02_ChEMBL_SIMPD_Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "8028f691",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T10:29:14.387795Z",
     "start_time": "2023-08-16T10:29:14.383434Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The sql extension is already loaded. To reload it, use:\n",
      "  %reload_ext sql\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import gzip\n",
    "import pickle\n",
    "from collections import defaultdict\n",
    "\n",
    "%load_ext sql\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "plt.style.use('tableau-colorblind10')\n",
    "plt.rcParams['font.size'] = '16'\n",
    "\n",
    "figdir = '/home/glandrum/Papers/MolecularTimeSeries/images/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "908f42a8",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T07:37:12.057009Z",
     "start_time": "2023-08-16T07:37:11.705991Z"
    }
   },
   "outputs": [],
   "source": [
    "import intake\n",
    "import numpy as np\n",
    "import ga_lib_3 as ga_lib\n",
    "from rdkit import DataStructs\n",
    "\n",
    "def score_pareto_solutions(Fs,weights,q=0.9):\n",
    "    weights = np.array(weights)\n",
    "    Fs = np.copy(Fs)\n",
    "    qs = np.quantile(Fs,q,axis=0)\n",
    "    for i in range(len(qs)):\n",
    "        if qs[i]==0.0:\n",
    "            qs[i] = 1.0\n",
    "    Fs /= qs\n",
    "    Fs = np.exp(Fs*-1)\n",
    "    weights = np.array(weights,float)\n",
    "    # normalize:\n",
    "    weights /= np.sum(weights)\n",
    "    \n",
    "    Fs *= weights\n",
    "    return np.sum(Fs,axis=1)\n",
    "\n",
    "\n",
    "from sklearn.model_selection import StratifiedShuffleSplit\n",
    "def alt_compliance_analysis(target,strategy='CLUSTERS_SPLIT',popSize=500,nGens=300,\n",
    "                           isBalanced=True, noGap=True):\n",
    "\n",
    "    sel_strategy = getattr(ga_lib.SelectionStrategy,strategy)\n",
    "\n",
    "    bioactivities = intake.open_catalog('./datasets/targets.yaml')\n",
    "\n",
    "    if noGap:\n",
    "        dirname = \"nogap_\"\n",
    "        window = 0.0\n",
    "    else:\n",
    "        dirname = \"\"\n",
    "        window = 0.3\n",
    "    if not isBalanced:\n",
    "        dirname += 'alt_results_imbalanced'\n",
    "    else:\n",
    "        dirname += 'alt_results_balanced'\n",
    "    \n",
    "    \n",
    "    for scen in range(2):\n",
    "        fname = f'./{dirname}/{target}_{strategy}.{popSize}.{nGens}.altscenario_{scen}.pkl.gz'\n",
    "        if not os.path.exists(fname):\n",
    "            # short-circuit if we can't find results for one scenario\n",
    "            return {}\n",
    "\n",
    "    \n",
    "    ds = bioactivities[target]\n",
    "    print('------------\\nLoading data')\n",
    "    df = ds.read()\n",
    "    # randomize the order:\n",
    "    df = df.sample(frac=1,random_state=0xf00d)\n",
    "    \n",
    "    \n",
    "    # Use the binning scheme\n",
    "    # generate the binned activity values\n",
    "    pActs = df.pchembl_value\n",
    "    if not isBalanced:\n",
    "        upper, lower = ga_lib.get_imbalanced_bins(pActs)\n",
    "    else:\n",
    "        upper, lower = ga_lib.get_imbalanced_bins(pActs,tgt_frac=0.4,active_inactive_offset=window)\n",
    "    print(f'  bins are: {lower:.2f} {upper:.2f}')\n",
    "    df = df[(pActs >= upper) | (pActs <= lower)]\n",
    "    binned = [1 if x >= upper else 0 for x in df.pchembl_value]\n",
    "    df['active'] = ['active' if x == 1 else 'inactive' for x in binned]\n",
    "    print(f'  post-binning data size {len(df)}')\n",
    "    \n",
    "    dvals = np.array([ga_lib.calc_descrs(m) for m in df.mol])\n",
    "    dtgts = ga_lib.get_descr_targets()\n",
    "    fps = ga_lib.get_fps(df.mol)\n",
    "    dmat = np.zeros((len(fps),len(fps)),float)\n",
    "    for i,fp in enumerate(fps):\n",
    "        if i==len(fps)-1:\n",
    "            break\n",
    "        ds = np.array(DataStructs.BulkTanimotoSimilarity(fp,fps[i+1:],returnDistance=1))\n",
    "        dmat[i,i+1:] = ds\n",
    "        dmat[i+1:,i] = ds\n",
    "    print(len(df))\n",
    "       \n",
    "    max_pts = len(binned)\n",
    "    keep = max_pts\n",
    "    n_max = int(0.2*keep)\n",
    "    print(f'working with {keep} points and picking {n_max}')\n",
    "\n",
    "    \n",
    "    binned = np.array(binned)\n",
    "\n",
    "    res = {}\n",
    "\n",
    "    for scen in range(2):\n",
    "        fname = f'./{dirname}/{target}_{strategy}.{popSize}.{nGens}.altscenario_{scen}.pkl.gz'\n",
    "        if not os.path.exists(fname):\n",
    "            break\n",
    "        with gzip.open(fname,'rb') as inf:\n",
    "            train_inds,test_inds = pickle.load(inf)\n",
