Aula05_Desafios_Victória_F.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Aula05_Desafios_Victória_F.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "hg2Lz3uoYUPx",
        "X_Lg2XTu20ND"
      ],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/victoriafabris/ImersaoDados/blob/main/Aula05_Desafios_Vict%C3%B3ria_F.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hg2Lz3uoYUPx"
      },
      "source": [
        "# **Aula 5**\n",
        "\n",
        "Nesta aula veremos sobre Machine Learning e a biblioteca scikit-learn."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JzKTV39Rhw_a"
      },
      "source": [
        "# **Importando biblioteca e dados**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CEHxcFL_YTuv"
      },
      "source": [
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "#banco de dados com os dados dos ensaios/experimentos\n",
        "url_dados = 'https://github.com/alura-cursos/imersaodados3/blob/main/dados/dados_experimentos.zip?raw=true'\n",
        "\n",
        "#banco de dados com os dados dos resultados dos ensaios/experimentos\n",
        "url_dados_resultados = 'https://github.com/alura-cursos/imersaodados3/blob/main/dados/dados_resultados.csv?raw=true'\n",
        "\n",
        "dados = pd.read_csv(url_dados, compression = 'zip')\n",
        "dados_resultados = pd.read_csv(url_dados_resultados)"
      ],
      "execution_count": 63,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ah5WcaJXiL6c"
      },
      "source": [
        "Primeiramente, vamos renomear a coluna 'droga' pelo nome 'composto', como já feito nas aulas anteriores:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rZwPw5WWdbl1"
      },
      "source": [
        "mapa = {'droga': 'composto'}\n",
        "dados.rename(columns=mapa, inplace=True)"
      ],
      "execution_count": 64,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wBa8ABGrcc1t"
      },
      "source": [
        "contagem_moa = dados_resultados.drop('id', axis=1).sum().sort_values(ascending=False)"
      ],
      "execution_count": 65,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HS0ZHSeTjFrr"
      },
      "source": [
        "dados_resultados['n_moa'] = dados_resultados.drop('id', axis=1).sum(axis=1)"
      ],
      "execution_count": 66,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MNMsak-ih9PQ",
        "outputId": "2efea06c-a626-4e67-cbbe-10248c900238"
      },
      "source": [
        "dados_resultados['n_moa'] != 0"
      ],
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0         True\n",
              "1        False\n",
              "2         True\n",
              "3        False\n",
              "4         True\n",
              "         ...  \n",
              "23809     True\n",
              "23810     True\n",
              "23811    False\n",
              "23812     True\n",
              "23813    False\n",
              "Name: n_moa, Length: 23814, dtype: bool"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 67
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "id": "bba-i1D8hcu8",
        "outputId": "a27eb044-af56-4ad5-fd5b-55bf162b19e3"
      },
      "source": [
        "dados_resultados['ativo_moa'] = (dados_resultados['n_moa'] != 0)\n",
        "dados_resultados.head()"
      ],
      "execution_count": 68,
      "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>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>adrenergic_receptor_agonist</th>\n",
              "      <th>adrenergic_receptor_antagonist</th>\n",
              "      <th>akt_inhibitor</th>\n",
              "      <th>aldehyde_dehydrogenase_inhibitor</th>\n",
              "      <th>alk_inhibitor</th>\n",
              "      <th>ampk_activator</th>\n",
              "      <th>analgesic</th>\n",
              "      <th>androgen_receptor_agonist</th>\n",
              "      <th>androgen_receptor_antagonist</th>\n",
              "      <th>anesthetic_-_local</th>\n",
              "      <th>angiogenesis_inhibitor</th>\n",
              "      <th>angiotensin_receptor_antagonist</th>\n",
              "      <th>anti-inflammatory</th>\n",
              "      <th>antiarrhythmic</th>\n",
              "      <th>antibiotic</th>\n",
              "      <th>anticonvulsant</th>\n",
              "      <th>antifungal</th>\n",
              "      <th>antihistamine</th>\n",
              "      <th>antimalarial</th>\n",
              "      <th>antioxidant</th>\n",
              "      <th>antiprotozoal</th>\n",
              "      <th>antiviral</th>\n",
              "      <th>apoptosis_stimulant</th>\n",
              "      <th>aromatase_inhibitor</th>\n",
              "      <th>atm_kinase_inhibitor</th>\n",
              "      <th>atp-sensitive_potassium_channel_antagonist</th>\n",
              "      <th>atp_synthase_inhibitor</th>\n",
              "      <th>atpase_inhibitor</th>\n",
              "      <th>atr_kinase_inhibitor</th>\n",
              "      <th>aurora_kinase_inhibitor</th>\n",
              "      <th>...</th>\n",
              "      <th>radiopaque_medium</th>\n",
              "      <th>raf_inhibitor</th>\n",
              "      <th>ras_gtpase_inhibitor</th>\n",
              "      <th>retinoid_receptor_agonist</th>\n",
              "      <th>retinoid_receptor_antagonist</th>\n",
              "      <th>rho_associated_kinase_inhibitor</th>\n",
              "      <th>ribonucleoside_reductase_inhibitor</th>\n",
              "      <th>rna_polymerase_inhibitor</th>\n",
              "      <th>serotonin_receptor_agonist</th>\n",
              "      <th>serotonin_receptor_antagonist</th>\n",
              "      <th>serotonin_reuptake_inhibitor</th>\n",
              "      <th>sigma_receptor_agonist</th>\n",
              "      <th>sigma_receptor_antagonist</th>\n",
              "      <th>smoothened_receptor_antagonist</th>\n",
              "      <th>sodium_channel_inhibitor</th>\n",
              "      <th>sphingosine_receptor_agonist</th>\n",
              "      <th>src_inhibitor</th>\n",
              "      <th>steroid</th>\n",
              "      <th>syk_inhibitor</th>\n",
              "      <th>tachykinin_antagonist</th>\n",
              "      <th>tgf-beta_receptor_inhibitor</th>\n",
              "      <th>thrombin_inhibitor</th>\n",
              "      <th>thymidylate_synthase_inhibitor</th>\n",
              "      <th>tlr_agonist</th>\n",
              "      <th>tlr_antagonist</th>\n",
              "      <th>tnf_inhibitor</th>\n",
              "      <th>topoisomerase_inhibitor</th>\n",
              "      <th>transient_receptor_potential_channel_antagonist</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",
              "      <th>n_moa</th>\n",
              "      <th>ativo_moa</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>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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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>1</td>\n",
              "      <td>True</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>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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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>False</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>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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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>3</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>id_0015fd391</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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>id_001626bd3</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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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>1</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 209 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "             id  5-alpha_reductase_inhibitor  ...  n_moa  ativo_moa\n",
              "0  id_000644bb2                            0  ...      1       True\n",
              "1  id_000779bfc                            0  ...      0      False\n",
              "2  id_000a6266a                            0  ...      3       True\n",
              "3  id_0015fd391                            0  ...      0      False\n",
              "4  id_001626bd3                            0  ...      1       True\n",
              "\n",
              "[5 rows x 209 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 68
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 251
        },
        "id": "0r4V83bujiao",
        "outputId": "41096654-5386-448b-a509-75f5d5cc6a18"
      },
      "source": [
        " dados_combinados = pd.merge(dados, dados_resultados[['id','n_moa', 'ativo_moa']], on='id')\n",
        " dados_combinados.head()"
      ],
      "execution_count": 69,
      "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>id</th>\n",
              "      <th>tratamento</th>\n",
              "      <th>tempo</th>\n",
              "      <th>dose</th>\n",
              "      <th>composto</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>g-7</th>\n",
              "      <th>g-8</th>\n",
              "      <th>g-9</th>\n",
              "      <th>g-10</th>\n",
              "      <th>g-11</th>\n",
              "      <th>g-12</th>\n",
              "      <th>g-13</th>\n",
              "      <th>g-14</th>\n",
              "      <th>g-15</th>\n",
              "      <th>g-16</th>\n",
              "      <th>g-17</th>\n",
              "      <th>g-18</th>\n",
              "      <th>g-19</th>\n",
              "      <th>g-20</th>\n",
              "      <th>g-21</th>\n",
              "      <th>g-22</th>\n",
              "      <th>g-23</th>\n",
              "      <th>g-24</th>\n",
              "      <th>g-25</th>\n",
              "      <th>g-26</th>\n",
              "      <th>g-27</th>\n",
              "      <th>g-28</th>\n",
              "      <th>g-29</th>\n",
              "      <th>g-30</th>\n",
              "      <th>g-31</th>\n",
              "      <th>g-32</th>\n",
              "      <th>g-33</th>\n",
              "      <th>g-34</th>\n",
              "      <th>...</th>\n",
              "      <th>c-62</th>\n",
              "      <th>c-63</th>\n",
              "      <th>c-64</th>\n",
              "      <th>c-65</th>\n",
              "      <th>c-66</th>\n",
              "      <th>c-67</th>\n",
              "      <th>c-68</th>\n",
              "      <th>c-69</th>\n",
              "      <th>c-70</th>\n",
              "      <th>c-71</th>\n",
              "      <th>c-72</th>\n",
              "      <th>c-73</th>\n",
              "      <th>c-74</th>\n",
              "      <th>c-75</th>\n",
              "      <th>c-76</th>\n",
              "      <th>c-77</th>\n",
              "      <th>c-78</th>\n",
              "      <th>c-79</th>\n",
              "      <th>c-80</th>\n",
              "      <th>c-81</th>\n",
              "      <th>c-82</th>\n",
              "      <th>c-83</th>\n",
              "      <th>c-84</th>\n",
              "      <th>c-85</th>\n",
              "      <th>c-86</th>\n",
              "      <th>c-87</th>\n",
              "      <th>c-88</th>\n",
              "      <th>c-89</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",
              "      <th>n_moa</th>\n",
              "      <th>ativo_moa</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>id_000644bb2</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>24</td>\n",
              "      <td>D1</td>\n",
              "      <td>b68db1d53</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>-0.0326</td>\n",
              "      <td>0.5548</td>\n",
              "      <td>-0.0921</td>\n",
              "      <td>1.1830</td>\n",
              "      <td>0.1530</td>\n",
              "      <td>0.5574</td>\n",
              "      <td>-0.4015</td>\n",
              "      <td>0.1789</td>\n",
              "      <td>-0.6528</td>\n",
              "      <td>-0.7969</td>\n",
              "      <td>0.6342</td>\n",
              "      <td>0.1778</td>\n",
              "      <td>-0.3694</td>\n",
              "      <td>-0.5688</td>\n",
              "      <td>-1.1360</td>\n",
              "      <td>-1.1880</td>\n",
              "      <td>0.6940</td>\n",
              "      <td>0.4393</td>\n",
              "      <td>0.2664</td>\n",
              "      <td>0.1907</td>\n",
              "      <td>0.1628</td>\n",
              "      <td>-0.2853</td>\n",
              "      <td>0.5819</td>\n",
              "      <td>0.2934</td>\n",
              "      <td>-0.5584</td>\n",
              "      <td>-0.0916</td>\n",
              "      <td>-0.3010</td>\n",
              "      <td>-0.1537</td>\n",
              "      <td>...</td>\n",
              "      <td>0.3680</td>\n",
              "      <td>0.8427</td>\n",
              "      <td>0.1042</td>\n",
              "      <td>0.1403</td>\n",
              "      <td>0.1758</td>\n",
              "      <td>1.2570</td>\n",
              "      <td>-0.5979</td>\n",
              "      <td>1.2250</td>\n",
              "      <td>-0.0553</td>\n",
              "      <td>0.7351</td>\n",
              "      <td>0.5810</td>\n",
              "      <td>0.9590</td>\n",
              "      <td>0.2427</td>\n",
              "      <td>0.0495</td>\n",
              "      <td>0.4141</td>\n",
              "      <td>0.8432</td>\n",
              "      <td>0.6162</td>\n",
              "      <td>-0.7318</td>\n",
              "      <td>1.2120</td>\n",
              "      <td>0.6362</td>\n",
              "      <td>-0.4427</td>\n",
              "      <td>0.1288</td>\n",
              "      <td>1.4840</td>\n",
              "      <td>0.1799</td>\n",
              "      <td>0.5367</td>\n",
              "      <td>-0.1111</td>\n",
              "      <td>-1.0120</td>\n",
              "      <td>0.6685</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",
              "      <td>1</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>id_000779bfc</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>72</td>\n",
              "      <td>D1</td>\n",
              "      <td>df89a8e5a</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>0.3372</td>\n",
              "      <td>-0.4047</td>\n",
              "      <td>0.8507</td>\n",
              "      <td>-1.1520</td>\n",
              "      <td>-0.4201</td>\n",
              "      <td>-0.0958</td>\n",
              "      <td>0.4590</td>\n",
              "      <td>0.0803</td>\n",
              "      <td>0.2250</td>\n",
              "      <td>0.5293</td>\n",
              "      <td>0.2839</td>\n",
              "      <td>-0.3494</td>\n",
              "      <td>0.2883</td>\n",
              "      <td>0.9449</td>\n",
              "      <td>-0.1646</td>\n",
              "      <td>-0.2657</td>\n",
              "      <td>-0.3372</td>\n",
              "      <td>0.3135</td>\n",
              "      <td>-0.4316</td>\n",
              "      <td>0.4773</td>\n",
              "      <td>0.2075</td>\n",
              "      <td>-0.4216</td>\n",
              "      <td>-0.1161</td>\n",
              "      <td>-0.0499</td>\n",
              "      <td>-0.2627</td>\n",
              "      <td>0.9959</td>\n",
              "      <td>-0.2483</td>\n",
              "      <td>0.2655</td>\n",
              "      <td>...</td>\n",
              "      <td>0.3905</td>\n",
              "      <td>0.7099</td>\n",
              "      <td>0.2912</td>\n",
              "      <td>0.4151</td>\n",
              "      <td>-0.2840</td>\n",
              "      <td>-0.3104</td>\n",
              "      <td>-0.6373</td>\n",
              "      <td>0.2887</td>\n",
              "      <td>-0.0765</td>\n",
              "      <td>0.2539</td>\n",
              "      <td>0.4443</td>\n",
              "      <td>0.5932</td>\n",
              "      <td>0.2031</td>\n",
              "      <td>0.7639</td>\n",
              "      <td>0.5499</td>\n",
              "      <td>-0.3322</td>\n",
              "      <td>-0.0977</td>\n",
              "      <td>0.4329</td>\n",
              "      <td>-0.2782</td>\n",
              "      <td>0.7827</td>\n",
              "      <td>0.5934</td>\n",
              "      <td>0.3402</td>\n",
              "      <td>0.1499</td>\n",
              "      <td>0.4420</td>\n",
              "      <td>0.9366</td>\n",
              "      <td>0.8193</td>\n",
              "      <td>-0.4236</td>\n",
              "      <td>0.3192</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",
              "      <td>0</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>id_000a6266a</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>48</td>\n",
              "      <td>D1</td>\n",
              "      <td>18bb41b2c</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>0.2155</td>\n",
              "      <td>0.0065</td>\n",
              "      <td>1.2300</td>\n",
              "      <td>-0.4797</td>\n",
              "      <td>-0.5631</td>\n",
              "      <td>-0.0366</td>\n",
              "      <td>-1.8300</td>\n",
              "      <td>0.6057</td>\n",
              "      <td>-0.3278</td>\n",
              "      <td>0.6042</td>\n",
              "      <td>-0.3075</td>\n",
              "      <td>-0.1147</td>\n",
              "      <td>-0.0570</td>\n",
              "      <td>-0.0799</td>\n",
              "      <td>-0.8181</td>\n",
              "      <td>-1.5320</td>\n",
              "      <td>0.2307</td>\n",
              "      <td>0.4901</td>\n",
              "      <td>0.4780</td>\n",
              "      <td>-1.3970</td>\n",
              "      <td>4.6240</td>\n",
              "      <td>-0.0437</td>\n",
              "      <td>1.2870</td>\n",
              "      <td>-1.8530</td>\n",
              "      <td>0.6069</td>\n",
              "      <td>0.4290</td>\n",
              "      <td>0.1783</td>\n",
              "      <td>0.0018</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.0444</td>\n",
              "      <td>0.1894</td>\n",
              "      <td>-0.0014</td>\n",
              "      <td>-2.3640</td>\n",
              "      <td>-0.4682</td>\n",
              "      <td>0.1210</td>\n",
              "      <td>-0.5177</td>\n",
              "      <td>-0.0604</td>\n",
              "      <td>0.1682</td>\n",
              "      <td>-0.4436</td>\n",
              "      <td>0.4963</td>\n",
              "      <td>0.1363</td>\n",
              "      <td>0.3335</td>\n",
              "      <td>0.9760</td>\n",
              "      <td>-0.0427</td>\n",
              "      <td>-0.1235</td>\n",
              "      <td>0.0959</td>\n",
              "      <td>0.0690</td>\n",
              "      <td>-0.9416</td>\n",
              "      <td>-0.7548</td>\n",
              "      <td>-0.1109</td>\n",
              "      <td>-0.6272</td>\n",
              "      <td>0.3019</td>\n",
              "      <td>0.1172</td>\n",
              "      <td>0.1093</td>\n",
              "      <td>-0.3113</td>\n",
              "      <td>0.3019</td>\n",
              "      <td>-0.0873</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",
              "      <td>3</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>id_0015fd391</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>48</td>\n",
              "      <td>D1</td>\n",
              "      <td>8c7f86626</td>\n",
              "      <td>-0.5138</td>\n",
              "      <td>-0.2491</td>\n",
              "      <td>-0.2656</td>\n",
              "      <td>0.5288</td>\n",
              "      <td>4.0620</td>\n",
              "      <td>-0.8095</td>\n",
              "      <td>-1.9590</td>\n",
              "      <td>0.1792</td>\n",
              "      <td>-0.1321</td>\n",
              "      <td>-1.0600</td>\n",
              "      <td>-0.8269</td>\n",
              "      <td>-0.3584</td>\n",
              "      <td>-0.8511</td>\n",
              "      <td>-0.5844</td>\n",
              "      <td>-2.5690</td>\n",
              "      <td>0.8183</td>\n",
              "      <td>-0.0532</td>\n",
              "      <td>-0.8554</td>\n",
              "      <td>0.1160</td>\n",
              "      <td>-2.3520</td>\n",
              "      <td>2.1200</td>\n",
              "      <td>-1.1580</td>\n",
              "      <td>-0.7191</td>\n",
              "      <td>-0.8004</td>\n",
              "      <td>-1.4670</td>\n",
              "      <td>-0.0107</td>\n",
              "      <td>-0.8995</td>\n",
              "      <td>0.2406</td>\n",
              "      <td>-0.2479</td>\n",
              "      <td>-1.0890</td>\n",
              "      <td>-0.7575</td>\n",
              "      <td>0.0881</td>\n",
              "      <td>-2.7370</td>\n",
              "      <td>0.8745</td>\n",
              "      <td>0.5787</td>\n",
              "      <td>...</td>\n",
              "      <td>-2.3820</td>\n",
              "      <td>-3.7350</td>\n",
              "      <td>-2.9740</td>\n",
              "      <td>-1.4930</td>\n",
              "      <td>-1.6600</td>\n",
              "      <td>-3.1660</td>\n",
              "      <td>0.2816</td>\n",
              "      <td>-0.2990</td>\n",
              "      <td>-1.1870</td>\n",
              "      <td>-0.5044</td>\n",
              "      <td>-1.7750</td>\n",
              "      <td>-1.6120</td>\n",
              "      <td>-0.9215</td>\n",
              "      <td>-1.0810</td>\n",
              "      <td>-3.0520</td>\n",
              "      <td>-3.4470</td>\n",
              "      <td>-2.7740</td>\n",
              "      <td>-1.8460</td>\n",
              "      <td>-0.5568</td>\n",
              "      <td>-3.3960</td>\n",
              "      <td>-2.9510</td>\n",
              "      <td>-1.1550</td>\n",
              "      <td>-3.2620</td>\n",
              "      <td>-1.5390</td>\n",
              "      <td>-2.4600</td>\n",
              "      <td>-0.9417</td>\n",
              "      <td>-1.5550</td>\n",
              "      <td>0.2431</td>\n",
              "      <td>-2.0990</td>\n",
              "      <td>-0.6441</td>\n",
              "      <td>-5.6300</td>\n",
              "      <td>-1.3780</td>\n",
              "      <td>-0.8632</td>\n",
              "      <td>-1.2880</td>\n",
              "      <td>-1.6210</td>\n",
              "      <td>-0.8784</td>\n",
              "      <td>-0.3876</td>\n",
              "      <td>-0.8154</td>\n",
              "      <td>0</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>id_001626bd3</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>72</td>\n",
              "      <td>D2</td>\n",
              "      <td>7cbed3131</td>\n",
              "      <td>-0.3254</td>\n",
              "      <td>-0.4009</td>\n",
              "      <td>0.9700</td>\n",
              "      <td>0.6919</td>\n",
              "      <td>1.4180</td>\n",
              "      <td>-0.8244</td>\n",
              "      <td>-0.2800</td>\n",
              "      <td>-0.1498</td>\n",
              "      <td>-0.8789</td>\n",
              "      <td>0.8630</td>\n",
              "      <td>-0.2219</td>\n",
              "      <td>-0.5121</td>\n",
              "      <td>-0.9577</td>\n",
              "      <td>1.1750</td>\n",
              "      <td>0.2042</td>\n",
              "      <td>0.1970</td>\n",
              "      <td>0.1244</td>\n",
              "      <td>-1.7090</td>\n",
              "      <td>-0.3543</td>\n",
              "      <td>-0.5160</td>\n",
              "      <td>-0.3330</td>\n",
              "      <td>-0.2685</td>\n",
              "      <td>0.7649</td>\n",
              "      <td>0.2057</td>\n",
              "      <td>1.3720</td>\n",
              "      <td>0.6835</td>\n",
              "      <td>0.8056</td>\n",
              "      <td>-0.3754</td>\n",
              "      <td>-1.2090</td>\n",
              "      <td>0.2965</td>\n",
              "      <td>-0.0712</td>\n",
              "      <td>0.6389</td>\n",
              "      <td>0.6674</td>\n",
              "      <td>-0.0783</td>\n",
              "      <td>1.1740</td>\n",
              "      <td>...</td>\n",
              "      <td>0.1535</td>\n",
              "      <td>-0.4640</td>\n",
              "      <td>-0.5943</td>\n",
              "      <td>0.3973</td>\n",
              "      <td>0.1500</td>\n",
              "      <td>0.5178</td>\n",
              "      <td>0.5159</td>\n",
              "      <td>0.6091</td>\n",
              "      <td>0.1813</td>\n",
              "      <td>-0.4249</td>\n",
              "      <td>0.7832</td>\n",
              "      <td>0.6529</td>\n",
              "      <td>0.5648</td>\n",
              "      <td>0.4817</td>\n",
              "      <td>0.0587</td>\n",
              "      <td>0.5303</td>\n",
              "      <td>0.6376</td>\n",
              "      <td>-0.3966</td>\n",
              "      <td>-1.4950</td>\n",
              "      <td>-0.9625</td>\n",
              "      <td>-0.0541</td>\n",
              "      <td>0.6273</td>\n",
              "      <td>0.4563</td>\n",
              "      <td>0.0698</td>\n",
              "      <td>0.8134</td>\n",
              "      <td>0.1924</td>\n",
              "      <td>0.6054</td>\n",
              "      <td>-0.1824</td>\n",
              "      <td>0.0042</td>\n",
              "      <td>0.0048</td>\n",
              "      <td>0.6670</td>\n",
              "      <td>1.0690</td>\n",
              "      <td>0.5523</td>\n",
              "      <td>-0.3031</td>\n",
              "      <td>0.1094</td>\n",
              "      <td>0.2885</td>\n",
              "      <td>-0.3786</td>\n",
              "      <td>0.7125</td>\n",
              "      <td>1</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 879 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "             id tratamento  tempo dose  ...    c-98    c-99  n_moa  ativo_moa\n",
              "0  id_000644bb2  com_droga     24   D1  ...  0.3801  0.4176      1       True\n",
              "1  id_000779bfc  com_droga     72   D1  ...  0.6077  0.7371      0      False\n",
              "2  id_000a6266a  com_droga     48   D1  ... -1.4080  0.6931      3       True\n",
              "3  id_0015fd391  com_droga     48   D1  ... -0.3876 -0.8154      0      False\n",
              "4  id_001626bd3  com_droga     72   D2  ... -0.3786  0.7125      1       True\n",
              "\n",
              "[5 rows x 879 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 69
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ti3leOz_V259"
      },
      "source": [
        "Olá Cientista!\n",
        "\n",
        "\n",
        "Seja bem vindo e bem vinda à aula 05. Parabéns por ter chegado até aqui!\n",
        "\n",
        "Na última aula nós analisamos separadamente, através de algumas técnicas, os efeitos em cada gene do grupo \"com_tratamento\" usados.\n",
        "Além disso utilizamos o boxplot para nos auxiliar na análise de forma visual tirando algumas conclusões muito interessantes.\n",
        "\n",
        "Você conseguiu fazer todos os desafios? Então vamos mergulhar juntos na aula 05!\n",
        "\n",
        "Nesta aula, vamos entender e aplicar o Machine Learning. Primeiramente, vamos entender o que o ML significa. Existem muitas definições para o termo, mas aqui, vamos trazer a definição clássica de um renomado e importante cientista da computação, o Arthur Samuel.\n",
