Projects/Churn_Prediction_para_empresa_de_Telecomunicações.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "[PROJETO - DSNP] Churn Prediction para empresa de Telecomunicações.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyPUPDLlwQ5DcuvnzUaPHfcw",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/matheusbuniotto/portfolio/blob/main/Churn_Prediction_para_empresa_de_Telecomunica%C3%A7%C3%B5es.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0qeZYbxIuJNc"
      },
      "source": [
        "![img](https://github.com/matheusbuniotto/presets/blob/main/sf.png?raw=true) \n",
        "\n",
        "#### **Data Science na Prática**\n",
        "#### *Projeto - Prevendo o Churn em uma empresa de Telecoomunicação*\n",
        "*por Matheus Buniotto*\n",
        "\n",
        "---\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dBfsnEOkuTzb"
      },
      "source": [
        "# Projeto - Churn Rate Prediction\n",
        "> \n",
        "O Churn rate é uma métrica utilizada para detectar a perda de clientes em um determinado período e é muito utilizada em empresas SaaS (Software as a Services) ou que trabalham com produtos ou serviços por assinatura.\n",
        "\n",
        "O cálculo do Churn Rate é representado pela fórmula:\n",
        "\n",
        "![Curn](https://agendor-blog-uploads.s3.sa-east-1.amazonaws.com/2017/06/11181759/churn-rate-03.png)\n",
        "Fonte: [Meet](https://meetime.com.br/blog/gestao-empresarial/reducao-de-churn/)\n",
        "\n",
        "**A importância do Churn Rate**\n",
        "\n",
        "A fidelização de um cliente é fundamental o sucesso das empresas. \n",
        "\n",
        "> “Manter um cliente é muito mais barato do que atrair um novo”\n",
        "\n",
        "Isso faz sentido, afinal, você não precisa gastar tempo e recursos para atrair alguém que já é seu cliente.\n",
        "\n",
        "Hoje, essa máxima é ainda mais importante e as cada vez mais as empresas estão despendendo esforços para que haja uma maior retenção dos consumidores.\n",
        "\n",
        "### **O problema**\n",
        "![Churn](https://ikarus.digital/wp-content/uploads/5d0136e22309e039a15a700f_churn-14-original.jpg)\n",
        "Manter uma taxa de Churn o mais baixo possível representa um aumento significativo nos lucros da empresa. \n",
        "\n",
        "Se você ainda não está convencido, um estudo conduzido por Frederick Reichheld da Bain & Company(o inventor do NPS), revelou que melhorar a retenção de clientes em apenas 5% pode aumentar os lucros da empresa de 25 a 95%!\n",
        "\n",
        "Fonte: [Harvard Business Review](https://hbr.org/2014/10/the-value-of-keeping-the-right-customers)\n",
        "\n",
        "\n",
        "* Você já imaginou se pudessemos prever um cliente que provavelmente cancelará sua assinatura?\n",
        "\n",
        "\n",
        "### **Hipótese**\n",
        "\n",
        "Utilizando dados históricos de clientes e através de algorítimos de Machine Learning podemos prever a probabilidade de um cliente cancelar sua assinatura.\n",
        "\n",
        "Sabendo qual cliente terá maior chance de desistir de sua assinatura podemos direcionar os esforços para mante-lo fiel ao produto ou serviço. \n",
        "\n",
        "Podemos ainda, utilizar os dados do cliente para fazermos uma oferta especial e personalizada à ele, dessa forma podemos aumentar a satisfação do cliente em relação ao serviço ou produto ofertado. \n",
        "\n",
        "> **91%** dos consumidores dizem que são mais propensos a comprar com marcas que fornecem ofertas e recomendações que são relevantes para eles. - *Accenture*\n",
        "\n",
        ">\n",
        "### **O Projeto**\n",
        "\n",
        "Nesse projeto irei utilizar os dados fornecidos originalmente na plataforma de ensino da IBM Developer, e refletem um problema típico de Churn em empresas de telecomunicação.\n",
        "\n",
        "Utilizando os dados, iremos prever os clientes que serão mais propensos à cancelar o serviço. "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gCYglGqV9MrX"
      },
      "source": [
        "## 1. Aquisição dos dados"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kMy4CBVd9S4H"
      },
      "source": [
        "Como citado acima, os dados foram disponibilizados pela plataforma IBM Developer, e podem ser encontrados através do link: [IBM](https://developer.ibm.com/technologies/data-science/patterns/predict-customer-churn-using-watson-studio-and-jupyter-notebooks/#)\n",
