gpboost_model.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/samuel-haddad/MBAUSP/blob/main/gpboost_model.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VPQRN6RISNBZ"
      },
      "source": [
        "# **Modelo Multinível**\n",
        "\n",
        "#### MBA Data Science & Analytics - USP & ESALQ\n",
        "Trabalho de Conclusão de Curso - 02/2022 <br>\n",
        "**Título:** Um estudo comparativo entre modelos Multinível, Logístico e XGBoost na detecção de fraudes ideológicas no setor financeiro. <br>\n",
        "**Autor:** Samuel Haddad Simões Machado\n",
        "<br>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7rZkhfNPrJk5"
      },
      "source": [
        "## Bibliotecas"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%%capture\n",
        "# instalar as bibliotecas\n",
        "!pip install gpboost"
      ],
      "metadata": {
        "id": "XbqkbzHNySKB"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "G-u5MiPtpZ2N"
      },
      "outputs": [],
      "source": [
        "# pacotes básicos\n",
        "import pandas as pd \n",
        "import numpy as np \n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# preprocessamento\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.model_selection import cross_val_score\n",
        "\n",
        "# modelagem\n",
        "from sklearn import metrics\n",
        "from sklearn.metrics import f1_score\n",
        "from sklearn.metrics import roc_auc_score\n",
        "from sklearn.metrics import log_loss\n",
        "from sklearn.metrics import confusion_matrix\n",
        "from sklearn.metrics import roc_curve\n",
        "from numpy import sqrt\n",
        "from numpy import argmax\n",
        "from sklearn.metrics import precision_recall_curve\n",
        "import gpboost as gpb\n",
        "from gpboost import GPBoostClassifier\n",
        "\n",
        "# bibliotecas google & auth\n",
        "from pydrive.auth import GoogleAuth\n",
        "from pydrive.drive import GoogleDrive\n",
        "from google.colab import auth\n",
        "from oauth2client.client import GoogleCredentials\n",
        "\n",
        "# autenticar e criar o cliente com PyDrive\n",
        "auth.authenticate_user()\n",
        "gauth = GoogleAuth()\n",
        "gauth.credentials = GoogleCredentials.get_application_default()\n",
        "drive = GoogleDrive(gauth)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sgaRIhgopZB1"
      },
      "source": [
        "## Dataset"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# link do arquivo no Drive\n",
        "link = 'https://drive.google.com/file/d/1vtTBtunb8fqbr-b6AAkEZwNFr0QdDcqX/view?usp=sharing'\n",
        " \n",
        "# extraindo o ID do link\n",
        "id = link.split(\"/\")[-2]\n",
        " \n",
        "downloaded = drive.CreateFile({'id':id})\n",
        "downloaded.GetContentFile('base_tcc_xgboost_group20220927.csv') \n",
        " \n",
        "# criando o dataframe \n",
        "df = pd.read_csv('base_tcc_xgboost_group20220927.csv', sep=';')\n",
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 236
        },
        "id": "9zE0MTgc6b-B",
        "outputId": "8fbec3f8-5376-4a8e-863f-0585024f6273"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   target      var2      var3  var4  var5  var6  var7  var8  var9  var10  ...  \\\n",
              "0       0  0.125744 -0.204706     0     0     0     0     0     0      0  ...   \n",
              "1       1 -0.194939  0.209826     0     0     0     0     0     0      0  ...   \n",
              "2       0  0.067438  1.079848     0     0     0     0     0     0      0  ...   \n",
              "3       0  0.424562 -0.454262     0     0     0     0     0     0      0  ...   \n",
              "4       0  0.424562 -0.211115     0     0     0     0     0     1      0  ...   \n",
              "\n",
              "   var36  var37  var38  var39  var40  var41  var42  var43  group  id  \n",
              "0      1      1      0      1      0      0      0      1    192   2  \n",
              "1      0      1      0      0      0      1      0      1    279  10  \n",
              "2      0      0      1      1      0      0      0      0    198  26  \n",
              "3      0      1      0      1      0      1      1      0    200  48  \n",
              "4      0      1      0      1      0      1      0      0    223  59  \n",
              "\n",
              "[5 rows x 45 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-6203655c-b5b4-4133-a2a8-a52fe559aa59\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>target</th>\n",
              "      <th>var2</th>\n",