        "\n",
        "***“Machine learning is the field of study that gives computers the ability to learn without being explicitly programmed” – (definição clássica de Arthur Samuel, cientista da computação, 1959)***\n",
        "\n",
        "*Tradução livre: \"Machine Learning é um campo de estudo que dá aos computadores a habilidade de aprender sem terem sido programados para tal.\"*\n",
        "\n",
        "Podemos interpretar a definição de Samuel, da seguinte forma: para que um computador reconheça determinada imagem, não precisamos passar um conjunto de regras, mostrar vários exemplos é suficiente para que o computador classifique suas imagens. Sendo assim, temos dois fatores determinantes neste processo: um conjunto de dados suficientemente grande e um poder computacional elevado. Por isso, até os anos 2000, era bem difícil encontrarmos problemas significantes resolvidos através do Machine Learning. Mas, a partir de 2009, empresas como a Netflix e Amazon, começaram a resolver problemas reais com o processamento dos dados através no ML, isso porque tinham uma infinidade de dados disponíveis e alto poder computacional.\n",
        "\n",
        "E quais problemas podemos resolver com esse tipo de algoritmo? Existem diversos casos em que podemos aplicá-los mas, vamos subdividir em dois grandes grupos:\n",
        "\n",
        "    - Problemas em que somos muito bons em resolver mas o Machine Learning, leva essa produção à um nível industrial, como é o caso da classificação de imagens\n",
        "    - Problemas em que não somos tão bons em resolver e o Machine Learning pode tomar uma melhor decisão em relação à nossa (imparcialidade, pode ser um fator), como é o caso da concessão de crédito.\n",
        "\n",
        "Mas como o Machine Learning vai ser aplicado no nosso problema?\n",
        "Bom, utilizando a nossa base de dados ```dados_combinados```, o primeiro problema que pensamos em resolver é: **dado os experimentos, queremos saber se qualquer mecanismo de ação foi ativado ou não ativado.** Ou seja, este caso é um problema de classificação binária.\n",
        "\n",
        "Mas, também podemos pensar se **dado os experimentos, queremos saber se um determinado mecanismo de ação foi ativado, recebendo como resposta, o composto relacionado**. E aqui, temos um problema de classificação multiclasse, já que cada composto químico é considerado uma classe diferente. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 251
        },
        "id": "bo0AdZGPoyX_",
        "outputId": "f7bc13a3-0ee1-4aa5-e20f-a85369396a9f"
      },
      "source": [
        "dados_combinados.head()"
      ],
      "execution_count": 70,
      "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>id</th>\n",
              "      <th>tratamento</th>\n",
              "      <th>tempo</th>\n",
              "      <th>dose</th>\n",
              "      <th>composto</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>g-7</th>\n",
              "      <th>g-8</th>\n",
              "      <th>g-9</th>\n",
              "      <th>g-10</th>\n",
              "      <th>g-11</th>\n",
              "      <th>g-12</th>\n",
              "      <th>g-13</th>\n",
              "      <th>g-14</th>\n",
              "      <th>g-15</th>\n",
              "      <th>g-16</th>\n",
              "      <th>g-17</th>\n",
              "      <th>g-18</th>\n",
              "      <th>g-19</th>\n",
              "      <th>g-20</th>\n",
              "      <th>g-21</th>\n",
              "      <th>g-22</th>\n",
              "      <th>g-23</th>\n",
              "      <th>g-24</th>\n",
              "      <th>g-25</th>\n",
              "      <th>g-26</th>\n",
              "      <th>g-27</th>\n",
              "      <th>g-28</th>\n",
              "      <th>g-29</th>\n",
              "      <th>g-30</th>\n",
              "      <th>g-31</th>\n",
              "      <th>g-32</th>\n",
              "      <th>g-33</th>\n",
              "      <th>g-34</th>\n",
              "      <th>...</th>\n",
              "      <th>c-62</th>\n",
              "      <th>c-63</th>\n",
              "      <th>c-64</th>\n",
              "      <th>c-65</th>\n",
              "      <th>c-66</th>\n",
              "      <th>c-67</th>\n",
              "      <th>c-68</th>\n",
              "      <th>c-69</th>\n",
              "      <th>c-70</th>\n",
              "      <th>c-71</th>\n",
              "      <th>c-72</th>\n",
              "      <th>c-73</th>\n",
              "      <th>c-74</th>\n",
              "      <th>c-75</th>\n",
              "      <th>c-76</th>\n",
              "      <th>c-77</th>\n",
              "      <th>c-78</th>\n",
              "      <th>c-79</th>\n",
              "      <th>c-80</th>\n",
              "      <th>c-81</th>\n",
              "      <th>c-82</th>\n",
              "      <th>c-83</th>\n",
              "      <th>c-84</th>\n",
              "      <th>c-85</th>\n",
              "      <th>c-86</th>\n",
              "      <th>c-87</th>\n",
              "      <th>c-88</th>\n",
              "      <th>c-89</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",
              "      <th>n_moa</th>\n",
              "      <th>ativo_moa</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>id_000644bb2</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>24</td>\n",
              "      <td>D1</td>\n",
              "      <td>b68db1d53</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>-0.0326</td>\n",
              "      <td>0.5548</td>\n",
              "      <td>-0.0921</td>\n",
              "      <td>1.1830</td>\n",
              "      <td>0.1530</td>\n",
              "      <td>0.5574</td>\n",
              "      <td>-0.4015</td>\n",
              "      <td>0.1789</td>\n",
              "      <td>-0.6528</td>\n",
              "      <td>-0.7969</td>\n",
              "      <td>0.6342</td>\n",
              "      <td>0.1778</td>\n",
              "      <td>-0.3694</td>\n",
              "      <td>-0.5688</td>\n",
              "      <td>-1.1360</td>\n",
              "      <td>-1.1880</td>\n",
              "      <td>0.6940</td>\n",
              "      <td>0.4393</td>\n",
              "      <td>0.2664</td>\n",
              "      <td>0.1907</td>\n",
              "      <td>0.1628</td>\n",
              "      <td>-0.2853</td>\n",
              "      <td>0.5819</td>\n",
              "      <td>0.2934</td>\n",
              "      <td>-0.5584</td>\n",
              "      <td>-0.0916</td>\n",
              "      <td>-0.3010</td>\n",
              "      <td>-0.1537</td>\n",
              "      <td>...</td>\n",
              "      <td>0.3680</td>\n",
              "      <td>0.8427</td>\n",
              "      <td>0.1042</td>\n",
              "      <td>0.1403</td>\n",
              "      <td>0.1758</td>\n",
              "      <td>1.2570</td>\n",
              "      <td>-0.5979</td>\n",
              "      <td>1.2250</td>\n",
              "      <td>-0.0553</td>\n",
              "      <td>0.7351</td>\n",
              "      <td>0.5810</td>\n",
              "      <td>0.9590</td>\n",
              "      <td>0.2427</td>\n",
              "      <td>0.0495</td>\n",
              "      <td>0.4141</td>\n",
              "      <td>0.8432</td>\n",
              "      <td>0.6162</td>\n",
              "      <td>-0.7318</td>\n",
              "      <td>1.2120</td>\n",
              "      <td>0.6362</td>\n",
              "      <td>-0.4427</td>\n",
              "      <td>0.1288</td>\n",
              "      <td>1.4840</td>\n",
              "      <td>0.1799</td>\n",
              "      <td>0.5367</td>\n",
              "      <td>-0.1111</td>\n",
              "      <td>-1.0120</td>\n",
              "      <td>0.6685</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",
              "      <td>1</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>id_000779bfc</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>72</td>\n",
              "      <td>D1</td>\n",
              "      <td>df89a8e5a</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>0.3372</td>\n",
              "      <td>-0.4047</td>\n",
              "      <td>0.8507</td>\n",
              "      <td>-1.1520</td>\n",
              "      <td>-0.4201</td>\n",
              "      <td>-0.0958</td>\n",
              "      <td>0.4590</td>\n",
              "      <td>0.0803</td>\n",
              "      <td>0.2250</td>\n",
              "      <td>0.5293</td>\n",
              "      <td>0.2839</td>\n",
              "      <td>-0.3494</td>\n",
              "      <td>0.2883</td>\n",
              "      <td>0.9449</td>\n",
              "      <td>-0.1646</td>\n",
              "      <td>-0.2657</td>\n",
              "      <td>-0.3372</td>\n",
              "      <td>0.3135</td>\n",
              "      <td>-0.4316</td>\n",
              "      <td>0.4773</td>\n",
              "      <td>0.2075</td>\n",
              "      <td>-0.4216</td>\n",
              "      <td>-0.1161</td>\n",
              "      <td>-0.0499</td>\n",
              "      <td>-0.2627</td>\n",
              "      <td>0.9959</td>\n",
              "      <td>-0.2483</td>\n",
              "      <td>0.2655</td>\n",
              "      <td>...</td>\n",
              "      <td>0.3905</td>\n",
              "      <td>0.7099</td>\n",
              "      <td>0.2912</td>\n",
              "      <td>0.4151</td>\n",
              "      <td>-0.2840</td>\n",
              "      <td>-0.3104</td>\n",
              "      <td>-0.6373</td>\n",
              "      <td>0.2887</td>\n",
              "      <td>-0.0765</td>\n",
              "      <td>0.2539</td>\n",
              "      <td>0.4443</td>\n",
              "      <td>0.5932</td>\n",
              "      <td>0.2031</td>\n",
              "      <td>0.7639</td>\n",
              "      <td>0.5499</td>\n",
              "      <td>-0.3322</td>\n",
              "      <td>-0.0977</td>\n",
              "      <td>0.4329</td>\n",
              "      <td>-0.2782</td>\n",
              "      <td>0.7827</td>\n",
              "      <td>0.5934</td>\n",
              "      <td>0.3402</td>\n",
              "      <td>0.1499</td>\n",
              "      <td>0.4420</td>\n",
              "      <td>0.9366</td>\n",
              "      <td>0.8193</td>\n",
              "      <td>-0.4236</td>\n",
              "      <td>0.3192</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",
              "      <td>0</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>id_000a6266a</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>48</td>\n",
              "      <td>D1</td>\n",
              "      <td>18bb41b2c</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>0.2155</td>\n",
              "      <td>0.0065</td>\n",
              "      <td>1.2300</td>\n",
              "      <td>-0.4797</td>\n",
              "      <td>-0.5631</td>\n",
              "      <td>-0.0366</td>\n",
              "      <td>-1.8300</td>\n",
              "      <td>0.6057</td>\n",
              "      <td>-0.3278</td>\n",
              "      <td>0.6042</td>\n",
              "      <td>-0.3075</td>\n",
              "      <td>-0.1147</td>\n",
              "      <td>-0.0570</td>\n",
              "      <td>-0.0799</td>\n",
              "      <td>-0.8181</td>\n",
              "      <td>-1.5320</td>\n",
              "      <td>0.2307</td>\n",
              "      <td>0.4901</td>\n",
              "      <td>0.4780</td>\n",
              "      <td>-1.3970</td>\n",
              "      <td>4.6240</td>\n",
              "      <td>-0.0437</td>\n",
              "      <td>1.2870</td>\n",
              "      <td>-1.8530</td>\n",
              "      <td>0.6069</td>\n",
              "      <td>0.4290</td>\n",
              "      <td>0.1783</td>\n",
              "      <td>0.0018</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.0444</td>\n",
              "      <td>0.1894</td>\n",
              "      <td>-0.0014</td>\n",
              "      <td>-2.3640</td>\n",
              "      <td>-0.4682</td>\n",
              "      <td>0.1210</td>\n",
              "      <td>-0.5177</td>\n",
              "      <td>-0.0604</td>\n",
              "      <td>0.1682</td>\n",
              "      <td>-0.4436</td>\n",
              "      <td>0.4963</td>\n",
              "      <td>0.1363</td>\n",
              "      <td>0.3335</td>\n",
              "      <td>0.9760</td>\n",
              "      <td>-0.0427</td>\n",
              "      <td>-0.1235</td>\n",
              "      <td>0.0959</td>\n",
              "      <td>0.0690</td>\n",
              "      <td>-0.9416</td>\n",
              "      <td>-0.7548</td>\n",
              "      <td>-0.1109</td>\n",
              "      <td>-0.6272</td>\n",
              "      <td>0.3019</td>\n",
              "      <td>0.1172</td>\n",
              "      <td>0.1093</td>\n",
              "      <td>-0.3113</td>\n",
              "      <td>0.3019</td>\n",
              "      <td>-0.0873</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",
              "      <td>3</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>id_0015fd391</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>48</td>\n",
              "      <td>D1</td>\n",
              "      <td>8c7f86626</td>\n",
              "      <td>-0.5138</td>\n",
              "      <td>-0.2491</td>\n",
              "      <td>-0.2656</td>\n",
              "      <td>0.5288</td>\n",
              "      <td>4.0620</td>\n",
              "      <td>-0.8095</td>\n",
              "      <td>-1.9590</td>\n",
              "      <td>0.1792</td>\n",
              "      <td>-0.1321</td>\n",
              "      <td>-1.0600</td>\n",
              "      <td>-0.8269</td>\n",
              "      <td>-0.3584</td>\n",
              "      <td>-0.8511</td>\n",
              "      <td>-0.5844</td>\n",
              "      <td>-2.5690</td>\n",
              "      <td>0.8183</td>\n",
              "      <td>-0.0532</td>\n",
              "      <td>-0.8554</td>\n",
              "      <td>0.1160</td>\n",
              "      <td>-2.3520</td>\n",
              "      <td>2.1200</td>\n",
              "      <td>-1.1580</td>\n",
              "      <td>-0.7191</td>\n",
              "      <td>-0.8004</td>\n",
              "      <td>-1.4670</td>\n",
              "      <td>-0.0107</td>\n",
              "      <td>-0.8995</td>\n",
              "      <td>0.2406</td>\n",
              "      <td>-0.2479</td>\n",
              "      <td>-1.0890</td>\n",
              "      <td>-0.7575</td>\n",
              "      <td>0.0881</td>\n",
              "      <td>-2.7370</td>\n",
              "      <td>0.8745</td>\n",
              "      <td>0.5787</td>\n",
              "      <td>...</td>\n",
              "      <td>-2.3820</td>\n",
              "      <td>-3.7350</td>\n",
              "      <td>-2.9740</td>\n",
              "      <td>-1.4930</td>\n",
              "      <td>-1.6600</td>\n",
              "      <td>-3.1660</td>\n",
              "      <td>0.2816</td>\n",
              "      <td>-0.2990</td>\n",
              "      <td>-1.1870</td>\n",
              "      <td>-0.5044</td>\n",
              "      <td>-1.7750</td>\n",
              "      <td>-1.6120</td>\n",
              "      <td>-0.9215</td>\n",
              "      <td>-1.0810</td>\n",
              "      <td>-3.0520</td>\n",
              "      <td>-3.4470</td>\n",
              "      <td>-2.7740</td>\n",
              "      <td>-1.8460</td>\n",
              "      <td>-0.5568</td>\n",
              "      <td>-3.3960</td>\n",
              "      <td>-2.9510</td>\n",
              "      <td>-1.1550</td>\n",
              "      <td>-3.2620</td>\n",
              "      <td>-1.5390</td>\n",
              "      <td>-2.4600</td>\n",
              "      <td>-0.9417</td>\n",
              "      <td>-1.5550</td>\n",
              "      <td>0.2431</td>\n",
              "      <td>-2.0990</td>\n",
              "      <td>-0.6441</td>\n",
              "      <td>-5.6300</td>\n",
              "      <td>-1.3780</td>\n",
              "      <td>-0.8632</td>\n",
              "      <td>-1.2880</td>\n",
              "      <td>-1.6210</td>\n",
              "      <td>-0.8784</td>\n",
              "      <td>-0.3876</td>\n",
              "      <td>-0.8154</td>\n",
              "      <td>0</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>id_001626bd3</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>72</td>\n",
              "      <td>D2</td>\n",
              "      <td>7cbed3131</td>\n",
              "      <td>-0.3254</td>\n",
              "      <td>-0.4009</td>\n",
              "      <td>0.9700</td>\n",
              "      <td>0.6919</td>\n",
              "      <td>1.4180</td>\n",
              "      <td>-0.8244</td>\n",
              "      <td>-0.2800</td>\n",
              "      <td>-0.1498</td>\n",
              "      <td>-0.8789</td>\n",
              "      <td>0.8630</td>\n",
              "      <td>-0.2219</td>\n",
              "      <td>-0.5121</td>\n",
              "      <td>-0.9577</td>\n",
              "      <td>1.1750</td>\n",
              "      <td>0.2042</td>\n",
              "      <td>0.1970</td>\n",
              "      <td>0.1244</td>\n",
              "      <td>-1.7090</td>\n",
              "      <td>-0.3543</td>\n",
              "      <td>-0.5160</td>\n",
              "      <td>-0.3330</td>\n",
              "      <td>-0.2685</td>\n",
              "      <td>0.7649</td>\n",
              "      <td>0.2057</td>\n",
              "      <td>1.3720</td>\n",
              "      <td>0.6835</td>\n",
              "      <td>0.8056</td>\n",
              "      <td>-0.3754</td>\n",
              "      <td>-1.2090</td>\n",
              "      <td>0.2965</td>\n",
              "      <td>-0.0712</td>\n",
              "      <td>0.6389</td>\n",
              "      <td>0.6674</td>\n",
              "      <td>-0.0783</td>\n",
              "      <td>1.1740</td>\n",
              "      <td>...</td>\n",
              "      <td>0.1535</td>\n",
              "      <td>-0.4640</td>\n",
              "      <td>-0.5943</td>\n",
              "      <td>0.3973</td>\n",
              "      <td>0.1500</td>\n",
              "      <td>0.5178</td>\n",
              "      <td>0.5159</td>\n",
              "      <td>0.6091</td>\n",
              "      <td>0.1813</td>\n",
              "      <td>-0.4249</td>\n",
              "      <td>0.7832</td>\n",
              "      <td>0.6529</td>\n",
              "      <td>0.5648</td>\n",
              "      <td>0.4817</td>\n",
              "      <td>0.0587</td>\n",
              "      <td>0.5303</td>\n",
              "      <td>0.6376</td>\n",
              "      <td>-0.3966</td>\n",
              "      <td>-1.4950</td>\n",
              "      <td>-0.9625</td>\n",
              "      <td>-0.0541</td>\n",
              "      <td>0.6273</td>\n",
              "      <td>0.4563</td>\n",
              "      <td>0.0698</td>\n",
              "      <td>0.8134</td>\n",
              "      <td>0.1924</td>\n",
              "      <td>0.6054</td>\n",
              "      <td>-0.1824</td>\n",
              "      <td>0.0042</td>\n",
              "      <td>0.0048</td>\n",
              "      <td>0.6670</td>\n",
              "      <td>1.0690</td>\n",
              "      <td>0.5523</td>\n",
              "      <td>-0.3031</td>\n",
              "      <td>0.1094</td>\n",
              "      <td>0.2885</td>\n",
              "      <td>-0.3786</td>\n",
              "      <td>0.7125</td>\n",
              "      <td>1</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 879 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "             id tratamento  tempo dose  ...    c-98    c-99  n_moa  ativo_moa\n",
              "0  id_000644bb2  com_droga     24   D1  ...  0.3801  0.4176      1       True\n",
              "1  id_000779bfc  com_droga     72   D1  ...  0.6077  0.7371      0      False\n",
              "2  id_000a6266a  com_droga     48   D1  ... -1.4080  0.6931      3       True\n",
              "3  id_0015fd391  com_droga     48   D1  ... -0.3876 -0.8154      0      False\n",
              "4  id_001626bd3  com_droga     72   D2  ... -0.3786  0.7125      1       True\n",
              "\n",
              "[5 rows x 879 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 70
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NpoTdeuoV9A3"
      },
      "source": [
        "Mas, na nossa base ```dados_resultados``` nós encontramos o ID do composto e quais mecanismos de ação foram acionados por determinado experimento. Aqui, como um composto pode acionar mais de um mecanismo, nós teremos uma classificação multilabel (quando uma única instância pode ter mais de uma classe associada) e, através de um algoritmo de Machine Learning nós também podemos resolver este problema: **não somente ter como retorno se um mecanismo de ação foi ativado mas qual ou quais são estes mecanismos**.\n",
        "\n",
        "É importante lembrar que nossas bases de dados estão separadas com um propósito. A segunda ```dados_resultados``` é como se fosse a resposta da primeira e foi construída por cientistas que analisaram o efeito de um determinado composto nos mais de 700 genes. Esse conhecimento, virou uma base de dados com o passar dos anos e é fruto de análise científica.\n",
        "\n",
        "E o papel de um modelo de ML neste caso é, a partir de todo esforço humano empenhado por vários anos para entender e tabular qual efeito de um composto quimíco em uma determinada expressão gênica, ser treinado e aprender, ou seja, ser capaz de produzir resultados compatíveis com aqueles feitos pelos(as) próprios(as) cientistas.\n",
        "\n",
        "Aqui na imersão, vamos começar resolvendo o problema menos complexo, a classificação binária: **dado os experimentos, queremos saber se QUALQUER mecanismo de ação foi ativado ou não**. Basicamente, queremos que o modelo faça predições compatível com a última coluna da nossa base de dados ```dados_combinados```, a ```ativo_moa```."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TjlKUBNmpql6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "outputId": "cb281855-a07a-4048-e688-72d6d141bf6d"
      },
      "source": [
        "dados_resultados.head()"
      ],
      "execution_count": 10,
      "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>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>adrenergic_receptor_agonist</th>\n",
              "      <th>adrenergic_receptor_antagonist</th>\n",
              "      <th>akt_inhibitor</th>\n",
              "      <th>aldehyde_dehydrogenase_inhibitor</th>\n",
              "      <th>alk_inhibitor</th>\n",
              "      <th>ampk_activator</th>\n",
              "      <th>analgesic</th>\n",
              "      <th>androgen_receptor_agonist</th>\n",
              "      <th>androgen_receptor_antagonist</th>\n",
              "      <th>anesthetic_-_local</th>\n",
              "      <th>angiogenesis_inhibitor</th>\n",
              "      <th>angiotensin_receptor_antagonist</th>\n",
              "      <th>anti-inflammatory</th>\n",
              "      <th>antiarrhythmic</th>\n",
              "      <th>antibiotic</th>\n",
              "      <th>anticonvulsant</th>\n",
              "      <th>antifungal</th>\n",
              "      <th>antihistamine</th>\n",
              "      <th>antimalarial</th>\n",
              "      <th>antioxidant</th>\n",
              "      <th>antiprotozoal</th>\n",
              "      <th>antiviral</th>\n",
              "      <th>apoptosis_stimulant</th>\n",
              "      <th>aromatase_inhibitor</th>\n",
              "      <th>atm_kinase_inhibitor</th>\n",
              "      <th>atp-sensitive_potassium_channel_antagonist</th>\n",
              "      <th>atp_synthase_inhibitor</th>\n",
              "      <th>atpase_inhibitor</th>\n",
              "      <th>atr_kinase_inhibitor</th>\n",
              "      <th>aurora_kinase_inhibitor</th>\n",
              "      <th>...</th>\n",
              "      <th>radiopaque_medium</th>\n",
              "      <th>raf_inhibitor</th>\n",
              "      <th>ras_gtpase_inhibitor</th>\n",
              "      <th>retinoid_receptor_agonist</th>\n",
              "      <th>retinoid_receptor_antagonist</th>\n",
              "      <th>rho_associated_kinase_inhibitor</th>\n",
              "      <th>ribonucleoside_reductase_inhibitor</th>\n",
              "      <th>rna_polymerase_inhibitor</th>\n",
              "      <th>serotonin_receptor_agonist</th>\n",
              "      <th>serotonin_receptor_antagonist</th>\n",
              "      <th>serotonin_reuptake_inhibitor</th>\n",
              "      <th>sigma_receptor_agonist</th>\n",
              "      <th>sigma_receptor_antagonist</th>\n",
              "      <th>smoothened_receptor_antagonist</th>\n",
              "      <th>sodium_channel_inhibitor</th>\n",
              "      <th>sphingosine_receptor_agonist</th>\n",
              "      <th>src_inhibitor</th>\n",
              "      <th>steroid</th>\n",
              "      <th>syk_inhibitor</th>\n",
              "      <th>tachykinin_antagonist</th>\n",
              "      <th>tgf-beta_receptor_inhibitor</th>\n",
              "      <th>thrombin_inhibitor</th>\n",
              "      <th>thymidylate_synthase_inhibitor</th>\n",
              "      <th>tlr_agonist</th>\n",
              "      <th>tlr_antagonist</th>\n",
              "      <th>tnf_inhibitor</th>\n",
              "      <th>topoisomerase_inhibitor</th>\n",
              "      <th>transient_receptor_potential_channel_antagonist</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",
              "      <th>n_moa</th>\n",
              "      <th>ativo_moa</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>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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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>1</td>\n",
              "      <td>True</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>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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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>False</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>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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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>3</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>id_0015fd391</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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>id_001626bd3</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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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",
              "      <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>1</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 209 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "             id  5-alpha_reductase_inhibitor  ...  n_moa  ativo_moa\n",
              "0  id_000644bb2                            0  ...      1       True\n",
              "1  id_000779bfc                            0  ...      0      False\n",
              "2  id_000a6266a                            0  ...      3       True\n",
              "3  id_0015fd391                            0  ...      0      False\n",
              "4  id_001626bd3                            0  ...      1       True\n",
