ML_validation_optimization.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 339
        },
        "colab_type": "code",
        "id": "yku6GxC9PujF",
        "outputId": "03c85f56-cc2c-44cc-f4e0-2ef493d04e05"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: graphviz==0.9 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (0.9)\n",
            "Requirement already satisfied: pydot in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (1.4.2)\n",
            "Requirement already satisfied: pyparsing>=2.1.4 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from pydot) (3.0.9)\n",
            "Requirement already satisfied: seaborn==0.9.0 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (0.9.0)\n",
            "Requirement already satisfied: matplotlib>=1.4.3 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from seaborn==0.9.0) (3.5.2)\n",
            "Requirement already satisfied: scipy>=0.14.0 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from seaborn==0.9.0) (1.9.1)\n",
            "Requirement already satisfied: numpy>=1.9.3 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from seaborn==0.9.0) (1.21.5)\n",
            "Requirement already satisfied: pandas>=0.15.2 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from seaborn==0.9.0) (1.4.4)\n",
            "Requirement already satisfied: pyparsing>=2.2.1 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (3.0.9)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (1.4.2)\n",
            "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (2.8.2)\n",
            "Requirement already satisfied: packaging>=20.0 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (21.3)\n",
            "Requirement already satisfied: fonttools>=4.22.0 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (4.25.0)\n",
            "Requirement already satisfied: cycler>=0.10 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (0.11.0)\n",
            "Requirement already satisfied: pillow>=6.2.0 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (9.2.0)\n",
            "Requirement already satisfied: pytz>=2020.1 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from pandas>=0.15.2->seaborn==0.9.0) (2022.1)\n",
            "Requirement already satisfied: six>=1.5 in c:\\users\\felipe martins\\anaconda3\\lib\\site-packages (from python-dateutil>=2.7->matplotlib>=1.4.3->seaborn==0.9.0) (1.16.0)\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "'apt-get' n�o � reconhecido como um comando interno\n",
            "ou externo, um programa oper�vel ou um arquivo em lotes.\n"
          ]
        }
      ],
      "source": [
        "!pip install graphviz==0.9\n",
        "!pip install pydot\n",
        "!pip install seaborn==0.9.0\n",
        "\n",
        "!apt-get install graphviz"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "colab_type": "code",
        "id": "y7ik04NlDZMA",
        "outputId": "3e60f174-f855-49bd-d378-60b6d840638b"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>preco</th>\n",
              "      <th>vendido</th>\n",
              "      <th>idade_do_modelo</th>\n",
              "      <th>km_por_ano</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>30941.02</td>\n",
              "      <td>1</td>\n",
              "      <td>18</td>\n",
              "      <td>35085.22134</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>40557.96</td>\n",
              "      <td>1</td>\n",
              "      <td>20</td>\n",
              "      <td>12622.05362</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>89627.50</td>\n",
              "      <td>0</td>\n",
              "      <td>12</td>\n",
              "      <td>11440.79806</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>95276.14</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>43167.32682</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>117384.68</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>12770.11290</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       preco  vendido  idade_do_modelo   km_por_ano\n",
              "0   30941.02        1               18  35085.22134\n",
              "1   40557.96        1               20  12622.05362\n",
              "2   89627.50        0               12  11440.79806\n",
              "3   95276.14        0                3  43167.32682\n",
              "4  117384.68        1                4  12770.11290"
            ]
          },
          "execution_count": 10,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "import pandas as pd\n",
        "\n",
        "uri = \"https://gist.githubusercontent.com/guilhermesilveira/e99a526b2e7ccc6c3b70f53db43a87d2/raw/1605fc74aa778066bf2e6695e24d53cf65f2f447/machine-learning-carros-simulacao.csv\"\n",
        "dados = pd.read_csv(uri).drop(columns=[\"Unnamed: 0\"], axis=1)\n",
        "dados.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "colab_type": "code",
        "id": "b52E4e1hjegN",
        "outputId": "46e32461-13ce-4b78-d64f-96e0a89576ba"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>preco</th>\n",
              "      <th>vendido</th>\n",
              "      <th>idade_do_modelo</th>\n",
              "      <th>km_por_ano</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>4999</th>\n",
              "      <td>74023.29</td>\n",
              "      <td>0</td>\n",
              "      <td>12</td>\n",
              "      <td>24812.80412</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5322</th>\n",
              "      <td>84843.49</td>\n",
              "      <td>0</td>\n",
              "      <td>13</td>\n",
              "      <td>23095.63834</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5319</th>\n",
              "      <td>83100.27</td>\n",
              "      <td>0</td>\n",
              "      <td>19</td>\n",
              "      <td>36240.72746</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5316</th>\n",
              "      <td>87932.13</td>\n",
              "      <td>0</td>\n",
              "      <td>16</td>\n",
              "      <td>32249.56426</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5315</th>\n",
              "      <td>77937.01</td>\n",
              "      <td>0</td>\n",
              "      <td>15</td>\n",
              "      <td>28414.50704</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         preco  vendido  idade_do_modelo   km_por_ano\n",
              "4999  74023.29        0               12  24812.80412\n",
              "5322  84843.49        0               13  23095.63834\n",
              "5319  83100.27        0               19  36240.72746\n",
              "5316  87932.13        0               16  32249.56426\n",
              "5315  77937.01        0               15  28414.50704"
            ]
          },
          "execution_count": 11,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# situação horrível de \"azar\" onde as classes estão ordenadas por padrão\n",
        "\n",
        "dados_azar = dados.sort_values(\"vendido\", ascending=True)\n",
        "x_azar = dados_azar[[\"preco\", \"idade_do_modelo\",\"km_por_ano\"]]\n",
        "y_azar = dados_azar[\"vendido\"]\n",
        "dados_azar.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "colab_type": "code",
        "id": "ZNT26vFEeYmz",
        "outputId": "5cd367dd-e297-40cc-c81d-e09bbc7bf41f"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Accuracy com dummy stratified, 10 = [58.00, 58.00]\n"
          ]
        }
      ],
      "source": [
        "from sklearn.model_selection import cross_validate\n",
        "from sklearn.dummy import DummyClassifier\n",
        "import numpy as np\n",
        "\n",
        "SEED = 301\n",
        "np.random.seed(SEED)\n",
        "\n",
        "modelo = DummyClassifier()\n",
        "results = cross_validate(modelo, x_azar, y_azar, cv = 10, return_train_score=False)\n",
        "media = results['test_score'].mean()\n",
        "desvio_padrao = results['test_score'].std()\n",
        "print(\"Accuracy com dummy stratified, 10 = [%.2f, %.2f]\" % ((media - 2 * desvio_padrao)*100, (media + 2 * desvio_padrao) * 100))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "colab_type": "code",
        "id": "73mCcFA_eG_K",
        "outputId": "6f585741-6e82-4e9e-9d6a-d74ed8224503"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Accuracy com cross validation, 10 = [73.83, 77.73]\n"
          ]
        }
      ],
      "source": [
        "from sklearn.model_selection import cross_validate\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "\n",
        "SEED = 301\n",
        "np.random.seed(SEED)\n",
        "\n",
        "modelo = DecisionTreeClassifier(max_depth=2)\n",
        "results = cross_validate(modelo, x_azar, y_azar, cv = 10, return_train_score=False)\n",
        "media = results['test_score'].mean()\n",
        "desvio_padrao = results['test_score'].std()\n",
        "print(\"Accuracy com cross validation, 10 = [%.2f, %.2f]\" % ((media - 2 * desvio_padrao)*100, (media + 2 * desvio_padrao) * 100))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "colab_type": "code",
        "id": "5C8Y6J-PGpYf",
        "outputId": "dcf93fa7-ff87-4bd9-b4aa-a3dfab869f84"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>preco</th>\n",
              "      <th>vendido</th>\n",
              "      <th>idade_do_modelo</th>\n",
              "      <th>km_por_ano</th>\n",
              "      <th>modelo</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>30941.02</td>\n",
              "      <td>1</td>\n",
              "      <td>18</td>\n",
              "      <td>35085.22134</td>\n",
              "      <td>18</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>40557.96</td>\n",
              "      <td>1</td>\n",
              "      <td>20</td>\n",
              "      <td>12622.05362</td>\n",
              "      <td>24</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>89627.50</td>\n",
              "      <td>0</td>\n",
              "      <td>12</td>\n",
              "      <td>11440.79806</td>\n",
              "      <td>14</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>95276.14</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>43167.32682</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>117384.68</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>12770.11290</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       preco  vendido  idade_do_modelo   km_por_ano  modelo\n",
              "0   30941.02        1               18  35085.22134      18\n",
              "1   40557.96        1               20  12622.05362      24\n",
              "2   89627.50        0               12  11440.79806      14\n",
              "3   95276.14        0                3  43167.32682       6\n",
              "4  117384.68        1                4  12770.11290       5"
            ]
          },
          "execution_count": 14,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# gerando dados elatorios de modelo de carro para simulacao de agrupamento ao usar nosso estimador\n",
        "\n",
        "np.random.seed(SEED)\n",
        "dados['modelo'] = dados.idade_do_modelo + np.random.randint(-2, 3, size=10000)\n",
        "dados.modelo = dados.modelo + abs(dados.modelo.min()) + 1\n",
        "dados.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {},
        "colab_type": "code",
        "id": "3hmjt7qPHOZY"
      },
      "outputs": [],
      "source": [
        "def imprime_resultados(results):\n",
        "  media = results['test_score'].mean() * 100\n",
        "  desvio = results['test_score'].std() * 100\n",
        "  print(\"Accuracy médio %.2f\" % media)\n",
        "  print(\"Intervalo [%.2f, %.2f]\" % (media - 2 * desvio, media + 2 * desvio))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 50
        },
        "colab_type": "code",
        "id": "goijy0rSS7n-",
        "outputId": "bf0df175-81db-4ee7-8ae4-591a31eb66db"
      },
      "outputs": [
        {
          "ename": "NameError",
          "evalue": "name 'GroupKFold' is not defined",
          "output_type": "error",
          "traceback": [
            "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_13724\\1052820869.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     13\u001b[0m \u001b[0mpipeline\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mPipeline\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'transformacao'\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mscaler\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m(\u001b[0m\u001b[1;34m'estimador'\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mmodelo\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     14\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 15\u001b[1;33m \u001b[0mcv\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mGroupKFold\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mn_splits\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;36m10\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     16\u001b[0m \u001b[0mresults\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mcross_validate\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mpipeline\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mx_azar\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my_azar\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcv\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mcv\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mgroups\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mdados\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmodelo\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mreturn_train_score\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mFalse\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     17\u001b[0m \u001b[0mimprime_resultados\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mresults\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
            "\u001b[1;31mNameError\u001b[0m: name 'GroupKFold' is not defined"
          ]
        }
      ],
      "source": [
        "# GroupKFold em um pipeline com StandardScaler e SVC\n",
        "\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.pipeline import Pipeline\n",
        "\n",
        "SEED = 301\n",
        "np.random.seed(SEED)\n",
        "\n",
        "scaler = StandardScaler()\n",
        "modelo = SVC()\n",
        "\n",
        "pipeline = Pipeline([('transformacao',scaler), ('estimador',modelo)])\n",
        "\n",
        "cv = GroupKFold(n_splits = 10)\n",
        "results = cross_validate(pipeline, x_azar, y_azar, cv = cv, groups = dados.modelo, return_train_score=False)\n",
        "imprime_resultados(results)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 50
        },
        "colab_type": "code",
        "id": "IKD8uaWkNcUL",
        "outputId": "c5430f13-067a-4d6a-e840-46d4669e931a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Accuracy médio 75.78\n",
            "Intervalo [73.67, 77.90]\n"
          ]
        }
      ],
      "source": [
        "# GroupKFold para analisar como o modelo se comporta com novos grupos\n",
        "\n",
        "from sklearn.model_selection import GroupKFold\n",
        "\n",
        "SEED = 301\n",
        "np.random.seed(SEED)\n",
        "\n",
        "cv = GroupKFold(n_splits = 10)\n",
        "modelo = DecisionTreeClassifier(max_depth=2)\n",
        "results = cross_validate(modelo, x_azar, y_azar, cv = cv, groups = dados.modelo, return_train_score=False)\n",
        "imprime_resultados(results)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 440
        },
        "colab_type": "code",
        "id": "WhJEIZ4bO3J6",
        "outputId": "9ac2fe01-dbbc-41d8-e4c4-b32ffeb75fd7"
      },
      "outputs": [
        {
          "data": {
            "image/svg+xml": [
              "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
              "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
              " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
              "<!-- Generated by graphviz version 2.40.1 (20161225.0304)\n",
              " -->\n",
              "<!-- Title: Tree Pages: 1 -->\n",
              "<svg width=\"628pt\" height=\"314pt\"\n",
              " viewBox=\"0.00 0.00 628.00 314.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
              "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 310)\">\n",
              "<title>Tree</title>\n",
              "<polygon fill=\"#ffffff\" stroke=\"transparent\" points=\"-4,4 -4,-310 624,-310 624,4 -4,4\"/>\n",
              "<!-- 0 -->\n",
              "<g id=\"node1\" class=\"node\">\n",
              "<title>0</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.274510\" stroke=\"#000000\" d=\"M361.5,-306C361.5,-306 236.5,-306 236.5,-306 230.5,-306 224.5,-300 224.5,-294 224.5,-294 224.5,-235 224.5,-235 224.5,-229 230.5,-223 236.5,-223 236.5,-223 361.5,-223 361.5,-223 367.5,-223 373.5,-229 373.5,-235 373.5,-235 373.5,-294 373.5,-294 373.5,-300 367.5,-306 361.5,-306\"/>\n",
              "<text text-anchor=\"middle\" x=\"299\" y=\"-290.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">preco &lt;= 59982.246</text>\n",
              "<text text-anchor=\"middle\" x=\"299\" y=\"-275.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.487</text>\n",
              "<text text-anchor=\"middle\" x=\"299\" y=\"-260.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 10000</text>\n",
              "<text text-anchor=\"middle\" x=\"299\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [4200, 5800]</text>\n",
              "<text text-anchor=\"middle\" x=\"299\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 1 -->\n",
              "<g id=\"node2\" class=\"node\">\n",
              "<title>1</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.874510\" stroke=\"#000000\" d=\"M266.5,-187C266.5,-187 143.5,-187 143.5,-187 137.5,-187 131.5,-181 131.5,-175 131.5,-175 131.5,-116 131.5,-116 131.5,-110 137.5,-104 143.5,-104 143.5,-104 266.5,-104 266.5,-104 272.5,-104 278.5,-110 278.5,-116 278.5,-116 278.5,-175 278.5,-175 278.5,-181 272.5,-187 266.5,-187\"/>\n",
              "<text text-anchor=\"middle\" x=\"205\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">preco &lt;= 40083.383</text>\n",
              "<text text-anchor=\"middle\" x=\"205\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.198</text>\n",
              "<text text-anchor=\"middle\" x=\"205\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 4347</text>\n",
              "<text text-anchor=\"middle\" x=\"205\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [484, 3863]</text>\n",
              "<text text-anchor=\"middle\" x=\"205\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 0&#45;&gt;1 -->\n",
              "<g id=\"edge1\" class=\"edge\">\n",
              "<title>0&#45;&gt;1</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M266.1234,-222.8796C259.0803,-213.9633 251.5707,-204.4565 244.3126,-195.268\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"246.9621,-192.9757 238.017,-187.2981 241.4691,-197.3147 246.9621,-192.9757\"/>\n",
              "<text text-anchor=\"middle\" x=\"235.1027\" y=\"-208.4276\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">True</text>\n",
              "</g>\n",
              "<!-- 4 -->\n",
              "<g id=\"node5\" class=\"node\">\n",
              "<title>4</title>\n",
              "<path fill=\"#e58139\" fill-opacity=\"0.478431\" stroke=\"#000000\" d=\"M477,-187C477,-187 309,-187 309,-187 303,-187 297,-181 297,-175 297,-175 297,-116 297,-116 297,-110 303,-104 309,-104 309,-104 477,-104 477,-104 483,-104 489,-110 489,-116 489,-116 489,-175 489,-175 489,-181 483,-187 477,-187\"/>\n",
              "<text text-anchor=\"middle\" x=\"393\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">km_por_ano &lt;= 24124.006</text>\n",
              "<text text-anchor=\"middle\" x=\"393\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.45</text>\n",
              "<text text-anchor=\"middle\" x=\"393\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 5653</text>\n",
              "<text text-anchor=\"middle\" x=\"393\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [3716, 1937]</text>\n",
              "<text text-anchor=\"middle\" x=\"393\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = não</text>\n",
              "</g>\n",
              "<!-- 0&#45;&gt;4 -->\n",
              "<g id=\"edge4\" class=\"edge\">\n",
              "<title>0&#45;&gt;4</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M331.8766,-222.8796C338.9197,-213.9633 346.4293,-204.4565 353.6874,-195.268\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"356.5309,-197.3147 359.983,-187.2981 351.0379,-192.9757 356.5309,-197.3147\"/>\n",
              "<text text-anchor=\"middle\" x=\"362.8973\" y=\"-208.4276\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">False</text>\n",
              "</g>\n",
              "<!-- 2 -->\n",
              "<g id=\"node3\" class=\"node\">\n",
              "<title>2</title>\n",
              "<path fill=\"#399de5\" stroke=\"#000000\" d=\"M112,-68C112,-68 12,-68 12,-68 6,-68 0,-62 0,-56 0,-56 0,-12 0,-12 0,-6 6,0 12,0 12,0 112,0 112,0 118,0 124,-6 124,-12 124,-12 124,-56 124,-56 124,-62 118,-68 112,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"62\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.0</text>\n",