              "      <th>var3</th>\n",
              "      <th>var4</th>\n",
              "      <th>var5</th>\n",
              "      <th>var6</th>\n",
              "      <th>var7</th>\n",
              "      <th>var8</th>\n",
              "      <th>var9</th>\n",
              "      <th>var10</th>\n",
              "      <th>...</th>\n",
              "      <th>var36</th>\n",
              "      <th>var37</th>\n",
              "      <th>var38</th>\n",
              "      <th>var39</th>\n",
              "      <th>var40</th>\n",
              "      <th>var41</th>\n",
              "      <th>var42</th>\n",
              "      <th>var43</th>\n",
              "      <th>group</th>\n",
              "      <th>id</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>0.125744</td>\n",
              "      <td>-0.204706</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>192</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>-0.194939</td>\n",
              "      <td>0.209826</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>279</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>0.067438</td>\n",
              "      <td>1.079848</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>198</td>\n",
              "      <td>26</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>0.424562</td>\n",
              "      <td>-0.454262</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>200</td>\n",
              "      <td>48</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>0.424562</td>\n",
              "      <td>-0.211115</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>223</td>\n",
              "      <td>59</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 45 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6203655c-b5b4-4133-a2a8-a52fe559aa59')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-6203655c-b5b4-4133-a2a8-a52fe559aa59 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-6203655c-b5b4-4133-a2a8-a52fe559aa59');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xlvCNVdKrRJ6"
      },
      "source": [
        "## Treino, teste e validação"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NyUtscEOauuN",
        "outputId": "27a0c16f-bf43-4c7e-bc6b-2ca01752dd8d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "+ ------------------------- RESUMO ------------------------- +\n",
            "X_train: (15860, 42) | y_train: 15860 | m_train: 38        \n",
            "X_test:  (5287, 42) | y_test: 5287  | m_test: 35        \n",
            "X_val:   (5287, 42) | y_val: 5287 | m_val: 38\n",
            "Train dist: 35.34% | Test dist: 35.2%\n",
            "+ ---------------------------------------------------------- +\n"
          ]
        }
      ],
      "source": [
        "# remover a coluna de id\n",
        "df_silver = df.drop(['id'], axis=1)\n",
        "\n",
        "# extrair as variáveis explicativas e resposta\n",
        "X, y, group = df_silver.drop(['group', 'target'], axis=1), df_silver['target'], df_silver['group']\n",
        "\n",
        "# separar a base de treino\n",
        "X_train, X_rem, y_train, y_rem, group_train, group_rem = train_test_split(X, y, group, train_size=0.6, random_state=42)\n",
        " \n",
        "# separar as bases de teste e validação\n",
        "X_val, X_test, y_val, y_test, group_val, group_test = train_test_split(X_rem, y_rem, group_rem, test_size=0.5, random_state=42)\n",
        "\n",
        "# distribuição da variável resposta nas bases\n",
        "y_train_dist = y_train.sum()/len(y_train)\n",
        "y_test_dist = y_test.sum()/len(y_test)\n",
        "\n",
        "# contagem dos grupos\n",
        "m_train = len(np.unique(group_train))\n",
        "m_val = len(np.unique(group_val))\n",
        "m_test = len(np.unique(group_test))\n",
        "\n",
        "\n",
        "# exibir a volumetria das bases\n",
        "print('+', 25*'-','RESUMO', 25*'-','+')\n",
        "print(f'X_train: {X_train.shape} | y_train: {len(y_train)} | m_train: {m_train}\\\n",
        "        \\nX_test:  {X_test.shape} | y_test: {len(y_test)}  | m_test: {m_test}\\\n",
        "        \\nX_val:   {X_val.shape} | y_val: {len(y_val)} | m_val: {m_val}'\n",
        "        )\n",
        "print(f'Train dist: {round(y_train_dist*100, 2)}% | Test dist: {round(y_test_dist*100, 2)}%')\n",
        "print('+', 58*'-', '+')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-FnS4IierTkY"
      },
      "source": [
        "## Modelagem"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "G_XKlDmqjG11"
      },
      "outputs": [],