    "            Fs,Gs,Xs = pickle.load(inf)\n",
    "\n",
    "        tgts = [x.target_value for x in ga_lib.descrs]\n",
    "        descr_names = [x.descriptor for x in ga_lib.descrs]\n",
    "        nDescrs = len(descr_names)\n",
    "        descr_names.append('delta_frac_active')\n",
    "        tgts.append(ga_lib.delta_test_active_frac_vals[scen])\n",
    "\n",
    "        vals = []\n",
    "        for x in Xs:\n",
    "            train = np.median(dvals[~x], axis=0)\n",
    "            test = np.median(dvals[x], axis=0)\n",
    "            descr_deltas = test - train\n",
    "            lres = list(descr_deltas)\n",
    "            \n",
    "            train_acts = binned[~x]\n",
    "            train_frac = np.sum(train_acts, axis=0) / len(train_acts)\n",
    "            test_acts = binned[x]\n",
    "            test_frac = np.sum(test_acts, axis=0) / len(test_acts)\n",
    "            dTestFracActive = test_frac - np.sum(binned, axis=0)/len(binned)\n",
    "            lres.append(dTestFracActive)\n",
    "            \n",
    "            allIdx = np.arange(0, len(x), dtype=int)\n",
    "            testIdx = allIdx[x]\n",
    "            trainIdx = allIdx[~x]\n",
    "            _, g_vals, f_vals, _ = ga_lib.modified_spatial_stats_dmat(\n",
    "                dmat, testIdx, trainIdx, includeTestInBootstrap=False)\n",
    "            sum_F = np.sum(f_vals)\n",
    "            sum_G = np.sum(g_vals)\n",
    "            lres.append(sum_F)\n",
    "            lres.append(sum_G)           \n",
    "            vals.append(lres)\n",
    "        res[scen] = (tgts,np.array(vals))\n",
    "    # short circuit if we couldn't read in all the results\n",
    "    if len(res)<2:\n",
    "        return res\n",
    "    \n",
    "    # random split results\n",
    "    sss = StratifiedShuffleSplit(n_splits = 10, test_size=0.2, random_state=0xf00d)\n",
    "    fvs = []\n",
    "    gvs = []\n",
    "    for train,test in sss.split(fps,binned):\n",
    "        _, g_vals, f_vals, _ = ga_lib.modified_spatial_stats_dmat(\n",
    "            dmat, test, train, includeTestInBootstrap=False)\n",
    "        sum_F = np.sum(f_vals)\n",
    "        sum_G = np.sum(g_vals)\n",
    "        fvs.append(sum_F)\n",
    "        gvs.append(sum_G)\n",
    "    res[-1] = (fvs,gvs)\n",
    "\n",
    "    \n",
    "    return res\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "dca91325",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T07:40:37.482135Z",
     "start_time": "2023-08-16T07:37:14.129878Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "------------\n",
      "Loading data\n",
      "  bins are: 7.10 7.10\n",
      "  post-binning data size 996\n",
      "996\n",
      "working with 996 points and picking 199\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.50 7.50\n",
      "  post-binning data size 969\n",
      "969\n",
      "working with 969 points and picking 193\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.30 7.30\n",
      "  post-binning data size 925\n",
      "925\n",
      "working with 925 points and picking 185\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.20 7.20\n",
      "  post-binning data size 907\n",
      "907\n",
      "working with 907 points and picking 181\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.80 7.80\n",
      "  post-binning data size 881\n",
      "881\n",
      "working with 881 points and picking 176\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.50 7.50\n",
      "  post-binning data size 861\n",
      "861\n",
      "working with 861 points and picking 172\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.60 7.60\n",
      "  post-binning data size 858\n",
      "858\n",
      "working with 858 points and picking 171\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.60 7.60\n",
      "  post-binning data size 837\n",
      "837\n",
      "working with 837 points and picking 167\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.40 7.40\n",
      "  post-binning data size 830\n",
      "830\n",
      "working with 830 points and picking 166\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.10 8.10\n",
      "  post-binning data size 749\n",
      "749\n",
      "working with 749 points and picking 149\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.90 6.90\n",
      "  post-binning data size 732\n",
      "732\n",
      "working with 732 points and picking 146\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.40 8.40\n",
      "  post-binning data size 723\n",
      "723\n",
      "working with 723 points and picking 144\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.80 7.80\n",
      "  post-binning data size 687\n",
      "687\n",
