TabNetMultiLabelClassifier.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "ea5520f8-6ee6-4c3a-8d5e-56f2f623bf26",
   "metadata": {},
   "source": [
    "## DL Model on Tabular Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bd8efa9e-b51a-4f91-be4d-1f3732ca16e8",
   "metadata": {},
   "source": [
    "- \"in most cases the DL models performed worse on unseen datasets\"\n",
    "- \"XGBoost/LightGBM models generally outperformed the deep models\"\n",
    "- \"no DL model consistently outperformed the others\"\n",
    "- \"ensemble of deep models and XGBoost outperforms the other models in most cases\"\n",
    "\n",
    "And here goes..."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "17f48504-c06e-40fd-8643-2184ef070f16",
   "metadata": {
    "tags": []
   },
   "source": [
    "## TabNet\n",
    "- Paper: https://arxiv.org/abs/1908.07442\n",
    "- Torch Framework: https://github.com/dreamquark-ai/tabnet\n",
    "- Google Cloud: https://cloud.google.com/ai-platform/training/docs/algorithms/tab-net-start"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "727b183f-626c-4655-a9c2-f1fd686c85f6",
   "metadata": {
    "tags": []
   },
   "source": [
    "### TabNet Architecture"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "623ec378-3247-4eb4-b286-830cddf452f2",
   "metadata": {
    "tags": []
   },
   "source": [
    "<img src=\"images/tabnet_arch.png\" width=\"1000\">"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dd056a01-d6d2-4441-9ea2-b72fdda2be7b",
   "metadata": {},
   "source": [
    "#### Entmax for different alphas (Sparsemax: alpha=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6ae6eec5-ccda-402c-8961-c1d7b8414c8f",
   "metadata": {},
   "source": [
    "<img src=\"images/entmax.png\" width=\"400\">"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "192b38a0-f77d-445c-9ea9-cdbbad4417ab",
   "metadata": {
    "tags": []
   },
   "source": [
    "### TabNet Advantages"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ddb63c5d-2c78-4925-b59a-8bc1968bd78e",
   "metadata": {
    "tags": []
   },
   "source": [
    "- MultiTaskClassifier, MultiRegressor\n",
    "- interpretability using Masks\n",
    "- use all the deep-learning concepts like LR scheduler, Custom loss, etc\n",
    "- no feature selection\n",
    "- it achieves high scores without feature engineering and finetuning\n",
    "- self-supervised learning (pre-training)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4d21919-8864-4771-a77d-7ae938bedcba",
   "metadata": {
    "tags": []
   },
   "source": [
    "#### Installing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "48a50016-c77d-4f06-b255-6c4fa5e638f0",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "#!pip install pytorch-tabnet"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dafc40dd-718f-4453-98e9-b757dd017c20",
   "metadata": {
    "tags": []
   },
   "source": [
    "#### Import packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "0bc528da-ac71-4b3e-806f-82db8c1b2a8f",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import random\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd \n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.multioutput import MultiOutputClassifier\n",
    "import torch\n",
    "from pytorch_tabnet.multitask import TabNetMultiTaskClassifier\n",
    "from pytorch_tabnet.pretraining import TabNetPretrainer\n",
    "from xgboost import XGBClassifier\n",
    "\n",
    "from utils import (\n",
    "    deterministic, encode_categorical, fit_and_predict, fit_and_predict_xgb, \n",
    "    extract_fold, feature_importances, feature_importances_xgb, get_diff_distribution_features\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "690d72c3-0105-4829-9779-663af00cba00",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "215ccc1a-82a9-4530-a12d-8b3743db073d",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "#configs\n",
    "NUM_FOLDS = 5\n",
    "seed = 2021"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "58c3ddeb-2351-422f-8c32-79c5b1880a45",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "deterministic(seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "44785b4c-d80e-4021-baaf-e880a7cd8542",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "name                          size  creationDate         \n",
      "---------------------------  -----  -------------------  \n",
      "test_features.csv             25MB  2020-11-03 20:33:37  \n",
      "train_targets_scored.csv      10MB  2020-11-03 20:33:37  \n",
      "sample_submission.csv          3MB  2020-11-03 20:33:37  \n",
      "train_features.csv           149MB  2020-11-03 20:33:37  \n",
      "train_targets_nonscored.csv   19MB  2020-11-03 20:33:37  \n",
      "train_drug.csv               535KB  2020-11-03 20:33:37  \n"
     ]
    }
   ],
   "source": [
    "!kaggle competitions files -c lish-moa"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "5c7b0fcb-746a-4c34-b5f5-90dd8ca81875",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading lish-moa.zip to data\n",
      "100%|██████████████████████████████████████| 64.7M/64.7M [00:49<00:00, 1.44MB/s]\n",
      "100%|██████████████████████████████████████| 64.7M/64.7M [00:49<00:00, 1.38MB/s]\n",
      "Archive:  lish-moa.zip\n",
      "  inflating: sample_submission.csv   \n",
      "  inflating: test_features.csv       \n",
      "  inflating: train_drug.csv          \n",
      "  inflating: train_features.csv      \n",
      "  inflating: train_targets_nonscored.csv  \n",
      "  inflating: train_targets_scored.csv  \n"
     ]
    }
   ],
   "source": [
    "!kaggle competitions download -c lish-moa -p data/ && cd data/ && unzip lish-moa.zip"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6cd9aed-c8ce-4b9f-bb19-6227fa457b20",
   "metadata": {
    "tags": []
   },
   "source": [
    "#### Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "1a643bfd-62c3-45d8-8ef8-bb7975171e2d",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X shape: (23814, 876) \n",
      "y shape: (23814, 207)\n"
     ]
    }
   ],
   "source": [
    "X = pd.read_csv(\"data/train_features.csv\")\n",
    "y = pd.read_csv('data/train_targets_scored.csv')\n",
    "print(f\"X shape: {X.shape} \\ny shape: {y.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "a044bd9f-6d4e-4885-9da0-67bc9afe1559",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>sig_id</th>\n",
       "      <th>cp_type</th>\n",
       "      <th>cp_time</th>\n",
       "      <th>cp_dose</th>\n",
       "      <th>g-0</th>\n",
       "      <th>g-1</th>\n",
       "      <th>g-2</th>\n",
       "      <th>g-3</th>\n",
       "      <th>g-4</th>\n",
       "      <th>g-5</th>\n",
       "      <th>...</th>\n",
       "      <th>c-90</th>\n",
       "      <th>c-91</th>\n",
       "      <th>c-92</th>\n",
       "      <th>c-93</th>\n",
       "      <th>c-94</th>\n",
       "      <th>c-95</th>\n",
       "      <th>c-96</th>\n",
       "      <th>c-97</th>\n",
       "      <th>c-98</th>\n",
       "      <th>c-99</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>id_000644bb2</td>\n",
       "      <td>trt_cp</td>\n",
       "      <td>24</td>\n",
       "      <td>D1</td>\n",
       "      <td>1.0620</td>\n",