      "source": [
        "%%capture\n",
        "# parâmetros iniciais \n",
        "likelihood = \"bernoulli_probit\"\n",
        "num_boost_round = 500\n",
        "\n",
        "#---------- Treinamento  ----------#\n",
        "\n",
        "# Criação do modelo multinível\n",
        "gp_model = gpb.GPModel(group_data=group_train, likelihood=likelihood)\n",
        "\n",
        "# Detalhamento das etapas para monitoramento da convergência do modelo passo a passo\n",
        "gp_model.set_optim_params(params={\"trace\": False})\n",
        "\n",
        "# Criar o dataset de treinamento\n",
        "data_train = gpb.Dataset(X_train, y_train)\n",
        "\n",
        "# Especificar o parâmetroa iniciais de boosting\n",
        "params = {'learning_rate': 0.1\n",
        "          ,'min_data_in_leaf': 1\n",
        "          ,'objective': 'binary'\n",
        "          }\n",
        "\n",
        "# treinando o modelo com os parâmtros de boosting\n",
        "bst = gpb.train(params=params\n",
        "                ,train_set=data_train\n",
        "                ,gp_model=gp_model\n",
        "                ,num_boost_round=num_boost_round\n",
        "                )\n",
        "\n",
        "# GPB Classifier\n",
        "model = gpb.GPBoostClassifier(objective='binary')\n",
        "\n",
        "# init_model=bst\n",
        "model.fit(X=X_train, y=y_train, gp_model=gp_model, init_model=bst)\n",
        "\n",
        "# previsão - teste\n",
        "y_pred_test = model.predict(X_test,  group_data_pred=group_test)\n",
        "\n",
        "# extrair as probabilidades\n",
        "yhat = model.predict_proba(X_test, group_data_pred=group_test)['response_mean']\n",
        "\n",
        "# extrair as métricas - test\n",
        "acuracy_test = (metrics.accuracy_score(y_test, y_pred_test)*100).round(2)\n",
        "precision_test = (metrics.precision_score(y_test, y_pred_test)*100).round(2)\n",
        "recall_test = (metrics.recall_score(y_test, y_pred_test)*100).round(2)\n",
        "aucroc_test = (roc_auc_score(y_test, yhat)*100).round(2)\n",
        "f1_score_test = (f1_score(y_test, y_pred_test)*100).round(2)\n",
        "logloss_test = (metrics.log_loss(y_test, yhat, normalize=True))\n",
        "\n",
        "#---------- OTIMIZAÇÃO ----------#\n",
        "\n",
        "# Escolhendo os parâmetros de otimização\n",
        "param_grid = {'learning_rate': [1,0.1,0.01]\n",
        "              ,'min_data_in_leaf': [1,10,100]\n",
        "              ,'max_depth': [1,3,5,10,-1]\n",
        "              }\n",
        "\n",
        "opt_params = gpb.grid_search_tune_parameters(param_grid=param_grid\n",
        "                                             ,params=params\n",
        "                                             ,num_try_random=None\n",
        "                                             ,nfold=4\n",
        "                                             ,gp_model=gp_model\n",
        "                                             ,use_gp_model_for_validation=True\n",
        "                                             ,train_set=data_train\n",
        "                                             ,verbose_eval=0\n",
        "                                             ,num_boost_round=1000\n",
        "                                             ,early_stopping_rounds=10\n",
        "                                             ,seed=1000\n",
        "                                             ,metrics='auc'\n",
        "                                             )\n",
        "\n",
        "\n",
        "params = {**opt_params['best_params'], 'objective': 'binary'}\n",
        "\n",
        "# Cross-validation para determinar o número de iterações\n",
        "cvbst = gpb.cv(params=params\n",
        "               ,train_set=data_train\n",
        "               ,gp_model=gp_model\n",
        "               ,use_gp_model_for_validation=True\n",
        "               ,num_boost_round=1000\n",
        "               ,nfold=4\n",
        "               ,verbose_eval=False\n",
        "               ,show_stdv=False\n",
        "               ,seed=1\n",
        "               )\n",
        "metric_name = list(cvbst.keys())[0]\n",
        "num_iteration = np.argmin(cvbst[metric_name])"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"Best parameters: \" + str(opt_params['best_params']))\n",
        "print(\"Best number of iterations(grid search): \" + str(opt_params['best_iter']))\n",
        "print(\"Best score: \" + str(opt_params['best_score']))\n",
