test more.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\ASUS\\anaconda3\\envs\\Pandas\\lib\\site-packages\\tqdm\\auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "import sys\n",
    "sys.path.append('./')\n",
    "sys.path.append('/Users/ASUS/Documents/GitHub/retrospective_t2e_analysis/auton-survival/')\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import torch\n",
    "from tqdm import tqdm \n",
    "from examples.cmhe_demo_utils import * \n",
    "from metrics import treatment_effect"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(9361, 27)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\ASUS\\Documents\\GitHub\\retrospective_t2e_analysis\\auton-survival\\auton_survival\\datasets_retrospective.py:14: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df.loc[:, col] = df[col].values.astype(str)\n",
      "c:\\Users\\ASUS\\Documents\\GitHub\\retrospective_t2e_analysis\\auton-survival\\auton_survival\\datasets_retrospective.py:117: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  intervention[intervention == 1.0] = 'SBP target <120mm Hg'\n",
      "c:\\Users\\ASUS\\Documents\\GitHub\\retrospective_t2e_analysis\\auton-survival\\auton_survival\\datasets_retrospective.py:117: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  intervention[intervention == 1.0] = 'SBP target <120mm Hg'\n",
      "c:\\Users\\ASUS\\Documents\\GitHub\\retrospective_t2e_analysis\\auton-survival\\auton_survival\\datasets_retrospective.py:118: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  intervention[intervention == 0.0] = 'SBP target <140mm Hg'\n",
      "c:\\Users\\ASUS\\Documents\\GitHub\\retrospective_t2e_analysis\\auton-survival\\auton_survival\\datasets_retrospective.py:118: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  intervention[intervention == 0.0] = 'SBP target <140mm Hg'\n",
      "c:\\Users\\ASUS\\Documents\\GitHub\\retrospective_t2e_analysis\\auton-survival\\auton_survival\\datasets_retrospective.py:122: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  outcomes.event = outcomes.event == 1.0\n"
     ]
    }
   ],
   "source": [
    "# Load dataset\n",
    "from datasets_retrospective import load_sprint_dataset\n",
    "outcomes, features, intervention, cat_feats, num_feats = load_sprint_dataset(location='datasets/SPRINT_POP/Datasets/', outcome='PRIMARY')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# intervention = intervention.dropna()\n",
    "# features = features.loc[intervention.dropna().index]\n",
    "# outcomes = outcomes.loc[intervention.dropna().index]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# best sprint\n",
    "# # Specify categorical and numeric features used for training the model\n",
    "# cat_feats=['Baseline_Smoke_status', 'Baseline_History_Cardiovascular_Disease']\n",
    "# num_feats=['Baseline_Glucose']\n",
    "# Oat \n",
    "# # Specify categorical and numeric features used for training the model\n",
    "# cat_feats=['Sex', 'Race', 'Baseline_New_York_Heart_Association_Highest_Class']\n",
    "# num_feats=['Baseline_Age']\n",
    "\n",
    "# # Specify categorical and numeric features used for training the model\n",
    "# cat_feats=['Sex', 'Race', 'Baseline_History_Diabetes']\n",
    "# num_feats=['Baseline_Age']\n",
    "\n",
    "# # Specify categorical and numeric features used for training the model\n",
    "# cat_feats=['Sex', 'Race', 'Baseline_Ejection_Fraction_50']\n",
    "# num_feats=['Baseline_Age']\n",
    "\n",
    "# # Specify categorical and numeric features used for training the model\n",
    "# cat_feats=['Sex', 'Race', 'Baseline_Two_or_more_Aspirin_Warfarin_Thienopyridine']\n",
    "# num_feats=['Baseline_Age']\n",
    "\n",
    "# best\n",
    "# Specify categorical and numeric features used for training the model\n",
    "# cat_feats=['Sex', 'Race', 'Baseline_One_or_more_Aspirin_Warfarin_Thienopyridine']\n",
    "# num_feats=['Baseline_Age']\n",
    "\n",
    "# best division\n",
    "# Specify categorical and numeric features used for training the model\n",
    "# cat_feats=['Sex', 'Race', 'Baseline_One_or_more_Aspirin_Warfarin_Thienopyridine', 'Baseline_History_Diabetes']\n",
    "# num_feats=['Baseline_Age']\n",
    "\n",
    "# best bari 2d\n",
    "# Specify categorical and numeric features used for training the model\n",
    "# cat_feats=['Sex', 'Race', 'Baseline_Beta_Blocker', \n",
    "#           'Baseline_Calcium_Blocker']\n",
    "# num_feats=['Baseline_Age', 'Baseline_Standing_Diastolic_Blood_Pressure', 'Baseline_Standing_Systolic_Blood_Pressure']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # Specify categorical and numeric features used for training the model\n",
    "# cat_feats=['Race', 'Baseline_Smoke_status']\n",
    "# num_feats=['Baseline_Triglycerides']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Preprocess feature data\n",
    "from preprocessing import Preprocessor\n",
    "features_prep = Preprocessor().fit_transform(data=features[cat_feats+num_feats], cat_feats=cat_feats, num_feats=num_feats)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # Hyper-parameters to train CMHE model\n",
    "# k = 1 # number of underlying base survival phenotypes\n",
    "# g = 2 # number of underlying treatment effect phenotypes.\n",
    "# layers = [] # number of neurons in each hidden layer.\n",
