45, 19.03it/s]\u001B[A\n",
      " 46%|████▌     | 4568/10000 [03:35<04:40, 19.38it/s]\u001B[A\n",
      " 46%|████▌     | 4576/10000 [03:35<04:37, 19.51it/s]\u001B[A\n",
      " 46%|████▌     | 4584/10000 [03:36<04:27, 20.24it/s]\u001B[A\n",
      " 46%|████▌     | 4592/10000 [03:36<04:13, 21.34it/s]\u001B[A\n",
      " 46%|████▌     | 4600/10000 [03:36<04:28, 20.12it/s]\u001B[A\n",
      " 46%|████▌     | 4608/10000 [03:37<04:26, 20.26it/s]\u001B[A\n",
      " 46%|████▌     | 4616/10000 [03:37<04:24, 20.36it/s]\u001B[A\n",
      " 46%|████▌     | 4624/10000 [03:38<04:46, 18.77it/s]\u001B[A\n",
      " 46%|████▋     | 4632/10000 [03:38<04:18, 20.77it/s]\u001B[A\n",
      " 46%|████▋     | 4640/10000 [03:38<04:19, 20.69it/s]\u001B[A\n",
      " 46%|████▋     | 4648/10000 [03:39<04:16, 20.83it/s]\u001B[A\n",
      " 47%|████▋     | 4656/10000 [03:39<04:20, 20.54it/s]\u001B[A\n",
      " 47%|████▋     | 4664/10000 [03:40<04:17, 20.75it/s]\u001B[A\n",
      " 47%|████▋     | 4672/10000 [03:40<04:12, 21.13it/s]\u001B[A\n",
      " 47%|████▋     | 4680/10000 [03:40<04:18, 20.55it/s]\u001B[A\n",
      " 47%|████▋     | 4688/10000 [03:41<04:49, 18.34it/s]\u001B[A\n",
      " 47%|████▋     | 4696/10000 [03:41<04:31, 19.56it/s]\u001B[A\n",
      " 47%|████▋     | 4704/10000 [03:42<04:19, 20.39it/s]\u001B[A\n",
      " 47%|████▋     | 4712/10000 [03:42<04:25, 19.91it/s]\u001B[A\n",
      " 47%|████▋     | 4720/10000 [03:42<04:15, 20.69it/s]\u001B[A\n",
      " 47%|████▋     | 4728/10000 [03:43<04:44, 18.52it/s]\u001B[A\n",
      " 47%|████▋     | 4736/10000 [03:43<04:24, 19.91it/s]\u001B[A\n",
      " 47%|████▋     | 4744/10000 [03:44<04:14, 20.65it/s]\u001B[A\n",
      " 48%|████▊     | 4752/10000 [03:44<04:10, 20.96it/s]\u001B[A\n",
      " 48%|████▊     | 4760/10000 [03:44<04:07, 21.19it/s]\u001B[A\n",
      " 48%|████▊     | 4768/10000 [03:45<04:23, 19.88it/s]\u001B[A\n",
      " 48%|████▊     | 4776/10000 [03:45<04:00, 21.76it/s]\u001B[A\n",
      " 48%|████▊     | 4784/10000 [03:47<07:55, 10.96it/s]\u001B[A\n",
      " 48%|████▊     | 4792/10000 [03:48<10:19,  8.40it/s]\u001B[A\n",
      " 48%|████▊     | 4800/10000 [03:49<09:17,  9.33it/s]\u001B[A\n",
      " 48%|████▊     | 4808/10000 [03:51<12:16,  7.05it/s]\u001B[A\n",
      " 48%|████▊     | 4816/10000 [03:52<11:47,  7.33it/s]\u001B[A\n",
      " 48%|████▊     | 4824/10000 [03:52<09:59,  8.64it/s]\u001B[A\n",
      " 48%|████▊     | 4832/10000 [03:52<08:12, 10.50it/s]\u001B[A\n",
      " 48%|████▊     | 4840/10000 [03:53<07:06, 12.11it/s]\u001B[A\n",
      " 48%|████▊     | 4848/10000 [03:53<06:06, 14.05it/s]\u001B[A\n",
      " 49%|████▊     | 4856/10000 [03:54<05:39, 15.16it/s]\u001B[A\n",
      " 49%|████▊     | 4864/10000 [03:54<04:59, 17.17it/s]\u001B[A\n",
      " 49%|████▊     | 4872/10000 [03:54<04:48, 17.77it/s]\u001B[A\n",
      " 49%|████▉     | 4880/10000 [03:55<04:40, 18.28it/s]\u001B[A\n",
      " 49%|████▉     | 4888/10000 [03:55<04:39, 18.26it/s]\u001B[A\n",
      " 49%|████▉     | 4896/10000 [03:56<04:22, 19.44it/s]\u001B[A\n",
      " 49%|████▉     | 4904/10000 [03:56<04:12, 20.16it/s]\u001B[A\n",
      " 49%|████▉     | 4912/10000 [03:56<04:07, 20.60it/s]\u001B[A\n",
      " 49%|████▉     | 4920/10000 [03:57<04:13, 20.07it/s]\u001B[A\n",
      " 49%|████▉     | 4928/10000 [03:57<04:06, 20.54it/s]\u001B[A\n",
      " 49%|████▉     | 4936/10000 [03:58<04:12, 20.07it/s]\u001B[A\n",
      " 49%|████▉     | 4944/10000 [03:58<04:06, 20.47it/s]\u001B[A\n",
      " 50%|████▉     | 4952/10000 [03:58<04:02, 20.78it/s]\u001B[A\n",
      " 50%|████▉     | 4960/10000 [03:59<03:55, 21.39it/s]\u001B[A\n",
      " 50%|████▉     | 4968/10000 [03:59<04:01, 20.79it/s]\u001B[A\n",
      " 50%|████▉     | 4976/10000 [03:59<03:39, 22.88it/s]\u001B[A\n",
      " 50%|████▉     | 4984/10000 [04:00<03:55, 21.33it/s]\u001B[A\n",
      " 50%|████▉     | 4992/10000 [04:00<03:55, 21.24it/s]\u001B[A\n",
      " 50%|█████     | 5000/10000 [04:00<03:41, 22.56it/s]\u001B[A\n",
      " 50%|█████     | 5008/10000 [04:01<03:57, 21.00it/s]\u001B[A\n",
      " 50%|█████     | 5016/10000 [04:01<03:59, 20.85it/s]\u001B[A\n",
      " 50%|█████     | 5024/10000 [04:02<03:46, 21.99it/s]\u001B[A\n",
      " 50%|█████     | 5032/10000 [04:02<03:59, 20.71it/s]\u001B[A\n",
      " 50%|█████     | 5040/10000 [04:02<03:42, 22.32it/s]\u001B[A\n",
      " 50%|█████     | 5048/10000 [04:03<03:52, 21.34it/s]\u001B[A\n",
      " 51%|█████     | 5056/10000 [04:03<03:48, 21.60it/s]\u001B[A\n",
      " 51%|█████     | 5064/10000 [04:03<03:45, 21.91it/s]\u001B[A\n",
      " 51%|█████     | 5072/10000 [04:04<03:39, 22.48it/s]\u001B[A\n",
      " 51%|█████     | 5080/10000 [04:04<03:42, 22.13it/s]\u001B[A\n",
      " 51%|█████     | 5088/10000 [04:04<03:36, 22.71it/s]\u001B[A\n",
      " 51%|█████     | 5096/10000 [04:05<03:31, 23.15it/s]\u001B[A\n",
      " 51%|█████     | 5104/10000 [04:05<03:46, 21.58it/s]\u001B[A\n",
      " 51%|█████     | 5112/10000 [04:06<03:31, 23.12it/s]\u001B[A\n",
      " 51%|█████     | 5120/10000 [04:06<03:32, 22.97it/s]\u001B[A\n",
      " 51%|█████▏    | 5128/10000 [04:06<03:39, 22.21it/s]\u001B[A\n",
      " 51%|█████▏    | 5136/10000 [04:07<03:36, 22.45it/s]\u001B[A\n",
      " 51%|█████▏    | 5144/10000 [04:07<03:23, 23.80it/s]\u001B[A\n",
      " 52%|█████▏    | 5152/10000 [04:07<03:42, 21.77it/s]\u001B[A\n",
      " 52%|█████▏    | 5160/10000 [04:08<03:28, 23.23it/s]\u001B[A\n",
      " 52%|█████▏    | 5168/10000 [04:08<03:26, 23.42it/s]\u001B[A\n",
      " 52%|█████▏    | 5176/10000 [04:08<03:29, 23.04it/s]\u001B[A\n",
      " 52%|█████▏    | 5184/10000 [04:09<03:51, 20.78it/s]\u001B[A\n",
      " 52%|█████▏    | 5192/10000 [04:09<03:49, 20.93it/s]\u001B[A\n",
      " 52%|█████▏    | 5200/10000 [04:09<03:33, 22.49it/s]\u001B[A\n",
      " 52%|█████▏    | 5208/10000 [04:10<03:38, 21.90it/s]\u001B[A\n",
      " 52%|█████▏    | 5216/10000 [04:10<03:45, 21.25it/s]\u001B[A\n",
      " 52%|█████▏    | 5224/10000 [04:11<03:31, 22.61it/s]\u001B[A\n",
      " 52%|█████▏    | 5232/10000 [04:11<03:29, 22.80it/s]\u001B[A\n",
      " 52%|█████▏    | 5240/10000 [04:11<03:29, 22.77it/s]\u001B[A\n",
      " 52%|█████▏    | 5248/10000 [04:12<03:18, 23.89it/s]\u001B[A\n",
      " 53%|█████▎    | 5256/10000 [04:12<03:23, 23.28it/s]\u001B[A\n",
      " 53%|█████▎    | 5264/10000 [04:12<03:32, 22.29it/s]\u001B[A\n",
      " 53%|█████▎    | 5272/10000 [04:13<03:27, 22.75it/s]\u001B[A\n",
      " 53%|█████▎    | 5280/10000 [04:13<03:23, 23.21it/s]\u001B[A\n",
      " 53%|█████▎    | 5288/10000 [04:13<03:21, 23.41it/s]\u001B[A\n",
      " 53%|█████▎    | 5296/10000 [04:14<03:27, 22.68it/s]\u001B[A\n",
      " 53%|█████▎    | 5304/10000 [04:14<03:15, 24.07it/s]\u001B[A\n",
      " 53%|█████▎    | 5312/10000 [04:14<03:24, 22.97it/s]\u001B[A\n",
      " 53%|█████▎    | 5320/10000 [04:15<03:28, 22.48it/s]\u001B[A\n",
      " 53%|█████▎    | 5328/10000 [04:15<03:23, 22.93it/s]\u001B[A\n",
      " 53%|█████▎    | 5336/10000 [04:15<03:15, 23.89it/s]\u001B[A\n",
      " 53%|█████▎    | 5344/10000 [04:16<03:10, 24.38it/s]\u001B[A\n",
      " 54%|█████▎    | 5352/10000 [04:16<03:09, 24.50it/s]\u001B[A\n",
      " 54%|█████▎    | 5360/10000 [04:16<03:18, 23.41it/s]\u001B[A\n",
      " 54%|█████▎    | 5368/10000 [04:17<03:16, 23.57it/s]\u001B[A\n",
      " 54%|█████▍    | 5376/10000 [04:17<03:11, 24.09it/s]\u001B[A\n",
      " 54%|█████▍    | 5384/10000 [04:17<03:14, 23.71it/s]\u001B[A\n",
      " 54%|█████▍    | 5392/10000 [04:18<03:07, 24.62it/s]\u001B[A\n",
      " 54%|█████▍    | 5400/10000 [04:18<03:11, 24.07it/s]\u001B[A\n",
      " 54%|█████▍    | 5408/10000 [04:18<03:00, 25.48it/s]\u001B[A\n",
      " 54%|█████▍    | 5416/10000 [04:19<03:08, 24.29it/s]\u001B[A\n",
      " 54%|█████▍    | 5424/10000 [04:19<03:07, 24.38it/s]\u001B[A\n",
      " 54%|█████▍    | 5432/10000 [04:19<03:03, 24.93it/s]\u001B[A\n",
      " 54%|█████▍    | 5440/10000 [04:20<02:57, 25.69it/s]\u001B[A\n",
      " 54%|█████▍    | 5448/10000 [04:20<03:07, 24.26it/s]\u001B[A\n",
      " 55%|█████▍    | 5456/10000 [04:20<03:05, 24.54it/s]\u001B[A\n",
      " 55%|█████▍    | 5464/10000 [04:21<03:13, 23.42it/s]\u001B[A\n",
      " 55%|█████▍    | 5472/10000 [04:21<03:05, 24.37it/s]\u001B[A\n",
      " 55%|█████▍    | 5480/10000 [04:21<03:09, 23.80it/s]\u001B[A\n",
      " 55%|█████▍    | 5488/10000 [04:22<03:00, 24.99it/s]\u001B[A\n",
      " 55%|█████▍    | 5496/10000 [04:22<03:17, 22.78it/s]\u001B[A\n",
      " 55%|█████▌    | 5504/10000 [04:22<03:18, 22.66it/s]\u001B[A\n",
      " 55%|█████▌    | 5512/10000 [04:23<03:09, 23.72it/s]\u001B[A\n",
      " 55%|█████▌    | 5520/10000 [04:23<03:11, 23.38it/s]\u001B[A\n",
      " 55%|█████▌    | 5528/10000 [04:23<03:03, 24.40it/s]\u001B[A\n",