              "\n",
              "[5 rows x 209 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eMsWg80Za2Jf"
      },
      "source": [
        "As etapas lógicas do nosso processo é:\n",
        "\n",
        "    Dado um composto -> temos a nossa assinatura celular (```g e c```) -> analisamos o(s) mecanismo(s) de ação ativados (MoA) \n",
        "\n",
        "Mas, a nossa pergunta a ser respondida é, **\"Dado um composto e uma assinatura celular, houve algum MoA ativado?\"** Então, as etapas lógicas passam a ser:\n",
        "\n",
        "    Composto -> assinatura celular -> MoA ativado?\n",
        "\n",
        "E como nosso modelo vai aprender a partir da nosa base de dados?\n",
        "\n",
        "    Composto 1 -> Assinatura A -> MoA = 1\n",
        "    Composto 2 -> Assinatura B -> MoA = 0\n",
        "    Composto 3 -> Assinatura C -> MoA = 1\n",
        "    ...\n",
        "    Composto N -> Assinatura N -> MoA = 0\n",
        "\n",
        "Ou seja, toda a nossa base de dados ```dados_combinados``` será a fonte de aprendizado do nosso modelo. Onde, cada experimento (linha) representa um exemplo (assinatura) já que temos a reunião de todas as informações necessárias para criar a nossa classificação binária e, terá como variável resposta, também chamada de target, a coluna ```ativo_moa```, uma variável binária (0, 1).\n",
        "\n",
        "E, por fim, o nosso modelo deverá ser capaz de resolver o seguinte problema:\n",
        "  \n",
        "    Composto candidato -> Assinatura X -> MoA = 1 OU MoA = 0?\n",
        "\n",
        "Existem diversas técnicas de ML que são capazes de solucionar o nosso problema mas aqui, vamos usar a **Regressão Logística** e, essa técnica já está implementada em uma biblioteca bastante usada e bem importante para ML no Python, o [Scikit-Learn](https://scikit-learn.org/stable/).\n",
        "\n",
        "Sugerimos que você leia essa documentação e se familiarize com esta biblioteca, ela será muito importante na sua trajetória como Cientista de Dados. E, clicando em \"Classification\" na página inicial, encontramos, no primeiro capítulo, chamado \"Linear Models\", o nosso subtítulo de interesse: [**\"1.1.11 Logistic Regression\"**](https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression).\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fpf69vPeh-eI"
      },
      "source": [
        "Quando implementamos um modelo de classificação, precisamos separar a nossa base de dados em base de treino e base de teste. A base de treino, vai ser onde o nosso modelo vai aprender e, a base de teste serve para a verificação do aprendizado do modelo, ou seja, poderemos entender se, de fato, as previsões do nosso modelo estão de acordo com o esperado. \n",
        "\n",
        "Não podemos treinar e testar o modelo com o mesmo conjunto de dados porque, seria muito mais fácil dele prever corretamente. A base de dados de teste, funciona como um novo conjunto onde poderemos medir o quanto o nosso modelo está acertando, dado que já temos a variável resposta real.\n",
        "\n",
        "Para fazer esta divisão, usamos uma função do Scikit-Learn chamada ```train_teste_split``` (como já aprendemos, para usar uma biblioteca ou função, precisamos, primeiramente, fazer a importação padrão e, neste caso, usaremos ```from sklearn.model_selection import train_test_split```).\n",
        "\n",
        "Para fazer a divisão da base de dados, é necessário definir qual será o nosso ```x``` e o nosso ```y```. Um modelo matemático, define um ```y``` em função de um ```x```, ou seja, ```f(x) = y```. Neste projeto, o nosso ```y``` será o target, ```ativo_moa``` e, ele será definido a partir de uma assinatura (conjunto composto por ```g``` + ```c``` ou , expressão gênica + viabilidade celular).\n",
        "\n",
        "Assim, teremos: \n",
        "\n",
        "    x = dados_combinados.select_dtypes('float64')  ->  toda a base de dados que tem os valores com o formato, ```float64```, no nosso caso, todas as variáveis ```g``` e ```c```\n",
        "\n",
        "    y = dados_combinados['ativo_moa']  ->  apenas a coluna target, ou seja, ```ativo_moa```\n",
        "\n",
        "Para o ```train_test_split``` definiremos:\n",
        "    \n",
        "    x_treino, x_teste, y_treino, y_teste\n",
        "\n",
        "E a função receberá como parâmetro:\n",
        "\n",
        "    train_test_split(x, y, test_size = 0.2)\n",
        "\n",
        "Onde, \n",
        "\n",
        "    x -> base de dados definida como x, no nosso caso, todas as variáveis ```g``` e ```c```\n",
        "\n",
        "    y -> variável resposta, no nosso caso, ```ativo_moa```\n",
        "\n",
        "    teste_size = 0.2 -> tamanho que definimos para a base de teste\n",
        "\n",
        "A proporção entre treino e teste, varia de acordo com o volume de dados que temos. Mas, usualmente, encontramos proporções 30/70, 25/75 ou 20/80. Aqui, definimos como 0,2 ou 20%, isso significa que a base de treino será composta pelos 80% restante.\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1xoN4eBFj072"
      },
      "source": [
        "###Separação de dados em treino e teste"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JajA78DhvirK"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "x = dados_combinados.select_dtypes('float64')\n",
        "y = dados_combinados['ativo_moa']\n",
        "\n",
        "x_treino, x_teste, y_treino, y_teste = train_test_split(x, y, test_size = 0.2)"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y_M6jQpXtY8k"
      },
      "source": [
        "#variáveis independentes: expressao genica e viabilidade celular (só os float)\n",
        "x = dados_combinados.select_dtypes('float64')\n",
        "#variável dependente: ativo_moa\n",
        "y = dados_combinados['ativo_moa'] \n",
        "#extrai as amostras de treino e teste, teste: 20%, treino: 80%\n",
        "x_treino, x_teste, y_treino, y_teste = train_test_split(x, y, test_size = 0.2, stratify=y, random_state=376)"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NNV5yEZIkllB"
      },
      "source": [
        "###Teste de acurácia"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LacGLcHapkeT"
      },
      "source": [
        "Agora, podemos fazer a nossa primeira regressão logística.\n",
        "\n",
        "O primeiro passo, é definir o nosso ```x```, ```y``` e o ```x_treino, x_teste, y_treino, y_teste``` do ```train_test_split```.\n",
        "\n",
        "Após isso, definimos um nome para o nosso modelo (```modelo_rlogistica```) e declaramos a função ```LogisticRegression()```. Acrescentamos um parâmetro ```max_iter = 1000``` no modelo para que ele seja capaz de convergir para um resultado e assim, não recebemos nenhum warning no resultado.\n",
        "\n",
        "Em seguida, ajustamos o modelo com o ```.fit(x_treino, y_treino)``` (lembrando que os argumentos desta função são ```x``` e ```y``` e, usamos ```x_treino``` e ```y_treino``` porque será através dessas bases que o algoritmo aprenderá).\n",
        "\n",
        "E, por fim, usamos a função ```.score(x_teste, y_teste)``` para verificar como o nosso modelo ajustado está se saindo nos dados de teste. O ```score()```calcula a acurácia, ou seja, quantas predições o nosso modelo acertou na base de teste.\n",
        "\n",
        "Como resultado, temos o valor da acurácia: ~0,627. Então, concluímos que o nosso modelo está acertando ~62% das previsões (caso tivéssemos 100 experimentos, o modelo acertaria 62 casos e erraria 38 casos).\n",
        "\n",
        "Mas, como podemos estabelecer se este é um bom valor de acurácia ou não?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fTi1yY7jkke7",
        "outputId": "ec93086c-8155-4a8a-e084-f45f1f0818c7"
      },
      "source": [
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "#modelo de regressão logística\n",
        "modelo_rlogistica = LogisticRegression(max_iter=1000)\n",
        "#treinando modelo\n",
        "modelo_rlogistica.fit(x_treino, y_treino)\n",
        "#score do modelo: como o modelo está se saindo no teste: retorna a acurácia\n",
        "modelo_rlogistica.score(x_teste, y_teste)"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6267058576527399"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "blWKcXpwlfsN"
      },
      "source": [
        "###Dummy Classifier"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Mr_zNLw2s7VC"
      },
      "source": [
        "Para estabelecer se temos um bom valor de acurácia, precisamos de uma base comparativa, quer dizer, preciso dos resultados de um outro modelo para entender se estamos acertando mais ou menos casos.\n",
        "\n",
        "Para isso, o Scikit-Learn, já tem implementado alguns algoritmos que fazem modelos menos complexos e que podem ser usados como base comparativa. Neste caso, vamos usar o ```DummyClassifier```.\n",
        "\n",
        "O processo para ajustar este modelo é muito parecido com aquele que usamos para a regressão logística: dividimos a base em treino e teste, definimos um nome para esse modelo e o instanciamos, o ajustamos com os dados de treino e, ao final, verificamos a acurácia com o conjunto teste.\n",
        "\n",
        "Ao instanciarmos o ```DummyClassifier```, precisamos definir a estratégia que ele usará para fazer o ajuste do modelo. No nosso caso, definimos a estratégia será os dados mais frequentes, isso quer dizer que: dado o valor mais frequente da nossa variável resposta (```ativo_moa```), o modelo Dummy vai chutar que todos os eventos da base de dados assumem aquele valor no target. Assim, temos: ```DummyClassifier('most_frequent')```.\n",
        "\n",
        "Nesse momento, para calcular a acurácia, vamos usar outra estratégia, a função ```accuracy_score``` também do Scikit-Learn. Ela receberá como parâmetros a base teste da variável resposta (```y_teste```) e as previsões do modelo que, até o momento ainda não calculamos.\n",
        "Para calcular as previsões, usamos o ```.predict()``` da mesma biblioteca que, receberá como parâmetro a base ```x_teste```.\n",
        "Por fim, o cálculo da acurácia será: \n",
        "\n",
        "    previsao_dummy = modelo_dummy.predict(x_teste)\n",
        "    accuracy_score(y_teste, previsao_dummy)\n",
        "\n",
        "A acurácia do ```DummyClassifier```foi de ~60,7%, isso significa que nosso modelo ```LogisticRegression``` teve um desempenho um pouco melhor."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zEy9U9aiyOPq",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f4971dff-1bb8-47f1-a244-778f2c2c3231"
      },
      "source": [
        "from sklearn.dummy import DummyClassifier\n",
        "from sklearn.metrics import accuracy_score\n",
        "\n",
        "modelo_dummy = DummyClassifier('most_frequent')\n",
        "modelo_dummy.fit(x_treino, y_treino)\n",
        "previsao_dummy = modelo_dummy.predict(x_teste)\n",
        "accuracy_score(y_teste, previsao_dummy)"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6067604450976275"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ilf09FzQzwbs"
      },
      "source": [
        "Podemos entender a acurácia do ```DummyClassifier``` se fizermos um ```value_counts``` normalizado e então verificamos que a proporção entre a classe 0 e a classe 1 é 60/30, ou seja, o valor mais frequente (estratégia usada no ```DummyClassifier```) representa 60% da base de dados.\n",
        "Isso quer dizer que, caso o modelo chutasse a mesma classe para todos os eventos da base de teste, teríamos um acerto de 60% dos casos. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vMtsJKUvziAl",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "72524604-d5ca-411e-d80f-613a28192e91"
      },
      "source": [
        "dados_combinados['ativo_moa'].value_counts(normalize=True) #base de dados não é balanceada"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True     0.60666\n",
              "False    0.39334\n",
              "Name: ativo_moa, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "riKEiA2d1Vqs"
      },
      "source": [
        "Entretanto, toda vez que rodamos tanto a nossa regressão logística, quanto o nosso dummy classifier, recebemos como retorno um valor diferente de acurácia, apesar de em muitos casos, serem próximos.\n",
        "\n",
        "O ```train_test_split``` a base de dados em subconjuntos aleatórios de treino e teste. Isso significa que toda vez que você executá-lo, você obterá um resultado diferente.\n",
        "\n",
        "Por outro lado, se você declarar o parâmetro ```random_state = n```, poderá garantir que a saída de todas as execuções será igual, ou seja, sua divisão será sempre a mesma. Não importa qual número você escolherá para o ```random_state```, o importante é que toda vez que você usá-lo, sempre terá a mesma saída. \n",
        "\n",
        "Além do ```random_state```, existe mais um parâmetro bem importante que deve ser levado em consideração na divisão da nossa base de dados, o ```stratify```. \n",
        "\n",
        "Se repararmos na divisão da variável ```ativo_moa```, entre ```true``` e ```false```, percebemos que segue um balanceamento 60/40. E, é do nosso interesse que as bases de treino e teste sigam, além de uma divisão reprodutível (garantida com ```random_state```), a proporção original do conjunto de dados. Assim, atribuimos ```stratify = coluna```, onde ```coluna = nome da coluna que o stratify deve levar em consideração a proporção de classes```. No nosso caso, usamos ```stratify = y```, pois ```y``` é a nossa variável resposta e, é dela que o stratify deve considerar a proporção: 60/40.\n",
        "\n",
        "Esses parâmetros são úteis, porque garantimos que os resultados são reproduzíveis e compatíveis, independente do modelo (qual técnica, por exemplo, regressão logística, dummy classifier, árvore de decisão e etc) que vamos treinar e testar a partir desta divisão.\n",
        "\n",
        "Ou seja, ao final, nosso ```train_teste_split``` ficará:\n",
        "\n",
        "    x_treino, x_teste, y_treino, y_teste = train_test_split(x, y, test_size = 0.2, stratify=y, random_state=376)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fu1PUlJyl_lY"
      },
      "source": [
        "###Decision Tree"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "20QwzZ879C8Q"
      },
      "source": [
        "Nesse momento, temos interesse a explorar outras técnicas de modelagem para comparar os resultados obtidos e ver qual o melhor algoritmo que conseguimos produzir. \n",
        "\n",
        "Sendo assim, vamos testar uma nova técnica: a árvore de decisão.\n",
        "\n",
        "Novamente, a sequência lógica para ajustar esse novo modelo é muito parecida com que estávamos vendo até o momento: definimos as bases ```x``` e ```y```, fazemos o ```train_test_split```, declaramos o novo modelo, ajustamos e verificamos a acurácia. \n",
        "\n",
        "Aqui, vamos usar o ```DecisionTreeClassifier```, também da biblioteca Scikit-Learn. O único parâmetro que vamos declarar para esta função será o ```max_depth = 3``` mas, vamos explorar seu significado a seguir. \n",
        "\n",
        "Como resultado, temos uma acurácia (calculada a partir da base de teste) de ~0,61% e, é um pouco menos do que a acurácia obtida com a regressão logística."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tXMM8M0-23We",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d6b3dbc9-2bff-4c3e-d54c-87e7f5d58e40"
      },
      "source": [
        "from sklearn.tree import DecisionTreeClassifier\n",
        "\n",
        "\n",
        "x = dados_combinados.select_dtypes('float64')\n",
        "y = dados_combinados['ativo_moa'] \n",
        "x_treino, x_teste, y_treino, y_teste = train_test_split(x, y, test_size = 0.2, stratify=y, random_state=376)\n",
        "\n",
        "#max_depth: parâmetro que determina a quantidade de níveis da árvore\n",
        "modelo_arvore = DecisionTreeClassifier(max_depth = 3)\n",
        "modelo_arvore.fit(x_treino, y_treino) #treinando modelo da árvore\n",
        "modelo_arvore.score(x_teste, y_teste) #vendo a acurácia do modelo da árvore"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6105395758975436"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QFA5dfcPCh-2"
      },
      "source": [
        "Agora, vamos entender melhor como funciona uma árvore de decisão e o que significa o parâmetro ```max_depth = 3```. \n",
        "Plotamos uma imagem que representa o nosso modelo e, nesse momento, é importante que você não se apegue ao código que usamos para isso. O que realmente é relevante é entender o algoritmo!\n",
        "\n",
        "A figura é dividida em 4 níveis e, o primeiro nível (estamos considerando a orientação de cima para baixo), é composto por apenas um quadro que é chamado de **nó raíz** ou somente **raíz** e ele tem como regras, os melhores atributos que podem dividir a nossa base de dados em dois conjuntos distintos (lembrando que nosso objetivo é separar a nossa base de dados em **mecanismo de ação ativado ou não ativado**).\n",
        "\n",
        "Entre o primeiro e o segundo nível (e nos demais níveis), temos duas setas: uma para a direita e outra para a esquerda. Essas setas são chamadas de **ramos** e subdividem o nível acima em dois outros conjuntos que separam novamente os dados em **mecanismo de ação ativado ou não ativado**, a partir de uma decisão.\n",
        "\n",
        "O último nível da árvore (mais abaixo), apresenta n quadros que apresentam características que devem ser levadas em conta na hora da classificação e se dividem em ```class = ativado``` e ```class = não ativado```. Esse nível não apresenta mais ramificações e, esses quadros são chamados de **folhas**.\n",
        "\n",
        "Se reparmos, a nossa árvore tem 3 níveis de decisão a partir da raíz e é exatamente isso que o parâmetro ```max_depth```representa: a profundidade da árvore."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pkwCz7pw3rnX",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 575
        },
        "outputId": "f6984109-a1af-40cf-aca6-b8057ad8ed3a"
      },
      "source": [
        "from sklearn import tree\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(15, 10), facecolor='k')\n",
        "tree.plot_tree(modelo_arvore,\n",
        "               ax=ax,\n",
        "               fontsize=10,\n",
        "               rounded=True, #imagem redonda e com fonte diferente\n",
        "               filled=True, #configura cores\n",
        "               feature_names=x_treino.columns,\n",
        "               class_names=['Não Ativado', 'Ativado'])\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZZIdxEN-GWGL"
      },
      "source": [
        "Como vimos, o desempenho do nosso novo modelo com apenas 3 camadas de decisão não foi tão satisfatório assim. Por isso, vamos variar o ```max_depth``` dentro de um intervalo de valores (range) e comparar os resultados obtidos. \n",
        "\n",
        "O processo para obter os valores que queremos, é sempre o mesmo. Por isso, vamos criar um loop ou um laço de repetição (um loop, executa um bloco de código em repetição enquanto uma condição é atendida). \n",
        "\n",
        "Então, após fazermos a separação da base de dados em ```x``` e ```y``` e dividí-las no ```train_test_split```, vamos criar um ```for``` para que nosso modelo seja treinado e testado em repetição dentro de um range. Para isso, faremos:\n",
        "\n",
        "    teste=[]  ->  criando lista para armazenar os resultados da acurácia nos dados de teste \n",
        "    treino=[]  ->  criando lista para armazenar os resultados da acurácia nos dados de treino \n",
        "    for i in range(1,15):  ->  estabelecendo condição que deve ser atendida (i) e range de valor (range(1, 15))\n",
        "        modelo_arvore = DecisionTreeClassifier(max_depth = i)  ->  instanciando modelo que será treinado e definindo max_depth = i, ou seja, estamos variando a profundidade da nossa árvore\n",
        "        modelo_arvore.fit(x_treino, y_treino)  ->  ajustando o modelo a partir das bases de treino\n",
        "        teste.append(modelo_arvore.score(x_teste, y_teste))  ->  armazenando os resultados de acurácia de teste dos modelos produzidos na lista 'teste'\n",
        "        treino.append(modelo_arvore.score(x_treino, y_treino))  ->  armazenando os resultados de acurácia de treino dos modelos produzidos na lista 'treino'"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nWgI7qld4vIf"
      },
      "source": [
        "from sklearn.tree import DecisionTreeClassifier\n",
        "\n",
        "\n",
        "x = dados_combinados.select_dtypes('float64')\n",
        "y = dados_combinados['ativo_moa'] \n",
        "x_treino, x_teste, y_treino, y_teste = train_test_split(x, y, test_size = 0.2, stratify=y, random_state=376)\n",
        "\n",
        "teste=[]\n",
        "treino=[]\n",
        "for i in range(1,15):\n",
        "    modelo_arvore = DecisionTreeClassifier(max_depth = i)\n",
        "    modelo_arvore.fit(x_treino, y_treino)\n",
        "    teste.append(modelo_arvore.score(x_teste, y_teste))\n",
        "    treino.append(modelo_arvore.score(x_treino, y_treino))"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7Ps2wFC3KYb9"
      },
      "source": [
        "Quando chamamos as saídas do nosso ```for``` (```teste``` e ```treino```), podemos observar um comportamento um pouco estranho: enquanto nos dados de teste, o valor da nossa acurácia vai caindo, nos dados de treino, esse valor vai aumentando."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uSSxDS3y5kgB",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5094fa57-3212-4a30-cab6-b96e8a31f6f3"
      },
      "source": [
        "teste"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.6067604450976275,\n",
              " 0.6067604450976275,\n",
              " 0.6105395758975436,\n",
              " 0.6204073063195465,\n",
              " 0.6199874028973336,\n",
              " 0.6180978374973756,\n",
              " 0.6210371614528658,\n",
              " 0.6210371614528658,\n",
              " 0.6162082720974176,\n",
              " 0.6132689481419273,\n",
              " 0.6099097207642242,\n",
              " 0.5973126180978375,\n",
              " 0.6010917488977535,\n",
              " 0.5992021834977955]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 84
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Sg69s_mP6hb9",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "91396f13-c38c-44f2-9a25-3cdb9ee3f964"
      },
      "source": [
        "treino"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.6066348223190384,\n",
              " 0.6066348223190384,\n",
              " 0.6211222508004829,\n",
              " 0.6360820954280615,\n",
              " 0.6412261823526324,\n",
              " 0.6489423127394888,\n",
              " 0.6605427536612252,\n",
              " 0.6707259461445594,\n",
              " 0.6896225919899218,\n",
              " 0.7068395359823632,\n",
              " 0.7294105296309905,\n",
              " 0.7529788462547897,\n",
              " 0.7770195790247231,\n",
              " 0.799223137893024]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 78
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cXMgQLmnLMAF"
      },
      "source": [
        "Para investigar o que aconteceu, vamos plotar um gráfico para entender o comportamento combinado dessas listas.\n",
        "\n",
        "Usaremos o ```lineplot``` do Seaborn para construir essa visualização. \n",
        "Como parâmetros, teremos: \n",
        "\n",
        "    x = range(1, 15)   ->  intervalo que usamos no for\n",
        "    y = teste e y = treino  ->  listas criamos no for e será usada no eixo y do gráfico\n",
        "    label = 'teste' e label = 'teste'  ->  etiqueta para identificar as duas linhas que estarão no gráfico \n",
        "\n",
        "Observando o gráfico produzido, podemos perceber que, inicialmente, os valores de treino e teste são muito próximos mas, conforme vamos aumentando o range, o valor da acurácia para os dados de treino aumentam muito enquanto, esse mesmo valor para os dados de teste, vai caindo (as linhas vão se afastando).\n",
        "\n",
        "Isso significa que, conforme aumentamos a profundidade da nossa árvore de decisão, a classificação para os dados de treino ficam muito boas pois, ela consegue captar muito bem as características desses dados. Mas, em contrapartida, o modelo fica tão bom os dados de treino que, quando apresentamos um novo conjunto de dados (teste), ele não consegue generalizar tão bem e assim, temos uma acurácia em queda.\n",
        "\n",
        "Esse problema de um modelo com performance muito boa nos dados de treino mas performance em queda nos dados de teste, é muito conhecido na ciência de dados e chamamos de **overfitting**.\n",
        "\n",
        "Então, conseguimos perceber que, para melhorar o nosso modelo, não é suficiente somente aumentar a profundidade da árvore de decisão e assim, precisamos traçar uma outra etratégia para melhorar a performance do nosso modelo.\n",
        "\n",
        "Nesse momento, vamos testar uma nova técnica de modelagem: o Random Forest. \n",
        "O que este modelo faz é criar várias árvores de decisões na qual ele considera amostras aleatórias do nosso conjunto total.  "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kB58AYEV6pxC",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "outputId": "bdd3b59a-51e3-43c3-f0ef-477dc8eb940c"
      },
      "source": [
        "sns.lineplot(x=range(1,15), y = teste, label='teste')\n",
        "sns.lineplot(x=range(1,15), y = treino, label='treino')"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f5f3ffdb410>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "r7Ke5shLSsJv"
      },
      "source": [
        "A função que treina um modelo de Random Forest (```RandomForestClassifier()```) não aceita que tenhamos variáveis em formato de string. \n",
        "Por isso, precisamos tratar as variáveis que são strings e transformá-las em números (ou no formato float64).\n",
        "\n",
        "Para fazer essa transformação, vamos usar uma função do Pandas, chamada ```get_dummies``` e nela, declaramos a coluna que queremos que seja transformada. Por exemplo, aqui, usamos:\n",
        "\n",
        "    pd.get_dummies(dados_combinados['tratamento'])\n",
        "\n",
        "Isso porque, nossa intenção é transformar a coluna ```tratamento``` da base ```dados_combinados``` em dados numéricos.\n",
        "\n",
        "Para fazer a transformação desejada, o ```get_dummies``` pega todas as categorias de uma variável e as transforma em novas colunas binárias. \n",
        "Por exemplo:\n",
        "\n",
        "A variável ```tratamento``` apresenta duas categorias ```com_controle``` e ```sem_controle```. Quando aplicamos o ```get_dummies```, ele cria duas novas colunas binárias e análogas as categorias. \n",
        "Se temos um evento que apresentava na coluna original a marcação ```com_droga```, agora ela apresentará valor 0 na coluna ```com_controle``` e valor 1 na coluna ```com_droga``` e vice-versa.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cv5kuY9l9BZy",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 415
        },
        "outputId": "284a3102-380a-4c93-828a-b3e0ab7a48a9"
      },
      "source": [
        "pd.get_dummies(dados_combinados['tratamento'])"
      ],
      "execution_count": 20,
      "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>com_controle</th>\n",