      "working with 687 points and picking 137\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.70 8.70\n",
      "  post-binning data size 694\n",
      "694\n",
      "working with 694 points and picking 138\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.40 8.40\n",
      "  post-binning data size 690\n",
      "690\n",
      "working with 690 points and picking 138\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.80 7.80\n",
      "  post-binning data size 690\n",
      "690\n",
      "working with 690 points and picking 138\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.60 8.60\n",
      "  post-binning data size 686\n",
      "686\n",
      "working with 686 points and picking 137\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.10 6.10\n",
      "  post-binning data size 678\n",
      "678\n",
      "working with 678 points and picking 135\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.20 7.20\n",
      "  post-binning data size 642\n",
      "642\n",
      "working with 642 points and picking 128\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.50 7.50\n",
      "  post-binning data size 640\n",
      "640\n",
      "working with 640 points and picking 128\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.90 7.90\n",
      "  post-binning data size 630\n",
      "630\n",
      "working with 630 points and picking 126\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.60 6.60\n",
      "  post-binning data size 627\n",
      "627\n",
      "working with 627 points and picking 125\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.70 7.70\n",
      "  post-binning data size 619\n",
      "619\n",
      "working with 619 points and picking 123\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 5.80 5.80\n",
      "  post-binning data size 606\n",
      "606\n",
      "working with 606 points and picking 121\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.50 6.50\n",
      "  post-binning data size 592\n",
      "592\n",
      "working with 592 points and picking 118\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.60 7.60\n",
      "  post-binning data size 585\n",
      "585\n",
      "working with 585 points and picking 117\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.70 8.70\n",
      "  post-binning data size 583\n",
      "583\n",
      "working with 583 points and picking 116\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.00 7.00\n",
      "  post-binning data size 580\n",
      "580\n",
      "working with 580 points and picking 116\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.10 6.10\n",
      "  post-binning data size 579\n",
      "579\n",
      "working with 579 points and picking 115\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.90 7.90\n",
      "  post-binning data size 593\n",
      "593\n",
      "working with 593 points and picking 118\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.00 8.00\n",
      "  post-binning data size 576\n",
      "576\n",
      "working with 576 points and picking 115\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.70 6.70\n",
      "  post-binning data size 543\n",
      "543\n",
      "working with 543 points and picking 108\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.40 6.40\n",
      "  post-binning data size 542\n",
      "542\n",
      "working with 542 points and picking 108\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.20 7.20\n",
      "  post-binning data size 524\n",
      "524\n",
      "working with 524 points and picking 104\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.30 7.30\n",
      "  post-binning data size 669\n",
      "669\n",
      "working with 669 points and picking 133\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.90 7.90\n",
      "  post-binning data size 518\n",
      "518\n",
      "working with 518 points and picking 103\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.80 7.80\n",
      "  post-binning data size 515\n",
      "515\n",
      "working with 515 points and picking 103\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.10 8.10\n",
      "  post-binning data size 507\n",
      "507\n",
      "working with 507 points and picking 101\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.40 8.40\n",
      "  post-binning data size 505\n",
      "505\n",
      "working with 505 points and picking 101\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.60 6.60\n",
      "  post-binning data size 498\n",
      "498\n",
      "working with 498 points and picking 99\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.50 7.50\n",
      "  post-binning data size 485\n",
      "485\n",
      "working with 485 points and picking 97\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.80 6.80\n",