       "      <td>0.5577</td>\n",
       "      <td>-0.2479</td>\n",
       "      <td>-0.6208</td>\n",
       "      <td>-0.1944</td>\n",
       "      <td>-1.0120</td>\n",
       "      <td>...</td>\n",
       "      <td>0.2862</td>\n",
       "      <td>0.2584</td>\n",
       "      <td>0.8076</td>\n",
       "      <td>0.5523</td>\n",
       "      <td>-0.1912</td>\n",
       "      <td>0.6584</td>\n",
       "      <td>-0.3981</td>\n",
       "      <td>0.2139</td>\n",
       "      <td>0.3801</td>\n",
       "      <td>0.4176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>id_000779bfc</td>\n",
       "      <td>trt_cp</td>\n",
       "      <td>72</td>\n",
       "      <td>D1</td>\n",
       "      <td>0.0743</td>\n",
       "      <td>0.4087</td>\n",
       "      <td>0.2991</td>\n",
       "      <td>0.0604</td>\n",
       "      <td>1.0190</td>\n",
       "      <td>0.5207</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.4265</td>\n",
       "      <td>0.7543</td>\n",
       "      <td>0.4708</td>\n",
       "      <td>0.0230</td>\n",
       "      <td>0.2957</td>\n",
       "      <td>0.4899</td>\n",
       "      <td>0.1522</td>\n",
       "      <td>0.1241</td>\n",
       "      <td>0.6077</td>\n",
       "      <td>0.7371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>id_000a6266a</td>\n",
       "      <td>trt_cp</td>\n",
       "      <td>48</td>\n",
       "      <td>D1</td>\n",
       "      <td>0.6280</td>\n",
       "      <td>0.5817</td>\n",
       "      <td>1.5540</td>\n",
       "      <td>-0.0764</td>\n",
       "      <td>-0.0323</td>\n",
       "      <td>1.2390</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.7250</td>\n",
       "      <td>-0.6297</td>\n",
       "      <td>0.6103</td>\n",
       "      <td>0.0223</td>\n",
       "      <td>-1.3240</td>\n",
       "      <td>-0.3174</td>\n",
       "      <td>-0.6417</td>\n",
       "      <td>-0.2187</td>\n",
       "      <td>-1.4080</td>\n",
       "      <td>0.6931</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 876 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         sig_id cp_type  cp_time cp_dose     g-0     g-1     g-2     g-3  \\\n",
       "0  id_000644bb2  trt_cp       24      D1  1.0620  0.5577 -0.2479 -0.6208   \n",
       "1  id_000779bfc  trt_cp       72      D1  0.0743  0.4087  0.2991  0.0604   \n",
       "2  id_000a6266a  trt_cp       48      D1  0.6280  0.5817  1.5540 -0.0764   \n",
       "\n",
       "      g-4     g-5  ...    c-90    c-91    c-92    c-93    c-94    c-95  \\\n",
       "0 -0.1944 -1.0120  ...  0.2862  0.2584  0.8076  0.5523 -0.1912  0.6584   \n",
       "1  1.0190  0.5207  ... -0.4265  0.7543  0.4708  0.0230  0.2957  0.4899   \n",
       "2 -0.0323  1.2390  ... -0.7250 -0.6297  0.6103  0.0223 -1.3240 -0.3174   \n",
       "\n",
       "     c-96    c-97    c-98    c-99  \n",
       "0 -0.3981  0.2139  0.3801  0.4176  \n",
       "1  0.1522  0.1241  0.6077  0.7371  \n",
       "2 -0.6417 -0.2187 -1.4080  0.6931  \n",
       "\n",
       "[3 rows x 876 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "bf9ea952-62a4-4ff6-b73a-c05e84c52f4d",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>sig_id</th>\n",
       "      <th>5-alpha_reductase_inhibitor</th>\n",
       "      <th>11-beta-hsd1_inhibitor</th>\n",
       "      <th>acat_inhibitor</th>\n",
       "      <th>acetylcholine_receptor_agonist</th>\n",
       "      <th>acetylcholine_receptor_antagonist</th>\n",
       "      <th>acetylcholinesterase_inhibitor</th>\n",
       "      <th>adenosine_receptor_agonist</th>\n",
       "      <th>adenosine_receptor_antagonist</th>\n",
       "      <th>adenylyl_cyclase_activator</th>\n",
       "      <th>...</th>\n",
       "      <th>tropomyosin_receptor_kinase_inhibitor</th>\n",
       "      <th>trpv_agonist</th>\n",
       "      <th>trpv_antagonist</th>\n",
       "      <th>tubulin_inhibitor</th>\n",
       "      <th>tyrosine_kinase_inhibitor</th>\n",
       "      <th>ubiquitin_specific_protease_inhibitor</th>\n",
       "      <th>vegfr_inhibitor</th>\n",
       "      <th>vitamin_b</th>\n",
       "      <th>vitamin_d_receptor_agonist</th>\n",
       "      <th>wnt_inhibitor</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>id_000644bb2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>id_000779bfc</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>id_000a6266a</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 207 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         sig_id  5-alpha_reductase_inhibitor  11-beta-hsd1_inhibitor  \\\n",
       "0  id_000644bb2                            0                       0   \n",
       "1  id_000779bfc                            0                       0   \n",
       "2  id_000a6266a                            0                       0   \n",
       "\n",
       "   acat_inhibitor  acetylcholine_receptor_agonist  \\\n",
       "0               0                               0   \n",
       "1               0                               0   \n",
       "2               0                               0   \n",
       "\n",
       "   acetylcholine_receptor_antagonist  acetylcholinesterase_inhibitor  \\\n",
       "0                                  0                               0   \n",
       "1                                  0                               0   \n",
       "2                                  0                               0   \n",
       "\n",
       "   adenosine_receptor_agonist  adenosine_receptor_antagonist  \\\n",
       "0                           0                              0   \n",
       "1                           0                              0   \n",
       "2                           0                              0   \n",
       "\n",
       "   adenylyl_cyclase_activator  ...  tropomyosin_receptor_kinase_inhibitor  \\\n",
       "0                           0  ...                                      0   \n",
       "1                           0  ...                                      0   \n",
       "2                           0  ...                                      0   \n",
       "\n",
       "   trpv_agonist  trpv_antagonist  tubulin_inhibitor  \\\n",
       "0             0                0                  0   \n",
       "1             0                0                  0   \n",
       "2             0                0                  0   \n",
       "\n",
       "   tyrosine_kinase_inhibitor  ubiquitin_specific_protease_inhibitor  \\\n",
       "0                          0                                      0   \n",
       "1                          0                                      0   \n",
       "2                          0                                      0   \n",
       "\n",
       "   vegfr_inhibitor  vitamin_b  vitamin_d_receptor_agonist  wnt_inhibitor  \n",
       "0                0          0                           0              0  \n",
       "1                0          0                           0              0  \n",
       "2                0          0                           0              0  \n",
       "\n",
