Logistic_Regression.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/SampMark/Machine-Learn/blob/main/Logistic_Regression.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HoHOcw5wcrSE"
      },
      "source": [
        "# **Modelo de Regressão Logística**\n",
        "\n",
        "### **O que é uma regressão logística?**\n",
        "\n",
        "A regressão logística é um modelo estatístico que, em essência, modela a relação entre uma ou mais variáveis independentes para prever um resultado de uma variável de saída/dependente categórica, isto é, que assume apenas dois valores, tais como: \"sim\" ou \"não\", \"verdadeiro\" ou \"falso\", \"0\" ou \"1\".\n",
        "\n",
        "Em outras palavras, a regressão logística é usada para classificar registros de um conjunto de dados com base nos valores dos campos de entrada. Tem como finalidade para prever a probabilidade de uma variável dependente binária, podendo ser usada tanto para classificação binária, quanto para classificação de múltiplas classes.\n",
        "\n",
        "### **Quando é interessante aplicar o Modelo de Regressão Logística?**\n",
        "\n",
        "A seguir quatro modalidades em que a regressão logística é uma boa escolha:\n",
        "\n",
        "* Quando o campo alvo dos dados for categórico, especificamente binário, como um/zero, sim/não, emigrou/não migrou, positivo/negativo, etc.\n",
        "* Quando for interessante estimar probabilidades para previsões. Por exemplo, quando é interessante saber qual a probabilidade de um cliente comprar um produto, a regressão logística retornará uma pontuação de probabilidade entre zero, conforme a amostra de dados.\n",
        "* Quando os dados forem linearmente separáveis, assim um classificador classificará todos os pontos de um lado da fronteira de decisão como pertencentes a uma classe e os do outro lado como pertencentes à outra classe. A fronteira de decisão da regressão logística pode ser uma linha, plano ou hiperplano.\n",
        "* Quando se pretende compreender o impacto de determinada característica. É possível selecionar as melhores características com base na significância estatística dos coeficientes ou parâmetros do modelo de regressão logística. Isso permite compreender o impacto de uma variável independente na variável dependente, controlando as outras variáveis independentes."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "apN9RDNchhYD"
      },
      "source": [
        "## **Instalando e importando as bibliotecas Python**\n",
        "\n",
        "---\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 44,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HcrbaPg6QqCU",
        "outputId": "cc7adb6d-cfff-48f7-8cc1-88ce40bd5581"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.10/dist-packages (1.5.2)\n",
            "Requirement already satisfied: numpy>=1.19.5 in /usr/local/lib/python3.10/dist-packages (from scikit-learn) (1.26.4)\n",
            "Requirement already satisfied: scipy>=1.6.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn) (1.13.1)\n",
            "Requirement already satisfied: joblib>=1.2.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn) (1.4.2)\n",
            "Requirement already satisfied: threadpoolctl>=3.1.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn) (3.5.0)\n",
            "Requirement already satisfied: matplotlib in /usr/local/lib/python3.10/dist-packages (3.8.0)\n",
            "Requirement already satisfied: contourpy>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (1.3.0)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (0.12.1)\n",
            "Requirement already satisfied: fonttools>=4.22.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (4.54.1)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (1.4.7)\n",
            "Requirement already satisfied: numpy<2,>=1.21 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (1.26.4)\n",
            "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (24.2)\n",
            "Requirement already satisfied: pillow>=6.2.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (11.0.0)\n",
            "Requirement already satisfied: pyparsing>=2.3.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (3.2.0)\n",
            "Requirement already satisfied: python-dateutil>=2.7 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (2.8.2)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.7->matplotlib) (1.16.0)\n",
            "Requirement already satisfied: pandas in /usr/local/lib/python3.10/dist-packages (2.2.2)\n",
            "Requirement already satisfied: numpy>=1.22.4 in /usr/local/lib/python3.10/dist-packages (from pandas) (1.26.4)\n",
            "Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.10/dist-packages (from pandas) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas) (2024.2)\n",
            "Requirement already satisfied: tzdata>=2022.7 in /usr/local/lib/python3.10/dist-packages (from pandas) (2024.2)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.2->pandas) (1.16.0)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (1.26.4)\n",
            "Requirement already satisfied: seaborn in /usr/local/lib/python3.10/dist-packages (0.13.2)\n",
            "Requirement already satisfied: numpy!=1.24.0,>=1.20 in /usr/local/lib/python3.10/dist-packages (from seaborn) (1.26.4)\n",
            "Requirement already satisfied: pandas>=1.2 in /usr/local/lib/python3.10/dist-packages (from seaborn) (2.2.2)\n",
            "Requirement already satisfied: matplotlib!=3.6.1,>=3.4 in /usr/local/lib/python3.10/dist-packages (from seaborn) (3.8.0)\n",
            "Requirement already satisfied: contourpy>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.3.0)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.10/dist-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (0.12.1)\n",
            "Requirement already satisfied: fonttools>=4.22.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (4.54.1)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.4.7)\n",
            "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (24.2)\n",
            "Requirement already satisfied: pillow>=6.2.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (11.0.0)\n",
            "Requirement already satisfied: pyparsing>=2.3.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (3.2.0)\n",
            "Requirement already satisfied: python-dateutil>=2.7 in /usr/local/lib/python3.10/dist-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas>=1.2->seaborn) (2024.2)\n",
            "Requirement already satisfied: tzdata>=2022.7 in /usr/local/lib/python3.10/dist-packages (from pandas>=1.2->seaborn) (2024.2)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.7->matplotlib!=3.6.1,>=3.4->seaborn) (1.16.0)\n"
          ]
        }
      ],
      "source": [
        "!pip install scikit-learn # Instala o Scikit-learn para o aprendizado de máquina\n",
        "!pip install matplotlib # Instala o Matplotlib para plotagem gráfica em vários formatos\n",
        "!pip install pandas # Instala o Pandas para análise e limpeza de dados, estruturas de dados e operações\n",
        "!pip install numpy # Instala o NumPy para suporte a arrays e matrizes multidimensionais\n",
        "!pip install seaborn # Instala o Seaborn para visualização de dados baseada no Matplotlib"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 53,
      "metadata": {
        "id": "3P8lTwrlSTnT"
      },
      "outputs": [],
      "source": [
        "import pandas as pd # Manipulação e análise de dados\n",
        "import pylab as pl # Computação numérica e plotagem (Pylab, NumPy)\n",
        "import numpy as np\n",
        "import seaborn as sns # Visualização de dados baseada no Matplotlib\n",
        "import matplotlib.pyplot as plt # Plotagem de gráficos\n",
        "import scipy.optimize as opt # Otimização de funções\n",
        "from sklearn import preprocessing # Pré-processamento de dados"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Estudo de caso: previsão de cancelamento (*churn*) ou rotatividade de clientes de uma empresa de telecomunicações hipotética**\n",
        "\n",
        "Neste estudo de caso, com base em dados hipotéticos (`ChurnData.csv`) de uma empresa de telecomunicações, iremos aplicar a regressão logística para analise de rotatividade dos clientes ou \"previsão de churn\", isto é, prever a probabilidade/risco de um cliente cancelar ou não seus serviços.\n",
        "\n",
        "Ao analisar variáveis como tempo de contrato, uso de serviços e histórico de pagamentos, é possível identificar clientes com maior \"risco de churn\", assim a empresas poderá fortalecer a estratégia de marketing para fidelização (retenção) de determinadas categorias de clientes.\n",
        "\n",
        "**Outros exemplos de aplicações:**\n",
        "\n",
        "1. **Economia e Finanças**:\n",
        "  * **Avaliação do risco de crédito de clientes**, ao analisar variáveis como histórico de crédito, renda e dívidas existentes, é possível determinar a probabilidade de um cliente não cumprir com suas obrigações financeiras, para se determinar se um empréstimo deve ou não ser concedido (risco de inadimplência).\n",
        "  * **Detecção de fraudes em transações**, instituições financeiras tem utilizado modelagens capazes de analisar padrões de comportamento dos clientes, e a partir do monitoramento das transações, o modelo irá sinalizar atividades anômalas passíveis de investigação.\n",
        "  * **Marketing**:\n",
        "  * **Identificar clientes em potencial**, que têm maior probabilidade de responder a uma campanha publicitária.\n",
        "2. **Industria**:\n",
        "  * **Previsão de falhas em equipamentos industriais**, com base m dados de sensores e histórico de manutenção é possível prever falhas em máquinas, auxiliando as equipes de manutenção preditiva, reduzindo custos e evitando paradas não planejadas.\n",
        "3. **Medicina**:\n",
        "  * **Previsão de pacientes desenvolverem determinadas doenças**, com base em base de dados de resultados de exames e fatores de risco em pacientes, modelos como o de regressão logística podem analisar vários fatores, como faixa eatária, índice de massa corporal, pressão arterial, histórico familiar etc, e realizar as predições."