      "  post-binning data size 491\n",
      "491\n",
      "working with 491 points and picking 98\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.20 7.20\n",
      "  post-binning data size 480\n",
      "480\n",
      "working with 480 points and picking 96\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.90 7.90\n",
      "  post-binning data size 476\n",
      "476\n",
      "working with 476 points and picking 95\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 9.10 9.10\n",
      "  post-binning data size 463\n",
      "463\n",
      "working with 463 points and picking 92\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 9.30 9.30\n",
      "  post-binning data size 467\n",
      "467\n",
      "working with 467 points and picking 93\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.00 7.00\n",
      "  post-binning data size 443\n",
      "443\n",
      "working with 443 points and picking 88\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.30 8.30\n",
      "  post-binning data size 442\n",
      "442\n",
      "working with 442 points and picking 88\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.10 8.10\n",
      "  post-binning data size 434\n",
      "434\n",
      "working with 434 points and picking 86\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.70 7.70\n",
      "  post-binning data size 431\n",
      "431\n",
      "working with 431 points and picking 86\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.10 8.10\n",
      "  post-binning data size 455\n",
      "455\n",
      "working with 455 points and picking 91\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.40 7.40\n",
      "  post-binning data size 423\n",
      "423\n",
      "working with 423 points and picking 84\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.40 8.40\n",
      "  post-binning data size 587\n",
      "587\n",
      "working with 587 points and picking 117\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.40 8.40\n",
      "  post-binning data size 413\n",
      "413\n",
      "working with 413 points and picking 82\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.30 7.30\n",
      "  post-binning data size 585\n",
      "585\n",
      "working with 585 points and picking 117\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.60 7.60\n",
      "  post-binning data size 524\n",
      "524\n",
      "working with 524 points and picking 104\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.80 7.80\n",
      "  post-binning data size 407\n",
      "407\n",
      "working with 407 points and picking 81\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.80 7.80\n",
      "  post-binning data size 399\n",
      "399\n",
      "working with 399 points and picking 79\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.30 7.30\n",
      "  post-binning data size 397\n",
      "397\n",
      "working with 397 points and picking 79\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.60 7.60\n",
      "  post-binning data size 397\n",
      "397\n",
      "working with 397 points and picking 79\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.20 7.20\n",
      "  post-binning data size 393\n",
      "393\n",
      "working with 393 points and picking 78\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.30 8.30\n",
      "  post-binning data size 386\n",
      "386\n",
      "working with 386 points and picking 77\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.90 7.90\n",
      "  post-binning data size 380\n",
      "380\n",
      "working with 380 points and picking 76\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.70 7.70\n",
      "  post-binning data size 438\n",
      "438\n",
      "working with 438 points and picking 87\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.20 7.20\n",
      "  post-binning data size 374\n",
      "374\n",
      "working with 374 points and picking 74\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.20 7.20\n",
      "  post-binning data size 371\n",
      "371\n",
      "working with 371 points and picking 74\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.00 6.00\n",
      "  post-binning data size 364\n",
      "364\n",
      "working with 364 points and picking 72\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.20 7.20\n",
      "  post-binning data size 365\n",
      "365\n",
      "working with 365 points and picking 73\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "------------\n",
      "Loading data\n",
      "  bins are: 7.50 7.50\n",
      "  post-binning data size 364\n",
      "364\n",
      "working with 364 points and picking 72\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.70 7.70\n",
      "  post-binning data size 360\n",
      "360\n",