       "[3 rows x 207 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "be00f59a-9436-4686-bca4-f691eb22669d",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Same ID in X and y: 23814'"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# is ID order the same\n",
    "f\"Same ID in X and y: {(X['sig_id'] == y['sig_id']).sum()}\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "61f33087-b378-4542-bb27-0a6bbc87aa6e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# drop id column\n",
    "X.drop(columns=[\"sig_id\"], inplace=True)\n",
    "y.drop(columns=[\"sig_id\"], inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ffb2ea6-6122-4d74-a92d-04fcccfc25d7",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Categorical variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "7316a92c-f8b6-48a9-a7ad-a1a1e50f3585",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Categorical features:\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "cp_type    2\n",
       "cp_time    3\n",
       "cp_dose    2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Categorical features:\")\n",
    "train_unique_ds = X.nunique()\n",
    "train_unique_ds[train_unique_ds < 200]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "6bccf848-a35d-4cff-a483-0430c9ca8eb5",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Categorical targets:\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "5-alpha_reductase_inhibitor              2\n",
       "11-beta-hsd1_inhibitor                   2\n",
       "acat_inhibitor                           2\n",
       "acetylcholine_receptor_agonist           2\n",
       "acetylcholine_receptor_antagonist        2\n",
       "                                        ..\n",
       "ubiquitin_specific_protease_inhibitor    2\n",
       "vegfr_inhibitor                          2\n",
       "vitamin_b                                2\n",
       "vitamin_d_receptor_agonist               2\n",
       "wnt_inhibitor                            2\n",
       "Length: 206, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Categorical targets:\")\n",
    "train_unique_ds = y.nunique()\n",
    "train_unique_ds[train_unique_ds < 200]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "7a43573b-185f-4e7b-899f-b8940b92877d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unique values for cp_type: 2\n",
      "Unique values for cp_time: 3\n",
      "Unique values for cp_dose: 2\n"
     ]
    }
   ],
   "source": [
    "X, cat_ids, cat_dimensions, encoders = encode_categorical(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "815d62a8-3ded-4251-afb5-05e9627b831b",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>cp_type</th>\n",
       "      <th>cp_time</th>\n",
       "      <th>cp_dose</th>\n",
       "      <th>g-0</th>\n",
       "      <th>g-1</th>\n",
       "      <th>g-2</th>\n",
       "      <th>g-3</th>\n",
       "      <th>g-4</th>\n",
       "      <th>g-5</th>\n",
       "      <th>g-6</th>\n",
       "      <th>...</th>\n",
       "      <th>c-90</th>\n",
       "      <th>c-91</th>\n",
       "      <th>c-92</th>\n",
       "      <th>c-93</th>\n",
       "      <th>c-94</th>\n",
       "      <th>c-95</th>\n",
       "      <th>c-96</th>\n",
       "      <th>c-97</th>\n",
       "      <th>c-98</th>\n",
       "      <th>c-99</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0620</td>\n",
       "      <td>0.5577</td>\n",
       "      <td>-0.2479</td>\n",
       "      <td>-0.6208</td>\n",
       "      <td>-0.1944</td>\n",
       "      <td>-1.0120</td>\n",
       "      <td>-1.0220</td>\n",
       "      <td>...</td>\n",
       "      <td>0.2862</td>\n",
       "      <td>0.2584</td>\n",
       "      <td>0.8076</td>\n",
       "      <td>0.5523</td>\n",
       "      <td>-0.1912</td>\n",
       "      <td>0.6584</td>\n",
       "      <td>-0.3981</td>\n",
       "      <td>0.2139</td>\n",
       "      <td>0.3801</td>\n",
       "      <td>0.4176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0743</td>\n",
       "      <td>0.4087</td>\n",
       "      <td>0.2991</td>\n",
       "      <td>0.0604</td>\n",
       "      <td>1.0190</td>\n",
       "      <td>0.5207</td>\n",
       "      <td>0.2341</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.4265</td>\n",
       "      <td>0.7543</td>\n",
       "      <td>0.4708</td>\n",
       "      <td>0.0230</td>\n",
       "      <td>0.2957</td>\n",
       "      <td>0.4899</td>\n",
       "      <td>0.1522</td>\n",
       "      <td>0.1241</td>\n",
       "      <td>0.6077</td>\n",
       "      <td>0.7371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.6280</td>\n",
       "      <td>0.5817</td>\n",
       "      <td>1.5540</td>\n",
       "      <td>-0.0764</td>\n",
       "      <td>-0.0323</td>\n",
       "      <td>1.2390</td>\n",
       "      <td>0.1715</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.7250</td>\n",
       "      <td>-0.6297</td>\n",
       "      <td>0.6103</td>\n",
       "      <td>0.0223</td>\n",
       "      <td>-1.3240</td>\n",
       "      <td>-0.3174</td>\n",
       "      <td>-0.6417</td>\n",
       "      <td>-0.2187</td>\n",
       "      <td>-1.4080</td>\n",
       "      <td>0.6931</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 875 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   cp_type  cp_time  cp_dose     g-0     g-1     g-2     g-3     g-4     g-5  \\\n",
       "0        1        0        0  1.0620  0.5577 -0.2479 -0.6208 -0.1944 -1.0120   \n",
       "1        1        2        0  0.0743  0.4087  0.2991  0.0604  1.0190  0.5207   \n",
       "2        1        1        0  0.6280  0.5817  1.5540 -0.0764 -0.0323  1.2390   \n",
       "\n",
       "      g-6  ...    c-90    c-91    c-92    c-93    c-94    c-95    c-96  \\\n",
       "0 -1.0220  ...  0.2862  0.2584  0.8076  0.5523 -0.1912  0.6584 -0.3981   \n",
       "1  0.2341  ... -0.4265  0.7543  0.4708  0.0230  0.2957  0.4899  0.1522   \n",
       "2  0.1715  ... -0.7250 -0.6297  0.6103  0.0223 -1.3240 -0.3174 -0.6417   \n",
       "\n",
       "     c-97    c-98    c-99  \n",
       "0  0.2139  0.3801  0.4176  \n",
       "1  0.1241  0.6077  0.7371  \n",
       "2 -0.2187 -1.4080  0.6931  \n",
       "\n",
       "[3 rows x 875 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ec1a12f4-f2dd-462d-b4f1-1ff32eb1f6f5",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "b4bae570-0c26-4561-9b9c-565d7ac49ab5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Number of NAs in dataset: 0'"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "is_na = lambda x: f\"Number of NAs in dataset: {x.isna().sum().sum()}\"\n",
    "is_na(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "d0768a46-f0ce-4e29-916b-d07857e1ac9f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Number of NAs in dataset: 9892'"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# make some NAs\n",
    "for col in random.sample(list(range(X.shape[1])), 8):\n",
    "    for row in random.sample(list(range(X.shape[0])), random.randint(0, int(0.1 * X.shape[0]))):\n",
    "        X.iat[row, col] = np.nan\n",
    "is_na(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "bb38ab67-d626-42a8-8dc3-39ab370924d5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"Columns with NAs: ['g-32', 'g-250', 'g-280', 'g-410', 'g-554', 'g-641', 'g-647', 'c-81']\""