        "\n",
        "Apesar de não termos informações explícitas sobre o que cada variável representa, os nomes das colunas nos permite entender o que cada uma representa. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Yb6IedO69MSx"
      },
      "source": [
        "# importar os pacotes necessários\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import warnings\n",
        "warnings.simplefilter(action='ignore', category=FutureWarning)\n",
        "warnings.simplefilter(action='ignore', category=UserWarning)\n",
        "sns.set(style=\"white\")\n",
        "%matplotlib inline\n",
        "\n",
        "\n",
        "# importar os dados\n",
        "PATH = \"https://raw.githubusercontent.com/carlosfab/dsnp2/master/datasets/WA_Fn-UseC_-Telco-Customer-Churn.csv\"\n",
        "df = pd.read_csv(PATH)\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UQZBxaHo-Xx4"
      },
      "source": [
        "## 2. Análise explorátória\n",
        "Nessa etapa iremos entender os dados, sua composição, formatos, estrutura, etc. Além disso, vamos transforma-los em gráficos que facilitem a compreensão e nos ajude a extrair informações relevantes para contrução do nosso modelo."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aVZkG_Vj9_D5"
      },
      "source": [
        " Verificando as 5 primeiras entradas do conjunto de dados. Vamos ver qual é a \"cara\" dos nossos dados."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 309
        },
        "id": "1eftdEaYuCiV",
        "outputId": "60190970-d23b-4d16-9cc5-a3f7837e8d0b"
      },
      "source": [
        "# ver as 5 primeiras entradas\n",
        "df.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>customerID</th>\n",
              "      <th>gender</th>\n",
              "      <th>SeniorCitizen</th>\n",
              "      <th>Partner</th>\n",
              "      <th>Dependents</th>\n",
              "      <th>tenure</th>\n",
              "      <th>PhoneService</th>\n",
              "      <th>MultipleLines</th>\n",
              "      <th>InternetService</th>\n",
              "      <th>OnlineSecurity</th>\n",
              "      <th>OnlineBackup</th>\n",
              "      <th>DeviceProtection</th>\n",
              "      <th>TechSupport</th>\n",
              "      <th>StreamingTV</th>\n",
              "      <th>StreamingMovies</th>\n",
              "      <th>Contract</th>\n",
              "      <th>PaperlessBilling</th>\n",
              "      <th>PaymentMethod</th>\n",
              "      <th>MonthlyCharges</th>\n",
              "      <th>TotalCharges</th>\n",
              "      <th>Churn</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>7590-VHVEG</td>\n",
              "      <td>Female</td>\n",
              "      <td>0</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>1</td>\n",
              "      <td>No</td>\n",
              "      <td>No phone service</td>\n",
              "      <td>DSL</td>\n",
              "      <td>No</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>Month-to-month</td>\n",
              "      <td>Yes</td>\n",
              "      <td>Electronic check</td>\n",
              "      <td>29.85</td>\n",
              "      <td>29.85</td>\n",
              "      <td>No</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5575-GNVDE</td>\n",
              "      <td>Male</td>\n",
              "      <td>0</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>34</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>DSL</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>One year</td>\n",
              "      <td>No</td>\n",
              "      <td>Mailed check</td>\n",
              "      <td>56.95</td>\n",
              "      <td>1889.5</td>\n",
              "      <td>No</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3668-QPYBK</td>\n",
              "      <td>Male</td>\n",
              "      <td>0</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>2</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>DSL</td>\n",
              "      <td>Yes</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>Month-to-month</td>\n",
              "      <td>Yes</td>\n",
              "      <td>Mailed check</td>\n",
              "      <td>53.85</td>\n",
              "      <td>108.15</td>\n",
              "      <td>Yes</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>7795-CFOCW</td>\n",
              "      <td>Male</td>\n",