              "      <th>com_droga</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23809</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23810</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23811</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23812</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23813</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>23814 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "       com_controle  com_droga\n",
              "0                 0          1\n",
              "1                 0          1\n",
              "2                 0          1\n",
              "3                 0          1\n",
              "4                 0          1\n",
              "...             ...        ...\n",
              "23809             0          1\n",
              "23810             0          1\n",
              "23811             1          0\n",
              "23812             0          1\n",
              "23813             0          1\n",
              "\n",
              "[23814 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Q_He8_XFUvfP"
      },
      "source": [
        "Agora que entendemos como o ```get_dummies``` funciona, podemos transformar todas as nossas variáveis de interesse: ```tratamento```, ```dose``` e ```tempo```.\n",
        "\n",
        "Para isso, declaramos a base de dados completa e, no parâmetro ```columns```, listamos as variáveis que devem ser transformadas. Assim teremos:\n",
        "\n",
        "    pd.get_dummies(dados_combinados, columns=['tratamento', 'dose', 'tempo'])"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MUNHu8om9SSz",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 435
        },
        "outputId": "efda3e05-ddf1-46fe-b49d-02a07f23457d"
      },
      "source": [
        "pd.get_dummies(dados_combinados, columns=['tratamento', 'dose', 'tempo'])"
      ],
      "execution_count": 21,
      "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>id</th>\n",
              "      <th>composto</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>g-7</th>\n",
              "      <th>g-8</th>\n",
              "      <th>g-9</th>\n",
              "      <th>g-10</th>\n",
              "      <th>g-11</th>\n",
              "      <th>g-12</th>\n",
              "      <th>g-13</th>\n",
              "      <th>g-14</th>\n",
              "      <th>g-15</th>\n",
              "      <th>g-16</th>\n",
              "      <th>g-17</th>\n",
              "      <th>g-18</th>\n",
              "      <th>g-19</th>\n",
              "      <th>g-20</th>\n",
              "      <th>g-21</th>\n",
              "      <th>g-22</th>\n",
              "      <th>g-23</th>\n",
              "      <th>g-24</th>\n",
              "      <th>g-25</th>\n",
              "      <th>g-26</th>\n",
              "      <th>g-27</th>\n",
              "      <th>g-28</th>\n",
              "      <th>g-29</th>\n",
              "      <th>g-30</th>\n",
              "      <th>g-31</th>\n",
              "      <th>g-32</th>\n",
              "      <th>g-33</th>\n",
              "      <th>g-34</th>\n",
              "      <th>g-35</th>\n",
              "      <th>g-36</th>\n",
              "      <th>g-37</th>\n",
              "      <th>...</th>\n",
              "      <th>c-69</th>\n",
              "      <th>c-70</th>\n",
              "      <th>c-71</th>\n",
              "      <th>c-72</th>\n",
              "      <th>c-73</th>\n",
              "      <th>c-74</th>\n",
              "      <th>c-75</th>\n",
              "      <th>c-76</th>\n",
              "      <th>c-77</th>\n",
              "      <th>c-78</th>\n",
              "      <th>c-79</th>\n",
              "      <th>c-80</th>\n",
              "      <th>c-81</th>\n",
              "      <th>c-82</th>\n",
              "      <th>c-83</th>\n",
              "      <th>c-84</th>\n",
              "      <th>c-85</th>\n",
              "      <th>c-86</th>\n",
              "      <th>c-87</th>\n",
              "      <th>c-88</th>\n",
              "      <th>c-89</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",
              "      <th>n_moa</th>\n",
              "      <th>ativo_moa</th>\n",
              "      <th>tratamento_com_controle</th>\n",
              "      <th>tratamento_com_droga</th>\n",
              "      <th>dose_D1</th>\n",
              "      <th>dose_D2</th>\n",
              "      <th>tempo_24</th>\n",
              "      <th>tempo_48</th>\n",
              "      <th>tempo_72</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>id_000644bb2</td>\n",
              "      <td>b68db1d53</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>-0.0326</td>\n",
              "      <td>0.5548</td>\n",
              "      <td>-0.0921</td>\n",
              "      <td>1.1830</td>\n",
              "      <td>0.1530</td>\n",
              "      <td>0.5574</td>\n",
              "      <td>-0.4015</td>\n",
              "      <td>0.1789</td>\n",
              "      <td>-0.6528</td>\n",
              "      <td>-0.7969</td>\n",
              "      <td>0.6342</td>\n",
              "      <td>0.1778</td>\n",
              "      <td>-0.3694</td>\n",
              "      <td>-0.5688</td>\n",
              "      <td>-1.1360</td>\n",
              "      <td>-1.1880</td>\n",
              "      <td>0.6940</td>\n",
              "      <td>0.4393</td>\n",
              "      <td>0.2664</td>\n",
              "      <td>0.1907</td>\n",
              "      <td>0.1628</td>\n",
              "      <td>-0.2853</td>\n",
              "      <td>0.5819</td>\n",
              "      <td>0.2934</td>\n",
              "      <td>-0.5584</td>\n",
              "      <td>-0.0916</td>\n",
              "      <td>-0.3010</td>\n",
              "      <td>-0.1537</td>\n",
              "      <td>0.2198</td>\n",
              "      <td>0.2965</td>\n",
              "      <td>-0.5055</td>\n",
              "      <td>...</td>\n",
              "      <td>1.2250</td>\n",
              "      <td>-0.0553</td>\n",
              "      <td>0.7351</td>\n",
              "      <td>0.5810</td>\n",
              "      <td>0.9590</td>\n",
              "      <td>0.2427</td>\n",
              "      <td>0.0495</td>\n",
              "      <td>0.4141</td>\n",
              "      <td>0.8432</td>\n",
              "      <td>0.6162</td>\n",
              "      <td>-0.7318</td>\n",
              "      <td>1.2120</td>\n",
              "      <td>0.6362</td>\n",
              "      <td>-0.4427</td>\n",
              "      <td>0.1288</td>\n",
              "      <td>1.4840</td>\n",
              "      <td>0.1799</td>\n",
              "      <td>0.5367</td>\n",
              "      <td>-0.1111</td>\n",
              "      <td>-1.0120</td>\n",
              "      <td>0.6685</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",
              "      <td>1</td>\n",
              "      <td>True</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>id_000779bfc</td>\n",
              "      <td>df89a8e5a</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>0.3372</td>\n",
              "      <td>-0.4047</td>\n",
              "      <td>0.8507</td>\n",
              "      <td>-1.1520</td>\n",
              "      <td>-0.4201</td>\n",
              "      <td>-0.0958</td>\n",
              "      <td>0.4590</td>\n",
              "      <td>0.0803</td>\n",
              "      <td>0.2250</td>\n",
              "      <td>0.5293</td>\n",
              "      <td>0.2839</td>\n",
              "      <td>-0.3494</td>\n",
              "      <td>0.2883</td>\n",
              "      <td>0.9449</td>\n",
              "      <td>-0.1646</td>\n",
              "      <td>-0.2657</td>\n",
              "      <td>-0.3372</td>\n",
              "      <td>0.3135</td>\n",
              "      <td>-0.4316</td>\n",
              "      <td>0.4773</td>\n",
              "      <td>0.2075</td>\n",
              "      <td>-0.4216</td>\n",
              "      <td>-0.1161</td>\n",
              "      <td>-0.0499</td>\n",
              "      <td>-0.2627</td>\n",
              "      <td>0.9959</td>\n",
              "      <td>-0.2483</td>\n",
              "      <td>0.2655</td>\n",
              "      <td>-0.2102</td>\n",
              "      <td>0.1656</td>\n",
              "      <td>0.5300</td>\n",
              "      <td>...</td>\n",
              "      <td>0.2887</td>\n",
              "      <td>-0.0765</td>\n",
              "      <td>0.2539</td>\n",
              "      <td>0.4443</td>\n",
              "      <td>0.5932</td>\n",
              "      <td>0.2031</td>\n",
              "      <td>0.7639</td>\n",
              "      <td>0.5499</td>\n",
              "      <td>-0.3322</td>\n",
              "      <td>-0.0977</td>\n",
              "      <td>0.4329</td>\n",
              "      <td>-0.2782</td>\n",
              "      <td>0.7827</td>\n",
              "      <td>0.5934</td>\n",
              "      <td>0.3402</td>\n",
              "      <td>0.1499</td>\n",
              "      <td>0.4420</td>\n",
              "      <td>0.9366</td>\n",
              "      <td>0.8193</td>\n",
              "      <td>-0.4236</td>\n",
              "      <td>0.3192</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",
              "      <td>0</td>\n",
              "      <td>False</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>id_000a6266a</td>\n",
              "      <td>18bb41b2c</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>0.2155</td>\n",
              "      <td>0.0065</td>\n",
              "      <td>1.2300</td>\n",
              "      <td>-0.4797</td>\n",
              "      <td>-0.5631</td>\n",
              "      <td>-0.0366</td>\n",
              "      <td>-1.8300</td>\n",
              "      <td>0.6057</td>\n",
              "      <td>-0.3278</td>\n",
              "      <td>0.6042</td>\n",
              "      <td>-0.3075</td>\n",
              "      <td>-0.1147</td>\n",
              "      <td>-0.0570</td>\n",
              "      <td>-0.0799</td>\n",
              "      <td>-0.8181</td>\n",
              "      <td>-1.5320</td>\n",
              "      <td>0.2307</td>\n",
              "      <td>0.4901</td>\n",
              "      <td>0.4780</td>\n",
              "      <td>-1.3970</td>\n",
              "      <td>4.6240</td>\n",
              "      <td>-0.0437</td>\n",
              "      <td>1.2870</td>\n",
              "      <td>-1.8530</td>\n",
              "      <td>0.6069</td>\n",
              "      <td>0.4290</td>\n",
              "      <td>0.1783</td>\n",
              "      <td>0.0018</td>\n",
              "      <td>-1.1800</td>\n",
              "      <td>0.1256</td>\n",
              "      <td>-0.1219</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.0604</td>\n",
              "      <td>0.1682</td>\n",
              "      <td>-0.4436</td>\n",
              "      <td>0.4963</td>\n",
              "      <td>0.1363</td>\n",
              "      <td>0.3335</td>\n",
              "      <td>0.9760</td>\n",
              "      <td>-0.0427</td>\n",
              "      <td>-0.1235</td>\n",
              "      <td>0.0959</td>\n",
              "      <td>0.0690</td>\n",
              "      <td>-0.9416</td>\n",
              "      <td>-0.7548</td>\n",
              "      <td>-0.1109</td>\n",
              "      <td>-0.6272</td>\n",
              "      <td>0.3019</td>\n",
              "      <td>0.1172</td>\n",
              "      <td>0.1093</td>\n",
              "      <td>-0.3113</td>\n",
              "      <td>0.3019</td>\n",
              "      <td>-0.0873</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",
              "      <td>3</td>\n",
              "      <td>True</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>id_0015fd391</td>\n",
              "      <td>8c7f86626</td>\n",
              "      <td>-0.5138</td>\n",
              "      <td>-0.2491</td>\n",
              "      <td>-0.2656</td>\n",
              "      <td>0.5288</td>\n",
              "      <td>4.0620</td>\n",
              "      <td>-0.8095</td>\n",
              "      <td>-1.9590</td>\n",
              "      <td>0.1792</td>\n",
              "      <td>-0.1321</td>\n",
              "      <td>-1.0600</td>\n",
              "      <td>-0.8269</td>\n",
              "      <td>-0.3584</td>\n",
              "      <td>-0.8511</td>\n",
              "      <td>-0.5844</td>\n",
              "      <td>-2.5690</td>\n",
              "      <td>0.8183</td>\n",
              "      <td>-0.0532</td>\n",
              "      <td>-0.8554</td>\n",
              "      <td>0.1160</td>\n",
              "      <td>-2.3520</td>\n",
              "      <td>2.1200</td>\n",
              "      <td>-1.1580</td>\n",
              "      <td>-0.7191</td>\n",
              "      <td>-0.8004</td>\n",
              "      <td>-1.4670</td>\n",
              "      <td>-0.0107</td>\n",
              "      <td>-0.8995</td>\n",
              "      <td>0.2406</td>\n",
              "      <td>-0.2479</td>\n",
              "      <td>-1.0890</td>\n",
              "      <td>-0.7575</td>\n",
              "      <td>0.0881</td>\n",
              "      <td>-2.7370</td>\n",
              "      <td>0.8745</td>\n",
              "      <td>0.5787</td>\n",
              "      <td>-1.6740</td>\n",
              "      <td>-1.6720</td>\n",
              "      <td>-1.2690</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.2990</td>\n",
              "      <td>-1.1870</td>\n",
              "      <td>-0.5044</td>\n",
              "      <td>-1.7750</td>\n",
              "      <td>-1.6120</td>\n",
              "      <td>-0.9215</td>\n",
              "      <td>-1.0810</td>\n",
              "      <td>-3.0520</td>\n",
              "      <td>-3.4470</td>\n",
              "      <td>-2.7740</td>\n",
              "      <td>-1.8460</td>\n",
              "      <td>-0.5568</td>\n",
              "      <td>-3.3960</td>\n",
              "      <td>-2.9510</td>\n",
              "      <td>-1.1550</td>\n",
              "      <td>-3.2620</td>\n",
              "      <td>-1.5390</td>\n",
              "      <td>-2.4600</td>\n",
              "      <td>-0.9417</td>\n",
              "      <td>-1.5550</td>\n",
              "      <td>0.2431</td>\n",
              "      <td>-2.0990</td>\n",
              "      <td>-0.6441</td>\n",
              "      <td>-5.6300</td>\n",
              "      <td>-1.3780</td>\n",
              "      <td>-0.8632</td>\n",
              "      <td>-1.2880</td>\n",
              "      <td>-1.6210</td>\n",
              "      <td>-0.8784</td>\n",
              "      <td>-0.3876</td>\n",
              "      <td>-0.8154</td>\n",
              "      <td>0</td>\n",
              "      <td>False</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>id_001626bd3</td>\n",
              "      <td>7cbed3131</td>\n",
              "      <td>-0.3254</td>\n",
              "      <td>-0.4009</td>\n",
              "      <td>0.9700</td>\n",
              "      <td>0.6919</td>\n",
              "      <td>1.4180</td>\n",
              "      <td>-0.8244</td>\n",
              "      <td>-0.2800</td>\n",
              "      <td>-0.1498</td>\n",
              "      <td>-0.8789</td>\n",
              "      <td>0.8630</td>\n",
              "      <td>-0.2219</td>\n",
              "      <td>-0.5121</td>\n",
              "      <td>-0.9577</td>\n",
              "      <td>1.1750</td>\n",
              "      <td>0.2042</td>\n",
              "      <td>0.1970</td>\n",
              "      <td>0.1244</td>\n",
              "      <td>-1.7090</td>\n",
              "      <td>-0.3543</td>\n",
              "      <td>-0.5160</td>\n",
              "      <td>-0.3330</td>\n",
              "      <td>-0.2685</td>\n",
              "      <td>0.7649</td>\n",
              "      <td>0.2057</td>\n",
              "      <td>1.3720</td>\n",
              "      <td>0.6835</td>\n",
              "      <td>0.8056</td>\n",
              "      <td>-0.3754</td>\n",
              "      <td>-1.2090</td>\n",
              "      <td>0.2965</td>\n",
              "      <td>-0.0712</td>\n",
              "      <td>0.6389</td>\n",
              "      <td>0.6674</td>\n",
              "      <td>-0.0783</td>\n",
              "      <td>1.1740</td>\n",
              "      <td>-0.7110</td>\n",
              "      <td>-1.4470</td>\n",
              "      <td>1.0620</td>\n",
              "      <td>...</td>\n",
              "      <td>0.6091</td>\n",
              "      <td>0.1813</td>\n",
              "      <td>-0.4249</td>\n",
              "      <td>0.7832</td>\n",
              "      <td>0.6529</td>\n",
              "      <td>0.5648</td>\n",
              "      <td>0.4817</td>\n",
              "      <td>0.0587</td>\n",
              "      <td>0.5303</td>\n",
              "      <td>0.6376</td>\n",
              "      <td>-0.3966</td>\n",
              "      <td>-1.4950</td>\n",
              "      <td>-0.9625</td>\n",
              "      <td>-0.0541</td>\n",
              "      <td>0.6273</td>\n",
              "      <td>0.4563</td>\n",
              "      <td>0.0698</td>\n",
              "      <td>0.8134</td>\n",
              "      <td>0.1924</td>\n",
              "      <td>0.6054</td>\n",
              "      <td>-0.1824</td>\n",
              "      <td>0.0042</td>\n",
              "      <td>0.0048</td>\n",
              "      <td>0.6670</td>\n",
              "      <td>1.0690</td>\n",
              "      <td>0.5523</td>\n",
              "      <td>-0.3031</td>\n",
              "      <td>0.1094</td>\n",
              "      <td>0.2885</td>\n",
              "      <td>-0.3786</td>\n",
              "      <td>0.7125</td>\n",
              "      <td>1</td>\n",
              "      <td>True</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23809</th>\n",
              "      <td>id_fffb1ceed</td>\n",
              "      <td>df1d0a5a1</td>\n",
              "      <td>0.1394</td>\n",
              "      <td>-0.0636</td>\n",
              "      <td>-0.1112</td>\n",
              "      <td>-0.5080</td>\n",
              "      <td>-0.4713</td>\n",
              "      <td>0.7201</td>\n",
              "      <td>0.5773</td>\n",
              "      <td>0.3055</td>\n",
              "      <td>-0.4726</td>\n",
              "      <td>0.1269</td>\n",
              "      <td>0.2531</td>\n",
              "      <td>0.1730</td>\n",
              "      <td>-0.4532</td>\n",
              "      <td>-1.0790</td>\n",
              "      <td>0.2474</td>\n",
              "      <td>-0.4550</td>\n",
              "      <td>0.3588</td>\n",
              "      <td>0.1600</td>\n",
              "      <td>-0.7362</td>\n",
              "      <td>-0.1103</td>\n",
              "      <td>0.8550</td>\n",
              "      <td>-0.4139</td>\n",
              "      <td>0.5541</td>\n",
              "      <td>0.2310</td>\n",
              "      <td>-0.5573</td>\n",
              "      <td>-0.4397</td>\n",
              "      <td>-0.9260</td>\n",
              "      <td>-0.2424</td>\n",
              "      <td>-0.6686</td>\n",
              "      <td>0.2326</td>\n",
              "      <td>0.6456</td>\n",
              "      <td>0.0136</td>\n",
              "      <td>-0.5141</td>\n",
              "      <td>-0.6320</td>\n",
              "      <td>0.7166</td>\n",
              "      <td>-0.1736</td>\n",
              "      <td>0.3686</td>\n",
              "      <td>-0.1565</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.6158</td>\n",
              "      <td>-0.2977</td>\n",
              "      <td>0.0992</td>\n",
              "      <td>0.6838</td>\n",
              "      <td>0.5259</td>\n",
              "      <td>0.7882</td>\n",
              "      <td>0.3119</td>\n",
              "      <td>-0.7697</td>\n",
              "      <td>0.2203</td>\n",
              "      <td>-1.0710</td>\n",
              "      <td>0.5979</td>\n",
              "      <td>0.0848</td>\n",
              "      <td>-0.2555</td>\n",
              "      <td>0.6293</td>\n",
              "      <td>1.1660</td>\n",
              "      <td>0.3329</td>\n",
              "      <td>0.2754</td>\n",
              "      <td>0.4108</td>\n",
              "      <td>-0.1252</td>\n",
              "      <td>-0.2340</td>\n",
              "      <td>0.2267</td>\n",
              "      <td>0.1969</td>\n",
              "      <td>0.0262</td>\n",
              "      <td>-0.8121</td>\n",
              "      <td>0.3434</td>\n",
              "      <td>0.5372</td>\n",
              "      <td>-0.3246</td>\n",
              "      <td>0.0631</td>\n",
              "      <td>0.9171</td>\n",
              "      <td>0.5258</td>\n",
              "      <td>0.4680</td>\n",
              "      <td>1</td>\n",
              "      <td>True</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23810</th>\n",
              "      <td>id_fffb70c0c</td>\n",
              "      <td>ecf3b6b74</td>\n",
              "      <td>-1.3260</td>\n",
              "      <td>0.3478</td>\n",
              "      <td>-0.3743</td>\n",
              "      <td>0.9905</td>\n",
              "      <td>-0.7178</td>\n",
              "      <td>0.6621</td>\n",
              "      <td>-0.2252</td>\n",
              "      <td>-0.5565</td>\n",
              "      <td>0.5112</td>\n",
              "      <td>0.6727</td>\n",
              "      <td>-0.1851</td>\n",
              "      <td>2.8650</td>\n",
              "      <td>-0.2140</td>\n",
              "      <td>-0.6153</td>\n",
              "      <td>0.8362</td>\n",
              "      <td>0.5584</td>\n",
              "      <td>-0.2589</td>\n",
              "      <td>0.1292</td>\n",
              "      <td>0.0148</td>\n",
              "      <td>0.0949</td>\n",
              "      <td>-0.2182</td>\n",
              "      <td>-0.9235</td>\n",
              "      <td>0.0749</td>\n",
              "      <td>-1.5910</td>\n",
              "      <td>-0.8359</td>\n",
              "      <td>-0.9217</td>\n",
              "      <td>0.3013</td>\n",
              "      <td>0.1716</td>\n",
              "      <td>0.0880</td>\n",
              "      <td>0.1842</td>\n",
              "      <td>0.1835</td>\n",
              "      <td>0.5436</td>\n",
              "      <td>-0.0533</td>\n",
              "      <td>-0.0491</td>\n",
              "      <td>0.9543</td>\n",
              "      <td>0.4626</td>\n",
              "      <td>0.0819</td>\n",
              "      <td>0.1586</td>\n",
              "      <td>...</td>\n",
              "      <td>1.0570</td>\n",
              "      <td>-0.3355</td>\n",
              "      <td>-0.2555</td>\n",
              "      <td>0.8219</td>\n",
              "      <td>0.6995</td>\n",
              "      <td>-0.2942</td>\n",
              "      <td>0.2408</td>\n",
              "      <td>-0.7781</td>\n",
              "      <td>-0.0929</td>\n",
              "      <td>-0.0329</td>\n",
              "      <td>0.0781</td>\n",
              "      <td>-1.4440</td>\n",
              "      <td>-0.0921</td>\n",
              "      <td>0.3188</td>\n",
              "      <td>-1.1080</td>\n",
              "      <td>0.4895</td>\n",
              "      <td>-0.2144</td>\n",
              "      <td>1.0960</td>\n",
              "      <td>0.3335</td>\n",
              "      <td>0.4444</td>\n",
              "      <td>-1.1130</td>\n",
              "      <td>0.4286</td>\n",
              "      <td>0.4426</td>\n",
              "      <td>0.0423</td>\n",
              "      <td>-0.3195</td>\n",
              "      <td>-0.8086</td>\n",
              "      <td>-0.9798</td>\n",
              "      <td>-0.2084</td>\n",
              "      <td>-0.1224</td>\n",
              "      <td>-0.2715</td>\n",
              "      <td>0.3689</td>\n",
              "      <td>1</td>\n",
              "      <td>True</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23811</th>\n",
              "      <td>id_fffc1c3f4</td>\n",
              "      <td>cacb2b860</td>\n",
              "      <td>0.3942</td>\n",
              "      <td>0.3756</td>\n",
              "      <td>0.3109</td>\n",
              "      <td>-0.7389</td>\n",
              "      <td>0.5505</td>\n",
              "      <td>-0.0159</td>\n",
              "      <td>-0.2541</td>\n",
              "      <td>0.1745</td>\n",
              "      <td>-0.0340</td>\n",
              "      <td>0.4865</td>\n",
              "      <td>-0.1854</td>\n",
              "      <td>0.0716</td>\n",
              "      <td>0.1729</td>\n",
              "      <td>-0.0434</td>\n",
              "      <td>0.1542</td>\n",
              "      <td>-0.2192</td>\n",
              "      <td>-0.0302</td>\n",
              "      <td>-0.4218</td>\n",
              "      <td>0.4057</td>\n",
              "      <td>-0.5372</td>\n",
              "      <td>0.1521</td>\n",
              "      <td>-0.2651</td>\n",
              "      <td>0.2310</td>\n",
              "      <td>-0.8101</td>\n",
              "      <td>0.4943</td>\n",
              "      <td>0.6905</td>\n",
              "      <td>-0.3720</td>\n",
              "      <td>-1.4110</td>\n",
              "      <td>0.4516</td>\n",
              "      <td>1.2300</td>\n",
              "      <td>-0.1949</td>\n",
              "      <td>-1.3280</td>\n",
              "      <td>-0.4276</td>\n",
              "      <td>-0.0040</td>\n",
              "      <td>-0.3086</td>\n",
              "      <td>-0.2355</td>\n",
              "      <td>-0.5222</td>\n",
              "      <td>-1.1470</td>\n",
              "      <td>...</td>\n",
              "      <td>1.4690</td>\n",
              "      <td>-0.6142</td>\n",
              "      <td>0.6068</td>\n",
              "      <td>0.3434</td>\n",
              "      <td>0.9880</td>\n",
              "      <td>-0.0468</td>\n",
              "      <td>-0.1882</td>\n",
              "      <td>-0.0087</td>\n",
              "      <td>-0.0356</td>\n",
              "      <td>0.5718</td>\n",
              "      <td>0.4971</td>\n",
              "      <td>0.1057</td>\n",
              "      <td>0.2111</td>\n",
              "      <td>0.6992</td>\n",
              "      <td>0.0708</td>\n",
              "      <td>0.6169</td>\n",
              "      <td>0.2248</td>\n",
              "      <td>0.5994</td>\n",
              "      <td>0.2689</td>\n",
              "      <td>0.0305</td>\n",
              "      <td>1.2320</td>\n",
              "      <td>0.5409</td>\n",
              "      <td>0.3755</td>\n",
              "      <td>0.7343</td>\n",
              "      <td>0.2807</td>\n",
              "      <td>0.4116</td>\n",
              "      <td>0.6422</td>\n",
              "      <td>0.2256</td>\n",
              "      <td>0.7592</td>\n",
              "      <td>0.6656</td>\n",
              "      <td>0.3808</td>\n",
              "      <td>0</td>\n",
              "      <td>False</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23812</th>\n",
              "      <td>id_fffcb9e7c</td>\n",
              "      <td>8b87a7a83</td>\n",
              "      <td>0.6660</td>\n",
              "      <td>0.2324</td>\n",
              "      <td>0.4392</td>\n",
              "      <td>0.2044</td>\n",
              "      <td>0.8531</td>\n",
              "      <td>-0.0343</td>\n",
              "      <td>0.0323</td>\n",
              "      <td>0.0463</td>\n",
              "      <td>0.4299</td>\n",
              "      <td>-0.7985</td>\n",
              "      <td>0.5742</td>\n",
