28 - Prétraitement de données (Corrigé).ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "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": "3.7.7"
    },
    "colab": {
      "name": "Untitled5-Copy5.ipynb",
      "provenance": [],
      "include_colab_link": true
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/MachineLearnia/Python-Machine-Learning/blob/master/28%20-%20Pr%C3%A9traitement%20de%20donn%C3%A9es%20(Corrig%C3%A9).ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EOS91mievJyr",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "outputId": "ea146814-de14-4270-d37c-b6fbe54471bc"
      },
      "source": [
        "%%javascript\n",
        "IPython.OutputArea.auto_scroll_threshold = 9999"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/javascript": [
              "IPython.OutputArea.auto_scroll_threshold = 9999"
            ],
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Vfl6Jxw6vJyu",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "f0e00250-5cd1-41c6-f30e-bb47335454d7"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
            "  import pandas.util.testing as tm\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7ReZHhe0vJyw",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 386
        },
        "outputId": "96e49d8e-cedb-43de-9bae-678f6fc6f24c"
      },
      "source": [
        "url = 'https://raw.githubusercontent.com/MachineLearnia/Python-Machine-Learning/master/Dataset/dataset.csv'\n",
        "data = pd.read_csv(url, index_col=0, encoding = \"ISO-8859-1\")\n",
        "data.head()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Patient age quantile</th>\n",
              "      <th>SARS-Cov-2 exam result</th>\n",
              "      <th>Patient addmited to regular ward (1=yes, 0=no)</th>\n",
              "      <th>Patient addmited to semi-intensive unit (1=yes, 0=no)</th>\n",
              "      <th>Patient addmited to intensive care unit (1=yes, 0=no)</th>\n",
              "      <th>Hematocrit</th>\n",
              "      <th>Hemoglobin</th>\n",
              "      <th>Platelets</th>\n",
              "      <th>Mean platelet volume</th>\n",
              "      <th>Red blood Cells</th>\n",
              "      <th>Lymphocytes</th>\n",
              "      <th>Mean corpuscular hemoglobin concentration (MCHC)</th>\n",
              "      <th>Leukocytes</th>\n",
              "      <th>Basophils</th>\n",
              "      <th>Mean corpuscular hemoglobin (MCH)</th>\n",
              "      <th>Eosinophils</th>\n",
              "      <th>Mean corpuscular volume (MCV)</th>\n",
              "      <th>Monocytes</th>\n",
              "      <th>Red blood cell distribution width (RDW)</th>\n",
              "      <th>Serum Glucose</th>\n",
              "      <th>Respiratory Syncytial Virus</th>\n",
              "      <th>Influenza A</th>\n",
              "      <th>Influenza B</th>\n",
              "      <th>Parainfluenza 1</th>\n",
              "      <th>CoronavirusNL63</th>\n",
              "      <th>Rhinovirus/Enterovirus</th>\n",
              "      <th>Mycoplasma pneumoniae</th>\n",
              "      <th>Coronavirus HKU1</th>\n",
              "      <th>Parainfluenza 3</th>\n",
              "      <th>Chlamydophila pneumoniae</th>\n",
              "      <th>Adenovirus</th>\n",
              "      <th>Parainfluenza 4</th>\n",
              "      <th>Coronavirus229E</th>\n",
              "      <th>CoronavirusOC43</th>\n",
              "      <th>Inf A H1N1 2009</th>\n",
              "      <th>Bordetella pertussis</th>\n",
              "      <th>Metapneumovirus</th>\n",
              "      <th>Parainfluenza 2</th>\n",
              "      <th>Neutrophils</th>\n",
              "      <th>Urea</th>\n",
              "      <th>...</th>\n",
              "      <th>Urine - Esterase</th>\n",
              "      <th>Urine - Aspect</th>\n",
              "      <th>Urine - pH</th>\n",
              "      <th>Urine - Hemoglobin</th>\n",
              "      <th>Urine - Bile pigments</th>\n",
              "      <th>Urine - Ketone Bodies</th>\n",
              "      <th>Urine - Nitrite</th>\n",
              "      <th>Urine - Density</th>\n",
              "      <th>Urine - Urobilinogen</th>\n",
              "      <th>Urine - Protein</th>\n",
              "      <th>Urine - Sugar</th>\n",
              "      <th>Urine - Leukocytes</th>\n",
              "      <th>Urine - Crystals</th>\n",
              "      <th>Urine - Red blood cells</th>\n",
              "      <th>Urine - Hyaline cylinders</th>\n",
              "      <th>Urine - Granular cylinders</th>\n",
              "      <th>Urine - Yeasts</th>\n",
              "      <th>Urine - Color</th>\n",
              "      <th>Partial thromboplastin time (PTT)</th>\n",