      ],
      "metadata": {
        "id": "Cm1AiD8DvXBd"
      }
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "09ktvzahTq-K"
      },
      "source": [
        "## **Importando e explorando os dados**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 46,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 292
        },
        "id": "O7WkSWNzTwv7",
        "outputId": "7b928fd1-60a4-437b-b210-41c7126d754f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Número de linhas e colunas: (200, 28)\n",
            "Variáveis: ['tenure', 'age', 'address', 'income', 'ed', 'employ', 'equip', 'callcard', 'wireless', 'longmon', 'tollmon', 'equipmon', 'cardmon', 'wiremon', 'longten', 'tollten', 'cardten', 'voice', 'pager', 'internet', 'callwait', 'confer', 'ebill', 'loglong', 'logtoll', 'lninc', 'custcat', 'churn']\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   tenure   age  address  income   ed  employ  equip  callcard  wireless  \\\n",
              "0    11.0  33.0      7.0   136.0  5.0     5.0    0.0       1.0       1.0   \n",
              "1    33.0  33.0     12.0    33.0  2.0     0.0    0.0       0.0       0.0   \n",
              "2    23.0  30.0      9.0    30.0  1.0     2.0    0.0       0.0       0.0   \n",
              "3    38.0  35.0      5.0    76.0  2.0    10.0    1.0       1.0       1.0   \n",
              "4     7.0  35.0     14.0    80.0  2.0    15.0    0.0       1.0       0.0   \n",
              "\n",
              "   longmon  ...  pager  internet  callwait  confer  ebill  loglong  logtoll  \\\n",
              "0     4.40  ...    1.0       0.0       1.0     1.0    0.0    1.482    3.033   \n",
              "1     9.45  ...    0.0       0.0       0.0     0.0    0.0    2.246    3.240   \n",
              "2     6.30  ...    0.0       0.0       0.0     1.0    0.0    1.841    3.240   \n",
              "3     6.05  ...    1.0       1.0       1.0     1.0    1.0    1.800    3.807   \n",
              "4     7.10  ...    0.0       0.0       1.0     1.0    0.0    1.960    3.091   \n",
              "\n",
              "   lninc  custcat  churn  \n",
              "0  4.913      4.0    1.0  \n",
              "1  3.497      1.0    1.0  \n",
              "2  3.401      3.0    0.0  \n",
              "3  4.331      4.0    0.0  \n",
              "4  4.382      3.0    0.0  \n",
              "\n",
              "[5 rows x 28 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-752d8680-0124-4091-bacc-24c4650a5a31\" 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>tenure</th>\n",
              "      <th>age</th>\n",
              "      <th>address</th>\n",
              "      <th>income</th>\n",
              "      <th>ed</th>\n",
              "      <th>employ</th>\n",
              "      <th>equip</th>\n",
              "      <th>callcard</th>\n",
              "      <th>wireless</th>\n",
              "      <th>longmon</th>\n",
              "      <th>...</th>\n",
              "      <th>pager</th>\n",
              "      <th>internet</th>\n",
              "      <th>callwait</th>\n",
              "      <th>confer</th>\n",
              "      <th>ebill</th>\n",
              "      <th>loglong</th>\n",
              "      <th>logtoll</th>\n",
              "      <th>lninc</th>\n",
              "      <th>custcat</th>\n",
              "      <th>churn</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>11.0</td>\n",
              "      <td>33.0</td>\n",
              "      <td>7.0</td>\n",
              "      <td>136.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>4.40</td>\n",
              "      <td>...</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.482</td>\n",
              "      <td>3.033</td>\n",
              "      <td>4.913</td>\n",
              "      <td>4.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>33.0</td>\n",
              "      <td>33.0</td>\n",
              "      <td>12.0</td>\n",
              "      <td>33.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>9.45</td>\n",
              "      <td>...</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.246</td>\n",
              "      <td>3.240</td>\n",
              "      <td>3.497</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>23.0</td>\n",
              "      <td>30.0</td>\n",
              "      <td>9.0</td>\n",
              "      <td>30.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>6.30</td>\n",
              "      <td>...</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.841</td>\n",
              "      <td>3.240</td>\n",
              "      <td>3.401</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>38.0</td>\n",
              "      <td>35.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>76.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>10.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>6.05</td>\n",
              "      <td>...</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.800</td>\n",
              "      <td>3.807</td>\n",
              "      <td>4.331</td>\n",
              "      <td>4.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>7.0</td>\n",
              "      <td>35.0</td>\n",
              "      <td>14.0</td>\n",
              "      <td>80.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>15.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.10</td>\n",
              "      <td>...</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.960</td>\n",
              "      <td>3.091</td>\n",
              "      <td>4.382</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 28 columns</p>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-752d8680-0124-4091-bacc-24c4650a5a31')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-752d8680-0124-4091-bacc-24c4650a5a31 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-752d8680-0124-4091-bacc-24c4650a5a31');\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",
              "\n",
              "\n",
              "<div id=\"df-e2bb0da6-0382-437a-863e-c6bf566f41a9\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-e2bb0da6-0382-437a-863e-c6bf566f41a9')\"\n",
              "            title=\"Suggest charts\"\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",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-e2bb0da6-0382-437a-863e-c6bf566f41a9 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "churn_df"
            }
          },
          "metadata": {},
          "execution_count": 46
        }
      ],
      "source": [
        "# importando e explorando os dados\n",
        "churn_df = pd.read_csv(\"https://raw.githubusercontent.com/SampMark/files/refs/heads/main/ChurnData.csv\")\n",
        "\n",
        "# Imprime o número de linhas e colunas\n",
        "print(\"Número de linhas e colunas:\", churn_df.shape)\n",
        "\n",
        "# Imprime os nomes das colunas/variáveis\n",
        "print(\"Variáveis:\", churn_df.columns.tolist())\n",
        "churn_df.head()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-rQtjh3XmSwH"
      },
      "source": [
        "O arquivo ChurnData.csv contém as seguintes variáveis:\n",
        "\n",
        "* `tenure`: Tempo de permanência do cliente.\n",
        "* `age`: Idade do cliente.\n",
        "* `address`: Número de endereços associados ao cliente.\n",
        "* `income`: Renda do cliente.\n",
        "* `ed`: Nível educacional do cliente.\n",
        "* `employ`: Tempo de emprego do cliente.\n",
        "* `equip`: Indica se o cliente possui equipamento específico.\n",
        "* `callcard`: Indica se o cliente possui cartão de chamadas.\n",
        "* `wireless`: Indica se o cliente possui serviço sem fio.\n",
        "* `longmon`: Gastos mensais com chamadas de longa distância.\n",
        "* `tollmon`: Gastos mensais com tarifas adicionais.\n",
        "* `equipmon`: Gastos mensais com equipamentos.\n",
        "* `cardmon`: Gastos mensais com cartão de chamadas.\n",
        "* `wiremon`: Gastos mensais com serviços sem fio.\n",
        "* `longten`: Gastos totais com chamadas de longa distância.\n",
        "* `tollten`: Gastos totais com tarifas adicionais.\n",
        "* `cardten`: Gastos totais com cartão de chamadas.\n",
        "* `voice`: Indica se o cliente possui serviço de correio de voz.\n",
        "* `pager`: Indica se o cliente possui pager.\n",
        "* `internet`: Indica se o cliente possui serviço de internet.\n",
        "* `callwait`: Indica se o cliente possui serviço de chamada em espera.\n",
        "* `confer`: Indica se o cliente possui serviço de conferência.\n",
        "* `ebill`: Indica se o cliente utiliza faturamento eletrônico.\n",
        "* `loglong`: Logaritmo dos gastos com chamadas de longa distância.\n",
        "* `logtoll`: Logaritmo dos gastos com tarifas adicionais.\n",
        "* `lninc`: Logaritmo da renda do cliente.\n",
        "* `custcat`: Categoria do cliente.\n",
        "* `churn`: total de clientes cancelados em determinado período de tempo (rotatividade de clientes)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 51,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "8Gcjome_UDPM",
        "outputId": "5aacb6ef-6bf7-4855-d02b-fdb30cc95e05"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   tenure   age  address  income  income   ed  employ  equip  callcard  \\\n",
              "0    11.0  33.0      7.0   136.0   136.0  5.0     5.0    0.0       1.0   \n",
              "1    33.0  33.0     12.0    33.0    33.0  2.0     0.0    0.0       0.0   \n",
              "2    23.0  30.0      9.0    30.0    30.0  1.0     2.0    0.0       0.0   \n",
              "3    38.0  35.0      5.0    76.0    76.0  2.0    10.0    1.0       1.0   \n",
              "4     7.0  35.0     14.0    80.0    80.0  2.0    15.0    0.0       1.0   \n",
              "\n",
              "   wireless  income  income  custcat  churn  \n",
              "0       1.0   136.0   136.0      4.0    1.0  \n",
              "1       0.0    33.0    33.0      1.0    1.0  \n",
              "2       0.0    30.0    30.0      3.0    0.0  \n",
              "3       1.0    76.0    76.0      4.0    0.0  \n",
              "4       0.0    80.0    80.0      3.0    0.0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-01f9e945-3905-44d3-89a7-542a4fe44780\" 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>tenure</th>\n",
              "      <th>age</th>\n",
              "      <th>address</th>\n",
              "      <th>income</th>\n",
              "      <th>income</th>\n",
              "      <th>ed</th>\n",
              "      <th>employ</th>\n",
              "      <th>equip</th>\n",
              "      <th>callcard</th>\n",
              "      <th>wireless</th>\n",
              "      <th>income</th>\n",
              "      <th>income</th>\n",
              "      <th>custcat</th>\n",
              "      <th>churn</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>11.0</td>\n",
              "      <td>33.0</td>\n",
              "      <td>7.0</td>\n",
              "      <td>136.0</td>\n",
              "      <td>136.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>136.0</td>\n",
              "      <td>136.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>33.0</td>\n",