        "print(\"Best number of iterations (cv): \" + str(np.argmin(cvbst[metric_name])))"
      ],
      "metadata": {
        "id": "oFR15T95lJBj",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ad033554-dd4d-457d-cbff-33f0f9b0b06a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Best parameters: {'learning_rate': 0.1, 'min_data_in_leaf': 1, 'max_depth': 3}\n",
            "Best number of iterations(grid search): 250\n",
            "Best score: 0.8106621551738119\n",
            "Best number of iterations (cv): 255\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#---------- VALIDAÇÃO ----------#\n",
        "\n",
        "# construir o modelo com parâmetros otimizados\n",
        "\n",
        "# gpb Classifier\n",
        "model = gpb.GPBoostClassifier(**opt_params['best_params'], objective='binary')\n",
        "\n",
        "# init_model=bst\n",
        "model.fit(X=X_train, y=y_train, gp_model=gp_model, init_model=bst)\n",
        "\n",
        "# previsão - validação\n",
        "y_pred = model.predict(X_val, num_iteration=num_iteration, group_data_pred=group_val)\n",
        "\n",
        "# extrair as probabilidades\n",
        "yhat = model.predict_proba(X_val, num_iteration=num_iteration, group_data_pred=group_val)['response_mean']\n",
        "\n",
        "# extrair as métricas - validação\n",
        "acuracy_val = (metrics.accuracy_score(y_val, y_pred)*100).round(2)\n",
        "precision_val = (metrics.precision_score(y_val, y_pred)*100).round(2)\n",
        "recall_val = (metrics.recall_score(y_val, y_pred)*100).round(2)\n",
        "aucroc_val = (roc_auc_score(y_val, yhat)*100).round(2)\n",
        "f1_score_val = (f1_score(y_val, y_pred)*100).round(2)\n",
        "logloss_val = (metrics.log_loss(y_val, yhat, normalize=True))\n",
        "\n",
        "# criar o dataframe comparativo\n",
        "columns= ['amostra', 'acuracia', 'precisao', 'recall', 'auc_roc', 'f1_score', 'logloss(normalizada)']\n",
        "data = [('teste', acuracy_test, precision_test, recall_test, aucroc_test, f1_score_test, logloss_test)\n",
        "        ,('validacao', acuracy_val, precision_val, recall_val, aucroc_val, f1_score_val, logloss_val)\n",
        "        ]\n",
        "df_metrics = pd.DataFrame(data=data, columns=columns)\n",
        "\n",
        "# parâmetro aleatório (covariância)\n",
        "print(gp_model.summary()) \n",
        "\n",
        "# visualizar resultados do grid search\n",
        "print('+', 18*'-','RESULTADO GRID SEARCH', 18*'-','+')\n",
        "print('Parâmetros: \\n', model.get_params())\n",
        "print(\"Melhor número de iterações (cv): \" + str(np.argmin(cvbst[metric_name])))\n",
        "print('+', 59*'-', '+')\n",
        "\n",
        "# visualizar tabela comparativa\n",
        "df_metrics"
      ],
      "metadata": {
        "id": "nD_Z7rDGl0HN",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 347
        },
        "outputId": "027f2857-7f76-4aa0-c973-399c8779f7e8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[GPBoost] [Warning] min_data_in_leaf is set=1, min_child_samples=20 will be ignored. Current value: min_data_in_leaf=1\n",
            "===================================================\n",
            "Covariance parameters (random effects):\n",
            "       Param.\n",
            "group  0.0607\n",
            "===================================================\n",
            "<gpboost.basic.GPModel object at 0x7f49adbab450>\n",
            "+ ------------------ RESULTADO GRID SEARCH ------------------ +\n",
            "Parâmetros: \n",
            " {'boosting_type': 'gbdt', 'class_weight': None, 'colsample_bytree': 1.0, 'importance_type': 'split', 'learning_rate': 0.1, 'max_depth': 3, 'min_child_samples': 20, 'min_child_weight': 0.001, 'min_split_gain': 0.0, 'n_estimators': 100, 'n_jobs': -1, 'num_leaves': 31, 'objective': 'binary', 'random_state': None, 'reg_alpha': 0.0, 'reg_lambda': 0.0, 'silent': True, 'subsample': 1.0, 'subsample_for_bin': 200000, 'subsample_freq': 0, 'min_data_in_leaf': 1}\n",
            "Melhor número de iterações (cv): 255\n",
            "+ ----------------------------------------------------------- +\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     amostra  acuracia  precisao  recall  auc_roc  f1_score  \\\n",
              "0      teste     76.30     71.11   55.02    80.27     62.04   \n",
              "1  validacao     78.19     73.48   60.00    83.03     66.06   \n",
              "\n",
              "   logloss(normalizada)  \n",
              "0              0.505169  \n",