              "      <td>0.1421</td>\n",
              "      <td>2.2700</td>\n",
              "      <td>0.2046</td>\n",
              "      <td>0.5363</td>\n",
              "      <td>-1.7330</td>\n",
              "      <td>0.1450</td>\n",
              "      <td>0.6097</td>\n",
              "      <td>0.2024</td>\n",
              "      <td>0.9865</td>\n",
              "      <td>-0.7805</td>\n",
              "      <td>0.9608</td>\n",
              "      <td>0.3440</td>\n",
              "      <td>2.7650</td>\n",
              "      <td>0.4925</td>\n",
              "      <td>0.6698</td>\n",
              "      <td>0.2374</td>\n",
              "      <td>-0.3372</td>\n",
              "      <td>0.8771</td>\n",
              "      <td>-2.6560</td>\n",
              "      <td>-0.2000</td>\n",
              "      <td>-0.2043</td>\n",
              "      <td>0.6797</td>\n",
              "      <td>-0.0248</td>\n",
              "      <td>-0.0927</td>\n",
              "      <td>1.8480</td>\n",
              "      <td>-2.7180</td>\n",
              "      <td>-5.3160</td>\n",
              "      <td>...</td>\n",
              "      <td>0.4524</td>\n",
              "      <td>0.4135</td>\n",
              "      <td>0.1356</td>\n",
              "      <td>-0.5801</td>\n",
              "      <td>0.0411</td>\n",
              "      <td>1.0240</td>\n",
              "      <td>1.0340</td>\n",
              "      <td>-0.0270</td>\n",
              "      <td>-0.4194</td>\n",
              "      <td>0.7403</td>\n",
              "      <td>-0.6793</td>\n",
              "      <td>-0.1423</td>\n",
              "      <td>0.7307</td>\n",
              "      <td>0.7946</td>\n",
              "      <td>-0.0650</td>\n",
              "      <td>0.9038</td>\n",
              "      <td>0.2324</td>\n",
              "      <td>0.9676</td>\n",
              "      <td>1.0940</td>\n",
              "      <td>0.4290</td>\n",
              "      <td>0.5187</td>\n",
              "      <td>-0.1105</td>\n",
              "      <td>0.4258</td>\n",
              "      <td>-0.2012</td>\n",
              "      <td>0.1506</td>\n",
              "      <td>1.5230</td>\n",
              "      <td>0.7101</td>\n",
              "      <td>0.1732</td>\n",
              "      <td>0.7015</td>\n",
              "      <td>-0.6290</td>\n",
              "      <td>0.0740</td>\n",
              "      <td>1</td>\n",
              "      <td>True</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23813</th>\n",
              "      <td>id_ffffdd77b</td>\n",
              "      <td>972f41291</td>\n",
              "      <td>-0.8598</td>\n",
              "      <td>1.0240</td>\n",
              "      <td>-0.1361</td>\n",
              "      <td>0.7952</td>\n",
              "      <td>-0.3611</td>\n",
              "      <td>-3.6750</td>\n",
              "      <td>-1.2420</td>\n",
              "      <td>0.9146</td>\n",
              "      <td>3.0790</td>\n",
              "      <td>1.2460</td>\n",
              "      <td>1.9460</td>\n",
              "      <td>1.4370</td>\n",
              "      <td>2.9780</td>\n",
              "      <td>2.2370</td>\n",
              "      <td>-0.6818</td>\n",
              "      <td>0.6870</td>\n",
              "      <td>-1.1060</td>\n",
              "      <td>0.0182</td>\n",
              "      <td>-0.9247</td>\n",
              "      <td>-0.0738</td>\n",
              "      <td>-0.1919</td>\n",
              "      <td>-0.7722</td>\n",
              "      <td>-1.4050</td>\n",
              "      <td>-1.0050</td>\n",
              "      <td>-1.1170</td>\n",
              "      <td>-0.5293</td>\n",
              "      <td>-1.1720</td>\n",
              "      <td>-0.2885</td>\n",
              "      <td>0.1599</td>\n",
              "      <td>-0.4250</td>\n",
              "      <td>0.3591</td>\n",
              "      <td>-0.1420</td>\n",
              "      <td>-0.9530</td>\n",
              "      <td>-0.2005</td>\n",
              "      <td>-1.8340</td>\n",
              "      <td>-0.5768</td>\n",
              "      <td>0.5145</td>\n",
              "      <td>-0.9913</td>\n",
              "      <td>...</td>\n",
              "      <td>-1.1690</td>\n",
              "      <td>-4.1460</td>\n",
              "      <td>-1.2670</td>\n",
              "      <td>-1.1300</td>\n",
              "      <td>-2.4390</td>\n",
              "      <td>0.1591</td>\n",
              "      <td>-2.2490</td>\n",
              "      <td>-2.5860</td>\n",
              "      <td>-1.9520</td>\n",
              "      <td>-2.1810</td>\n",
              "      <td>-4.6690</td>\n",
              "      <td>-3.9450</td>\n",
              "      <td>-2.9560</td>\n",
              "      <td>-2.7930</td>\n",
              "      <td>-2.1560</td>\n",
              "      <td>-2.4100</td>\n",
              "      <td>-1.8190</td>\n",
              "      <td>-3.3480</td>\n",
              "      <td>-0.1414</td>\n",
              "      <td>-2.6430</td>\n",
              "      <td>-2.5810</td>\n",
              "      <td>-3.3890</td>\n",
              "      <td>-1.7450</td>\n",
              "      <td>-6.6300</td>\n",
              "      <td>-4.0950</td>\n",
              "      <td>-7.3860</td>\n",
              "      <td>-1.4160</td>\n",
              "      <td>-3.5770</td>\n",
              "      <td>-0.4775</td>\n",
              "      <td>-2.1500</td>\n",
              "      <td>-4.2520</td>\n",
              "      <td>0</td>\n",
              "      <td>False</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>23814 rows × 883 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                 id   composto     g-0  ...  tempo_24  tempo_48  tempo_72\n",
              "0      id_000644bb2  b68db1d53  1.0620  ...         1         0         0\n",
              "1      id_000779bfc  df89a8e5a  0.0743  ...         0         0         1\n",
              "2      id_000a6266a  18bb41b2c  0.6280  ...         0         1         0\n",
              "3      id_0015fd391  8c7f86626 -0.5138  ...         0         1         0\n",
              "4      id_001626bd3  7cbed3131 -0.3254  ...         0         0         1\n",
              "...             ...        ...     ...  ...       ...       ...       ...\n",
              "23809  id_fffb1ceed  df1d0a5a1  0.1394  ...         1         0         0\n",
              "23810  id_fffb70c0c  ecf3b6b74 -1.3260  ...         1         0         0\n",
              "23811  id_fffc1c3f4  cacb2b860  0.3942  ...         0         1         0\n",
              "23812  id_fffcb9e7c  8b87a7a83  0.6660  ...         1         0         0\n",
              "23813  id_ffffdd77b  972f41291 -0.8598  ...         0         0         1\n",
              "\n",
              "[23814 rows x 883 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "l_eaIzNtVAv0"
      },
      "source": [
        "Aqui, estamos estabelecendo qual será a base de dados usada para modelar o novo algoritmo. Agora, não mais consideraremos apenas os dados genéticos, também levaremos em conta as demais variáveis, exceto as seguintes variáveis:\n",
        "\n",
        "    'id'  ->  não será relevante nesse momento pois é somente um número de identificação\n",
        "    'n_moa'  ->  não será relevante nesse momento pois é a quantidade de MoAs ativados\n",
        "    'ativo_moa'  ->  retiramos da base pois é a nossa variável resposta\n",
        "    'composto'  ->  não será relevante nesse momento pois identifica o composto usado e foge do que estamos buscando fazer\n",
        "\n",
        "Para retirar essas variáveis da nossa base de dados, usamos a função ```drop```, declaramos as variáveis a serem retiradas em forma de uma lista e identificamos que são colunas através do parâmetro ```axis = 1```."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZWczCFua9ZYb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 435
        },
        "outputId": "b8c41346-cef2-4757-ad76-7a24ae7a40fd"
      },
      "source": [
        "dados_combinados.drop(['id', 'n_moa', 'ativo_moa', 'composto'], axis=1)"
      ],
      "execution_count": 22,
      "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>tratamento</th>\n",
              "      <th>tempo</th>\n",
              "      <th>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>g-7</th>\n",
              "      <th>g-8</th>\n",
              "      <th>g-9</th>\n",
              "      <th>g-10</th>\n",
              "      <th>g-11</th>\n",
              "      <th>g-12</th>\n",
              "      <th>g-13</th>\n",
              "      <th>g-14</th>\n",
              "      <th>g-15</th>\n",
              "      <th>g-16</th>\n",
              "      <th>g-17</th>\n",
              "      <th>g-18</th>\n",
              "      <th>g-19</th>\n",
              "      <th>g-20</th>\n",
              "      <th>g-21</th>\n",
              "      <th>g-22</th>\n",
              "      <th>g-23</th>\n",
              "      <th>g-24</th>\n",
              "      <th>g-25</th>\n",
              "      <th>g-26</th>\n",
              "      <th>g-27</th>\n",
              "      <th>g-28</th>\n",
              "      <th>g-29</th>\n",
              "      <th>g-30</th>\n",
              "      <th>g-31</th>\n",
              "      <th>g-32</th>\n",
              "      <th>g-33</th>\n",
              "      <th>g-34</th>\n",
              "      <th>g-35</th>\n",
              "      <th>g-36</th>\n",
              "      <th>...</th>\n",
              "      <th>c-60</th>\n",
              "      <th>c-61</th>\n",
              "      <th>c-62</th>\n",
              "      <th>c-63</th>\n",
              "      <th>c-64</th>\n",
              "      <th>c-65</th>\n",
              "      <th>c-66</th>\n",
              "      <th>c-67</th>\n",
              "      <th>c-68</th>\n",
              "      <th>c-69</th>\n",
              "      <th>c-70</th>\n",
              "      <th>c-71</th>\n",
              "      <th>c-72</th>\n",
              "      <th>c-73</th>\n",
              "      <th>c-74</th>\n",
              "      <th>c-75</th>\n",
              "      <th>c-76</th>\n",
              "      <th>c-77</th>\n",
              "      <th>c-78</th>\n",
              "      <th>c-79</th>\n",
              "      <th>c-80</th>\n",
              "      <th>c-81</th>\n",
              "      <th>c-82</th>\n",
              "      <th>c-83</th>\n",
              "      <th>c-84</th>\n",
              "      <th>c-85</th>\n",
              "      <th>c-86</th>\n",
              "      <th>c-87</th>\n",
              "      <th>c-88</th>\n",
              "      <th>c-89</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>com_droga</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>-1.0220</td>\n",
              "      <td>-0.0326</td>\n",
              "      <td>0.5548</td>\n",
              "      <td>-0.0921</td>\n",
              "      <td>1.1830</td>\n",
              "      <td>0.1530</td>\n",
              "      <td>0.5574</td>\n",
              "      <td>-0.4015</td>\n",
              "      <td>0.1789</td>\n",
              "      <td>-0.6528</td>\n",
              "      <td>-0.7969</td>\n",
              "      <td>0.6342</td>\n",
              "      <td>0.1778</td>\n",
              "      <td>-0.3694</td>\n",
              "      <td>-0.5688</td>\n",
              "      <td>-1.1360</td>\n",
              "      <td>-1.1880</td>\n",
              "      <td>0.6940</td>\n",
              "      <td>0.4393</td>\n",
              "      <td>0.2664</td>\n",
              "      <td>0.1907</td>\n",
              "      <td>0.1628</td>\n",
              "      <td>-0.2853</td>\n",
              "      <td>0.5819</td>\n",
              "      <td>0.2934</td>\n",
              "      <td>-0.5584</td>\n",
              "      <td>-0.0916</td>\n",
              "      <td>-0.3010</td>\n",
              "      <td>-0.1537</td>\n",
              "      <td>0.2198</td>\n",
              "      <td>0.2965</td>\n",
              "      <td>...</td>\n",
              "      <td>0.4805</td>\n",
              "      <td>0.4965</td>\n",
              "      <td>0.3680</td>\n",
              "      <td>0.8427</td>\n",
              "      <td>0.1042</td>\n",
              "      <td>0.1403</td>\n",
              "      <td>0.1758</td>\n",
              "      <td>1.2570</td>\n",
              "      <td>-0.5979</td>\n",
              "      <td>1.2250</td>\n",
              "      <td>-0.0553</td>\n",
              "      <td>0.7351</td>\n",
              "      <td>0.5810</td>\n",
              "      <td>0.9590</td>\n",
              "      <td>0.2427</td>\n",
              "      <td>0.0495</td>\n",
              "      <td>0.4141</td>\n",
              "      <td>0.8432</td>\n",
              "      <td>0.6162</td>\n",
              "      <td>-0.7318</td>\n",
              "      <td>1.2120</td>\n",
              "      <td>0.6362</td>\n",
              "      <td>-0.4427</td>\n",
              "      <td>0.1288</td>\n",
              "      <td>1.4840</td>\n",
              "      <td>0.1799</td>\n",
              "      <td>0.5367</td>\n",
              "      <td>-0.1111</td>\n",
              "      <td>-1.0120</td>\n",
              "      <td>0.6685</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>com_droga</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>0.2341</td>\n",
              "      <td>0.3372</td>\n",
              "      <td>-0.4047</td>\n",
              "      <td>0.8507</td>\n",
              "      <td>-1.1520</td>\n",
              "      <td>-0.4201</td>\n",
              "      <td>-0.0958</td>\n",
              "      <td>0.4590</td>\n",
              "      <td>0.0803</td>\n",
              "      <td>0.2250</td>\n",
              "      <td>0.5293</td>\n",
              "      <td>0.2839</td>\n",
              "      <td>-0.3494</td>\n",
              "      <td>0.2883</td>\n",
              "      <td>0.9449</td>\n",
              "      <td>-0.1646</td>\n",
              "      <td>-0.2657</td>\n",
              "      <td>-0.3372</td>\n",
              "      <td>0.3135</td>\n",
              "      <td>-0.4316</td>\n",
              "      <td>0.4773</td>\n",
              "      <td>0.2075</td>\n",
              "      <td>-0.4216</td>\n",
              "      <td>-0.1161</td>\n",
              "      <td>-0.0499</td>\n",
              "      <td>-0.2627</td>\n",
              "      <td>0.9959</td>\n",
              "      <td>-0.2483</td>\n",
              "      <td>0.2655</td>\n",
              "      <td>-0.2102</td>\n",
              "      <td>0.1656</td>\n",
              "      <td>...</td>\n",
              "      <td>0.4083</td>\n",
              "      <td>0.0319</td>\n",
              "      <td>0.3905</td>\n",
              "      <td>0.7099</td>\n",
              "      <td>0.2912</td>\n",
              "      <td>0.4151</td>\n",
              "      <td>-0.2840</td>\n",
              "      <td>-0.3104</td>\n",
              "      <td>-0.6373</td>\n",
              "      <td>0.2887</td>\n",
              "      <td>-0.0765</td>\n",
              "      <td>0.2539</td>\n",
              "      <td>0.4443</td>\n",
              "      <td>0.5932</td>\n",
              "      <td>0.2031</td>\n",
              "      <td>0.7639</td>\n",
              "      <td>0.5499</td>\n",
              "      <td>-0.3322</td>\n",
              "      <td>-0.0977</td>\n",
              "      <td>0.4329</td>\n",
              "      <td>-0.2782</td>\n",
              "      <td>0.7827</td>\n",
              "      <td>0.5934</td>\n",
              "      <td>0.3402</td>\n",
              "      <td>0.1499</td>\n",
              "      <td>0.4420</td>\n",
              "      <td>0.9366</td>\n",
              "      <td>0.8193</td>\n",
              "      <td>-0.4236</td>\n",
              "      <td>0.3192</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>com_droga</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>0.1715</td>\n",
              "      <td>0.2155</td>\n",
              "      <td>0.0065</td>\n",
              "      <td>1.2300</td>\n",
              "      <td>-0.4797</td>\n",
              "      <td>-0.5631</td>\n",
              "      <td>-0.0366</td>\n",
              "      <td>-1.8300</td>\n",
              "      <td>0.6057</td>\n",
              "      <td>-0.3278</td>\n",
              "      <td>0.6042</td>\n",
              "      <td>-0.3075</td>\n",
              "      <td>-0.1147</td>\n",
              "      <td>-0.0570</td>\n",
              "      <td>-0.0799</td>\n",
              "      <td>-0.8181</td>\n",
              "      <td>-1.5320</td>\n",
              "      <td>0.2307</td>\n",
              "      <td>0.4901</td>\n",
              "      <td>0.4780</td>\n",
              "      <td>-1.3970</td>\n",
              "      <td>4.6240</td>\n",
              "      <td>-0.0437</td>\n",
              "      <td>1.2870</td>\n",
              "      <td>-1.8530</td>\n",
              "      <td>0.6069</td>\n",
              "      <td>0.4290</td>\n",
              "      <td>0.1783</td>\n",
              "      <td>0.0018</td>\n",
              "      <td>-1.1800</td>\n",
              "      <td>0.1256</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.5477</td>\n",
              "      <td>-0.7576</td>\n",
              "      <td>-0.0444</td>\n",
              "      <td>0.1894</td>\n",
              "      <td>-0.0014</td>\n",
              "      <td>-2.3640</td>\n",
              "      <td>-0.4682</td>\n",
              "      <td>0.1210</td>\n",
              "      <td>-0.5177</td>\n",
              "      <td>-0.0604</td>\n",
              "      <td>0.1682</td>\n",
              "      <td>-0.4436</td>\n",
              "      <td>0.4963</td>\n",
              "      <td>0.1363</td>\n",
              "      <td>0.3335</td>\n",
              "      <td>0.9760</td>\n",
              "      <td>-0.0427</td>\n",
              "      <td>-0.1235</td>\n",
              "      <td>0.0959</td>\n",
              "      <td>0.0690</td>\n",
              "      <td>-0.9416</td>\n",
              "      <td>-0.7548</td>\n",
              "      <td>-0.1109</td>\n",
              "      <td>-0.6272</td>\n",
              "      <td>0.3019</td>\n",
              "      <td>0.1172</td>\n",
              "      <td>0.1093</td>\n",
              "      <td>-0.3113</td>\n",
              "      <td>0.3019</td>\n",
              "      <td>-0.0873</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",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>com_droga</td>\n",
              "      <td>48</td>\n",
              "      <td>D1</td>\n",
              "      <td>-0.5138</td>\n",
              "      <td>-0.2491</td>\n",
              "      <td>-0.2656</td>\n",
              "      <td>0.5288</td>\n",
              "      <td>4.0620</td>\n",
              "      <td>-0.8095</td>\n",
              "      <td>-1.9590</td>\n",
              "      <td>0.1792</td>\n",
              "      <td>-0.1321</td>\n",
              "      <td>-1.0600</td>\n",
              "      <td>-0.8269</td>\n",
              "      <td>-0.3584</td>\n",
              "      <td>-0.8511</td>\n",
              "      <td>-0.5844</td>\n",
              "      <td>-2.5690</td>\n",
              "      <td>0.8183</td>\n",
              "      <td>-0.0532</td>\n",
              "      <td>-0.8554</td>\n",
              "      <td>0.1160</td>\n",
              "      <td>-2.3520</td>\n",
              "      <td>2.1200</td>\n",
              "      <td>-1.1580</td>\n",
              "      <td>-0.7191</td>\n",
              "      <td>-0.8004</td>\n",
              "      <td>-1.4670</td>\n",
              "      <td>-0.0107</td>\n",
              "      <td>-0.8995</td>\n",
              "      <td>0.2406</td>\n",
              "      <td>-0.2479</td>\n",
              "      <td>-1.0890</td>\n",
              "      <td>-0.7575</td>\n",
              "      <td>0.0881</td>\n",
              "      <td>-2.7370</td>\n",
              "      <td>0.8745</td>\n",
              "      <td>0.5787</td>\n",
              "      <td>-1.6740</td>\n",
              "      <td>-1.6720</td>\n",
              "      <td>...</td>\n",
              "      <td>-2.1220</td>\n",
              "      <td>-0.3752</td>\n",
              "      <td>-2.3820</td>\n",
              "      <td>-3.7350</td>\n",
              "      <td>-2.9740</td>\n",
              "      <td>-1.4930</td>\n",
              "      <td>-1.6600</td>\n",
              "      <td>-3.1660</td>\n",
              "      <td>0.2816</td>\n",
              "      <td>-0.2990</td>\n",
              "      <td>-1.1870</td>\n",
              "      <td>-0.5044</td>\n",
              "      <td>-1.7750</td>\n",
              "      <td>-1.6120</td>\n",
              "      <td>-0.9215</td>\n",
              "      <td>-1.0810</td>\n",
              "      <td>-3.0520</td>\n",
              "      <td>-3.4470</td>\n",
              "      <td>-2.7740</td>\n",
              "      <td>-1.8460</td>\n",
              "      <td>-0.5568</td>\n",
              "      <td>-3.3960</td>\n",
              "      <td>-2.9510</td>\n",
              "      <td>-1.1550</td>\n",
              "      <td>-3.2620</td>\n",
              "      <td>-1.5390</td>\n",
              "      <td>-2.4600</td>\n",
              "      <td>-0.9417</td>\n",
              "      <td>-1.5550</td>\n",
              "      <td>0.2431</td>\n",
              "      <td>-2.0990</td>\n",
              "      <td>-0.6441</td>\n",
              "      <td>-5.6300</td>\n",
              "      <td>-1.3780</td>\n",
              "      <td>-0.8632</td>\n",
              "      <td>-1.2880</td>\n",
              "      <td>-1.6210</td>\n",
              "      <td>-0.8784</td>\n",
              "      <td>-0.3876</td>\n",
              "      <td>-0.8154</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>com_droga</td>\n",
              "      <td>72</td>\n",
              "      <td>D2</td>\n",
              "      <td>-0.3254</td>\n",
              "      <td>-0.4009</td>\n",
              "      <td>0.9700</td>\n",
              "      <td>0.6919</td>\n",
              "      <td>1.4180</td>\n",
              "      <td>-0.8244</td>\n",
              "      <td>-0.2800</td>\n",
              "      <td>-0.1498</td>\n",
              "      <td>-0.8789</td>\n",
              "      <td>0.8630</td>\n",
              "      <td>-0.2219</td>\n",
              "      <td>-0.5121</td>\n",
              "      <td>-0.9577</td>\n",
              "      <td>1.1750</td>\n",
              "      <td>0.2042</td>\n",
              "      <td>0.1970</td>\n",
              "      <td>0.1244</td>\n",
              "      <td>-1.7090</td>\n",
              "      <td>-0.3543</td>\n",
              "      <td>-0.5160</td>\n",
              "      <td>-0.3330</td>\n",
              "      <td>-0.2685</td>\n",
              "      <td>0.7649</td>\n",
              "      <td>0.2057</td>\n",
              "      <td>1.3720</td>\n",
              "      <td>0.6835</td>\n",
              "      <td>0.8056</td>\n",
              "      <td>-0.3754</td>\n",
              "      <td>-1.2090</td>\n",
              "      <td>0.2965</td>\n",
              "      <td>-0.0712</td>\n",
              "      <td>0.6389</td>\n",
              "      <td>0.6674</td>\n",
              "      <td>-0.0783</td>\n",
              "      <td>1.1740</td>\n",
              "      <td>-0.7110</td>\n",
              "      <td>-1.4470</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.2274</td>\n",
              "      <td>0.3215</td>\n",
              "      <td>0.1535</td>\n",
              "      <td>-0.4640</td>\n",
              "      <td>-0.5943</td>\n",
              "      <td>0.3973</td>\n",
              "      <td>0.1500</td>\n",
              "      <td>0.5178</td>\n",
              "      <td>0.5159</td>\n",
              "      <td>0.6091</td>\n",
              "      <td>0.1813</td>\n",
              "      <td>-0.4249</td>\n",
              "      <td>0.7832</td>\n",
              "      <td>0.6529</td>\n",
              "      <td>0.5648</td>\n",
              "      <td>0.4817</td>\n",
              "      <td>0.0587</td>\n",
              "      <td>0.5303</td>\n",
              "      <td>0.6376</td>\n",
              "      <td>-0.3966</td>\n",
              "      <td>-1.4950</td>\n",
              "      <td>-0.9625</td>\n",
              "      <td>-0.0541</td>\n",
              "      <td>0.6273</td>\n",
              "      <td>0.4563</td>\n",
              "      <td>0.0698</td>\n",
              "      <td>0.8134</td>\n",
              "      <td>0.1924</td>\n",
              "      <td>0.6054</td>\n",
              "      <td>-0.1824</td>\n",
              "      <td>0.0042</td>\n",
              "      <td>0.0048</td>\n",
              "      <td>0.6670</td>\n",
              "      <td>1.0690</td>\n",
              "      <td>0.5523</td>\n",
              "      <td>-0.3031</td>\n",
              "      <td>0.1094</td>\n",
              "      <td>0.2885</td>\n",
              "      <td>-0.3786</td>\n",
              "      <td>0.7125</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23809</th>\n",
              "      <td>com_droga</td>\n",
              "      <td>24</td>\n",
              "      <td>D2</td>\n",
              "      <td>0.1394</td>\n",
              "      <td>-0.0636</td>\n",
              "      <td>-0.1112</td>\n",
              "      <td>-0.5080</td>\n",
              "      <td>-0.4713</td>\n",
              "      <td>0.7201</td>\n",
              "      <td>0.5773</td>\n",
              "      <td>0.3055</td>\n",
              "      <td>-0.4726</td>\n",
              "      <td>0.1269</td>\n",
              "      <td>0.2531</td>\n",
              "      <td>0.1730</td>\n",
              "      <td>-0.4532</td>\n",
              "      <td>-1.0790</td>\n",
              "      <td>0.2474</td>\n",
              "      <td>-0.4550</td>\n",
              "      <td>0.3588</td>\n",
              "      <td>0.1600</td>\n",
              "      <td>-0.7362</td>\n",
              "      <td>-0.1103</td>\n",
              "      <td>0.8550</td>\n",
              "      <td>-0.4139</td>\n",
              "      <td>0.5541</td>\n",
              "      <td>0.2310</td>\n",
              "      <td>-0.5573</td>\n",
              "      <td>-0.4397</td>\n",
              "      <td>-0.9260</td>\n",
              "      <td>-0.2424</td>\n",
              "      <td>-0.6686</td>\n",
              "      <td>0.2326</td>\n",
              "      <td>0.6456</td>\n",
              "      <td>0.0136</td>\n",
              "      <td>-0.5141</td>\n",
              "      <td>-0.6320</td>\n",
              "      <td>0.7166</td>\n",
              "      <td>-0.1736</td>\n",
              "      <td>0.3686</td>\n",
              "      <td>...</td>\n",
              "      <td>0.0807</td>\n",
              "      <td>0.4024</td>\n",
              "      <td>-0.0895</td>\n",
              "      <td>0.4420</td>\n",
              "      <td>0.9641</td>\n",
              "      <td>-0.1846</td>\n",
              "      <td>-0.0014</td>\n",
              "      <td>0.3154</td>\n",
              "      <td>-0.2071</td>\n",
              "      <td>-0.6158</td>\n",
              "      <td>-0.2977</td>\n",
              "      <td>0.0992</td>\n",
              "      <td>0.6838</td>\n",
              "      <td>0.5259</td>\n",