              "      <td>33.0</td>\n",
              "      <td>12.0</td>\n",
              "      <td>33.0</td>\n",
              "      <td>33.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>33.0</td>\n",
              "      <td>33.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>23.0</td>\n",
              "      <td>30.0</td>\n",
              "      <td>9.0</td>\n",
              "      <td>30.0</td>\n",
              "      <td>30.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>30.0</td>\n",
              "      <td>30.0</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>38.0</td>\n",
              "      <td>35.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>76.0</td>\n",
              "      <td>76.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>10.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>76.0</td>\n",
              "      <td>76.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>7.0</td>\n",
              "      <td>35.0</td>\n",
              "      <td>14.0</td>\n",
              "      <td>80.0</td>\n",
              "      <td>80.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>15.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>80.0</td>\n",
              "      <td>80.0</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-01f9e945-3905-44d3-89a7-542a4fe44780')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-01f9e945-3905-44d3-89a7-542a4fe44780 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-01f9e945-3905-44d3-89a7-542a4fe44780');\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",
              "\n",
              "\n",
              "<div id=\"df-c1ecf5d2-ca6e-44a5-9b14-d08cf9602b85\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-c1ecf5d2-ca6e-44a5-9b14-d08cf9602b85')\"\n",
              "            title=\"Suggest charts\"\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",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-c1ecf5d2-ca6e-44a5-9b14-d08cf9602b85 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "churn_df",
              "summary": "{\n  \"name\": \"churn_df\",\n  \"rows\": 200,\n  \"fields\": [\n    {\n      \"column\": \"tenure\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 21.640971169654776,\n        \"min\": 1.0,\n        \"max\": 72.0,\n        \"num_unique_values\": 68,\n        \"samples\": [\n          67.0,\n          52.0,\n          7.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"age\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 13.076803468153965,\n        \"min\": 19.0,\n        \"max\": 76.0,\n        \"num_unique_values\": 51,\n        \"samples\": [\n          41.0,\n          32.0,\n          53.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"address\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 10.158418545063164,\n        \"min\": 0.0,\n        \"max\": 48.0,\n        \"num_unique_values\": 40,\n        \"samples\": [\n          25.0,\n          13.0,\n          23.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"income\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 128.43046766684935,\n        \"min\": 9.0,\n        \"max\": 1668.0,\n        \"num_unique_values\": 101,\n        \"samples\": [\n          72.0,\n          12.0,\n          60.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"income\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 128.43046766684935,\n        \"min\": 9.0,\n        \"max\": 1668.0,\n        \"num_unique_values\": 101,\n        \"samples\": [\n          72.0,\n          12.0,\n          60.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ed\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.2855497621464422,\n        \"min\": 1.0,\n        \"max\": 5.0,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          2.0,\n          3.0,\n          1.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"employ\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 8.957430031198628,\n        \"min\": 0.0,\n        \"max\": 44.0,\n        \"num_unique_values\": 35,\n        \"samples\": [\n          27.0,\n          1.0,\n          17.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"equip\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.4955835096887065,\n        \"min\": 0.0,\n        \"max\": 1.0,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          1.0,\n          0.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"callcard\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.45718716134917026,\n        \"min\": 0.0,\n        \"max\": 1.0,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          0.0,\n          1.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"wireless\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.4549007291342079,\n        \"min\": 0.0,\n        \"max\": 1.0,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          0.0,\n          1.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"income\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 128.43046766684935,\n        \"min\": 9.0,\n        \"max\": 1668.0,\n        \"num_unique_values\": 101,\n        \"samples\": [\n          72.0,\n          12.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"income\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 128.43046766684935,\n        \"min\": 9.0,\n        \"max\": 1668.0,\n        \"num_unique_values\": 101,\n        \"samples\": [\n          72.0,\n          12.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"custcat\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1.0794447670099316,\n        \"min\": 1.0,\n        \"max\": 4.0,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          1.0,\n          2.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"churn\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.454900729134208,\n        \"min\": 0.0,\n        \"max\": 1.0,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          0.0,\n          1.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 51
        }
      ],
      "source": [
        "# Selecionando as variáveis de interesse\n",
        "churn_df = churn_df[['tenure', 'age', 'address', 'income', 'ed', 'employ', 'equip', 'callcard', 'wireless', 'income', 'custcat', 'churn']]\n",
        "# Convertendo a coluna 'churn' para o tipo inteiro\n",
        "churn_df.loc[:, 'churn'] = churn_df['churn'].astype('int')\n",
        "churn_df.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GOlDj0plUTFv",
        "outputId": "f78cbffb-22f9-424e-b8f0-6a0e47f7640b"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Colunas selecionadas: ['tenure', 'age', 'address', 'income', 'ed', 'employ', 'equip', 'callcard', 'wireless', 'churn']\n",
            "Número de linhas e colunas selecionadas: (200, 10)\n"
          ]
        }
      ],
      "source": [
        "# Checa os nomes das colunas selecionadas\n",
        "print(\"Colunas selecionadas:\", churn_df.columns.tolist())\n",
        "\n",
        "# Imprime o número de linhas e colunas selecionadas\n",
        "print(\"Número de linhas e colunas selecionadas:\", churn_df.shape)"
      ]
    },
    {
      "source": [
        "# @title Consumidores por idade vs Taxa de Cancelamento\n",
        "\n",
        "plt.figure(figsize=(10, 6))\n",
        "plt.hist([churn_df[churn_df['churn'] == 0]['age'], churn_df[churn_df['churn'] == 1]['age']],\n",
        "         bins=10, label=['Não Cancelado', 'Cancelado'], color=['blue', 'orange'], alpha=0.7)\n",
        "plt.xlabel('Idade')\n",
        "plt.ylabel('Número de Consumidores')\n",
        "plt.title('Distribuição de consumidores por idade versus taxa de cancelamento (rotatividade)')\n",
        "_ = plt.legend()"
      ],
      "cell_type": "code",
      "execution_count": 65,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 564
        },
        "id": "iWjgsXeF1Yiu",
        "outputId": "dfd00008-3406-48aa-cc28-bf16863a52b7"
      }
    },
    {
      "source": [
        "# @title Duração no emprego vs Permanência\n",
        "\n",
        "churn_df.plot(kind='scatter', x='employ', y='tenure', s=32, alpha=.8)\n",
        "plt.gca().spines[['top', 'right',]].set_visible(False)\n",
        "plt.xlabel('Tempo no emprego')\n",
        "plt.ylabel('Tempo permanência do cliente')\n",
        "plt.title('Duração no emprego versus rotatividade de clientes')\n",
        "plt.show()"
      ],
      "cell_type": "code",
      "execution_count": 123,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "5fCRa9HwyKGq",
        "outputId": "565bb62c-476d-47ea-9e41-6f8341c1c53b"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Etapa de pré-processamento de dados**\n",
        "\n",
        "---\n",
        "\n",
        "**Por que o pré-processamento é necessário?**\n",
        "\n",
        "No aprendizado de máquina, a etapa de pré-processamento de dados é fundamental para garantir a qualidade e a eficácia dos modelos desenvolvidos. Uma das práticas essenciais nesse processo é a **padronização dos dados**, que envolve transformar as variáveis para que tenham média zero e desvio padrão igual a um.\n",
        "Além da padronização, existem outras técnicas como a **normalização**, **codificação de variáveis categóricas** e o **tratamento de valores faltantes**.\n",
        "\n",
        "1. **Melhorar a performance do modelo**: Algoritmos de aprendizado de máquina são sensíveis à escala das features. Quando as features têm escalas diferentes, aquelas com valores maiores podem dominar o processo de treinamento, levando a um modelo enviesado e com performance inferior. A padronização coloca todas as features na mesma escala, evitando problemas numéricos e permitindo que o algoritmo aprenda de forma mais eficaz.\n",
        "\n",
        "2. **Convergência mais rápida e estável em algoritmos de otimização**:  Em algoritmos que usam gradiente descendente (como regressão logística, redes neurais, etc.), sensíveis à escala das variáveis, a padronização ajuda a acelerar o processo de convergência, permitindo que o gradiente descendente convirja mais rapidamente quando as features estão na mesma escala. Quando as variáveis não são padronizadas, o processo de otimização pode se tornar ineficiente, resultando em uma convergência mais lenta ou até mesmo na incapacidade de encontrar o mínimo global da função de custo.\n",
        "\n",
        "3. **Facilita a interpretação dos coeficientes**:  Em modelos lineares como a regressão logística, os coeficientes representam a importância de cada variável dependente (*feature*). Quando as features são padronizadas, os coeficientes se tornam comparáveis, facilitando a interpretação sobre a importância relativa de cada feature.\n",
        "\n",
        "4. **Uniformização das Escalas das Variáveis**: Em conjuntos de dados, é comum que as variáveis apresentem diferentes escalas e unidades de medida. Por exemplo, enquanto a renda pode ser expressa em milhares de reais, a idade é medida em anos. Modelos de aprendizado de máquina que utilizam medidas de distância, como k-Nearest Neighbors (k-NN) ou algoritmos baseados em gradiente, podem atribuir maior importância a variáveis com maior magnitude, independentemente de sua relevância real. A padronização coloca todas as variáveis na mesma escala, evitando vieses indesejados.\n",