              "<text text-anchor=\"middle\" x=\"62\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 1935</text>\n",
              "<text text-anchor=\"middle\" x=\"62\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [0, 1935]</text>\n",
              "<text text-anchor=\"middle\" x=\"62\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 1&#45;&gt;2 -->\n",
              "<g id=\"edge2\" class=\"edge\">\n",
              "<title>1&#45;&gt;2</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M151.7521,-103.9815C139.3756,-94.3313 126.2172,-84.0714 113.8997,-74.4673\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"115.9818,-71.6525 105.9435,-68.2637 111.6775,-77.1728 115.9818,-71.6525\"/>\n",
              "</g>\n",
              "<!-- 3 -->\n",
              "<g id=\"node4\" class=\"node\">\n",
              "<title>3</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.749020\" stroke=\"#000000\" d=\"M270,-68C270,-68 154,-68 154,-68 148,-68 142,-62 142,-56 142,-56 142,-12 142,-12 142,-6 148,0 154,0 154,0 270,0 270,0 276,0 282,-6 282,-12 282,-12 282,-56 282,-56 282,-62 276,-68 270,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"212\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.321</text>\n",
              "<text text-anchor=\"middle\" x=\"212\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 2412</text>\n",
              "<text text-anchor=\"middle\" x=\"212\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [484, 1928]</text>\n",
              "<text text-anchor=\"middle\" x=\"212\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 1&#45;&gt;3 -->\n",
              "<g id=\"edge3\" class=\"edge\">\n",
              "<title>1&#45;&gt;3</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M207.6065,-103.9815C208.1316,-95.618 208.6854,-86.7965 209.2163,-78.3409\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"212.7154,-78.4634 209.8489,-68.2637 205.7291,-78.0247 212.7154,-78.4634\"/>\n",
              "</g>\n",
              "<!-- 5 -->\n",
              "<g id=\"node6\" class=\"node\">\n",
              "<title>5</title>\n",
              "<path fill=\"#e58139\" fill-opacity=\"0.129412\" stroke=\"#000000\" d=\"M449.5,-68C449.5,-68 324.5,-68 324.5,-68 318.5,-68 312.5,-62 312.5,-56 312.5,-56 312.5,-12 312.5,-12 312.5,-6 318.5,0 324.5,0 324.5,0 449.5,0 449.5,0 455.5,0 461.5,-6 461.5,-12 461.5,-12 461.5,-56 461.5,-56 461.5,-62 455.5,-68 449.5,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"387\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.498</text>\n",
              "<text text-anchor=\"middle\" x=\"387\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 3513</text>\n",
              "<text text-anchor=\"middle\" x=\"387\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [1879, 1634]</text>\n",
              "<text text-anchor=\"middle\" x=\"387\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = não</text>\n",
              "</g>\n",
              "<!-- 4&#45;&gt;5 -->\n",
              "<g id=\"edge5\" class=\"edge\">\n",
              "<title>4&#45;&gt;5</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M390.7658,-103.9815C390.3158,-95.618 389.8411,-86.7965 389.3861,-78.3409\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"392.8762,-78.0611 388.8438,-68.2637 385.8863,-78.4373 392.8762,-78.0611\"/>\n",
              "</g>\n",
              "<!-- 6 -->\n",
              "<g id=\"node7\" class=\"node\">\n",
              "<title>6</title>\n",
              "<path fill=\"#e58139\" fill-opacity=\"0.835294\" stroke=\"#000000\" d=\"M608,-68C608,-68 492,-68 492,-68 486,-68 480,-62 480,-56 480,-56 480,-12 480,-12 480,-6 486,0 492,0 492,0 608,0 608,0 614,0 620,-6 620,-12 620,-12 620,-56 620,-56 620,-62 614,-68 608,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"550\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.243</text>\n",
              "<text text-anchor=\"middle\" x=\"550\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 2140</text>\n",
              "<text text-anchor=\"middle\" x=\"550\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [1837, 303]</text>\n",
              "<text text-anchor=\"middle\" x=\"550\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = não</text>\n",
              "</g>\n",
              "<!-- 4&#45;&gt;6 -->\n",
              "<g id=\"edge6\" class=\"edge\">\n",
              "<title>4&#45;&gt;6</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M451.461,-103.9815C465.2508,-94.1881 479.9247,-83.7668 493.6205,-74.0402\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"495.8785,-76.7295 502.005,-68.0856 491.8253,-71.0223 495.8785,-76.7295\"/>\n",
              "</g>\n",
              "</g>\n",
              "</svg>\n"
            ],
            "text/plain": [
              "<graphviz.files.Source at 0x7f29012c59e8>"
            ]
          },
          "execution_count": 9,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "from sklearn.tree import export_graphviz\n",
        "import graphviz\n",
        "\n",
        "modelo.fit(x_azar, y_azar)\n",
        "features = x_azar.columns\n",
        "dot_data = export_graphviz(modelo, out_file=None, filled=True, rounded=True,\n",
        "                          class_names=[\"não\",\"sim\"],\n",
        "                          feature_names = features)\n",
        "graph = graphviz.Source(dot_data)\n",
        "graph"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 50
        },
        "colab_type": "code",
        "id": "zdfgLexRPQ54",
        "outputId": "c8c75555-27f2-4ceb-e31b-57d7938723f3"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Accuracy médio 78.67\n",
            "Intervalo [76.40, 80.94]\n"
          ]
        }
      ],
      "source": [
        "# GroupKFold para analisar como o modelo se comporta com novos grupos\n",
        "\n",
        "from sklearn.model_selection import GroupKFold\n",
        "\n",
        "SEED = 301\n",
        "np.random.seed(SEED)\n",
        "\n",
        "cv = GroupKFold(n_splits = 10)\n",
        "modelo = DecisionTreeClassifier(max_depth=3)\n",
        "results = cross_validate(modelo, x_azar, y_azar, cv = cv, groups = dados.modelo, return_train_score=False)\n",
        "imprime_resultados(results)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 618
        },
        "colab_type": "code",
        "id": "kSe6rPpdQ4Wy",
        "outputId": "e93cc869-3790-4c65-d712-fb1dc82111dc"
      },
      "outputs": [
        {
          "data": {
            "image/svg+xml": [
              "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
              "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
              " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
              "<!-- Generated by graphviz version 2.40.1 (20161225.0304)\n",
              " -->\n",
              "<!-- Title: Tree Pages: 1 -->\n",
              "<svg width=\"912pt\" height=\"433pt\"\n",
              " viewBox=\"0.00 0.00 912.00 433.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
              "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 429)\">\n",
              "<title>Tree</title>\n",
              "<polygon fill=\"#ffffff\" stroke=\"transparent\" points=\"-4,4 -4,-429 908,-429 908,4 -4,4\"/>\n",
              "<!-- 0 -->\n",
              "<g id=\"node1\" class=\"node\">\n",
              "<title>0</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.274510\" stroke=\"#000000\" d=\"M372.5,-425C372.5,-425 247.5,-425 247.5,-425 241.5,-425 235.5,-419 235.5,-413 235.5,-413 235.5,-354 235.5,-354 235.5,-348 241.5,-342 247.5,-342 247.5,-342 372.5,-342 372.5,-342 378.5,-342 384.5,-348 384.5,-354 384.5,-354 384.5,-413 384.5,-413 384.5,-419 378.5,-425 372.5,-425\"/>\n",
              "<text text-anchor=\"middle\" x=\"310\" y=\"-409.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">preco &lt;= 59982.246</text>\n",
              "<text text-anchor=\"middle\" x=\"310\" y=\"-394.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.487</text>\n",
              "<text text-anchor=\"middle\" x=\"310\" y=\"-379.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 10000</text>\n",
              "<text text-anchor=\"middle\" x=\"310\" y=\"-364.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [4200, 5800]</text>\n",
              "<text text-anchor=\"middle\" x=\"310\" y=\"-349.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 1 -->\n",
              "<g id=\"node2\" class=\"node\">\n",
              "<title>1</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.874510\" stroke=\"#000000\" d=\"M277.5,-306C277.5,-306 154.5,-306 154.5,-306 148.5,-306 142.5,-300 142.5,-294 142.5,-294 142.5,-235 142.5,-235 142.5,-229 148.5,-223 154.5,-223 154.5,-223 277.5,-223 277.5,-223 283.5,-223 289.5,-229 289.5,-235 289.5,-235 289.5,-294 289.5,-294 289.5,-300 283.5,-306 277.5,-306\"/>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-290.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">preco &lt;= 40083.383</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-275.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.198</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-260.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 4347</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [484, 3863]</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 0&#45;&gt;1 -->\n",
              "<g id=\"edge1\" class=\"edge\">\n",
              "<title>0&#45;&gt;1</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M277.1234,-341.8796C270.0803,-332.9633 262.5707,-323.4565 255.3126,-314.268\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"257.9621,-311.9757 249.017,-306.2981 252.4691,-316.3147 257.9621,-311.9757\"/>\n",
              "<text text-anchor=\"middle\" x=\"246.1027\" y=\"-327.4276\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">True</text>\n",
              "</g>\n",
              "<!-- 6 -->\n",
              "<g id=\"node7\" class=\"node\">\n",
              "<title>6</title>\n",
              "<path fill=\"#e58139\" fill-opacity=\"0.478431\" stroke=\"#000000\" d=\"M545,-306C545,-306 377,-306 377,-306 371,-306 365,-300 365,-294 365,-294 365,-235 365,-235 365,-229 371,-223 377,-223 377,-223 545,-223 545,-223 551,-223 557,-229 557,-235 557,-235 557,-294 557,-294 557,-300 551,-306 545,-306\"/>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-290.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">km_por_ano &lt;= 24124.006</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-275.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.45</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-260.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 5653</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [3716, 1937]</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = não</text>\n",
              "</g>\n",
              "<!-- 0&#45;&gt;6 -->\n",
              "<g id=\"edge6\" class=\"edge\">\n",
              "<title>0&#45;&gt;6</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M362.8125,-341.8796C374.8818,-332.368 387.8039,-322.1843 400.1788,-312.432\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"402.5296,-315.0356 408.2173,-306.0969 398.1968,-309.5377 402.5296,-315.0356\"/>\n",
              "<text text-anchor=\"middle\" x=\"405.275\" y=\"-327.2232\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">False</text>\n",
              "</g>\n",
              "<!-- 2 -->\n",
              "<g id=\"node3\" class=\"node\">\n",
              "<title>2</title>\n",
              "<path fill=\"#399de5\" stroke=\"#000000\" d=\"M112,-179.5C112,-179.5 12,-179.5 12,-179.5 6,-179.5 0,-173.5 0,-167.5 0,-167.5 0,-123.5 0,-123.5 0,-117.5 6,-111.5 12,-111.5 12,-111.5 112,-111.5 112,-111.5 118,-111.5 124,-117.5 124,-123.5 124,-123.5 124,-167.5 124,-167.5 124,-173.5 118,-179.5 112,-179.5\"/>\n",
              "<text text-anchor=\"middle\" x=\"62\" y=\"-164.3\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.0</text>\n",
              "<text text-anchor=\"middle\" x=\"62\" y=\"-149.3\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 1935</text>\n",
              "<text text-anchor=\"middle\" x=\"62\" y=\"-134.3\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [0, 1935]</text>\n",
              "<text text-anchor=\"middle\" x=\"62\" y=\"-119.3\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 1&#45;&gt;2 -->\n",
              "<g id=\"edge2\" class=\"edge\">\n",
              "<title>1&#45;&gt;2</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M162.1383,-222.8796C146.7696,-211.0038 130.0453,-198.0804 114.7441,-186.2568\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"116.4605,-183.1599 106.4076,-179.8149 112.1804,-188.699 116.4605,-183.1599\"/>\n",
              "</g>\n",
              "<!-- 3 -->\n",
              "<g id=\"node4\" class=\"node\">\n",
              "<title>3</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.749020\" stroke=\"#000000\" d=\"M277.5,-187C277.5,-187 154.5,-187 154.5,-187 148.5,-187 142.5,-181 142.5,-175 142.5,-175 142.5,-116 142.5,-116 142.5,-110 148.5,-104 154.5,-104 154.5,-104 277.5,-104 277.5,-104 283.5,-104 289.5,-110 289.5,-116 289.5,-116 289.5,-175 289.5,-175 289.5,-181 283.5,-187 277.5,-187\"/>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">preco &lt;= 41083.031</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.321</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 2412</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [484, 1928]</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 1&#45;&gt;3 -->\n",
              "<g id=\"edge3\" class=\"edge\">\n",
              "<title>1&#45;&gt;3</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M216,-222.8796C216,-214.6838 216,-205.9891 216,-197.5013\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"219.5001,-197.298 216,-187.2981 212.5001,-197.2981 219.5001,-197.298\"/>\n",
              "</g>\n",
              "<!-- 4 -->\n",
              "<g id=\"node5\" class=\"node\">\n",
              "<title>4</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.533333\" stroke=\"#000000\" d=\"M116,-68C116,-68 24,-68 24,-68 18,-68 12,-62 12,-56 12,-56 12,-12 12,-12 12,-6 18,0 24,0 24,0 116,0 116,0 122,0 128,-6 128,-12 128,-12 128,-56 128,-56 128,-62 122,-68 116,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"70\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.434</text>\n",
              "<text text-anchor=\"middle\" x=\"70\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 110</text>\n",
              "<text text-anchor=\"middle\" x=\"70\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [35, 75]</text>\n",
              "<text text-anchor=\"middle\" x=\"70\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 3&#45;&gt;4 -->\n",
              "<g id=\"edge4\" class=\"edge\">\n",
              "<title>3&#45;&gt;4</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M161.635,-103.9815C148.9989,-94.3313 135.5644,-84.0714 122.9885,-74.4673\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"124.9372,-71.5515 114.8654,-68.2637 120.6886,-77.1148 124.9372,-71.5515\"/>\n",
              "</g>\n",
              "<!-- 5 -->\n",
              "<g id=\"node6\" class=\"node\">\n",
              "<title>5</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.756863\" stroke=\"#000000\" d=\"M274,-68C274,-68 158,-68 158,-68 152,-68 146,-62 146,-56 146,-56 146,-12 146,-12 146,-6 152,0 158,0 158,0 274,0 274,0 280,0 286,-6 286,-12 286,-12 286,-56 286,-56 286,-62 280,-68 274,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.314</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 2302</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [449, 1853]</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 3&#45;&gt;5 -->\n",
              "<g id=\"edge5\" class=\"edge\">\n",
              "<title>3&#45;&gt;5</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M216,-103.9815C216,-95.618 216,-86.7965 216,-78.3409\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"219.5001,-78.2636 216,-68.2637 212.5001,-78.2637 219.5001,-78.2636\"/>\n",
              "</g>\n",
              "<!-- 7 -->\n",
              "<g id=\"node8\" class=\"node\">\n",
              "<title>7</title>\n",
              "<path fill=\"#e58139\" fill-opacity=\"0.129412\" stroke=\"#000000\" d=\"M523.5,-187C523.5,-187 398.5,-187 398.5,-187 392.5,-187 386.5,-181 386.5,-175 386.5,-175 386.5,-116 386.5,-116 386.5,-110 392.5,-104 398.5,-104 398.5,-104 523.5,-104 523.5,-104 529.5,-104 535.5,-110 535.5,-116 535.5,-116 535.5,-175 535.5,-175 535.5,-181 529.5,-187 523.5,-187\"/>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">preco &lt;= 99974.406</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.498</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 3513</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [1879, 1634]</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = não</text>\n",
              "</g>\n",
              "<!-- 6&#45;&gt;7 -->\n",
              "<g id=\"edge7\" class=\"edge\">\n",
              "<title>6&#45;&gt;7</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M461,-222.8796C461,-214.6838 461,-205.9891 461,-197.5013\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"464.5001,-197.298 461,-187.2981 457.5001,-197.2981 464.5001,-197.298\"/>\n",
              "</g>\n",
              "<!-- 10 -->\n",
              "<g id=\"node11\" class=\"node\">\n",
              "<title>10</title>\n",
              "<path fill=\"#e58139\" fill-opacity=\"0.835294\" stroke=\"#000000\" d=\"M753.5,-187C753.5,-187 622.5,-187 622.5,-187 616.5,-187 610.5,-181 610.5,-175 610.5,-175 610.5,-116 610.5,-116 610.5,-110 616.5,-104 622.5,-104 622.5,-104 753.5,-104 753.5,-104 759.5,-104 765.5,-110 765.5,-116 765.5,-116 765.5,-175 765.5,-175 765.5,-181 759.5,-187 753.5,-187\"/>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">preco &lt;= 100076.953</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.243</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 2140</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [1837, 303]</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = não</text>\n",
              "</g>\n",
              "<!-- 6&#45;&gt;10 -->\n",
              "<g id=\"edge10\" class=\"edge\">\n",
              "<title>6&#45;&gt;10</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M540.3936,-222.8796C560.0179,-212.592 581.1418,-201.5182 601.1025,-191.0542\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"602.9504,-194.0373 610.1822,-186.2944 599.7003,-187.8375 602.9504,-194.0373\"/>\n",
              "</g>\n",
              "<!-- 8 -->\n",
              "<g id=\"node9\" class=\"node\">\n",
              "<title>8</title>\n",
              "<path fill=\"#e58139\" fill-opacity=\"0.250980\" stroke=\"#000000\" d=\"M441.5,-68C441.5,-68 316.5,-68 316.5,-68 310.5,-68 304.5,-62 304.5,-56 304.5,-56 304.5,-12 304.5,-12 304.5,-6 310.5,0 316.5,0 316.5,0 441.5,0 441.5,0 447.5,0 453.5,-6 453.5,-12 453.5,-12 453.5,-56 453.5,-56 453.5,-62 447.5,-68 441.5,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"379\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.49</text>\n",
              "<text text-anchor=\"middle\" x=\"379\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 2944</text>\n",
              "<text text-anchor=\"middle\" x=\"379\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [1682, 1262]</text>\n",
              "<text text-anchor=\"middle\" x=\"379\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = não</text>\n",
              "</g>\n",
              "<!-- 7&#45;&gt;8 -->\n",
              "<g id=\"edge8\" class=\"edge\">\n",
              "<title>7&#45;&gt;8</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M430.4662,-103.9815C423.8424,-94.9747 416.8279,-85.4367 410.1795,-76.3965\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"412.9426,-74.2461 404.1984,-68.2637 407.3034,-78.3933 412.9426,-74.2461\"/>\n",
              "</g>\n",
              "<!-- 9 -->\n",
              "<g id=\"node10\" class=\"node\">\n",
              "<title>9</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.470588\" stroke=\"#000000\" d=\"M592,-68C592,-68 484,-68 484,-68 478,-68 472,-62 472,-56 472,-56 472,-12 472,-12 472,-6 478,0 484,0 484,0 592,0 592,0 598,0 604,-6 604,-12 604,-12 604,-56 604,-56 604,-62 598,-68 592,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"538\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.453</text>\n",
              "<text text-anchor=\"middle\" x=\"538\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 569</text>\n",
              "<text text-anchor=\"middle\" x=\"538\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [197, 372]</text>\n",
              "<text text-anchor=\"middle\" x=\"538\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 7&#45;&gt;9 -->\n",
              "<g id=\"edge9\" class=\"edge\">\n",
              "<title>7&#45;&gt;9</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M489.672,-103.9815C495.8285,-95.0666 502.3443,-85.6313 508.5305,-76.6734\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"511.5355,-78.4811 514.3381,-68.2637 505.7755,-74.5034 511.5355,-78.4811\"/>\n",
              "</g>\n",
              "<!-- 11 -->\n",