    "# random_seed = 3\n",
    "# iters = 100 # number of training epochs\n",
    "# learning_rate = 0.001\n",
    "# batch_size = 256 \n",
    "# patience = 5\n",
    "# optimizer = \"Adam\"\n",
    "# gate_l2_penalty = 1e-3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Hyper-parameters to train SPARSE model\n",
    "k = 2 # number of underlying base survival phenotypes\n",
    "random_seed = 105\n",
    "n_epochs = 100 # number of training epochs\n",
    "step_size = 1e-5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Restarts:   0%|          | 0/1 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 6508.711344934423 6925.493312645704 13434.204657580127\n",
      "268725.25117528246 4.218991794141097e-12\n",
      "\n",
      " 6002.584590188411 6606.164720713565 12608.749310901976\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "190953.23785403362 4.061365097787432e-22\n",
      "\n",
      " 5851.966995702225 6600.014751911833 12451.981747614058\n",
      "146594.09291958396 5.965299180732514e-25\n",
      "\n",
      " 5773.588155987114 6594.002097248267 12367.590253235381\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "113112.4126132048 5.143563501579123e-24\n",
      "\n",
      " 5717.9589071798155 6588.216549207175 12306.17545638699\n",
      "86485.80640862497 2.356542512142384e-22\n",
      "\n",
      " 5671.90267521349 6582.760149120031 12254.662824333522\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "69449.64860193766 9.20841400148631e-21\n",
      "\n",
      " 5632.128606793431 6577.603013740869 12209.7316205343\n",
      "59660.40319668294 2.6588845658803023e-19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 5597.220861905027 6572.723584015403 12169.94444592043\n",
      "52837.699241866634 6.4952554573495524e-18\n",
      "\n",
      " 5565.967606892147 6568.133030216476 12134.100637108622\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "47566.791794763885 1.34010055307262e-16\n",
      "\n",
      " 5537.596183017746 6563.824792680427 12101.420975698173\n",
      "43414.50981581746 2.268335120992925e-15\n",
      "\n",
      " 5511.607173714771 6559.784128587603 12071.391302302374\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "40183.79746675597 3.0935261906821176e-14\n",
      "\n",
      " 5487.484297034742 6555.994405124433 12043.478702159175\n",
      "37758.26654045576 3.3486377245335026e-13\n",
      "\n",
      " 5464.996555620155 6552.43862248328 12017.435178103435\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "36016.61797505677 2.8670898045812933e-12\n",
      "\n",
      " 5443.724046085757 6549.100594377304 11992.824640463063\n",
      "34699.242078953495 1.9784050486555304e-11\n",
      "\n",
      " 5423.858913688975 6545.967736782364 11969.82665047134\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "Gradient Update Steps:  17%|█▋        | 17/100 [00:02<00:12,  6.52it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "33378.8319033408 1.1861535075282735e-10\n",
      "\n",
      " 5405.470238798329 6543.033089592542 11948.50332839087\n",
      "31786.30471864617 6.713320528829526e-10\n",
      "\n",
      " 5388.415293062188 6540.289813657466 11928.705106719653\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[A\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "30022.79317101799 3.621206684054446e-09\n",
      "\n",
      " 5372.4918224306975 6537.7248512665665 11910.216673697265\n",
      "28277.264796428106 1.8015773117285113e-08\n",
      "\n",
      " 5357.480316933373 6535.322850266047 11892.80316719942\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "26640.582838609665 8.107845189936822e-08\n",
      "\n",
      " 5343.25167564241 6533.070570884802 11876.322246527212\n",
      "25133.268985983847 3.29233027919021e-07\n",
      "\n",
      " 5329.739202711868 6530.956910045592 11860.69611275746\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "23751.44871022119 1.210967959615378e-06\n",
      "\n",
      " 5316.863721076983 6528.9720806825035 11845.835801759487\n",
      "22485.22676712084 4.060569846415276e-06\n",
      "\n",
      " 5304.530673506817 6527.107165662166 11831.637839168983\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "21323.516713046953 1.249188438114238e-05\n",
      "\n",
      " 5292.6891147170245 6525.353954235448 11818.043068952473\n",
      "20255.70573268008 3.5485227304965726e-05\n",
      "\n",
      " 5281.350368985925 6523.704836101042 11805.055205086966\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19271.979363973696 9.043894784768675e-05\n",
      "\n",
      " 5270.4794023314025 6522.152792833298 11792.6321951647\n",
      "18363.36375732875 0.0002028727282030648\n",
      "\n",
      " 5260.098838757738 6520.691291667521 11780.79013042526\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "17521.85225022439 0.000433337096719966\n",
      "\n",
      " 5250.032419949712 6519.3143217977995 11769.346741747511\n",
      "16740.327021408157 0.0007001441792001263\n",
      "\n",
      " 5240.383668249213 6518.01628410139 11758.399952350603\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16012.554721985041 0.0007620746249855751\n",
      "\n",
      " 5231.034390234176 6516.792002132001 11747.826392366176\n",