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "columns_with_nas = X.columns[X.isnull().any()].to_list()\n",
    "f\"Columns with NAs: {columns_with_nas}\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "3b0c8f67-2715-4484-85dd-bbba2b0000f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# draw NAs\n",
    "import missingno as msno\n",
    "msno.matrix(X[columns_with_nas].sample(1000))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "f3f39056-3c97-4ef3-b58f-6686ac323557",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Number of NAs in dataset: 0'"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# impute missing values\n",
    "X.fillna(X.mean(), inplace=True)\n",
    "is_na(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e9643095-6b65-43d5-9cbe-31378ec4b5db",
   "metadata": {
    "tags": []
   },
   "source": [
    "#### Make K folds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "b4d1410a-f791-40f4-ae11-10d112241371",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "kf = KFold(n_splits=5, shuffle=True)\n",
    "folds = {fn: (train_idx, val_idx) for fn, (train_idx, val_idx) in enumerate(kf.split(X, y))}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "17c64b32-0ed0-4770-8ea0-e244f5d72995",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of targets to predict (after targets with one value is removed): 177\n"
     ]
    }
   ],
   "source": [
    "# simplification trick - keep only tasks with more than one label in train and valid\n",
    "targets = set(y.columns)\n",
    "for fn in folds:\n",
    "    y_train_check = y.iloc[folds[fn][0]]\n",
    "    one_cat_train_columns = y_train_check.columns[y_train_check.nunique() < 2]\n",
    "    y_valid_check = y.iloc[folds[fn][1]]\n",
    "    one_cat_valid_columns = y_valid_check.columns[y_valid_check.nunique() < 2]\n",
    "    targets = targets - set(one_cat_train_columns) - set(one_cat_valid_columns)\n",
    "targets = list(targets)\n",
    "print(f\"Number of targets to predict (after targets with one value is removed): {len(targets)}\")\n",
    "y_to_predict = y[targets]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c4461ba5-704e-4879-9d70-1e12cf4f1ce0",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Define Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "22eb6616-a7bc-419a-b9bd-a09bc440cf6a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def init_tabnet():\n",
    "    print(\"New TabNetMultiTaskClassifier initialized.\")\n",
    "    return TabNetMultiTaskClassifier(\n",
    "        cat_idxs=cat_ids,\n",
    "        cat_dims=cat_dimensions,\n",
    "        cat_emb_dim=2,\n",
    "        optimizer_fn=torch.optim.Adam,\n",
    "        optimizer_params=dict(lr=2e-2),\n",
    "        scheduler_params={\"step_size\":50, \"gamma\":0.9},\n",
    "        scheduler_fn=torch.optim.lr_scheduler.StepLR,\n",
    "        mask_type=\"entmax\",\n",
    "        lambda_sparse=0,\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "d94bee9b-a38a-4c32-85ad-83632f04827e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "New TabNetMultiTaskClassifier initialized.\n",
      "Device used : cuda\n",
      "Training for fold 0:\n",
      "epoch 0  | loss: 0.542   | train_auc: 0.5029  | train_logloss: 0.11152 | valid_auc: 0.48501 | valid_logloss: 0.11227 |  0:00:03s\n",
      "epoch 40 | loss: 0.02035 | train_auc: 0.68521 | train_logloss: 0.02018 | valid_auc: 0.66247 | valid_logloss: 0.02011 |  0:02:36s\n",
      "epoch 80 | loss: 0.01913 | train_auc: 0.74811 | train_logloss: 0.01891 | valid_auc: 0.69773 | valid_logloss: 0.01908 |  0:05:10s\n",
      "epoch 120| loss: 0.0184  | train_auc: 0.78263 | train_logloss: 0.01811 | valid_auc: 0.71675 | valid_logloss: 0.01857 |  0:07:40s\n",
      "epoch 160| loss: 0.01764 | train_auc: 0.81751 | train_logloss: 0.01735 | valid_auc: 0.72617 | valid_logloss: 0.01844 |  0:10:14s\n",
      "\n",
      "Early stopping occurred at epoch 199 with best_epoch = 149 and best_valid_logloss = 0.0184\n",
      "Best weights from best epoch are automatically used!\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "clf_0, valid_aucs_0 = fit_and_predict(init_tabnet(), X, y_to_predict, folds, fold_ix=0, max_epochs=201, verbose=40)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "9355e745-1b30-4b46-8ab5-e7992981a2eb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "New TabNetMultiTaskClassifier initialized.\n",
      "Device used : cuda\n",
      "Training for fold 1:\n",
      "epoch 0  | loss: 0.55225 | train_auc: 0.50971 | train_logloss: 0.12263 | valid_auc: 0.51129 | valid_logloss: 0.12207 |  0:00:03s\n",
      "epoch 40 | loss: 0.02074 | train_auc: 0.67912 | train_logloss: 0.02049 | valid_auc: 0.65774 | valid_logloss: 0.02104 |  0:02:37s\n",
      "epoch 80 | loss: 0.0193  | train_auc: 0.76081 | train_logloss: 0.01899 | valid_auc: 0.71428 | valid_logloss: 0.01977 |  0:05:12s\n",
      "epoch 120| loss: 0.01789 | train_auc: 0.80153 | train_logloss: 0.01757 | valid_auc: 0.72921 | valid_logloss: 0.01907 |  0:07:48s\n",
      "epoch 160| loss: 0.01712 | train_auc: 0.83004 | train_logloss: 0.01679 | valid_auc: 0.73918 | valid_logloss: 0.01901 |  0:10:24s\n",
      "\n",
      "Early stopping occurred at epoch 194 with best_epoch = 144 and best_valid_logloss = 0.01891\n",
      "Best weights from best epoch are automatically used!\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "clf_1, valid_aucs_1 = fit_and_predict(init_tabnet(), X, y_to_predict, folds, fold_ix=1, max_epochs=201, verbose=40)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4756482e-e40e-46d8-8fc0-bb237ea2e2ae",
   "metadata": {
    "tags": []
   },
   "source": [
    "#### Feature importances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "3bae3956-cc7c-4bed-90af-cb41047178a2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "feature_importances(clf_0, X, 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "65803467-4da8-4b24-a186-a856d900f216",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "feature_importances(clf_1, X, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "c64ebdf1-8fff-4b8f-9007-0d171033c959",
   "metadata": {},
   "outputs": [],
   "source": [
    "_, _, X_fold_3, _ = extract_fold(X, y_to_predict, folds, fold_ix=3)\n",
    "explain_matrix, masks = clf_0.explain(X_fold_3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "b5520ef5-fdfd-4efe-95a9-8063294bbbe4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TabNet Explain Matrix for fold 3:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>865</th>\n",
       "      <th>866</th>\n",
       "      <th>867</th>\n",
       "      <th>868</th>\n",
       "      <th>869</th>\n",
       "      <th>870</th>\n",
       "      <th>871</th>\n",
       "      <th>872</th>\n",
       "      <th>873</th>\n",
       "      <th>874</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4763.000000</td>\n",
       "      <td>4763.000000</td>\n",
       "      <td>4763.000000</td>\n",
       "      <td>4763.0</td>\n",
       "      <td>4763.0</td>\n",
       "      <td>4763.000000</td>\n",
       "      <td>4763.0</td>\n",