              "<g id=\"node12\" class=\"node\">\n",
              "<title>11</title>\n",
              "<path fill=\"#e58139\" fill-opacity=\"0.956863\" stroke=\"#000000\" d=\"M742,-68C742,-68 634,-68 634,-68 628,-68 622,-62 622,-56 622,-56 622,-12 622,-12 622,-6 628,0 634,0 634,0 742,0 742,0 748,0 754,-6 754,-12 754,-12 754,-56 754,-56 754,-62 748,-68 742,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.082</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 1809</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [1732, 77]</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = não</text>\n",
              "</g>\n",
              "<!-- 10&#45;&gt;11 -->\n",
              "<g id=\"edge11\" class=\"edge\">\n",
              "<title>10&#45;&gt;11</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M688,-103.9815C688,-95.618 688,-86.7965 688,-78.3409\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"691.5001,-78.2636 688,-68.2637 684.5001,-78.2637 691.5001,-78.2636\"/>\n",
              "</g>\n",
              "<!-- 12 -->\n",
              "<g id=\"node13\" class=\"node\">\n",
              "<title>12</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.537255\" stroke=\"#000000\" d=\"M892,-68C892,-68 784,-68 784,-68 778,-68 772,-62 772,-56 772,-56 772,-12 772,-12 772,-6 778,0 784,0 784,0 892,0 892,0 898,0 904,-6 904,-12 904,-12 904,-56 904,-56 904,-62 898,-68 892,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"838\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.433</text>\n",
              "<text text-anchor=\"middle\" x=\"838\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 331</text>\n",
              "<text text-anchor=\"middle\" x=\"838\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [105, 226]</text>\n",
              "<text text-anchor=\"middle\" x=\"838\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 10&#45;&gt;12 -->\n",
              "<g id=\"edge12\" class=\"edge\">\n",
              "<title>10&#45;&gt;12</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M743.8545,-103.9815C756.8368,-94.3313 770.6394,-84.0714 783.5597,-74.4673\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"785.9678,-77.0384 791.9054,-68.2637 781.7917,-71.4204 785.9678,-77.0384\"/>\n",
              "</g>\n",
              "</g>\n",
              "</svg>\n"
            ],
            "text/plain": [
              "<graphviz.files.Source at 0x7f291bbe0470>"
            ]
          },
          "execution_count": 11,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "from sklearn.tree import export_graphviz\n",
        "import graphviz\n",
        "\n",
        "modelo.fit(x_azar, y_azar)\n",
        "features = x_azar.columns\n",
        "dot_data = export_graphviz(modelo, out_file=None, filled=True, rounded=True,\n",
        "                          class_names=[\"não\",\"sim\"],\n",
        "                          feature_names = features)\n",
        "graph = graphviz.Source(dot_data)\n",
        "graph"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 50
        },
        "colab_type": "code",
        "id": "2mZNqHhCQ7DO",
        "outputId": "e74e04e4-ddf8-442d-f530-ab40b3db04dc"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Accuracy médio 77.19\n",
            "Intervalo [75.26, 79.13]\n"
          ]
        }
      ],
      "source": [
        "# GroupKFold para analisar como o modelo se comporta com novos grupos\n",
        "\n",
        "from sklearn.model_selection import GroupKFold\n",
        "\n",
        "SEED = 301\n",
        "np.random.seed(SEED)\n",
        "\n",
        "cv = GroupKFold(n_splits = 10)\n",
        "modelo = DecisionTreeClassifier(max_depth=10)\n",
        "results = cross_validate(modelo, x_azar, y_azar, cv = cv, groups = dados.modelo, return_train_score=False)\n",
        "imprime_resultados(results)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "GMXJcGqWTiIp"
      },
      "source": [
        "# Explorando hiper parâmetros em uma dimensão"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 733
        },
        "colab_type": "code",
        "id": "xTV7Tc4aRFTF",
        "outputId": "a5d355cc-fb81-4c99-d67c-bf445de7d9fc"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Arvore max_depth = 1, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 2, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 3, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 4, treino = 78.79, teste = 78.63\n",
            "Arvore max_depth = 5, treino = 78.94, teste = 78.56\n",
            "Arvore max_depth = 6, treino = 79.17, teste = 78.12\n",
            "Arvore max_depth = 7, treino = 79.50, teste = 77.96\n",
            "Arvore max_depth = 8, treino = 79.89, teste = 77.86\n",
            "Arvore max_depth = 9, treino = 80.46, teste = 77.38\n",
            "Arvore max_depth = 10, treino = 81.16, teste = 77.19\n",
            "Arvore max_depth = 11, treino = 82.03, teste = 76.97\n",
            "Arvore max_depth = 12, treino = 82.89, teste = 76.49\n",
            "Arvore max_depth = 13, treino = 83.95, teste = 75.81\n",
            "Arvore max_depth = 14, treino = 85.03, teste = 75.66\n",
            "Arvore max_depth = 15, treino = 86.14, teste = 75.16\n",
            "Arvore max_depth = 16, treino = 87.19, teste = 75.11\n",
            "Arvore max_depth = 17, treino = 88.31, teste = 74.74\n",
            "Arvore max_depth = 18, treino = 89.26, teste = 74.33\n",
            "Arvore max_depth = 19, treino = 90.32, teste = 74.34\n",
            "Arvore max_depth = 20, treino = 91.32, teste = 74.22\n",
            "Arvore max_depth = 21, treino = 92.14, teste = 73.80\n",
            "Arvore max_depth = 22, treino = 93.02, teste = 73.81\n",
            "Arvore max_depth = 23, treino = 93.80, teste = 73.38\n",
            "Arvore max_depth = 24, treino = 94.47, teste = 73.43\n",
            "Arvore max_depth = 25, treino = 95.16, teste = 73.14\n",
            "Arvore max_depth = 26, treino = 95.74, teste = 73.04\n",
            "Arvore max_depth = 27, treino = 96.33, teste = 72.91\n",
            "Arvore max_depth = 28, treino = 96.75, teste = 72.66\n",
            "Arvore max_depth = 29, treino = 97.10, teste = 72.73\n",
            "Arvore max_depth = 30, treino = 97.43, teste = 72.81\n",
            "Arvore max_depth = 31, treino = 97.80, teste = 72.86\n",
            "Arvore max_depth = 32, treino = 98.10, teste = 72.52\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>max_depth</th>\n",
              "      <th>train</th>\n",
              "      <th>test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>78.787628</td>\n",
              "      <td>78.632803</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>78.941007</td>\n",
              "      <td>78.555912</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   max_depth      train       test\n",
              "0          1  75.791169  75.784219\n",
              "1          2  75.791169  75.784219\n",
              "2          3  78.750993  78.672633\n",
              "3          4  78.787628  78.632803\n",
              "4          5  78.941007  78.555912"
            ]
          },
          "execution_count": 13,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "def roda_arvore_de_decisao(max_depth):\n",
        "  SEED = 301\n",
        "  np.random.seed(SEED)\n",
        "\n",
        "  cv = GroupKFold(n_splits = 10)\n",
        "  modelo = DecisionTreeClassifier(max_depth=max_depth)\n",
        "  results = cross_validate(modelo, x_azar, y_azar, cv = cv, groups = dados.modelo, return_train_score=True)\n",
        "  train_score = results['train_score'].mean() * 100\n",
        "  test_score = results['test_score'].mean() * 100\n",
        "  print(\"Arvore max_depth = %d, treino = %.2f, teste = %.2f\" % (max_depth, train_score, test_score))\n",
        "  tabela = [max_depth, train_score, test_score]\n",
        "  return tabela\n",
        "  \n",
        "resultados = [roda_arvore_de_decisao(i) for i in range(1,33)]\n",
        "resultados = pd.DataFrame(resultados, columns= [\"max_depth\",\"train\",\"test\"])\n",
        "resultados.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 301
        },
        "colab_type": "code",
        "id": "6HBLCc7QTsLt",
        "outputId": "5c14e473-e4ac-4913-8e17-89a417896d2a"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f29012c5550>"
            ]
          },
          "execution_count": 14,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f28fe797cf8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "import seaborn as sns\n",
        "\n",
        "sns.lineplot(x = \"max_depth\", y = \"train\", data = resultados)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "j-o6VtTrXbhu"
      },
      "source": [
        "# OVERFIT: ficou \"perfeito\" para o treino mas ruim para o teste"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 301
        },
        "colab_type": "code",
        "id": "OhwJ_FAnWOlq",
        "outputId": "b136f667-4257-48eb-9bc5-b7f0fe53f4fb"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f28fbda5cc0>"
            ]
          },
          "execution_count": 16,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f28fbe3f3c8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "sns.lineplot(x = \"max_depth\", y = \"train\", data = resultados)\n",
        "sns.lineplot(x = \"max_depth\", y = \"test\", data = resultados)\n",
        "plt.legend([\"Treino\", \"Teste\"])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "colab_type": "code",
        "id": "wXmQpoGaXXqP",
        "outputId": "40f9ea94-6de4-4dff-fff1-d8ebc7739f63"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>max_depth</th>\n",
              "      <th>train</th>\n",
              "      <th>test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>78.787628</td>\n",
              "      <td>78.632803</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>78.941007</td>\n",
              "      <td>78.555912</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>6</td>\n",
              "      <td>79.170115</td>\n",
              "      <td>78.123266</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>7</td>\n",
              "      <td>79.496806</td>\n",
              "      <td>77.963185</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   max_depth      train       test\n",
              "2          3  78.750993  78.672633\n",
              "3          4  78.787628  78.632803\n",
              "4          5  78.941007  78.555912\n",
              "5          6  79.170115  78.123266\n",
              "6          7  79.496806  77.963185"
            ]
          },
          "execution_count": 18,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "resultados.sort_values(\"test\", ascending=False).head()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "SnW9-CTjcpWp"
      },
      "source": [
        "# Explorando hiper parâmetros em 2 dimensões"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 2346
        },
        "colab_type": "code",
        "id": "Odrf-gsWXl9u",
        "outputId": "5cd04740-d6c5-40f9-9032-21a12a3f0207"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Arvore max_depth = 1, min_samples_leaf = 32, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 1, min_samples_leaf = 64, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 1, min_samples_leaf = 128, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 1, min_samples_leaf = 256, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 2, min_samples_leaf = 32, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 2, min_samples_leaf = 64, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 2, min_samples_leaf = 128, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 2, min_samples_leaf = 256, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 3, min_samples_leaf = 32, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 3, min_samples_leaf = 64, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 3, min_samples_leaf = 128, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 3, min_samples_leaf = 256, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 4, min_samples_leaf = 32, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 4, min_samples_leaf = 64, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 4, min_samples_leaf = 128, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 4, min_samples_leaf = 256, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 5, min_samples_leaf = 32, treino = 78.82, teste = 78.50\n",
            "Arvore max_depth = 5, min_samples_leaf = 64, treino = 78.79, teste = 78.45\n",
            "Arvore max_depth = 5, min_samples_leaf = 128, treino = 78.77, teste = 78.44\n",
            "Arvore max_depth = 5, min_samples_leaf = 256, treino = 78.77, teste = 78.44\n",
            "Arvore max_depth = 6, min_samples_leaf = 32, treino = 78.96, teste = 78.02\n",
            "Arvore max_depth = 6, min_samples_leaf = 64, treino = 78.92, teste = 78.09\n",
            "Arvore max_depth = 6, min_samples_leaf = 128, treino = 78.85, teste = 78.27\n",
            "Arvore max_depth = 6, min_samples_leaf = 256, treino = 78.79, teste = 78.27\n",
            "Arvore max_depth = 7, min_samples_leaf = 32, treino = 79.24, teste = 77.94\n",
            "Arvore max_depth = 7, min_samples_leaf = 64, treino = 79.18, teste = 78.05\n",
            "Arvore max_depth = 7, min_samples_leaf = 128, treino = 79.01, teste = 78.21\n",
            "Arvore max_depth = 7, min_samples_leaf = 256, treino = 78.85, teste = 78.08\n",
            "Arvore max_depth = 8, min_samples_leaf = 32, treino = 79.47, teste = 77.93\n",
            "Arvore max_depth = 8, min_samples_leaf = 64, treino = 79.30, teste = 77.99\n",
            "Arvore max_depth = 8, min_samples_leaf = 128, treino = 79.05, teste = 78.37\n",
            "Arvore max_depth = 8, min_samples_leaf = 256, treino = 78.89, teste = 77.94\n",
            "Arvore max_depth = 9, min_samples_leaf = 32, treino = 79.80, teste = 77.32\n",
            "Arvore max_depth = 9, min_samples_leaf = 64, treino = 79.46, teste = 77.55\n",
            "Arvore max_depth = 9, min_samples_leaf = 128, treino = 79.12, teste = 78.30\n",
            "Arvore max_depth = 9, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 10, min_samples_leaf = 32, treino = 80.05, teste = 77.45\n",
            "Arvore max_depth = 10, min_samples_leaf = 64, treino = 79.55, teste = 77.64\n",
            "Arvore max_depth = 10, min_samples_leaf = 128, treino = 79.16, teste = 78.29\n",
            "Arvore max_depth = 10, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 11, min_samples_leaf = 32, treino = 80.21, teste = 77.20\n",
            "Arvore max_depth = 11, min_samples_leaf = 64, treino = 79.60, teste = 77.53\n",
            "Arvore max_depth = 11, min_samples_leaf = 128, treino = 79.19, teste = 78.31\n",
            "Arvore max_depth = 11, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 12, min_samples_leaf = 32, treino = 80.34, teste = 77.00\n",
            "Arvore max_depth = 12, min_samples_leaf = 64, treino = 79.72, teste = 77.29\n",
            "Arvore max_depth = 12, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 12, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 13, min_samples_leaf = 32, treino = 80.51, teste = 76.85\n",
            "Arvore max_depth = 13, min_samples_leaf = 64, treino = 79.74, teste = 77.34\n",
            "Arvore max_depth = 13, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 13, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 14, min_samples_leaf = 32, treino = 80.60, teste = 76.69\n",
            "Arvore max_depth = 14, min_samples_leaf = 64, treino = 79.77, teste = 77.26\n",
            "Arvore max_depth = 14, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 14, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 15, min_samples_leaf = 32, treino = 80.70, teste = 76.88\n",
            "Arvore max_depth = 15, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 15, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 15, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 16, min_samples_leaf = 32, treino = 80.76, teste = 76.87\n",
            "Arvore max_depth = 16, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 16, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 16, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 17, min_samples_leaf = 32, treino = 80.82, teste = 76.98\n",
            "Arvore max_depth = 17, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 17, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 17, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 18, min_samples_leaf = 32, treino = 80.85, teste = 76.86\n",
            "Arvore max_depth = 18, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 18, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 18, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 19, min_samples_leaf = 32, treino = 80.86, teste = 76.89\n",
            "Arvore max_depth = 19, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 19, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 19, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 20, min_samples_leaf = 32, treino = 80.88, teste = 76.87\n",
            "Arvore max_depth = 20, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 20, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 20, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 21, min_samples_leaf = 32, treino = 80.88, teste = 76.92\n",
            "Arvore max_depth = 21, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 21, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 21, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 22, min_samples_leaf = 32, treino = 80.88, teste = 76.87\n",
            "Arvore max_depth = 22, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 22, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 22, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 23, min_samples_leaf = 32, treino = 80.88, teste = 76.87\n",
            "Arvore max_depth = 23, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 23, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 23, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 24, min_samples_leaf = 32, treino = 80.89, teste = 76.84\n",
            "Arvore max_depth = 24, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 24, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 24, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 25, min_samples_leaf = 32, treino = 80.89, teste = 76.84\n",
            "Arvore max_depth = 25, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 25, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 25, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 26, min_samples_leaf = 32, treino = 80.89, teste = 76.84\n",
            "Arvore max_depth = 26, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 26, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 26, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 27, min_samples_leaf = 32, treino = 80.89, teste = 76.84\n",
            "Arvore max_depth = 27, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 27, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 27, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 28, min_samples_leaf = 32, treino = 80.89, teste = 76.84\n",
            "Arvore max_depth = 28, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 28, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 28, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 29, min_samples_leaf = 32, treino = 80.89, teste = 76.84\n",
            "Arvore max_depth = 29, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 29, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 29, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 30, min_samples_leaf = 32, treino = 80.89, teste = 76.84\n",
            "Arvore max_depth = 30, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 30, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 30, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 31, min_samples_leaf = 32, treino = 80.89, teste = 76.84\n",
            "Arvore max_depth = 31, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 31, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 31, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 32, min_samples_leaf = 32, treino = 80.89, teste = 76.84\n",
            "Arvore max_depth = 32, min_samples_leaf = 64, treino = 79.79, teste = 77.30\n",
            "Arvore max_depth = 32, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 32, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>max_depth</th>\n",
              "      <th>min_samples_leaf</th>\n",
              "      <th>train</th>\n",
              "      <th>test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>32</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>64</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>128</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>256</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2</td>\n",
              "      <td>32</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   max_depth  min_samples_leaf      train       test\n",
              "0          1                32  75.791169  75.784219\n",
              "1          1                64  75.791169  75.784219\n",
              "2          1               128  75.791169  75.784219\n",