      "15332.982175896386 0.0008288753833066265\n",
      "\n",
      " 5221.965772319985 6515.636687795504 11737.60246011549\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14696.706678725572 0.0009006930223165983\n",
      "\n",
      " 5213.247927332958 6514.5458922937605 11727.793819626719\n",
      "14099.467394880698 0.0009778074873979606\n",
      "\n",
      " 5204.797567100371 6513.515527089154 11718.313094189525\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "13537.426561329657 0.001060352648371874\n",
      "\n",
      " 5196.657236962834 6512.541797593261 11709.199034556095\n",
      "13007.270784202705 0.0011485341615999742\n",
      "\n",
      " 5188.791092961332 6511.621221519273 11700.412314480604\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "12506.010920585897 0.0012424053781235618\n",
      "\n",
      " 5181.090218448304 6510.750566633244 11691.840785081547\n",
      "12031.076597476113 0.0013420907018076669\n",
      "\n",
      " 5173.660777367604 6509.926826613647 11683.58760398125\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "11580.220845635587 0.001447612297342909\n",
      "\n",
      " 5166.446968100887 6509.1472015368345 11675.594169637721\n",
      "11151.533287037948 0.0015590637429904652\n",
      "\n",
      " 5159.437854846321 6508.408975622967 11667.846830469287\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10743.408045771921 0.0016765890086471362\n",
      "\n",
      " 5152.627571150112 6507.7095475550195 11660.33711870513\n",
      "10354.5203600849 0.0018002581457729034\n",
      "\n",
      " 5145.980447639728 6507.046356403722 11653.02680404345\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "Gradient Update Steps:  43%|████▎     | 43/100 [00:06<00:08,  6.46it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9983.933445981984 0.0019303621261634425\n",
      "\n",
      " 5139.531559692296 6506.416871002805 11645.948430695102\n",
      "9630.86765388076 0.002067257677834194\n",
      "\n",
      " 5133.233140347234 6505.818687794155 11639.05182814139\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9294.678517757653 0.002211442754811603\n",
      "\n",
      " 5127.097596859483 6505.249563206154 11632.347160065638\n",
      "8974.75280606476 0.0023634127429300526\n",
      "\n",
      " 5121.144293207233 6504.707492378606 11625.851785585839\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "8670.409416260458 0.002523617440921255\n",
      "\n",
      " 5115.313705863768 6504.190712233544 11619.504418097313\n",
      "8380.885522702456 0.002692438671822201\n",
      "\n",
      " 5109.676783111092 6503.697711892773 11613.374495003865\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "8105.343377586987 0.002870233214298961\n",
      "\n",
      " 5104.14842898608 6503.227151540008 11607.375580526088\n",
      "7842.87194084045 0.0030571121529065714\n",
      "\n",
      " 5098.7478715636225 6502.777867302497 11601.52573886612\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7592.590916129778 0.003253255256972803\n",
      "\n",
      " 5093.499290153424 6502.348774975941 11595.848065129365\n",
      "7353.632657870527 0.0034586654607911466\n",
      "\n",
      " 5088.312508045963 6501.938875434594 11590.251383480558\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7125.19623563963 0.0036464542860774906\n",
      "\n",
      " 5083.313029474122 6501.547232582985 11584.860262057107\n",
      "6906.535761785631 0.0037370043401461694\n",
      "\n",
      " 5078.355387722987 6501.172925851217 11579.528313574203\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6696.993470927675 0.0038336699092468385\n",
      "\n",
      " 5073.543841389304 6500.815104455172 11574.358945844477\n",
      "6495.960699482267 0.003936445973515\n",
      "\n",
      " 5068.869738990563 6500.4729273729 11569.342666363464\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6302.909987028708 0.0040453202184339305\n",
      "\n",
      " 5064.232442022183 6500.145601335134 11564.378043357317\n",
      "6117.347638295381 0.004160301781741692\n",
      "\n",
      " 5059.767617555484 6499.832378641448 11559.599996196932\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5938.851877202903 0.004281519858639698\n",
      "\n",
      " 5055.373592078096 6499.532524680875 11554.906116758972\n",
      "5767.019995387126 0.004408972860087989\n",
      "\n",
      " 5051.034491269618 6499.245367707953 11550.27985897757\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5601.49821710516 0.00454265182940356\n",
      "\n",
      " 5046.828718176167 6498.970259651728 11545.798977827895\n",
      "5441.965000283394 0.0046828131011725825\n",
      "\n",
      " 5042.710039763195 6498.706569617895 11541.416609381089\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5288.118742474354 0.004829487785749489\n",
      "\n",
      " 5038.65150031986 6498.453739270924 11537.105239590785\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5139.691547769468 0.004982751280344185\n",
      "\n",
      " 5034.648030511377 6498.211197819789 11532.859228331166\n",
      "4996.443906076295 0.005142718497604533\n",
      "\n",
      " 5030.756034816115 6497.978429083941 11528.734463900055\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4858.134683543713 0.005309543068568193\n",