      " 55%|█████▌    | 5536/10000 [04:24<03:18, 22.46it/s]\u001B[A\n",
      " 55%|█████▌    | 5544/10000 [04:24<03:12, 23.18it/s]\u001B[A\n",
      " 56%|█████▌    | 5552/10000 [04:24<03:05, 23.98it/s]\u001B[A\n",
      " 56%|█████▌    | 5560/10000 [04:25<03:08, 23.53it/s]\u001B[A\n",
      " 56%|█████▌    | 5568/10000 [04:25<03:08, 23.52it/s]\u001B[A\n",
      " 56%|█████▌    | 5576/10000 [04:25<03:13, 22.91it/s]\u001B[A\n",
      " 56%|█████▌    | 5584/10000 [04:26<03:12, 22.90it/s]\u001B[A\n",
      " 56%|█████▌    | 5592/10000 [04:26<03:24, 21.50it/s]\u001B[A\n",
      " 56%|█████▌    | 5600/10000 [04:26<03:08, 23.37it/s]\u001B[A\n",
      " 56%|█████▌    | 5608/10000 [04:27<03:14, 22.57it/s]\u001B[A\n",
      " 56%|█████▌    | 5616/10000 [04:27<03:01, 24.16it/s]\u001B[A\n",
      " 56%|█████▌    | 5624/10000 [04:27<03:02, 24.01it/s]\u001B[A\n",
      " 56%|█████▋    | 5632/10000 [04:28<03:25, 21.30it/s]\u001B[A\n",
      " 56%|█████▋    | 5640/10000 [04:28<03:16, 22.19it/s]\u001B[A\n",
      " 56%|█████▋    | 5648/10000 [04:29<03:28, 20.86it/s]\u001B[A\n",
      " 57%|█████▋    | 5656/10000 [04:29<03:10, 22.85it/s]\u001B[A\n",
      " 57%|█████▋    | 5664/10000 [04:29<03:10, 22.80it/s]\u001B[A\n",
      " 57%|█████▋    | 5672/10000 [04:30<03:01, 23.89it/s]\u001B[A\n",
      " 57%|█████▋    | 5680/10000 [04:30<03:03, 23.59it/s]\u001B[A\n",
      " 57%|█████▋    | 5688/10000 [04:30<02:57, 24.34it/s]\u001B[A\n",
      " 57%|█████▋    | 5696/10000 [04:31<03:14, 22.10it/s]\u001B[A\n",
      " 57%|█████▋    | 5704/10000 [04:31<03:19, 21.59it/s]\u001B[A\n",
      " 57%|█████▋    | 5712/10000 [04:31<03:16, 21.84it/s]\u001B[A\n",
      " 57%|█████▋    | 5720/10000 [04:32<03:04, 23.24it/s]\u001B[A\n",
      " 57%|█████▋    | 5728/10000 [04:32<02:57, 24.06it/s]\u001B[A\n",
      " 57%|█████▋    | 5736/10000 [04:32<03:00, 23.64it/s]\u001B[A\n",
      " 57%|█████▋    | 5744/10000 [04:33<03:05, 22.97it/s]\u001B[A\n",
      " 58%|█████▊    | 5752/10000 [04:33<03:16, 21.66it/s]\u001B[A\n",
      " 58%|█████▊    | 5760/10000 [04:33<02:54, 24.37it/s]\u001B[A\n",
      " 58%|█████▊    | 5768/10000 [04:34<02:51, 24.71it/s]\u001B[A\n",
      " 58%|█████▊    | 5776/10000 [04:34<02:51, 24.60it/s]\u001B[A\n",
      " 58%|█████▊    | 5784/10000 [04:34<02:52, 24.49it/s]\u001B[A\n",
      " 58%|█████▊    | 5792/10000 [04:35<02:57, 23.76it/s]\u001B[A\n",
      " 58%|█████▊    | 5800/10000 [04:35<02:50, 24.60it/s]\u001B[A\n",
      " 58%|█████▊    | 5808/10000 [04:35<02:44, 25.43it/s]\u001B[A\n",
      " 58%|█████▊    | 5816/10000 [04:36<02:49, 24.66it/s]\u001B[A\n",
      " 58%|█████▊    | 5824/10000 [04:36<03:07, 22.30it/s]\u001B[A\n",
      " 58%|█████▊    | 5832/10000 [04:36<02:54, 23.92it/s]\u001B[A\n",
      " 58%|█████▊    | 5840/10000 [04:37<02:50, 24.38it/s]\u001B[A\n",
      " 58%|█████▊    | 5848/10000 [04:37<02:46, 24.89it/s]\u001B[A\n",
      " 59%|█████▊    | 5856/10000 [04:37<02:46, 24.82it/s]\u001B[A\n",
      " 59%|█████▊    | 5864/10000 [04:38<02:45, 24.94it/s]\u001B[A\n",
      " 59%|█████▊    | 5872/10000 [04:38<02:44, 25.11it/s]\u001B[A\n",
      " 59%|█████▉    | 5880/10000 [04:39<03:10, 21.62it/s]\u001B[A\n",
      " 59%|█████▉    | 5888/10000 [04:39<03:00, 22.75it/s]\u001B[A\n",
      " 59%|█████▉    | 5896/10000 [04:39<02:51, 23.99it/s]\u001B[A\n",
      " 59%|█████▉    | 5904/10000 [04:39<02:46, 24.65it/s]\u001B[A\n",
      " 59%|█████▉    | 5912/10000 [04:40<02:51, 23.90it/s]\u001B[A\n",
      " 59%|█████▉    | 5920/10000 [04:40<03:49, 17.79it/s]\u001B[A\n",
      " 59%|█████▉    | 5928/10000 [04:41<03:43, 18.25it/s]\u001B[A\n",
      " 59%|█████▉    | 5936/10000 [04:41<03:25, 19.80it/s]\u001B[A\n",
      " 59%|█████▉    | 5944/10000 [04:42<03:42, 18.23it/s]\u001B[A\n",
      " 60%|█████▉    | 5952/10000 [04:42<03:47, 17.78it/s]\u001B[A\n",
      " 60%|█████▉    | 5960/10000 [04:43<03:35, 18.74it/s]\u001B[A\n",
      " 60%|█████▉    | 5968/10000 [04:43<03:33, 18.86it/s]\u001B[A\n",
      " 60%|█████▉    | 5976/10000 [04:43<03:11, 21.05it/s]\u001B[A\n",
      " 60%|█████▉    | 5984/10000 [04:44<02:56, 22.71it/s]\u001B[A\n",
      " 60%|█████▉    | 5992/10000 [04:44<03:00, 22.18it/s]\u001B[A\n",
      " 60%|██████    | 6000/10000 [04:44<02:51, 23.32it/s]\u001B[A\n",
      " 60%|██████    | 6008/10000 [04:45<02:57, 22.53it/s]\u001B[A\n",
      " 60%|██████    | 6016/10000 [04:45<02:47, 23.74it/s]\u001B[A\n",
      " 60%|██████    | 6024/10000 [04:45<02:45, 24.09it/s]\u001B[A\n",
      " 60%|██████    | 6032/10000 [04:46<02:40, 24.68it/s]\u001B[A\n",
      " 60%|██████    | 6040/10000 [04:46<02:41, 24.59it/s]\u001B[A\n",
      " 60%|██████    | 6048/10000 [04:46<02:39, 24.84it/s]\u001B[A\n",
      " 61%|██████    | 6056/10000 [04:47<02:44, 23.92it/s]\u001B[A\n",
      " 61%|██████    | 6064/10000 [04:47<02:40, 24.59it/s]\u001B[A\n",
      " 61%|██████    | 6072/10000 [04:47<02:38, 24.74it/s]\u001B[A\n",
      " 61%|██████    | 6080/10000 [04:48<02:37, 24.97it/s]\u001B[A\n",
      " 61%|██████    | 6088/10000 [04:48<02:41, 24.17it/s]\u001B[A\n",
      " 61%|██████    | 6096/10000 [04:48<02:34, 25.31it/s]\u001B[A\n",
      " 61%|██████    | 6104/10000 [04:48<02:38, 24.61it/s]\u001B[A\n",
      " 61%|██████    | 6112/10000 [04:49<02:35, 24.97it/s]\u001B[A\n",
      " 61%|██████    | 6120/10000 [04:49<02:39, 24.31it/s]\u001B[A\n",
      " 61%|██████▏   | 6128/10000 [04:49<02:36, 24.76it/s]\u001B[A\n",
      " 61%|██████▏   | 6136/10000 [04:50<02:34, 25.02it/s]\u001B[A\n",
      " 61%|██████▏   | 6144/10000 [04:50<02:32, 25.28it/s]\u001B[A\n",
      " 62%|██████▏   | 6152/10000 [04:50<02:33, 25.13it/s]\u001B[A\n",
      " 62%|██████▏   | 6160/10000 [04:51<02:36, 24.50it/s]\u001B[A\n",
      " 62%|██████▏   | 6168/10000 [04:51<02:32, 25.17it/s]\u001B[A\n",
      " 62%|██████▏   | 6176/10000 [04:51<02:35, 24.55it/s]\u001B[A\n",
      " 62%|██████▏   | 6184/10000 [04:52<02:44, 23.13it/s]\u001B[A\n",
      " 62%|██████▏   | 6192/10000 [04:52<02:47, 22.75it/s]\u001B[A\n",
      " 62%|██████▏   | 6200/10000 [04:52<02:40, 23.65it/s]\u001B[A\n",
      " 62%|██████▏   | 6208/10000 [04:53<02:42, 23.35it/s]\u001B[A\n",
      " 62%|██████▏   | 6216/10000 [04:53<02:48, 22.48it/s]\u001B[A\n",
      " 62%|██████▏   | 6224/10000 [04:54<02:48, 22.39it/s]\u001B[A\n",
      " 62%|██████▏   | 6232/10000 [04:54<02:46, 22.66it/s]\u001B[A\n",
      " 62%|██████▏   | 6240/10000 [04:54<02:44, 22.81it/s]\u001B[A\n",
      " 62%|██████▏   | 6248/10000 [04:55<02:42, 23.07it/s]\u001B[A\n",
      " 63%|██████▎   | 6256/10000 [04:55<02:45, 22.59it/s]\u001B[A\n",
      " 63%|██████▎   | 6264/10000 [04:55<02:38, 23.60it/s]\u001B[A\n",
      " 63%|██████▎   | 6272/10000 [04:56<02:33, 24.32it/s]\u001B[A\n",
      " 63%|██████▎   | 6280/10000 [04:56<02:33, 24.22it/s]\u001B[A\n",
      " 63%|██████▎   | 6288/10000 [04:56<02:29, 24.85it/s]\u001B[A\n",
      " 63%|██████▎   | 6296/10000 [04:56<02:25, 25.52it/s]\u001B[A\n",
      " 63%|██████▎   | 6304/10000 [04:57<02:31, 24.39it/s]\u001B[A\n",
      " 63%|██████▎   | 6312/10000 [04:57<02:32, 24.14it/s]\u001B[A\n",
      " 63%|██████▎   | 6320/10000 [04:57<02:27, 24.87it/s]\u001B[A\n",
      " 63%|██████▎   | 6328/10000 [04:58<02:31, 24.31it/s]\u001B[A\n",
      " 63%|██████▎   | 6336/10000 [04:58<02:34, 23.78it/s]\u001B[A\n",
      " 63%|██████▎   | 6344/10000 [04:59<02:30, 24.33it/s]\u001B[A\n",
      " 64%|██████▎   | 6352/10000 [04:59<02:28, 24.58it/s]\u001B[A\n",
      " 64%|██████▎   | 6360/10000 [04:59<02:38, 23.02it/s]\u001B[A\n",
      " 64%|██████▎   | 6368/10000 [05:00<02:33, 23.64it/s]\u001B[A\n",
      " 64%|██████▍   | 6376/10000 [05:00<02:29, 24.18it/s]\u001B[A\n",
      " 64%|██████▍   | 6384/10000 [05:00<02:29, 24.26it/s]\u001B[A\n",
      " 64%|██████▍   | 6392/10000 [05:01<02:37, 22.97it/s]\u001B[A\n",
      " 64%|██████▍   | 6400/10000 [05:01<02:24, 24.89it/s]\u001B[A\n",
      " 64%|██████▍   | 6408/10000 [05:01<02:24, 24.83it/s]\u001B[A\n",
      " 64%|██████▍   | 6416/10000 [05:02<02:27, 24.25it/s]\u001B[A\n",
      " 64%|██████▍   | 6424/10000 [05:02<02:25, 24.56it/s]\u001B[A\n",
      " 64%|██████▍   | 6432/10000 [05:02<02:19, 25.51it/s]\u001B[A\n",
      " 64%|██████▍   | 6440/10000 [05:02<02:21, 25.20it/s]\u001B[A\n",
      " 64%|██████▍   | 6448/10000 [05:03<02:19, 25.37it/s]\u001B[A\n",
      " 65%|██████▍   | 6456/10000 [05:03<02:21, 25.11it/s]\u001B[A\n",
      " 65%|██████▍   | 6464/10000 [05:03<02:27, 24.04it/s]\u001B[A\n",
      " 65%|██████▍   | 6472/10000 [05:04<02:23, 24.61it/s]\u001B[A\n",
      " 65%|██████▍   | 6480/10000 [05:04<02:21, 24.90it/s]\u001B[A\n",
      " 65%|██████▍   | 6488/10000 [05:04<02:25, 24.19it/s]\u001B[A\n",
      " 65%|██████▍   | 6496/10000 [05:05<02:24, 24.30it/s]\u001B[A\n",