        "\n",
        "  * **Redução da Influência de Outliers**: Embora a padronização não elimine outliers, ela pode reduzir seu impacto relativo. Ao centralizar os dados em torno de uma média zero e normalizar a variância, os outliers se tornam menos proeminentes, permitindo que o modelo aprenda padrões mais representativos dos dados.\n",
        "\n",
        "**Padronização dos Dados**\n",
        "\n",
        "Conforme mencionado a padronização transforma os dados para que tenham média 0 e desvio padrão 1, de modo a preservar o padrão de distribuição original dos dados, centralizando e escalonando os valores. O cálculo é feito subtraindo-se a média e dividindo pelo desvio padrão de cada feature. A padronização é útil em algoritmos sensíveis à escala das features, como:\n",
        "\n",
        "* Regressão logística\n",
        "* Support Vector Machines (SVMs)\n",
        "* K-Nearest Neighbors (KNN)\n",
        "* Redes neurais\n",
        "* Principal Component Analysis (PCA)\n",
        "\n",
        "**Outras técnicas de pré-processamento:**\n",
        "\n",
        "* **Normalização**: Escala os dados para um intervalo específico, como [0, 1].\n",
        "* **Codificação de variáveis categóricas**: Converte variáveis categóricas em numéricas para que possam ser usadas pelos algoritmos.\n",
        "* **Tratamento de valores faltantes**: Preenche ou remove valores faltantes nos dados.\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "---\n",
        "\n",
        "\n",
        "Aplicar as técnicas de pré-processamento de dados é essencial para estabilidade, eficiência e interpretabilidade dos modelos de aprendizado de máquina. Negligenciá-la pode resultar em modelos tendenciosos, com desempenho inferior e interpretações equivocadas.\n"
      ],
      "metadata": {
        "id": "VqkqnKDi8bOt"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Convertendo as features (variáveis independentes) em um array NumPy\n",
        "X = np.asarray(churn_df[['tenure', 'age', 'address', 'income', 'ed', 'employ', 'equip']])\n",
        "\n",
        "# Exibindo as primeiras 5 linhas das features\n",
        "print(\"Variáveis independentes (X):\\n\", X[0:5])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8YBUcQbH7G8b",
        "outputId": "3659040d-e687-4dea-dbc3-d671b134ce11"
      },
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Variáveis independentes (X):\n",
            " [[ 11.  33.   7. 136. 136. 136. 136.   5.   5.   0.]\n",
            " [ 33.  33.  12.  33.  33.  33.  33.   2.   0.   0.]\n",
            " [ 23.  30.   9.  30.  30.  30.  30.   1.   2.   0.]\n",
            " [ 38.  35.   5.  76.  76.  76.  76.   2.  10.   1.]\n",
            " [  7.  35.  14.  80.  80.  80.  80.   2.  15.   0.]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Convertendo os rótulos (variável dependente) em um array NumPy\n",
        "y = np.asarray(churn_df['churn'])\n",
        "\n",
        "# Exibindo os primeiros 5 rótulos\n",
        "print(\"\\nRótulos (y):\\n\", y[0:5])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "r0HJBVy97W3k",
        "outputId": "899f0c62-0c28-42a4-c127-6bab4c47046b"
      },
      "execution_count": 63,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Rótulos (y):\n",
            " [1. 1. 0. 0. 0.]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 64,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VI7-wVX2VGaY",
        "outputId": "8446d101-20fe-400a-b8b8-e494946fa2e6"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Features padronizadas (X):\n",
            " [[-1.13518441 -0.62595491 -0.4588971   0.4751423   0.4751423   0.4751423\n",
            "   0.4751423   1.6961288  -0.58477841 -0.85972695]\n",
            " [-0.11604313 -0.62595491  0.03454064 -0.32886061 -0.32886061 -0.32886061\n",
            "  -0.32886061 -0.6433592  -1.14437497 -0.85972695]\n",
            " [-0.57928917 -0.85594447 -0.261522   -0.35227817 -0.35227817 -0.35227817\n",
            "  -0.35227817 -1.42318853 -0.92053635 -0.85972695]\n",
            " [ 0.11557989 -0.47262854 -0.65627219  0.00679109  0.00679109  0.00679109\n",
            "   0.00679109 -0.6433592  -0.02518185  1.16316   ]\n",
            " [-1.32048283 -0.47262854  0.23191574  0.03801451  0.03801451  0.03801451\n",
            "   0.03801451 -0.6433592   0.53441472 -0.85972695]]\n"
          ]
        }
      ],
      "source": [
        "from sklearn import preprocessing\n",
        "# Padronizando as features (transformando para média 0 e desvio padrão 1)\n",
        "X = preprocessing.StandardScaler().fit(X).transform(X)\n",
        "\n",
        "# Exibindo as primeiras 5 linhas das features após a padronização\n",
        "print(\"\\nFeatures padronizadas (X):\\n\", X[0:5])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rn0YPtalVM0X"
      },
      "source": [
        "## **Divisão dos dados em conjuntos de treinamento e teste**\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 71,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "o96icaRBVOgl",
        "outputId": "d0fd3397-08c5-4ef8-fa60-86aa3ecaef95"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Conjunto de Treinamento: (160, 10) (160,)\n",
            "Conjunto de Teste: (40, 10) (40,)\n"
          ]
        }
      ],
      "source": [
        "# Importando a função train_test_split do Scikit-learn\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "# Dividindo os dados em conjuntos de treinamento e teste\n",
        "# 80% para treinamento e 20% para teste, com random_state=4 para reprodutibilidade\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=4)\n",
        "\n",
        "# Exibindo o tamanho do conjunto de treinamento\n",
        "print('Conjunto de Treinamento:', X_train.shape, y_train.shape)\n",
        "\n",
        "# Exibindo o tamanho do conjunto de teste\n",
        "print('Conjunto de Teste:', X_test.shape, y_test.shape)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6ARglUYIVVGV"
      },
      "source": [
        "## **Modelando uma Regressão Logística com Scikit-learn**\n",
        "\n",
        "---\n",
        "Para construir o modelo, começamos implementando a função `LogisticRegression` do pacote Scikit-learn, que possui diferentes otimizadores numéricos para encontrar parâmetros, tais como: `'newton-cg'`, `'lbfgs'`, `'liblinear'`, `'sag'`, `'saga'`.\n",
        "\n",
        "A versão da Regressão Logística no Scikit-learn suporta regularização, esta é uma técnica usada para resolver o problema de *overfitting* em modelos de aprendizado de máquina. O parâmetro C indica o inverso da força de regularização, que deve ser um float positivo. Valores menores especificam uma regularização mais forte. Em seuida, ajusta-se o modelo com o conjunto de treinamento.\n",
        "\n",
        "* `LogisticRegression(C=0.01, solver='liblinear')`: Cria o objeto `LogisticRegression` com os seguintes parâmetros:\n",
        "\n",
        "  * `C=0.01`: Parâmetro de regularização, que ajuda a prevenir o *overfitting*, valores menores especificam uma regularização mais forte.\n",
        "\n",
        "  * `solver='liblinear'`: Especifica o parâmetro de otimização `'liblinear'`, boa opção para conjuntos de dados pequenos e médios.\n",
        "\n",
        "* `.fit(X_train,y_train)`: Treina o modelo de regressão logística usando os dados de treinamento `X_train` (*features*) e `y_train` (rótulos)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 105,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 80
        },
        "id": "XRw1CTnnVXgo",
        "outputId": "b7288078-816d-4c42-ea74-db3787cd2a46"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression(C=0.01, solver='liblinear')"
            ],
            "text/html": [
              "<style>#sk-container-id-4 {\n",
              "  /* Definition of color scheme common for light and dark mode */\n",
              "  --sklearn-color-text: black;\n",
              "  --sklearn-color-line: gray;\n",
              "  /* Definition of color scheme for unfitted estimators */\n",
              "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
              "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
              "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
              "  --sklearn-color-unfitted-level-3: chocolate;\n",
              "  /* Definition of color scheme for fitted estimators */\n",
              "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
              "  --sklearn-color-fitted-level-1: #d4ebff;\n",
              "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
              "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
              "\n",
              "  /* Specific color for light theme */\n",
              "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
              "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
              "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
              "  --sklearn-color-icon: #696969;\n",
              "\n",
              "  @media (prefers-color-scheme: dark) {\n",
              "    /* Redefinition of color scheme for dark theme */\n",
              "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
              "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
              "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
              "    --sklearn-color-icon: #878787;\n",
              "  }\n",
              "}\n",
              "\n",
              "#sk-container-id-4 {\n",
              "  color: var(--sklearn-color-text);\n",
              "}\n",
              "\n",
              "#sk-container-id-4 pre {\n",
              "  padding: 0;\n",
              "}\n",
              "\n",
              "#sk-container-id-4 input.sk-hidden--visually {\n",
              "  border: 0;\n",
              "  clip: rect(1px 1px 1px 1px);\n",
              "  clip: rect(1px, 1px, 1px, 1px);\n",
              "  height: 1px;\n",
              "  margin: -1px;\n",
              "  overflow: hidden;\n",
              "  padding: 0;\n",
              "  position: absolute;\n",
              "  width: 1px;\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-dashed-wrapped {\n",
              "  border: 1px dashed var(--sklearn-color-line);\n",
              "  margin: 0 0.4em 0.5em 0.4em;\n",
              "  box-sizing: border-box;\n",
              "  padding-bottom: 0.4em;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-container {\n",
              "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
              "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
              "     so we also need the `!important` here to be able to override the\n",
              "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
              "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
              "  display: inline-block !important;\n",
              "  position: relative;\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-text-repr-fallback {\n",
              "  display: none;\n",
              "}\n",
              "\n",
              "div.sk-parallel-item,\n",
              "div.sk-serial,\n",
              "div.sk-item {\n",