      "\n",
      " 5026.954175039139 6497.754950075776 11524.709125114914\n",
      "4724.541587548433 0.005483347614996295\n",
      "\n",
      " 5023.17655921583 6497.540289977638 11520.716849193468\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4595.461870682921 0.005664206381551552\n",
      "\n",
      " 5019.499758544831 6497.334009261478 11516.833767806309\n",
      "4470.725549106681 0.0058523317968720776\n",
      "\n",
      " 5015.905383888721 6497.13569304716 11513.041076935882\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4350.137001676413 0.006047802984490563\n",
      "\n",
      " 5012.324792346499 6496.944949237451 11509.26974158395\n",
      "4233.5298285397585 0.006250776050115515\n",
      "\n",
      " 5008.850288780779 6496.7614132201625 11505.61170200094\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4120.763330472021 0.006461379117905833\n",
      "\n",
      " 5005.437067390316 6496.584724832435 11502.021792222751\n",
      "4011.669929756142 0.006679733188156741\n",
      "\n",
      " 5002.065300010198 6496.414564929674 11498.479864939873\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3906.1200620344052 0.006905916647265435\n",
      "\n",
      " 4998.765593738261 6496.250617770631 11495.016211508892\n",
      "3803.975469517789 0.0071401250549351985\n",
      "\n",
      " 4995.515302498987 6496.092584413821 11491.607886912807\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3705.116398172737 0.007382587380052765\n",
      "\n",
      " 4992.336945653312 6495.940184132296 11488.277129785609\n",
      "3609.4072078859804 0.0076333992133715\n",
      "\n",
      " 4989.172417141226 6495.793166791191 11484.965583932417\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3516.7387186245537 0.0078925763179112\n",
      "\n",
      " 4986.119164903856 6495.6512645057055 11481.77042940956\n",
      "3427.002344252885 0.008160355372945304\n",
      "\n",
      " 4983.113026007183 6495.51426179821 11478.627287805393\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3340.0789309429374 0.008436890007847248\n",
      "\n",
      " 4980.135301354196 6495.381917780854 11475.51721913505\n",
      "3255.8900015643862 0.008722248353801369\n",
      "\n",
      " 4977.186963747291 6495.254034110192 11472.440997857484\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3195.827254581115 0.009016611280908152\n",
      "\n",
      " 4974.293607703012 6495.130403062196 11469.424010765208\n",
      "3140.485529692283 0.00932005131532537\n",
      "\n",
      " 4971.482384472234 6495.010837933685 11466.49322240592\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3086.495872682638 0.00963273048469071\n",
      "\n",
      " 4968.691299954407 6494.895156220707 11463.586456175115\n",
      "3033.818081423486 0.009954887232482954\n",
      "\n",
      " 4965.953281772579 6494.783187059758 11460.736468832338\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2982.4085861825642 0.010286367945368685\n",
      "\n",
      " 4963.236136844644 6494.674774178513 11457.910911023157\n",
      "2932.2258073053063 0.01062750235808082\n",
      "\n",
      " 4960.5802229608025 6494.569758091175 11455.149981051978\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2883.2249807767203 0.010978423976566336\n",
      "\n",
      " 4957.954862056795 6494.468005103909 11452.422867160703\n",
      "2835.378017862166 0.011339003053281603\n",
      "\n",
      " 4955.392105610148 6494.369368433046 11449.761474043193\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2788.6429682316934 0.011709719533785665\n",
      "\n",
      " 4952.814481810056 6494.273717414711 11447.088199224767\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2742.987940594861 0.012090409654667704\n",
      "\n",
      " 4950.320093583088 6494.1809315739365 11444.501025157024\n",
      "2698.3764675457023 0.01248124485774489\n",
      "\n",
      " 4947.854054146886 6494.090900487464 11441.94495463435\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2654.774280372998 0.01288231098301554\n",
      "\n",
      " 4945.42240263909 6494.003493095774 11439.425895734865\n",
      "2612.1618422589527 0.013293742180648676\n",
      "\n",
      " 4943.034097711684 6493.91861906303 11436.952716774715\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2570.4973181997543 0.013715652940875601\n",
      "\n",
      " 4940.68797686534 6493.836165472598 11434.524142337938\n",
      "2529.7570684908023 0.014147970169747026\n",
      "\n",
      " 4938.361133671378 6493.756062160844 11432.117195832223\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2489.9161735598923 0.01459098555595356\n",
      "\n",
      " 4936.056681906592 6493.67818625174 11429.734868158332\n",
      "2450.9462359042654 0.015044578073969151\n",
      "\n",
      " 4933.764692589297 6493.602462819812 11427.367155409109\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2412.8207401521304 0.015508807504622172\n",
      "\n",
      " 4931.533640502366 6493.528813682264 11425.06245418463\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Gradient Update Steps: 100%|██████████| 100/100 [00:16<00:00,  6.21it/s]\n",
      "Restarts: 100%|██████████| 1/1 [00:16<00:00, 16.12s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2375.5184580658897 0.015983906278393564\n",