      " 65%|██████▌   | 6504/10000 [05:05<02:38, 22.12it/s]\u001B[A\n",
      " 65%|██████▌   | 6512/10000 [05:05<02:18, 25.11it/s]\u001B[A\n",
      " 65%|██████▌   | 6520/10000 [05:06<02:20, 24.77it/s]\u001B[A\n",
      " 65%|██████▌   | 6528/10000 [05:06<02:21, 24.48it/s]\u001B[A\n",
      " 65%|██████▌   | 6536/10000 [05:06<02:21, 24.41it/s]\u001B[A\n",
      " 65%|██████▌   | 6544/10000 [05:07<02:19, 24.85it/s]\u001B[A\n",
      " 66%|██████▌   | 6552/10000 [05:07<02:22, 24.27it/s]\u001B[A\n",
      " 66%|██████▌   | 6560/10000 [05:07<02:21, 24.24it/s]\u001B[A\n",
      " 66%|██████▌   | 6568/10000 [05:08<02:25, 23.58it/s]\u001B[A\n",
      " 66%|██████▌   | 6576/10000 [05:08<02:21, 24.24it/s]\u001B[A\n",
      " 66%|██████▌   | 6584/10000 [05:08<02:22, 24.02it/s]\u001B[A\n",
      " 66%|██████▌   | 6592/10000 [05:09<02:17, 24.84it/s]\u001B[A\n",
      " 66%|██████▌   | 6600/10000 [05:09<02:22, 23.86it/s]\u001B[A\n",
      " 66%|██████▌   | 6608/10000 [05:09<02:21, 23.96it/s]\u001B[A\n",
      " 66%|██████▌   | 6616/10000 [05:10<02:18, 24.41it/s]\u001B[A\n",
      " 66%|██████▌   | 6624/10000 [05:10<02:19, 24.13it/s]\u001B[A\n",
      " 66%|██████▋   | 6632/10000 [05:10<02:22, 23.58it/s]\u001B[A\n",
      " 66%|██████▋   | 6640/10000 [05:11<02:18, 24.34it/s]\u001B[A\n",
      " 66%|██████▋   | 6648/10000 [05:11<02:20, 23.85it/s]\u001B[A\n",
      " 67%|██████▋   | 6656/10000 [05:11<02:15, 24.68it/s]\u001B[A\n",
      " 67%|██████▋   | 6664/10000 [05:12<02:15, 24.71it/s]\u001B[A\n",
      " 67%|██████▋   | 6672/10000 [05:12<02:15, 24.56it/s]\u001B[A\n",
      " 67%|██████▋   | 6680/10000 [05:12<02:25, 22.87it/s]\u001B[A\n",
      " 67%|██████▋   | 6688/10000 [05:13<02:21, 23.43it/s]\u001B[A\n",
      " 67%|██████▋   | 6696/10000 [05:13<02:16, 24.15it/s]\u001B[A\n",
      " 67%|██████▋   | 6704/10000 [05:13<02:17, 24.04it/s]\u001B[A\n",
      " 67%|██████▋   | 6712/10000 [05:14<02:14, 24.41it/s]\u001B[A\n",
      " 67%|██████▋   | 6720/10000 [05:14<02:19, 23.59it/s]\u001B[A\n",
      " 67%|██████▋   | 6728/10000 [05:14<02:15, 24.12it/s]\u001B[A\n",
      " 67%|██████▋   | 6736/10000 [05:15<02:15, 24.12it/s]\u001B[A\n",
      " 67%|██████▋   | 6744/10000 [05:15<02:18, 23.58it/s]\u001B[A\n",
      " 68%|██████▊   | 6752/10000 [05:15<02:23, 22.60it/s]\u001B[A\n",
      " 68%|██████▊   | 6760/10000 [05:16<02:22, 22.67it/s]\u001B[A\n",
      " 68%|██████▊   | 6768/10000 [05:16<02:18, 23.40it/s]\u001B[A\n",
      " 68%|██████▊   | 6776/10000 [05:16<02:21, 22.79it/s]\u001B[A\n",
      " 68%|██████▊   | 6784/10000 [05:18<04:08, 12.93it/s]\u001B[A\n",
      " 68%|██████▊   | 6792/10000 [05:19<04:29, 11.92it/s]\u001B[A\n",
      " 68%|██████▊   | 6800/10000 [05:19<04:21, 12.23it/s]\u001B[A\n",
      " 68%|██████▊   | 6808/10000 [05:20<04:08, 12.85it/s]\u001B[A\n",
      " 68%|██████▊   | 6816/10000 [05:20<04:07, 12.87it/s]\u001B[A\n",
      " 68%|██████▊   | 6824/10000 [05:21<03:51, 13.71it/s]\u001B[A\n",
      " 68%|██████▊   | 6832/10000 [05:21<03:48, 13.85it/s]\u001B[A\n",
      " 68%|██████▊   | 6840/10000 [05:22<03:49, 13.76it/s]\u001B[A\n",
      " 68%|██████▊   | 6848/10000 [05:23<03:49, 13.75it/s]\u001B[A\n",
      " 69%|██████▊   | 6856/10000 [05:23<04:20, 12.06it/s]\u001B[A\n",
      " 69%|██████▊   | 6864/10000 [05:24<04:03, 12.89it/s]\u001B[A\n",
      " 69%|██████▊   | 6872/10000 [05:24<03:47, 13.78it/s]\u001B[A\n",
      " 69%|██████▉   | 6880/10000 [05:25<03:59, 13.05it/s]\u001B[A\n",
      " 69%|██████▉   | 6888/10000 [05:26<03:45, 13.80it/s]\u001B[A\n",
      " 69%|██████▉   | 6896/10000 [05:26<03:39, 14.15it/s]\u001B[A\n",
      " 69%|██████▉   | 6904/10000 [05:27<03:56, 13.09it/s]\u001B[A\n",
      " 69%|██████▉   | 6912/10000 [05:27<03:39, 14.06it/s]\u001B[A\n",
      " 69%|██████▉   | 6920/10000 [05:28<03:21, 15.28it/s]\u001B[A\n",
      " 69%|██████▉   | 6928/10000 [05:28<03:09, 16.17it/s]\u001B[A\n",
      " 69%|██████▉   | 6936/10000 [05:29<02:55, 17.47it/s]\u001B[A\n",
      " 69%|██████▉   | 6944/10000 [05:29<02:44, 18.57it/s]\u001B[A\n",
      " 70%|██████▉   | 6952/10000 [05:29<02:40, 18.97it/s]\u001B[A\n",
      " 70%|██████▉   | 6960/10000 [05:30<02:31, 20.07it/s]\u001B[A\n",
      " 70%|██████▉   | 6968/10000 [05:30<02:31, 20.00it/s]\u001B[A\n",
      " 70%|██████▉   | 6976/10000 [05:30<02:31, 19.96it/s]\u001B[A\n",
      " 70%|██████▉   | 6984/10000 [05:31<02:21, 21.28it/s]\u001B[A\n",
      " 70%|██████▉   | 6992/10000 [05:31<02:17, 21.94it/s]\u001B[A\n",
      " 70%|███████   | 7000/10000 [05:32<02:27, 20.40it/s]\u001B[A\n",
      " 70%|███████   | 7008/10000 [05:32<02:25, 20.61it/s]\u001B[A\n",
      " 70%|███████   | 7016/10000 [05:32<02:25, 20.46it/s]\u001B[A\n",
      " 70%|███████   | 7024/10000 [05:33<02:23, 20.72it/s]\u001B[A\n",
      " 70%|███████   | 7032/10000 [05:33<02:19, 21.32it/s]\u001B[A\n",
      " 70%|███████   | 7040/10000 [05:33<02:12, 22.33it/s]\u001B[A\n",
      " 70%|███████   | 7048/10000 [05:34<02:14, 22.01it/s]\u001B[A\n",
      " 71%|███████   | 7056/10000 [05:34<02:09, 22.66it/s]\u001B[A\n",
      " 71%|███████   | 7064/10000 [05:35<02:40, 18.33it/s]\u001B[A\n",
      " 71%|███████   | 7072/10000 [05:35<02:25, 20.19it/s]\u001B[A\n",
      " 71%|███████   | 7080/10000 [05:35<02:18, 21.05it/s]\u001B[A\n",
      " 71%|███████   | 7088/10000 [05:36<02:19, 20.81it/s]\u001B[A\n",
      " 71%|███████   | 7096/10000 [05:36<02:14, 21.53it/s]\u001B[A\n",
      " 71%|███████   | 7104/10000 [05:36<02:07, 22.77it/s]\u001B[A\n",
      " 71%|███████   | 7112/10000 [05:37<02:28, 19.45it/s]\u001B[A\n",
      " 71%|███████   | 7120/10000 [05:37<02:09, 22.17it/s]\u001B[A\n",
      " 71%|███████▏  | 7128/10000 [05:38<02:15, 21.26it/s]\u001B[A\n",
      " 71%|███████▏  | 7136/10000 [05:38<02:09, 22.18it/s]\u001B[A\n",
      " 71%|███████▏  | 7144/10000 [05:38<02:12, 21.55it/s]\u001B[A\n",
      " 72%|███████▏  | 7152/10000 [05:39<02:05, 22.70it/s]\u001B[A\n",
      " 72%|███████▏  | 7160/10000 [05:39<02:10, 21.82it/s]\u001B[A\n",
      " 72%|███████▏  | 7168/10000 [05:39<02:08, 22.08it/s]\u001B[A\n",
      " 72%|███████▏  | 7176/10000 [05:40<02:10, 21.64it/s]\u001B[A\n",
      " 72%|███████▏  | 7184/10000 [05:40<02:03, 22.86it/s]\u001B[A\n",
      " 72%|███████▏  | 7192/10000 [05:40<02:00, 23.22it/s]\u001B[A\n",
      " 72%|███████▏  | 7200/10000 [05:41<02:11, 21.35it/s]\u001B[A\n",
      " 72%|███████▏  | 7208/10000 [05:41<02:08, 21.65it/s]\u001B[A\n",
      " 72%|███████▏  | 7216/10000 [05:42<02:04, 22.41it/s]\u001B[A\n",
      " 72%|███████▏  | 7224/10000 [05:42<02:07, 21.74it/s]\u001B[A\n",
      " 72%|███████▏  | 7232/10000 [05:42<02:07, 21.78it/s]\u001B[A\n",
      " 72%|███████▏  | 7240/10000 [05:43<02:00, 22.82it/s]\u001B[A\n",
      " 72%|███████▏  | 7248/10000 [05:43<02:03, 22.26it/s]\u001B[A\n",
      " 73%|███████▎  | 7256/10000 [05:43<02:04, 22.10it/s]\u001B[A\n",
      " 73%|███████▎  | 7264/10000 [05:44<02:06, 21.60it/s]\u001B[A\n",
      " 73%|███████▎  | 7272/10000 [05:44<02:07, 21.32it/s]\u001B[A\n",
      " 73%|███████▎  | 7280/10000 [05:44<02:06, 21.43it/s]\u001B[A\n",
      " 73%|███████▎  | 7288/10000 [05:45<02:05, 21.69it/s]\u001B[A\n",
      " 73%|███████▎  | 7296/10000 [05:45<02:01, 22.25it/s]\u001B[A\n",
      " 73%|███████▎  | 7304/10000 [05:46<02:12, 20.33it/s]\u001B[A\n",
      " 73%|███████▎  | 7312/10000 [05:46<02:05, 21.41it/s]\u001B[A\n",
      " 73%|███████▎  | 7320/10000 [05:46<02:05, 21.30it/s]\u001B[A\n",
      " 73%|███████▎  | 7328/10000 [05:47<02:06, 21.17it/s]\u001B[A\n",
      " 73%|███████▎  | 7336/10000 [05:47<02:20, 18.92it/s]\u001B[A\n",
      " 73%|███████▎  | 7344/10000 [05:48<02:13, 19.92it/s]\u001B[A\n",
      " 74%|███████▎  | 7352/10000 [05:48<02:08, 20.57it/s]\u001B[A\n",
      " 74%|███████▎  | 7360/10000 [05:48<02:13, 19.81it/s]\u001B[A\n",
      " 74%|███████▎  | 7368/10000 [05:49<02:09, 20.37it/s]\u001B[A\n",
      " 74%|███████▍  | 7376/10000 [05:49<02:11, 20.02it/s]\u001B[A\n",
      " 74%|███████▍  | 7384/10000 [05:50<02:07, 20.48it/s]\u001B[A\n",
      " 74%|███████▍  | 7392/10000 [05:50<02:07, 20.51it/s]\u001B[A\n",
      " 74%|███████▍  | 7400/10000 [05:50<02:02, 21.26it/s]\u001B[A\n",
      " 74%|███████▍  | 7408/10000 [05:51<02:02, 21.22it/s]\u001B[A\n",
      " 74%|███████▍  | 7416/10000 [05:51<01:58, 21.76it/s]\u001B[A\n",
      " 74%|███████▍  | 7424/10000 [05:51<02:01, 21.16it/s]\u001B[A\n",
      " 74%|███████▍  | 7432/10000 [05:52<02:18, 18.50it/s]\u001B[A\n",
      " 74%|███████▍  | 7440/10000 [05:52<02:11, 19.50it/s]\u001B[A\n",
      " 74%|███████▍  | 7448/10000 [05:53<02:10, 19.59it/s]\u001B[A\n",
      " 75%|███████▍  | 7456/10000 [05:53<02:06, 20.10it/s]\u001B[A\n",
      " 75%|███████▍  | 7464/10000 [05:54<02:04, 20.41it/s]\u001B[A\n",
      " 75%|███████▍  | 7472/10000 [05:54<02:07, 19.90it/s]\u001B[A\n",