              "      <th>Relationship (Patient/Normal)</th>\n",
              "      <th>International normalized ratio (INR)</th>\n",
              "      <th>Lactic Dehydrogenase</th>\n",
              "      <th>Prothrombin time (PT), Activity</th>\n",
              "      <th>Vitamin B12</th>\n",
              "      <th>Creatine phosphokinase (CPK)</th>\n",
              "      <th>Ferritin</th>\n",
              "      <th>Arterial Lactic Acid</th>\n",
              "      <th>Lipase dosage</th>\n",
              "      <th>D-Dimer</th>\n",
              "      <th>Albumin</th>\n",
              "      <th>Hb saturation (arterial blood gases)</th>\n",
              "      <th>pCO2 (arterial blood gas analysis)</th>\n",
              "      <th>Base excess (arterial blood gas analysis)</th>\n",
              "      <th>pH (arterial blood gas analysis)</th>\n",
              "      <th>Total CO2 (arterial blood gas analysis)</th>\n",
              "      <th>HCO3 (arterial blood gas analysis)</th>\n",
              "      <th>pO2 (arterial blood gas analysis)</th>\n",
              "      <th>Arteiral Fio2</th>\n",
              "      <th>Phosphor</th>\n",
              "      <th>ctO2 (arterial blood gas analysis)</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Patient ID</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>44477f75e8169d2</th>\n",
              "      <td>13</td>\n",
              "      <td>negative</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>126e9dd13932f68</th>\n",
              "      <td>17</td>\n",
              "      <td>negative</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.236515</td>\n",
              "      <td>-0.02234</td>\n",
              "      <td>-0.517413</td>\n",
              "      <td>0.010677</td>\n",
              "      <td>0.102004</td>\n",
              "      <td>0.318366</td>\n",
              "      <td>-0.95079</td>\n",
              "      <td>-0.09461</td>\n",
              "      <td>-0.223767</td>\n",
              "      <td>-0.292269</td>\n",
              "      <td>1.482158</td>\n",
              "      <td>0.166192</td>\n",
              "      <td>0.357547</td>\n",
              "      <td>-0.625073</td>\n",
              "      <td>-0.140648</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>detected</td>\n",
              "      <td>NaN</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>-0.619086</td>\n",
              "      <td>1.198059</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>a46b4402a0e5696</th>\n",
              "      <td>8</td>\n",
              "      <td>negative</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>f7d619a94f97c45</th>\n",
              "      <td>5</td>\n",
              "      <td>negative</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>d9e41465789c2b5</th>\n",
              "      <td>15</td>\n",
              "      <td>negative</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>detected</td>\n",
              "      <td>NaN</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 110 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                 Patient age quantile  ... ctO2 (arterial blood gas analysis)\n",
              "Patient ID                             ...                                   \n",
              "44477f75e8169d2                    13  ...                                NaN\n",
              "126e9dd13932f68                    17  ...                                NaN\n",
              "a46b4402a0e5696                     8  ...                                NaN\n",
              "f7d619a94f97c45                     5  ...                                NaN\n",
              "d9e41465789c2b5                    15  ...                                NaN\n",
              "\n",
              "[5 rows x 110 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_AHnVIi5vJyy",
        "colab_type": "text"
      },
      "source": [
        "# PRE-PROCESSING"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nmdMdxp5vJyz",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 386
        },
        "outputId": "8a4b85ee-d4d4-4502-e616-8d1d2c2aeff5"
      },
      "source": [
        "df = data.copy()\n",
        "df.head()"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Patient age quantile</th>\n",
              "      <th>SARS-Cov-2 exam result</th>\n",
              "      <th>Patient addmited to regular ward (1=yes, 0=no)</th>\n",
              "      <th>Patient addmited to semi-intensive unit (1=yes, 0=no)</th>\n",
              "      <th>Patient addmited to intensive care unit (1=yes, 0=no)</th>\n",
              "      <th>Hematocrit</th>\n",
              "      <th>Hemoglobin</th>\n",
              "      <th>Platelets</th>\n",
              "      <th>Mean platelet volume</th>\n",