              "3          1               256  75.791169  75.784219\n",
              "4          2                32  75.791169  75.784219"
            ]
          },
          "execution_count": 21,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "def roda_arvore_de_decisao(max_depth, min_samples_leaf):\n",
        "  SEED = 301\n",
        "  np.random.seed(SEED)\n",
        "\n",
        "  cv = GroupKFold(n_splits = 10)\n",
        "  modelo = DecisionTreeClassifier(max_depth=max_depth, min_samples_leaf = min_samples_leaf)\n",
        "  results = cross_validate(modelo, x_azar, y_azar, cv = cv, groups = dados.modelo, return_train_score=True)\n",
        "  train_score = results['train_score'].mean() * 100\n",
        "  test_score = results['test_score'].mean() * 100\n",
        "  print(\"Arvore max_depth = %d, min_samples_leaf = %d, treino = %.2f, teste = %.2f\" % (max_depth, min_samples_leaf, train_score, test_score))\n",
        "  tabela = [max_depth, min_samples_leaf, train_score, test_score]\n",
        "  return tabela\n",
        "\n",
        "def busca():\n",
        "  resultados = []\n",
        "  for max_depth in range(1,33):\n",
        "    for min_samples_leaf in [32, 64, 128, 256]:\n",
        "      tabela = roda_arvore_de_decisao(max_depth, min_samples_leaf)\n",
        "      resultados.append(tabela)\n",
        "  resultados = pd.DataFrame(resultados, columns= [\"max_depth\",\"min_samples_leaf\",\"train\",\"test\"])\n",
        "  return resultados\n",
        "\n",
        "resultados = busca()\n",
        "resultados.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "colab_type": "code",
        "id": "FEl16TSwdQgW",
        "outputId": "fc9d64cb-cdce-4f25-ffa2-682fd581513a"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>max_depth</th>\n",
              "      <th>min_samples_leaf</th>\n",
              "      <th>train</th>\n",
              "      <th>test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>4</td>\n",
              "      <td>256</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>4</td>\n",
              "      <td>32</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>4</td>\n",
              "      <td>128</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>4</td>\n",
              "      <td>64</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>3</td>\n",
              "      <td>32</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    max_depth  min_samples_leaf      train       test\n",
              "15          4               256  78.750993  78.672633\n",
              "12          4                32  78.750993  78.672633\n",
              "14          4               128  78.750993  78.672633\n",
              "13          4                64  78.750993  78.672633\n",
              "8           3                32  78.750993  78.672633"
            ]
          },
          "execution_count": 22,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "resultados.sort_values(\"test\", ascending=False).head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 166
        },
        "colab_type": "code",
        "id": "lZsBCpgRd6Yh",
        "outputId": "31685b91-cbd8-4769-c9ce-aca5ec2cde09"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>max_depth</th>\n",
              "      <th>min_samples_leaf</th>\n",
              "      <th>train</th>\n",
              "      <th>test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>max_depth</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.536705</td>\n",
              "      <td>-0.027675</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min_samples_leaf</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.409676</td>\n",
              "      <td>0.349011</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>train</th>\n",
              "      <td>0.536705</td>\n",
              "      <td>-0.409676</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.116466</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>test</th>\n",
              "      <td>-0.027675</td>\n",
              "      <td>0.349011</td>\n",
              "      <td>0.116466</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                  max_depth  min_samples_leaf     train      test\n",
              "max_depth          1.000000          0.000000  0.536705 -0.027675\n",
              "min_samples_leaf   0.000000          1.000000 -0.409676  0.349011\n",
              "train              0.536705         -0.409676  1.000000  0.116466\n",
              "test              -0.027675          0.349011  0.116466  1.000000"
            ]
          },
          "execution_count": 24,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "corr = resultados.corr()\n",
        "corr"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 365
        },
        "colab_type": "code",
        "id": "AdxtFwmZfmck",
        "outputId": "ede38ee9-12de-45f4-b2a1-c1986982a8ce"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f28fbabdef0>"
            ]
          },
          "execution_count": 25,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f28fbb1ee48>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "sns.heatmap(corr)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 845
        },
        "colab_type": "code",
        "id": "UWQaVyE-gKHI",
        "outputId": "5d2fefbe-e789-478f-c464-1f564a09ebc6"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: FutureWarning: pandas.scatter_matrix is deprecated. Use pandas.plotting.scatter_matrix instead\n",
            "  \"\"\"Entry point for launching an IPython kernel.\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x7f28fb1511d0>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f28fb0c4cf8>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f28fb076c50>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f28fb032cc0>],\n",
              "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7f28fafedcc0>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f28fafedcf8>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f28fafd87f0>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f28faf92d68>],\n",
              "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7f28faf4cc50>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f28faf714a8>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f28faebb2b0>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f28fae72780>],\n",
              "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7f28fae2dcc0>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f28fae4d080>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f28fae12d30>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7f28fadd50f0>]],\n",
              "      dtype=object)"
            ]
          },
          "execution_count": 28,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f28fb12b470>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "pd.scatter_matrix(resultados, figsize = (14, 8), alpha = 0.3)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 746
        },
        "colab_type": "code",
        "id": "AcQnhOM3gnJ-",
        "outputId": "87f86d66-792a-4b17-fc84-993c75582a8d"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<seaborn.axisgrid.PairGrid at 0x7f28fb2e56a0>"
            ]
          },
          "execution_count": 29,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f28fb0f2e48>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "sns.pairplot(resultados)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 564
        },
        "colab_type": "code",
        "id": "sH5F3Gd8hAFG",
        "outputId": "bc6efbdc-f2f3-4ee5-9b96-6479fd8ad923"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f28fa639d30>"
            ]
          },
          "execution_count": 30,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f28fa6dc320>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "sns.set(style=\"white\")\n",
        "\n",
        "# Generate a mask for the upper triangle\n",
        "mask = np.zeros_like(corr, dtype=np.bool)\n",
        "mask[np.triu_indices_from(mask)] = True\n",
        "\n",
        "# Set up the matplotlib figure\n",
        "f, ax = plt.subplots(figsize=(11, 9))\n",
        "\n",
        "# Generate a custom diverging colormap\n",
        "cmap = sns.diverging_palette(220, 10, as_cmap=True)\n",
        "\n",
        "# Draw the heatmap with the mask and correct aspect ratio\n",
        "sns.heatmap(corr, mask=mask, cmap=cmap, vmax=.3, center=0,\n",
        "            square=True, linewidths=.5, cbar_kws={\"shrink\": .5})"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 2346
        },
        "colab_type": "code",
        "id": "fa3AUdNZiBEs",
        "outputId": "ca88efaa-d4e9-4494-85c0-46d404436308"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Arvore max_depth = 1, min_samples_leaf = 128, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 1, min_samples_leaf = 192, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 1, min_samples_leaf = 256, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 1, min_samples_leaf = 512, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 2, min_samples_leaf = 128, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 2, min_samples_leaf = 192, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 2, min_samples_leaf = 256, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 2, min_samples_leaf = 512, treino = 75.79, teste = 75.78\n",
            "Arvore max_depth = 3, min_samples_leaf = 128, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 3, min_samples_leaf = 192, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 3, min_samples_leaf = 256, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 3, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 4, min_samples_leaf = 128, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 4, min_samples_leaf = 192, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 4, min_samples_leaf = 256, treino = 78.75, teste = 78.67\n",
            "Arvore max_depth = 4, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 5, min_samples_leaf = 128, treino = 78.77, teste = 78.44\n",
            "Arvore max_depth = 5, min_samples_leaf = 192, treino = 78.77, teste = 78.44\n",
            "Arvore max_depth = 5, min_samples_leaf = 256, treino = 78.77, teste = 78.44\n",
            "Arvore max_depth = 5, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 6, min_samples_leaf = 128, treino = 78.85, teste = 78.27\n",
            "Arvore max_depth = 6, min_samples_leaf = 192, treino = 78.79, teste = 78.31\n",
            "Arvore max_depth = 6, min_samples_leaf = 256, treino = 78.79, teste = 78.27\n",
            "Arvore max_depth = 6, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 7, min_samples_leaf = 128, treino = 79.01, teste = 78.21\n",
            "Arvore max_depth = 7, min_samples_leaf = 192, treino = 78.92, teste = 78.10\n",
            "Arvore max_depth = 7, min_samples_leaf = 256, treino = 78.85, teste = 78.08\n",
            "Arvore max_depth = 7, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 8, min_samples_leaf = 128, treino = 79.05, teste = 78.37\n",
            "Arvore max_depth = 8, min_samples_leaf = 192, treino = 78.98, teste = 78.20\n",
            "Arvore max_depth = 8, min_samples_leaf = 256, treino = 78.89, teste = 77.94\n",
            "Arvore max_depth = 8, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 9, min_samples_leaf = 128, treino = 79.12, teste = 78.30\n",
            "Arvore max_depth = 9, min_samples_leaf = 192, treino = 78.99, teste = 78.11\n",
            "Arvore max_depth = 9, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 9, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 10, min_samples_leaf = 128, treino = 79.16, teste = 78.29\n",
            "Arvore max_depth = 10, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 10, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 10, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 11, min_samples_leaf = 128, treino = 79.19, teste = 78.31\n",
            "Arvore max_depth = 11, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 11, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 11, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 12, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 12, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 12, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 12, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 13, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 13, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 13, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 13, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 14, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 14, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 14, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 14, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 15, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 15, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 15, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 15, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 16, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 16, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 16, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 16, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 17, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 17, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 17, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 17, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 18, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 18, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 18, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 18, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 19, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 19, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 19, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 19, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 20, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 20, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 20, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 20, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 21, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 21, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 21, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 21, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 22, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 22, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 22, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 22, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 23, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 23, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 23, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 23, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 24, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 24, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 24, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 24, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 25, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 25, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 25, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 25, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 26, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 26, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 26, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 26, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 27, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 27, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 27, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 27, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 28, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 28, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 28, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 28, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 29, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 29, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 29, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 29, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 30, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 30, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 30, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 30, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 31, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 31, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 31, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 31, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n",
            "Arvore max_depth = 32, min_samples_leaf = 128, treino = 79.21, teste = 78.22\n",
            "Arvore max_depth = 32, min_samples_leaf = 192, treino = 79.01, teste = 78.05\n",
            "Arvore max_depth = 32, min_samples_leaf = 256, treino = 78.90, teste = 77.85\n",
            "Arvore max_depth = 32, min_samples_leaf = 512, treino = 77.53, teste = 77.50\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>max_depth</th>\n",
              "      <th>min_samples_leaf</th>\n",
              "      <th>train</th>\n",
              "      <th>test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>128</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>192</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>256</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>512</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2</td>\n",
              "      <td>128</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   max_depth  min_samples_leaf      train       test\n",
              "0          1               128  75.791169  75.784219\n",
              "1          1               192  75.791169  75.784219\n",
              "2          1               256  75.791169  75.784219\n",
              "3          1               512  75.791169  75.784219\n",
              "4          2               128  75.791169  75.784219"
            ]
          },
          "execution_count": 34,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "def busca():\n",
        "  resultados = []\n",
        "  for max_depth in range(1,33):\n",
        "    for min_samples_leaf in [128, 192, 256, 512]:\n",
        "      tabela = roda_arvore_de_decisao(max_depth, min_samples_leaf)\n",
        "      resultados.append(tabela)\n",
        "  resultados = pd.DataFrame(resultados, columns= [\"max_depth\",\"min_samples_leaf\",\"train\",\"test\"])\n",
        "  return resultados\n",
        "\n",
        "resultados = busca()\n",
        "resultados.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {},
        "colab_type": "code",
        "id": "GLMt71GYii3A"
      },
      "outputs": [],
      "source": [
        "corr = resultados.corr()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 564
        },
        "colab_type": "code",
        "id": "3XjB5-HyinEM",
        "outputId": "9a52e183-ddad-426a-c25f-894f9fa421df"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f28fa3f4518>"
            ]
          },