      " 75%|███████▍  | 7480/10000 [05:54<02:06, 19.98it/s]\u001B[A\n",
      " 75%|███████▍  | 7488/10000 [05:55<01:55, 21.66it/s]\u001B[A\n",
      " 75%|███████▍  | 7496/10000 [05:55<01:51, 22.37it/s]\u001B[A\n",
      " 75%|███████▌  | 7504/10000 [05:55<02:02, 20.32it/s]\u001B[A\n",
      " 75%|███████▌  | 7512/10000 [05:56<02:08, 19.35it/s]\u001B[A\n",
      " 75%|███████▌  | 7520/10000 [05:56<01:57, 21.03it/s]\u001B[A\n",
      " 75%|███████▌  | 7528/10000 [05:57<02:10, 18.93it/s]\u001B[A\n",
      " 75%|███████▌  | 7536/10000 [05:57<02:06, 19.52it/s]\u001B[A\n",
      " 75%|███████▌  | 7544/10000 [05:57<02:03, 19.82it/s]\u001B[A\n",
      " 76%|███████▌  | 7552/10000 [05:58<02:05, 19.55it/s]\u001B[A\n",
      " 76%|███████▌  | 7560/10000 [05:58<01:59, 20.44it/s]\u001B[A\n",
      " 76%|███████▌  | 7568/10000 [05:59<01:59, 20.38it/s]\u001B[A\n",
      " 76%|███████▌  | 7576/10000 [05:59<01:54, 21.15it/s]\u001B[A\n",
      " 76%|███████▌  | 7584/10000 [05:59<01:53, 21.34it/s]\u001B[A\n",
      " 76%|███████▌  | 7592/10000 [06:00<01:45, 22.89it/s]\u001B[A\n",
      " 76%|███████▌  | 7600/10000 [06:00<01:51, 21.56it/s]\u001B[A\n",
      " 76%|███████▌  | 7608/10000 [06:01<01:55, 20.77it/s]\u001B[A\n",
      " 76%|███████▌  | 7616/10000 [06:01<01:50, 21.51it/s]\u001B[A\n",
      " 76%|███████▌  | 7624/10000 [06:01<01:50, 21.44it/s]\u001B[A\n",
      " 76%|███████▋  | 7632/10000 [06:02<01:53, 20.83it/s]\u001B[A\n",
      " 76%|███████▋  | 7640/10000 [06:02<02:04, 18.90it/s]\u001B[A\n",
      " 76%|███████▋  | 7648/10000 [06:02<01:54, 20.57it/s]\u001B[A\n",
      " 77%|███████▋  | 7656/10000 [06:03<01:53, 20.63it/s]\u001B[A\n",
      " 77%|███████▋  | 7664/10000 [06:03<01:48, 21.53it/s]\u001B[A\n",
      " 77%|███████▋  | 7672/10000 [06:04<01:47, 21.62it/s]\u001B[A\n",
      " 77%|███████▋  | 7680/10000 [06:04<01:47, 21.63it/s]\u001B[A\n",
      " 77%|███████▋  | 7688/10000 [06:04<01:44, 22.10it/s]\u001B[A\n",
      " 77%|███████▋  | 7696/10000 [06:05<01:48, 21.22it/s]\u001B[A\n",
      " 77%|███████▋  | 7704/10000 [06:05<01:46, 21.50it/s]\u001B[A\n",
      " 77%|███████▋  | 7712/10000 [06:05<01:42, 22.27it/s]\u001B[A\n",
      " 77%|███████▋  | 7720/10000 [06:06<01:41, 22.41it/s]\u001B[A\n",
      " 77%|███████▋  | 7728/10000 [06:06<01:39, 22.78it/s]\u001B[A\n",
      " 77%|███████▋  | 7736/10000 [06:06<01:41, 22.24it/s]\u001B[A\n",
      " 77%|███████▋  | 7744/10000 [06:07<01:35, 23.60it/s]\u001B[A\n",
      " 78%|███████▊  | 7752/10000 [06:07<01:39, 22.71it/s]\u001B[A\n",
      " 78%|███████▊  | 7760/10000 [06:08<01:47, 20.77it/s]\u001B[A\n",
      " 78%|███████▊  | 7768/10000 [06:08<01:42, 21.82it/s]\u001B[A\n",
      " 78%|███████▊  | 7776/10000 [06:08<01:44, 21.30it/s]\u001B[A\n",
      " 78%|███████▊  | 7784/10000 [06:09<01:41, 21.78it/s]\u001B[A\n",
      " 78%|███████▊  | 7792/10000 [06:09<01:38, 22.36it/s]\u001B[A\n",
      " 78%|███████▊  | 7800/10000 [06:09<01:35, 23.03it/s]\u001B[A\n",
      " 78%|███████▊  | 7808/10000 [06:10<01:42, 21.45it/s]\u001B[A\n",
      " 78%|███████▊  | 7816/10000 [06:10<01:37, 22.38it/s]\u001B[A\n",
      " 78%|███████▊  | 7824/10000 [06:10<01:41, 21.39it/s]\u001B[A\n",
      " 78%|███████▊  | 7832/10000 [06:11<01:38, 21.98it/s]\u001B[A\n",
      " 78%|███████▊  | 7840/10000 [06:11<01:38, 21.93it/s]\u001B[A\n",
      " 78%|███████▊  | 7848/10000 [06:11<01:33, 22.90it/s]\u001B[A\n",
      " 79%|███████▊  | 7856/10000 [06:12<01:38, 21.69it/s]\u001B[A\n",
      " 79%|███████▊  | 7864/10000 [06:12<01:37, 21.94it/s]\u001B[A\n",
      " 79%|███████▊  | 7872/10000 [06:13<01:37, 21.87it/s]\u001B[A\n",
      " 79%|███████▉  | 7880/10000 [06:13<01:32, 22.82it/s]\u001B[A\n",
      " 79%|███████▉  | 7888/10000 [06:13<01:31, 22.96it/s]\u001B[A\n",
      " 79%|███████▉  | 7896/10000 [06:14<01:31, 23.02it/s]\u001B[A\n",
      " 79%|███████▉  | 7904/10000 [06:14<01:32, 22.56it/s]\u001B[A\n",
      " 79%|███████▉  | 7912/10000 [06:14<01:32, 22.62it/s]\u001B[A\n",
      " 79%|███████▉  | 7920/10000 [06:15<01:33, 22.32it/s]\u001B[A\n",
      " 79%|███████▉  | 7928/10000 [06:15<01:34, 21.95it/s]\u001B[A\n",
      " 79%|███████▉  | 7936/10000 [06:15<01:30, 22.92it/s]\u001B[A\n",
      " 79%|███████▉  | 7944/10000 [06:16<01:34, 21.65it/s]\u001B[A\n",
      " 80%|███████▉  | 7952/10000 [06:16<01:27, 23.45it/s]\u001B[A\n",
      " 80%|███████▉  | 7960/10000 [06:17<02:32, 13.36it/s]\u001B[A\n",
      " 80%|███████▉  | 7968/10000 [06:18<03:15, 10.40it/s]\u001B[A\n",
      " 80%|███████▉  | 7976/10000 [06:19<02:40, 12.60it/s]\u001B[A\n",
      " 80%|███████▉  | 7984/10000 [06:19<02:18, 14.57it/s]\u001B[A\n",
      " 80%|███████▉  | 7992/10000 [06:19<02:00, 16.71it/s]\u001B[A\n",
      " 80%|████████  | 8000/10000 [06:20<01:47, 18.58it/s]\u001B[A\n",
      " 80%|████████  | 8008/10000 [06:20<01:39, 20.04it/s]\u001B[A\n",
      " 80%|████████  | 8016/10000 [06:20<01:35, 20.76it/s]\u001B[A\n",
      " 80%|████████  | 8024/10000 [06:21<01:34, 20.95it/s]\u001B[A\n",
      " 80%|████████  | 8032/10000 [06:21<01:32, 21.18it/s]\u001B[A\n",
      " 80%|████████  | 8040/10000 [06:22<01:29, 21.82it/s]\u001B[A\n",
      " 80%|████████  | 8048/10000 [06:22<01:53, 17.25it/s]\u001B[A\n",
      " 81%|████████  | 8056/10000 [06:23<01:42, 19.01it/s]\u001B[A\n",
      " 81%|████████  | 8064/10000 [06:23<01:38, 19.62it/s]\u001B[A\n",
      " 81%|████████  | 8072/10000 [06:23<01:36, 20.07it/s]\u001B[A\n",
      " 81%|████████  | 8080/10000 [06:24<01:36, 19.92it/s]\u001B[A\n",
      " 81%|████████  | 8088/10000 [06:24<01:30, 21.07it/s]\u001B[A\n",
      " 81%|████████  | 8096/10000 [06:24<01:27, 21.65it/s]\u001B[A\n",
      " 81%|████████  | 8104/10000 [06:25<01:25, 22.24it/s]\u001B[A\n",
      " 81%|████████  | 8112/10000 [06:25<01:21, 23.09it/s]\u001B[A\n",
      " 81%|████████  | 8120/10000 [06:25<01:22, 22.68it/s]\u001B[A\n",
      " 81%|████████▏ | 8128/10000 [06:26<01:31, 20.49it/s]\u001B[A\n",
      " 81%|████████▏ | 8136/10000 [06:26<01:27, 21.28it/s]\u001B[A\n",
      " 81%|████████▏ | 8144/10000 [06:27<01:24, 22.02it/s]\u001B[A\n",
      " 82%|████████▏ | 8152/10000 [06:27<01:21, 22.59it/s]\u001B[A\n",
      " 82%|████████▏ | 8160/10000 [06:27<01:20, 22.90it/s]\u001B[A\n",
      " 82%|████████▏ | 8168/10000 [06:28<01:23, 22.05it/s]\u001B[A\n",
      " 82%|████████▏ | 8176/10000 [06:28<01:19, 22.81it/s]\u001B[A\n",
      " 82%|████████▏ | 8184/10000 [06:28<01:18, 22.99it/s]\u001B[A\n",
      " 82%|████████▏ | 8192/10000 [06:29<01:18, 23.05it/s]\u001B[A\n",
      " 82%|████████▏ | 8200/10000 [06:29<01:15, 23.82it/s]\u001B[A\n",
      " 82%|████████▏ | 8208/10000 [06:29<01:18, 22.92it/s]\u001B[A\n",
      " 82%|████████▏ | 8216/10000 [06:30<01:17, 23.05it/s]\u001B[A\n",
      " 82%|████████▏ | 8224/10000 [06:30<01:18, 22.58it/s]\u001B[A\n",
      " 82%|████████▏ | 8232/10000 [06:30<01:19, 22.37it/s]\u001B[A\n",
      " 82%|████████▏ | 8240/10000 [06:31<01:15, 23.19it/s]\u001B[A\n",
      " 82%|████████▏ | 8248/10000 [06:31<01:17, 22.70it/s]\u001B[A\n",
      " 83%|████████▎ | 8256/10000 [06:31<01:16, 22.85it/s]\u001B[A\n",
      " 83%|████████▎ | 8264/10000 [06:32<01:13, 23.69it/s]\u001B[A\n",
      " 83%|████████▎ | 8272/10000 [06:32<01:18, 22.03it/s]\u001B[A\n",
      " 83%|████████▎ | 8280/10000 [06:32<01:15, 22.72it/s]\u001B[A\n",
      " 83%|████████▎ | 8288/10000 [06:33<01:15, 22.79it/s]\u001B[A\n",
      " 83%|████████▎ | 8296/10000 [06:33<01:14, 22.89it/s]\u001B[A\n",
      " 83%|████████▎ | 8304/10000 [06:33<01:11, 23.86it/s]\u001B[A\n",
      " 83%|████████▎ | 8312/10000 [06:34<01:12, 23.39it/s]\u001B[A\n",
      " 83%|████████▎ | 8320/10000 [06:34<01:10, 23.86it/s]\u001B[A\n",
      " 83%|████████▎ | 8328/10000 [06:35<01:13, 22.85it/s]\u001B[A\n",
      " 83%|████████▎ | 8336/10000 [06:35<01:15, 21.92it/s]\u001B[A\n",
      " 83%|████████▎ | 8344/10000 [06:35<01:14, 22.16it/s]\u001B[A\n",
      " 84%|████████▎ | 8352/10000 [06:36<01:08, 24.11it/s]\u001B[A\n",
      " 84%|████████▎ | 8360/10000 [06:36<01:11, 23.04it/s]\u001B[A\n",
      " 84%|████████▎ | 8368/10000 [06:36<01:14, 21.91it/s]\u001B[A\n",
      " 84%|████████▍ | 8376/10000 [06:37<01:11, 22.62it/s]\u001B[A\n",
      " 84%|████████▍ | 8384/10000 [06:37<01:13, 21.95it/s]\u001B[A\n",
      " 84%|████████▍ | 8392/10000 [06:37<01:12, 22.27it/s]\u001B[A\n",
      " 84%|████████▍ | 8400/10000 [06:38<01:10, 22.74it/s]\u001B[A\n",
      " 84%|████████▍ | 8408/10000 [06:38<01:11, 22.23it/s]\u001B[A\n",
      " 84%|████████▍ | 8416/10000 [06:38<01:09, 22.63it/s]\u001B[A\n",
      " 84%|████████▍ | 8424/10000 [06:39<01:10, 22.46it/s]\u001B[A\n",
      " 84%|████████▍ | 8432/10000 [06:39<01:09, 22.65it/s]\u001B[A\n",
      " 84%|████████▍ | 8440/10000 [06:40<01:08, 22.93it/s]\u001B[A\n",