              "      <th>Red blood Cells</th>\n",
              "      <th>Lymphocytes</th>\n",
              "      <th>Mean corpuscular hemoglobin concentration (MCHC)</th>\n",
              "      <th>Leukocytes</th>\n",
              "      <th>Basophils</th>\n",
              "      <th>Mean corpuscular hemoglobin (MCH)</th>\n",
              "      <th>Eosinophils</th>\n",
              "      <th>Mean corpuscular volume (MCV)</th>\n",
              "      <th>Monocytes</th>\n",
              "      <th>Red blood cell distribution width (RDW)</th>\n",
              "      <th>Serum Glucose</th>\n",
              "      <th>Respiratory Syncytial Virus</th>\n",
              "      <th>Influenza A</th>\n",
              "      <th>Influenza B</th>\n",
              "      <th>Parainfluenza 1</th>\n",
              "      <th>CoronavirusNL63</th>\n",
              "      <th>Rhinovirus/Enterovirus</th>\n",
              "      <th>Mycoplasma pneumoniae</th>\n",
              "      <th>Coronavirus HKU1</th>\n",
              "      <th>Parainfluenza 3</th>\n",
              "      <th>Chlamydophila pneumoniae</th>\n",
              "      <th>Adenovirus</th>\n",
              "      <th>Parainfluenza 4</th>\n",
              "      <th>Coronavirus229E</th>\n",
              "      <th>CoronavirusOC43</th>\n",
              "      <th>Inf A H1N1 2009</th>\n",
              "      <th>Bordetella pertussis</th>\n",
              "      <th>Metapneumovirus</th>\n",
              "      <th>Parainfluenza 2</th>\n",
              "      <th>Neutrophils</th>\n",
              "      <th>Urea</th>\n",
              "      <th>...</th>\n",
              "      <th>Urine - Esterase</th>\n",
              "      <th>Urine - Aspect</th>\n",
              "      <th>Urine - pH</th>\n",
              "      <th>Urine - Hemoglobin</th>\n",
              "      <th>Urine - Bile pigments</th>\n",
              "      <th>Urine - Ketone Bodies</th>\n",
              "      <th>Urine - Nitrite</th>\n",
              "      <th>Urine - Density</th>\n",
              "      <th>Urine - Urobilinogen</th>\n",
              "      <th>Urine - Protein</th>\n",
              "      <th>Urine - Sugar</th>\n",
              "      <th>Urine - Leukocytes</th>\n",
              "      <th>Urine - Crystals</th>\n",
              "      <th>Urine - Red blood cells</th>\n",
              "      <th>Urine - Hyaline cylinders</th>\n",
              "      <th>Urine - Granular cylinders</th>\n",
              "      <th>Urine - Yeasts</th>\n",
              "      <th>Urine - Color</th>\n",
              "      <th>Partial thromboplastin time (PTT)</th>\n",
              "      <th>Relationship (Patient/Normal)</th>\n",
              "      <th>International normalized ratio (INR)</th>\n",
              "      <th>Lactic Dehydrogenase</th>\n",
              "      <th>Prothrombin time (PT), Activity</th>\n",
              "      <th>Vitamin B12</th>\n",
              "      <th>Creatine phosphokinase (CPK)</th>\n",
              "      <th>Ferritin</th>\n",
              "      <th>Arterial Lactic Acid</th>\n",
              "      <th>Lipase dosage</th>\n",
              "      <th>D-Dimer</th>\n",
              "      <th>Albumin</th>\n",
              "      <th>Hb saturation (arterial blood gases)</th>\n",
              "      <th>pCO2 (arterial blood gas analysis)</th>\n",
              "      <th>Base excess (arterial blood gas analysis)</th>\n",
              "      <th>pH (arterial blood gas analysis)</th>\n",
              "      <th>Total CO2 (arterial blood gas analysis)</th>\n",
              "      <th>HCO3 (arterial blood gas analysis)</th>\n",
              "      <th>pO2 (arterial blood gas analysis)</th>\n",
              "      <th>Arteiral Fio2</th>\n",
              "      <th>Phosphor</th>\n",
              "      <th>ctO2 (arterial blood gas analysis)</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Patient ID</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>44477f75e8169d2</th>\n",
              "      <td>13</td>\n",
              "      <td>negative</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>126e9dd13932f68</th>\n",
              "      <td>17</td>\n",
              "      <td>negative</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.236515</td>\n",
              "      <td>-0.02234</td>\n",
              "      <td>-0.517413</td>\n",
              "      <td>0.010677</td>\n",
              "      <td>0.102004</td>\n",
              "      <td>0.318366</td>\n",
              "      <td>-0.95079</td>\n",
              "      <td>-0.09461</td>\n",
              "      <td>-0.223767</td>\n",
              "      <td>-0.292269</td>\n",
              "      <td>1.482158</td>\n",
              "      <td>0.166192</td>\n",
              "      <td>0.357547</td>\n",