       "      <td>4.763000e+03</td>\n",
       "      <td>4763.000000</td>\n",
       "      <td>4763.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>4763.0</td>\n",
       "      <td>4763.000000</td>\n",
       "      <td>4.763000e+03</td>\n",
       "      <td>4763.000000</td>\n",
       "      <td>4763.0</td>\n",
       "      <td>4763.0</td>\n",
       "      <td>4763.0</td>\n",
       "      <td>4763.0</td>\n",
       "      <td>4763.0</td>\n",
       "      <td>4763.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.622300</td>\n",
       "      <td>0.000001</td>\n",
       "      <td>0.000004</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000078</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.882887e-07</td>\n",
       "      <td>0.000011</td>\n",
       "      <td>0.000006</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.020278</td>\n",
       "      <td>8.879488e-10</td>\n",
       "      <td>0.000103</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4.003972</td>\n",
       "      <td>0.000063</td>\n",
       "      <td>0.000192</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.004393</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.989610e-05</td>\n",
       "      <td>0.000288</td>\n",
       "      <td>0.000241</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.164763</td>\n",
       "      <td>5.783930e-08</td>\n",
       "      <td>0.001733</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.358058</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.476332</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.592175</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>18.291238</td>\n",
       "      <td>0.004210</td>\n",
       "      <td>0.013069</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.295266</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.373119e-03</td>\n",
       "      <td>0.015323</td>\n",
       "      <td>0.014694</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.756405</td>\n",
       "      <td>3.984274e-06</td>\n",
       "      <td>0.067063</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 875 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               0            1            2       3       4            5    \\\n",
       "count  4763.000000  4763.000000  4763.000000  4763.0  4763.0  4763.000000   \n",
       "mean      1.622300     0.000001     0.000004     0.0     0.0     0.000078   \n",
       "std       4.003972     0.000063     0.000192     0.0     0.0     0.004393   \n",
       "min       0.000000     0.000000     0.000000     0.0     0.0     0.000000   \n",
       "25%       0.358058     0.000000     0.000000     0.0     0.0     0.000000   \n",
       "50%       0.476332     0.000000     0.000000     0.0     0.0     0.000000   \n",
       "75%       0.592175     0.000000     0.000000     0.0     0.0     0.000000   \n",
       "max      18.291238     0.004210     0.013069     0.0     0.0     0.295266   \n",
       "\n",
       "          6             7            8            9    ...     865  \\\n",
       "count  4763.0  4.763000e+03  4763.000000  4763.000000  ...  4763.0   \n",
       "mean      0.0  2.882887e-07     0.000011     0.000006  ...     0.0   \n",
       "std       0.0  1.989610e-05     0.000288     0.000241  ...     0.0   \n",
       "min       0.0  0.000000e+00     0.000000     0.000000  ...     0.0   \n",
       "25%       0.0  0.000000e+00     0.000000     0.000000  ...     0.0   \n",
       "50%       0.0  0.000000e+00     0.000000     0.000000  ...     0.0   \n",
       "75%       0.0  0.000000e+00     0.000000     0.000000  ...     0.0   \n",
       "max       0.0  1.373119e-03     0.015323     0.014694  ...     0.0   \n",
       "\n",
       "               866           867          868     869     870     871     872  \\\n",
       "count  4763.000000  4.763000e+03  4763.000000  4763.0  4763.0  4763.0  4763.0   \n",
       "mean      0.020278  8.879488e-10     0.000103     0.0     0.0     0.0     0.0   \n",
       "std       0.164763  5.783930e-08     0.001733     0.0     0.0     0.0     0.0   \n",
       "min       0.000000  0.000000e+00     0.000000     0.0     0.0     0.0     0.0   \n",
       "25%       0.000000  0.000000e+00     0.000000     0.0     0.0     0.0     0.0   \n",
       "50%       0.000000  0.000000e+00     0.000000     0.0     0.0     0.0     0.0   \n",
       "75%       0.000000  0.000000e+00     0.000000     0.0     0.0     0.0     0.0   \n",
       "max       4.756405  3.984274e-06     0.067063     0.0     0.0     0.0     0.0   \n",
       "\n",
       "          873     874  \n",
       "count  4763.0  4763.0  \n",
       "mean      0.0     0.0  \n",
       "std       0.0     0.0  \n",
       "min       0.0     0.0  \n",
       "25%       0.0     0.0  \n",
       "50%       0.0     0.0  \n",
       "75%       0.0     0.0  \n",
       "max       0.0     0.0  \n",
       "\n",
       "[8 rows x 875 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"TabNet Explain Matrix for fold 3:\")\n",
    "pd.DataFrame(explain_matrix).describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "a8c7cbfd-cb5a-44b1-beb9-b780c180b73a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TabNet Masks:\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"TabNet Masks:\")\n",
    "fig, axs = plt.subplots(1, 3, figsize=(20, 20))\n",
    "\n",
    "around_idx_shift = 10\n",
    "for i in range(3):\n",
    "    strong_feature_idx = int(np.median(masks[i].argmax(axis=1)))\n",
    "    axs[i].imshow(masks[i][:20, max(0, strong_feature_idx - around_idx_shift): strong_feature_idx + around_idx_shift])\n",
    "    xticks_labels = list(range(max(0, strong_feature_idx - around_idx_shift), strong_feature_idx + around_idx_shift))\n",
    "    axs[i].set_title(f\"mask {i}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ce0d7b1-7df0-4f6b-9b29-cddf4d058aef",
   "metadata": {
    "tags": []
   },
   "source": [
    "### XGBoost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "f1d3a98b-98bb-4d22-837b-59ef9a217f3e",
   "metadata": {},
   "outputs": [],
   "source": [
    "def init_xgb():\n",
    "    print(\"New MultiOutputClassifier(XGBClassifier) initialized.\")\n",
    "    xgb_clf = MultiOutputClassifier(XGBClassifier(tree_method=\"gpu_hist\", use_label_encoder=False, verbosity=0))\n",
    "    params = {\n",
    "        \"estimator__colsample_bytree\": 0.6522,\n",
    "        \"estimator__gamma\": 3.6975,\n",
    "        \"estimator__learning_rate\": 0.0503,\n",
    "        \"estimator__max_delta_step\": 2.0706,\n",
    "        \"estimator__max_depth\": 10,\n",
    "        \"estimator__min_child_weight\": 31.5800,\n",
    "        \"estimator__n_estimators\": 166,\n",
    "        \"estimator__subsample\": 0.8639\n",
    "    }\n",
    "    xgb_clf = xgb_clf.set_params(**params)\n",
    "    return xgb_clf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "260ef34d-9757-44b9-93ba-2c34e44797a0",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "New MultiOutputClassifier(XGBClassifier) initialized.\n",
      "Training for fold 0:\n",
      "CPU times: user 5min 8s, sys: 9.43 s, total: 5min 18s\n",
      "Wall time: 2min 57s\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%time\n",