      "working with 360 points and picking 72\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.60 6.60\n",
      "  post-binning data size 360\n",
      "360\n",
      "working with 360 points and picking 72\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.10 8.10\n",
      "  post-binning data size 358\n",
      "358\n",
      "working with 358 points and picking 71\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.30 6.30\n",
      "  post-binning data size 358\n",
      "358\n",
      "working with 358 points and picking 71\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.00 7.00\n",
      "  post-binning data size 357\n",
      "357\n",
      "working with 357 points and picking 71\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.50 7.50\n",
      "  post-binning data size 355\n",
      "355\n",
      "working with 355 points and picking 71\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.60 6.60\n",
      "  post-binning data size 354\n",
      "354\n",
      "working with 354 points and picking 70\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.60 7.60\n",
      "  post-binning data size 351\n",
      "351\n",
      "working with 351 points and picking 70\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.60 7.60\n",
      "  post-binning data size 350\n",
      "350\n",
      "working with 350 points and picking 70\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.10 7.10\n",
      "  post-binning data size 348\n",
      "348\n",
      "working with 348 points and picking 69\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.50 8.50\n",
      "  post-binning data size 347\n",
      "347\n",
      "working with 347 points and picking 69\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.80 7.80\n",
      "  post-binning data size 347\n",
      "347\n",
      "working with 347 points and picking 69\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 6.60 6.60\n",
      "  post-binning data size 346\n",
      "346\n",
      "working with 346 points and picking 69\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 5.80 5.80\n",
      "  post-binning data size 350\n",
      "350\n",
      "working with 350 points and picking 70\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.10 7.10\n",
      "  post-binning data size 338\n",
      "338\n",
      "working with 338 points and picking 67\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.40 7.40\n",
      "  post-binning data size 333\n",
      "333\n",
      "working with 333 points and picking 66\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.10 7.10\n",
      "  post-binning data size 396\n",
      "396\n",
      "working with 396 points and picking 79\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.90 7.90\n",
      "  post-binning data size 467\n",
      "467\n",
      "working with 467 points and picking 93\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.90 7.90\n",
      "  post-binning data size 431\n",
      "431\n",
      "working with 431 points and picking 86\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.10 8.10\n",
      "  post-binning data size 326\n",
      "326\n",
      "working with 326 points and picking 65\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.10 8.10\n",
      "  post-binning data size 322\n",
      "322\n",
      "working with 322 points and picking 64\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.10 7.10\n",
      "  post-binning data size 318\n",
      "318\n",
      "working with 318 points and picking 63\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.50 7.50\n",
      "  post-binning data size 316\n",
      "316\n",
      "working with 316 points and picking 63\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.20 7.20\n",
      "  post-binning data size 378\n",
      "378\n",
      "working with 378 points and picking 75\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.90 7.90\n",
      "  post-binning data size 311\n",
      "311\n",
      "working with 311 points and picking 62\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 8.30 8.30\n",
      "  post-binning data size 404\n",
      "404\n",
      "working with 404 points and picking 80\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.20 7.20\n",
      "  post-binning data size 308\n",
      "308\n",
      "working with 308 points and picking 61\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.60 7.60\n",
      "  post-binning data size 307\n",
      "307\n",
      "working with 307 points and picking 61\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.50 7.50\n",
      "  post-binning data size 301\n",
      "301\n",
      "working with 301 points and picking 60\n",
      "------------\n",
      "Loading data\n",