          "execution_count": 36,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlsAAAISCAYAAAD7kgrVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzt3XtcVOXe///3AAIBYlBmqPmzw9ZUVEoTvbvNcmeRQYDHytS07GCYlpqUe6tZt2bqXSoeOqeZZXnIdJs7LQ+1Vdpu22mltTVRPB8RMASGWd8/+jV3iskanctpZl7Px2MeD2fNmrU+4yi+/VzXupbDsixLAAAAMCLE1wUAAAAEMsIWAACAQYQtAAAAgwhbAAAABhG2AAAADCJsAQAAGETYAgAAMIiwBQAAYBBhCwAAwCDCFgAAgEGELQAAAIMIWwAAAAYRtgAAAAwibAEAABhE2AIAADCIsAUAAGAQYQsAAMAgwhYAAIBBhC0AAACDCFsAAAAGEbYAAAAMImwBAAAYRNgCAAAwiLAFAABgEGELAADAIMIWAACAQYQtAAAAgwhbAAAABhG2AAAADCJsAQAAGETYAgAAMIiwBQAAYBBhCwAAwCDCFgAAgEGELQAAAIMIWwAAAAYRtgAAAAwibAEAABhE2AIAADCIsAUAAGAQYQsAAMAgwhYAAIBBhC0AAACDCFsAAAAGEbYAAAAMImwBAAAYRNgCAAAwiLAFAABgEGELAADAIMIWAACAQYQtAAAAgwhbAAAABhG2AAAADCJsAQAAGETYAgAAMIiwBQAAYBBhCwAAwCDCFgAAgEGELQAAAIMIWwAAAAYRtgAAAAwibAEAABhE2AIAADCIsAUAAGAQYQsAAMAgwhYAAIBBhC0AAACDCFsAAAAGEbYAAAAMImwBAAAYRNgCAAAwiLAFAABgUJjdHX/66SdNnz5d+fn5cjqd7u3z5s0zUhgAAEAgcFiWZdnZMSMjQykpKWrevLlCQ0Pd21u1amWsOAAAAH9nu7Plcrn0yCOPmKwFAAAg4Nies5WUlKStW7earAUAACDgVDmM2LlzZzkcDjmdTm3btk1XXnmlIiIi3K8zZwsAAOD3VRm2vvrqq7MegDlbAAAAv6/KOVu/hqlFixYpPT39lNcWLVpkpioAAIAAYXvO1ttvv21rGwAAAP5PlZ2tzZs3a9OmTTp27Jjeffdd9/bi4mKVl5cbLQ4AAMDfVRm2Dhw4oG+//VYlJSX69ttv3dujo6M1duxYo8UBAAD4O9uLmn755Zf67//+b9P1AAAABBTbYcvpdGru3LnKzc2VJLVu3VrdunVTWJjtdVEBAACCju2wNWLECO3Zs0cZGRmSfrkSsXbt2ho9erTRAgEAAPyZ7bbUV199paVLlyok5JcLGO+44w7deeedxgoDAAAIBLaXfrj44otVVlbmfu50OhUfH2+kKAAAgEBhexjxr3/9qzZt2qSOHTtKkpYtW6amTZuqYcOGkqQePXqYqxIAAMBP2R5GdDqdaty4sfLy8iRJ1157rcrLy09ZDgIAAACnst3ZAgAAgOdsz9kqKSnRyy+/rMGDB0uStm/frhUrVhgrDAAAIBDYDlujRo2S0+nU1q1bJUmXX365cnJyjBUGAAAQCGyHrR9++EFDhgxRtWrVJP1yux6Xy2WsMAAAgEBgO2yFh4ef8ry0tFRM9wIAADg721cjtmzZUjNmzFBZWZlyc3P11ltvqX379iZrAwAA8Hu2r0YsLy/X66+/rs8//1yWZal9+/Z66KGHuDciAADAWbD0AwAAgEFVtqVefPHFs77+1FNPea0YAACAQFPlBPmoqChFRUXp8OHD+uSTT+R0OuV0OrVs2TIdOXLkQtQIAADgt2wPI/bq1UuTJk1SXFycJOnYsWMaOHCgZs2aZbRAAAAAf2Z76YfDhw+7g5YkxcXF6fDhw0aKAgAACBS2LyW85pprNHz4cHXp0kWStGDBAl1zzTXGCgMAAAgEtocRi4uLNXXqVOXm5kqSkpOT9dhjjykmJsZogQAAAP7Ma0s/TJs2Tf379/fGoQAAAAKG7TlbVVm+fLm3DgUAABAwvBa2WBsVAACgMq+FLYfD4a1DAQAABAyvhS0AAABUxjAiAACAQbbD1tatWytt++KLL9y/fvPNN71TEQAAQACxHbYee+wxvf/++5Ikl8uliRMnaty4ce7X4+PjvV8dAACAn7O9ztbhw4c1dOhQxcbG6vDhw6pXr55GjBihiy66yHSNAAAAfst2Z+vSSy9Vjx49tGbNGuXn5+vRRx8laAEAAFTB9r0Rx44dq/Xr12v+/Pnatm2b7r//fg0ZMkQdO3Y0WR8AAIBfsx22Tpw4oQ8++EARERG66qqr1KhRIz355JOELQAAgLM4r3sjlpeXq1q1at6sBwAAIKDY7mxJ0pdffqktW7aotLTUvS0rK8vrRQEAAAQK22FrwoQJ2rx5s7Zt26Y///nP+uyzz9SmTRuTtQEAAPg921cjrl69Wm+88YYuueQSjR49WgsWLNDx48dN1gYAAOD3bIet8PBwhYWFyeFwqLy8XLVq1dL+/ftN1gYAAOD3bA8jRkdHq6SkRNddd52ys7NVs2ZNRUZGmqwNAADA73m0gnxsbKwqKir01ltvqaioSD179lTt2rVN1wgAAOC3zmvpBwAAAJyd7WHEn376STNmzNCuXbvkdDrd2+fNm2ekMAAAgEBgu7OVlpam9PR0NWnSRKGhoe7trVq1MlYcAACAv7Pd2QoLC9ODDz5oshYAAICAY3vph7Zt22r16tUmawEAAAg4tjtbbdq0Uf/+/RUSEqLw8HBZliWHw6F169aZrA8AAMCv2Z6z1aFDBw0ePFhNmjRRSMj/NcTq1KljrDgAAAB/Z7uzVaNGDaWkpJisBQAAIODYnrN166236r333lNBQYFKSkrcDwAAAPw+28OI11577f+9yeFwz9nasmWLseIAAAD8HSvIAwAAGGR7GBEAAACesz1BHgAAwJQjr8/yaP9LHuxlqBLvu+Bhq6io6EKfEhdY9erVfV0CAAB/GHS2AACA7zkcvq7AGMIWAADwOUcAhy0myAMAABhEZwsAAPheSOD2fwhbAADA9wJ4GJGwBQAAfC+AO1uB+8kAAAD+AOhsAQAAn3OEMIwIAABgDnO2AAAADHIE7swmwhYAAPA5hhEBAABMYhgRAADAoAAeRgzcTwYAAPAHQGcLAAD4HnO2AAAAzHEwZwsAAMCgAO5sMWcLAADAIDpbAADA9wL4RtSELQAA4HsBvPQDYQsAAPgcE+QBAAD81I4dO5Sdna2CggJdfPHFGjdunOrXr3/KPlOnTtXSpUsVEhKiatWq6YknnlDbtm29cn7CFgAA8D2DVyOOHDlS9957r9LT07Vo0SKNGDFCs2bNOmWfZs2aqW/fvrrooou0detW3Xffffryyy8VGRl53ucP3AFSAADgPxwOzx42HTlyRN9//71SU1MlSampqfr+++919OjRU/Zr27atLrroIklSw4YNZVmWCgoKvPLR6GwBAAC/U1hYqMLCwkrbY2NjFRsb636+b98+1apVS6GhoZKk0NBQXXbZZdq3b5/i4+PPeOyPPvpI9erV0+WXX+6VWglbAADA9zy8GnHmzJnKycmptD0rK0sDBgw45zK++uorTZo0SW+++eY5H+N0hC0AAOBzDg/nbPXu3VuZmZmVtv+2qyVJCQkJOnDggCoqKhQaGqqKigodPHhQCQkJld779ddfa+jQoZo2bZquuuoqzz7AWRC2AACA73m49MPpw4W/55JLLlGjRo20ZMkSpaena8mSJWrUqFGlIcRNmzbpiSee0OTJk9WkSROPaqmKw7Isy6tHrEJRUdGFPB18oHr16r4uAQDgZ4qWr/Ro/+odbrG97/bt25Wdna3CwkLFxsZq3Lhxuuqqq9SvXz89/vjjatq0qTp37qw9e/aoVq1a7ve9+OKLatiwoUd1nQlhC15H2AIAeKros9Ue7V/9z+0MVeJ9DCMCAACfc/z/VwsGItbZAgAAMIjOFgAA8D3ujQgAAGCQwdv1+BphCwAA+JzDw0VN/UngfjIAAIA/ADpbAADA95izBQAAYBBztgAAAAxizhYAAADOBZ0tAADgcw6GEQEAAAwK4AnyDCMCAAAYRGcLAAD4XgB3tghbAADA5xwhgTvYRtgCAAC+F8BhK3A/GQAAwB8AnS0AAOB7zNkCAAAwiHW2AAAAzHFwux4AAACcCzpbAADA95izBQAAYFBo4A62Be4nAwAA+AOgswUAAHwukCfIE7YAAIDvsfQDAACAQcF8u54nn3xSkjRz5kzjxQAAAASaKsPWf/7zH0nSRx99ZLwYAAAQnBwOh0cPf1LlMGJiYqJatGih0tJStWnTxr3dsiw5HA6tW7fOaIEAACAIBPAwosOyLKuqnQ4fPqzevXvr1VdfrfRanTp1PDphUVGRR/vD/1SvXt3XJQAA/EzZznyP9g///64wVIn32QpbknTixAlFR0ef9wkJW4GPsAUA8FQghy3bVyNGRERo7ty52rJli0pLS93bx44da6QwAAAQRPxsHpYnbA+QjhgxQhs3btSqVatUv359ffvtt4qMjDRZGwAACBKOEIdHD39iO2xt3rxZ48aNU/Xq1fXwww9rzpw52rZtm8naAAAA/J5Hw4iSFBoaqpKSElWvXl1HjhwxVhgAAAgi3K5HqlGjho4fP662bduqX79+iouLU61atUzWBgAAgkUAz9myfTViRUWFQkND5XK5tHjxYhUVFSkjI0MxMTEenZCrEQMfVyMCADxVvv+AR/tXu9x/Gj62O1uhoaEqLi7Wzp07lZ6ebrImAACAgGF7gHT16tW68847NWDAAEm/TJh/5JFHjBUGAACCh8MR4tHDn9iudvLkyZo3b55iY2MlSU2bNtWuXbuMFQYAAIJIiMOzhx+xPYwoSTVr1jzleXh4uFeLAQAAwakkMsKj/f1pdrDtzlZ0dLQOHz7svtN2bm4uE6EBAACqYLuzNWTIEPXr10+7d+9Wz549lZeXp+nTp5usDQAA4Lzt2LFD2dnZKigo0MUXX6xx48apfv36p+xTUVGh559/Xl988YUcDoceeughde3a1Svnt730g/TLsg0bN26UJF133XXu+VueYOmHwEfHEwDgKU/zgSf/1vTq1UudO3dWenq6Fi1apPnz52vWrFmn7PPRRx9p8eLFeu2111RQUKCMjAzNmTNHdevW9aiuM/FoOn/16tXVrl07tWvX7pyCFgAAwIV05MgRff/990pNTZUkpaam6vvvv9fRo0dP2W/p0qXq2rWrQkJCFB8fr1tvvVXLli3zSg1VDiO2bt3aPU/rtyzLksPh0Lp167xSCAAAgF2FhYUqLCystD02NvaUhtC+fftUq1YthYaGSvpl3dDLLrtM+/btU3x8/Cn71a5d2/08ISFB+/fv90qtVYat+fPne+VEAAAA3jJz5kzl5ORU2p6VleVeE/SPosqwVadOHVsH6tKli+bNm3feBQEAAFSld+/eyszMrLT99GlOCQkJOnDggPu2gxUVFTp48KASEhIq7bd37141a9ZMUuVO1/nw2hKsTqfTW4cCAAA4q9jYWNWtW7fS4/Swdckll6hRo0ZasmSJJGnJkiVq1KjRKUOIkpSSkqIPP/xQLpdLR48e1YoVK3T77bd7pVavha0zzesCAADwtVGjRmn27Nm6/fbbNXv2bD377LOSpH79+mnz5s2SpPT0dNWtW1e33XabunXrpscee0xXXHGFV87v0dIPZ5OZmamFCxdWuR9LPwQ+ln4AAHjK5NIPvua1zpaXMhsAAEBA8VrYat68ubcOBQAAEDBsDyMuXbpUN910k2JiYjRp0iRt2rRJTzzxhBITEz06IcOIgc+fWrsAgD+Goz+f9Gj/+KhIQ5V4n+3O1vTp0xUTE6NNmzbpyy+/VEZGhp5//nmTtQEAgCBhWZ49/IntsBUW9suSXP/4xz/UtWtXpaWlqbS01FhhAAAAgcB22HI4HFq6dKmWLl2qNm3aSJLKy8uNFQYAAIKHy7I8evgT22Hrr3/9q5YsWaIuXbroiiuuUF5enpKTk03WBgAAgoRlWR49/InX1tmyiwnygY8J8gAATx0oPOHR/rViow1V4n22O1t5eXm655571L59e0nSd999pylTphgrDAAAIBDYDlujRo3So48+6u5aNGrUSMuWLTNWGAAACB7M2dIvw3833XST+x6IISEhqlatmrHCAABA8AjkpR/C7O4YGhqq8vJyd9g6cOCAQkK8tgA9AAAIYv426d0TttPSvffeq6ysLB07dkxTpkzRvffeq759+5qsDQAAwO95dDXihg0btHLlSlmWpfbt26tly5Yen5CrEQMfVyMCADy16+hxj/avF1/DUCXex9IP8DrCFgDAU3lHCjzav/4lFxuqxPuqnLP1+OOPu+dpncmkSZO8WhAAAEAgqTJs3XLLLReiDgAAEMRcrsCdIM8wIryOYUQAgKe2Hzzm0f5XXxZnqBLvs3014sGDB/X4448rOTlZycnJGjhwoA4ePGiyNgAAECQC+d6ItsPWU089pQYNGujjjz/Wxx9/rIYNG+qpp54yWRsAAIDfs72o6aFDh5SVleV+3r9/f/3tb38zUhQAAAguLvlXt8oTtjtb9erV086dO93Pd+3apfr165uoCQAABJlAHka03dkqLS1Venq6WrRoIUnauHGjrr/+eg0cOFASS0AAAACcie2wlZaWprS0NPfz1NRUIwUBAIDg42/dKk/YDluZmZkm6wAAAEEsgJfZsh+2jhw5otmzZ2vXrl1yOp3u7QwfAgCA80VnS79cfdi4cWO1adNGoaGhJmsCAAAIGLbDVklJiUaOHGmyFgAAEKQCubNle+mH5s2b64cffjBZCwAACFIuy/Lo4U9sd7buvvtu3Xfffbr88ssVERHh3j5v3jwjhQEAgODhbwHKE7bD1tChQ/XII4+ocePGzNkCAACwyXbYioiI0AMPPGCyFgAAEKSYsyWpbdu2WrNmjclaAABAkArkOVsOy2aUbN26tQoKChQdHa3w8HBZliWHw6F169Z5dMKioqJzKhT+o3r16r4uAQDgZzbs2OPR/i2vrGOoEu+zPYw4f/58k3UAAIAgFsjDiLbDVp06/pMgAQCAf/G3oUFP2A5b+/bt0/jx47V161aVlpa6t3/22WdGCgMAAMEjkDtbtifIP/PMM2rTpo0sy9KECRPUokULbk4NAABQBdudrWPHjqlr166aNWuWrrvuOjVv3lzdu3dXVlaWRydk8jQAADhdADe27He2qlWrJkmKiorS3r175XQ6dfToUWOFAQCA4BHISz/Y7my1bNlSBQUFuueee9SpUyeFh4crJSXF4xMeeX2Wx++Bf7nkwV6SpJtH5fi4Epi0apRnXW0ACFa2w9awYcMkSRkZGWrVqpWKi4vVoEEDY4UBAIDgwQR5STt27HBfhbh9+3atWrVKx48fN1YYAAAIHoE8jGg7bA0aNEghISHKz8/XyJEjlZ+f7+52AQAAnA9fha2SkhINGjRIHTp0UEpKilauXHnG/VasWKFOnTopNTVVd955p958803b57A9jBgSEqJq1app9erVuueee9SvXz+lp6fbPhEAAMAfzRtvvKGYmBgtX75ceXl56tGjhz799FNFR0efsl/NmjU1ffp01apVS0VFRerUqZOaNWumli1bVnkO252t0tJSHT58WCtXrlTr1q0lBfb4KgAAuHAsy/Lo4S2ffPKJunfvLkmqX7++EhMTtWbNmkr7NW/eXLVq1ZL0yzJWV199tfbssXc/R9udrd69eyslJUVt2rRR06ZNlZ+fz5pZAADAKzwNUIWFhSosLKy0PTY2VrGxsbaPs3fv3lNuSZiQkKD9+/ef9T3bt2/Xv//9bz377LO2zmE7bHXv3t2d/CSpdu3aeuutt9zP582bpy5dutg9HAAAgJvLw2bVzJkzlZNTeYmhrKwsDRgwwP08MzNTe/fuPeMx1q5d69lJJR08eFD9+/fXyJEj3Z2uqtgOW6cLDQ1VaGio+/m7775L2AIAABdE7969z3jbwNO7WgsXLjzrcWrXrq09e/YoPj5e0i/3gk5OTj7jvkeOHFGfPn304IMP6o477rBd6zmHrdMxfwsAAJwrT3OEp8OFvyclJUVz585V06ZNlZeXp82bN2vixImV9jt27Jj69OmjHj16qGvXrh6dw/YE+ao4HA5vHQoAAAQZX02Qf+CBB1RYWKgOHTro4Ycf1ujRoxUTEyNJmjRpkt577z1J0quvvqq8vDzNnTtX6enpSk9P1/z5822dw2udLQAAgHPlkm9GyKKiojR58uQzvjZw4ED3r4cNG3bO64t6rbPFMCIAAEBlXutsvfDCC946FAAACDKB3LSxHbY2btyo8ePHKz8/XxUVFbIsSw6HQ+vWrZMkXXvttcaKBAAAgc3TpR/8ie2wNXz4cPXv319JSUkKCfHa6CMAAIBcAZy2bIetyMhIpaWlmawFAAAEqUAeRrTdorrpppu0evVqk7UAAIAg5aulHy4E252tuXPn6pVXXlF0dLTCw8MrzdkCAAA4V75a+uFCsB227C7cBQAAgP9jO2z99o7YAAAA3uRvQ4OeqDJsDR06VOPHj1fnzp3PeEueefPmGSkMAAAEjwDOWlWHrd69e0vSOS9RDwAAEMyqDFuJiYmSpFatWhkvBgAABCdXALe2bM/Z+umnnzRjxgzt2rVLTqfTvZ1hRAAAcL6Ces7WrwYOHKj09HRlZmYqNDTUZE0AACDIELYkhYWF6cEHHzRZCwAAQMCxvYJ827ZtWUEeAAAY4bIsjx7+xHZnq02bNurfv79CQkJYQR4AAHiVvwUoT9gOWyNGjNDYsWPVpEkThYTYbogBAABUiTlbkmrUqKGUlBSTtQAAAAQc2y2qW2+9Ve+9954KCgpUUlLifgAAAJwvl+XZw5/Y7my9/PLLkqRnn31WDofDPWdry5YtxooDAADBgWFESVu3bjVZBwAACGKBHLaY6Q4AAGCQ7c4WAACAKSz9AAAAYFAAZy3CFgAA8L1AnrNF2AIAAD7HMCIAAIBBgdzZ4mpEAAAAg+hsAQAAn2MYEQAAwCDCFgAAgEHM2QIAAMA5obMFAAB8LoAbW4QtAADge8zZAgAAMIg5WwAAADgndLYAAIDPBXJni7AFAAB8jjlbAAAABgVu1GLOFgAAgFF0tgAAgM8xjAgAAGBQIE+QZxgRAAD4nMtlefTwlpKSEg0aNEgdOnRQSkqKVq5cedb9S0tLdeedd6pTp062z0HYAgAAQeuNN95QTEyMli9frhkzZugvf/mLTpw48bv7v/TSS2revLlH5yBsAQAAn7Msy6OHt3zyySfq3r27JKl+/fpKTEzUmjVrzrjvhg0blJeXp/T0dI/OwZwtAADgc55OkC8sLFRhYWGl7bGxsYqNjbV9nL1796pOnTru5wkJCdq/f3+l/X7++WeNGTNG06dPV15enke12gpbGzdu1Pjx45Wfn6+KigpZliWHw6F169Z5dDIAAIAz8bRXNXPmTOXk5FTanpWVpQEDBrifZ2Zmau/evWc8xtq1a22f78UXX9S9996rWrVqmQlbw4cPV//+/ZWUlKSQEEYeAQCAd3k6NNi7d29lZmZW2n56V2vhwoVnPU7t2rW1Z88excfHS5L27dun5OTkSvv961//0po1azRt2jSVlpbq+PHjSktL0+LFi6us1VbYioyMVFpamp1dAQAAPObpMKKnw4W/JyUlRXPnzlXTpk2Vl5enzZs3a+LEiZX2+22oys3N1bhx47RgwQJb57DVprrpppu0evVqm2UDAAD4hwceeECFhYXq0KGDHn74YY0ePVoxMTGSpEmTJum9994773PY6mzNnTtXr7zyiqKjoxUeHs6cLQAA4FW+WtQ0KipKkydPPuNrAwcOPOP25ORk210tyWbYmj9/vu0DAgAAeCrob9fz20siAQAAvC2As9bZw9bQoUM1fvx4de7cWQ6Ho9Lr8+bNM1YYAABAIDhr2Ordu7ckadiwYRekGAAAEJwC+UbUZw1biYmJkqRWrVpdkGIAAEBwCvo5W0VFRXrttde0ZcsWlZaWurfPmjXLWGEAACB4BHLYsrXO1jPPPKOQkBDl5eWpW7duCg0NVbNmzUzXBgAA4Pdsha2dO3dq0KBBioyMVGpqql555RVt2LDBdG0AACBIWJbl0cOf2BpGDA8PlyRVq1ZNBQUFqlGjho4ePWq0MAAAEDz8LUB5wlbYql+/vgoKCpSWlqbu3burevXqatKkienaAABAkHAFbtayF7YmTJggSerTp4+aNm2qoqIitW3b1mhhAAAAgaDKOVsVFRXKzMx0P2/ZsqVuueUWhYXZymkAAABVCuo5W6GhoYqKilJpaakiIiIuRE0AACDI+FuA8oSt9tSVV16pHj166Pbbb1dUVJR7e48ePYwVBgAAgkcgr7NlK2wdO3ZMf/rTn/TTTz+5txUVFRG2AAAAqmArbO3du1dTp049Zdtv53EBAACcj6AdRnQ6nSovL5fL5dLJkyfdvxFFRUUqKSm5IAUCAIDAF7RLP8yYMUM5OTmSpKSkJPf2mJgY9enTx2xlAAAgaLgsl69LMOasYSsrK0tZWVkaPXq0RowYcaFqAgAAQSaARxHtzdkiaAEAAJMCec6WrRtRAwAA4NywDDwAAPC5oF9nCwAAwKRAHkYkbAEAAJ8L5LDFnC0AAACD6GwBAACfC9pFTQEAAC6EQB5GJGwBAACfcylwwxZztgAAAAyiswUAAHyOYUQAAACDXAE8Q56wBQAAfC6QO1vM2QIAADCIzhYAAPC5AB5FJGwBAADfC+RhRMIWAADwOYt1tgAAAHAu6GwBAACfczGMCAAAYA5ztgAAAAziakQAAACD6GwBAAAYFMhhi6sRAQAADCJsAQAAn3NZlkcPbykpKdGgQYPUoUMHpaSkaOXKlb+775YtW9SjRw917NhRHTt21OrVq22dg2FEAADgc75a+uGNN95QTEyMli9frry8PPXo0UOffvqpoqOjT9nv559/VlZWliZOnKikpCQ5nU4VFRXZOgedLQAA4HOWZXn08JZPPvlE3bt3lyTVr18FiGCfAAASrUlEQVRfiYmJWrNmTaX9lixZohYtWigpKUmSFBYWpri4OFvnoLMFAAD8TmFhoQoLCyttj42NVWxsrO3j7N27V3Xq1HE/T0hI0P79+yvtt23bNoWFhalfv346ePCgmjRpomHDhqlGjRpVnoOwBQAAfM7TZtXMmTOVk5NTaXtWVpYGDBjgfp6Zmam9e/ee8Rhr1661fT6Xy6X169fr/fff16WXXqqxY8fqhRde0NixY6t8L2ELAAD4nKdztnr37q3MzMxK20/vai1cuPCsx6ldu7b27Nmj+Ph4SdK+ffuUnJxcab+EhAQlJyfrsssukySlpaXpmWeesVUrc7YAAIDPeTpnKzY2VnXr1q308GQIUZJSUlI0d+5cSVJeXp42b96stm3bVtrvjjvu0KZNm1RcXCxJWrNmjRo2bGjrHHS2AABA0HrggQeUnZ2tDh06KCQkRKNHj1ZMTIwkadKkSbrssst0zz33qHbt2urXr5/uvvtuORwO1a1bV88995ytcxC2AACAz/lq6YeoqChNnjz5jK8NHDjwlOcZGRnKyMjw+ByELQAA4HOBfLsewhYAAPC5AM5aTJAHAAAwic4WAADwOV/N2boQLnjYuuTBXhf6lPCRVaOyfF0CAMBPBPKcLYcVyJ8OAADAxy54Z6tg/scX+pS4wC7ufJckadicJT6uBCaNuzdVknTk9Vk+rgSmMSIBnB8myAMAABhE2AIAADCIsAUAAGAQYQsAAMAgwhYAAIBBhC0AAACDCFsAAAAGEbYAAAAMImwBAAAYRNgCAAAwiLAFAABgEGELAADAIMIWAACAQYQtAAAAgwhbAAAABhG2AAAADCJsAQAAGETYAgAAMIiwBQAAYBBhCwAAwCDCFgAAgEGELQAAAIMIWwAAAAYRtgAAAAwibAEAABhE2AIAADCIsAUAAGAQYQsAAMAgwhYAAIBBhC0AAACDCFsAAAAGEbYAAAAMImwBAAAYRNgCAAAwiLAFAABgEGELAADAIMIWAACAQYQtAAAAgwhbAAAABhG2AAAADCJsAQAAGETYAgAAMIiwBQAAYBBhCwAAwCDCFgAAgEGELQAAAIMIWwAAAAYRtgAAAAwibAEAABhE2AIAADCIsAUAAGAQYQsAAMAgwhYAAIBBhC0AAACDCFsAAAAGEbYAAAAMImwBAAAYZCtsFRcX29oGAACAU9kKWz179rS1DQAAAKcKO9uLTqdT5eXlcrlcOnnypCzLkiQVFRWppKTkghQIAADgz84atmbMmKGcnBw5HA4lJSW5t8fExKhPnz7GiwMAAPB3Zw1bWVlZysrK0ujRozVixIgLVRMAAEDAsDVn68knn5TL5ZIk/fjjj/rb3/6msrIyo4UBAAAEAlthq1evXjp58qQOHTqkBx54QAsWLKDTBQAAYIOtsGVZlqKiorRq1Sp169ZNb7zxhr777jvTtQEAAPg9W2GrtLRUZWVl+sc//qE2bdr88sYQ1kMFAACoiq3E1LFjR914443avXu3rr/+eh06dEgRERGmawMAAPB7Z70a8VdZWVnq2bOnqlevrpCQEEVFRWnKlCmmawMAAPB7tudsffrpp5o4caIk6dixY9qzZ4/RwgAAAAKBrbA1duxYrV+/Xp999pkkKTo6WmPGjDFaGAAAQCCwFbZyc3M1YcIERUZGSpLi4uJUWlpqtDAAAIBAYCtsRUREyOFwuJ//usApAAAAzs7WBPkGDRro448/lmVZ2r17t1599VW1aNHCdG0AAAB+z1Zn67HHHtNXX32lQ4cOqVu3bnK5XHr00UdN1wYAAOD3bHW2+vfvr4ULF56yLTMzs9I2AAAAnOqsYcvpdKq8vFwul0snT56UZVlyOBwqLCxUSUnJhaoRAADAb501bM2YMUM5OTlyOBxKSkpyb4+JiVGfPn2MFwcAAODvzhq2srKylJWVpdGjR2vEiBEXqiYAAICAYWuCPEELAADg3NgKWwAAADg3hC0AAACDCFsAAAAGEbYAAAAMImwBAAAYRNgCAAAwiLAFAABgEGELAADAIMIWAACAQYQtAAAAgwhbAAAABhG2AAAADCJsAQAAGETYAgAAMIiwBQAAYBBhCwAAwCDCFgAAgEGELQAAAIMIWwAAAAYRtgAAAAwibAEAABhE2AIAADCIsAUAAGAQYQsAAMAgwhYAAIBBhC0AAACDCFsAAAAGEbYAAAAMImwBAAAYRNgCAAAwiLAFAABgEGELAADAIMIWAACAQYQtAAAAgwhbAAAABhG2AAAADCJsAQAAGETYAgAAMIiwBQAAYBBhCwAAwCDCFgAAgEEOy7IsXxcBAAAQqOhsAQAAGETYAgAAMIiwBQAAYBBhCwAAwCDCFgAAgEGELQAAAIMIWwAAAAYRtgAAAAwibAEAABhE2AIQ9KZMmaKysjKP37d582YNHjzYQEUw6Vy/b2+9H8GHsHWBTZkyRePGjTuvY6xYsUKbNm1yP8/NzVWnTp3OtzS/4a//wC1YsECPP/64V4+5e/duJScnn/dxnn76ad15550aNGiQF6ryPzk5OSovL6+03el0nvV9TZs21cSJE02VBUN+7/u+UO9H8AnzdQHw3IoVK5SYmKhmzZr5uhSf4B847zp8+LD+/ve/a8OGDQoJCb7/fz377LOSpLvvvlshISGqU6eO4uLitGPHDp04cUKLFi3S4MGDtWPHDpWXl6tevXoaM2aMatSoodzcXI0bN04LFizQ7t271blzZ919991avXq1SkpK9D//8z9q2bKljz8hfuv073v69OmaOnWqfvjhB5WWlio5OVlPP/20QkNDlZOToyVLligiIkIOh0OzZs3SSy+9dMr733nnHcXGxvryI8EPBNWNqBs2bKhBgwZpxYoVKigo0PPPP6+1a9fqiy++kNPp1KRJk3T11Vfr0KFDevLJJ3XixAmVlpaqXbt2euqppyRJw4cPV3R0tJ555hkdPnxY3bp109SpU9WoUaMznrOoqEjDhw/Xjz/+qJo1a+ryyy/XpZdeqmHDhqmsrEwvvfSS/vnPf6qsrEwNGzbUqFGjFB0drezsbIWFhWnbtm06duyYbrjhBo0YMUK5ubkaPHiwIiMjFRcXpz59+ighIUFjxoxR8+bN9fXXX8vhcOill17S1VdffSF/e8+b3e/nfP+Bmzt3rt5++22Fh4fL5XLp5Zdf1tVXX61x48bpq6++Unl5ueLi4jRmzBjVqVPHfY5u3brpiy++0MmTJzVhwgS9//77+uabbxQZGalp06apZs2aWrBggRYvXqyIiAjt2rVLl156qcaPH69atWppwYIFWrVqlSZPnixJWrhwoebMmaOKigrFxMRo1KhRuuqqq7Rx40Y999xzcrlccjqdevTRR5WamnrGz/Jrbbm5uZKkb775RhMmTNCJEyckSY8//rhuvvlmOZ1OPfzwwzp27JhKS0vVrFkzPfvssyorK1O3bt2Ul5enP/3pT8rMzNT999/v3S/WDzRs2FAbN250/9378ccfNXv2bEVFRUmSjh49qvj4eEnSSy+9pIqKCg0ZMqTSn8U///nPmjFjhm655RZ9/PHHmjNnjt5//31ffjScwW+/7+HDh+uGG25QRkaGXC6XhgwZotatW+u2227Trbfeqi+//FKRkZEqLi5WZGSkwsLCTnk/YIsVRBo0aGDNnj3bsizLWrp0qZWUlGR9/vnnlmVZ1quvvmoNHjzYsizLOnnypFVcXGxZlmWVlZVZPXv2tFavXm1ZlmWVlJRYqamp1vLly63evXu7j/d7xo4da2VnZ1uWZVlHjhyx2rVrZ73wwguWZVnW1KlTralTp7r3ffHFF63//d//tSzLsoYNG2alpqZaxcXFVnl5udWnTx/rnXfecb/2668ty7LWr19vNW7c2Pruu+8sy7KsadOmWU8++eR5/E75ht3vZ/369VZmZqZlWZaVn59vNWjQwL3fokWLrO7du5/1PNdff7114MABy7Isq7S01Pr5558ty/rl+/nVBx98YA0aNOiUc6xcudKyLMt67bXXrBYtWljff/+9ZVmWNXLkSPf3Nn/+fKtp06bW9u3bLcuyrClTplgDBgxwv/brr//5z39a/fr1s0pLSy3LsqxVq1a5637kkUesxYsXW5ZlWS6Xyzp+/Pjvfpb8/HyrVatWlmVZ1vHjx6309HT3Zztw4IDVtm1b6/jx45bL5bKOHj3qPubQoUOtOXPmVDpGsGrQoIH77/ywYcOsV1555ZTX33zzTSszM9NKTU212rdvb/Xt29eyrMp/FpOSktzv2blzp/Vf//VfF+gTwBO//b5bt25tpaamWnfddZd11113Wbfddpv1wgsvWE6n08rMzLQGDBhgzZ071/336vT3A3YE3TDiHXfcIUlq0qSJJOmWW26RJCUmJmr58uWSpIqKCr344ov6+uuvZVmWDh8+rK1bt+qmm25SZGSkXn75ZXXp0kVt27ZVjx49znq+3Nxc/eUvf5EkxcfHq0OHDu7XPv/8cxUXF+vvf/+7JKmsrEzXXnut+/WOHTu6/+eUkZGhTz/9VPfdd98Zz3PllVeqcePGkqSkpCStXLnSs9+YPwg738/poqKi3PslJSVVOSeudevWys7O1i233KKbb75ZV1xxhSRpzZo1mjNnjn7++edKc3WioqJ08803u2u7/PLL3d3MJk2aaO3ate59W7RooauuukqS1LVrV6WlpVWq4fPPP9fWrVvVtWtXSZJlWSosLJQkJScna/r06dq1a5duvPFGNW/e/Kyf51dff/21du/erX79+rm3ORwO7dy5U40bN9abb76pNWvWyOVy6fjx44qMjLR13GD0a0dLkjZs2KD33ntP77//vuLj47V48WJ98MEHZ3xfeHi4+9chISFVzvmC71mWpWnTprl/DvzWBx98oI0bN2r9+vXq1KmTXn/99VN+RgN2BV3YioiIkPTLD8Lf+8H41ltvqbCwUB9++KEiIiL017/+VaWlpe59t2/frujoaB06dEhOp1NhYef222hZlkaOHKk2bdqcxyf6RaD8kLfz/ZzO08+ek5OjzZs3a/369erVq5dGjRqla665RmPHjtW8efN0xRVXaOPGjRoyZMjvnuO3z0NDQ1VRUeHR57QsS507d9bAgQMrvXb//ferffv2Wrt2rZ577jndeOONeuKJJ2wds2HDhnr33XcrvfbRRx/pX//6l959913FxMRoxowZysvL86jmQBYdHa3i4uIzDgsVFhYqJiZGF198scrKyjR//nwfVAhv+u333b59e7366qsaNWqUQkNDdfToUZ04cUJxcXH6+eef1apVK7Vq1Ur//ve/9Z///EfXXnvtWf+8AGcSfLNhbSgqKlLNmjUVERGhAwcO6LPPPnO/lp+frzFjxmj27NmqV6+eXn755bMeq3Xr1lqwYIEk6dixY1qxYoX7tfbt2+vtt9/WyZMnJUnFxcXavn27+/Vly5a5uyyLFi1S69atJUkxMTEqKiry2ucNJk6nU/n5+WrWrJkeeugh3XjjjdqyZYuKi4tVrVo11axZUy6X67zm2WzcuNEdZObPn+/+3n6rffv2WrRokfbv3y/pl27qt99+K0nasWOH6tWrp7vvvlu9evXS5s2bbZ33uuuu086dO7V+/Xr3tk2bNsmyLBUVFSkuLs79Z2fJkiXn/PkCUd++fdWrVy+lp6e7O4y/atu2rerVq6fbb79d9913n7uDDP/12+97wIABCgkJUXp6utLS0vTggw/qwIEDKi4u1mOPPaa0tDSlpqbq0ksv1W233Vbp/af/eQHOJOg6W3b07NlTAwcOVGpqqmrVquXuPJWVlemJJ57Q4MGDVb9+fY0cOVJdunTRDTfcoHbt2p3xWP3799czzzyjlJQU1axZ85SJ2w899JBycnLUpUsXORwOORwOZWVluSe2N23aVH379tXRo0fVqlUrdevWTZJ011136emnn9ayZcvcE+Rhj8vlUnZ2toqKiuRwOJSQkKDBgwcrLi5OKSkp6tixo+Li4tSuXTtt2LDhnM5x/fXXa9y4cdq5c6d7gvzpbrjhBg0aNEiPPvqoKioqVF5erpSUFCUmJuqdd95Rbm6uqlWrpvDwcPcwdFVq1KihadOmafz48RozZozKy8t1xRVXaMaMGcrIyNBnn32mlJQUXXLJJWrRosUp3dpgl5WVpaysrDO+Vq1atd/9T1VycrL7P1N169Z1X6hwpuf44zj9+/71CsXTffjhh7beD1QlqK5G9CfZ2dlKTEz83Tla+GM6/YpDAAAYRgQAADCIzpYXHDlyRH379q20vUOHDrSafWTLli3Kzs6utP2+++5zXwHoT0aMGKFvvvnmlG2hoaHuISwAwB8XYQsAAMAghhEBAAAMImwBAAAYRNgCAAAwiLAFAABgEGELAADAoP8Hi+h+DTFhrN0AAAAASUVORK5CYII=",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f28fa579550>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "sns.set(style=\"white\")\n",
        "\n",
        "# Generate a mask for the upper triangle\n",
        "mask = np.zeros_like(corr, dtype=np.bool)\n",
        "mask[np.triu_indices_from(mask)] = True\n",
        "\n",
        "# Set up the matplotlib figure\n",
        "f, ax = plt.subplots(figsize=(11, 9))\n",
        "\n",
        "# Generate a custom diverging colormap\n",
        "cmap = sns.diverging_palette(220, 10, as_cmap=True)\n",
        "\n",
        "# Draw the heatmap with the mask and correct aspect ratio\n",
        "sns.heatmap(corr, mask=mask, cmap=cmap, vmax=.3, center=0,\n",
        "            square=True, linewidths=.5, cbar_kws={\"shrink\": .5})"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "colab_type": "code",
        "id": "7_yHHDkEiqCC",
        "outputId": "5569f2aa-acf9-4a40-89f0-45e56ed70a21"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>max_depth</th>\n",
              "      <th>min_samples_leaf</th>\n",
              "      <th>train</th>\n",
              "      <th>test</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>4</td>\n",
              "      <td>192</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>3</td>\n",
              "      <td>128</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>3</td>\n",
              "      <td>192</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>3</td>\n",
              "      <td>256</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>4</td>\n",
              "      <td>256</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    max_depth  min_samples_leaf      train       test\n",
              "13          4               192  78.750993  78.672633\n",
              "8           3               128  78.750993  78.672633\n",
              "9           3               192  78.750993  78.672633\n",
              "10          3               256  78.750993  78.672633\n",
              "14          4               256  78.750993  78.672633"
            ]
          },
          "execution_count": 37,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "resultados.sort_values(\"test\", ascending=False).head()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "pwwC1wHakOvC"
      },
      "source": [
        "# Explorando 3 dimensões de hiper parâmetros"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "colab_type": "code",
        "id": "jlCg0np0kSsa",
        "outputId": "c4b1c47c-ba31-49a3-a68f-40804b2df126"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>max_depth</th>\n",
              "      <th>min_samples_leaf</th>\n",
              "      <th>min_samples_split</th>\n",
              "      <th>train</th>\n",
              "      <th>test</th>\n",
              "      <th>fit_time</th>\n",
              "      <th>score_time</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>32</td>\n",
              "      <td>32</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "      <td>0.005269</td>\n",
              "      <td>0.000704</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>32</td>\n",
              "      <td>64</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "      <td>0.005171</td>\n",
              "      <td>0.000776</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>32</td>\n",
              "      <td>128</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "      <td>0.005182</td>\n",
              "      <td>0.000680</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>32</td>\n",
              "      <td>256</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "      <td>0.005308</td>\n",
              "      <td>0.000796</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>64</td>\n",
              "      <td>32</td>\n",
              "      <td>75.791169</td>\n",
              "      <td>75.784219</td>\n",
              "      <td>0.005331</td>\n",
              "      <td>0.000676</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   max_depth  min_samples_leaf  min_samples_split      train       test  \\\n",
              "0          1                32                 32  75.791169  75.784219   \n",
              "1          1                32                 64  75.791169  75.784219   \n",
              "2          1                32                128  75.791169  75.784219   \n",
              "3          1                32                256  75.791169  75.784219   \n",
              "4          1                64                 32  75.791169  75.784219   \n",
              "\n",
              "   fit_time  score_time  \n",
              "0  0.005269    0.000704  \n",
              "1  0.005171    0.000776  \n",
              "2  0.005182    0.000680  \n",
              "3  0.005308    0.000796  \n",
              "4  0.005331    0.000676  "
            ]
          },
          "execution_count": 49,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "def roda_arvore_de_decisao(max_depth, min_samples_leaf, min_samples_split):\n",
        "  SEED = 301\n",
        "  np.random.seed(SEED)\n",
        "\n",
        "  cv = GroupKFold(n_splits = 10)\n",
        "  modelo = DecisionTreeClassifier(max_depth=max_depth, min_samples_leaf = min_samples_leaf, min_samples_split = min_samples_split)\n",
        "  results = cross_validate(modelo, x_azar, y_azar, cv = cv, groups = dados.modelo, return_train_score=True)\n",
        "  fit_time = results['fit_time'].mean()\n",
        "  score_time = results['score_time'].mean()\n",
        "  train_score = results['train_score'].mean() * 100\n",
        "  test_score = results['test_score'].mean() * 100\n",
        "  tabela = [max_depth, min_samples_leaf, min_samples_split, train_score, test_score, fit_time, score_time]\n",
        "  return tabela\n",
        "\n",
        "def busca():\n",
        "  resultados = []\n",
        "  for max_depth in range(1,33):\n",