              "      <td>-0.625073</td>\n",
              "      <td>-0.140648</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>detected</td>\n",
              "      <td>NaN</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>-0.619086</td>\n",
              "      <td>1.198059</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>a46b4402a0e5696</th>\n",
              "      <td>8</td>\n",
              "      <td>negative</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>f7d619a94f97c45</th>\n",
              "      <td>5</td>\n",
              "      <td>negative</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>d9e41465789c2b5</th>\n",
              "      <td>15</td>\n",
              "      <td>negative</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>detected</td>\n",
              "      <td>NaN</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>...</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 110 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                 Patient age quantile  ... ctO2 (arterial blood gas analysis)\n",
              "Patient ID                             ...                                   \n",
              "44477f75e8169d2                    13  ...                                NaN\n",
              "126e9dd13932f68                    17  ...                                NaN\n",
              "a46b4402a0e5696                     8  ...                                NaN\n",
              "f7d619a94f97c45                     5  ...                                NaN\n",
              "d9e41465789c2b5                    15  ...                                NaN\n",
              "\n",
              "[5 rows x 110 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Utxp9WEcvJy1",
        "colab_type": "text"
      },
      "source": [
        "## Création des sous-ensembles (suite au EDA)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ziegnMu3vJy1",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "missing_rate = df.isna().sum()/df.shape[0]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tJsNA3EIvJy3",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "blood_columns = list(df.columns[(missing_rate < 0.9) & (missing_rate >0.88)])\n",
        "viral_columns = list(df.columns[(missing_rate < 0.80) & (missing_rate > 0.75)])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kyF1e3FHvJy5",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "key_columns = ['Patient age quantile', 'SARS-Cov-2 exam result']"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6weBpPP3vJy7",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 306
        },
        "outputId": "b2dca5f6-5644-4a08-bef2-09683e392a13"
      },
      "source": [
        "df = df[key_columns + blood_columns + viral_columns]\n",
        "df.head()"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Patient age quantile</th>\n",
              "      <th>SARS-Cov-2 exam result</th>\n",
              "      <th>Hematocrit</th>\n",
              "      <th>Hemoglobin</th>\n",
              "      <th>Platelets</th>\n",
              "      <th>Mean platelet volume</th>\n",
              "      <th>Red blood Cells</th>\n",
              "      <th>Lymphocytes</th>\n",
              "      <th>Mean corpuscular hemoglobin concentration (MCHC)</th>\n",
              "      <th>Leukocytes</th>\n",
              "      <th>Basophils</th>\n",
              "      <th>Mean corpuscular hemoglobin (MCH)</th>\n",
              "      <th>Eosinophils</th>\n",
              "      <th>Mean corpuscular volume (MCV)</th>\n",
              "      <th>Monocytes</th>\n",
              "      <th>Red blood cell distribution width (RDW)</th>\n",
              "      <th>Respiratory Syncytial Virus</th>\n",
              "      <th>Influenza A</th>\n",
              "      <th>Influenza B</th>\n",
              "      <th>Parainfluenza 1</th>\n",
              "      <th>CoronavirusNL63</th>\n",
              "      <th>Rhinovirus/Enterovirus</th>\n",
              "      <th>Coronavirus HKU1</th>\n",
              "      <th>Parainfluenza 3</th>\n",
              "      <th>Chlamydophila pneumoniae</th>\n",
              "      <th>Adenovirus</th>\n",
              "      <th>Parainfluenza 4</th>\n",
              "      <th>Coronavirus229E</th>\n",
              "      <th>CoronavirusOC43</th>\n",
              "      <th>Inf A H1N1 2009</th>\n",
              "      <th>Bordetella pertussis</th>\n",
              "      <th>Metapneumovirus</th>\n",
              "      <th>Parainfluenza 2</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Patient ID</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>44477f75e8169d2</th>\n",
              "      <td>13</td>\n",
              "      <td>negative</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>126e9dd13932f68</th>\n",