      "  bins are: 7.00 7.00\n",
      "  post-binning data size 300\n",
      "300\n",
      "working with 300 points and picking 60\n"
     ]
    }
   ],
   "source": [
    "bioactivities = intake.open_catalog('./datasets/targets.yaml')\n",
    "\n",
    "alt_alltgts = defaultdict(list)\n",
    "alt_alldesc = defaultdict(list)\n",
    "random_fgvals = defaultdict(list)\n",
    "for target in bioactivities:\n",
    "    ares = alt_compliance_analysis(target)\n",
    "    if 0 not in ares or 1 not in ares:\n",
    "        print(f'skipping {target}')\n",
    "        continue\n",
    "    random_fgvals[-1].append(ares[-1])\n",
    "    \n",
    "    for scen in (0,1):\n",
    "        alt_alltgts[scen].append(ares[scen][0])\n",
    "        alt_alldesc[scen].append(ares[scen][1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "acfa044a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T07:40:37.616830Z",
     "start_time": "2023-08-16T07:40:37.486699Z"
    }
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('./chembl_alt_moo_objectives.pkl','wb+') as outf:\n",
    "    pickle.dump((alt_alltgts,alt_alldesc,random_fgvals),outf)\n",
    "!chmod gou-w ./chembl_alt_moo_objectives.pkl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3a249596",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-15T13:24:39.453278Z",
     "start_time": "2023-08-15T13:24:39.453270Z"
    }
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "with open('./chembl_alt_moo_objectives.pkl','rb') as inf:\n",
    "    (alt_alltgts,alt_alldesc,random_fgvals) = pickle.load(inf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ac86c212",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T07:41:56.158814Z",
     "start_time": "2023-08-16T07:41:55.557377Z"
    }
   },
   "outputs": [],
   "source": [
    "targets = intake.open_catalog('./datasets/targets.yaml')\n",
    "\n",
    "strategy='CLUSTERS_SPLIT'\n",
    "popSize=500\n",
    "nGens=300\n",
    "ares0 = defaultdict(list)\n",
    "best_F0 = defaultdict(list)\n",
    "ares1 = defaultdict(list)\n",
    "best_F1 = defaultdict(list)\n",
    "ares2 = defaultdict(list)\n",
    "best_F2 = defaultdict(list)\n",
    "idx = [0]\n",
    "for target in targets:\n",
    "    for scen in range(len(idx)):\n",
    "        fname = f'./nogap_alt_results_balanced/{target}_{strategy}.{popSize}.{nGens}.altscenario_{scen}.pkl.gz'\n",
    "        if not os.path.exists(fname):\n",
    "            continue\n",
    "        with gzip.open(fname,'rb') as inf:\n",
    "            train_inds,test_inds = pickle.load(inf)\n",
    "            Fs,Gs,Xs = pickle.load(inf)\n",
    "        scores1 = score_pareto_solutions(Fs,weights=[1.]*7)\n",
    "        scores2 = score_pareto_solutions(Fs,weights=[1.]*5+[10.,5.])\n",
    "        best1 = np.argmax(scores1)\n",
    "        best2 = np.argmax(scores2)\n",
    "        \n",
    "        best_F0[scen].append(Fs[0])\n",
    "        best_F1[scen].append(Fs[best1])\n",
    "        best_F2[scen].append(Fs[best2])\n",
    "        \n",
    "        ares0[scen].append(alt_alldesc[scen][idx[scen]][0])\n",
    "        ares1[scen].append(alt_alldesc[scen][idx[scen]][best1])\n",
    "        ares2[scen].append(alt_alldesc[scen][idx[scen]][best2])\n",
    "        idx[scen] += 1\n",
    "\n",
    "for k in ares1:\n",
    "    for d in (ares0,ares1,ares2,best_F0,best_F1,best_F2):\n",
    "        d[k] = np.array(d[k])\n",
    "        \n",
    "\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "2d59b606",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T07:41:58.540048Z",
     "start_time": "2023-08-16T07:41:58.529376Z"
    }
   },
   "outputs": [],
   "source": [
    "randomsplit_fvals = [np.median(x) for x,y in random_fgvals[-1]]\n",
    "randomsplit_gvals = [np.median(y) for x,y in random_fgvals[-1]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "187db3c6",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T07:41:59.950882Z",
     "start_time": "2023-08-16T07:41:59.698062Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "       \n",
    "plt.figure(figsize=(6,6))\n",
    "plt.scatter(ares0[0][:,6],ares0[0][:,5],marker='s',color='C0',label='Scenario 1');\n",
    "\n",
    "plt.plot((20,100),(20,100),'k--')\n",
    "\n",
    "    \n",
    "    \n",
    "plt.xlabel('$\\sum G$');\n",
    "plt.ylabel('$\\sum F\\'$');\n",
    "plt.title('First solution');\n",
    "plt.legend();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "e7d63316",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T07:42:01.118983Z",
     "start_time": "2023-08-16T07:42:01.028572Z"
    },
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "       \n",
    "plt.figure(figsize=(6,6))\n",
    "plt.scatter(ares1[0][:,6],ares1[0][:,5],marker='s',color='C0',label='Scenario 1');\n",