      "\n",
      " 4929.333352113388 6493.457157780264 11422.790509893654\n",
      "2339.010937037581 0.0164699709462225\n",
      "\n",
      " 4927.186446326536 6493.387412706562 11420.573859033098\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "from models.scs.model import SparseCoxSubgroup\n",
    "\n",
    "torch.manual_seed(random_seed)\n",
    "np.random.seed(random_seed)\n",
    "\n",
    "# Instantiate the CMHE model on all data\n",
    "model = SparseCoxSubgroup(k=k, random_seed=random_seed, n_epochs=n_epochs, n_restarts=1, step_size=step_size, alpha=0.)\n",
    "\n",
    "model.fit(features_prep.values, outcomes.time.values, outcomes.event.values, (intervention=='SBP target <120mm Hg').values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [00:15<00:00,  6.58it/s]\n"
     ]
    }
   ],
   "source": [
    "# from models.cmhe import DeepCoxMixturesHeterogenousEffects\n",
    "\n",
    "# torch.manual_seed(random_seed)\n",
    "# np.random.seed(random_seed)\n",
    "\n",
    "# # Instantiate the CMHE model on all data\n",
    "# model = DeepCoxMixturesHeterogenousEffects(random_seed=10, k=k, g=g, layers=layers, gate_l2_penalty=gate_l2_penalty)\n",
    "\n",
    "# model = model.fit(features_prep, outcomes.time, outcomes.event, intervention=='Niacin', val_data=None, iters=iters, \n",
    "#                   learning_rate=learning_rate, batch_size=batch_size, \n",
    "#                   optimizer=optimizer, patience=patience)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [],
   "source": [
    "# zeta_probs = model.predict_latent_phi(features_prep)\n",
    "# zeta = np.argmax(zeta_probs, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "zeta = model.predict_proba(features_prep.values).argmax(axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5022\n",
      "4800\n"
     ]
    }
   ],
   "source": [
    "print((zeta == 0).sum())\n",
    "print((zeta == 1).sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from auton_survival import reporting\n",
    "fig, ((ax1, ax2)) = plt.subplots(1, 2, figsize=(10, 5))\n",
    "fig.suptitle('Patient with max probability of survival scores')\n",
    "reporting.plot_kaplanmeier(outcomes[zeta == 0], intervention[zeta == 0], ax=ax1)\n",
    "reporting.plot_kaplanmeier(outcomes[zeta == 1], intervention[zeta == 1], ax=ax2)\n",
    "for ax in fig.get_axes():\n",
    "    ax.label_outer()\n",
    "    h, l = ax.get_legend_handles_labels()\n",
    "    ax.set_xlabel('Time (Days)')\n",
    "    ax.set_ylabel('Event-Free Survival Probability')\n",
    "    ax.legend([h[0], h[1]], ['SBP target <120mm Hg', 'SBP target <140mm Hg'], loc=\"upper right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of training data points: 3694\n",
      "Number of test data points: 1584\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Split the PEACE data into training and test data\n",
    "x_train, x_test, y_train, y_test = train_test_split(features_prep, outcomes, test_size=0.3, random_state=1, \n",
    "                                          stratify = outcomes.event + 2*(intervention=='Lipid Fibrate'))\n",
    "\n",
    "print(f'Number of training data points: {len(x_train)}')\n",
    "print(f'Number of test data points: {len(x_test)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_tr = x_train\n",
    "t_tr = y_train['time']\n",
    "e_tr = y_train['event']\n",
    "a_tr = intervention[y_train.index]=='Lipid Fibrate'\n",
    "\n",
    "x_te = x_test\n",
    "t_te = y_test['time']\n",
    "e_te = y_test['event']\n",
    "a_te = intervention[y_test.index]=='Lipid Fibrate'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Restarts:   0%|          | 0/1 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 6766.56256514811 2645.770743080898 9412.333308229008\n",
      "293961126.0192749 1.0590847313572378e-158\n",
      "\n",
      " 6679.28317479637 2584.037954131119 9263.32112892749\n",
      "176638729.71025002 3.13755411157731e-71\n",
      "\n",
      " 6512.998159111286 2583.713157106172 9096.711316217457\n",
      "116140278.63893168 2.1786978046342056e-70\n",
      "\n",
      " 6343.814025748962 2583.303449422854 8927.117475171817\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "76089478.37289758 1.9864218219873842e-69\n",
      "\n",
      " 6199.532814165825 2582.904550000942 8782.437364166766\n",
      "49648853.514992565 2.099964524606141e-68\n",
      "\n",
      " 6043.989943957702 2582.5169478514176 8626.50689180912\n",
      "32289496.016926922 2.3674873293366508e-67\n",
      "\n",
      " 5903.978061174568 2582.1412016106106 8486.119262785178\n",
      "20924511.243392527 4.885707887639669e-66\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 5756.3190975264615 2581.7777419489107 8338.096839475373\n",
      "13511169.484512204 1.1054033932014936e-64\n",
      "\n",
      " 5604.89692623701 2581.4279143537933 8186.324840590803\n",
      "8711041.977223223 2.9434468983917206e-63\n",
      "\n",
      " 5464.77307880882 2581.092792584035 8045.865871392855\n",
      "5618593.1462175045 1.0665541202777072e-61\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 5317.190591304094 2580.7734336323692 7897.964024936463\n",
      "3633402.897011301 4.439488986715005e-60\n",
      "\n",