              "  /* draw centered vertical line to link estimators */\n",
              "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
              "  background-size: 2px 100%;\n",
              "  background-repeat: no-repeat;\n",
              "  background-position: center center;\n",
              "}\n",
              "\n",
              "/* Parallel-specific style estimator block */\n",
              "\n",
              "#sk-container-id-4 div.sk-parallel-item::after {\n",
              "  content: \"\";\n",
              "  width: 100%;\n",
              "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
              "  flex-grow: 1;\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-parallel {\n",
              "  display: flex;\n",
              "  align-items: stretch;\n",
              "  justify-content: center;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "  position: relative;\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-parallel-item {\n",
              "  display: flex;\n",
              "  flex-direction: column;\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-parallel-item:first-child::after {\n",
              "  align-self: flex-end;\n",
              "  width: 50%;\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-parallel-item:last-child::after {\n",
              "  align-self: flex-start;\n",
              "  width: 50%;\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-parallel-item:only-child::after {\n",
              "  width: 0;\n",
              "}\n",
              "\n",
              "/* Serial-specific style estimator block */\n",
              "\n",
              "#sk-container-id-4 div.sk-serial {\n",
              "  display: flex;\n",
              "  flex-direction: column;\n",
              "  align-items: center;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "  padding-right: 1em;\n",
              "  padding-left: 1em;\n",
              "}\n",
              "\n",
              "\n",
              "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
              "clickable and can be expanded/collapsed.\n",
              "- Pipeline and ColumnTransformer use this feature and define the default style\n",
              "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
              "*/\n",
              "\n",
              "/* Pipeline and ColumnTransformer style (default) */\n",
              "\n",
              "#sk-container-id-4 div.sk-toggleable {\n",
              "  /* Default theme specific background. It is overwritten whether we have a\n",
              "  specific estimator or a Pipeline/ColumnTransformer */\n",
              "  background-color: var(--sklearn-color-background);\n",
              "}\n",
              "\n",
              "/* Toggleable label */\n",
              "#sk-container-id-4 label.sk-toggleable__label {\n",
              "  cursor: pointer;\n",
              "  display: block;\n",
              "  width: 100%;\n",
              "  margin-bottom: 0;\n",
              "  padding: 0.5em;\n",
              "  box-sizing: border-box;\n",
              "  text-align: center;\n",
              "}\n",
              "\n",
              "#sk-container-id-4 label.sk-toggleable__label-arrow:before {\n",
              "  /* Arrow on the left of the label */\n",
              "  content: \"▸\";\n",
              "  float: left;\n",
              "  margin-right: 0.25em;\n",
              "  color: var(--sklearn-color-icon);\n",
              "}\n",
              "\n",
              "#sk-container-id-4 label.sk-toggleable__label-arrow:hover:before {\n",
              "  color: var(--sklearn-color-text);\n",
              "}\n",
              "\n",
              "/* Toggleable content - dropdown */\n",
              "\n",
              "#sk-container-id-4 div.sk-toggleable__content {\n",
              "  max-height: 0;\n",
              "  max-width: 0;\n",
              "  overflow: hidden;\n",
              "  text-align: left;\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-toggleable__content.fitted {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-toggleable__content pre {\n",
              "  margin: 0.2em;\n",
              "  border-radius: 0.25em;\n",
              "  color: var(--sklearn-color-text);\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-toggleable__content.fitted pre {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-4 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
              "  /* Expand drop-down */\n",
              "  max-height: 200px;\n",
              "  max-width: 100%;\n",
              "  overflow: auto;\n",
              "}\n",
              "\n",
              "#sk-container-id-4 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
              "  content: \"▾\";\n",
              "}\n",
              "\n",
              "/* Pipeline/ColumnTransformer-specific style */\n",
              "\n",
              "#sk-container-id-4 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
              "  color: var(--sklearn-color-text);\n",
              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
              "  background-color: var(--sklearn-color-fitted-level-2);\n",
              "}\n",
              "\n",
              "/* Estimator-specific style */\n",
              "\n",
              "/* Colorize estimator box */\n",
              "#sk-container-id-4 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-2);\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-label label.sk-toggleable__label,\n",
              "#sk-container-id-4 div.sk-label label {\n",
              "  /* The background is the default theme color */\n",
              "  color: var(--sklearn-color-text-on-default-background);\n",
              "}\n",
              "\n",
              "/* On hover, darken the color of the background */\n",
              "#sk-container-id-4 div.sk-label:hover label.sk-toggleable__label {\n",
              "  color: var(--sklearn-color-text);\n",
              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
              "}\n",
              "\n",
              "/* Label box, darken color on hover, fitted */\n",
              "#sk-container-id-4 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
              "  color: var(--sklearn-color-text);\n",
              "  background-color: var(--sklearn-color-fitted-level-2);\n",
              "}\n",
              "\n",
              "/* Estimator label */\n",
              "\n",
              "#sk-container-id-4 div.sk-label label {\n",
              "  font-family: monospace;\n",
              "  font-weight: bold;\n",
              "  display: inline-block;\n",
              "  line-height: 1.2em;\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-label-container {\n",
              "  text-align: center;\n",
              "}\n",
              "\n",
              "/* Estimator-specific */\n",
              "#sk-container-id-4 div.sk-estimator {\n",
              "  font-family: monospace;\n",
              "  border: 1px dotted var(--sklearn-color-border-box);\n",
              "  border-radius: 0.25em;\n",
              "  box-sizing: border-box;\n",
              "  margin-bottom: 0.5em;\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-0);\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-estimator.fitted {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-0);\n",
              "}\n",
              "\n",
              "/* on hover */\n",
              "#sk-container-id-4 div.sk-estimator:hover {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-2);\n",
              "}\n",
              "\n",
              "#sk-container-id-4 div.sk-estimator.fitted:hover {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-2);\n",
              "}\n",
              "\n",
              "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
              "\n",
              "/* Common style for \"i\" and \"?\" */\n",
              "\n",
              ".sk-estimator-doc-link,\n",
              "a:link.sk-estimator-doc-link,\n",
              "a:visited.sk-estimator-doc-link {\n",
              "  float: right;\n",
              "  font-size: smaller;\n",
              "  line-height: 1em;\n",
              "  font-family: monospace;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "  border-radius: 1em;\n",
              "  height: 1em;\n",
              "  width: 1em;\n",
              "  text-decoration: none !important;\n",
              "  margin-left: 1ex;\n",
              "  /* unfitted */\n",
              "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
              "  color: var(--sklearn-color-unfitted-level-1);\n",
              "}\n",
              "\n",
              ".sk-estimator-doc-link.fitted,\n",
              "a:link.sk-estimator-doc-link.fitted,\n",
              "a:visited.sk-estimator-doc-link.fitted {\n",
              "  /* fitted */\n",
              "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
              "  color: var(--sklearn-color-fitted-level-1);\n",
              "}\n",
              "\n",
              "/* On hover */\n",
              "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
              ".sk-estimator-doc-link:hover,\n",
              "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
              ".sk-estimator-doc-link:hover {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-3);\n",
              "  color: var(--sklearn-color-background);\n",
              "  text-decoration: none;\n",
              "}\n",
              "\n",
              "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
              ".sk-estimator-doc-link.fitted:hover,\n",
              "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
              ".sk-estimator-doc-link.fitted:hover {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-3);\n",
              "  color: var(--sklearn-color-background);\n",
              "  text-decoration: none;\n",
              "}\n",
              "\n",
              "/* Span, style for the box shown on hovering the info icon */\n",
              ".sk-estimator-doc-link span {\n",
              "  display: none;\n",
              "  z-index: 9999;\n",
              "  position: relative;\n",
              "  font-weight: normal;\n",
              "  right: .2ex;\n",
              "  padding: .5ex;\n",
              "  margin: .5ex;\n",
              "  width: min-content;\n",
              "  min-width: 20ex;\n",
              "  max-width: 50ex;\n",
              "  color: var(--sklearn-color-text);\n",
              "  box-shadow: 2pt 2pt 4pt #999;\n",
              "  /* unfitted */\n",
              "  background: var(--sklearn-color-unfitted-level-0);\n",
              "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
              "}\n",
              "\n",
              ".sk-estimator-doc-link.fitted span {\n",
              "  /* fitted */\n",
              "  background: var(--sklearn-color-fitted-level-0);\n",
              "  border: var(--sklearn-color-fitted-level-3);\n",
              "}\n",
              "\n",
              ".sk-estimator-doc-link:hover span {\n",
              "  display: block;\n",
              "}\n",
              "\n",