      " 84%|████████▍ | 8448/10000 [06:40<01:11, 21.71it/s]\u001B[A\n",
      " 85%|████████▍ | 8456/10000 [06:40<01:07, 22.97it/s]\u001B[A\n",
      " 85%|████████▍ | 8464/10000 [06:41<01:10, 21.92it/s]\u001B[A\n",
      " 85%|████████▍ | 8472/10000 [06:41<01:08, 22.22it/s]\u001B[A\n",
      " 85%|████████▍ | 8480/10000 [06:41<01:08, 22.35it/s]\u001B[A\n",
      " 85%|████████▍ | 8488/10000 [06:42<01:08, 22.06it/s]\u001B[A\n",
      " 85%|████████▍ | 8496/10000 [06:42<01:07, 22.36it/s]\u001B[A\n",
      " 85%|████████▌ | 8504/10000 [06:42<01:09, 21.63it/s]\u001B[A\n",
      " 85%|████████▌ | 8512/10000 [06:43<01:09, 21.36it/s]\u001B[A\n",
      " 85%|████████▌ | 8520/10000 [06:43<01:09, 21.30it/s]\u001B[A\n",
      " 85%|████████▌ | 8528/10000 [06:44<01:09, 21.26it/s]\u001B[A\n",
      " 85%|████████▌ | 8536/10000 [06:44<01:03, 23.00it/s]\u001B[A\n",
      " 85%|████████▌ | 8544/10000 [06:44<01:05, 22.21it/s]\u001B[A\n",
      " 86%|████████▌ | 8552/10000 [06:45<01:05, 22.01it/s]\u001B[A\n",
      " 86%|████████▌ | 8560/10000 [06:45<01:05, 22.10it/s]\u001B[A\n",
      " 86%|████████▌ | 8568/10000 [06:45<01:05, 21.82it/s]\u001B[A\n",
      " 86%|████████▌ | 8576/10000 [06:46<01:06, 21.47it/s]\u001B[A\n",
      " 86%|████████▌ | 8584/10000 [06:46<01:01, 22.84it/s]\u001B[A\n",
      " 86%|████████▌ | 8592/10000 [06:46<01:00, 23.23it/s]\u001B[A\n",
      " 86%|████████▌ | 8600/10000 [06:47<01:00, 23.24it/s]\u001B[A\n",
      " 86%|████████▌ | 8608/10000 [06:47<01:00, 22.88it/s]\u001B[A\n",
      " 86%|████████▌ | 8616/10000 [06:47<01:00, 22.79it/s]\u001B[A\n",
      " 86%|████████▌ | 8624/10000 [06:48<01:03, 21.76it/s]\u001B[A\n",
      " 86%|████████▋ | 8632/10000 [06:48<01:01, 22.18it/s]\u001B[A\n",
      " 86%|████████▋ | 8640/10000 [06:49<01:04, 21.21it/s]\u001B[A\n",
      " 86%|████████▋ | 8648/10000 [06:49<01:01, 22.13it/s]\u001B[A\n",
      " 87%|████████▋ | 8656/10000 [06:49<01:00, 22.23it/s]\u001B[A\n",
      " 87%|████████▋ | 8664/10000 [06:50<00:59, 22.31it/s]\u001B[A\n",
      " 87%|████████▋ | 8672/10000 [06:50<01:02, 21.12it/s]\u001B[A\n",
      " 87%|████████▋ | 8680/10000 [06:50<01:00, 21.93it/s]\u001B[A\n",
      " 87%|████████▋ | 8688/10000 [06:51<00:58, 22.49it/s]\u001B[A\n",
      " 87%|████████▋ | 8696/10000 [06:51<00:57, 22.54it/s]\u001B[A\n",
      " 87%|████████▋ | 8704/10000 [06:51<00:58, 22.24it/s]\u001B[A\n",
      " 87%|████████▋ | 8712/10000 [06:52<01:09, 18.53it/s]\u001B[A\n",
      " 87%|████████▋ | 8720/10000 [06:52<01:08, 18.56it/s]\u001B[A\n",
      " 87%|████████▋ | 8728/10000 [06:53<01:09, 18.32it/s]\u001B[A\n",
      " 87%|████████▋ | 8736/10000 [06:53<01:03, 19.80it/s]\u001B[A\n",
      " 87%|████████▋ | 8744/10000 [06:54<01:04, 19.51it/s]\u001B[A\n",
      " 88%|████████▊ | 8752/10000 [06:54<01:00, 20.61it/s]\u001B[A\n",
      " 88%|████████▊ | 8760/10000 [06:54<01:01, 20.02it/s]\u001B[A\n",
      " 88%|████████▊ | 8768/10000 [06:55<00:57, 21.34it/s]\u001B[A\n",
      " 88%|████████▊ | 8776/10000 [06:55<00:56, 21.60it/s]\u001B[A\n",
      " 88%|████████▊ | 8784/10000 [06:56<00:56, 21.39it/s]\u001B[A\n",
      " 88%|████████▊ | 8792/10000 [06:56<00:57, 21.17it/s]\u001B[A\n",
      " 88%|████████▊ | 8800/10000 [06:56<00:54, 22.01it/s]\u001B[A\n",
      " 88%|████████▊ | 8808/10000 [06:57<00:54, 21.92it/s]\u001B[A\n",
      " 88%|████████▊ | 8816/10000 [06:57<00:54, 21.78it/s]\u001B[A\n",
      " 88%|████████▊ | 8824/10000 [06:57<00:54, 21.71it/s]\u001B[A\n",
      " 88%|████████▊ | 8832/10000 [06:58<00:55, 21.05it/s]\u001B[A\n",
      " 88%|████████▊ | 8840/10000 [06:58<00:56, 20.50it/s]\u001B[A\n",
      " 88%|████████▊ | 8848/10000 [06:58<00:52, 21.77it/s]\u001B[A\n",
      " 89%|████████▊ | 8856/10000 [06:59<00:54, 21.10it/s]\u001B[A\n",
      " 89%|████████▊ | 8864/10000 [06:59<00:52, 21.46it/s]\u001B[A\n",
      " 89%|████████▊ | 8872/10000 [07:00<00:53, 21.09it/s]\u001B[A\n",
      " 89%|████████▉ | 8880/10000 [07:00<01:00, 18.37it/s]\u001B[A\n",
      " 89%|████████▉ | 8888/10000 [07:01<00:55, 20.04it/s]\u001B[A\n",
      " 89%|████████▉ | 8896/10000 [07:01<00:53, 20.68it/s]\u001B[A\n",
      " 89%|████████▉ | 8904/10000 [07:01<00:52, 20.69it/s]\u001B[A\n",
      " 89%|████████▉ | 8912/10000 [07:02<00:54, 19.95it/s]\u001B[A\n",
      " 89%|████████▉ | 8920/10000 [07:02<00:53, 20.03it/s]\u001B[A\n",
      " 89%|████████▉ | 8928/10000 [07:02<00:52, 20.41it/s]\u001B[A\n",
      " 89%|████████▉ | 8936/10000 [07:03<00:48, 21.89it/s]\u001B[A\n",
      " 89%|████████▉ | 8944/10000 [07:03<00:55, 19.02it/s]\u001B[A\n",
      " 90%|████████▉ | 8952/10000 [07:04<00:53, 19.68it/s]\u001B[A\n",
      " 90%|████████▉ | 8960/10000 [07:04<00:50, 20.73it/s]\u001B[A\n",
      " 90%|████████▉ | 8968/10000 [07:04<00:50, 20.63it/s]\u001B[A\n",
      " 90%|████████▉ | 8976/10000 [07:05<00:47, 21.61it/s]\u001B[A\n",
      " 90%|████████▉ | 8984/10000 [07:05<00:48, 21.07it/s]\u001B[A\n",
      " 90%|████████▉ | 8992/10000 [07:05<00:46, 21.58it/s]\u001B[A\n",
      " 90%|█████████ | 9000/10000 [07:06<00:46, 21.32it/s]\u001B[A\n",
      " 90%|█████████ | 9008/10000 [07:06<00:48, 20.54it/s]\u001B[A\n",
      " 90%|█████████ | 9016/10000 [07:07<00:46, 21.33it/s]\u001B[A\n",
      " 90%|█████████ | 9024/10000 [07:07<00:45, 21.27it/s]\u001B[A\n",
      " 90%|█████████ | 9032/10000 [07:07<00:44, 21.54it/s]\u001B[A\n",
      " 90%|█████████ | 9040/10000 [07:08<00:47, 20.39it/s]\u001B[A\n",
      " 90%|█████████ | 9048/10000 [07:08<00:44, 21.23it/s]\u001B[A\n",
      " 91%|█████████ | 9056/10000 [07:09<00:45, 20.88it/s]\u001B[A\n",
      " 91%|█████████ | 9064/10000 [07:09<00:42, 21.78it/s]\u001B[A\n",
      " 91%|█████████ | 9072/10000 [07:09<00:44, 21.03it/s]\u001B[A\n",
      " 91%|█████████ | 9080/10000 [07:10<00:41, 22.03it/s]\u001B[A\n",
      " 91%|█████████ | 9088/10000 [07:10<00:42, 21.66it/s]\u001B[A\n",
      " 91%|█████████ | 9096/10000 [07:10<00:42, 21.42it/s]\u001B[A\n",
      " 91%|█████████ | 9104/10000 [07:11<00:40, 21.91it/s]\u001B[A\n",
      " 91%|█████████ | 9112/10000 [07:11<00:39, 22.23it/s]\u001B[A\n",
      " 91%|█████████ | 9120/10000 [07:12<00:41, 21.00it/s]\u001B[A\n",
      " 91%|█████████▏| 9128/10000 [07:12<00:40, 21.68it/s]\u001B[A\n",
      " 91%|█████████▏| 9136/10000 [07:12<00:40, 21.33it/s]\u001B[A\n",
      " 91%|█████████▏| 9144/10000 [07:13<00:37, 23.02it/s]\u001B[A\n",
      " 92%|█████████▏| 9152/10000 [07:13<00:38, 21.75it/s]\u001B[A\n",
      " 92%|█████████▏| 9160/10000 [07:13<00:37, 22.16it/s]\u001B[A\n",
      " 92%|█████████▏| 9168/10000 [07:14<00:37, 22.26it/s]\u001B[A\n",
      " 92%|█████████▏| 9176/10000 [07:14<00:35, 23.14it/s]\u001B[A\n",
      " 92%|█████████▏| 9184/10000 [07:14<00:34, 23.40it/s]\u001B[A\n",
      " 92%|█████████▏| 9192/10000 [07:15<00:39, 20.28it/s]\u001B[A\n",
      " 92%|█████████▏| 9200/10000 [07:15<00:38, 20.85it/s]\u001B[A\n",
      " 92%|█████████▏| 9208/10000 [07:16<00:37, 21.37it/s]\u001B[A\n",
      " 92%|█████████▏| 9216/10000 [07:16<00:37, 20.91it/s]\u001B[A\n",
      " 92%|█████████▏| 9224/10000 [07:16<00:36, 21.17it/s]\u001B[A\n",
      " 92%|█████████▏| 9232/10000 [07:17<00:36, 20.79it/s]\u001B[A\n",
      " 92%|█████████▏| 9240/10000 [07:17<00:35, 21.40it/s]\u001B[A\n",
      " 92%|█████████▏| 9248/10000 [07:17<00:35, 21.30it/s]\u001B[A\n",
      " 93%|█████████▎| 9256/10000 [07:18<00:33, 22.02it/s]\u001B[A\n",
      " 93%|█████████▎| 9264/10000 [07:18<00:32, 22.34it/s]\u001B[A\n",
      " 93%|█████████▎| 9272/10000 [07:18<00:33, 21.86it/s]\u001B[A\n",
      " 93%|█████████▎| 9280/10000 [07:19<00:33, 21.36it/s]\u001B[A\n",
      " 93%|█████████▎| 9288/10000 [07:19<00:31, 22.50it/s]\u001B[A\n",
      " 93%|█████████▎| 9296/10000 [07:20<00:31, 22.01it/s]\u001B[A\n",
      " 93%|█████████▎| 9304/10000 [07:20<00:33, 21.07it/s]\u001B[A\n",
      " 93%|█████████▎| 9312/10000 [07:20<00:30, 22.36it/s]\u001B[A\n",
      " 93%|█████████▎| 9320/10000 [07:21<00:29, 22.78it/s]\u001B[A\n",
      " 93%|█████████▎| 9328/10000 [07:21<00:34, 19.20it/s]\u001B[A\n",
      " 93%|█████████▎| 9336/10000 [07:22<00:33, 19.75it/s]\u001B[A\n",
      " 93%|█████████▎| 9344/10000 [07:22<00:30, 21.79it/s]\u001B[A\n",
      " 94%|█████████▎| 9352/10000 [07:22<00:30, 21.41it/s]\u001B[A\n",
      " 94%|█████████▎| 9360/10000 [07:23<00:28, 22.11it/s]\u001B[A\n",
      " 94%|█████████▎| 9368/10000 [07:23<00:28, 22.13it/s]\u001B[A\n",
      " 94%|█████████▍| 9376/10000 [07:23<00:27, 22.93it/s]\u001B[A\n",
      " 94%|█████████▍| 9384/10000 [07:24<00:27, 22.51it/s]\u001B[A\n",
      " 94%|█████████▍| 9392/10000 [07:24<00:26, 23.08it/s]\u001B[A\n",
      " 94%|█████████▍| 9400/10000 [07:24<00:25, 23.44it/s]\u001B[A\n",
      " 94%|█████████▍| 9408/10000 [07:25<00:25, 23.26it/s]\u001B[A\n",
      " 94%|█████████▍| 9416/10000 [07:25<00:25, 23.15it/s]\u001B[A\n",