      " 5179.669763435003 2580.4710176153512 7760.140781050354\n",
      "2364287.3971564025 2.9149457698056005e-58\n",
      "\n",
      " 5054.254124740593 2580.185397625365 7634.4395223659585\n",
      "1552012.6769745555 3.0154729381789535e-56\n",
      "\n",
      " 4924.118832630205 2579.9156855872398 7504.034518217445\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1030496.6824920584 3.62578487225238e-54\n",
      "\n",
      " 4806.933313012141 2579.6596762372674 7386.592989249409\n",
      "695659.895691913 6.443643737274991e-52\n",
      "\n",
      " 4692.507114242548 2579.4130076955435 7271.920121938091\n",
      "478780.6419968276 1.5546259574850953e-49\n",
      "\n",
      " 4577.295493190066 2579.171889687307 7156.467382877373\n",
      "337357.1866715467 5.929297337119198e-47\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 4482.546290788159 2578.934095094125 7061.480385882284\n",
      "244001.1270467896 3.8850129302173526e-44\n",
      "\n",
      " 4389.696665906878 2578.699846167017 6968.396512073894\n",
      "181157.22378373853 3.496892448441344e-41\n",
      "\n",
      " 4301.756681527166 2578.469860997494 6880.22654252466\n",
      "138067.78772337496 3.4876443509591387e-38\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 4223.9364878255965 2578.2426414799393 6802.179129305536\n",
      "107881.46201111574 4.6591164528514325e-35\n",
      "\n",
      " 4155.656191719681 2578.0130537292657 6733.669245448948\n",
      "85780.56067893458 1.340942886549834e-32\n",
      "\n",
      " 4096.2607654905 2577.7740790963426 6674.034844586842\n",
      "69116.54896388343 1.4093184009999366e-30\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 4044.2447318369686 2577.5211938951484 6621.7659257321175\n",
      "56824.829555734425 1.4199224114683244e-28\n",
      "\n",
      " 3991.449675278499 2577.2646543274477 6568.714329605947\n",
      "47886.510817553215 1.6743397398712517e-26\n",
      "\n",
      " 3944.960403599755 2577.015980512098 6521.976384111853\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41133.07259668134 1.8978546921194415e-24\n",
      "\n",
      " 3912.343595919588 2576.7725083386913 6489.116104258279\n",
      "35794.24293120293 1.6214579800424955e-22\n",
      "\n",
      " 3873.5917921118958 2576.5311688033325 6450.122960915229\n",
      "31460.556985305633 9.762538610825634e-21\n",
      "\n",
      " 3832.088975285457 2576.2913275595956 6408.380302845053\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "27881.609094092368 3.916607962928151e-19\n",
      "\n",
      " 3811.9544900487595 2576.053332495371 6388.00782254413\n",
      "24886.297280120834 1.1843542221837156e-17\n",
      "\n",
      " 3781.7361491701386 2575.8170361137804 6357.553185283919\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "22332.510826317946 2.5336341906468526e-16\n",
      "\n",
      " 3752.7560327285714 2575.583459617059 6328.339492345631\n",
      "20143.795964425713 4.007970903012233e-15\n",
      "\n",
      " 3734.327409017854 2575.352631576054 6309.680040593908\n",
      "18254.985096271586 4.8653124853480584e-14\n",
      "\n",
      " 3703.732476800763 2575.1246548108734 6278.857131611636\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16613.338290834658 4.5383354721980413e-13\n",
      "\n",
      " 3690.5056000866493 2574.9000203590554 6265.405620445705\n",
      "15170.27301595012 7.100017239213959e-13\n",
      "\n",
      " 3670.6906459648435 2574.6785540298724 6245.369199994716\n",
      "13897.769772847403 4.067287491116637e-13\n",
      "\n",
      " 3650.4208420896357 2574.4606720801485 6224.881514169784\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "12767.744393763745 2.529066718812403e-13\n",
      "\n",
      " 3629.793947592886 2574.246250034335 6204.040197627221\n",
      "11764.114809775316 1.6848645042140623e-13\n",
      "\n",
      " 3612.655936994696 2574.0356170414307 6186.691554036126\n",
      "10867.950779820529 1.2052289259383786e-13\n",
      "\n",
      " 3594.502078789633 2573.828364211824 6168.330443001457\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10066.411176051366 9.088120908708271e-14\n",
      "\n",
      " 3581.8013970196093 2573.624879248474 6155.426276268083\n",
      "9344.441373616153 7.29827632047724e-14\n",
      "\n",
      " 3563.4468773049525 2573.424869279179 6136.871746584131\n",
      "8697.13227621705 6.13246598171003e-14\n",
      "\n",
      " 3547.250024362529 2573.228635285442 6120.478659647972\n",
      "8108.212362850037 5.393547241764802e-14\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 3523.084857947537 2573.035823055661 6096.120681003198\n",
      "7574.73418700655 4.855213503146499e-14\n",
      "\n",
      " 3506.3519810603484 2572.8467404396984 6079.198721500046\n",
      "7091.191986027727 4.6024386224509304e-14\n",
      "\n",
      " 3501.213248940451 2572.6609962377906 6073.874245178242\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6649.3365548366355 4.5224750964542215e-14\n",
      "\n",
      " 3484.7906968503994 2572.478708545091 6057.269405395491\n",
      "6247.027836726399 4.5779947082088845e-14\n",
      "\n",
      " 3469.6381179281066 2572.299909738907 6041.938027667014\n",
      "5879.797265366545 4.725912083579512e-14\n",
      "\n",