              "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
              "\n",
              "#sk-container-id-4 a.estimator_doc_link {\n",
              "  float: right;\n",
              "  font-size: 1rem;\n",
              "  line-height: 1em;\n",
              "  font-family: monospace;\n",
              "  background-color: var(--sklearn-color-background);\n",
              "  border-radius: 1rem;\n",
              "  height: 1rem;\n",
              "  width: 1rem;\n",
              "  text-decoration: none;\n",
              "  /* unfitted */\n",
              "  color: var(--sklearn-color-unfitted-level-1);\n",
              "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
              "}\n",
              "\n",
              "#sk-container-id-4 a.estimator_doc_link.fitted {\n",
              "  /* fitted */\n",
              "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
              "  color: var(--sklearn-color-fitted-level-1);\n",
              "}\n",
              "\n",
              "/* On hover */\n",
              "#sk-container-id-4 a.estimator_doc_link:hover {\n",
              "  /* unfitted */\n",
              "  background-color: var(--sklearn-color-unfitted-level-3);\n",
              "  color: var(--sklearn-color-background);\n",
              "  text-decoration: none;\n",
              "}\n",
              "\n",
              "#sk-container-id-4 a.estimator_doc_link.fitted:hover {\n",
              "  /* fitted */\n",
              "  background-color: var(--sklearn-color-fitted-level-3);\n",
              "}\n",
              "</style><div id=\"sk-container-id-4\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>LogisticRegression(C=0.01, solver=&#x27;liblinear&#x27;)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" checked><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;&nbsp;LogisticRegression<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.5/modules/generated/sklearn.linear_model.LogisticRegression.html\">?<span>Documentation for LogisticRegression</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></label><div class=\"sk-toggleable__content fitted\"><pre>LogisticRegression(C=0.01, solver=&#x27;liblinear&#x27;)</pre></div> </div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 105
        }
      ],
      "source": [
        "# Importa a classe LogisticRegression da biblioteca Scikit-learn\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "\n",
        "LR = LogisticRegression(C=0.01, solver='liblinear').fit(X_train,y_train)\n",
        "LR # Exibe o modelo de regressão logística treinado e mostra os parâmetros escolhidos"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Visão geral dos otimizadores**\n",
        "Cada otimizador tem suas características, vantagens e desvantagens, tornando a escolha do otimizador um passo importante no processo de modelagem.\n",
        "\n",
        "1. `newton-cg`: **Método de Newton com Conjugate Gradient**:\n",
        "  * Adequado para problemas de pequena a média escala com features densas.\n",
        "  Requer o cálculo da Hessiana (matriz de derivadas segundas), o que pode ser computacionalmente caro para conjuntos de dados grandes.\n",
        "\n",
        "2. `lbfgs`: **Limited-memory Broyden–Fletcher–Goldfarb–Shanno**:\n",
        "  * Método Quasi-Newton que aproxima a Hessiana. É eficiente em termos de memória, tornando-o adequado para problemas de grande escala, mas pode ser mais lento que `'newton-cg'` em alguns casos.\n",
        "\n",
        "3. `liblinear`: **Utiliza a biblioteca LIBLINEAR**:\n",
        "  * Adequado para problemas de grande escala com features esparsas.\n",
        "  Suporta apenas regularização L1 e L2. Pode ser mais rápido que `'newton-cg'` e `'lbfgs'` para dados esparsos.\n",
        "\n",
        "4. `sag`: **Stochastic Average Gradient descent**:\n",
        "  * Método de gradiente descendente estocástico que usa uma média dos gradientes.\n",
        "  Adequado para problemas de grande escala com features densas. Converge mais rápido que o gradiente descendente estocástico padrão.\n",
        "\n",
        "5. `saga`: **Variante do 'sag' que suporta regularização L1**:\n",
        "  * Combina as vantagens do 'sag' com a capacidade de lidar com regularização L1."
      ],
      "metadata": {
        "id": "MpphKD9RGaeB"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 77,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "q38HDQirVszB",
        "outputId": "12294436-dcce-4858-e40f-f925e953428e"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Previsões (yhat): [0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 1. 1. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0.\n",
            " 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 1. 0. 0. 0.]\n"
          ]
        }
      ],
      "source": [
        "# Usa o modelo de regressão logística treinado LR para fazer previsões nos dados de teste X_test\n",
        "yhat = LR.predict(X_test)\n",
        "yhat # Armazena as previsões geradas pelo modelo\n",
        "# Exibindo as previsões\n",
        "print(\"Previsões (yhat):\", yhat)"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* `yhat` é um array contendo as classes previstas (0 ou 1) para cada amostra no conjunto de teste.\n",
        "\n",
        "\n",
        "---\n",
        "\n",
        "* `predizer_proba` retorna estimativas para todas as classes, ordenadas pelo rótulo das classes. Portanto, a primeira coluna é a probabilidade da `classe 0`, P(Y=0|X), e a segunda coluna é a probabilidade da `classe `1, P(Y=1|X):"
      ],
      "metadata": {
        "id": "3bBlJXPXLpGf"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 79,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OtbiD_93VxZe",
        "outputId": "56ddf0fa-a049-4b5b-8dff-d20e18b9e812"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Probabilidades (yhat_prob):\n",
            " [[0.54341504 0.45658496]\n",
            " [0.60605901 0.39394099]\n",
            " [0.56109377 0.43890623]\n",
            " [0.63583688 0.36416312]\n",
            " [0.56326336 0.43673664]\n",
            " [0.55319206 0.44680794]\n",
            " [0.52196501 0.47803499]\n",
            " [0.60373312 0.39626688]\n",
            " [0.40928918 0.59071082]\n",
            " [0.63207988 0.36792012]\n",
            " [0.58212024 0.41787976]\n",
            " [0.63211043 0.36788957]\n",
            " [0.47461064 0.52538936]\n",
            " [0.42455698 0.57544302]\n",
            " [0.66046238 0.33953762]\n",
            " [0.55002811 0.44997189]\n",
            " [0.5164954  0.4835046 ]\n",
            " [0.48473034 0.51526966]\n",
            " [0.48904711 0.51095289]\n",
            " [0.52204537 0.47795463]\n",
            " [0.61497219 0.38502781]\n",
            " [0.52493835 0.47506165]\n",
            " [0.63867398 0.36132602]\n",
            " [0.52093508 0.47906492]\n",
            " [0.50741434 0.49258566]\n",
            " [0.70476794 0.29523206]\n",
            " [0.56065001 0.43934999]\n",
            " [0.52132975 0.47867025]\n",
            " [0.51523421 0.48476579]\n",
            " [0.72077758 0.27922242]\n",
            " [0.67591405 0.32408595]\n",
            " [0.50978046 0.49021954]\n",
            " [0.42362683 0.57637317]\n",
            " [0.71698663 0.28301337]\n",
            " [0.59714599 0.40285401]\n",
            " [0.63607667 0.36392333]\n",
            " [0.4029295  0.5970705 ]\n",
            " [0.52120939 0.47879061]\n",
            " [0.65884179 0.34115821]\n",
            " [0.509909   0.490091  ]]\n"
          ]
        }
      ],
      "source": [
        "# Calculando a probabilidade de cada classe para as amostras no conjunto de teste\n",
        "yhat_prob = LR.predict_proba(X_test)\n",
        "yhat_prob # Probabilidades calculadas para cada classe\n",
        "\n",
        "# Exibindo as probabilidades\n",
        "print(\"Probabilidades (yhat_prob):\\n\", yhat_prob)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "M18Ya0vJXLzT"
      },
      "source": [
        "## **Avaliação do Modelo**\n",
        "\n",
        "### **Índice Jaccard**\n",
        "\n",
        "O índice Jaccard pode ser utilizado para a avaliação da precisão do modelo. Podemos definir Jaccard como o tamanho da intersecção dividido pelo tamanho da união dos dois conjuntos de rótulos. Se o conjunto inteiro de rótulos previstos para uma amostra corresponder estritamente ao conjunto verdadeiro de rótulos, então a precisão do subconjunto é 1.0; caso contrário, é 0.0."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 118,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "blBqXOUfXMEb",
        "outputId": "37fe202a-151f-4fb4-f7d1-d8c133f3302a"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7058823529411765"
            ]
          },
          "metadata": {},
          "execution_count": 118
        }
      ],
      "source": [
        "from sklearn.metrics import jaccard_score\n",
        "jaccard_score(y_test, yhat,pos_label=0)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "r0Z66FHNXzRo"
      },
      "source": [
        "**Análise do Resultado:**\n",
        "\n",
        "De acordo com o índice Jaccard para a `classe 1` (*churn*) é 0,7059 ou 70,59%, o que indica uma boa interseção entre as previsões corretas e os casos reais positivos. No entanto, ainda há uma margem para melhorar a identificação de clientes propensos ao cancelamento (*churn*)."
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Log Loss (perda logarítmica)**\n",
        "\n",
        "A `log_loss` é uma métrica que penaliza fortemente os modelos que fazem previsões com alta confiança, mas que estão erradas. É frequentemente aplicado na avaliação de modelos de classificação em problemas diversos, como previsão de cancelamento (*churn*), detecção de fraudes, diagnóstico médico etc, quando a probabilidade da previsão é crucial.\n",
        "\n",
        "**Observações**:\n",
        "* Quanto menor a Log Loss, melhor é o desempenho do modelo (valores próximos de zero são ideais).\n",
        "* Ao calcular a Log Loss, é fundamental usar as probabilidades previstas pelo modelo e não as classes previstas.\n",
        "* É importante usar a Log Loss em conjunto com outras métricas de avaliação, como acurácia, precisão, recall e F1-score, para obter uma visão completa do desempenho do modelo."
      ],
      "metadata": {
        "id": "KHkN1SG3YfQT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import log_loss\n",
        "log_loss(y_test, yhat_prob)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "By-wK_g-YArW",
        "outputId": "9b116509-ae28-441d-999f-bb6919f9011a"
      },
      "execution_count": 120,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.60083907009908"
            ]
          },
          "metadata": {},
          "execution_count": 120
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Análise do Resultado:**\n",
        "\n",
        "A métrica Log Loss de 0,6008 ou 60,08% sugere que o modelo tem um desempenho entre bom a razoável, havendo espaço para melhorias pois as previsões probabilísticas não estão perfeitamente calibradas. Quanto menor o Log Loss, mais confiáveis são as probabilidades atribuídas pelo modelo."