              "      <td>17</td>\n",
              "      <td>negative</td>\n",
              "      <td>0.236515</td>\n",
              "      <td>-0.02234</td>\n",
              "      <td>-0.517413</td>\n",
              "      <td>0.010677</td>\n",
              "      <td>0.102004</td>\n",
              "      <td>0.318366</td>\n",
              "      <td>-0.95079</td>\n",
              "      <td>-0.09461</td>\n",
              "      <td>-0.223767</td>\n",
              "      <td>-0.292269</td>\n",
              "      <td>1.482158</td>\n",
              "      <td>0.166192</td>\n",
              "      <td>0.357547</td>\n",
              "      <td>-0.625073</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>a46b4402a0e5696</th>\n",
              "      <td>8</td>\n",
              "      <td>negative</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>f7d619a94f97c45</th>\n",
              "      <td>5</td>\n",
              "      <td>negative</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>d9e41465789c2b5</th>\n",
              "      <td>15</td>\n",
              "      <td>negative</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "      <td>not_detected</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                 Patient age quantile  ... Parainfluenza 2\n",
              "Patient ID                             ...                \n",
              "44477f75e8169d2                    13  ...             NaN\n",
              "126e9dd13932f68                    17  ...    not_detected\n",
              "a46b4402a0e5696                     8  ...             NaN\n",
              "f7d619a94f97c45                     5  ...             NaN\n",
              "d9e41465789c2b5                    15  ...    not_detected\n",
              "\n",
              "[5 rows x 33 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kjDxmab7vJy9",
        "colab_type": "text"
      },
      "source": [
        "## TrainTest - Nettoyage - Encodage"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FGoTxrGhvJy9",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.model_selection import train_test_split"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dNOA3tuNvJy_",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "trainset, testset = train_test_split(df, test_size=0.2, random_state=0)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CRoPWXQovJzB",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "843f4ca3-37fb-4447-aca5-a988ad2c4a7b"
      },
      "source": [
        "trainset['SARS-Cov-2 exam result'].value_counts()"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "negative    4068\n",
              "positive     447\n",
              "Name: SARS-Cov-2 exam result, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "85m56b32vJzD",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "08751b73-6286-4175-f9c9-6e7f9351fe3f"
      },
      "source": [
        "testset['SARS-Cov-2 exam result'].value_counts()"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "negative    1018\n",
              "positive     111\n",
              "Name: SARS-Cov-2 exam result, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3jgKiQcyvJzF",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def encodage(df):\n",
        "    code = {'negative':0,\n",
        "            'positive':1,\n",
        "            'not_detected':0,\n",
        "            'detected':1}\n",
        "    \n",
        "    for col in df.select_dtypes('object').columns:\n",
        "        df.loc[:,col] = df[col].map(code)\n",
        "        \n",
        "    return df"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RKdIkxtHvJzH",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def feature_engineering(df):\n",
        "    df['est malade'] = df[viral_columns].sum(axis=1) >= 1\n",
        "    df = df.drop(viral_columns, axis=1)\n",
        "    return df"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qcp9JvE5vJzJ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def imputation(df):\n",
        "    #df['is na'] = (df['Parainfluenza 3'].isna()) | (df['Leukocytes'].isna())\n",
        "    #df = df.fillna(-999)\n",
        "    df = df.dropna(axis=0)\n",
        "    return  df"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AKYCi3NnvJzK",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def preprocessing(df):\n",
        "    \n",
        "    df = encodage(df)\n",
        "    df = feature_engineering(df)\n",
        "    df = imputation(df)\n",
        "    \n",
        "    X = df.drop('SARS-Cov-2 exam result', axis=1)\n",
        "    y = df['SARS-Cov-2 exam result']\n",
        "    \n",
        "    print(y.value_counts())\n",
        "    \n",