              "      <td>0.7882</td>\n",
              "      <td>0.3119</td>\n",
              "      <td>-0.7697</td>\n",
              "      <td>0.2203</td>\n",
              "      <td>-1.0710</td>\n",
              "      <td>0.5979</td>\n",
              "      <td>0.0848</td>\n",
              "      <td>-0.2555</td>\n",
              "      <td>0.6293</td>\n",
              "      <td>1.1660</td>\n",
              "      <td>0.3329</td>\n",
              "      <td>0.2754</td>\n",
              "      <td>0.4108</td>\n",
              "      <td>-0.1252</td>\n",
              "      <td>-0.2340</td>\n",
              "      <td>0.2267</td>\n",
              "      <td>0.1969</td>\n",
              "      <td>0.0262</td>\n",
              "      <td>-0.8121</td>\n",
              "      <td>0.3434</td>\n",
              "      <td>0.5372</td>\n",
              "      <td>-0.3246</td>\n",
              "      <td>0.0631</td>\n",
              "      <td>0.9171</td>\n",
              "      <td>0.5258</td>\n",
              "      <td>0.4680</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23810</th>\n",
              "      <td>com_droga</td>\n",
              "      <td>24</td>\n",
              "      <td>D2</td>\n",
              "      <td>-1.3260</td>\n",
              "      <td>0.3478</td>\n",
              "      <td>-0.3743</td>\n",
              "      <td>0.9905</td>\n",
              "      <td>-0.7178</td>\n",
              "      <td>0.6621</td>\n",
              "      <td>-0.2252</td>\n",
              "      <td>-0.5565</td>\n",
              "      <td>0.5112</td>\n",
              "      <td>0.6727</td>\n",
              "      <td>-0.1851</td>\n",
              "      <td>2.8650</td>\n",
              "      <td>-0.2140</td>\n",
              "      <td>-0.6153</td>\n",
              "      <td>0.8362</td>\n",
              "      <td>0.5584</td>\n",
              "      <td>-0.2589</td>\n",
              "      <td>0.1292</td>\n",
              "      <td>0.0148</td>\n",
              "      <td>0.0949</td>\n",
              "      <td>-0.2182</td>\n",
              "      <td>-0.9235</td>\n",
              "      <td>0.0749</td>\n",
              "      <td>-1.5910</td>\n",
              "      <td>-0.8359</td>\n",
              "      <td>-0.9217</td>\n",
              "      <td>0.3013</td>\n",
              "      <td>0.1716</td>\n",
              "      <td>0.0880</td>\n",
              "      <td>0.1842</td>\n",
              "      <td>0.1835</td>\n",
              "      <td>0.5436</td>\n",
              "      <td>-0.0533</td>\n",
              "      <td>-0.0491</td>\n",
              "      <td>0.9543</td>\n",
              "      <td>0.4626</td>\n",
              "      <td>0.0819</td>\n",
              "      <td>...</td>\n",
              "      <td>0.1410</td>\n",
              "      <td>1.2640</td>\n",
              "      <td>-0.8663</td>\n",
              "      <td>0.8129</td>\n",
              "      <td>-0.1514</td>\n",
              "      <td>-0.4652</td>\n",
              "      <td>-0.7390</td>\n",
              "      <td>-1.3270</td>\n",
              "      <td>0.9925</td>\n",
              "      <td>1.0570</td>\n",
              "      <td>-0.3355</td>\n",
              "      <td>-0.2555</td>\n",
              "      <td>0.8219</td>\n",
              "      <td>0.6995</td>\n",
              "      <td>-0.2942</td>\n",
              "      <td>0.2408</td>\n",
              "      <td>-0.7781</td>\n",
              "      <td>-0.0929</td>\n",
              "      <td>-0.0329</td>\n",
              "      <td>0.0781</td>\n",
              "      <td>-1.4440</td>\n",
              "      <td>-0.0921</td>\n",
              "      <td>0.3188</td>\n",
              "      <td>-1.1080</td>\n",
              "      <td>0.4895</td>\n",
              "      <td>-0.2144</td>\n",
              "      <td>1.0960</td>\n",
              "      <td>0.3335</td>\n",
              "      <td>0.4444</td>\n",
              "      <td>-1.1130</td>\n",
              "      <td>0.4286</td>\n",
              "      <td>0.4426</td>\n",
              "      <td>0.0423</td>\n",
              "      <td>-0.3195</td>\n",
              "      <td>-0.8086</td>\n",
              "      <td>-0.9798</td>\n",
              "      <td>-0.2084</td>\n",
              "      <td>-0.1224</td>\n",
              "      <td>-0.2715</td>\n",
              "      <td>0.3689</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23811</th>\n",
              "      <td>com_controle</td>\n",
              "      <td>48</td>\n",
              "      <td>D2</td>\n",
              "      <td>0.3942</td>\n",
              "      <td>0.3756</td>\n",
              "      <td>0.3109</td>\n",
              "      <td>-0.7389</td>\n",
              "      <td>0.5505</td>\n",
              "      <td>-0.0159</td>\n",
              "      <td>-0.2541</td>\n",
              "      <td>0.1745</td>\n",
              "      <td>-0.0340</td>\n",
              "      <td>0.4865</td>\n",
              "      <td>-0.1854</td>\n",
              "      <td>0.0716</td>\n",
              "      <td>0.1729</td>\n",
              "      <td>-0.0434</td>\n",
              "      <td>0.1542</td>\n",
              "      <td>-0.2192</td>\n",
              "      <td>-0.0302</td>\n",
              "      <td>-0.4218</td>\n",
              "      <td>0.4057</td>\n",
              "      <td>-0.5372</td>\n",
              "      <td>0.1521</td>\n",
              "      <td>-0.2651</td>\n",
              "      <td>0.2310</td>\n",
              "      <td>-0.8101</td>\n",
              "      <td>0.4943</td>\n",
              "      <td>0.6905</td>\n",
              "      <td>-0.3720</td>\n",
              "      <td>-1.4110</td>\n",
              "      <td>0.4516</td>\n",
              "      <td>1.2300</td>\n",
              "      <td>-0.1949</td>\n",
              "      <td>-1.3280</td>\n",
              "      <td>-0.4276</td>\n",
              "      <td>-0.0040</td>\n",
              "      <td>-0.3086</td>\n",
              "      <td>-0.2355</td>\n",
              "      <td>-0.5222</td>\n",
              "      <td>...</td>\n",
              "      <td>0.6845</td>\n",
              "      <td>0.7127</td>\n",
              "      <td>0.7294</td>\n",
              "      <td>0.4718</td>\n",
              "      <td>-0.2020</td>\n",
              "      <td>0.2783</td>\n",
              "      <td>0.4934</td>\n",
              "      <td>0.4144</td>\n",
              "      <td>0.5449</td>\n",
              "      <td>1.4690</td>\n",
              "      <td>-0.6142</td>\n",
              "      <td>0.6068</td>\n",
              "      <td>0.3434</td>\n",
              "      <td>0.9880</td>\n",
              "      <td>-0.0468</td>\n",
              "      <td>-0.1882</td>\n",
              "      <td>-0.0087</td>\n",
              "      <td>-0.0356</td>\n",
              "      <td>0.5718</td>\n",
              "      <td>0.4971</td>\n",
              "      <td>0.1057</td>\n",
              "      <td>0.2111</td>\n",
              "      <td>0.6992</td>\n",
              "      <td>0.0708</td>\n",
              "      <td>0.6169</td>\n",
              "      <td>0.2248</td>\n",
              "      <td>0.5994</td>\n",
              "      <td>0.2689</td>\n",
              "      <td>0.0305</td>\n",
              "      <td>1.2320</td>\n",
              "      <td>0.5409</td>\n",
              "      <td>0.3755</td>\n",
              "      <td>0.7343</td>\n",
              "      <td>0.2807</td>\n",
              "      <td>0.4116</td>\n",
              "      <td>0.6422</td>\n",
              "      <td>0.2256</td>\n",
              "      <td>0.7592</td>\n",
              "      <td>0.6656</td>\n",
              "      <td>0.3808</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23812</th>\n",
              "      <td>com_droga</td>\n",
              "      <td>24</td>\n",
              "      <td>D1</td>\n",
              "      <td>0.6660</td>\n",
              "      <td>0.2324</td>\n",
              "      <td>0.4392</td>\n",
              "      <td>0.2044</td>\n",
              "      <td>0.8531</td>\n",
              "      <td>-0.0343</td>\n",
              "      <td>0.0323</td>\n",
              "      <td>0.0463</td>\n",
              "      <td>0.4299</td>\n",
              "      <td>-0.7985</td>\n",
              "      <td>0.5742</td>\n",
              "      <td>0.1421</td>\n",
              "      <td>2.2700</td>\n",
              "      <td>0.2046</td>\n",
              "      <td>0.5363</td>\n",
              "      <td>-1.7330</td>\n",
              "      <td>0.1450</td>\n",
              "      <td>0.6097</td>\n",
              "      <td>0.2024</td>\n",
              "      <td>0.9865</td>\n",
              "      <td>-0.7805</td>\n",
              "      <td>0.9608</td>\n",
              "      <td>0.3440</td>\n",
              "      <td>2.7650</td>\n",
              "      <td>0.4925</td>\n",
              "      <td>0.6698</td>\n",
              "      <td>0.2374</td>\n",
              "      <td>-0.3372</td>\n",
              "      <td>0.8771</td>\n",
              "      <td>-2.6560</td>\n",
              "      <td>-0.2000</td>\n",
              "      <td>-0.2043</td>\n",
              "      <td>0.6797</td>\n",
              "      <td>-0.0248</td>\n",
              "      <td>-0.0927</td>\n",
              "      <td>1.8480</td>\n",
              "      <td>-2.7180</td>\n",
              "      <td>...</td>\n",
              "      <td>0.3360</td>\n",
              "      <td>-0.6136</td>\n",
              "      <td>0.5011</td>\n",
              "      <td>0.9261</td>\n",
              "      <td>0.4419</td>\n",
              "      <td>0.0295</td>\n",
              "      <td>0.4220</td>\n",
              "      <td>0.4677</td>\n",
              "      <td>-0.1184</td>\n",
              "      <td>0.4524</td>\n",
              "      <td>0.4135</td>\n",
              "      <td>0.1356</td>\n",
              "      <td>-0.5801</td>\n",
              "      <td>0.0411</td>\n",
              "      <td>1.0240</td>\n",
              "      <td>1.0340</td>\n",
              "      <td>-0.0270</td>\n",
              "      <td>-0.4194</td>\n",
              "      <td>0.7403</td>\n",
              "      <td>-0.6793</td>\n",
              "      <td>-0.1423</td>\n",
              "      <td>0.7307</td>\n",
              "      <td>0.7946</td>\n",
              "      <td>-0.0650</td>\n",
              "      <td>0.9038</td>\n",
              "      <td>0.2324</td>\n",
              "      <td>0.9676</td>\n",
              "      <td>1.0940</td>\n",
              "      <td>0.4290</td>\n",
              "      <td>0.5187</td>\n",
              "      <td>-0.1105</td>\n",
              "      <td>0.4258</td>\n",
              "      <td>-0.2012</td>\n",
              "      <td>0.1506</td>\n",
              "      <td>1.5230</td>\n",
              "      <td>0.7101</td>\n",
              "      <td>0.1732</td>\n",
              "      <td>0.7015</td>\n",
              "      <td>-0.6290</td>\n",
              "      <td>0.0740</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23813</th>\n",
              "      <td>com_droga</td>\n",
              "      <td>72</td>\n",
              "      <td>D1</td>\n",
              "      <td>-0.8598</td>\n",
              "      <td>1.0240</td>\n",
              "      <td>-0.1361</td>\n",
              "      <td>0.7952</td>\n",
              "      <td>-0.3611</td>\n",
              "      <td>-3.6750</td>\n",
              "      <td>-1.2420</td>\n",
              "      <td>0.9146</td>\n",
              "      <td>3.0790</td>\n",
              "      <td>1.2460</td>\n",
              "      <td>1.9460</td>\n",
              "      <td>1.4370</td>\n",
              "      <td>2.9780</td>\n",
              "      <td>2.2370</td>\n",
              "      <td>-0.6818</td>\n",
              "      <td>0.6870</td>\n",
              "      <td>-1.1060</td>\n",
              "      <td>0.0182</td>\n",
              "      <td>-0.9247</td>\n",
              "      <td>-0.0738</td>\n",
              "      <td>-0.1919</td>\n",
              "      <td>-0.7722</td>\n",
              "      <td>-1.4050</td>\n",
              "      <td>-1.0050</td>\n",
              "      <td>-1.1170</td>\n",
              "      <td>-0.5293</td>\n",
              "      <td>-1.1720</td>\n",
              "      <td>-0.2885</td>\n",
              "      <td>0.1599</td>\n",
              "      <td>-0.4250</td>\n",
              "      <td>0.3591</td>\n",
              "      <td>-0.1420</td>\n",
              "      <td>-0.9530</td>\n",
              "      <td>-0.2005</td>\n",
              "      <td>-1.8340</td>\n",
              "      <td>-0.5768</td>\n",
              "      <td>0.5145</td>\n",
              "      <td>...</td>\n",
              "      <td>-6.9170</td>\n",
              "      <td>-1.8640</td>\n",
              "      <td>-2.5090</td>\n",
              "      <td>-4.7130</td>\n",
              "      <td>-1.7250</td>\n",
              "      <td>-3.8650</td>\n",
              "      <td>-3.0800</td>\n",
              "      <td>-4.1530</td>\n",
              "      <td>-1.2030</td>\n",
              "      <td>-1.1690</td>\n",
              "      <td>-4.1460</td>\n",
              "      <td>-1.2670</td>\n",
              "      <td>-1.1300</td>\n",
              "      <td>-2.4390</td>\n",
              "      <td>0.1591</td>\n",
              "      <td>-2.2490</td>\n",
              "      <td>-2.5860</td>\n",
              "      <td>-1.9520</td>\n",
              "      <td>-2.1810</td>\n",
              "      <td>-4.6690</td>\n",
              "      <td>-3.9450</td>\n",
              "      <td>-2.9560</td>\n",
              "      <td>-2.7930</td>\n",
              "      <td>-2.1560</td>\n",
              "      <td>-2.4100</td>\n",
              "      <td>-1.8190</td>\n",
              "      <td>-3.3480</td>\n",
              "      <td>-0.1414</td>\n",
              "      <td>-2.6430</td>\n",
              "      <td>-2.5810</td>\n",
              "      <td>-3.3890</td>\n",
              "      <td>-1.7450</td>\n",
              "      <td>-6.6300</td>\n",
              "      <td>-4.0950</td>\n",
              "      <td>-7.3860</td>\n",
              "      <td>-1.4160</td>\n",
              "      <td>-3.5770</td>\n",
              "      <td>-0.4775</td>\n",
              "      <td>-2.1500</td>\n",
              "      <td>-4.2520</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>23814 rows × 875 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "         tratamento  tempo dose     g-0  ...    c-96    c-97    c-98    c-99\n",
              "0         com_droga     24   D1  1.0620  ... -0.3981  0.2139  0.3801  0.4176\n",
              "1         com_droga     72   D1  0.0743  ...  0.1522  0.1241  0.6077  0.7371\n",
              "2         com_droga     48   D1  0.6280  ... -0.6417 -0.2187 -1.4080  0.6931\n",
              "3         com_droga     48   D1 -0.5138  ... -1.6210 -0.8784 -0.3876 -0.8154\n",
              "4         com_droga     72   D2 -0.3254  ...  0.1094  0.2885 -0.3786  0.7125\n",
              "...             ...    ...  ...     ...  ...     ...     ...     ...     ...\n",
              "23809     com_droga     24   D2  0.1394  ...  0.0631  0.9171  0.5258  0.4680\n",
              "23810     com_droga     24   D2 -1.3260  ... -0.2084 -0.1224 -0.2715  0.3689\n",
              "23811  com_controle     48   D2  0.3942  ...  0.2256  0.7592  0.6656  0.3808\n",
              "23812     com_droga     24   D1  0.6660  ...  0.1732  0.7015 -0.6290  0.0740\n",
              "23813     com_droga     72   D1 -0.8598  ... -3.5770 -0.4775 -2.1500 -4.2520\n",
              "\n",
              "[23814 rows x 875 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8kdH6cklncP-"
      },
      "source": [
        "###Random Forest"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "c83SdyYDWQQx"
      },
      "source": [
        "Enfim, podemos partir para o ajuste do nosso modelo de Random Forest.\n",
        "\n",
        "Novamente, vamos seguir as mesmas etapas lógicas anteriores. Entretando, precisamos modificar a nossa base ```x```. \n",
        "Então faremos: \n",
        "\n",
        "    x = dados_combinados.drop(['id', 'n_moa', 'ativo_moa', 'composto'], axis=1)  ->  base de dados que deve ser considerada com a exclusão das colunas que determinamos anteriormente\n",
        "    x = pd.get_dummies(x, columns=['tratamento', 'dose', 'tempo'])  ->  a partir do x definido acima, fazemos a transformação com o get_dummies nas colunas com o formato de string\n",
        "\n",
        "Por fim, temos como retorno uma acurácia de ~68,0% e, ela é, aproximadamente, 7% melhor que a acurácia que conseguimos com o ```DummyClassifier```. Aqui, não podemos comparar com a regressão logística porque estamos considerando um outro conjunto de dados. \n",
        "\n",
        "É muito importante ressaltar que, apesar de termos melhorado a acurácia do nosso modelo, pensando na vida real, talvez a nossa melhora ainda não seja suficiente. Ou seja, desenvolvemos um modelo com um algoritmo de Machine Learning mas, ele ainda não é aplicável à vida real (isso quer dizer que, ainda não nos traz ganhos suficientes para sua implantação).\n",
        "\n",
        "Por outro lado, durante todo o nosso projeto, desenvolvemos modelos simples e não nos aprofundamos em todos os pontos que podem ser modificados (variáveis, parâmetros de modelos, por exemplo) a fim de otimizar a nossa generalização. \n",
        "Deixamos em destaque também a sua trajetória até aqui, certamente ela deve ser seu foco (e é também o nosso)! O resultado é apenas uma consequência disso tudo e sabemos que, para melhorá-lo você precisa persistir no estudo dessa área. \n",
        "\n",
        "A Imersão de Dados 3 foi o seu primeiro passo na ciência de dados e aqui, deixamos o convite para que você continue estudando, explorando e aprendendo todos os processos e áreas desse mundo gigantesco de possibilidades!\n",
        "\n",
        "Nos vemos em breve, cientistas!"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qrE_ADZW7lG2",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7815e459-f14b-41a8-d9ed-1332feb37538"
      },
      "source": [
        "from sklearn.ensemble import RandomForestClassifier\n",
        "\n",
        "x = dados_combinados.drop(['id', 'n_moa', 'ativo_moa', 'composto'], axis=1)\n",
        "x = pd.get_dummies(x, columns=['tratamento', 'dose', 'tempo'])\n",
        "y = dados_combinados['ativo_moa'] \n",
        "x_treino, x_teste, y_treino, y_teste = train_test_split(x, y, test_size = 0.2, stratify=y, random_state=376)\n",
        "\n",
        "modelo_randomforest = RandomForestClassifier()\n",
        "modelo_randomforest.fit(x_treino, y_treino)\n",
        "modelo_randomforest.score(x_teste, y_teste)"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6836027713625866"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2-X3Y44VVuUV"
      },
      "source": [
        "# **Desafio 1**\n",
        "\n",
        "Tente outros modelos ou mude os parâmetros dos modelos que testamos em aula."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "muZtG5GvzqSu"
      },
      "source": [
        "Existem muitos modelos para machine learning, e cada modelo tem seus pontos fortes e fracos. Vamos verificar alguns deles:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vEBL7psffwrw",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ec5c4c11-c2dc-479f-c959-87614e53eedd"
      },
      "source": [
        "#Teste 1: Regressãp Logística\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "x = dados_combinados.select_dtypes('float64')\n",
        "y = dados_combinados['ativo_moa'] \n",
        "x_treino, x_teste, y_treino, y_teste = train_test_split(x, y, test_size = 0.2, stratify=y, random_state=42)\n",
        "\n",
        "#modelo de regressão logística\n",
        "modelo_rlogistica = LogisticRegression(max_iter=1000, C=10)\n",
        "#treinando modelo\n",
        "modelo_rlogistica.fit(x_treino, y_treino)\n",
        "#score do modelo: como o modelo está se saindo no teste: retorna a acurácia\n",
        "modelo_rlogistica.score(x_teste, y_teste)"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6348939743858912"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "O7-Dim62Bkq3"
      },
      "source": [
        "Mudando o valor do random_state e parâmetro C para 10. O parâmetro controla a complexidade do modelo. Valores baixos indicam que desejamos um modelo mais complexo."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_YKGL6fKoUeF",
        "outputId": "41a71278-85ec-4182-e076-70dd4e0da3d8"
      },
      "source": [
        "#Teste 2: Random Forest\n",
        "\n",
        "x = dados_combinados.drop(['id', 'n_moa', 'ativo_moa', 'composto'], axis=1)\n",
        "#variáveis independentes\n",
        "x = pd.get_dummies(x, columns=['tratamento', 'dose', 'tempo'])\n",
        "#variável dependente\n",
        "y = dados_combinados['ativo_moa'] \n",
        "x_treino, x_teste, y_treino, y_teste = train_test_split(x, \n",
        "                                                        y, \n",
        "                                                        test_size = 0.2, \n",
        "                                                        stratify=y, \n",
        "                                                        random_state=42)\n",
        "\n",
        "modelo_randomforest = RandomForestClassifier()\n",
        "modelo_randomforest.fit(x_treino, y_treino)\n",
        "modelo_randomforest.score(x_teste, y_teste)"
      ],
      "execution_count": 136,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6970396808733991"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 136
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S9OJG185zE_5",
        "outputId": "f4deb06d-08e5-4f2b-a94c-0c5283d4bbcf"
      },
      "source": [
        "#Teste 3: Bayesian Ridge Regression\n",
        "from sklearn import linear_model\n",
        "\n",
        "x = dados_combinados.drop(['id', 'n_moa', 'ativo_moa', 'composto'], axis=1)\n",
        "#variáveis independentes\n",
        "x = pd.get_dummies(x, columns=['tratamento', 'dose', 'tempo'])\n",
        "#variável dependente\n",
        "y = dados_combinados['ativo_moa'] \n",
        "x_treino, x_teste, y_treino, y_teste = train_test_split(x,\n",
        "                                                          y,\n",
        "                                                          test_size=0.2,\n",
        "                                                          stratify=y,\n",
        "                                                          random_state=42)\n",
        "# Inicializar nosso classificador\n",
        "modelo_bayesian = linear_model.BayesianRidge()\n",
        "\n",
        "# Treinar nosso classificador\n",
        "modelo_bayesian.fit(x_treino, y_treino)\n",
        "\n",
        "# Fazer previsões\n",
        "preds = modelo_bayesian.predict(x_teste, y_teste)\n",
        "\n",
        "# Avaliar a precisão\n",
        "modelo_bayesian.score(x_teste, y_teste)"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.1580082164249076"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BjqzGCtlg2Tt"
      },
      "source": [
        "# **Desafio 02**\n",
        "\n",
        "Estude a [documentação](https://scikit-learn.org/stable/) do Sklearn."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_Ljm7Zxyg7-L"
      },
      "source": [
        "# **Desafio 03**\n",
        "\n",
        "Pesquisar o fluxograma 'Escolhendo o melhor modelo' choosing the right estimator.\n",
        "\n",
        "https://scikit-learn.org/stable/tutorial/machine_learning_map/index.html\n",
        "\n",
        "Artigo interessante no medium sobre: https://medium.com/@t.rosen2101/choosing-the-right-estimator-ae054809b693\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7P3FwdqIhA2o"
      },
      "source": [
        "# **Desafio 4**\n",
        "\n",
        "Teste outras problemas (perguntas). 0, 1 ou +."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 251
        },
        "id": "UmR8Z49lyJHO",
        "outputId": "03ab902c-4ff3-46f6-d7ca-6ae8b5fa3f6c"
      },
      "source": [
        "dados_combinados.head()"
      ],
      "execution_count": 157,
      "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>id</th>\n",
              "      <th>tratamento</th>\n",
              "      <th>tempo</th>\n",
              "      <th>dose</th>\n",
              "      <th>composto</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>g-7</th>\n",
              "      <th>g-8</th>\n",
              "      <th>g-9</th>\n",
              "      <th>g-10</th>\n",
              "      <th>g-11</th>\n",
              "      <th>g-12</th>\n",
              "      <th>g-13</th>\n",
              "      <th>g-14</th>\n",
              "      <th>g-15</th>\n",
              "      <th>g-16</th>\n",
              "      <th>g-17</th>\n",
              "      <th>g-18</th>\n",
              "      <th>g-19</th>\n",
              "      <th>g-20</th>\n",
              "      <th>g-21</th>\n",
              "      <th>g-22</th>\n",
              "      <th>g-23</th>\n",
              "      <th>g-24</th>\n",
              "      <th>g-25</th>\n",
              "      <th>g-26</th>\n",
              "      <th>g-27</th>\n",
              "      <th>g-28</th>\n",
              "      <th>g-29</th>\n",
              "      <th>g-30</th>\n",
              "      <th>g-31</th>\n",
              "      <th>g-32</th>\n",
              "      <th>g-33</th>\n",
              "      <th>g-34</th>\n",
              "      <th>...</th>\n",
              "      <th>c-62</th>\n",
              "      <th>c-63</th>\n",
              "      <th>c-64</th>\n",
              "      <th>c-65</th>\n",
              "      <th>c-66</th>\n",
              "      <th>c-67</th>\n",
              "      <th>c-68</th>\n",
              "      <th>c-69</th>\n",
              "      <th>c-70</th>\n",
              "      <th>c-71</th>\n",
              "      <th>c-72</th>\n",
              "      <th>c-73</th>\n",
              "      <th>c-74</th>\n",
              "      <th>c-75</th>\n",
              "      <th>c-76</th>\n",
              "      <th>c-77</th>\n",
              "      <th>c-78</th>\n",
              "      <th>c-79</th>\n",
              "      <th>c-80</th>\n",
              "      <th>c-81</th>\n",
              "      <th>c-82</th>\n",
              "      <th>c-83</th>\n",
              "      <th>c-84</th>\n",
              "      <th>c-85</th>\n",
              "      <th>c-86</th>\n",
              "      <th>c-87</th>\n",
              "      <th>c-88</th>\n",
              "      <th>c-89</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",
              "      <th>n_moa</th>\n",
              "      <th>ativo_moa</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>id_000644bb2</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>24</td>\n",
              "      <td>D1</td>\n",
              "      <td>b68db1d53</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>-0.0326</td>\n",
              "      <td>0.5548</td>\n",
              "      <td>-0.0921</td>\n",
              "      <td>1.1830</td>\n",
              "      <td>0.1530</td>\n",
              "      <td>0.5574</td>\n",
              "      <td>-0.4015</td>\n",
              "      <td>0.1789</td>\n",
              "      <td>-0.6528</td>\n",