      " 94%|█████████▍| 9424/10000 [07:25<00:24, 23.05it/s]\u001B[A\n",
      " 94%|█████████▍| 9432/10000 [07:26<00:25, 22.69it/s]\u001B[A\n",
      " 94%|█████████▍| 9440/10000 [07:26<00:23, 23.74it/s]\u001B[A\n",
      " 94%|█████████▍| 9448/10000 [07:26<00:24, 22.14it/s]\u001B[A\n",
      " 95%|█████████▍| 9456/10000 [07:27<00:23, 22.84it/s]\u001B[A\n",
      " 95%|█████████▍| 9464/10000 [07:27<00:25, 20.76it/s]\u001B[A\n",
      " 95%|█████████▍| 9472/10000 [07:28<00:25, 20.96it/s]\u001B[A\n",
      " 95%|█████████▍| 9480/10000 [07:28<00:24, 20.90it/s]\u001B[A\n",
      " 95%|█████████▍| 9488/10000 [07:28<00:22, 22.31it/s]\u001B[A\n",
      " 95%|█████████▍| 9496/10000 [07:29<00:23, 21.91it/s]\u001B[A\n",
      " 95%|█████████▌| 9504/10000 [07:29<00:21, 22.68it/s]\u001B[A\n",
      " 95%|█████████▌| 9512/10000 [07:29<00:21, 22.22it/s]\u001B[A\n",
      " 95%|█████████▌| 9520/10000 [07:30<00:22, 21.40it/s]\u001B[A\n",
      " 95%|█████████▌| 9528/10000 [07:30<00:21, 21.90it/s]\u001B[A\n",
      " 95%|█████████▌| 9536/10000 [07:30<00:20, 22.32it/s]\u001B[A\n",
      " 95%|█████████▌| 9544/10000 [07:31<00:20, 22.45it/s]\u001B[A\n",
      " 96%|█████████▌| 9552/10000 [07:31<00:19, 23.18it/s]\u001B[A\n",
      " 96%|█████████▌| 9560/10000 [07:31<00:18, 23.39it/s]\u001B[A\n",
      " 96%|█████████▌| 9568/10000 [07:32<00:19, 22.54it/s]\u001B[A\n",
      " 96%|█████████▌| 9576/10000 [07:32<00:18, 22.51it/s]\u001B[A\n",
      " 96%|█████████▌| 9584/10000 [07:33<00:17, 23.56it/s]\u001B[A\n",
      " 96%|█████████▌| 9592/10000 [07:33<00:16, 24.90it/s]\u001B[A\n",
      " 96%|█████████▌| 9600/10000 [07:33<00:16, 24.28it/s]\u001B[A\n",
      " 96%|█████████▌| 9608/10000 [07:34<00:16, 23.28it/s]\u001B[A\n",
      " 96%|█████████▌| 9616/10000 [07:34<00:17, 22.30it/s]\u001B[A\n",
      " 96%|█████████▌| 9624/10000 [07:34<00:15, 23.63it/s]\u001B[A\n",
      " 96%|█████████▋| 9632/10000 [07:35<00:15, 23.42it/s]\u001B[A\n",
      " 96%|█████████▋| 9640/10000 [07:35<00:15, 23.53it/s]\u001B[A\n",
      " 96%|█████████▋| 9648/10000 [07:35<00:14, 23.80it/s]\u001B[A\n",
      " 97%|█████████▋| 9656/10000 [07:36<00:14, 24.21it/s]\u001B[A\n",
      " 97%|█████████▋| 9664/10000 [07:36<00:14, 23.09it/s]\u001B[A\n",
      " 97%|█████████▋| 9672/10000 [07:36<00:13, 23.77it/s]\u001B[A\n",
      " 97%|█████████▋| 9680/10000 [07:37<00:13, 23.80it/s]\u001B[A\n",
      " 97%|█████████▋| 9688/10000 [07:37<00:12, 24.05it/s]\u001B[A\n",
      " 97%|█████████▋| 9696/10000 [07:37<00:12, 23.83it/s]\u001B[A\n",
      " 97%|█████████▋| 9704/10000 [07:38<00:12, 23.41it/s]\u001B[A\n",
      " 97%|█████████▋| 9712/10000 [07:38<00:12, 23.28it/s]\u001B[A\n",
      " 97%|█████████▋| 9720/10000 [07:38<00:12, 22.62it/s]\u001B[A\n",
      " 97%|█████████▋| 9728/10000 [07:39<00:12, 22.65it/s]\u001B[A\n",
      " 97%|█████████▋| 9736/10000 [07:39<00:11, 23.31it/s]\u001B[A\n",
      " 97%|█████████▋| 9744/10000 [07:39<00:11, 22.83it/s]\u001B[A\n",
      " 98%|█████████▊| 9752/10000 [07:40<00:10, 23.19it/s]\u001B[A\n",
      " 98%|█████████▊| 9760/10000 [07:40<00:10, 23.35it/s]\u001B[A\n",
      " 98%|█████████▊| 9768/10000 [07:40<00:09, 24.08it/s]\u001B[A\n",
      " 98%|█████████▊| 9776/10000 [07:41<00:09, 23.95it/s]\u001B[A\n",
      " 98%|█████████▊| 9784/10000 [07:41<00:09, 23.58it/s]\u001B[A\n",
      " 98%|█████████▊| 9792/10000 [07:41<00:08, 23.19it/s]\u001B[A\n",
      " 98%|█████████▊| 9800/10000 [07:42<00:08, 23.85it/s]\u001B[A\n",
      " 98%|█████████▊| 9808/10000 [07:42<00:08, 23.55it/s]\u001B[A\n",
      " 98%|█████████▊| 9816/10000 [07:42<00:08, 22.56it/s]\u001B[A\n",
      " 98%|█████████▊| 9824/10000 [07:43<00:07, 23.56it/s]\u001B[A\n",
      " 98%|█████████▊| 9832/10000 [07:43<00:07, 23.13it/s]\u001B[A\n",
      " 98%|█████████▊| 9840/10000 [07:43<00:07, 22.64it/s]\u001B[A\n",
      " 98%|█████████▊| 9848/10000 [07:44<00:06, 22.33it/s]\u001B[A\n",
      " 99%|█████████▊| 9856/10000 [07:44<00:06, 23.36it/s]\u001B[A\n",
      " 99%|█████████▊| 9864/10000 [07:45<00:06, 22.44it/s]\u001B[A\n",
      " 99%|█████████▊| 9872/10000 [07:45<00:05, 22.41it/s]\u001B[A\n",
      " 99%|█████████▉| 9880/10000 [07:45<00:05, 23.37it/s]\u001B[A\n",
      " 99%|█████████▉| 9888/10000 [07:46<00:04, 23.27it/s]\u001B[A\n",
      " 99%|█████████▉| 9896/10000 [07:46<00:04, 23.35it/s]\u001B[A\n",
      " 99%|█████████▉| 9904/10000 [07:46<00:04, 22.84it/s]\u001B[A\n",
      " 99%|█████████▉| 9912/10000 [07:47<00:03, 23.00it/s]\u001B[A\n",
      " 99%|█████████▉| 9920/10000 [07:47<00:03, 22.68it/s]\u001B[A\n",
      " 99%|█████████▉| 9928/10000 [07:47<00:03, 22.97it/s]\u001B[A\n",
      " 99%|█████████▉| 9936/10000 [07:48<00:02, 22.73it/s]\u001B[A\n",
      " 99%|█████████▉| 9944/10000 [07:48<00:02, 22.18it/s]\u001B[A\n",
      "100%|█████████▉| 9952/10000 [07:48<00:02, 22.99it/s]\u001B[A\n",
      "100%|█████████▉| 9960/10000 [07:49<00:01, 22.18it/s]\u001B[A\n",
      "100%|█████████▉| 9968/10000 [07:49<00:01, 21.42it/s]\u001B[A\n",
      "100%|█████████▉| 9976/10000 [07:49<00:01, 22.40it/s]\u001B[A\n",
      "100%|█████████▉| 9984/10000 [07:50<00:00, 21.19it/s]\u001B[A\n",
      "100%|█████████▉| 9992/10000 [07:50<00:00, 21.35it/s]\u001B[A\n",
      "100%|██████████| 10000/10000 [07:51<00:00, 21.23it/s]\u001B[A\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(85, 5)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/aaronw/Desktop/PhD/Research/QMUL/Research/scBench/venv/lib/python3.10/site-packages/statsmodels/stats/multitest.py:186: RuntimeWarning: divide by zero encountered in log1p\n",
      "  np.log1p(-pvals))\n"
     ]
    }
   ],
   "execution_count": 28
  },
  {
   "cell_type": "code",
   "source": [
    "df.head()"
   ],
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "inoUq2y16K-s",
    "outputId": "7f3f8849-ad6c-425a-a697-23f18cbc6913"
   },
   "execution_count": null,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "(153, 5)\n"
     ]
    },
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "              edist  pvalue  significant  pvalue_adj  significant_adj\n",
       "AHR       21.986925  0.0001         True    0.015184             True\n",
       "AHR_KLF1  35.566693  0.0001         True    0.015184             True\n",
       "ARRDC3     6.685922  0.0001         True    0.015184             True\n",
       "ATL1      57.990839  0.0001         True    0.015184             True\n",
       "BAK1       0.725031  0.0282         True    0.082237            False"
      ],
      "text/html": [
       "\n",
       "  <div id=\"df-49d0f958-7d17-4089-b3bb-3baebd3724d4\" class=\"colab-df-container\">\n",
       "    <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>edist</th>\n",
       "      <th>pvalue</th>\n",
       "      <th>significant</th>\n",
       "      <th>pvalue_adj</th>\n",
       "      <th>significant_adj</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AHR</th>\n",
       "      <td>21.986925</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>True</td>\n",
       "      <td>0.015184</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AHR_KLF1</th>\n",
       "      <td>35.566693</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>True</td>\n",
       "      <td>0.015184</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ARRDC3</th>\n",
       "      <td>6.685922</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>True</td>\n",
       "      <td>0.015184</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ATL1</th>\n",
       "      <td>57.990839</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>True</td>\n",
       "      <td>0.015184</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BAK1</th>\n",
       "      <td>0.725031</td>\n",
       "      <td>0.0282</td>\n",
       "      <td>True</td>\n",
       "      <td>0.082237</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "    <div class=\"colab-df-buttons\">\n",
       "\n",
       "  <div class=\"colab-df-container\">\n",
       "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-49d0f958-7d17-4089-b3bb-3baebd3724d4')\"\n",
       "            title=\"Convert this dataframe to an interactive table.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
       "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "\n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    .colab-df-buttons div {\n",
       "      margin-bottom: 4px;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "    <script>\n",
       "      const buttonEl =\n",
       "        document.querySelector('#df-49d0f958-7d17-4089-b3bb-3baebd3724d4 button.colab-df-convert');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      async function convertToInteractive(key) {\n",