      " 3462.015934595687 2572.1245508122684 6034.140485407956\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5540.326195509723 5.0197383277152106e-14\n",
      "\n",
      " 3444.5218095617756 2571.9523061999953 6016.474115761771\n",
      "5229.373646124469 5.3836823619542374e-14\n",
      "\n",
      " 3430.7126631177052 2571.7835013596105 6002.496164477316\n",
      "4939.996464679888 5.938576420754074e-14\n",
      "\n",
      " 3421.6564443519164 2571.6177387209864 5993.274183072903\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4675.029096538934 6.657636856176658e-14\n",
      "\n",
      " 3404.2251309004896 2571.455343467772 5975.6804743682615\n",
      "4429.3455875573945 7.538111728445097e-14\n",
      "\n",
      " 3396.156099213661 2571.2959644408966 5967.452063654558\n",
      "4201.125165542088 8.729082973384233e-14\n",
      "\n",
      " 3381.5958245834136 2571.139427563634 5952.7352521470475\n",
      "3989.69073295395 1.019902600632457e-13\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 3363.768279202156 2570.9859669716457 5934.754246173801\n",
      "3794.2603320764606 1.1923688649850554e-13\n",
      "\n",
      " 3356.713867467271 2570.835438975825 5927.549306443096\n",
      "3611.453019793287 1.4272124817568124e-13\n",
      "\n",
      " 3347.4107959908883 2570.6873530768057 5918.098149067694\n",
      "3441.2220678440985 1.7110961627784717e-13\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 3333.451194641717 2570.5421568208676 5903.9933514625845\n",
      "3284.2057028170198 2.0758165691070022e-13\n",
      "\n",
      " 3322.387938777875 2570.3996468351997 5892.787585613075\n",
      "3135.424205987544 2.5424304779471683e-13\n",
      "\n",
      " 3312.749006195336 2570.259467486974 5883.00847368231\n",
      "2997.575049856391 3.152932762674512e-13\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 3300.1298639827232 2570.121912991781 5870.2517769745045\n",
      "2868.17411263819 3.9176568590501715e-13\n",
      "\n",
      " 3287.119279139793 2569.986797914863 5857.1060770546555\n",
      "2745.8307412167137 4.922864216176046e-13\n",
      "\n",
      " 3281.725624199721 2569.853835429699 5851.57945962942\n",
      "2631.8396990215897 6.188741836768943e-13\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 3266.857062988138 2569.723303113389 5836.580366101527\n",
      "2522.1640984109113 7.771001529124283e-13\n",
      "\n",
      " 3260.511433481714 2569.5947515442895 5830.106185026003\n",
      "2420.5847239608497 9.875232015137923e-13\n",
      "\n",
      " 3249.7374417967935 2569.468558906955 5819.206000703749\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2353.981030558902 1.2631929708851308e-12\n",
      "\n",
      " 3238.492805117843 2569.3443263812924 5807.8371314991355\n",
      "2297.086228204407 1.6143573157932767e-12\n",
      "\n",
      " 3227.6710550399857 2569.2223298229796 5796.893384862966\n",
      "2241.908482275624 2.0887065316688887e-12\n",
      "\n",
      " 3223.079345303483 2569.1020405327263 5792.181385836209\n",
      "2188.469794636802 2.7105010374426685e-12\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 3215.8423838245412 2568.983817224535 5784.826201049076\n",
      "2136.686372702226 3.5061051731511717e-12\n",
      "\n",
      " 3207.8768678840656 2568.867487486642 5776.744355370707\n",
      "2086.508445154604 4.550208982739766e-12\n",
      "\n",
      " 3193.3236590719566 2568.753024191635 5762.076683263592\n",
      "2037.9911243068339 5.9089974594659715e-12\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 3187.8607045184217 2568.6404553385246 5756.501159856946\n",
      "1990.923996551217 7.733673212299115e-12\n",
      "\n",
      " 3177.1236783909253 2568.529636580415 5745.6533149713405\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1945.3821779741386 1.0095458940201325e-11\n",
      "\n",
      " 3169.2044485009365 2568.4206234910926 5737.625071992029\n",
      "1901.2683862662434 1.3257898110570056e-11\n",
      "\n",
      " 3164.218040368703 2568.313291727031 5732.531332095734\n",
      "1858.4535706858126 1.7366033123182145e-11\n",
      "\n",
      " 3143.5538887635303 2568.207813443287 5711.761702206817\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1817.192430907018 2.2765753715137632e-11\n",
      "\n",
      " 3145.7556891431036 2568.10400338783 5713.859692530934\n",
      "1777.0030691455759 2.995028019154241e-11\n",
      "\n",
      " 3137.051490247978 2568.0017842770303 5705.053274525008\n",
      "1738.1387882540112 3.945866332556135e-11\n",
      "\n",
      " 3127.588605332935 2567.9013409794443 5695.489946312379\n",
      "1700.5057819157164 5.1842979219566757e-11\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 3120.634690295843 2567.8025049724038 5688.437195268247\n",
      "1663.7083570358561 6.808041109740185e-11\n",
      "\n",
      " 3109.0726967660903 2567.705268061285 5676.777964827375\n",
      "1628.2221265698731 8.918759509190202e-11\n",
      "\n",
      " 3101.286627269514 2567.6096451921035 5668.896272461618\n",
      "1593.8591285892935 1.170782757621685e-10\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 3095.626185909925 2567.51553974485 5663.141725654775\n",
      "1560.5104717216186 1.5432366198807032e-10\n",
      "\n",