    "xgb_clf_0 = fit_and_predict_xgb(init_xgb(), X, y_to_predict, folds, 0, valid_aucs_0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "634c39d5-3799-4c06-b95e-bb465f6f45a3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Number of XGB Estimators: 177'"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f\"Number of XGB Estimators: {len(xgb_clf_0.estimators_)}\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "a274c7b7-a5d9-49c5-9b8e-52e4287c761a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "New MultiOutputClassifier(XGBClassifier) initialized.\n",
      "Training for fold 1:\n",
      "CPU times: user 5min 15s, sys: 9.48 s, total: 5min 25s\n",
      "Wall time: 2min 59s\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%time\n",
    "xgb_clf_1 = fit_and_predict_xgb(init_xgb(), X, y_to_predict, folds, 1, valid_aucs_1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "decd2e18-5df2-4659-9642-54138e0c347a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "feature_importances_xgb(xgb_clf_0, X, 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "2946817c-a57b-426c-be60-2c56e944ba89",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "feature_importances_xgb(xgb_clf_1, X, 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8804a967-03bb-4e7c-85e6-ff83e979e861",
   "metadata": {
    "tags": []
   },
   "source": [
    "### TabNet Pretrainer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "ae383b07-2eeb-4f0a-9275-6dc60299220e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Encode categorical variables in test dataset the same way as they were encoded in train:\n",
      "Unique values for cp_type: 2\n",
      "Unique values for cp_time: 3\n",
      "Unique values for cp_dose: 2\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'X_test shape: (3982, 875)'"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Encode categorical variables in test dataset the same way as they were encoded in train:\")\n",
    "X_test = pd.read_csv(\"data/test_features.csv\")\n",
    "X_test.drop(columns=[\"sig_id\"], inplace=True)\n",
    "for cat_id, encoder in zip(cat_ids, encoders):\n",
    "    feature = X.columns[cat_id]\n",
    "    print(f\"Unique values for {feature}: {X_test.nunique()[feature]}\")\n",
    "    X_test[feature] = X_test[feature].fillna(\"Encoded_by_Raid\")\n",
    "    X_test[feature] = encoder.transform(X_test[feature].values)\n",
    "f\"X_test shape: {X_test.shape}\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "a3859b92-c84b-40fc-9f78-90e7d27c4bdd",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Device used : cuda\n",
      "epoch 0  | loss: 1.74594 | val_0_unsup_loss: 1.01608 |  0:00:00s\n",
      "epoch 40 | loss: 0.77336 | val_0_unsup_loss: 0.76229 |  0:00:21s\n",
      "epoch 80 | loss: 0.73183 | val_0_unsup_loss: 0.72503 |  0:00:43s\n",
      "epoch 120| loss: 0.71696 | val_0_unsup_loss: 0.70848 |  0:01:04s\n",
      "epoch 160| loss: 0.69659 | val_0_unsup_loss: 0.68931 |  0:01:25s\n",
      "epoch 200| loss: 0.69151 | val_0_unsup_loss: 0.66957 |  0:01:47s\n",
      "epoch 240| loss: 0.66101 | val_0_unsup_loss: 0.65105 |  0:02:08s\n",
      "epoch 280| loss: 0.65833 | val_0_unsup_loss: 0.64616 |  0:02:30s\n",
      "Stop training because you reached max_epochs = 300 with best_epoch = 299 and best_val_0_unsup_loss = 0.63536\n",
      "Best weights from best epoch are automatically used!\n",
      "Successfully saved model at ./test_pretrained.zip\n",
      "Device used : cuda\n",
      "Device used : cuda\n"
     ]
    }
   ],
   "source": [
    "# Split data on train and valid for Pretrainer\n",
    "train_valid_split_percentage = 0.8\n",
    "idx = random.sample(list(range(X_test.shape[0])), X_test.shape[0])\n",
    "last_train_row_idx = int(train_valid_split_percentage * X_test.shape[0])\n",
    "X_train_pre, X_valid_pre = X_test.iloc[idx[:last_train_row_idx]].to_numpy(), X_test.iloc[idx[last_train_row_idx:]].to_numpy()\n",
    "\n",
    "# TabNetPretrainer\n",
    "unsupervised_model_no_preproc = TabNetPretrainer(\n",
    "    cat_idxs=cat_ids,\n",
    "    cat_dims=cat_dimensions,\n",
    "    cat_emb_dim=2,\n",
    "    optimizer_fn=torch.optim.Adam,\n",
    "    optimizer_params=dict(lr=2e-2),\n",
    "    mask_type=\"entmax\",\n",
    ")\n",
    "unsupervised_model_no_preproc.verbose = 40\n",
    "\n",
    "# fit the model\n",
    "unsupervised_model_no_preproc.fit(\n",
    "    X_train_pre,\n",
    "    eval_set=[X_valid_pre],\n",
    "    max_epochs=300 , patience=50,\n",
    "    batch_size=256, virtual_batch_size=128,\n",
    "    num_workers=0,\n",
    "    drop_last=False,\n",
    "    pretraining_ratio=0.8\n",
    ")\n",
    "\n",
    "# Make reconstruction from a dataset\n",
    "reconstructed_X, embedded_X = unsupervised_model_no_preproc.predict(X_valid_pre)\n",
    "assert(reconstructed_X.shape==embedded_X.shape)\n",
    "\n",
    "unsupervised_model_no_preproc.save_model(\"./test_pretrained\")\n",
    "loaded_pretrain = TabNetPretrainer()\n",
    "loaded_pretrain.load_model(\"./test_pretrained.zip\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "b9258956-7d29-4bc1-86e3-78be20883464",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "New TabNetMultiTaskClassifier initialized.\n",
      "Device used : cuda\n",
      "Training for fold 0:\n",
      "Loading weights from unsupervised pretraining\n",
      "epoch 0  | loss: 0.5367  | train_auc: 0.51377 | train_logloss: 0.21357 | valid_auc: 0.51424 | valid_logloss: 0.21226 |  0:00:04s\n",
      "epoch 40 | loss: 0.0182  | train_auc: 0.75786 | train_logloss: 0.01791 | valid_auc: 0.69792 | valid_logloss: 0.01865 |  0:02:47s\n",
      "epoch 80 | loss: 0.01613 | train_auc: 0.86    | train_logloss: 0.01575 | valid_auc: 0.70468 | valid_logloss: 0.01949 |  0:05:28s\n",
      "\n",
      "Early stopping occurred at epoch 106 with best_epoch = 56 and best_valid_logloss = 0.0185\n",
      "Best weights from best epoch are automatically used!\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "clf_pre_0, valid_aucs_pre_0 = fit_and_predict(init_tabnet(), X, y_to_predict, folds, fold_ix=0, max_epochs=201, verbose=40., unsupervised_model=loaded_pretrain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "6ea40272-e8ee-4622-a48d-f356abf59b85",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AUC Difference Median: -0.004175501437580276. Mean: -0.01920771574146591. Min: -0.4785804283914322. Max: 0.33077015499219026\n"
     ]
    }
   ],
   "source": [
    "auc_diff = [auc_pre0 - auc0 for auc_pre0, auc0 in zip(valid_aucs_pre_0, valid_aucs_0)]\n",
    "print(f\"AUC Difference Median: {np.median(auc_diff)}. Mean: {np.mean(auc_diff)}. Min: {np.min(auc_diff)}. Max: {np.max(auc_diff)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5ae683c8-7809-4b6c-a0dc-c8b6386e3142",
   "metadata": {},
   "source": [
    "### Train/test features distribution similarity check (Kolmogorov–Smirnov test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "a7b068fc-242e-4ceb-9385-329e20ffcf6e",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All the features have the same distribution in train and test datasets!\n"
     ]
    }
   ],
   "source": [