       "        const element = document.querySelector('#df-49d0f958-7d17-4089-b3bb-3baebd3724d4');\n",
       "        const dataTable =\n",
       "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                    [key], {});\n",
       "        if (!dataTable) return;\n",
       "\n",
       "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "          + ' to learn more about interactive tables.';\n",
       "        element.innerHTML = '';\n",
       "        dataTable['output_type'] = 'display_data';\n",
       "        await google.colab.output.renderOutput(dataTable, element);\n",
       "        const docLink = document.createElement('div');\n",
       "        docLink.innerHTML = docLinkHtml;\n",
       "        element.appendChild(docLink);\n",
       "      }\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "\n",
       "<div id=\"df-775bc6ac-e761-4c9b-8082-c21dc13aad86\">\n",
       "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-775bc6ac-e761-4c9b-8082-c21dc13aad86')\"\n",
       "            title=\"Suggest charts\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "     width=\"24px\">\n",
       "    <g>\n",
       "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
       "    </g>\n",
       "</svg>\n",
       "  </button>\n",
       "\n",
       "<style>\n",
       "  .colab-df-quickchart {\n",
       "      --bg-color: #E8F0FE;\n",
       "      --fill-color: #1967D2;\n",
       "      --hover-bg-color: #E2EBFA;\n",
       "      --hover-fill-color: #174EA6;\n",
       "      --disabled-fill-color: #AAA;\n",
       "      --disabled-bg-color: #DDD;\n",
       "  }\n",
       "\n",
       "  [theme=dark] .colab-df-quickchart {\n",
       "      --bg-color: #3B4455;\n",
       "      --fill-color: #D2E3FC;\n",
       "      --hover-bg-color: #434B5C;\n",
       "      --hover-fill-color: #FFFFFF;\n",
       "      --disabled-bg-color: #3B4455;\n",
       "      --disabled-fill-color: #666;\n",
       "  }\n",
       "\n",
       "  .colab-df-quickchart {\n",
       "    background-color: var(--bg-color);\n",
       "    border: none;\n",
       "    border-radius: 50%;\n",
       "    cursor: pointer;\n",
       "    display: none;\n",
       "    fill: var(--fill-color);\n",
       "    height: 32px;\n",
       "    padding: 0;\n",
       "    width: 32px;\n",
       "  }\n",
       "\n",
       "  .colab-df-quickchart:hover {\n",
       "    background-color: var(--hover-bg-color);\n",
       "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "    fill: var(--button-hover-fill-color);\n",
       "  }\n",
       "\n",
       "  .colab-df-quickchart-complete:disabled,\n",
       "  .colab-df-quickchart-complete:disabled:hover {\n",
       "    background-color: var(--disabled-bg-color);\n",
       "    fill: var(--disabled-fill-color);\n",
       "    box-shadow: none;\n",
       "  }\n",
       "\n",
       "  .colab-df-spinner {\n",
       "    border: 2px solid var(--fill-color);\n",
       "    border-color: transparent;\n",
       "    border-bottom-color: var(--fill-color);\n",
       "    animation:\n",
       "      spin 1s steps(1) infinite;\n",
       "  }\n",
       "\n",
       "  @keyframes spin {\n",
       "    0% {\n",
       "      border-color: transparent;\n",
       "      border-bottom-color: var(--fill-color);\n",
       "      border-left-color: var(--fill-color);\n",
       "    }\n",
       "    20% {\n",
       "      border-color: transparent;\n",
       "      border-left-color: var(--fill-color);\n",
       "      border-top-color: var(--fill-color);\n",
       "    }\n",
       "    30% {\n",
       "      border-color: transparent;\n",
       "      border-left-color: var(--fill-color);\n",
       "      border-top-color: var(--fill-color);\n",
       "      border-right-color: var(--fill-color);\n",
       "    }\n",
       "    40% {\n",
       "      border-color: transparent;\n",
       "      border-right-color: var(--fill-color);\n",
       "      border-top-color: var(--fill-color);\n",
       "    }\n",
       "    60% {\n",
       "      border-color: transparent;\n",
       "      border-right-color: var(--fill-color);\n",
       "    }\n",
       "    80% {\n",
       "      border-color: transparent;\n",
       "      border-right-color: var(--fill-color);\n",
       "      border-bottom-color: var(--fill-color);\n",
       "    }\n",
       "    90% {\n",
       "      border-color: transparent;\n",
       "      border-bottom-color: var(--fill-color);\n",
       "    }\n",
       "  }\n",
       "</style>\n",
       "\n",
       "  <script>\n",
       "    async function quickchart(key) {\n",
       "      const quickchartButtonEl =\n",
       "        document.querySelector('#' + key + ' button');\n",
       "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
       "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
       "      try {\n",
       "        const charts = await google.colab.kernel.invokeFunction(\n",
       "            'suggestCharts', [key], {});\n",
       "      } catch (error) {\n",
       "        console.error('Error during call to suggestCharts:', error);\n",
       "      }\n",
       "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
       "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
       "    }\n",
       "    (() => {\n",
       "      let quickchartButtonEl =\n",
       "        document.querySelector('#df-775bc6ac-e761-4c9b-8082-c21dc13aad86 button');\n",
       "      quickchartButtonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "    })();\n",
       "  </script>\n",
       "</div>\n",
       "\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "dataframe",
       "variable_name": "df",
       "summary": "{\n  \"name\": \"df\",\n  \"rows\": 153,\n  \"fields\": [\n    {\n      \"column\": \"edist\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 39.188939487434894,\n        \"min\": -0.06326336520100995,\n        \"max\": 264.8508484992351,\n        \"num_unique_values\": 153,\n        \"samples\": [\n          35.4895776855996,\n          20.55357913573583,\n          29.361116291214458\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"pvalue\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.08991843371757663,\n        \"min\": 0.0001,\n        \"max\": 1.0,\n        \"num_unique_values\": 7,\n        \"samples\": [\n          0.0001,\n          0.0282,\n          0.0014\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"significant\",\n      \"properties\": {\n        \"dtype\": \"boolean\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          false,\n          true\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"pvalue_adj\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.09881447234228133,\n        \"min\": 0.015184303087739362,\n        \"max\": 1.0,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          0.08223670576799999,\n          1.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"significant_adj\",\n      \"properties\": {\n        \"dtype\": \"boolean\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          false,\n          true\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
      }
     },
     "metadata": {},
     "execution_count": 4
    }
   ]
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-23T22:52:55.515005Z",
     "start_time": "2024-09-23T22:52:55.492973Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# open etest results\n",
    "df = pd.read_csv('etest_results_single_perts.csv')\n",
    "df"
   ],
   "outputs": [
    {
     "data": {
      "text/plain": [
       "   Unnamed: 0      edist  pvalue  significant  pvalue_adj  significant_adj\n",
       "0         AHR  21.757839  0.0001         True    0.008464             True\n",
       "1      ARRDC3   6.931206  0.0001         True    0.008464             True\n",
       "2        ATL1  60.535701  0.0001         True    0.008464             True\n",
       "3        BAK1   0.994366  0.0104         True    0.030877             True\n",
       "4     BCL2L11   0.082736  0.2923        False    0.499161            False\n",
       "..        ...        ...     ...          ...         ...              ...\n",
       "80      ZBTB1  27.508356  0.0001         True    0.008464             True\n",
       "81     ZBTB25  13.570649  0.0001         True    0.008464             True\n",
       "82    ZC3HAV1   3.710965  0.0001         True    0.008464             True\n",
       "83     ZNF318   2.619759  0.0001         True    0.008464             True\n",
       "84    control   0.000000  1.0000        False    1.000000            False\n",
       "\n",
       "[85 rows x 6 columns]"
      ],