        "    for min_samples_leaf in [32, 64, 128, 256]:\n",
        "      for min_samples_split in [32, 64, 128, 256]:\n",
        "        tabela = roda_arvore_de_decisao(max_depth, min_samples_leaf, min_samples_split)\n",
        "        resultados.append(tabela)\n",
        "  resultados = pd.DataFrame(resultados, columns= [\"max_depth\",\"min_samples_leaf\",\"min_samples_split\",\"train\",\"test\", \"fit_time\", \"score_time\"])\n",
        "  return resultados\n",
        "\n",
        "resultados = busca()\n",
        "resultados.head()\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {},
        "colab_type": "code",
        "id": "vUmCKerTkqdS"
      },
      "outputs": [],
      "source": [
        "corr = resultados.corr()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 654
        },
        "colab_type": "code",
        "id": "a7d6zbMglbBJ",
        "outputId": "63e661b9-79da-4a7f-f13f-b3b2a8d110ee"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f28fa294b38>"
            ]
          },
          "execution_count": 51,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f28fa294438>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "sns.set(style=\"white\")\n",
        "\n",
        "# Generate a mask for the upper triangle\n",
        "mask = np.zeros_like(corr, dtype=np.bool)\n",
        "mask[np.triu_indices_from(mask)] = True\n",
        "\n",
        "# Set up the matplotlib figure\n",
        "f, ax = plt.subplots(figsize=(11, 9))\n",
        "\n",
        "# Generate a custom diverging colormap\n",
        "cmap = sns.diverging_palette(220, 10, as_cmap=True)\n",
        "\n",
        "# Draw the heatmap with the mask and correct aspect ratio\n",
        "sns.heatmap(corr, mask=mask, cmap=cmap, vmax=.3, center=0,\n",
        "            square=True, linewidths=.5, cbar_kws={\"shrink\": .5})"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "colab_type": "code",
        "id": "YY5vVBEGlctR",
        "outputId": "855cdcbd-8951-4c90-f307-d35fac841790"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>max_depth</th>\n",
              "      <th>min_samples_leaf</th>\n",
              "      <th>min_samples_split</th>\n",
              "      <th>train</th>\n",
              "      <th>test</th>\n",
              "      <th>fit_time</th>\n",
              "      <th>score_time</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>63</th>\n",
              "      <td>4</td>\n",
              "      <td>256</td>\n",
              "      <td>256</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "      <td>0.011748</td>\n",
              "      <td>0.000834</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>48</th>\n",
              "      <td>4</td>\n",
              "      <td>32</td>\n",
              "      <td>32</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "      <td>0.012379</td>\n",
              "      <td>0.000861</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>34</th>\n",
              "      <td>3</td>\n",
              "      <td>32</td>\n",
              "      <td>128</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "      <td>0.010429</td>\n",
              "      <td>0.000840</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35</th>\n",
              "      <td>3</td>\n",
              "      <td>32</td>\n",
              "      <td>256</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "      <td>0.010391</td>\n",
              "      <td>0.000839</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>36</th>\n",
              "      <td>3</td>\n",
              "      <td>64</td>\n",
              "      <td>32</td>\n",
              "      <td>78.750993</td>\n",
              "      <td>78.672633</td>\n",
              "      <td>0.010453</td>\n",
              "      <td>0.000818</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    max_depth  min_samples_leaf  min_samples_split      train       test  \\\n",
              "63          4               256                256  78.750993  78.672633   \n",
              "48          4                32                 32  78.750993  78.672633   \n",
              "34          3                32                128  78.750993  78.672633   \n",
              "35          3                32                256  78.750993  78.672633   \n",
              "36          3                64                 32  78.750993  78.672633   \n",
              "\n",
              "    fit_time  score_time  \n",
              "63  0.011748    0.000834  \n",
              "48  0.012379    0.000861  \n",
              "34  0.010429    0.000840  \n",
              "35  0.010391    0.000839  \n",
              "36  0.010453    0.000818  "
            ]
          },
          "execution_count": 52,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "resultados.sort_values(\"test\", ascending=False).head()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "at5ydyJOqQMv"
      },
      "source": [
        "# Explorando espaço de hiper parâmetros com GridSearchCV"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 983
        },
        "colab_type": "code",
        "id": "Zb69KkJwljln",
        "outputId": "5a5b77b6-5659-43de-d7ef-6c0b1bcbe64f"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('mean_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split0_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split1_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split2_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split3_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split4_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split5_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split6_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split7_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split8_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split9_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('std_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>mean_fit_time</th>\n",
              "      <th>mean_score_time</th>\n",
              "      <th>mean_test_score</th>\n",
              "      <th>mean_train_score</th>\n",
              "      <th>param_criterion</th>\n",
              "      <th>param_max_depth</th>\n",
              "      <th>param_min_samples_leaf</th>\n",
              "      <th>param_min_samples_split</th>\n",
              "      <th>params</th>\n",
              "      <th>rank_test_score</th>\n",
              "      <th>...</th>\n",
              "      <th>split7_test_score</th>\n",
              "      <th>split7_train_score</th>\n",
              "      <th>split8_test_score</th>\n",
              "      <th>split8_train_score</th>\n",
              "      <th>split9_test_score</th>\n",
              "      <th>split9_train_score</th>\n",
              "      <th>std_fit_time</th>\n",
              "      <th>std_score_time</th>\n",
              "      <th>std_test_score</th>\n",
              "      <th>std_train_score</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.011080</td>\n",
              "      <td>0.000818</td>\n",
              "      <td>0.7868</td>\n",
              "      <td>0.78751</td>\n",
              "      <td>gini</td>\n",
              "      <td>3</td>\n",
              "      <td>32</td>\n",
              "      <td>32</td>\n",
              "      <td>{'criterion': 'gini', 'max_depth': 3, 'min_sam...</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>0.781818</td>\n",
              "      <td>0.788124</td>\n",
              "      <td>0.77551</td>\n",
              "      <td>0.788803</td>\n",
              "      <td>0.790262</td>\n",
              "      <td>0.786834</td>\n",
              "      <td>0.001471</td>\n",
              "      <td>0.000067</td>\n",
              "      <td>0.011338</td>\n",
              "      <td>0.001303</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.010413</td>\n",
              "      <td>0.000825</td>\n",
              "      <td>0.7868</td>\n",
              "      <td>0.78751</td>\n",
              "      <td>gini</td>\n",
              "      <td>3</td>\n",
              "      <td>32</td>\n",
              "      <td>64</td>\n",
              "      <td>{'criterion': 'gini', 'max_depth': 3, 'min_sam...</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>0.781818</td>\n",
              "      <td>0.788124</td>\n",
              "      <td>0.77551</td>\n",
              "      <td>0.788803</td>\n",
              "      <td>0.790262</td>\n",
              "      <td>0.786834</td>\n",
              "      <td>0.000441</td>\n",
              "      <td>0.000092</td>\n",
              "      <td>0.011338</td>\n",
              "      <td>0.001303</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.010458</td>\n",
              "      <td>0.000863</td>\n",
              "      <td>0.7868</td>\n",
              "      <td>0.78751</td>\n",
              "      <td>gini</td>\n",
              "      <td>3</td>\n",
              "      <td>32</td>\n",
              "      <td>128</td>\n",
              "      <td>{'criterion': 'gini', 'max_depth': 3, 'min_sam...</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>0.781818</td>\n",
              "      <td>0.788124</td>\n",
              "      <td>0.77551</td>\n",
              "      <td>0.788803</td>\n",
              "      <td>0.790262</td>\n",
              "      <td>0.786834</td>\n",
              "      <td>0.000382</td>\n",
              "      <td>0.000077</td>\n",
              "      <td>0.011338</td>\n",
              "      <td>0.001303</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.010731</td>\n",
              "      <td>0.000807</td>\n",
              "      <td>0.7868</td>\n",
              "      <td>0.78751</td>\n",
              "      <td>gini</td>\n",
              "      <td>3</td>\n",
              "      <td>64</td>\n",
              "      <td>32</td>\n",
              "      <td>{'criterion': 'gini', 'max_depth': 3, 'min_sam...</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>0.781818</td>\n",
              "      <td>0.788124</td>\n",
              "      <td>0.77551</td>\n",
              "      <td>0.788803</td>\n",
              "      <td>0.790262</td>\n",
              "      <td>0.786834</td>\n",
              "      <td>0.000492</td>\n",
              "      <td>0.000057</td>\n",
              "      <td>0.011338</td>\n",
              "      <td>0.001303</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.010403</td>\n",
              "      <td>0.000820</td>\n",
              "      <td>0.7868</td>\n",
              "      <td>0.78751</td>\n",
              "      <td>gini</td>\n",
              "      <td>3</td>\n",
              "      <td>64</td>\n",
              "      <td>64</td>\n",
              "      <td>{'criterion': 'gini', 'max_depth': 3, 'min_sam...</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>0.781818</td>\n",
              "      <td>0.788124</td>\n",
              "      <td>0.77551</td>\n",
              "      <td>0.788803</td>\n",
              "      <td>0.790262</td>\n",
              "      <td>0.786834</td>\n",
              "      <td>0.000353</td>\n",
              "      <td>0.000045</td>\n",
              "      <td>0.011338</td>\n",
              "      <td>0.001303</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 34 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "   mean_fit_time  mean_score_time  mean_test_score  mean_train_score  \\\n",
              "0       0.011080         0.000818           0.7868           0.78751   \n",
              "1       0.010413         0.000825           0.7868           0.78751   \n",
              "2       0.010458         0.000863           0.7868           0.78751   \n",
              "3       0.010731         0.000807           0.7868           0.78751   \n",
              "4       0.010403         0.000820           0.7868           0.78751   \n",
              "\n",
              "  param_criterion param_max_depth param_min_samples_leaf  \\\n",
              "0            gini               3                     32   \n",
              "1            gini               3                     32   \n",
              "2            gini               3                     32   \n",
              "3            gini               3                     64   \n",
              "4            gini               3                     64   \n",
              "\n",
              "  param_min_samples_split                                             params  \\\n",
              "0                      32  {'criterion': 'gini', 'max_depth': 3, 'min_sam...   \n",
              "1                      64  {'criterion': 'gini', 'max_depth': 3, 'min_sam...   \n",
              "2                     128  {'criterion': 'gini', 'max_depth': 3, 'min_sam...   \n",
              "3                      32  {'criterion': 'gini', 'max_depth': 3, 'min_sam...   \n",
              "4                      64  {'criterion': 'gini', 'max_depth': 3, 'min_sam...   \n",
              "\n",
              "   rank_test_score       ...         split7_test_score  split7_train_score  \\\n",
              "0                1       ...                  0.781818            0.788124   \n",
              "1                1       ...                  0.781818            0.788124   \n",
              "2                1       ...                  0.781818            0.788124   \n",
              "3                1       ...                  0.781818            0.788124   \n",
              "4                1       ...                  0.781818            0.788124   \n",
              "\n",
              "   split8_test_score  split8_train_score  split9_test_score  \\\n",
              "0            0.77551            0.788803           0.790262   \n",
              "1            0.77551            0.788803           0.790262   \n",
              "2            0.77551            0.788803           0.790262   \n",
              "3            0.77551            0.788803           0.790262   \n",
              "4            0.77551            0.788803           0.790262   \n",
              "\n",
              "   split9_train_score  std_fit_time  std_score_time  std_test_score  \\\n",
              "0            0.786834      0.001471        0.000067        0.011338   \n",
              "1            0.786834      0.000441        0.000092        0.011338   \n",
              "2            0.786834      0.000382        0.000077        0.011338   \n",
              "3            0.786834      0.000492        0.000057        0.011338   \n",
              "4            0.786834      0.000353        0.000045        0.011338   \n",
              "\n",
              "   std_train_score  \n",
              "0         0.001303  \n",
              "1         0.001303  \n",
              "2         0.001303  \n",
              "3         0.001303  \n",
              "4         0.001303  \n",
              "\n",
              "[5 rows x 34 columns]"
            ]
          },
          "execution_count": 53,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "from sklearn.model_selection import GridSearchCV\n",
        "\n",
        "SEED=301\n",
        "np.random.seed(SEED)\n",
        "\n",
        "espaco_de_parametros = {\n",
        "    \"max_depth\" : [3, 5],\n",
        "    \"min_samples_split\" : [32, 64, 128],\n",
        "    \"min_samples_leaf\" : [32, 64, 128],\n",
        "    \"criterion\" : [\"gini\", \"entropy\"]\n",
        "}\n",
        "\n",
        "busca = GridSearchCV(DecisionTreeClassifier(),\n",
        "                    espaco_de_parametros,\n",
        "                    cv = GroupKFold(n_splits = 10))\n",
        "busca.fit(x_azar, y_azar, groups = dados.modelo)\n",
        "resultados = pd.DataFrame(busca.cv_results_)\n",
        "resultados.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 50
        },
        "colab_type": "code",
        "id": "xGEqRLcUrMuF",
        "outputId": "ada074b6-eedf-40eb-bc35-9af10deba1ea"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "{'criterion': 'gini', 'max_depth': 3, 'min_samples_leaf': 32, 'min_samples_split': 32}\n",
            "78.68\n"
          ]
        }
      ],
      "source": [
        "print(busca.best_params_)\n",
        "print(busca.best_score_ * 100)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 118
        },
        "colab_type": "code",
        "id": "EcXir723rbI5",
        "outputId": "46abd9f6-d8f6-4883-f730-478f8a116f6f"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=3,\n",
              "            max_features=None, max_leaf_nodes=None,\n",
              "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
              "            min_samples_leaf=32, min_samples_split=32,\n",
              "            min_weight_fraction_leaf=0.0, presort=False, random_state=None,\n",
              "            splitter='best')"
            ]
          },
          "execution_count": 57,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "melhor = busca.best_estimator_\n",
        "melhor"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "colab_type": "code",
        "id": "fBUiCF3krkZo",
        "outputId": "4c93a18a-4dd2-4e3d-bd5e-a7be0e881f03"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Accuracy para os dados foi 78.75%\n"
          ]
        }
      ],
      "source": [
        "from sklearn.metrics import accuracy_score\n",
        "\n",
        "# evitar essa abordagem pois estará sendo otimista\n",
        "\n",
        "predicoes = melhor.predict(x_azar)\n",
        "accuracy = accuracy_score(predicoes, y_azar) * 100\n",
        "print(\"Accuracy para os dados foi %.2f%%\" % accuracy)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "mM-vg1Mrtj7V"
      },
      "source": [
        "# Como ter uma estimativa sem esse vício nos dados que eu já vi?\n",
        "\n",
        "No caso de cross validation com busca de hiper parâmetros, fazemos uma nova validação cruzada. Chama-se nested cross validation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1980
        },
        "colab_type": "code",
        "id": "cqZznVDKr--h",
        "outputId": "f444dc7b-47ba-493c-deac-a0b618300f3f"
      },
      "outputs": [
        {
          "ename": "ValueError",
          "evalue": "ignored",