              "      <td>0</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>45</td>\n",
              "      <td>No</td>\n",
              "      <td>No phone service</td>\n",
              "      <td>DSL</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>Yes</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>One year</td>\n",
              "      <td>No</td>\n",
              "      <td>Bank transfer (automatic)</td>\n",
              "      <td>42.30</td>\n",
              "      <td>1840.75</td>\n",
              "      <td>No</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>9237-HQITU</td>\n",
              "      <td>Female</td>\n",
              "      <td>0</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>2</td>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>Fiber optic</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>Month-to-month</td>\n",
              "      <td>Yes</td>\n",
              "      <td>Electronic check</td>\n",
              "      <td>70.70</td>\n",
              "      <td>151.65</td>\n",
              "      <td>Yes</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   customerID  gender  SeniorCitizen  ... MonthlyCharges TotalCharges  Churn\n",
              "0  7590-VHVEG  Female              0  ...          29.85        29.85     No\n",
              "1  5575-GNVDE    Male              0  ...          56.95       1889.5     No\n",
              "2  3668-QPYBK    Male              0  ...          53.85       108.15    Yes\n",
              "3  7795-CFOCW    Male              0  ...          42.30      1840.75     No\n",
              "4  9237-HQITU  Female              0  ...          70.70       151.65    Yes\n",
              "\n",
              "[5 rows x 21 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 126
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ruiwOzzc-2fI"
      },
      "source": [
        "Tamanho,composição e formato dos dados."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bTKaN8UKuGFb",
        "outputId": "42ac1aae-76e6-4de3-b983-df1708c851b1"
      },
      "source": [
        "df.shape\n",
        "\n",
        "print('O conjunto de dados possui: {:} entradas'.format(df.shape[0]))\n",
        "print('O conjunto de dados possui: {:} variáveis'.format(df.shape[1]))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "O conjunto de dados possui: 7043 entradas\n",
            "O conjunto de dados possui: 21 variáveis\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IzSaz-EC_Bhh",
        "outputId": "b626481f-a264-4c99-a9d3-36b41b2e25f4"
      },
      "source": [
        "df.dtypes"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "customerID           object\n",
              "gender               object\n",
              "SeniorCitizen         int64\n",
              "Partner              object\n",
              "Dependents           object\n",
              "tenure                int64\n",
              "PhoneService         object\n",
              "MultipleLines        object\n",
              "InternetService      object\n",
              "OnlineSecurity       object\n",
              "OnlineBackup         object\n",
              "DeviceProtection     object\n",
              "TechSupport          object\n",
              "StreamingTV          object\n",
              "StreamingMovies      object\n",
              "Contract             object\n",
              "PaperlessBilling     object\n",
              "PaymentMethod        object\n",
              "MonthlyCharges      float64\n",
              "TotalCharges         object\n",
              "Churn                object\n",
              "dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 128
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8Bbm9omQ-7oJ"
      },
      "source": [
        "Vemos que a variável TotalCharges, que representa o valor total de recargas, está em formato *object*. Como a variável representa um valor numérico, vamos realizar uma conversão para tipo numérico.\n",
        "\n",
        "A coluna CostumerID não irá influenciar em nossa análise, por se tratar de uma identificação do cliente e não agregará em nosso modelo."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Qr9JS70WHJ3b"
      },
      "source": [
        "#converter TotalCharges para float64\n",
        "df.TotalCharges = df.TotalCharges.replace(\" \", 0).astype('float64')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "00ZXnY46HTFr",