    "diff_features = get_diff_distribution_features(X, X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "6ad74170-2c7e-41db-a758-e9172660e30b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "X[\"g-2\"].hist(bins=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "4125f38e-76d4-4a30-a215-5d3b6bfe02dc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD4CAYAAAAXUaZHAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAATqklEQVR4nO3df4zk9X3f8ecr4B+ETTksnC09Tj0q40SYq7HZYKduq92QHxii4EiJi0VcsF1dGoHrticl2FGVVCnStQl2HTlxezHUpCbeUIwDwpCGEM6R/wDbhzHLj7i52mdzW3KYGK5em9o6590/9nthOGZvf8zMzt5nnw9ptTOf7/c789qbudd+9zPf+U6qCklSW75v3AEkScNnuUtSgyx3SWqQ5S5JDbLcJalBJ487AMAZZ5xR27dvH3cMAL71rW9x6qmnjjvGksw3GPMNxnyDGXa+ffv2PV1Vr+y7sKrG/nXBBRfURnHfffeNO8JxmW8w5huM+QYz7HzA52uJXnVaRpIaZLlLUoMsd0lqkOUuSQ2y3CWpQZa7JDXIcpekBlnuktQgy12SGmS5a1OZmz/M9ms/xfZrPzXuKNJIWe6S1CDLXZIaZLlLUoMsd0lqkOUuSQ2y3CWpQRvik5ikUeo97HHXjjEGkdaRe+6S1CDLXZIaZLlLUoOWLfckL0/y2SRfTPJokn/fjZ+d5IEk+5P8YZKXduMv667v75ZvH/HPIEk6xkpeUP0O8GNVtZDkJcBnktwN/FvgA1U1m+S/AO8CPtx9f6aqXpXkcuA/Av9sRPmlvjx3jDa7Zffca9FCd/Ul3VcBPwbc2o3fBLylu3xZd51u+UVJMqzAkqTlpaqWXyk5CdgHvAr4HeA3gfur6lXd8m3A3VV1XpJHgIur6mC37H8Db6iqp4+5zZ3AToDJyckLZmdnh/dTDWBhYYGJiYlxx1iS+VZmbv5w3/HJU+DQc4uXd2w9bR0TrcxG+fdbivkGM+x8MzMz+6pqqt+yFR3nXlXfA85PsgX4JPDDg4aqqj3AHoCpqamanp4e9CaHYu/evWyULP2Yb2WuWmJaZteOI1w/t/i0P3DF9DomWpmN8u+3FPMNZj3zrepomap6FrgP+FFgS5KjvxzOAua7y/PANoBu+WnAXw8jrCRpZVZytMwruz12kpwC/ATwOIsl/3PdalcCt3eX7+iu0y3/s1rJ3I8kaWhWMi1zJnBTN+/+fcAtVXVnkseA2ST/AfgCcEO3/g3Af0+yH/gGcPkIckuSjmPZcq+qh4HX9Rn/MnBhn/H/B/z8UNJJktbEd6hKUoMsd0lqkOUuSQ2y3CWpQZa7JDXIcpekBlnuktQgP0NVm9axpwU+sPvSMSWRhs89d0lqkOUuSQ2y3CWpQZa7JDXIcpekBlnuktQgD4VUM449tFHazNxzl6QGWe6S1CDLXZIaZLlLUoMsd0lqkOUuSQ2y3CWpQZa7JDVo2XJPsi3JfUkeS/Jokvd047+eZD7JQ93XJT3bvDfJ/iRfSvJTo/wBJEkvtpJ3qB4BdlXVg0l+ANiX5J5u2Qeq6rd6V05yLnA58Brg7wF/muTVVfW9YQaXJC1t2T33qnqyqh7sLn8TeBzYepxNLgNmq+o7VfUVYD9w4TDCSpJWZlVz7km2A68DHuiGrknycJIbk5zejW0FnujZ7CDH/2UgSRqyVNXKVkwmgE8D11XVbUkmgaeBAn4DOLOq3pnkQ8D9VfWxbrsbgLur6tZjbm8nsBNgcnLygtnZ2WH9TANZWFhgYmJi3DGWZL6lzc0fXnadyVPg0HP9l+3YetqQE62ej+9gNlu+mZmZfVU11W/Zis4KmeQlwCeAm6vqNoCqOtSz/PeAO7ur88C2ns3P6sZeoKr2AHsApqamanp6eiVRRm7v3r1slCz9mG9pV63grJC7dhzh+rn+T/sDV0wPOdHq+fgOxnzPW7bckwS4AXi8qt7fM35mVT3ZXf1Z4JHu8h3AHyR5P4svqJ4DfHaoqaUR6D1l8IHdl44xiTS4ley5vwl4OzCX5KFu7H3A25Kcz+K0zAHgFwGq6tEktwCPsXikzdUeKSNJ62vZcq+qzwDps+iu42xzHXDdALkkSQPwk5h0QhvVpy85RaMTnacfkKQGWe6S1CDLXZIaZLlLUoMsd0lqkOUuSQ2y3CWpQZa7JDXIcpekBlnuktQgy12SGmS5S1KDLHdJapDlLkkNstwlqUGWuyQ1yHKXpAb5SUzSMvxUJp2I3HOXpAZZ7pLUIMtdkhpkuUtSg5Yt9yTbktyX5LEkjyZ5Tzf+iiT3JPnL7vvp3XiS/HaS/UkeTvL6Uf8QkqQXWsme+xFgV1WdC7wRuDrJucC1wL1VdQ5wb3cd4M3AOd3XTuDDQ08tSTquZcu9qp6sqge7y98EHge2ApcBN3Wr3QS8pbt8GfD7teh+YEuSM4cdXJK0tFXNuSfZDrwOeACYrKonu0V/BUx2l7cCT/RsdrAbkyStk1TVylZMJoBPA9dV1W1Jnq2qLT3Ln6mq05PcCeyuqs904/cCv1JVnz/m9nayOG3D5OTkBbOzs0P5gQa1sLDAxMTEuGMsyXwvNDd/eFXrT54Ch55b+/3t2Hra2jdeAR/fwWy2fDMzM/uqaqrfshW9QzXJS4BPADdX1W3d8KEkZ1bVk920y1Pd+DywrWfzs7qxF6iqPcAegKmpqZqenl5JlJHbu3cvGyVLP+Z7oat63j26Ert2HOH6ubW/MfvAFdNr3nYlfHwHY77nLfssTxLgBuDxqnp/z6I7gCuB3d3323vGr0kyC7wBONwzfSMNbPsqC13ajFayC/Mm4O3AXJKHurH3sVjqtyR5F/BV4K3dsruAS4D9wLeBdwwzsCRpecuWezd3niUWX9Rn/QKuHjCXJGkAvkNVkhpkuUtSgyx3SWqQ5S5JDbLcJalBlrskNchyl6QGWe6S1CDLXZIaZLlLUoMsd0lqkOUuSQ2y3CWpQZa7JDXIcpekBlnuktQgy12SGmS5S1KDLHdJapDlLkkNstwlqUGWuyQ1yHKXpAYtW+5JbkzyVJJHesZ+Pcl8koe6r0t6lr03yf4kX0ryU6MKLkla2kr23D8KXNxn/ANVdX73dRdAknOBy4HXdNv8bpKThhVWkrQyy5Z7Vf058I0V3t5lwGxVfaeqvgLsBy4cIJ8kaQ0GmXO/JsnD3bTN6d3YVuCJnnUOdmOSpHWUqlp+pWQ7cGdVndddnwSeBgr4DeDMqnpnkg8B91fVx7r1bgDurqpb+9zmTmAnwOTk5AWzs7PD+YkGtLCwwMTExLhjLMl8MDd/eM3bTp4Ch55b+33v2Hra2jdeAR/fwWy2fDMzM/uqaqrfspPXcoNVdejo5SS/B9zZXZ0HtvWselY31u829gB7AKampmp6enotUYZu7969bJQs/ZgPrrr2U2vedteOI1w/t6anPQAHrphe87Yr4eM7GPM9b03TMknO7Ln6s8DRI2nuAC5P8rIkZwPnAJ8dLKIkabWW3YVJ8nFgGjgjyUHg14DpJOezOC1zAPhFgKp6NMktwGPAEeDqqvreSJJrU9k+wN66tBktW+5V9bY+wzccZ/3rgOsGCSVJGozvUJWkBlnuktQgy12SGmS5S1KDLHdJapDlLkkNWvtb9aRNrvfY+wO7Lx1jEunF3HOXpAZZ7pLUIMtdkhpkuUtSg3xBVVoFT2CmE4V77pLUIMtdkhpkuUtSgyx3SWqQ5S5JDbLcJalBlrskNchyl6QGWe6S1CDLXZIaZLlLUoM8t4w0BH5whzaaZffck9yY5Kkkj/SMvSLJPUn+svt+ejeeJL+dZH+Sh5O8fpThJUn9rWRa5qPAxceMXQvcW1XnAPd21wHeDJzTfe0EPjycmJKk1Vi23Kvqz4FvHDN8GXBTd/km4C09479fi+4HtiQ5c0hZJUkrlKpafqVkO3BnVZ3XXX+2qrZ0lwM8U1VbktwJ7K6qz3TL7gV+pao+3+c2d7K4d8/k5OQFs7Ozw/mJBrSwsMDExMS4Yyxps+abmz88lNuZPAUOPTeUm1rSjq2nrXnbzfr4DstmyzczM7Ovqqb6LRv4BdWqqiTL/4Z48XZ7gD0AU1NTNT09PWiUodi7dy8bJUs/mzXfVUP6kIxdO45w/dxojyM4cMX0mrfdrI/vsJjveWs9FPLQ0emW7vtT3fg8sK1nvbO6MUnSOlprud8BXNldvhK4vWf8n3dHzbwROFxVTw6YUZK0Ssv+fZrk48A0cEaSg8CvAbuBW5K8C/gq8NZu9buAS4D9wLeBd4wgsyRpGcuWe1W9bYlFF/VZt4CrBw0lnch8Q5M2At+hqg1r+5BeRJU2I88tI0kNstwlqUGWuyQ1yHKXpAZZ7pLUIMtdkhpkuUtSgyx3SWqQ5S5JDbLcJalBlrskNchyl6QGWe6S1CDPCqkNxTNBSsPhnrskNchyl6QGOS0jjZCfyqRxcc9dkhrknru0TtyL13pyz12SGmS5S1KDLHdJatBAc+5JDgDfBL4HHKmqqSSvAP4Q2A4cAN5aVc8MFlOStBrD2HOfqarzq2qqu34tcG9VnQPc212XJK2jUUzLXAbc1F2+CXjLCO5DknQcqaq1b5x8BXgGKOC/VtWeJM9W1ZZueYBnjl4/ZtudwE6AycnJC2ZnZ9ecY5gWFhaYmJgYd4wltZ5vbv7wENO82OQpcOi5kd7FiuzYelrf8dYf31HbbPlmZmb29cyavMCgx7n/46qaT/KDwD1J/qJ3YVVVkr6/PapqD7AHYGpqqqanpweMMhx79+5lo2Tpp/V8V434xGG7dhzh+rnxv73jwBXTfcdbf3xHzXzPG2hapqrmu+9PAZ8ELgQOJTkToPv+1KAhJUmrs+ZyT3Jqkh84ehn4SeAR4A7gym61K4HbBw0pSVqdQf4+nQQ+uTitzsnAH1TVHyf5HHBLkncBXwXeOnhMSdJqrLncq+rLwGv7jP81cNEgoSRJgxn/K0va9Pz0JWn4PP2AJDXIcpekBjkto7FwKuZ5vf8WH7341DEmUUssd2kM/OWmUXNaRpIaZLlLUoMsd0lqkOUubSBz84fZfu2nnJPXwCx3SWqQR8topHr3QA/svnSMSaTNxT13SWqQe+7SBuVfPRqE5a5144uEa2fRa7Usd+kEY9FrJSx3qRGWvnpZ7ho6p1/Wj//WWorlrr5WshfYr1h27TiCTytp/DwUUpIaZLlLUoP8+1lq0FJz8b7QunlY7lqWL9q1w9LfPJyWkaQGjWzPPcnFwAeBk4CPVNXuUd2Xju94e97usQnWdoy8x9VvbCMp9yQnAb8D/ARwEPhckjuq6rFR3N9Gttr/AMNa36kUDcuwPsDbXwbra1R77hcC+6vqywBJZoHLgKGX+3o8YVZSoKst1rWus2vHEaYHuM1hbKO2He85MTd/mKtW8ZxZ6rYG+T+1nr8YRnW/6/HzpKqGf6PJzwEXV9W/6K6/HXhDVV3Ts85OYGd39YeALw09yNqcATw97hDHYb7BmG8w5hvMsPP9/ap6Zb8FYztapqr2AHvGdf9LSfL5qpoad46lmG8w5huM+QaznvlGdbTMPLCt5/pZ3ZgkaR2Mqtw/B5yT5OwkLwUuB+4Y0X1Jko4xkmmZqjqS5Brgf7J4KOSNVfXoKO5rBDbcVNExzDcY8w3GfINZt3wjeUFVkjRevkNVkhpkuUtSgyz3JSR5d5K/SPJokv807jz9JNmVpJKcMe4svZL8Zvdv93CSTybZMu5MsHhKjCRfSrI/ybXjztMrybYk9yV5rHvOvWfcmfpJclKSLyS5c9xZjpVkS5Jbu+fe40l+dNyZeiX5N91j+0iSjyd5+Sjvz3LvI8kMi++ofW1VvQb4rTFHepEk24CfBL427ix93AOcV1X/EPhfwHvHnKf3lBhvBs4F3pbk3PGmeoEjwK6qOhd4I3D1Bst31HuAx8cdYgkfBP64qn4YeC0bKGeSrcC/Aqaq6jwWDzS5fJT3abn390vA7qr6DkBVPTXmPP18APhlYMO9Il5Vf1JVR7qr97P4Podx+9tTYlTVd4Gjp8TYEKrqyap6sLv8TRaLaet4U71QkrOAS4GPjDvLsZKcBvxT4AaAqvpuVT071lAvdjJwSpKTge8H/s8o78xy7+/VwD9J8kCSTyf5kXEH6pXkMmC+qr447iwr8E7g7nGHYLEon+i5fpANVp5HJdkOvA54YMxRjvWfWdyh+Jsx5+jnbODrwH/rpo0+kmTtZzkbsqqaZ3EG4GvAk8DhqvqTUd7npv2wjiR/CvzdPot+lcV/l1ew+OfxjwC3JPkHtY7HjS6T730sTsmMzfHyVdXt3Tq/yuJ0w83rme1ElmQC+ATwr6vq/447z1FJfhp4qqr2JZkec5x+TgZeD7y7qh5I8kHgWuDfjTfWoiSns/iX4tnAs8D/SPILVfWxUd3npi33qvrxpZYl+SXgtq7MP5vkb1g84c/Xx50vyQ4WnyBfTAKLUx4PJrmwqv5q3PmOSnIV8NPARev5S/E4NvwpMZK8hMViv7mqbht3nmO8CfiZJJcALwf+TpKPVdUvjDnXUQeBg1V19K+dW1ks943ix4GvVNXXAZLcBvwjYGTl7rRMf38EzAAkeTXwUjbImeaqaq6qfrCqtlfVdhaf1K9fz2JfTvdBLb8M/ExVfXvceTob+pQYWfxNfQPweFW9f9x5jlVV762qs7rn3OXAn22gYqd7/j+R5Ie6oYsYwSnGB/A14I1Jvr97rC9ixC/4bto992XcCNyY5BHgu8CVG2Tv80TxIeBlwD3dXxf3V9W/HGegE+CUGG8C3g7MJXmoG3tfVd01vkgnnHcDN3e/vL8MvGPMef5WN1V0K/Agi1OVX2DEpyLw9AOS1CCnZSSpQZa7JDXIcpekBlnuktQgy12SGmS5S1KDLHdJatD/B65CEc6i8yFRAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "X_test[\"g-2\"].hist(bins=