      ],
      "metadata": {
        "id": "I62iZZbPaPVk"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Análise das métricas de avaliação (*precision*, *recall*, F1-score, *support* e *accuracy*)**"
      ],
      "metadata": {
        "id": "bav50XQwfLXL"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 122,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rHSq1ZdIZpc5",
        "outputId": "c6a6f7ae-8939-4677-9178-77f330596936"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "         0.0       0.73      0.96      0.83        25\n",
            "         1.0       0.86      0.40      0.55        15\n",
            "\n",
            "    accuracy                           0.75        40\n",
            "   macro avg       0.79      0.68      0.69        40\n",
            "weighted avg       0.78      0.75      0.72        40\n",
            "\n"
          ]
        }
      ],
      "source": [
        "print (classification_report(y_test, yhat))"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Análise dos resultados**:\n",
        "\n",
        "1. **Precision**: foram classificadas corretamente, respectivamente, 73% e 86%, das amostras classificadas como da `classe 0` e como da `classe 1`, sendo portanto melhor em identificar corretamente as amostras da classe 1, quando as prevê.\n",
        "\n",
        "2. **Recall**: foram classificadas corretamente, respectivamente, 96% e 40%, das amostras classificadas como da `classe 0` e como da `classe 1`. O `recall` para a classe 1 é baixo (40%), o modelo pode estar perdendo muitas amostras da `classe 1`.\n",
        "\n",
        "3. **F1-score**: apresenta média harmônica entre precisão e recall,respectivamente, de 83% para a `classe 0` e 55% para a `classe 1`, refletindo o bom recall e a precisão razoável para a `classe 0`.\n",
        "\n",
        "4. **Support**: respectivamente, 25 e 15, amostras pertencem à `classe 0` e a `classe 1` no conjunto de teste.\n",
        "\n",
        "5. **Accuracy**: 75% de todas as amostras foram classificadas, indicando um desempenho razoável.\n",
        "\n",
        "  * `macro avg`: média não ponderada das métricas para cada classe.\n",
        "  * `weighted avg`: média ponderada das métricas para cada classe, considerando o número de amostras em cada classe.\n"
      ],
      "metadata": {
        "id": "yuE0xrwrbpwK"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Conclusões**\n",
        "\n",
        "* A `classe 0` (não churn) possui alta precisão (96%) e um bom F1-score (83%), indicando que o modelo é mais eficaz em identificar clientes que não irão cancelar.\n",
        "* O baixo número de falsos positivos (1) sugere que o modelo raramente prevê `churn` (cancelamento) incorretamente para clientes que permaneceriam.\n",
        "* A `classe 1` (churn) apresenta um recall baixo (40%), indicando que o modelo não está conseguindo identificar adequadamente todos os clientes que cancelarão (9 falsos negativos).\n",
        "* O F1-score da `classe 1` (55%) reflete o desequilíbrio entre precisão e recall, sugerindo que o modelo tem dificuldade em lidar com a classe minoritária."
      ],
      "metadata": {
        "id": "-iauivbtnF7k"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Análise da Matriz de Confusão**\n",
        "\n",
        "A matriz de confusão fornece uma visão detalhada dos erros de classificação do modelo, o que pode ajudar a identificar áreas onde o modelo pode ser melhorado.\n",
        "\n",
        "**Dica**: com base na matriz de confusão é possível calcular as métricas acima para a `Classe 1` (`churn`).\n",
        "\n",
        "* `Precision`: VP / (VP + FP) = 6 / (6 + 9) = 0,4 ou 40%\n",
        "* `Recall`: VP / (VP + FN) = 6 / (6 + 1) = 0,86 ou 86%\n",
        "* `F1-score`: 2 * (Precisão * Recall) / (Precisão + Recall) = 2 * (0,4 * 0,86) / (0,4 + 0,86) = 0,55 ou 55%\n",
        "* `Accuracy`: (VP + VN) / Total = (6 + 24) / 40 = 0,75 ou 75%"
      ],
      "metadata": {
        "id": "82xIv458gVNA"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 115,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 561
        },
        "id": "CuxQQbSOYQga",
        "outputId": "d70fe0d4-1f40-49ac-ce6b-c1bdceb0c2e3"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Matriz de Confusão sem Normalização\n",
            "[[ 6  9]\n",
            " [ 1 24]]\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# Importa as funções 'classification_report' e 'confusion_matrix' do módulo metrics do Scikit-learn\n",
        "from sklearn.metrics import classification_report, confusion_matrix\n",
        "# Importa o módulo 'itertools' para criar iteradores eficientes\n",
        "import itertools\n",
        "\n",
        "# Matriz de confusão calculada\n",
        "def plot_confusion_matrix(cm, classes,\n",
        "                          normalize=False,\n",
        "                          title='Matriz de Confusão',\n",
        "                          cmap=plt.cm.Blues):\n",
        "    \"\"\"\n",
        "    Esta função gera a matriz de confusão.\n",
        "    A normalização pode ser aplicada definindo `normalize=True`.\n",
        "    \"\"\"\n",
        "    if normalize:\n",
        "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
        "        print(\"Matriz de Confusão Normalizada\")\n",
        "    else:\n",
        "        print('Matriz de Confusão sem Normalização')\n",
        "\n",
        "    print(cm)\n",
        "\n",
        "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
        "    plt.title(title)\n",
        "    plt.colorbar()\n",
        "    tick_marks = np.arange(len(classes))\n",
        "    plt.xticks(tick_marks, classes, rotation=45)\n",
        "    plt.yticks(tick_marks, classes)\n",
        "\n",
        "    fmt = '.2f' if normalize else 'd'\n",
        "    thresh = cm.max() / 2.\n",
        "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
        "        plt.text(j, i, format(cm[i, j], fmt),\n",
        "                 horizontalalignment=\"center\",\n",
        "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
        "\n",
        "    plt.tight_layout()\n",
        "    plt.ylabel('Classe Real')\n",
        "    plt.xlabel('Classe Prevista')\n",
        "#print(confusion_matrix(y_test, yhat, labels=[1,0]))\n",
        "\n",
        "# Compute confusion matrix\n",
        "cnf_matrix = confusion_matrix(y_test, yhat, labels=[1,0])\n",
        "np.set_printoptions(precision=2)\n",
        "\n",
        "# Plota a matriz de confusão não-normalizada\n",
        "plt.figure()\n",
        "plot_confusion_matrix(cnf_matrix, classes=['churn=1','churn=0'],normalize= False,  title='Matriz de Confusão')"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Análise dos elementos da matriz**:\n",
        "\n",
        "1. **Verdadeiros Positivos** (VP): 6 casos em que o modelo previu corretamente a classe positiva (\"churn=1\"), ou seja, clientes que realmente iriam cancelar o serviço e foram identificados corretamente pelo modelo.\n",
        "\n",
        "2. **Falsos Positivos** (FP): 9 casos em que o modelo previu a classe positiva (\"churn=1\") incorretamente, ou seja, clientes que não iriam cancelar o serviço, mas foram classificados pelo modelo como se fossem.\n",
        "\n",
        "3. **Verdadeiros Negativos** (VN): 24 casos em que o modelo previu corretamente a classe negativa (\"churn=0\"), ou seja, clientes que realmente não iriam cancelar o serviço e foram identificados corretamente.\n",
        "4. **Falsos Negativos** (FN): 1 caso em que o modelo previu a classe negativa (\"churn=0\") incorretamente, ou seja, clientes que iriam cancelar o serviço, mas foram classificados pelo modelo como se não fossem."