        "outputId": "affed66f-5278-4c7e-f64c-2f14108c0669"
      },
      "source": [
        "df.TotalCharges.dtypes"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "dtype('float64')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 130
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wR22wG3p_7Bf"
      },
      "source": [
        "df.drop(('customerID'), axis=1, inplace=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "m6tLc-RWAida"
      },
      "source": [
        "Vamos investigar a nossa variável **Alvo - Churn!**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XW9oPIO5_nKj",
        "outputId": "4401b20c-ef9a-475d-829c-32a7aa9a7d37"
      },
      "source": [
        "df['Churn'].unique().tolist()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['No', 'Yes']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 132
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xPoZZ0_4Asbq",
        "outputId": "22f24543-6921-4f90-8769-887d0151ddc6"
      },
      "source": [
        "df['Churn'].value_counts()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "No     5174\n",
              "Yes    1869\n",
              "Name: Churn, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 133
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "t0Xf-rpZEZa6"
      },
      "source": [
        "A variável Churn é composta de por Sim e Não. Mais a frente iremos balancear os Churn para que nosso modelo possa aprender sem nenhum viés. \n",
        "\n",
        "Grande parte das variáveis do conjunto de dados são variáveis categóricas, vamos enteder como elas são compostas:\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qLX87ZA6FGlB"
      },
      "source": [
        "#criar df para ver as categorias das colunas\n",
        "df_categorias = df.drop(['tenure', 'MonthlyCharges', 'TotalCharges'], axis = 1) "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Vao154BeFc26",
        "outputId": "eb016f2e-3ea3-478c-e481-5447e91025fb"
      },
      "source": [
        "for c in df_categorias.columns:\n",
        "  print(c, ':', df[c].nunique(), 'categorias')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "gender : 2 categorias\n",
            "SeniorCitizen : 2 categorias\n",
            "Partner : 2 categorias\n",
            "Dependents : 2 categorias\n",
            "PhoneService : 2 categorias\n",
            "MultipleLines : 3 categorias\n",
            "InternetService : 3 categorias\n",
            "OnlineSecurity : 3 categorias\n",
            "OnlineBackup : 3 categorias\n",
            "DeviceProtection : 3 categorias\n",
            "TechSupport : 3 categorias\n",
            "StreamingTV : 3 categorias\n",
            "StreamingMovies : 3 categorias\n",
            "Contract : 3 categorias\n",
            "PaperlessBilling : 2 categorias\n",
            "PaymentMethod : 4 categorias\n",
            "Churn : 2 categorias\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xTl0REWbF8DC"
      },
      "source": [
        "Vamos criar 2 listas, representando as variáveis binárias e não binárias. Isso facilitará o tratamento das variáveis futuramente."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XfCYQkyZFgWo"
      },
      "source": [
        "categorias_binarias = []\n",
        "categorias_n_binarias = []\n",
        "\n",
        "for c in df_categorias.columns:\n",
        "  if df[c].nunique() == 2:\n",
        "    categorias_binarias.append(c)\n",
        "  else:\n",
        "    categorias_n_binarias.append(c)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "df-5IKhZFzP4",
        "outputId": "4fbf82a3-e769-41ae-8f23-6170b0a845b0"
      },
      "source": [
        "print('Categorias Binárias:')\n",
        "print(categorias_binarias)\n",
        "\n",
        "print('')\n",
        "\n",
        "print('Categorias Não Binárias:')\n",
        "print(categorias_n_binarias)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Categorias Binárias:\n",
            "['gender', 'SeniorCitizen', 'Partner', 'Dependents', 'PhoneService', 'PaperlessBilling', 'Churn']\n",
            "\n",
            "Categorias Não Binárias:\n",
            "['MultipleLines', 'InternetService', 'OnlineSecurity', 'OnlineBackup', 'DeviceProtection', 'TechSupport', 'StreamingTV', 'StreamingMovies', 'Contract', 'PaymentMethod']\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "44II2w3iHYSc"
      },
      "source": [