          "output_type": "error",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-61-5d0a4e0c67ad>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0msklearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel_selection\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mcross_val_score\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mscores\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcross_val_score\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbusca\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx_azar\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_azar\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcv\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mGroupKFold\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn_splits\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m10\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgroups\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdados\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodelo\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/model_selection/_validation.py\u001b[0m in \u001b[0;36mcross_val_score\u001b[0;34m(estimator, X, y, groups, scoring, cv, n_jobs, verbose, fit_params, pre_dispatch)\u001b[0m\n\u001b[1;32m    340\u001b[0m                                 \u001b[0mn_jobs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mn_jobs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mverbose\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    341\u001b[0m                                 \u001b[0mfit_params\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfit_params\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 342\u001b[0;31m                                 pre_dispatch=pre_dispatch)\n\u001b[0m\u001b[1;32m    343\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mcv_results\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'test_score'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    344\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/model_selection/_validation.py\u001b[0m in \u001b[0;36mcross_validate\u001b[0;34m(estimator, X, y, groups, scoring, cv, n_jobs, verbose, fit_params, pre_dispatch, return_train_score)\u001b[0m\n\u001b[1;32m    204\u001b[0m             \u001b[0mfit_params\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mreturn_train_score\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mreturn_train_score\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    205\u001b[0m             return_times=True)\n\u001b[0;32m--> 206\u001b[0;31m         for train, test in cv.split(X, y, groups))\n\u001b[0m\u001b[1;32m    207\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    208\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mreturn_train_score\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/externals/joblib/parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, iterable)\u001b[0m\n\u001b[1;32m    777\u001b[0m             \u001b[0;31m# was dispatched. In particular this covers the edge\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    778\u001b[0m             \u001b[0;31m# case of Parallel used with an exhausted iterator.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 779\u001b[0;31m             \u001b[0;32mwhile\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdispatch_one_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    780\u001b[0m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_iterating\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    781\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/externals/joblib/parallel.py\u001b[0m in \u001b[0;36mdispatch_one_batch\u001b[0;34m(self, iterator)\u001b[0m\n\u001b[1;32m    623\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    624\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 625\u001b[0;31m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dispatch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtasks\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    626\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    627\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/externals/joblib/parallel.py\u001b[0m in \u001b[0;36m_dispatch\u001b[0;34m(self, batch)\u001b[0m\n\u001b[1;32m    586\u001b[0m         \u001b[0mdispatch_timestamp\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    587\u001b[0m         \u001b[0mcb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mBatchCompletionCallBack\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdispatch_timestamp\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 588\u001b[0;31m         \u001b[0mjob\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply_async\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcallback\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    589\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_jobs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mjob\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    590\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/externals/joblib/_parallel_backends.py\u001b[0m in \u001b[0;36mapply_async\u001b[0;34m(self, func, callback)\u001b[0m\n\u001b[1;32m    109\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mapply_async\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcallback\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    110\u001b[0m         \u001b[0;34m\"\"\"Schedule a func to be run\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 111\u001b[0;31m         \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mImmediateResult\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    112\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mcallback\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    113\u001b[0m             \u001b[0mcallback\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresult\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/externals/joblib/_parallel_backends.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, batch)\u001b[0m\n\u001b[1;32m    330\u001b[0m         \u001b[0;31m# Don't delay the application, to avoid keeping the input\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    331\u001b[0m         \u001b[0;31m# arguments in memory\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 332\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mbatch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    333\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    334\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/externals/joblib/parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    129\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    130\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__call__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 131\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mitems\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    132\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    133\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__len__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/externals/joblib/parallel.py\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m    129\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    130\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__call__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 131\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mitems\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    132\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    133\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__len__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/model_selection/_validation.py\u001b[0m in \u001b[0;36m_fit_and_score\u001b[0;34m(estimator, X, y, scorer, train, test, verbose, parameters, fit_params, return_train_score, return_parameters, return_n_test_samples, return_times, error_score)\u001b[0m\n\u001b[1;32m    456\u001b[0m             \u001b[0mestimator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mfit_params\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    457\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 458\u001b[0;31m             \u001b[0mestimator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mfit_params\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    459\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    460\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/model_selection/_search.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, groups, **fit_params)\u001b[0m\n\u001b[1;32m    638\u001b[0m                                   error_score=self.error_score)\n\u001b[1;32m    639\u001b[0m           for parameters, (train, test) in product(candidate_params,\n\u001b[0;32m--> 640\u001b[0;31m                                                    cv.split(X, y, groups)))\n\u001b[0m\u001b[1;32m    641\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    642\u001b[0m         \u001b[0;31m# if one choose to see train score, \"out\" will contain train score info\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/model_selection/_split.py\u001b[0m in \u001b[0;36msplit\u001b[0;34m(self, X, y, groups)\u001b[0m\n\u001b[1;32m    330\u001b[0m                                                              n_samples))\n\u001b[1;32m    331\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 332\u001b[0;31m         \u001b[0;32mfor\u001b[0m \u001b[0mtrain\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest\u001b[0m \u001b[0;32min\u001b[0m \u001b[0msuper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_BaseKFold\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msplit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    333\u001b[0m             \u001b[0;32myield\u001b[0m \u001b[0mtrain\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    334\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/model_selection/_split.py\u001b[0m in \u001b[0;36msplit\u001b[0;34m(self, X, y, groups)\u001b[0m\n\u001b[1;32m     93\u001b[0m         \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgroups\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mindexable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     94\u001b[0m         \u001b[0mindices\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0marange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_num_samples\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 95\u001b[0;31m         \u001b[0;32mfor\u001b[0m \u001b[0mtest_index\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_iter_test_masks\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     96\u001b[0m             \u001b[0mtrain_index\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mindices\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogical_not\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtest_index\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     97\u001b[0m             \u001b[0mtest_index\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mindices\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mtest_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/model_selection/_split.py\u001b[0m in \u001b[0;36m_iter_test_masks\u001b[0;34m(self, X, y, groups)\u001b[0m\n\u001b[1;32m    105\u001b[0m         \u001b[0mBy\u001b[0m \u001b[0mdefault\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdelegates\u001b[0m \u001b[0mto\u001b[0m \u001b[0m_iter_test_indices\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    106\u001b[0m         \"\"\"\n\u001b[0;32m--> 107\u001b[0;31m         \u001b[0;32mfor\u001b[0m \u001b[0mtest_index\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_iter_test_indices\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    108\u001b[0m             \u001b[0mtest_mask\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mzeros\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_num_samples\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbool\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    109\u001b[0m             \u001b[0mtest_mask\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mtest_index\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/sklearn/model_selection/_split.py\u001b[0m in \u001b[0;36m_iter_test_indices\u001b[0;34m(self, X, y, groups)\u001b[0m\n\u001b[1;32m    487\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_iter_test_indices\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgroups\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    488\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mgroups\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 489\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"The 'groups' parameter should not be None.\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    490\u001b[0m         \u001b[0mgroups\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcheck_array\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgroups\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mensure_2d\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    491\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mValueError\u001b[0m: The 'groups' parameter should not be None."
          ]
        }
      ],
      "source": [
        "from sklearn.model_selection import cross_val_score\n",
        "\n",
        "scores = cross_val_score(busca, x_azar, y_azar, cv = GroupKFold(n_splits=10), groups = dados.modelo)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "E_MEMPRYuWYT"
      },
      "source": [
        "# Infelizmente como o Pandas não suporta nested validation com group k fold não conseguimos prever o resultado para novos grupos"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 815
        },
        "colab_type": "code",
        "id": "9aYIMS1At7ZW",
        "outputId": "796c67c2-85f0-43cd-fb4e-439162257201"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('mean_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split0_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split1_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split2_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split3_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('split4_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:122: FutureWarning: You are accessing a training score ('std_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
            "  warnings.warn(*warn_args, **warn_kwargs)\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>mean_fit_time</th>\n",
              "      <th>mean_score_time</th>\n",
              "      <th>mean_test_score</th>\n",
              "      <th>mean_train_score</th>\n",
              "      <th>param_criterion</th>\n",
              "      <th>param_max_depth</th>\n",
              "      <th>param_min_samples_leaf</th>\n",
              "      <th>param_min_samples_split</th>\n",
              "      <th>params</th>\n",
              "      <th>rank_test_score</th>\n",
              "      <th>...</th>\n",
              "      <th>split2_test_score</th>\n",
              "      <th>split2_train_score</th>\n",
              "      <th>split3_test_score</th>\n",
              "      <th>split3_train_score</th>\n",
              "      <th>split4_test_score</th>\n",
              "      <th>split4_train_score</th>\n",
              "      <th>std_fit_time</th>\n",
              "      <th>std_score_time</th>\n",
              "      <th>std_test_score</th>\n",
              "      <th>std_train_score</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.010791</td>\n",
              "      <td>0.001162</td>\n",
              "      <td>0.787</td>\n",
              "      <td>0.787525</td>\n",
              "      <td>gini</td>\n",
              "      <td>3</td>\n",
              "      <td>32</td>\n",
              "      <td>32</td>\n",
              "      <td>{'criterion': 'gini', 'max_depth': 3, 'min_sam...</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>0.8025</td>\n",
              "      <td>0.783625</td>\n",
              "      <td>0.793</td>\n",
              "      <td>0.786</td>\n",
              "      <td>0.7795</td>\n",
              "      <td>0.7895</td>\n",
              "      <td>0.002076</td>\n",
              "      <td>0.000303</td>\n",
              "      <td>0.009618</td>\n",
              "      <td>0.002405</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.009462</td>\n",
              "      <td>0.000985</td>\n",
              "      <td>0.787</td>\n",
              "      <td>0.787525</td>\n",
              "      <td>gini</td>\n",
              "      <td>3</td>\n",
              "      <td>32</td>\n",
              "      <td>64</td>\n",
              "      <td>{'criterion': 'gini', 'max_depth': 3, 'min_sam...</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>0.8025</td>\n",
              "      <td>0.783625</td>\n",
              "      <td>0.793</td>\n",
              "      <td>0.786</td>\n",
              "      <td>0.7795</td>\n",
              "      <td>0.7895</td>\n",
              "      <td>0.000497</td>\n",
              "      <td>0.000153</td>\n",
              "      <td>0.009618</td>\n",
              "      <td>0.002405</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.009499</td>\n",
              "      <td>0.000977</td>\n",
              "      <td>0.787</td>\n",
              "      <td>0.787525</td>\n",
              "      <td>gini</td>\n",
              "      <td>3</td>\n",
              "      <td>32</td>\n",
              "      <td>128</td>\n",
              "      <td>{'criterion': 'gini', 'max_depth': 3, 'min_sam...</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>0.8025</td>\n",
              "      <td>0.783625</td>\n",
              "      <td>0.793</td>\n",
              "      <td>0.786</td>\n",
              "      <td>0.7795</td>\n",
              "      <td>0.7895</td>\n",
              "      <td>0.000431</td>\n",
              "      <td>0.000115</td>\n",
              "      <td>0.009618</td>\n",
              "      <td>0.002405</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.009158</td>\n",
              "      <td>0.000896</td>\n",
              "      <td>0.787</td>\n",
              "      <td>0.787525</td>\n",
              "      <td>gini</td>\n",
              "      <td>3</td>\n",
              "      <td>64</td>\n",
              "      <td>32</td>\n",
              "      <td>{'criterion': 'gini', 'max_depth': 3, 'min_sam...</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>0.8025</td>\n",
              "      <td>0.783625</td>\n",
              "      <td>0.793</td>\n",
              "      <td>0.786</td>\n",
              "      <td>0.7795</td>\n",
              "      <td>0.7895</td>\n",
              "      <td>0.000085</td>\n",
              "      <td>0.000055</td>\n",
              "      <td>0.009618</td>\n",
              "      <td>0.002405</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.009352</td>\n",
              "      <td>0.000919</td>\n",
              "      <td>0.787</td>\n",
              "      <td>0.787525</td>\n",
              "      <td>gini</td>\n",
              "      <td>3</td>\n",
              "      <td>64</td>\n",
              "      <td>64</td>\n",
              "      <td>{'criterion': 'gini', 'max_depth': 3, 'min_sam...</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>0.8025</td>\n",
              "      <td>0.783625</td>\n",
              "      <td>0.793</td>\n",
              "      <td>0.786</td>\n",
              "      <td>0.7795</td>\n",
              "      <td>0.7895</td>\n",
              "      <td>0.000151</td>\n",
              "      <td>0.000079</td>\n",
              "      <td>0.009618</td>\n",
              "      <td>0.002405</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 24 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "   mean_fit_time  mean_score_time  mean_test_score  mean_train_score  \\\n",
              "0       0.010791         0.001162            0.787          0.787525   \n",
              "1       0.009462         0.000985            0.787          0.787525   \n",
              "2       0.009499         0.000977            0.787          0.787525   \n",
              "3       0.009158         0.000896            0.787          0.787525   \n",
              "4       0.009352         0.000919            0.787          0.787525   \n",
              "\n",
              "  param_criterion param_max_depth param_min_samples_leaf  \\\n",
              "0            gini               3                     32   \n",
              "1            gini               3                     32   \n",
              "2            gini               3                     32   \n",
              "3            gini               3                     64   \n",
              "4            gini               3                     64   \n",
              "\n",
              "  param_min_samples_split                                             params  \\\n",
              "0                      32  {'criterion': 'gini', 'max_depth': 3, 'min_sam...   \n",
              "1                      64  {'criterion': 'gini', 'max_depth': 3, 'min_sam...   \n",
              "2                     128  {'criterion': 'gini', 'max_depth': 3, 'min_sam...   \n",
              "3                      32  {'criterion': 'gini', 'max_depth': 3, 'min_sam...   \n",
              "4                      64  {'criterion': 'gini', 'max_depth': 3, 'min_sam...   \n",
              "\n",
              "   rank_test_score       ...         split2_test_score  split2_train_score  \\\n",
              "0                1       ...                    0.8025            0.783625   \n",
              "1                1       ...                    0.8025            0.783625   \n",
              "2                1       ...                    0.8025            0.783625   \n",
              "3                1       ...                    0.8025            0.783625   \n",
              "4                1       ...                    0.8025            0.783625   \n",
              "\n",
              "   split3_test_score  split3_train_score  split4_test_score  \\\n",
              "0              0.793               0.786             0.7795   \n",
              "1              0.793               0.786             0.7795   \n",
              "2              0.793               0.786             0.7795   \n",
              "3              0.793               0.786             0.7795   \n",
              "4              0.793               0.786             0.7795   \n",
              "\n",
              "   split4_train_score  std_fit_time  std_score_time  std_test_score  \\\n",
              "0              0.7895      0.002076        0.000303        0.009618   \n",
              "1              0.7895      0.000497        0.000153        0.009618   \n",
              "2              0.7895      0.000431        0.000115        0.009618   \n",
              "3              0.7895      0.000085        0.000055        0.009618   \n",
              "4              0.7895      0.000151        0.000079        0.009618   \n",
              "\n",
              "   std_train_score  \n",
              "0         0.002405  \n",
              "1         0.002405  \n",
              "2         0.002405  \n",
              "3         0.002405  \n",
              "4         0.002405  \n",
              "\n",
              "[5 rows x 24 columns]"
            ]
          },
          "execution_count": 63,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "from sklearn.model_selection import GridSearchCV, KFold\n",
        "\n",
        "SEED=301\n",
        "np.random.seed(SEED)\n",
        "\n",
        "espaco_de_parametros = {\n",
        "    \"max_depth\" : [3, 5],\n",
        "    \"min_samples_split\" : [32, 64, 128],\n",
        "    \"min_samples_leaf\" : [32, 64, 128],\n",
        "    \"criterion\" : [\"gini\", \"entropy\"]\n",
        "}\n",
        "\n",
        "busca = GridSearchCV(DecisionTreeClassifier(),\n",
        "                    espaco_de_parametros,\n",
        "                    cv = KFold(n_splits = 5, shuffle=True))\n",