        "    return X, y"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3KBs-ujBvJzM",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "outputId": "2bcacce4-762f-437e-c4ed-51fc3b0f8462"
      },
      "source": [
        "X_train, y_train = preprocessing(trainset)"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0    422\n",
            "1     65\n",
            "Name: SARS-Cov-2 exam result, dtype: int64\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/pandas/core/indexing.py:966: 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",
            "  self.obj[item] = s\n",
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:2: 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",
            "  \n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DVf4oBdnvJzP",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "outputId": "f96ad64d-7dfd-4c51-9f3e-bcc2fcdfd903"
      },
      "source": [
        "X_test, y_test = preprocessing(testset)"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/pandas/core/indexing.py:966: 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",
            "  self.obj[item] = s\n",
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:2: 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",
            "  \n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "0    95\n",
            "1    16\n",
            "Name: SARS-Cov-2 exam result, dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yLFEfMcmvJzQ",
        "colab_type": "text"
      },
      "source": [
        "## Modellisation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DaYCentPvJzR",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.pipeline import make_pipeline\n",
        "from sklearn.feature_selection import SelectKBest, f_classif\n",
        "from sklearn.preprocessing import PolynomialFeatures\n",
        "from sklearn.decomposition import PCA"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "A85d9oArvJzS",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model_1 = RandomForestClassifier(random_state=0)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "N-VDXhe2vJzU",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model_2 = make_pipeline(PolynomialFeatures(2), SelectKBest(f_classif, k=10),\n",
        "                      RandomForestClassifier(random_state=0))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_iwPzFFKvJzV",
        "colab_type": "text"
      },
      "source": [
        "## Procédure d'évaluation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "thqK6zyxvJzW",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.metrics import f1_score, confusion_matrix, classification_report\n",
        "from sklearn.model_selection import learning_curve"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ig2M_p3FvJzX",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def evaluation(model):\n",
        "    \n",
        "    model.fit(X_train, y_train)\n",
        "    ypred = model.predict(X_test)\n",
        "    \n",
        "    print(confusion_matrix(y_test, ypred))\n",
        "    print(classification_report(y_test, ypred))\n",
        "    \n",
        "    N, train_score, val_score = learning_curve(model, X_train, y_train,\n",
        "                                              cv=4, scoring='f1',\n",
        "                                               train_sizes=np.linspace(0.1, 1, 10))\n",
        "    \n",
        "    \n",
        "    plt.figure(figsize=(12, 8))\n",
        "    plt.plot(N, train_score.mean(axis=1), label='train score')\n",
        "    plt.plot(N, val_score.mean(axis=1), label='validation score')\n",
        "    plt.legend()\n",
        "    \n",
        "    "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "94u9GhSyvJzZ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 670
        },
        "outputId": "1387dd13-97b5-463a-a11c-f298acd2364a"
      },
      "source": [
        "evaluation(model_1)"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[92  3]\n",
            " [13  3]]\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.88      0.97      0.92        95\n",
            "           1       0.50      0.19      0.27        16\n",
            "\n",
            "    accuracy                           0.86       111\n",
            "   macro avg       0.69      0.58      0.60       111\n",
            "weighted avg       0.82      0.86      0.83       111\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VdNHtoRSvJza",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 754
        },
        "outputId": "33960ced-edb0-4109-8e26-5fc9293f7b53"
      },
      "source": [
        "pd.DataFrame(model_1.feature_importances_, index=X_train.columns).plot.bar(figsize=(12, 8))"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f0b668fbcc0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}