      " 3088.841935391739 2567.4229617501946 5656.264897141933\n",
      "1528.0396459823448 2.0175801376862935e-10\n",
      "\n",
      " 3079.4056015719116 2567.3320335074277 5646.737635079339\n",
      "1496.5238074634283 2.6343666778907064e-10\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 3068.717497175561 2567.2425210355154 5635.960018211077\n",
      "1465.9483517753613 3.434006787745634e-10\n",
      "\n",
      " 3067.6212471096514 2567.154519118055 5634.775766227706\n",
      "1436.3701705879932 4.482129667194338e-10\n",
      "\n",
      " 3058.434578043375 2567.0679338655077 5625.502511908882\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1407.4737582044763 5.824587410556298e-10\n",
      "\n",
      " 3052.0303824456228 2566.982791539119 5619.013173984742\n",
      "1379.5303710347491 7.570824123425861e-10\n",
      "\n",
      " 3049.796813542373 2566.899099354055 5616.695912896428\n",
      "1352.4722629392395 9.812261991318075e-10\n",
      "\n",
      " 3037.288170788217 2566.8167999536045 5604.1049707418215\n",
      "1326.1647847036065 1.2647352216132186e-09\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 3033.1882341849587 2566.7358774429 5599.924111627859\n",
      "1300.5362978367145 1.6318505591326285e-09\n",
      "\n",
      " 3020.8969784224237 2566.656253556947 5587.553231979371\n",
      "1275.713708340708 2.090120874172116e-09\n",
      "\n",
      " 3019.484905054727 2566.5780595542997 5586.0629646090265\n",
      "1251.5545161505459 2.6795448114805513e-09\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": []
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 3010.91102707344 2566.5011065539366 5577.4121336273765\n",
      "1228.1150038793276 3.413356552545915e-09\n",
      "\n",
      " 3005.2901354947358 2566.425490421963 5571.715625916699\n",
      "1205.3570296444884 4.3632509598192515e-09\n",
      "\n",
      " 2997.120614197931 2566.351060404224 5563.471674602155\n",
      "1183.1925893120524 5.527877546020783e-09\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Gradient Update Steps: 100%|██████████| 100/100 [00:07<00:00, 14.25it/s]\n",
      "Restarts: 100%|██████████| 1/1 [00:07<00:00,  7.03s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 2995.905070830392 2566.2779647344555 5562.183035564847\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "from models.scs.model import SparseCoxSubgroup\n",
    "\n",
    "torch.manual_seed(random_seed)\n",
    "np.random.seed(random_seed)\n",
    "\n",
    "# Instantiate the CMHE model on all data\n",
    "model = SparseCoxSubgroup(k=k, random_seed=random_seed, n_epochs=n_epochs, n_restarts=1, step_size=step_size, alpha=0.)\n",
    "\n",
    "model.fit(x_tr.values, t_tr.values, e_tr.values, a_tr.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [00:08<00:00, 11.17it/s]\n"
     ]
    }
   ],
   "source": [
    "# from models.cmhe import DeepCoxMixturesHeterogenousEffects\n",
    "\n",
    "# torch.manual_seed(random_seed)\n",
    "# np.random.seed(random_seed)\n",
    "\n",
    "# # Instantiate the CMHE model\n",
    "# model = DeepCoxMixturesHeterogenousEffects(random_seed=10, k=k, g=g, layers=layers, gate_l2_penalty=gate_l2_penalty)\n",
    "\n",
    "# model = model.fit(x_tr, t_tr, e_tr, a_tr, iters=iters, \n",
    "#                   learning_rate=learning_rate, batch_size=batch_size, \n",
    "#                   optimizer=optimizer, patience=patience)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "zeta_probs_train = model.predict_proba(x_tr.values)\n",
    "zeta_train =  np.argmax(zeta_probs_train, axis=1)\n",
    "\n",
    "zeta_probs_test = model.predict_proba(x_te.values)\n",
    "zeta_test =  np.argmax(zeta_probs_test, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [],
   "source": [
    "# zeta_probs_train = model.predict_latent_phi(x_tr)\n",
    "# zeta_train =  np.argmax(zeta_probs_train, axis=1)\n",
    "\n",
    "# zeta_probs_test = model.predict_latent_phi(x_te)\n",
    "# zeta_test =  np.argmax(zeta_probs_test, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2, 2, figsize= (10, 10))\n",
    "fig.suptitle('Patient with max probability of survival scores')\n",
    "reporting.plot_kaplanmeier(y_train[zeta_train == 0], a_tr[zeta_train == 0], ax=ax1)\n",
    "reporting.plot_kaplanmeier(y_train[zeta_train == 1], a_tr[zeta_train == 1], ax=ax2)\n",
    "reporting.plot_kaplanmeier(y_test[zeta_test == 0], a_te[zeta_test == 0], ax=ax3) \n",
    "reporting.plot_kaplanmeier(y_test[zeta_test == 1], a_te[zeta_test == 1], ax=ax4)\n",
    "for ax in fig.get_axes():\n",
    "    ax.label_outer()\n",
    "    h, l = ax.get_legend_handles_labels()\n",
    "    ax.set_xlabel('Time (Days)')\n",
    "    ax.set_ylabel('Event-Free Survival Probability')\n",
    "    ax.legend([h[0], h[1]], ['Lipid Fibrate', 'Lipid Placebo'], loc=\"upper right\")\n",
    "plt.show()\n",
    "# First row is training data\n",
    "# Second row is test data\n",
    "# First column is zeta 0\n",
    "# Second column is zeta 1"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.12 ('Pandas')",
   "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": "3.9.12"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "f08cc569b594e1720561ae63e75656c07c548e4f90e451cb99794b898c100aec"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}