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "0c46492e-49f3-41a7-9e95-863b28c558cd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "X[\"cp_type\"].hist(bins=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "8b35ddfd-b76d-47e1-bf86-4c7cea07a45c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD4CAYAAAAAczaOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAAUtklEQVR4nO3df6zd9X3f8ecr/EiYyQIp6ZVnvJkpjjYSVGBXQJVpu4QFDH/EVGsjkFociuYug6ndUFXSSSOFISXanEggSusIz1DREJY2tUXokEc4QplmwDQEYyjjFkix58AaG7c3rGxm7/1xvqSn3rXv8f1xbg6f50M6ut/z+X4+3/N53+/163zP93zPcaoKSVIb3rPcE5AkjY6hL0kNMfQlqSGGviQ1xNCXpIacuNwTOJYzzjij1qxZM+/xP/zhD1mxYsXiTWgMtFZza/WCNbdiITU/9dRTf1ZVH5pt3Y916K9Zs4Zdu3bNe3yv12NqamrxJjQGWqu5tXrBmluxkJqTfO9o6zy9I0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWqIoS9JDfmx/kSuJAHs3neIz9z0zeWexkhtXbc0Xzvhkb4kNcTQl6SGGPqS1JA5Qz/J+5I8keS7SfYk+Y2ufWuSl5M83d3O7dqT5PYk00meSXL+wLY2JHmxu21YsqokSbMa5o3ct4BPVNVMkpOAbyf5w27dr1bV14/ofzmwtrtdCNwFXJjkg8DNwCRQwFNJtlfVwcUoRJI0tzmP9Ktvprt7UnerYwxZD9zbjdsJnJZkJXAZsKOqDnRBvwNYt7DpS5KOx1CXbCY5AXgK+DBwZ1U9nuSzwG1J/i3wCHBTVb0FrAJeHRi+t2s7WvuRj7UR2AgwMTFBr9c73pp+ZGZmZkHjx1FrNbdWL7RZ88QpcOM5h5d7GiO1VPt5qNCvqreBc5OcBnwjyceAzwHfB04GNgO/Btyy0AlV1eZue0xOTtZC/rcc/7edd7/W6oU2a77jvm1s2t3Wx4q2rluxJPv5uK7eqao3gEeBdVW1vzuF8xbwH4ELum77gNUDw87s2o7WLkkakWGu3vlQd4RPklOATwJ/3J2nJ0mAK4FnuyHbgWu6q3guAg5V1X7gYeDSJKcnOR24tGuTJI3IMK+XVgL3dOf13wM8UFUPJvlWkg8BAZ4G/nnX/yHgCmAaeBO4FqCqDiS5FXiy63dLVR1YtEokSXOaM/Sr6hngvFnaP3GU/gVcf5R1W4AtxzlHSdIi8RO5ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkPmDP0k70vyRJLvJtmT5De69rOSPJ5kOsnXkpzctb+3uz/drV8zsK3Pde0vJLlsyaqSJM1qmCP9t4BPVNVPAecC65JcBHwR+HJVfRg4CFzX9b8OONi1f7nrR5KzgauAjwLrgN9McsIi1iJJmsOcoV99M93dk7pbAZ8Avt613wNc2S2v7+7Trb8kSbr2+6vqrap6GZgGLliMIiRJwzlxmE7dEflTwIeBO4E/Ad6oqsNdl73Aqm55FfAqQFUdTnII+ImufefAZgfHDD7WRmAjwMTEBL1e7/gqGjAzM7Og8eOotZpbqxfarHniFLjxnMNzd3wXWar9PFToV9XbwLlJTgO+Afy9RZ/JXz3WZmAzwOTkZE1NTc17W71ej4WMH0et1dxavdBmzXfct41Nu4eKq3eNretWLMl+Pq6rd6rqDeBR4KeB05K8sxfOBPZ1y/uA1QDd+g8APxhsn2WMJGkEhrl650PdET5JTgE+CTxPP/x/tuu2AdjWLW/v7tOt/1ZVVdd+VXd1z1nAWuCJRapDkjSEYV4vrQTu6c7rvwd4oKoeTPIccH+Sfwd8B7i763838DtJpoED9K/Yoar2JHkAeA44DFzfnTaSJI3InKFfVc8A583S/hKzXH1TVX8J/NxRtnUbcNvxT1OStBj8RK4kNcTQl6SGGPqS1BBDX5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhoyZ+gnWZ3k0STPJdmT5Je79s8n2Zfk6e52xcCYzyWZTvJCkssG2td1bdNJblqakiRJR3PiEH0OAzdW1R8leT/wVJId3bovV9V/GOyc5GzgKuCjwN8C/kuSj3Sr7wQ+CewFnkyyvaqeW4xCJElzmzP0q2o/sL9b/oskzwOrjjFkPXB/Vb0FvJxkGrigWzddVS8BJLm/62voS9KIDHOk/yNJ1gDnAY8DHwduSHINsIv+q4GD9J8Qdg4M28tfPUm8ekT7hbM8xkZgI8DExAS9Xu94pvjXzMzMLGj8OGqt5tbqhTZrnjgFbjzn8HJPY6SWaj8PHfpJTgV+D/iVqvrzJHcBtwLV/dwE/OJCJ1RVm4HNAJOTkzU1NTXvbfV6PRYyfhy1VnNr9UKbNd9x3zY27T6uY9Sxt3XdiiXZz0P9FpOcRD/w76uq3weoqtcG1n8FeLC7uw9YPTD8zK6NY7RLkkZgmKt3AtwNPF9VXxpoXznQ7WeAZ7vl7cBVSd6b5CxgLfAE8CSwNslZSU6m/2bv9sUpQ5I0jGGO9D8O/AKwO8nTXduvA1cnOZf+6Z1XgF8CqKo9SR6g/wbtYeD6qnobIMkNwMPACcCWqtqzaJVIkuY0zNU73wYyy6qHjjHmNuC2WdofOtY4SdLS8hO5ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkPmDP0kq5M8muS5JHuS/HLX/sEkO5K82P08vWtPktuTTCd5Jsn5A9va0PV/McmGpStLkjSbYY70DwM3VtXZwEXA9UnOBm4CHqmqtcAj3X2Ay4G13W0jcBf0nySAm4ELgQuAm995opAkjcacoV9V+6vqj7rlvwCeB1YB64F7um73AFd2y+uBe6tvJ3BakpXAZcCOqjpQVQeBHcC6xSxGknRsx3VOP8ka4DzgcWCiqvZ3q74PTHTLq4BXB4bt7dqO1i5JGpETh+2Y5FTg94Bfqao/T/KjdVVVSWoxJpRkI/3TQkxMTNDr9ea9rZmZmQWNH0et1dxavdBmzROnwI3nHF7uaYzUUu3noUI/yUn0A/++qvr9rvm1JCuran93+ub1rn0fsHpg+Jld2z5g6oj23pGPVVWbgc0Ak5OTNTU1dWSXofV6PRYyfhy1VnNr9UKbNd9x3zY27R76GPVdYeu6FUuyn4e5eifA3cDzVfWlgVXbgXeuwNkAbBtov6a7iuci4FB3Guhh4NIkp3dv4F7atUmSRmSYp86PA78A7E7ydNf268AXgAeSXAd8D/h0t+4h4ApgGngTuBagqg4kuRV4sut3S1UdWIwiJEnDmTP0q+rbQI6y+pJZ+hdw/VG2tQXYcjwTlCQtHj+RK0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDXE0JekhswZ+km2JHk9ybMDbZ9Psi/J093tioF1n0syneSFJJcNtK/r2qaT3LT4pUiS5jLMkf5WYN0s7V+uqnO720MASc4GrgI+2o35zSQnJDkBuBO4HDgbuLrrK0kaoRPn6lBVjyVZM+T21gP3V9VbwMtJpoELunXTVfUSQJL7u77PHf+UJUnzNWfoH8MNSa4BdgE3VtVBYBWwc6DP3q4N4NUj2i+cbaNJNgIbASYmJuj1evOe4MzMzILGj6PWam6tXmiz5olT4MZzDi/3NEZqqfbzfEP/LuBWoLqfm4BfXIwJVdVmYDPA5ORkTU1NzXtbvV6PhYwfR63V3Fq90GbNd9y3jU27F3KMOn62rluxJPt5Xr/FqnrtneUkXwEe7O7uA1YPdD2za+MY7ZKkEZnXJZtJVg7c/RngnSt7tgNXJXlvkrOAtcATwJPA2iRnJTmZ/pu92+c/bUnSfMx5pJ/kq8AUcEaSvcDNwFSSc+mf3nkF+CWAqtqT5AH6b9AeBq6vqre77dwAPAycAGypqj2LXYwk6diGuXrn6lma7z5G/9uA22Zpfwh46LhmJ0laVH4iV5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWrInKGfZEuS15M8O9D2wSQ7krzY/Ty9a0+S25NMJ3kmyfkDYzZ0/V9MsmFpypEkHcswR/pbgXVHtN0EPFJVa4FHuvsAlwNru9tG4C7oP0kANwMXAhcAN7/zRCFJGp05Q7+qHgMOHNG8HrinW74HuHKg/d7q2wmclmQlcBmwo6oOVNVBYAf//xOJJGmJzfec/kRV7e+Wvw9MdMurgFcH+u3t2o7WLkkaoRMXuoGqqiS1GJMBSLKR/qkhJiYm6PV6897WzMzMgsaPo9Zqbq1eaLPmiVPgxnMOL/c0Rmqp9vN8Q/+1JCuran93+ub1rn0fsHqg35ld2z5g6oj23mwbrqrNwGaAycnJmpqamq3bUHq9HgsZP45aq7m1eqHNmu+4bxubdi/4GHWsbF23Ykn283xP72wH3rkCZwOwbaD9mu4qnouAQ91poIeBS5Oc3r2Be2nXJkkaoTmfOpN8lf5R+hlJ9tK/CucLwANJrgO+B3y66/4QcAUwDbwJXAtQVQeS3Ao82fW7paqOfHNYkrTE5gz9qrr6KKsumaVvAdcfZTtbgC3HNTtJ0qLyE7mS1BBDX5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWrIgkI/yStJdid5Osmuru2DSXYkebH7eXrXniS3J5lO8kyS8xejAEnS8BbjSP/iqjq3qia7+zcBj1TVWuCR7j7A5cDa7rYRuGsRHluSdByW4vTOeuCebvke4MqB9nurbydwWpKVS/D4kqSjSFXNf3DyMnAQKOC3q2pzkjeq6rRufYCDVXVakgeBL1TVt7t1jwC/VlW7jtjmRvqvBJiYmPgH999//7znNzMzw6mnnjrv8eOotZpbqxfarPn1A4d47X8t9yxG66wPnDDv/XzxxRc/NXD25a85cUGzgn9YVfuS/CSwI8kfD66sqkpyXM8qVbUZ2AwwOTlZU1NT855cr9djIePHUWs1t1YvtFnzHfdtY9PuhcbVeNm6bsWS7OcFnd6pqn3dz9eBbwAXAK+9c9qm+/l6130fsHpg+JldmyRpROYd+klWJHn/O8vApcCzwHZgQ9dtA7CtW94OXNNdxXMRcKiq9s975pKk47aQ10sTwDf6p+05EfjdqvrPSZ4EHkhyHfA94NNd/4eAK4Bp4E3g2gU8tiRpHuYd+lX1EvBTs7T/ALhklvYCrp/v40mSFs5P5EpSQwx9SWrIu/oaqN37DvGZm7653NMYqa3rViz3FCT9GPNIX5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhoy8tBPsi7JC0mmk9w06seXpJaNNPSTnADcCVwOnA1cneTsUc5Bklo26iP9C4Dpqnqpqv43cD+wfsRzkKRmjfo/Rl8FvDpwfy9w4WCHJBuBjd3dmSQvLODxzgD+bAHjx87FX2yu5tbqBWtuwgL/Lf+do60YdejPqao2A5sXY1tJdlXV5GJsa1y0VnNr9YI1t2Kpah716Z19wOqB+2d2bZKkERh16D8JrE1yVpKTgauA7SOegyQ1a6Snd6rqcJIbgIeBE4AtVbVnCR9yUU4TjZnWam6tXrDmVixJzamqpdiuJOnHkJ/IlaSGGPqS1JCxD/25vtYhyXuTfK1b/3iSNcswzUU1RM3/OslzSZ5J8kiSo16zOy6G/fqOJP80SSUZ+8v7hqk5yae7fb0nye+Oeo6LbYi/7b+d5NEk3+n+vq9YjnkuliRbkrye5NmjrE+S27vfxzNJzl/wg1bV2N7ovxn8J8DfBU4GvgucfUSffwH8Vrd8FfC15Z73CGq+GPgb3fJnW6i56/d+4DFgJzC53PMewX5eC3wHOL27/5PLPe8R1LwZ+Gy3fDbwynLPe4E1/yPgfODZo6y/AvhDIMBFwOMLfcxxP9If5msd1gP3dMtfBy5JkhHOcbHNWXNVPVpVb3Z3d9L/PMQ4G/brO24Fvgj85Sgnt0SGqfmfAXdW1UGAqnp9xHNcbMPUXMDf7JY/APyPEc5v0VXVY8CBY3RZD9xbfTuB05KsXMhjjnvoz/a1DquO1qeqDgOHgJ8YyeyWxjA1D7qO/pHCOJuz5u5l7+qq+uYoJ7aEhtnPHwE+kuS/JtmZZN3IZrc0hqn588DPJ9kLPAT8y9FMbdkc77/3Of3YfQ2DFk+SnwcmgX+83HNZSkneA3wJ+MwyT2XUTqR/imeK/qu5x5KcU1VvLOekltjVwNaq2pTkp4HfSfKxqvq/yz2xcTHuR/rDfK3Dj/okOZH+S8IfjGR2S2Oor7JI8k+AfwN8qqreGtHclspcNb8f+BjQS/IK/XOf28f8zdxh9vNeYHtV/Z+qehn47/SfBMbVMDVfBzwAUFX/DXgf/S9je7da9K+uGffQH+ZrHbYDG7rlnwW+Vd07JGNqzpqTnAf8Nv3AH/fzvDBHzVV1qKrOqKo1VbWG/vsYn6qqXcsz3UUxzN/2H9A/yifJGfRP97w0wjkutmFq/lPgEoAkf59+6P/Pkc5ytLYD13RX8VwEHKqq/QvZ4Fif3qmjfK1DkluAXVW1Hbib/kvAafpvmFy1fDNeuCFr/vfAqcB/6t6z/tOq+tSyTXqBhqz5XWXImh8GLk3yHPA28KtVNbavYoes+UbgK0n+Ff03dT8zzgdxSb5K/4n7jO59ipuBkwCq6rfov29xBTANvAlcu+DHHOPflyTpOI376R1J0nEw9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JD/h+idbdGrcn//QAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "X_test[\"cp_type\"].hist(bins=5)"
   ]
  }
 ],
 "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}