      ],
      "metadata": {
        "id": "wSENY__JhV-F"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Conclusões**\n",
        "\n",
        "* O modelo apresenta melhor desempenho na identificação de clientes que **não irão cancelar o serviço** (`VN = 24`) do que naqueles que irão cancelar (`VP = 6`), ou seja, não é capaz de auxiliar numa estratégia de marketing voltada a reter clientes.\n",
        "\n",
        "* O número de falsos positivos (`FP = 9`) é relativamente alto, o que indica que o modelo está classificando erroneamente alguns clientes como propensos a cancelar o serviço quando na verdade não irão. No mundo real, o modelo poderia sugerir o aumento de custos (desnecessários) com ações de retenção para esses clientes.\n",
        "\n",
        "* O número de falsos negativos (`FN = 1`) é baixo, o que é positivo, pois indica que o modelo está perdendo poucos clientes que realmente irão cancelar. No entanto, dependendo do contexto, mesmo um falso negativo pode ser problemático.\n"
      ],
      "metadata": {
        "id": "EZOnZ03Opobj"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Análise Gráfica do Modelo Regressão Logística**: Fronteira de Decisão e da Curva ROC (*Receiver Operating Characteristic*)\n",
        "\n"
      ],
      "metadata": {
        "id": "1Ik1_4iG6Ubr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.decomposition import PCA\n",
        "\n",
        "# Reduzindo o modelo de regressão logística para duas dimensões usando PCA\n",
        "pca = PCA(n_components=2)\n",
        "X_test_pca = pca.fit_transform(X_test)\n",
        "\n",
        "# Plotar os pontos de dados\n",
        "plt.scatter(X_test_pca[:, 0], X_test_pca[:, 1], c=y_test, cmap='viridis')\n",
        "\n",
        "# Incluir legenda de cores\n",
        "legend1 = plt.legend(*scatter.legend_elements(),\n",
        "                    loc=\"upper right\", title=\"Classes\")\n",
        "plt.gca().add_artist(legend1)\n",
        "\n",
        "plt.xlabel('Classe 0')\n",
        "plt.ylabel('Classe 1')\n",
        "plt.title('Distribuição da Regressão Logística')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "cruCA9G5NUoX",
        "outputId": "a2b6b95e-ba6e-4bee-ee17-3951f98a02e1"
      },
      "execution_count": 142,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import roc_curve, auc\n",
        "\n",
        "#model = LR.fit(X_train, y_train)\n",
        "\n",
        "# Reduzindo para duas dimensões usando PCA\n",
        "pca = PCA(n_components=2)\n",
        "X_test_pca = pca.fit_transform(X_test)\n",
        "\n",
        "# Criar a figura e os eixos para os dois gráficos\n",
        "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(10, 4))\n",
        "\n",
        "# Plotando a fronteira de decisão no primeiro subplot (ax1)\n",
        "def plot_decision_boundary(clf, X, y, ax, pca):\n",
        "    # Gerar uma grade de pontos no espaço 2D\n",
        "    x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1\n",
        "    y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1\n",
        "    xx, yy = np.meshgrid(np.arange(x_min, x_max, 0.02),\n",
        "                         np.arange(y_min, y_max, 0.02))\n",
        "\n",
        "    # Transformar a grade de volta para o espaço original\n",
        "    grid = np.c_[xx.ravel(), yy.ravel()]\n",
        "    grid_original = pca.inverse_transform(grid)\n",
        "\n",
        "    # Obter as previsões para cada ponto na grade\n",
        "    Z = clf.predict(grid_original)\n",
        "    Z = Z.reshape(xx.shape)\n",
        "\n",
        "    # Plotar o contorno da fronteira de decisão\n",
        "    ax.contourf(xx, yy, Z, alpha=0.3, cmap='viridis')\n",
        "\n",
        "# Plotar a fronteira de decisão no primeiro subplot (ax1)\n",
        "plot_decision_boundary(model, X_test_pca, y_test, ax1, pca)\n",
        "\n",
        "# Plotar os pontos de dados no primeiro subplot (ax1)\n",
        "scatter = ax1.scatter(X_test_pca[:, 0], X_test_pca[:, 1], c=y_test, cmap='viridis', edgecolor='k')\n",
        "\n",
        "# Adicionar rótulos e título ao primeiro subplot (ax1)\n",
        "ax1.set_xlabel('Classe 0')\n",
        "ax1.set_ylabel('Classe 1')\n",
        "ax1.set_title('Gráfico 1: Fronteira de Decisão da Regressão Logística')\n",
        "\n",
        "# Adicionar legenda ao primeiro subplot (ax1)\n",
        "legend = ax1.legend(*scatter.legend_elements(), title=\"Classes\")\n",
        "ax1.add_artist(legend)\n",
        "\n",
        "# Plotando a curva ROC no segundo subplot (ax2)\n",
        "yhat_prob = LR.predict_proba(X_test)\n",
        "fpr, tpr, thresholds = roc_curve(y_test, yhat_prob[:, 1])\n",
        "roc_auc = auc(fpr, tpr)\n",
        "\n",
        "ax2.plot(fpr, tpr, label='ROC curve (area = %0.2f)' % roc_auc)\n",
        "ax2.plot([0, 1], [0, 1], 'k--')\n",
        "ax2.set_xlim([0.0, 1.0])\n",
        "ax2.set_ylim([0.0, 1.05])\n",
        "ax2.set_xlabel('Taxa de Falsos Positivos')\n",
        "ax2.set_ylabel('Taxa de Verdadeiros Positivos')\n",
        "ax2.set_title('Gráfico 2: Curva ROC')\n",
        "ax2.legend(loc=\"lower right\")\n",
        "\n",
        "# Ajustar o layout para evitar sobreposição\n",
        "plt.tight_layout()\n",
        "\n",
        "# Exibir o gráfico\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 407
        },
        "id": "LlT2ARfARWOW",
        "outputId": "541d493a-ee3d-40a7-cc27-31e63d182bba"
      },
      "execution_count": 145,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x400 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Análise gráfica**:\n",
        "\n",
        "1. **Fronteira de Decisão** (linha verde): representa o limite que o modelo usa para classificar os clientes como \"churn=1\" (cor amarela) ou \"churn=0\" (cor roxa).A posição da fronteira de decisão é influenciada pelos dados de treinamento e pelos parâmetros do modelo. Idealmente, a fronteira de decisão deveria separar perfeitamente os pontos das duas classes. No entanto, pode-se observar alguns pontos amarelos do lado roxo da linha e vice-versa, o que indica que o modelo cometeu alguns erros de classificação.\n",
        "\n",
        "2. **Curva ROC**: a curva ROC permite visualizar o *trade-off* entre sensibilidade (taxa de verdadeiros positivos) e especificidade (1 - taxa de falsos positivos) do modelo em diferentes limiares de decisão. Quanto mais a curva ROC se aproximar do canto superior esquerdo do gráfico, melhor será o desempenho do modelo, pois ele terá uma alta taxa de verdadeiros positivos e uma baixa taxa de falsos positivos.\n",
        "\n",
        "  * **Área sob a curva (AUC)**: é uma métrica que quantifica o desempenho geral do modelo, a AUC é de 0.77 o que indica um desempenho bom a razoável. Uma AUC de 0.5 representa um modelo aleatório, enquanto uma AUC de 1.0 representa um modelo perfeito.\n"
      ],
      "metadata": {
        "id": "w-3-q57i9nR4"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Ajustes ao modelo de Regressão Logística (LR)**\n",
        "\n",
        "Experimentação de diferentes hiperparâmetros, como a escolha do otimizador (`'solver'`) e parâmetro de regularização (C), buscando variações que apresentem melhor desempenho, especialmente na redução de falsos positivos."
      ],
      "metadata": {
        "id": "DK1A7hsmrpls"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 133,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Z_Aepp07afV6",
        "outputId": "52b2167b-6d3e-4098-a06f-ccbb8a1850aa"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[LR1] Jaccard-Score: 0.705882\n",
            "[LR1] LogLoss: 0.591510\n",
            "[LR1] Relatório de Classificação:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "         0.0       0.73      0.96      0.83        25\n",
            "         1.0       0.86      0.40      0.55        15\n",
            "\n",
            "    accuracy                           0.75        40\n",
            "   macro avg       0.79      0.68      0.69        40\n",
            "weighted avg       0.78      0.75      0.72        40\n",
            "\n"
          ]
        }
      ],
      "source": [
        "# Modelando 'LR1' com hiperparâmetros diferentes\n",
        "LR1 = LogisticRegression(C=0.1, solver='newton-cg').fit(X_train,y_train)\n",
        "yhat2 = LR1.predict(X_test)\n",
        "yhat_prob2 = LR1.predict_proba(X_test)\n",
        "print (\"[LR1] Jaccard-Score: %.6f\" %jaccard_score(y_test, yhat,pos_label=0))\n",
        "print (\"[LR1] LogLoss: %.6f\" % log_loss(y_test, yhat_prob2))\n",
        "print (\"[LR1] Relatório de Classificação:\")\n",
        "print (classification_report(y_test, yhat))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 136,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pQBBzf42ase1",
        "outputId": "535af8ad-3215-4183-8674-8d2a13314596"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[LR2] Jaccard-Score: 0.705882\n",
            "[LR2] LogLoss: 0.591240\n",
            "[LR2] Relatório de Classificação:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "         0.0       0.73      0.96      0.83        25\n",
            "         1.0       0.86      0.40      0.55        15\n",
            "\n",
            "    accuracy                           0.75        40\n",
            "   macro avg       0.79      0.68      0.69        40\n",
            "weighted avg       0.78      0.75      0.72        40\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/sklearn/linear_model/_sag.py:349: ConvergenceWarning: The max_iter was reached which means the coef_ did not converge\n",
            "  warnings.warn(\n"
          ]
        }
      ],
      "source": [
        "# Modelando 'LR2' com hiperparâmetros diferentes\n",
        "LR2 = LogisticRegression(C=0.1, solver='saga').fit(X_train,y_train)\n",
        "yhat2 = LR2.predict(X_test)\n",
        "yhat_prob2 = LR2.predict_proba(X_test)\n",
        "print (\"[LR2] Jaccard-Score: %.6f\" %jaccard_score(y_test, yhat,pos_label=0))\n",
        "print (\"[LR2] LogLoss: %.6f\" % log_loss(y_test, yhat_prob2))\n",
        "print (\"[LR2] Relatório de Classificação:\")\n",
        "print (classification_report(y_test, yhat))"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Referências**\n",
        "\n",
        "1. JAIN, Hemlata; KHUNTETA, Ajay; SRIVASTAVA, Sumit. **Churn prediction in telecommunication using logistic regression and logit boost**. Procedia Computer Science, v. 167, p. 101-112, 2020. Disponível em: https://www.sciencedirect.com/science/article/pii/S1877050920306529?via%3Dihub Acesso em: 11/10/2024.\n",
        "\n",
        "2. DIAS, Camila Alves. Previsão de Churn A3Data Telecomunicações.  https://github.com/cmdiasbr/Desafio-Telecom"
      ],
      "metadata": {
        "id": "ZJjS-rpw2AcT"
      }
    }
  ],
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyODHqn0aGnVYR/jg+Lp4p3P",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}