        "busca.fit(x_azar, y_azar)\n",
        "resultados = pd.DataFrame(busca.cv_results_)\n",
        "resultados.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "colab_type": "code",
        "id": "IHOi-91kulqx",
        "outputId": "8924bf96-8bc2-413b-ae11-2ce40000f072"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "array([0.783 , 0.781 , 0.79  , 0.781 , 0.7995])"
            ]
          },
          "execution_count": 65,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "from sklearn.model_selection import cross_val_score\n",
        "\n",
        "scores = cross_val_score(busca, x_azar, y_azar, cv = KFold(n_splits=5, shuffle=True))\n",
        "scores"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {},
        "colab_type": "code",
        "id": "Lc5EU1Uou07O"
      },
      "outputs": [],
      "source": [
        "def imprime_score(scores):\n",
        "  media = scores.mean() * 100\n",
        "  desvio = scores.std() * 100\n",
        "  print(\"Accuracy médio %.2f\" % media)\n",
        "  print(\"Intervalo [%.2f, %.2f]\" % (media - 2 * desvio, media + 2 * desvio))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "colab_type": "code",
        "id": "VDRWxxyevJiz",
        "outputId": "42022718-23a6-428b-a265-18107042a0f1"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Accuracy médio 78.69\n",
            "Intervalo [77.27, 80.11]\n"
          ]
        }
      ],
      "source": [
        "imprime_score(scores)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 119
        },
        "colab_type": "code",
        "id": "7fUVmolpvKco",
        "outputId": "4673d25f-b94a-4ba9-c67d-5cc211717a21"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=3,\n",
            "            max_features=None, max_leaf_nodes=None,\n",
            "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
            "            min_samples_leaf=32, min_samples_split=32,\n",
            "            min_weight_fraction_leaf=0.0, presort=False, random_state=None,\n",
            "            splitter='best')\n"
          ]
        }
      ],
      "source": [
        "melhor = busca.best_estimator_\n",
        "print(melhor)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 618
        },
        "colab_type": "code",
        "id": "eDBL3kBxvPPi",
        "outputId": "0c840ac7-4430-4450-ee9f-90488d10c2c7"
      },
      "outputs": [
        {
          "data": {
            "image/svg+xml": [
              "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
              "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
              " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
              "<!-- Generated by graphviz version 2.40.1 (20161225.0304)\n",
              " -->\n",
              "<!-- Title: Tree Pages: 1 -->\n",
              "<svg width=\"912pt\" height=\"433pt\"\n",
              " viewBox=\"0.00 0.00 912.00 433.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
              "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 429)\">\n",
              "<title>Tree</title>\n",
              "<polygon fill=\"#ffffff\" stroke=\"transparent\" points=\"-4,4 -4,-429 908,-429 908,4 -4,4\"/>\n",
              "<!-- 0 -->\n",
              "<g id=\"node1\" class=\"node\">\n",
              "<title>0</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.274510\" stroke=\"#000000\" d=\"M372.5,-425C372.5,-425 247.5,-425 247.5,-425 241.5,-425 235.5,-419 235.5,-413 235.5,-413 235.5,-354 235.5,-354 235.5,-348 241.5,-342 247.5,-342 247.5,-342 372.5,-342 372.5,-342 378.5,-342 384.5,-348 384.5,-354 384.5,-354 384.5,-413 384.5,-413 384.5,-419 378.5,-425 372.5,-425\"/>\n",
              "<text text-anchor=\"middle\" x=\"310\" y=\"-409.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">preco &lt;= 59982.246</text>\n",
              "<text text-anchor=\"middle\" x=\"310\" y=\"-394.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.487</text>\n",
              "<text text-anchor=\"middle\" x=\"310\" y=\"-379.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 10000</text>\n",
              "<text text-anchor=\"middle\" x=\"310\" y=\"-364.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [4200, 5800]</text>\n",
              "<text text-anchor=\"middle\" x=\"310\" y=\"-349.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 1 -->\n",
              "<g id=\"node2\" class=\"node\">\n",
              "<title>1</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.874510\" stroke=\"#000000\" d=\"M277.5,-306C277.5,-306 154.5,-306 154.5,-306 148.5,-306 142.5,-300 142.5,-294 142.5,-294 142.5,-235 142.5,-235 142.5,-229 148.5,-223 154.5,-223 154.5,-223 277.5,-223 277.5,-223 283.5,-223 289.5,-229 289.5,-235 289.5,-235 289.5,-294 289.5,-294 289.5,-300 283.5,-306 277.5,-306\"/>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-290.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">preco &lt;= 40083.383</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-275.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.198</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-260.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 4347</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [484, 3863]</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 0&#45;&gt;1 -->\n",
              "<g id=\"edge1\" class=\"edge\">\n",
              "<title>0&#45;&gt;1</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M277.1234,-341.8796C270.0803,-332.9633 262.5707,-323.4565 255.3126,-314.268\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"257.9621,-311.9757 249.017,-306.2981 252.4691,-316.3147 257.9621,-311.9757\"/>\n",
              "<text text-anchor=\"middle\" x=\"246.1027\" y=\"-327.4276\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">True</text>\n",
              "</g>\n",
              "<!-- 6 -->\n",
              "<g id=\"node7\" class=\"node\">\n",
              "<title>6</title>\n",
              "<path fill=\"#e58139\" fill-opacity=\"0.478431\" stroke=\"#000000\" d=\"M545,-306C545,-306 377,-306 377,-306 371,-306 365,-300 365,-294 365,-294 365,-235 365,-235 365,-229 371,-223 377,-223 377,-223 545,-223 545,-223 551,-223 557,-229 557,-235 557,-235 557,-294 557,-294 557,-300 551,-306 545,-306\"/>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-290.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">km_por_ano &lt;= 24124.006</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-275.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.45</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-260.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 5653</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-245.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [3716, 1937]</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-230.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = não</text>\n",
              "</g>\n",
              "<!-- 0&#45;&gt;6 -->\n",
              "<g id=\"edge6\" class=\"edge\">\n",
              "<title>0&#45;&gt;6</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M362.8125,-341.8796C374.8818,-332.368 387.8039,-322.1843 400.1788,-312.432\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"402.5296,-315.0356 408.2173,-306.0969 398.1968,-309.5377 402.5296,-315.0356\"/>\n",
              "<text text-anchor=\"middle\" x=\"405.275\" y=\"-327.2232\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">False</text>\n",
              "</g>\n",
              "<!-- 2 -->\n",
              "<g id=\"node3\" class=\"node\">\n",
              "<title>2</title>\n",
              "<path fill=\"#399de5\" stroke=\"#000000\" d=\"M112,-179.5C112,-179.5 12,-179.5 12,-179.5 6,-179.5 0,-173.5 0,-167.5 0,-167.5 0,-123.5 0,-123.5 0,-117.5 6,-111.5 12,-111.5 12,-111.5 112,-111.5 112,-111.5 118,-111.5 124,-117.5 124,-123.5 124,-123.5 124,-167.5 124,-167.5 124,-173.5 118,-179.5 112,-179.5\"/>\n",
              "<text text-anchor=\"middle\" x=\"62\" y=\"-164.3\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.0</text>\n",
              "<text text-anchor=\"middle\" x=\"62\" y=\"-149.3\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 1935</text>\n",
              "<text text-anchor=\"middle\" x=\"62\" y=\"-134.3\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [0, 1935]</text>\n",
              "<text text-anchor=\"middle\" x=\"62\" y=\"-119.3\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 1&#45;&gt;2 -->\n",
              "<g id=\"edge2\" class=\"edge\">\n",
              "<title>1&#45;&gt;2</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M162.1383,-222.8796C146.7696,-211.0038 130.0453,-198.0804 114.7441,-186.2568\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"116.4605,-183.1599 106.4076,-179.8149 112.1804,-188.699 116.4605,-183.1599\"/>\n",
              "</g>\n",
              "<!-- 3 -->\n",
              "<g id=\"node4\" class=\"node\">\n",
              "<title>3</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.749020\" stroke=\"#000000\" d=\"M277.5,-187C277.5,-187 154.5,-187 154.5,-187 148.5,-187 142.5,-181 142.5,-175 142.5,-175 142.5,-116 142.5,-116 142.5,-110 148.5,-104 154.5,-104 154.5,-104 277.5,-104 277.5,-104 283.5,-104 289.5,-110 289.5,-116 289.5,-116 289.5,-175 289.5,-175 289.5,-181 283.5,-187 277.5,-187\"/>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">preco &lt;= 41083.031</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.321</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 2412</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [484, 1928]</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 1&#45;&gt;3 -->\n",
              "<g id=\"edge3\" class=\"edge\">\n",
              "<title>1&#45;&gt;3</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M216,-222.8796C216,-214.6838 216,-205.9891 216,-197.5013\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"219.5001,-197.298 216,-187.2981 212.5001,-197.2981 219.5001,-197.298\"/>\n",
              "</g>\n",
              "<!-- 4 -->\n",
              "<g id=\"node5\" class=\"node\">\n",
              "<title>4</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.533333\" stroke=\"#000000\" d=\"M116,-68C116,-68 24,-68 24,-68 18,-68 12,-62 12,-56 12,-56 12,-12 12,-12 12,-6 18,0 24,0 24,0 116,0 116,0 122,0 128,-6 128,-12 128,-12 128,-56 128,-56 128,-62 122,-68 116,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"70\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.434</text>\n",
              "<text text-anchor=\"middle\" x=\"70\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 110</text>\n",
              "<text text-anchor=\"middle\" x=\"70\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [35, 75]</text>\n",
              "<text text-anchor=\"middle\" x=\"70\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 3&#45;&gt;4 -->\n",
              "<g id=\"edge4\" class=\"edge\">\n",
              "<title>3&#45;&gt;4</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M161.635,-103.9815C148.9989,-94.3313 135.5644,-84.0714 122.9885,-74.4673\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"124.9372,-71.5515 114.8654,-68.2637 120.6886,-77.1148 124.9372,-71.5515\"/>\n",
              "</g>\n",
              "<!-- 5 -->\n",
              "<g id=\"node6\" class=\"node\">\n",
              "<title>5</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.756863\" stroke=\"#000000\" d=\"M274,-68C274,-68 158,-68 158,-68 152,-68 146,-62 146,-56 146,-56 146,-12 146,-12 146,-6 152,0 158,0 158,0 274,0 274,0 280,0 286,-6 286,-12 286,-12 286,-56 286,-56 286,-62 280,-68 274,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.314</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 2302</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [449, 1853]</text>\n",
              "<text text-anchor=\"middle\" x=\"216\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 3&#45;&gt;5 -->\n",
              "<g id=\"edge5\" class=\"edge\">\n",
              "<title>3&#45;&gt;5</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M216,-103.9815C216,-95.618 216,-86.7965 216,-78.3409\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"219.5001,-78.2636 216,-68.2637 212.5001,-78.2637 219.5001,-78.2636\"/>\n",
              "</g>\n",
              "<!-- 7 -->\n",
              "<g id=\"node8\" class=\"node\">\n",
              "<title>7</title>\n",
              "<path fill=\"#e58139\" fill-opacity=\"0.129412\" stroke=\"#000000\" d=\"M523.5,-187C523.5,-187 398.5,-187 398.5,-187 392.5,-187 386.5,-181 386.5,-175 386.5,-175 386.5,-116 386.5,-116 386.5,-110 392.5,-104 398.5,-104 398.5,-104 523.5,-104 523.5,-104 529.5,-104 535.5,-110 535.5,-116 535.5,-116 535.5,-175 535.5,-175 535.5,-181 529.5,-187 523.5,-187\"/>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">preco &lt;= 99974.406</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.498</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 3513</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [1879, 1634]</text>\n",
              "<text text-anchor=\"middle\" x=\"461\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = não</text>\n",
              "</g>\n",
              "<!-- 6&#45;&gt;7 -->\n",
              "<g id=\"edge7\" class=\"edge\">\n",
              "<title>6&#45;&gt;7</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M461,-222.8796C461,-214.6838 461,-205.9891 461,-197.5013\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"464.5001,-197.298 461,-187.2981 457.5001,-197.2981 464.5001,-197.298\"/>\n",
              "</g>\n",
              "<!-- 10 -->\n",
              "<g id=\"node11\" class=\"node\">\n",
              "<title>10</title>\n",
              "<path fill=\"#e58139\" fill-opacity=\"0.835294\" stroke=\"#000000\" d=\"M753.5,-187C753.5,-187 622.5,-187 622.5,-187 616.5,-187 610.5,-181 610.5,-175 610.5,-175 610.5,-116 610.5,-116 610.5,-110 616.5,-104 622.5,-104 622.5,-104 753.5,-104 753.5,-104 759.5,-104 765.5,-110 765.5,-116 765.5,-116 765.5,-175 765.5,-175 765.5,-181 759.5,-187 753.5,-187\"/>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-171.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">preco &lt;= 100076.953</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-156.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.243</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-141.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 2140</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-126.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [1837, 303]</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-111.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = não</text>\n",
              "</g>\n",
              "<!-- 6&#45;&gt;10 -->\n",
              "<g id=\"edge10\" class=\"edge\">\n",
              "<title>6&#45;&gt;10</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M540.3936,-222.8796C560.0179,-212.592 581.1418,-201.5182 601.1025,-191.0542\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"602.9504,-194.0373 610.1822,-186.2944 599.7003,-187.8375 602.9504,-194.0373\"/>\n",
              "</g>\n",
              "<!-- 8 -->\n",
              "<g id=\"node9\" class=\"node\">\n",
              "<title>8</title>\n",
              "<path fill=\"#e58139\" fill-opacity=\"0.250980\" stroke=\"#000000\" d=\"M441.5,-68C441.5,-68 316.5,-68 316.5,-68 310.5,-68 304.5,-62 304.5,-56 304.5,-56 304.5,-12 304.5,-12 304.5,-6 310.5,0 316.5,0 316.5,0 441.5,0 441.5,0 447.5,0 453.5,-6 453.5,-12 453.5,-12 453.5,-56 453.5,-56 453.5,-62 447.5,-68 441.5,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"379\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.49</text>\n",
              "<text text-anchor=\"middle\" x=\"379\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 2944</text>\n",
              "<text text-anchor=\"middle\" x=\"379\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [1682, 1262]</text>\n",
              "<text text-anchor=\"middle\" x=\"379\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = não</text>\n",
              "</g>\n",
              "<!-- 7&#45;&gt;8 -->\n",
              "<g id=\"edge8\" class=\"edge\">\n",
              "<title>7&#45;&gt;8</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M430.4662,-103.9815C423.8424,-94.9747 416.8279,-85.4367 410.1795,-76.3965\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"412.9426,-74.2461 404.1984,-68.2637 407.3034,-78.3933 412.9426,-74.2461\"/>\n",
              "</g>\n",
              "<!-- 9 -->\n",
              "<g id=\"node10\" class=\"node\">\n",
              "<title>9</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.470588\" stroke=\"#000000\" d=\"M592,-68C592,-68 484,-68 484,-68 478,-68 472,-62 472,-56 472,-56 472,-12 472,-12 472,-6 478,0 484,0 484,0 592,0 592,0 598,0 604,-6 604,-12 604,-12 604,-56 604,-56 604,-62 598,-68 592,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"538\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.453</text>\n",
              "<text text-anchor=\"middle\" x=\"538\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 569</text>\n",
              "<text text-anchor=\"middle\" x=\"538\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [197, 372]</text>\n",
              "<text text-anchor=\"middle\" x=\"538\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 7&#45;&gt;9 -->\n",
              "<g id=\"edge9\" class=\"edge\">\n",
              "<title>7&#45;&gt;9</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M489.672,-103.9815C495.8285,-95.0666 502.3443,-85.6313 508.5305,-76.6734\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"511.5355,-78.4811 514.3381,-68.2637 505.7755,-74.5034 511.5355,-78.4811\"/>\n",
              "</g>\n",
              "<!-- 11 -->\n",
              "<g id=\"node12\" class=\"node\">\n",
              "<title>11</title>\n",
              "<path fill=\"#e58139\" fill-opacity=\"0.956863\" stroke=\"#000000\" d=\"M742,-68C742,-68 634,-68 634,-68 628,-68 622,-62 622,-56 622,-56 622,-12 622,-12 622,-6 628,0 634,0 634,0 742,0 742,0 748,0 754,-6 754,-12 754,-12 754,-56 754,-56 754,-62 748,-68 742,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.082</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 1809</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [1732, 77]</text>\n",
              "<text text-anchor=\"middle\" x=\"688\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = não</text>\n",
              "</g>\n",
              "<!-- 10&#45;&gt;11 -->\n",
              "<g id=\"edge11\" class=\"edge\">\n",
              "<title>10&#45;&gt;11</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M688,-103.9815C688,-95.618 688,-86.7965 688,-78.3409\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"691.5001,-78.2636 688,-68.2637 684.5001,-78.2637 691.5001,-78.2636\"/>\n",
              "</g>\n",
              "<!-- 12 -->\n",
              "<g id=\"node13\" class=\"node\">\n",
              "<title>12</title>\n",
              "<path fill=\"#399de5\" fill-opacity=\"0.537255\" stroke=\"#000000\" d=\"M892,-68C892,-68 784,-68 784,-68 778,-68 772,-62 772,-56 772,-56 772,-12 772,-12 772,-6 778,0 784,0 784,0 892,0 892,0 898,0 904,-6 904,-12 904,-12 904,-56 904,-56 904,-62 898,-68 892,-68\"/>\n",
              "<text text-anchor=\"middle\" x=\"838\" y=\"-52.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">gini = 0.433</text>\n",
              "<text text-anchor=\"middle\" x=\"838\" y=\"-37.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">samples = 331</text>\n",
              "<text text-anchor=\"middle\" x=\"838\" y=\"-22.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">value = [105, 226]</text>\n",
              "<text text-anchor=\"middle\" x=\"838\" y=\"-7.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\" fill=\"#000000\">class = sim</text>\n",
              "</g>\n",
              "<!-- 10&#45;&gt;12 -->\n",
              "<g id=\"edge12\" class=\"edge\">\n",
              "<title>10&#45;&gt;12</title>\n",
              "<path fill=\"none\" stroke=\"#000000\" d=\"M743.8545,-103.9815C756.8368,-94.3313 770.6394,-84.0714 783.5597,-74.4673\"/>\n",
              "<polygon fill=\"#000000\" stroke=\"#000000\" points=\"785.9678,-77.0384 791.9054,-68.2637 781.7917,-71.4204 785.9678,-77.0384\"/>\n",
              "</g>\n",
              "</g>\n",
              "</svg>\n"
            ],
            "text/plain": [
              "<graphviz.files.Source at 0x7f28fa10d3c8>"
            ]
          },
          "execution_count": 69,
          "metadata": {
            "tags": []
          },
          "output_type": "execute_result"
        }
      ],
      "source": [
        "from sklearn.tree import export_graphviz\n",
        "import graphviz\n",
        "\n",
        "features = x_azar.columns\n",
        "dot_data = export_graphviz(melhor, out_file=None, filled=True, rounded=True,\n",
        "                          class_names=[\"não\",\"sim\"],\n",
        "                          feature_names = features)\n",
        "graph = graphviz.Source(dot_data)\n",
        "graph"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {},
        "colab_type": "code",
        "id": "JiW754atvZ8N"
      },
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [],
      "name": "Introdução_a_Machine_Learning_Otimização.ipynb",
      "provenance": [],
      "version": "0.3.2"
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}