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>edist</th>\n",
       "      <th>pvalue</th>\n",
       "      <th>significant</th>\n",
       "      <th>pvalue_adj</th>\n",
       "      <th>significant_adj</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AHR</td>\n",
       "      <td>21.757839</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>True</td>\n",
       "      <td>0.008464</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ARRDC3</td>\n",
       "      <td>6.931206</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>True</td>\n",
       "      <td>0.008464</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ATL1</td>\n",
       "      <td>60.535701</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>True</td>\n",
       "      <td>0.008464</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BAK1</td>\n",
       "      <td>0.994366</td>\n",
       "      <td>0.0104</td>\n",
       "      <td>True</td>\n",
       "      <td>0.030877</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BCL2L11</td>\n",
       "      <td>0.082736</td>\n",
       "      <td>0.2923</td>\n",
       "      <td>False</td>\n",
       "      <td>0.499161</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80</th>\n",
       "      <td>ZBTB1</td>\n",
       "      <td>27.508356</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>True</td>\n",
       "      <td>0.008464</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81</th>\n",
       "      <td>ZBTB25</td>\n",
       "      <td>13.570649</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>True</td>\n",
       "      <td>0.008464</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82</th>\n",
       "      <td>ZC3HAV1</td>\n",
       "      <td>3.710965</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>True</td>\n",
       "      <td>0.008464</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>ZNF318</td>\n",
       "      <td>2.619759</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>True</td>\n",
       "      <td>0.008464</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>84</th>\n",
       "      <td>control</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0000</td>\n",
       "      <td>False</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>85 rows × 6 columns</p>\n",
       "</div>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 12
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-23T16:14:55.340826Z",
     "start_time": "2024-09-23T16:14:55.105446Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# plot histogram of edist values\n",
    "df['edist'].hist()"
   ],
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 7
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-23T16:23:20.837405Z",
     "start_time": "2024-09-23T16:23:20.824939Z"
    }
   },
   "cell_type": "code",
   "source": [
    "df = df.rename(columns={'Unnamed: 0': 'perturbation'})\n",
    "# remove perturbation 'control' from the df\n",
    "df = df[df['perturbation'] != 'control']\n",
    "\n",
    "# calculate the mean e-distance across the entire dataset (column 'edist') and the associated p-value (column 'pvalue_adj').\n",
    "df['edist'].mean()"
   ],
   "outputs": [
    {
     "data": {
      "text/plain": [
       "26.777577459631395"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 8
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-23T16:23:26.362464Z",
     "start_time": "2024-09-23T16:23:26.357064Z"
    }
   },
   "cell_type": "code",
   "source": [
    "from scipy.stats import combine_pvalues\n",
    "\n",
    "# Assuming df is your DataFrame and 'pvalue_adj' is the column with p-values\n",
    "p_values = df['pvalue_adj'].values\n",
    "\n",
    "# Use Fisher's method to combine p-values\n",
    "_, combined_p_value = combine_pvalues(p_values, method='fisher')\n",
    "\n",
    "print(f\"Combined p-value: {combined_p_value}\")"
   ],
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Combined p-value: 1.2252751342846089e-80\n"
     ]
    }
   ],
   "execution_count": 9
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "## Calculate sigma values for perturbation "
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-25T20:17:17.847121Z",
     "start_time": "2024-09-25T20:17:17.834899Z"
    }
   },
   "cell_type": "code",
   "source": [
    "boolean_idx = adata.obs['guide_ids'] == 'CEBPA'\n",
    "adata.shape"
   ],
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(111445, 17414)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 46
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-23T23:05:51.245997Z",
     "start_time": "2024-09-23T23:05:51.212804Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# extract the expression for 'CEBPA' from adata \n",
    "cebpa_expr = adata[boolean_idx,:].X.toarray()"
   ],
   "outputs": [],
   "execution_count": 31
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-25T20:18:53.972168Z",
     "start_time": "2024-09-25T20:18:52.630488Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# calculate the sum of the euclidean distances between all the cells\n",
    "all_edist = []\n",
    "for i in range(cebpa_expr.shape[0]):\n",
    "    for j in range(cebpa_expr.shape[0]):\n",
    "        if i != j:\n",
    "            all_edist.append(np.linalg.norm(cebpa_expr[i] - cebpa_expr[j]))\n",
    "\n",
    "sigma_y_cebpa = sum(all_edist) / (cebpa_expr.shape[0] ** 2)\n",
    "sigma_y_cebpa"
   ],
   "outputs": [
    {
     "data": {
      "text/plain": [
       "62.9345994717863"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 54
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-25T20:17:28.796238Z",
     "start_time": "2024-09-25T20:17:28.428254Z"
    }
   },
   "cell_type": "code",
   "source": [
    "boolean_idx_ikzf3 = adata.obs['guide_ids'] == 'IKZF3'\n",
    "# extract the expression for 'IKZF3' from adata\n",
    "ikzf3_expr = adata[boolean_idx_ikzf3,:].X.toarray()"
   ],
   "outputs": [],
   "execution_count": 48
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-25T20:17:38.798084Z",
     "start_time": "2024-09-25T20:17:38.776481Z"
    }
   },
   "cell_type": "code",
   "source": "ikzf3_expr.shape",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(681, 17414)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 50
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-25T20:18:50.718527Z",
     "start_time": "2024-09-25T20:18:47.867616Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# calculate the sum of the euclidean distances between all the cells\n",
    "all_edist_ikzf3 = []\n",
    "for i in range(ikzf3_expr.shape[0]):\n",
    "    for j in range(ikzf3_expr.shape[0]):\n",
    "        if i != j:\n",
    "            all_edist_ikzf3.append(np.linalg.norm(ikzf3_expr[i] - ikzf3_expr[j]))\n",
    "\n",
    "sigma_y_ikzf3 = sum(all_edist_ikzf3) / (ikzf3_expr.shape[0]) ** 2\n",
    "sigma_y_ikzf3"
   ],
   "outputs": [
    {
     "data": {
      "text/plain": [
       "54.687621580808994"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 53
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-23T23:15:42.816927Z",
     "start_time": "2024-09-23T23:15:42.764372Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# calculate the mean of cebpa and ikzf3 across all cells \n",
    "cebpa_mean = cebpa_expr.mean(axis=0)\n",
    "ikzf3_mean = ikzf3_expr.mean(axis=0)\n",
    "# sort the mean values descending\n",
    "cebpa_mean = np.sort(cebpa_mean)[::-1]\n",
    "ikzf3_mean = np.sort(ikzf3_mean)[::-1]"
   ],
   "outputs": [],
   "execution_count": 35
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-23T23:34:04.839382Z",
     "start_time": "2024-09-23T23:34:04.477241Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# plot a line plot of the mean expression of cebpa and ikzf3\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "x = np.arange(cebpa_mean.shape[0])\n",
    "plt.plot(x, cebpa_mean, label='CEBPA')\n",
    "plt.plot(x, ikzf3_mean, label='IKZF3')\n",
    "plt.yscale('log')\n",
    "plt.xlabel('Gene')\n",
    "plt.ylabel('Mean post-perturbation expression')\n",
    "plt.gca().set_facecolor('white')\n",
    "plt.grid(color='grey', linestyle='-', linewidth=0.1)\n",
    "# set all spines to black \n",
    "plt.gca().spines['bottom'].set_color('black')\n",
    "plt.gca().spines['top'].set_color('black')\n",
    "plt.gca().spines['right'].set_color('black')\n",
    "plt.gca().spines['left'].set_color('black')\n",
    "plt.legend(facecolor='white')\n",
    "plt.savefig('../plots/paper_figs/CEBPA_IKZF3.pdf', dpi=300)"
   ],
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 45
  },
  {
   "metadata": {},
   "cell_type": "code",
   "outputs": [],
   "execution_count": null,
   "source": ""
  }
 ]
}