              "1              0.473676  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-6e84391d-6cf7-42b3-a70f-5fd934c22a33\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>amostra</th>\n",
              "      <th>acuracia</th>\n",
              "      <th>precisao</th>\n",
              "      <th>recall</th>\n",
              "      <th>auc_roc</th>\n",
              "      <th>f1_score</th>\n",
              "      <th>logloss(normalizada)</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>teste</td>\n",
              "      <td>76.30</td>\n",
              "      <td>71.11</td>\n",
              "      <td>55.02</td>\n",
              "      <td>80.27</td>\n",
              "      <td>62.04</td>\n",
              "      <td>0.505169</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>validacao</td>\n",
              "      <td>78.19</td>\n",
              "      <td>73.48</td>\n",
              "      <td>60.00</td>\n",
              "      <td>83.03</td>\n",
              "      <td>66.06</td>\n",
              "      <td>0.473676</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6e84391d-6cf7-42b3-a70f-5fd934c22a33')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-6e84391d-6cf7-42b3-a70f-5fd934c22a33 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-6e84391d-6cf7-42b3-a70f-5fd934c22a33');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dSBd5XIChrSL"
      },
      "source": [
        "## Gráficos"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fig, axes = plt.subplots(1, 3, figsize=(25, 8), gridspec_kw=({'hspace':0.5}))\n",
        "fig.suptitle('GPBoost', fontsize=18, weight='semibold', y=0.98)\n",
        "pad = 15\n",
        "\n",
        "#---------- MATRIZ DE CONFUSÃO ----------#\n",
        "cnf_matrix = confusion_matrix(y_val, y_pred)\n",
        "plt.subplot(1, 3, 1)\n",
        "\n",
        "labels = [0, 1]\n",
        "tick_marks = np.arange(len(labels))\n",
        "plt.xticks(tick_marks, labels)\n",
        "plt.yticks(tick_marks, labels)\n",
        "\n",
        "group_names = ['True Neg','False Pos','False Neg','True Pos']\n",
        "group_counts = ['{0:0.0f}'.format(value) for value in\n",
        "                cnf_matrix.flatten()]\n",
        "group_percentages = ['{0:.2%}'.format(value) for value in\n",
        "                     cnf_matrix.flatten()/np.sum(cnf_matrix)]\n",
        "labels = [f'{v1}\\n{v2}\\n{v3}' for v1, v2, v3 in\n",
        "          zip(group_names,group_counts,group_percentages)]\n",
        "labels = np.asarray(labels).reshape(2,2)\n",
        "\n",
        "# criar o heatmap\n",
        "ax = sns.heatmap(data=pd.DataFrame(cnf_matrix), annot=labels, cmap=\"viridis\", fmt='')\n",
        "ax.xaxis.set_label_position(\"bottom\")\n",
        "ax.set_title('Matrix de confusão')\n",
        "plt.ylabel('True')\n",
        "plt.xlabel('Predicted');\n",
        "\n",
        "#---------- AUC ROC ----------#\n",
        "fpr, tpr, thresholds = roc_curve(y_val, yhat)\n",
        "\n",
        "# plot da curva roc\n",
        "plt.subplot(1, 3, 2)\n",
        "plt.plot([0,1], [0,1], linestyle='--', label='No Skill', color='#fde725')\n",
        "plt.plot(fpr, tpr,  color='#4ac16d', marker='.', label='XGBoost (area = {}%)'.format(aucroc_val))\n",
        "\n",
        "# axis labels\n",
        "plt.title('AUC ROC', pad=pad)\n",
        "plt.xlabel('False Positive Rate')\n",
        "plt.ylabel('True Positive Rate')\n",
        "plt.legend()\n",
        "\n",
        "#---------- PRECISÃO, RECALL X THRESHOLD ----------#\n",
        "precision, recall, thresholds = precision_recall_curve(y_val, yhat) \n",
        "\n",
        "plt.subplot(1, 3, 3)\n",
        "plt.title('Precisão, Recall x Threshold', fontsize=12, pad=pad)\n",
        "plt.plot(thresholds, precision[: -1], \"b--\", label=\"Precision\", color='#5ec962')\n",
        "plt.plot(thresholds, recall[: -1], \"r--\", label=\"Recall\", color='#440154')\n",
        "plt.ylabel(\"Precision, Recall\")\n",
        "plt.xlabel(\"Threshold\")\n",
        "plt.legend(loc=\"lower left\")\n",
        "plt.ylim([0,1]);"
      ],
      "metadata": {
        "id": "hc7TDX_Bxeh1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 554
        },
        "outputId": "e1705b63-fcfd-4a41-f9f9-4ea6321d1f2d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1800x576 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "U34IdRQFnRJ2"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [],
      "provenance": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyNUjvFQ/Yr6jQ+4KFa+7DPB",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "gpuClass": "standard"
  },
  "nbformat": 4,
  "nbformat_minor": 0
}