    "\n",
    "plt.plot((20,100),(20,100),'k--')\n",
    "\n",
    "    \n",
    "    \n",
    "plt.xlabel('$\\sum G$');\n",
    "plt.ylabel('$\\sum F\\'$');\n",
    "plt.title('Equal weights');\n",
    "plt.legend();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "db584176",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T07:42:02.437168Z",
     "start_time": "2023-08-16T07:42:02.352759Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "       \n",
    "plt.figure(figsize=(6,6))\n",
    "plt.scatter(ares2[0][:,6],ares2[0][:,5],marker='s',color='C0',label='Scenario 1');\n",
    "\n",
    "plt.plot((20,100),(20,100),'k--')\n",
    "\n",
    "    \n",
    "    \n",
    "plt.xlabel('$\\sum G$');\n",
    "plt.ylabel('$\\sum F\\'$');\n",
    "plt.title('Favor F&G');\n",
    "plt.legend();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "100c4126",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T11:18:32.854825Z",
     "start_time": "2023-08-16T11:18:32.573990Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "randomsplit_fvals = [np.median(x) for x,y in random_fgvals[-1]]\n",
    "randomsplit_gvals = [np.median(y) for x,y in random_fgvals[-1]]\n",
    "\n",
    "plt.figure(figsize=(6,6))\n",
    "plt.scatter(ares2[0][:,6],ares2[0][:,5],marker='o',color='C1',label='SIMPD');\n",
    "plt.scatter(randomsplit_gvals,randomsplit_fvals,marker='^',color='C2',label='Random');\n",
    "\n",
    "plt.plot((20,90),(20,90),'k-')\n",
    "\n",
    "    \n",
    "    \n",
    "plt.xlabel('$\\sum G$');\n",
    "plt.ylabel('$\\sum F\\'$');\n",
    "plt.legend();\n",
    "plt.savefig(f'{figdir}/chembl_assays_spatial_stats.pdf',bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "50c0663c",
   "metadata": {},
   "source": [
    "## Descriptors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "d60f8d1b",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T11:23:14.132468Z",
     "start_time": "2023-08-16T11:23:13.601539Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "descrs = [x.descriptor for x in ga_lib.descrs]\n",
    "plt.figure(figsize=(12,12))\n",
    "\n",
    "for i in range(4):\n",
    "    plt.subplot(2,2,i+1)\n",
    "    plt.hist([best_F2[0][:,i]],label=('F&G',))\n",
    "    plt.xlabel(f'$\\Delta$ {descrs[i]}');\n",
    "plt.savefig(f'{figdir}/chembl_assays_descriptor_match.pdf',bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "139b8095",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T12:07:28.200954Z",
     "start_time": "2023-08-16T12:07:27.535150Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x1296 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "randomsplit_fvals = [np.median(x) for x,y in random_fgvals[-1]]\n",
    "randomsplit_gvals = [np.median(y) for x,y in random_fgvals[-1]]\n",
    "descrs = [x.descriptor for x in ga_lib.descrs]\n",
    "\n",
    "\n",
    "from matplotlib.gridspec import GridSpec\n",
    "fig = plt.figure(figsize=(12,18))\n",
    "\n",
    "\n",
    "gs = GridSpec(3,4,figure=fig)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "ax1 = fig.add_subplot(gs[0,1:3])\n",
    "\n",
    "ax1.text(-.5,1,'A',transform=ax1.transAxes,fontsize=24)\n",
    "\n",
    "ax1.scatter(ares2[0][:,6],ares2[0][:,5],marker='o',color='C1',label='SIMPD');\n",
    "ax1.scatter(randomsplit_gvals,randomsplit_fvals,marker='^',color='C2',label='Random');\n",
    "\n",
    "ax1.plot((20,90),(20,90),'k-')\n",
    "\n",
    "    \n",
    "    \n",
    "ax1.set_xlabel('$\\sum G$');\n",
    "ax1.set_ylabel('$\\sum F\\'$');\n",
    "ax1.legend();\n",
    "# plt.savefig(f'{figdir}/chembl_assays_spatial_stats.pdf',bbox_inches='tight')\n",
    "\n",
    "\n",
    "ax1.text(-.5,-.15,'B',transform=ax1.transAxes,fontsize=24)\n",
    "\n",
    "ax2 = fig.add_subplot(gs[1,0:2])\n",
    "ax2.hist([best_F2[0][:,0]],label=('F&G',))\n",
    "ax2.set_xlabel(f'$\\Delta$ {descrs[0]}');\n",
    "ax2 = fig.add_subplot(gs[1,2:4])\n",
    "ax2.hist([best_F2[0][:,1]],label=('F&G',))\n",
    "ax2.set_xlabel(f'$\\Delta$ {descrs[1]}');\n",
    "\n",
    "ax2 = fig.add_subplot(gs[2,0:2])\n",
    "ax2.hist([best_F2[0][:,2]],label=('F&G',))\n",
    "ax2.set_xlabel(f'$\\Delta$ {descrs[2]}');\n",
    "ax2 = fig.add_subplot(gs[2,2:4])\n",
    "ax2.hist([best_F2[0][:,3]],label=('F&G',))\n",
    "ax2.set_xlabel(f'$\\Delta$ {descrs[3]}');\n",
    "\n",
    "plt.savefig(f'{figdir}/chembl_simpd_results_summary.pdf',bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "650320ba",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c7186830",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "cc438aaa",
   "metadata": {},
   "source": [
    "# ML Performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "01100410",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T07:42:11.934764Z",