              "      <td>-0.7969</td>\n",
              "      <td>0.6342</td>\n",
              "      <td>0.1778</td>\n",
              "      <td>-0.3694</td>\n",
              "      <td>-0.5688</td>\n",
              "      <td>-1.1360</td>\n",
              "      <td>-1.1880</td>\n",
              "      <td>0.6940</td>\n",
              "      <td>0.4393</td>\n",
              "      <td>0.2664</td>\n",
              "      <td>0.1907</td>\n",
              "      <td>0.1628</td>\n",
              "      <td>-0.2853</td>\n",
              "      <td>0.5819</td>\n",
              "      <td>0.2934</td>\n",
              "      <td>-0.5584</td>\n",
              "      <td>-0.0916</td>\n",
              "      <td>-0.3010</td>\n",
              "      <td>-0.1537</td>\n",
              "      <td>...</td>\n",
              "      <td>0.3680</td>\n",
              "      <td>0.8427</td>\n",
              "      <td>0.1042</td>\n",
              "      <td>0.1403</td>\n",
              "      <td>0.1758</td>\n",
              "      <td>1.2570</td>\n",
              "      <td>-0.5979</td>\n",
              "      <td>1.2250</td>\n",
              "      <td>-0.0553</td>\n",
              "      <td>0.7351</td>\n",
              "      <td>0.5810</td>\n",
              "      <td>0.9590</td>\n",
              "      <td>0.2427</td>\n",
              "      <td>0.0495</td>\n",
              "      <td>0.4141</td>\n",
              "      <td>0.8432</td>\n",
              "      <td>0.6162</td>\n",
              "      <td>-0.7318</td>\n",
              "      <td>1.2120</td>\n",
              "      <td>0.6362</td>\n",
              "      <td>-0.4427</td>\n",
              "      <td>0.1288</td>\n",
              "      <td>1.4840</td>\n",
              "      <td>0.1799</td>\n",
              "      <td>0.5367</td>\n",
              "      <td>-0.1111</td>\n",
              "      <td>-1.0120</td>\n",
              "      <td>0.6685</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",
              "      <td>1</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>id_000779bfc</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>72</td>\n",
              "      <td>D1</td>\n",
              "      <td>df89a8e5a</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>0.3372</td>\n",
              "      <td>-0.4047</td>\n",
              "      <td>0.8507</td>\n",
              "      <td>-1.1520</td>\n",
              "      <td>-0.4201</td>\n",
              "      <td>-0.0958</td>\n",
              "      <td>0.4590</td>\n",
              "      <td>0.0803</td>\n",
              "      <td>0.2250</td>\n",
              "      <td>0.5293</td>\n",
              "      <td>0.2839</td>\n",
              "      <td>-0.3494</td>\n",
              "      <td>0.2883</td>\n",
              "      <td>0.9449</td>\n",
              "      <td>-0.1646</td>\n",
              "      <td>-0.2657</td>\n",
              "      <td>-0.3372</td>\n",
              "      <td>0.3135</td>\n",
              "      <td>-0.4316</td>\n",
              "      <td>0.4773</td>\n",
              "      <td>0.2075</td>\n",
              "      <td>-0.4216</td>\n",
              "      <td>-0.1161</td>\n",
              "      <td>-0.0499</td>\n",
              "      <td>-0.2627</td>\n",
              "      <td>0.9959</td>\n",
              "      <td>-0.2483</td>\n",
              "      <td>0.2655</td>\n",
              "      <td>...</td>\n",
              "      <td>0.3905</td>\n",
              "      <td>0.7099</td>\n",
              "      <td>0.2912</td>\n",
              "      <td>0.4151</td>\n",
              "      <td>-0.2840</td>\n",
              "      <td>-0.3104</td>\n",
              "      <td>-0.6373</td>\n",
              "      <td>0.2887</td>\n",
              "      <td>-0.0765</td>\n",
              "      <td>0.2539</td>\n",
              "      <td>0.4443</td>\n",
              "      <td>0.5932</td>\n",
              "      <td>0.2031</td>\n",
              "      <td>0.7639</td>\n",
              "      <td>0.5499</td>\n",
              "      <td>-0.3322</td>\n",
              "      <td>-0.0977</td>\n",
              "      <td>0.4329</td>\n",
              "      <td>-0.2782</td>\n",
              "      <td>0.7827</td>\n",
              "      <td>0.5934</td>\n",
              "      <td>0.3402</td>\n",
              "      <td>0.1499</td>\n",
              "      <td>0.4420</td>\n",
              "      <td>0.9366</td>\n",
              "      <td>0.8193</td>\n",
              "      <td>-0.4236</td>\n",
              "      <td>0.3192</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",
              "      <td>0</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>id_000a6266a</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>48</td>\n",
              "      <td>D1</td>\n",
              "      <td>18bb41b2c</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>0.2155</td>\n",
              "      <td>0.0065</td>\n",
              "      <td>1.2300</td>\n",
              "      <td>-0.4797</td>\n",
              "      <td>-0.5631</td>\n",
              "      <td>-0.0366</td>\n",
              "      <td>-1.8300</td>\n",
              "      <td>0.6057</td>\n",
              "      <td>-0.3278</td>\n",
              "      <td>0.6042</td>\n",
              "      <td>-0.3075</td>\n",
              "      <td>-0.1147</td>\n",
              "      <td>-0.0570</td>\n",
              "      <td>-0.0799</td>\n",
              "      <td>-0.8181</td>\n",
              "      <td>-1.5320</td>\n",
              "      <td>0.2307</td>\n",
              "      <td>0.4901</td>\n",
              "      <td>0.4780</td>\n",
              "      <td>-1.3970</td>\n",
              "      <td>4.6240</td>\n",
              "      <td>-0.0437</td>\n",
              "      <td>1.2870</td>\n",
              "      <td>-1.8530</td>\n",
              "      <td>0.6069</td>\n",
              "      <td>0.4290</td>\n",
              "      <td>0.1783</td>\n",
              "      <td>0.0018</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.0444</td>\n",
              "      <td>0.1894</td>\n",
              "      <td>-0.0014</td>\n",
              "      <td>-2.3640</td>\n",
              "      <td>-0.4682</td>\n",
              "      <td>0.1210</td>\n",
              "      <td>-0.5177</td>\n",
              "      <td>-0.0604</td>\n",
              "      <td>0.1682</td>\n",
              "      <td>-0.4436</td>\n",
              "      <td>0.4963</td>\n",
              "      <td>0.1363</td>\n",
              "      <td>0.3335</td>\n",
              "      <td>0.9760</td>\n",
              "      <td>-0.0427</td>\n",
              "      <td>-0.1235</td>\n",
              "      <td>0.0959</td>\n",
              "      <td>0.0690</td>\n",
              "      <td>-0.9416</td>\n",
              "      <td>-0.7548</td>\n",
              "      <td>-0.1109</td>\n",
              "      <td>-0.6272</td>\n",
              "      <td>0.3019</td>\n",
              "      <td>0.1172</td>\n",
              "      <td>0.1093</td>\n",
              "      <td>-0.3113</td>\n",
              "      <td>0.3019</td>\n",
              "      <td>-0.0873</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",
              "      <td>3</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>id_0015fd391</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>48</td>\n",
              "      <td>D1</td>\n",
              "      <td>8c7f86626</td>\n",
              "      <td>-0.5138</td>\n",
              "      <td>-0.2491</td>\n",
              "      <td>-0.2656</td>\n",
              "      <td>0.5288</td>\n",
              "      <td>4.0620</td>\n",
              "      <td>-0.8095</td>\n",
              "      <td>-1.9590</td>\n",
              "      <td>0.1792</td>\n",
              "      <td>-0.1321</td>\n",
              "      <td>-1.0600</td>\n",
              "      <td>-0.8269</td>\n",
              "      <td>-0.3584</td>\n",
              "      <td>-0.8511</td>\n",
              "      <td>-0.5844</td>\n",
              "      <td>-2.5690</td>\n",
              "      <td>0.8183</td>\n",
              "      <td>-0.0532</td>\n",
              "      <td>-0.8554</td>\n",
              "      <td>0.1160</td>\n",
              "      <td>-2.3520</td>\n",
              "      <td>2.1200</td>\n",
              "      <td>-1.1580</td>\n",
              "      <td>-0.7191</td>\n",
              "      <td>-0.8004</td>\n",
              "      <td>-1.4670</td>\n",
              "      <td>-0.0107</td>\n",
              "      <td>-0.8995</td>\n",
              "      <td>0.2406</td>\n",
              "      <td>-0.2479</td>\n",
              "      <td>-1.0890</td>\n",
              "      <td>-0.7575</td>\n",
              "      <td>0.0881</td>\n",
              "      <td>-2.7370</td>\n",
              "      <td>0.8745</td>\n",
              "      <td>0.5787</td>\n",
              "      <td>...</td>\n",
              "      <td>-2.3820</td>\n",
              "      <td>-3.7350</td>\n",
              "      <td>-2.9740</td>\n",
              "      <td>-1.4930</td>\n",
              "      <td>-1.6600</td>\n",
              "      <td>-3.1660</td>\n",
              "      <td>0.2816</td>\n",
              "      <td>-0.2990</td>\n",
              "      <td>-1.1870</td>\n",
              "      <td>-0.5044</td>\n",
              "      <td>-1.7750</td>\n",
              "      <td>-1.6120</td>\n",
              "      <td>-0.9215</td>\n",
              "      <td>-1.0810</td>\n",
              "      <td>-3.0520</td>\n",
              "      <td>-3.4470</td>\n",
              "      <td>-2.7740</td>\n",
              "      <td>-1.8460</td>\n",
              "      <td>-0.5568</td>\n",
              "      <td>-3.3960</td>\n",
              "      <td>-2.9510</td>\n",
              "      <td>-1.1550</td>\n",
              "      <td>-3.2620</td>\n",
              "      <td>-1.5390</td>\n",
              "      <td>-2.4600</td>\n",
              "      <td>-0.9417</td>\n",
              "      <td>-1.5550</td>\n",
              "      <td>0.2431</td>\n",
              "      <td>-2.0990</td>\n",
              "      <td>-0.6441</td>\n",
              "      <td>-5.6300</td>\n",
              "      <td>-1.3780</td>\n",
              "      <td>-0.8632</td>\n",
              "      <td>-1.2880</td>\n",
              "      <td>-1.6210</td>\n",
              "      <td>-0.8784</td>\n",
              "      <td>-0.3876</td>\n",
              "      <td>-0.8154</td>\n",
              "      <td>0</td>\n",
              "      <td>False</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>id_001626bd3</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>72</td>\n",
              "      <td>D2</td>\n",
              "      <td>7cbed3131</td>\n",
              "      <td>-0.3254</td>\n",
              "      <td>-0.4009</td>\n",
              "      <td>0.9700</td>\n",
              "      <td>0.6919</td>\n",
              "      <td>1.4180</td>\n",
              "      <td>-0.8244</td>\n",
              "      <td>-0.2800</td>\n",
              "      <td>-0.1498</td>\n",
              "      <td>-0.8789</td>\n",
              "      <td>0.8630</td>\n",
              "      <td>-0.2219</td>\n",
              "      <td>-0.5121</td>\n",
              "      <td>-0.9577</td>\n",
              "      <td>1.1750</td>\n",
              "      <td>0.2042</td>\n",
              "      <td>0.1970</td>\n",
              "      <td>0.1244</td>\n",
              "      <td>-1.7090</td>\n",
              "      <td>-0.3543</td>\n",
              "      <td>-0.5160</td>\n",
              "      <td>-0.3330</td>\n",
              "      <td>-0.2685</td>\n",
              "      <td>0.7649</td>\n",
              "      <td>0.2057</td>\n",
              "      <td>1.3720</td>\n",
              "      <td>0.6835</td>\n",
              "      <td>0.8056</td>\n",
              "      <td>-0.3754</td>\n",
              "      <td>-1.2090</td>\n",
              "      <td>0.2965</td>\n",
              "      <td>-0.0712</td>\n",
              "      <td>0.6389</td>\n",
              "      <td>0.6674</td>\n",
              "      <td>-0.0783</td>\n",
              "      <td>1.1740</td>\n",
              "      <td>...</td>\n",
              "      <td>0.1535</td>\n",
              "      <td>-0.4640</td>\n",
              "      <td>-0.5943</td>\n",
              "      <td>0.3973</td>\n",
              "      <td>0.1500</td>\n",
              "      <td>0.5178</td>\n",
              "      <td>0.5159</td>\n",
              "      <td>0.6091</td>\n",
              "      <td>0.1813</td>\n",
              "      <td>-0.4249</td>\n",
              "      <td>0.7832</td>\n",
              "      <td>0.6529</td>\n",
              "      <td>0.5648</td>\n",
              "      <td>0.4817</td>\n",
              "      <td>0.0587</td>\n",
              "      <td>0.5303</td>\n",
              "      <td>0.6376</td>\n",
              "      <td>-0.3966</td>\n",
              "      <td>-1.4950</td>\n",
              "      <td>-0.9625</td>\n",
              "      <td>-0.0541</td>\n",
              "      <td>0.6273</td>\n",
              "      <td>0.4563</td>\n",
              "      <td>0.0698</td>\n",
              "      <td>0.8134</td>\n",
              "      <td>0.1924</td>\n",
              "      <td>0.6054</td>\n",
              "      <td>-0.1824</td>\n",
              "      <td>0.0042</td>\n",
              "      <td>0.0048</td>\n",
              "      <td>0.6670</td>\n",
              "      <td>1.0690</td>\n",
              "      <td>0.5523</td>\n",
              "      <td>-0.3031</td>\n",
              "      <td>0.1094</td>\n",
              "      <td>0.2885</td>\n",
              "      <td>-0.3786</td>\n",
              "      <td>0.7125</td>\n",
              "      <td>1</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 879 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "             id tratamento  tempo dose  ...    c-98    c-99  n_moa  ativo_moa\n",
              "0  id_000644bb2  com_droga     24   D1  ...  0.3801  0.4176      1       True\n",
              "1  id_000779bfc  com_droga     72   D1  ...  0.6077  0.7371      0      False\n",
              "2  id_000a6266a  com_droga     48   D1  ... -1.4080  0.6931      3       True\n",
              "3  id_0015fd391  com_droga     48   D1  ... -0.3876 -0.8154      0      False\n",
              "4  id_001626bd3  com_droga     72   D2  ... -0.3786  0.7125      1       True\n",
              "\n",
              "[5 rows x 879 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 157
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ufnGqzBmyT19",
        "outputId": "f24b92ff-1c88-4876-eea0-bbf8669ef7bb"
      },
      "source": [
        "#objetivo: prever n_moa\n",
        "x = dados_combinados.drop(['id', 'n_moa', 'ativo_moa', 'composto'], axis=1)\n",
        "#variáveis independentes\n",
        "x = pd.get_dummies(x, columns=['tratamento', 'dose', 'tempo'])\n",
        "#variável dependente\n",
        "y = dados_combinados['n_moa'] \n",
        "#separando dados em treino e teste, teste: 20% e treino: 80%\n",
        "x_treino, x_teste, y_treino, y_teste = train_test_split(x,\n",
        "                                                        y, \n",
        "                                                        test_size = 0.2, \n",
        "                                                        stratify = y, \n",
        "                                                        random_state = 42)\n",
        "\n",
        "modelo_randomforest = RandomForestClassifier()\n",
        "#treinando o modelo\n",
        "modelo_randomforest.fit(x_treino, y_treino)\n",
        "#acurácia do modelo\n",
        "modelo_randomforest.score(x_teste, y_teste)"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6376233466302751"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KoZqYvojhEod"
      },
      "source": [
        "# **Desafio 5**\n",
        "\n",
        "Supondo um cenário onde, num dado ensaio, as labels (referentes ao tipo de tratamento) dos frascos contendo os materiais usados foi perdida. Escolha a droga mais utilizada e tente criar um modelo para prever a label de tratamento."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mrgh8hUHwGLx"
      },
      "source": [
        "Primeiro vamos prever a label utilizando todas as drogas:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KbdGa4vur_DB",
        "outputId": "69a4d31c-38de-439e-d3ed-ce7d0ec48902"
      },
      "source": [
        "#retirando variavel tratamento do conjunto de variaveis independentes\n",
        "x = dados_combinados.drop(['id', 'n_moa', 'ativo_moa', 'composto', 'tratamento'], axis=1)\n",
        "#variáveis independentes\n",
        "x = pd.get_dummies(x, columns=['dose', 'tempo'])\n",
        "#variável dependente\n",
        "y = dados_combinados['tratamento'] \n",
        "#separando dados em treino e teste, teste: 20% e treino: 80%\n",
        "x_treino, x_teste, y_treino, y_teste = train_test_split(x, y, test_size = 0.2, stratify=y, random_state=42)\n",
        "\n",
        "modelo_randomforest = RandomForestClassifier()\n",
        "#treinando o modelo\n",
        "modelo_randomforest.fit(x_treino, y_treino)\n",
        "#acurácia do modelo\n",
        "modelo_randomforest.score(x_teste, y_teste)"
      ],
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9298761284904472"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yZmlgW9PEnYi"
      },
      "source": [
        "Vamos analisar a **curva ROC** deste modelo. A curva de ROC (Receiver Operator Characteristic) determina o desempenho de um modelo de classificação em um valor limite definido pelo usuário usando a área sob a curva ROC (Area Under Curve, AUC). Quanto maior a área, melhor o modelo. É um gráfico simples que permite estudar a variação da sensibilidade e especificidade, para diferentes pontos de corte na probabilidade estimada (thresholds). A AUC é uma medida de área que facilita a comparação entre Curvas ROC."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "i4_2q7mTDu5H",
        "outputId": "ec21a840-50e9-437a-c7d4-8f0e0941390c"
      },
      "source": [
        "from sklearn.metrics import plot_roc_curve\n",
        "plot_roc_curve(modelo_randomforest, x_teste,y_teste)\n",
        "plt.show()"
      ],
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sIcT3AfjFhs6"
      },
      "source": [
        "O valor do AUC varia de 0,0 até 1,0, ou de 0% a 100%. Quanto maior o AUC, melhor. Um modelo cujas previsões estão 100% erradas tem uma AUC de 0, enquanto um modelo cujas previsões são 100% corretas tem uma AUC de 1. Nosso modelo acima possui uma AUC de 0.93 = 93%, um valor ótimo!"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7-r7hGkkwJ-a"
      },
      "source": [
        "Agora, vamos filtrar as drogas e utilizar somente a droga mais utilizada:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WANuon5Xwd--",
        "outputId": "dd0dddec-5b2d-44bf-db2d-461bb1c0f32e"
      },
      "source": [
        "compostos_modelo = dados_combinados['composto'].value_counts().index[1:2] #excluindo a primeira droga que é a de controle\n",
        "compostos_modelo"
      ],
      "execution_count": 71,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['87d714366'], dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 71
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 435
        },
        "id": "oGQKHLO1w9OJ",
        "outputId": "f56facdd-19ec-4d53-f974-d3ea1ef1e72d"
      },
      "source": [
        "dados_filtro_composto = dados_combinados.query('composto in @compostos_modelo')\n",
        "dados_filtro_composto"
      ],
      "execution_count": 72,
      "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>id</th>\n",
              "      <th>tratamento</th>\n",
              "      <th>tempo</th>\n",
              "      <th>dose</th>\n",
              "      <th>composto</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>g-7</th>\n",
              "      <th>g-8</th>\n",
              "      <th>g-9</th>\n",
              "      <th>g-10</th>\n",
              "      <th>g-11</th>\n",
              "      <th>g-12</th>\n",
              "      <th>g-13</th>\n",
              "      <th>g-14</th>\n",
              "      <th>g-15</th>\n",
              "      <th>g-16</th>\n",
              "      <th>g-17</th>\n",
              "      <th>g-18</th>\n",
              "      <th>g-19</th>\n",
              "      <th>g-20</th>\n",
              "      <th>g-21</th>\n",
              "      <th>g-22</th>\n",
              "      <th>g-23</th>\n",
              "      <th>g-24</th>\n",
              "      <th>g-25</th>\n",
              "      <th>g-26</th>\n",
              "      <th>g-27</th>\n",
              "      <th>g-28</th>\n",
              "      <th>g-29</th>\n",
              "      <th>g-30</th>\n",
              "      <th>g-31</th>\n",
              "      <th>g-32</th>\n",
              "      <th>g-33</th>\n",
              "      <th>g-34</th>\n",
              "      <th>...</th>\n",
              "      <th>c-62</th>\n",
              "      <th>c-63</th>\n",
              "      <th>c-64</th>\n",
              "      <th>c-65</th>\n",
              "      <th>c-66</th>\n",
              "      <th>c-67</th>\n",
              "      <th>c-68</th>\n",
              "      <th>c-69</th>\n",
              "      <th>c-70</th>\n",
              "      <th>c-71</th>\n",
              "      <th>c-72</th>\n",
              "      <th>c-73</th>\n",
              "      <th>c-74</th>\n",
              "      <th>c-75</th>\n",
              "      <th>c-76</th>\n",
              "      <th>c-77</th>\n",
              "      <th>c-78</th>\n",
              "      <th>c-79</th>\n",
              "      <th>c-80</th>\n",
              "      <th>c-81</th>\n",
              "      <th>c-82</th>\n",
              "      <th>c-83</th>\n",
              "      <th>c-84</th>\n",
              "      <th>c-85</th>\n",
              "      <th>c-86</th>\n",
              "      <th>c-87</th>\n",
              "      <th>c-88</th>\n",
              "      <th>c-89</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",
              "      <th>n_moa</th>\n",
              "      <th>ativo_moa</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>id_002fb9c19</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>48</td>\n",
              "      <td>D1</td>\n",
              "      <td>87d714366</td>\n",
              "      <td>8.738</td>\n",
              "      <td>0.1914</td>\n",
              "      <td>2.4380</td>\n",
              "      <td>-0.2930</td>\n",
              "      <td>3.2010</td>\n",
              "      <td>1.4910</td>\n",
              "      <td>5.3920</td>\n",
              "      <td>-1.0420</td>\n",
              "      <td>-8.486</td>\n",
              "      <td>-2.5790</td>\n",
              "      <td>-6.747</td>\n",
              "      <td>3.6150</td>\n",
              "      <td>-1.4510</td>\n",
              "      <td>0.6909</td>\n",
              "      <td>-0.7623</td>\n",
              "      <td>-2.9400</td>\n",
              "      <td>-9.285</td>\n",
              "      <td>-6.7110</td>\n",
              "      <td>-0.4195</td>\n",
              "      <td>1.1600</td>\n",
              "      <td>-1.3180</td>\n",
              "      <td>1.3660</td>\n",
              "      <td>-1.4020</td>\n",
              "      <td>-2.6220</td>\n",
              "      <td>0.7204</td>\n",
              "      <td>-2.5930</td>\n",
              "      <td>-4.063</td>\n",
              "      <td>-0.8282</td>\n",
              "      <td>-3.500</td>\n",
              "      <td>0.9206</td>\n",
              "      <td>-5.145</td>\n",
              "      <td>9.804</td>\n",
              "      <td>-2.6840</td>\n",
              "      <td>-2.4390</td>\n",
              "      <td>-0.1427</td>\n",
              "      <td>...</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.511</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.887</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-4.902</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-5.552</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.796</td>\n",
              "      <td>-8.177</td>\n",
              "      <td>-9.899</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-9.984</td>\n",
              "      <td>-6.784</td>\n",
              "      <td>2</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>76</th>\n",
              "      <td>id_00cf039a7</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>48</td>\n",
              "      <td>D2</td>\n",
              "      <td>87d714366</td>\n",
              "      <td>4.546</td>\n",
              "      <td>0.4257</td>\n",
              "      <td>0.7670</td>\n",
              "      <td>2.1950</td>\n",
              "      <td>2.5960</td>\n",
              "      <td>0.3677</td>\n",
              "      <td>-0.3854</td>\n",
              "      <td>-1.1590</td>\n",
              "      <td>-8.180</td>\n",
              "      <td>-2.1310</td>\n",
              "      <td>-4.168</td>\n",
              "      <td>2.5630</td>\n",
              "      <td>-1.3410</td>\n",
              "      <td>5.2610</td>\n",
              "      <td>-2.5680</td>\n",
              "      <td>1.0840</td>\n",
              "      <td>-3.835</td>\n",
              "      <td>-4.8440</td>\n",
              "      <td>0.8522</td>\n",
              "      <td>0.2435</td>\n",
              "      <td>-2.6090</td>\n",
              "      <td>0.3067</td>\n",
              "      <td>0.9733</td>\n",
              "      <td>-1.6080</td>\n",
              "      <td>-2.9210</td>\n",
              "      <td>-1.9570</td>\n",
              "      <td>-2.644</td>\n",
              "      <td>2.4030</td>\n",
              "      <td>-3.036</td>\n",
              "      <td>-6.4520</td>\n",
              "      <td>-3.258</td>\n",
              "      <td>4.355</td>\n",
              "      <td>-0.7562</td>\n",
              "      <td>0.9930</td>\n",
              "      <td>-0.2854</td>\n",
              "      <td>...</td>\n",
              "      <td>-7.370</td>\n",
              "      <td>-9.995</td>\n",
              "      <td>-6.135</td>\n",
              "      <td>-9.986</td>\n",
              "      <td>-8.504</td>\n",
              "      <td>-8.528</td>\n",
              "      <td>-9.995</td>\n",
              "      <td>-6.254</td>\n",
              "      <td>-9.995</td>\n",
              "      <td>-6.955</td>\n",
              "      <td>-9.995</td>\n",
              "      <td>-9.743</td>\n",
              "      <td>-3.457</td>\n",
              "      <td>-8.712</td>\n",
              "      <td>-3.972</td>\n",
              "      <td>-8.709</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.461</td>\n",
              "      <td>-9.347</td>\n",
              "      <td>-8.698</td>\n",
              "      <td>-9.807</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-5.657</td>\n",
              "      <td>-5.513</td>\n",
              "      <td>-8.284</td>\n",
              "      <td>-9.490</td>\n",
              "      <td>-8.729</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-7.834</td>\n",
              "      <td>-9.538</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.520</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-7.436</td>\n",
              "      <td>-6.918</td>\n",