     "start_time": "2023-08-16T07:42:11.905739Z"
    }
   },
   "outputs": [],
   "source": [
    "from collections import namedtuple\n",
    "restuple = namedtuple('restuple',['rocs','scores','kappas','BAs'])\n",
    "\n",
    "import pickle\n",
    "import glob\n",
    "import os\n",
    "import gzip\n",
    "tmp_assays = glob.glob('./mfp0_nogap_ml_results_balanced/*.pkl.gz')\n",
    "mfp0_nogap_random_results_bal = {}\n",
    "for fn in tmp_assays:\n",
    "    assay = os.path.basename(fn).split('_')[0]\n",
    "    with gzip.open(fn,'rb') as inf:\n",
    "        tpl = restuple(*pickle.load(inf))\n",
    "    mfp0_nogap_random_results_bal[assay] = tpl\n",
    "\n",
    "tmp_assays = glob.glob('./nogap_ml_results_balanced/*.pkl.gz')\n",
    "nogap_random_results_bal = {}\n",
    "for fn in tmp_assays:\n",
    "    assay = os.path.basename(fn).split('_')[0]\n",
    "    with gzip.open(fn,'rb') as inf:\n",
    "        tpl = restuple(*pickle.load(inf))\n",
    "    nogap_random_results_bal[assay] = tpl\n",
    "    \n",
    "   \n",
    "fns = glob.glob('./nogap_alt_moo_ml_results_balanced/CHEMBL*.altscenario_0.favorfg_rf.pkl.gz')\n",
    "results_moo_fg_bal = {}\n",
    "for fn in fns:\n",
    "    assay = os.path.basename(fn).split('_')[0]\n",
    "    with gzip.open(fn,'rb') as inf:\n",
    "        tpl = restuple(*pickle.load(inf))\n",
    "    results_moo_fg_bal[assay] = tpl\n",
    "\n",
    "fns = glob.glob('./nogap_alt_moo_ml_results_balanced/mfp0_CHEMBL*.altscenario_0.favorfg_rf.pkl.gz')\n",
    "mfp0_results_moo_fg_bal = {}\n",
    "for fn in fns:\n",
    "    assay = os.path.basename(fn).split('_')[1]\n",
    "    with gzip.open(fn,'rb') as inf:\n",
    "        tpl = restuple(*pickle.load(inf))\n",
    "    mfp0_results_moo_fg_bal[assay] = tpl\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "b67a2739",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T07:42:14.428238Z",
     "start_time": "2023-08-16T07:42:14.290809Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "random_meds = [np.median(mfp0_nogap_random_results_bal[x].rocs) for x in nogap_random_results_bal]\n",
    "#moo_meds = [np.median(results_moo_bal[x].rocs) for x in moo_assays]\n",
    "moo_fg_first = [mfp0_results_moo_fg_bal[x].rocs[0] for x in nogap_random_results_bal]\n",
    "\n",
    "plt.figure(figsize=(12,6))\n",
    "\n",
    "plt.subplot(1,2,1)\n",
    "plt.scatter(random_meds,moo_fg_first,label='favor F&G');\n",
    "plt.plot((.5,1),(.5,1),'k');\n",
    "plt.xlabel('random median(AUC)')\n",
    "plt.ylabel('SIMPD AUC');\n",
    "plt.title('MFP0')\n",
    "\n",
    "random_meds = [np.median(nogap_random_results_bal[x].rocs) for x in nogap_random_results_bal]\n",
    "#moo_meds = [np.median(results_moo_bal[x].rocs) for x in moo_assays]\n",
    "moo_fg_first = [results_moo_fg_bal[x].rocs[0] for x in nogap_random_results_bal]\n",
    "\n",
    "plt.subplot(1,2,2)\n",
    "plt.scatter(random_meds,moo_fg_first,label='favor F&G');\n",
    "plt.plot((.5,1),(.5,1),'k');\n",
    "plt.xlabel('random median(AUC)')\n",
    "plt.ylabel('SIMPD AUC');\n",
    "plt.title('MFP2');\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "093d2af1",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T07:42:23.129266Z",
     "start_time": "2023-08-16T07:42:23.057032Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mfp0_moo_fg_first = [mfp0_results_moo_fg_bal[x].rocs[0] for x in nogap_random_results_bal]\n",
    "moo_fg_first = [results_moo_fg_bal[x].rocs[0] for x in nogap_random_results_bal]\n",
    "\n",
    "plt.figure(figsize=(6,6))\n",
    "\n",
    "plt.scatter(mfp0_moo_fg_first,moo_fg_first,label='favor F&G');\n",
    "plt.plot((.5,1),(.5,1),'k');\n",
    "plt.xlabel('MFP0 AUC')\n",
    "plt.ylabel('MFP2 AUC');\n",
    "plt.title('SIMPD');\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "ee0df862",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-08-16T11:31:40.128874Z",
     "start_time": "2023-08-16T11:31:39.986646Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "random_meds = [np.median(nogap_random_results_bal[x].rocs) for x in nogap_random_results_bal]\n",
    "#moo_meds = [np.median(results_moo_bal[x].rocs) for x in moo_assays]\n",
    "moo_fg_first = [results_moo_fg_bal[x].rocs[0] for x in nogap_random_results_bal]\n",
    "\n",
    "plt.figure(figsize=(6,6))\n",
    "\n",
    "plt.scatter(random_meds,moo_fg_first,label='favor F&G');\n",
    "plt.plot((.5,1),(.5,1),'k');\n",
    "plt.xlabel('random median(AUC)')\n",
    "plt.ylabel('SIMPD AUC');\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "plt.savefig(f'{figdir}/chembl_assays_ga_impact_on_model_performance.pdf',bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8ec24656",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}