              "      <td>-5.235</td>\n",
              "      <td>2</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>119</th>\n",
              "      <td>id_01534c4ee</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>72</td>\n",
              "      <td>D2</td>\n",
              "      <td>87d714366</td>\n",
              "      <td>10.000</td>\n",
              "      <td>-1.2200</td>\n",
              "      <td>1.5150</td>\n",
              "      <td>-2.1910</td>\n",
              "      <td>4.3530</td>\n",
              "      <td>-0.9594</td>\n",
              "      <td>2.5600</td>\n",
              "      <td>-3.3120</td>\n",
              "      <td>-9.507</td>\n",
              "      <td>-3.3930</td>\n",
              "      <td>-4.160</td>\n",
              "      <td>-2.9320</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>2.2650</td>\n",
              "      <td>-6.2560</td>\n",
              "      <td>0.1147</td>\n",
              "      <td>-4.209</td>\n",
              "      <td>-3.5580</td>\n",
              "      <td>3.2920</td>\n",
              "      <td>0.1551</td>\n",
              "      <td>-3.7200</td>\n",
              "      <td>0.1069</td>\n",
              "      <td>1.3000</td>\n",
              "      <td>1.9710</td>\n",
              "      <td>-9.9350</td>\n",
              "      <td>-1.8340</td>\n",
              "      <td>-1.072</td>\n",
              "      <td>0.2092</td>\n",
              "      <td>-1.989</td>\n",
              "      <td>-4.0390</td>\n",
              "      <td>-6.047</td>\n",
              "      <td>8.404</td>\n",
              "      <td>3.1040</td>\n",
              "      <td>-3.3390</td>\n",
              "      <td>-2.8780</td>\n",
              "      <td>...</td>\n",
              "      <td>-9.978</td>\n",
              "      <td>-9.978</td>\n",
              "      <td>-8.823</td>\n",
              "      <td>-9.978</td>\n",
              "      <td>-9.513</td>\n",
              "      <td>-9.832</td>\n",
              "      <td>-9.978</td>\n",
              "      <td>-6.568</td>\n",
              "      <td>-9.978</td>\n",
              "      <td>-8.664</td>\n",
              "      <td>-9.978</td>\n",
              "      <td>-9.978</td>\n",
              "      <td>-4.073</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-5.163</td>\n",
              "      <td>-7.797</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.297</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.019</td>\n",
              "      <td>-8.688</td>\n",
              "      <td>-9.658</td>\n",
              "      <td>-9.308</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-7.703</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-9.054</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.771</td>\n",
              "      <td>-9.694</td>\n",
              "      <td>-5.548</td>\n",
              "      <td>2</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>209</th>\n",
              "      <td>id_0236e3425</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>72</td>\n",
              "      <td>D1</td>\n",
              "      <td>87d714366</td>\n",
              "      <td>3.612</td>\n",
              "      <td>-1.7130</td>\n",
              "      <td>0.0313</td>\n",
              "      <td>-1.7360</td>\n",
              "      <td>3.3550</td>\n",
              "      <td>-1.7040</td>\n",
              "      <td>2.6010</td>\n",
              "      <td>-1.4530</td>\n",
              "      <td>-9.534</td>\n",
              "      <td>-2.8730</td>\n",
              "      <td>-4.465</td>\n",
              "      <td>-0.4308</td>\n",
              "      <td>-1.7920</td>\n",
              "      <td>3.8900</td>\n",
              "      <td>-3.4910</td>\n",
              "      <td>0.6124</td>\n",
              "      <td>-3.504</td>\n",
              "      <td>-4.4450</td>\n",
              "      <td>-1.4690</td>\n",
              "      <td>-2.8590</td>\n",
              "      <td>-1.4090</td>\n",
              "      <td>0.1560</td>\n",
              "      <td>-3.1610</td>\n",
              "      <td>-1.5710</td>\n",
              "      <td>0.0587</td>\n",
              "      <td>-0.0839</td>\n",
              "      <td>-3.619</td>\n",
              "      <td>-0.2986</td>\n",
              "      <td>-2.164</td>\n",
              "      <td>-2.1390</td>\n",
              "      <td>-3.929</td>\n",
              "      <td>7.451</td>\n",
              "      <td>-1.0450</td>\n",
              "      <td>4.2260</td>\n",
              "      <td>-3.5800</td>\n",
              "      <td>...</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-9.974</td>\n",
              "      <td>-9.062</td>\n",
              "      <td>-9.896</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-9.553</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-7.088</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-4.432</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-6.025</td>\n",
              "      <td>-8.882</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.023</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.517</td>\n",
              "      <td>-7.589</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.451</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.695</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.559</td>\n",
              "      <td>-8.025</td>\n",
              "      <td>-6.642</td>\n",
              "      <td>2</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>218</th>\n",
              "      <td>id_024a52260</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>24</td>\n",
              "      <td>D2</td>\n",
              "      <td>87d714366</td>\n",
              "      <td>2.119</td>\n",
              "      <td>0.1098</td>\n",
              "      <td>1.1040</td>\n",
              "      <td>5.0620</td>\n",
              "      <td>-0.7427</td>\n",
              "      <td>-0.3564</td>\n",
              "      <td>-0.8253</td>\n",
              "      <td>-3.6180</td>\n",
              "      <td>-1.127</td>\n",
              "      <td>-2.1360</td>\n",
              "      <td>-3.371</td>\n",
              "      <td>5.4820</td>\n",
              "      <td>0.0537</td>\n",
              "      <td>8.0940</td>\n",
              "      <td>-1.3650</td>\n",
              "      <td>-1.4740</td>\n",
              "      <td>-4.195</td>\n",
              "      <td>0.0338</td>\n",
              "      <td>-0.2759</td>\n",
              "      <td>0.3821</td>\n",
              "      <td>-0.6859</td>\n",
              "      <td>-2.6870</td>\n",
              "      <td>0.1784</td>\n",
              "      <td>-1.0710</td>\n",
              "      <td>-4.1710</td>\n",
              "      <td>1.5710</td>\n",
              "      <td>-1.949</td>\n",
              "      <td>-1.0180</td>\n",
              "      <td>-4.791</td>\n",
              "      <td>-5.3540</td>\n",
              "      <td>-5.922</td>\n",
              "      <td>2.569</td>\n",
              "      <td>-1.5100</td>\n",
              "      <td>-2.7960</td>\n",
              "      <td>3.0740</td>\n",
              "      <td>...</td>\n",
              "      <td>-6.928</td>\n",
              "      <td>-9.313</td>\n",
              "      <td>-2.972</td>\n",
              "      <td>-8.412</td>\n",
              "      <td>-6.681</td>\n",
              "      <td>-7.150</td>\n",
              "      <td>-6.109</td>\n",
              "      <td>-4.503</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-5.367</td>\n",
              "      <td>-5.762</td>\n",
              "      <td>-5.173</td>\n",
              "      <td>-1.988</td>\n",
              "      <td>-7.265</td>\n",
              "      <td>-6.189</td>\n",
              "      <td>-5.044</td>\n",
              "      <td>-8.035</td>\n",
              "      <td>-4.652</td>\n",
              "      <td>-4.068</td>\n",
              "      <td>-6.564</td>\n",
              "      <td>-5.805</td>\n",
              "      <td>-8.420</td>\n",
              "      <td>-5.524</td>\n",
              "      <td>-6.696</td>\n",
              "      <td>-2.777</td>\n",
              "      <td>-3.348</td>\n",
              "      <td>-4.720</td>\n",
              "      <td>-5.357</td>\n",
              "      <td>-8.619</td>\n",
              "      <td>-7.332</td>\n",
              "      <td>-7.828</td>\n",
              "      <td>-9.679</td>\n",
              "      <td>-8.439</td>\n",
              "      <td>-2.047</td>\n",
              "      <td>-7.430</td>\n",
              "      <td>-5.861</td>\n",
              "      <td>-4.591</td>\n",
              "      <td>-3.533</td>\n",
              "      <td>2</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23680</th>\n",
              "      <td>id_fe8abffe0</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>72</td>\n",
              "      <td>D1</td>\n",
              "      <td>87d714366</td>\n",
              "      <td>6.649</td>\n",
              "      <td>-0.4462</td>\n",
              "      <td>1.7370</td>\n",
              "      <td>0.0553</td>\n",
              "      <td>3.0330</td>\n",
              "      <td>-2.7360</td>\n",
              "      <td>0.4016</td>\n",
              "      <td>-4.5340</td>\n",
              "      <td>-9.746</td>\n",
              "      <td>-3.1160</td>\n",
              "      <td>-5.381</td>\n",
              "      <td>2.1980</td>\n",
              "      <td>-1.4490</td>\n",
              "      <td>6.8380</td>\n",
              "      <td>-5.0190</td>\n",
              "      <td>0.7382</td>\n",
              "      <td>-4.690</td>\n",
              "      <td>-4.5760</td>\n",
              "      <td>-1.5880</td>\n",
              "      <td>-3.1120</td>\n",
              "      <td>-5.5090</td>\n",
              "      <td>-1.4490</td>\n",
              "      <td>-2.2670</td>\n",
              "      <td>-1.7240</td>\n",
              "      <td>-1.2190</td>\n",
              "      <td>-2.5050</td>\n",
              "      <td>-5.049</td>\n",
              "      <td>0.9918</td>\n",
              "      <td>-4.635</td>\n",
              "      <td>-1.8910</td>\n",
              "      <td>-3.750</td>\n",
              "      <td>4.339</td>\n",
              "      <td>-0.5515</td>\n",
              "      <td>7.0970</td>\n",
              "      <td>-2.5690</td>\n",
              "      <td>...</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-9.839</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.596</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-5.098</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-5.222</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-8.969</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.152</td>\n",
              "      <td>-9.432</td>\n",
              "      <td>-8.110</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-8.830</td>\n",
              "      <td>-9.910</td>\n",
              "      <td>-7.807</td>\n",
              "      <td>2</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23727</th>\n",
              "      <td>id_ff1d43f96</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>24</td>\n",
              "      <td>D2</td>\n",
              "      <td>87d714366</td>\n",
              "      <td>3.263</td>\n",
              "      <td>1.4620</td>\n",
              "      <td>0.2311</td>\n",
              "      <td>3.3590</td>\n",
              "      <td>1.7250</td>\n",
              "      <td>0.4117</td>\n",
              "      <td>-0.4033</td>\n",
              "      <td>-0.6053</td>\n",
              "      <td>-3.445</td>\n",
              "      <td>-1.5280</td>\n",
              "      <td>-3.771</td>\n",
              "      <td>6.1560</td>\n",
              "      <td>1.4140</td>\n",
              "      <td>6.3520</td>\n",
              "      <td>0.2650</td>\n",
              "      <td>-1.3610</td>\n",
              "      <td>-4.642</td>\n",
              "      <td>-2.8910</td>\n",
              "      <td>0.6158</td>\n",
              "      <td>1.5470</td>\n",
              "      <td>-3.2930</td>\n",
              "      <td>0.3337</td>\n",
              "      <td>-0.6316</td>\n",
              "      <td>-2.0770</td>\n",
              "      <td>0.7569</td>\n",
              "      <td>0.7684</td>\n",
              "      <td>-3.064</td>\n",
              "      <td>-1.4090</td>\n",
              "      <td>-4.214</td>\n",
              "      <td>-5.5860</td>\n",
              "      <td>-4.665</td>\n",
              "      <td>1.429</td>\n",
              "      <td>0.2997</td>\n",
              "      <td>-2.0450</td>\n",
              "      <td>1.8980</td>\n",
              "      <td>...</td>\n",
              "      <td>-4.636</td>\n",
              "      <td>-9.106</td>\n",
              "      <td>-3.191</td>\n",
              "      <td>-7.704</td>\n",
              "      <td>-5.721</td>\n",
              "      <td>-6.229</td>\n",
              "      <td>-6.879</td>\n",
              "      <td>-4.248</td>\n",
              "      <td>-9.957</td>\n",
              "      <td>-5.353</td>\n",
              "      <td>-7.343</td>\n",
              "      <td>-5.936</td>\n",
              "      <td>-3.006</td>\n",
              "      <td>-8.881</td>\n",
              "      <td>-4.543</td>\n",
              "      <td>-6.545</td>\n",
              "      <td>-8.181</td>\n",
              "      <td>-4.239</td>\n",
              "      <td>-4.828</td>\n",
              "      <td>-5.770</td>\n",
              "      <td>-5.915</td>\n",
              "      <td>-8.183</td>\n",
              "      <td>-4.614</td>\n",
              "      <td>-5.767</td>\n",
              "      <td>-4.699</td>\n",
              "      <td>-4.526</td>\n",
              "      <td>-4.098</td>\n",
              "      <td>-5.579</td>\n",
              "      <td>-8.000</td>\n",
              "      <td>-9.344</td>\n",
              "      <td>-7.395</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-7.359</td>\n",
              "      <td>-1.574</td>\n",
              "      <td>-6.896</td>\n",
              "      <td>-5.001</td>\n",
              "      <td>-4.139</td>\n",
              "      <td>-3.271</td>\n",
              "      <td>2</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23776</th>\n",
              "      <td>id_ff9d0a022</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>24</td>\n",
              "      <td>D2</td>\n",
              "      <td>87d714366</td>\n",
              "      <td>1.227</td>\n",
              "      <td>0.3781</td>\n",
              "      <td>2.8860</td>\n",
              "      <td>7.1470</td>\n",
              "      <td>-0.4875</td>\n",
              "      <td>-1.2070</td>\n",
              "      <td>0.7124</td>\n",
              "      <td>-2.9770</td>\n",
              "      <td>-2.879</td>\n",
              "      <td>0.8712</td>\n",
              "      <td>10.000</td>\n",
              "      <td>5.5150</td>\n",
              "      <td>-1.9930</td>\n",
              "      <td>5.1930</td>\n",
              "      <td>-0.6509</td>\n",
              "      <td>-1.5610</td>\n",
              "      <td>-1.137</td>\n",
              "      <td>-1.4880</td>\n",
              "      <td>0.2867</td>\n",
              "      <td>0.2786</td>\n",
              "      <td>-0.9956</td>\n",
              "      <td>-1.3100</td>\n",
              "      <td>-0.6124</td>\n",
              "      <td>-1.4480</td>\n",
              "      <td>-3.2240</td>\n",
              "      <td>1.0280</td>\n",
              "      <td>-1.423</td>\n",
              "      <td>-2.7110</td>\n",
              "      <td>-4.183</td>\n",
              "      <td>-3.8860</td>\n",
              "      <td>-5.463</td>\n",
              "      <td>3.662</td>\n",
              "      <td>-1.1530</td>\n",
              "      <td>-1.2060</td>\n",
              "      <td>3.5390</td>\n",
              "      <td>...</td>\n",
              "      <td>-7.280</td>\n",
              "      <td>-9.326</td>\n",
              "      <td>-4.024</td>\n",
              "      <td>-8.476</td>\n",
              "      <td>-7.526</td>\n",
              "      <td>-9.182</td>\n",
              "      <td>-6.765</td>\n",
              "      <td>-5.493</td>\n",
              "      <td>-9.657</td>\n",
              "      <td>-5.448</td>\n",
              "      <td>-8.192</td>\n",
              "      <td>-5.703</td>\n",
              "      <td>-2.689</td>\n",
              "      <td>-7.787</td>\n",
              "      <td>-7.416</td>\n",
              "      <td>-4.585</td>\n",
              "      <td>-8.108</td>\n",
              "      <td>-4.620</td>\n",
              "      <td>-7.661</td>\n",
              "      <td>-8.309</td>\n",
              "      <td>-6.302</td>\n",
              "      <td>-9.355</td>\n",
              "      <td>-5.185</td>\n",
              "      <td>-8.082</td>\n",
              "      <td>-3.078</td>\n",
              "      <td>-3.840</td>\n",
              "      <td>-5.908</td>\n",
              "      <td>-5.867</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-9.328</td>\n",
              "      <td>-7.380</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-8.754</td>\n",
              "      <td>-1.938</td>\n",
              "      <td>-7.529</td>\n",
              "      <td>-6.318</td>\n",
              "      <td>-5.701</td>\n",
              "      <td>-3.561</td>\n",
              "      <td>2</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23791</th>\n",
              "      <td>id_ffd1fd119</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>72</td>\n",
              "      <td>D2</td>\n",
              "      <td>87d714366</td>\n",
              "      <td>5.312</td>\n",
              "      <td>0.4505</td>\n",
              "      <td>1.5320</td>\n",
              "      <td>0.8226</td>\n",
              "      <td>3.0250</td>\n",
              "      <td>-0.4502</td>\n",
              "      <td>-0.8196</td>\n",
              "      <td>-1.7550</td>\n",
              "      <td>-9.205</td>\n",
              "      <td>-2.4890</td>\n",
              "      <td>-2.743</td>\n",
              "      <td>0.3039</td>\n",
              "      <td>-3.1820</td>\n",
              "      <td>3.7120</td>\n",
              "      <td>-3.5190</td>\n",
              "      <td>1.3900</td>\n",
              "      <td>-2.796</td>\n",
              "      <td>-4.3670</td>\n",
              "      <td>-1.6070</td>\n",
              "      <td>0.6094</td>\n",
              "      <td>-5.0060</td>\n",
              "      <td>-0.6930</td>\n",
              "      <td>1.2240</td>\n",
              "      <td>0.5679</td>\n",
              "      <td>-3.9940</td>\n",
              "      <td>-3.0080</td>\n",
              "      <td>-1.883</td>\n",
              "      <td>3.5390</td>\n",
              "      <td>-2.899</td>\n",
              "      <td>-9.0610</td>\n",
              "      <td>-5.198</td>\n",
              "      <td>5.802</td>\n",
              "      <td>1.3620</td>\n",
              "      <td>0.8115</td>\n",
              "      <td>-2.0090</td>\n",
              "      <td>...</td>\n",
              "      <td>-9.033</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-9.795</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-9.436</td>\n",
              "      <td>-9.918</td>\n",
              "      <td>-7.790</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-9.802</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-10.000</td>\n",
              "      <td>-5.340</td>\n",
              "      <td>-9.852</td>\n",
              "      <td>-8.071</td>\n",
              "      <td>-8.019</td>\n",
              "      <td>-9.966</td>\n",
              "      <td>-9.410</td>\n",
              "      <td>-9.966</td>\n",
              "      <td>-9.817</td>\n",
              "      <td>-9.966</td>\n",
              "      <td>-9.966</td>\n",
              "      <td>-9.966</td>\n",
              "      <td>-9.966</td>\n",
              "      <td>-8.638</td>\n",
              "      <td>-8.966</td>\n",
              "      <td>-9.171</td>\n",
              "      <td>-8.631</td>\n",
              "      <td>-9.966</td>\n",
              "      <td>-9.966</td>\n",
              "      <td>-9.220</td>\n",
              "      <td>-9.373</td>\n",
              "      <td>-9.966</td>\n",
              "      <td>-9.362</td>\n",
              "      <td>-9.966</td>\n",
              "      <td>-8.642</td>\n",
              "      <td>-9.966</td>\n",
              "      <td>-4.605</td>\n",
              "      <td>2</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23802</th>\n",
              "      <td>id_fff3976bd</td>\n",
              "      <td>com_droga</td>\n",
              "      <td>24</td>\n",
              "      <td>D1</td>\n",
              "      <td>87d714366</td>\n",
              "      <td>3.289</td>\n",
              "      <td>2.1270</td>\n",
              "      <td>0.9770</td>\n",
              "      <td>2.0490</td>\n",
              "      <td>2.3170</td>\n",
              "      <td>-0.5355</td>\n",
              "      <td>-0.6357</td>\n",
              "      <td>-6.9060</td>\n",
              "      <td>-4.483</td>\n",
              "      <td>-2.8400</td>\n",
              "      <td>-5.331</td>\n",
              "      <td>0.6547</td>\n",
              "      <td>-2.2080</td>\n",
              "      <td>3.6030</td>\n",
              "      <td>0.6380</td>\n",
              "      <td>-2.1280</td>\n",
              "      <td>-3.168</td>\n",
              "      <td>-4.2180</td>\n",
              "      <td>-0.3170</td>\n",
              "      <td>0.2140</td>\n",
              "      <td>-1.7660</td>\n",
              "      <td>-1.9700</td>\n",
              "      <td>-1.4820</td>\n",
              "      <td>-1.3710</td>\n",
              "      <td>-0.2628</td>\n",
              "      <td>-0.5276</td>\n",
              "      <td>-2.416</td>\n",
              "      <td>-0.4638</td>\n",
              "      <td>-3.863</td>\n",
              "      <td>-0.8880</td>\n",
              "      <td>-1.985</td>\n",
              "      <td>2.053</td>\n",
              "      <td>-1.4700</td>\n",
              "      <td>-3.3220</td>\n",
              "      <td>1.8800</td>\n",
              "      <td>...</td>\n",
              "      <td>-6.756</td>\n",
              "      <td>-9.917</td>\n",
              "      <td>-4.124</td>\n",
              "      <td>-8.511</td>\n",
              "      <td>-5.750</td>\n",
              "      <td>-8.476</td>\n",
              "      <td>-6.947</td>\n",
              "      <td>-4.765</td>\n",
              "      <td>-9.769</td>\n",
              "      <td>-5.790</td>\n",
              "      <td>-9.917</td>\n",
              "      <td>-5.745</td>\n",
              "      <td>-2.357</td>\n",
              "      <td>-9.119</td>\n",
              "      <td>-4.598</td>\n",
              "      <td>-8.776</td>\n",
              "      <td>-7.798</td>\n",
              "      <td>-4.162</td>\n",
              "      <td>-6.151</td>\n",
              "      <td>-5.740</td>\n",
              "      <td>-6.870</td>\n",
              "      <td>-9.100</td>\n",
              "      <td>-3.336</td>\n",
              "      <td>-6.642</td>\n",
              "      <td>-2.275</td>\n",
              "      <td>-3.472</td>\n",
              "      <td>-6.267</td>\n",
              "      <td>-6.398</td>\n",
              "      <td>-8.630</td>\n",
              "      <td>-8.149</td>\n",
              "      <td>-6.290</td>\n",
              "      <td>-8.097</td>\n",
              "      <td>-8.733</td>\n",
              "      <td>-2.378</td>\n",
              "      <td>-8.448</td>\n",
              "      <td>-4.455</td>\n",
              "      <td>-5.793</td>\n",
              "      <td>-3.700</td>\n",
              "      <td>2</td>\n",
              "      <td>True</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>718 rows × 879 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                 id tratamento  tempo dose  ...   c-98   c-99  n_moa  ativo_moa\n",
              "16     id_002fb9c19  com_droga     48   D1  ... -9.984 -6.784      2       True\n",
              "76     id_00cf039a7  com_droga     48   D2  ... -6.918 -5.235      2       True\n",
              "119    id_01534c4ee  com_droga     72   D2  ... -9.694 -5.548      2       True\n",
              "209    id_0236e3425  com_droga     72   D1  ... -8.025 -6.642      2       True\n",
              "218    id_024a52260  com_droga     24   D2  ... -4.591 -3.533      2       True\n",
              "...             ...        ...    ...  ...  ...    ...    ...    ...        ...\n",
              "23680  id_fe8abffe0  com_droga     72   D1  ... -9.910 -7.807      2       True\n",
              "23727  id_ff1d43f96  com_droga     24   D2  ... -4.139 -3.271      2       True\n",
              "23776  id_ff9d0a022  com_droga     24   D2  ... -5.701 -3.561      2       True\n",
              "23791  id_ffd1fd119  com_droga     72   D2  ... -9.966 -4.605      2       True\n",
              "23802  id_fff3976bd  com_droga     24   D1  ... -5.793 -3.700      2       True\n",
              "\n",
              "[718 rows x 879 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 72
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xF7nuC2bxCbd",
        "outputId": "098d6906-4005-4afe-d5bf-d9bcf5ee3d57"
      },
      "source": [
        "#retirando variavel tratamento do conjunto de variaveis independentes\n",
        "x = dados_filtro_composto.drop(['id', 'n_moa', 'ativo_moa', 'composto', 'tratamento'], axis=1)\n",
        "#variáveis independentes\n",
        "x = pd.get_dummies(x, columns=['dose', 'tempo'])\n",
        "#variável dependente\n",
        "y = dados_filtro_composto['tratamento']\n",
        "#separando dados em treino e teste, teste: 20% e treino: 80%\n",
        "x_treino, x_teste, y_treino, y_teste = train_test_split(x, y, test_size = 0.2, stratify=y, random_state=42)\n",
        "\n",
        "modelo_randomforest = RandomForestClassifier()\n",
        "#treinando o modelo\n",
        "modelo_randomforest.fit(x_treino, y_treino)\n",
        "#acurácia do modelo\n",
        "modelo_randomforest.score(x_teste, y_teste)"
      ],
      "execution_count": 73,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 73
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AELMx0FVFzlG"
      },
      "source": [
        "Surpreendentemente, utilizando a drogra mais usada para prever seu tratamento, o modelo acerta 100% das previsões. Não tenho certeza se o modelo está correto, mas este resultado é de se espantar."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "X_Lg2XTu20ND"
      },
      "source": [
        "# **Desafio 6**\n",
        "\n",
        "Na aula 6 tivemos uma introdução aos métodos de Machine Learning no Python. Pudemos ver na prática como se faz a predição de valores num dataset e como manipular o dataset para tal procedimento. Pudemos testar diferentes métodos para prever parâmetros dos ensaios e observar que para cada problema, um modelo pode se adequar melhor, tendo uma acurácia mais alta. \n",
        "\n",
        "Assim, cabe ao cientista de dados conhecer as ferramentas disponíveis e aplicar da melhor forma  para se obter sucesso nos seus modelos preditivos."
      ]
    }
  ]
}