        "Sabendo que a maioria dos conjuntos de dados aprensentam algum problema em relação a valores nulos ou em brancos, vamos analisar nosso conjunto."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fGZcNoFGG-bi",
        "outputId": "244029a1-8f73-4417-9fda-009c1cfe6744"
      },
      "source": [
        "print('O conjunto de dados possui: {} valores nulos'.format(df.isnull().sum().sum()))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "O conjunto de dados possui: 0 valores nulos\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rEQKM7bnH3Ir"
      },
      "source": [
        "Vamos extrair as informações estatísticas relevantes dos dados númericos "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "id": "mdfRbZ-GH1U7",
        "outputId": "5d280509-29cd-463f-8f7f-cdcc80d2f61f"
      },
      "source": [
        "df[['tenure', 'MonthlyCharges', 'TotalCharges']].describe()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>tenure</th>\n",
              "      <th>MonthlyCharges</th>\n",
              "      <th>TotalCharges</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>7043.000000</td>\n",
              "      <td>7043.000000</td>\n",
              "      <td>7043.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>32.371149</td>\n",
              "      <td>64.761692</td>\n",
              "      <td>2279.734304</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>24.559481</td>\n",
              "      <td>30.090047</td>\n",
              "      <td>2266.794470</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>18.250000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>9.000000</td>\n",
              "      <td>35.500000</td>\n",
              "      <td>398.550000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>29.000000</td>\n",
              "      <td>70.350000</td>\n",
              "      <td>1394.550000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>55.000000</td>\n",
              "      <td>89.850000</td>\n",
              "      <td>3786.600000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>72.000000</td>\n",
              "      <td>118.750000</td>\n",
              "      <td>8684.800000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "            tenure  MonthlyCharges  TotalCharges\n",
              "count  7043.000000     7043.000000   7043.000000\n",
              "mean     32.371149       64.761692   2279.734304\n",
              "std      24.559481       30.090047   2266.794470\n",
              "min       0.000000       18.250000      0.000000\n",
              "25%       9.000000       35.500000    398.550000\n",
              "50%      29.000000       70.350000   1394.550000\n",
              "75%      55.000000       89.850000   3786.600000\n",
              "max      72.000000      118.750000   8684.800000"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 139
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "omexFCQxIfvs"
      },
      "source": [
        "Anteriormente descobrimos que a variável alvo (aquela que estamos tentando prever) está desbalanceada. Vamos analisar melhor:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 355
        },
        "id": "Nw9XbY8NIbC4",
        "outputId": "5d44b80f-7989-4caf-8e9f-d7dc151a9e00"
      },
      "source": [
        "ax = sns.countplot(x=\"Churn\", data=df)\n",
        "plt.title('Balanceamento dos dados - Churn')\n",
        "ax.set_yticklabels(['{:.0%}'.format(num) for num in ax.get_yticks()/len(df)])\n",
        "plt.show()\n",
        "\n",
        "churn_yes = df.Churn.value_counts()[1]/df.shape[0]\n",
        "churn_no = df.Churn.value_counts()[0]/df.shape[0]\n",
        "\n",
        "print('\\n')\n",
        "print('Os clientes Churn representam {:,.2%} do total'.format(churn_yes))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZAAAAEcCAYAAADpzeJvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAaaElEQVR4nO3deZhcVZ3/8XcSNkOQLUhUEHDhqwji4IYboILiiAojo0LQIPwYhVFcRqO4QMTAAA9rEEd+EBUJriNkFOOwjMqAwk9QQZHxiyCJCRAmBEFEIOvvj3Nbi6I76TrdXdWh36/nqaf7nnOXU5VOfeqcc+vecatXr0aSpE6N73UDJEnrJgNEklTFAJEkVTFAJElVDBBJUhUDRJJUxQDRoETE9hGxOiLW63Vbnmgi4tCIuGaEj7FXRCyq3HZU/ttHxFciYmav2zGWjao/CI2siJgPbA2sBJYDPwXel5kLe9isdU5EzACenZmH9LotT2QRMQ74APBPwA7AH4FrgeMz89e9bJsKeyBjz5szcxLwVOAe4Owet0cayFnAB4GjgS2AHYG5wJuG+0CjrXe1rvBFG6My85GI+HfgzL6yiHgTMBN4FvAAMDszZ/S3fUS8B5gObAMsAU7OzHObur2AOcAZwMcpPZ5PZuaXm/onNcc5ENgM+DWwT2Y+HBG7A6cDOwELgA9m5o87OOYs4KPNMY8EljXPcTJwamae2Kw/vtnXEU0b/ovSG7svIrYH7gAOBT4HTATOyMwTImJf4JPAuIjYH7g9M3eNiKcBXwReBdzXtO28AV67LYEvA3sBvwUua6t/BeXNc0fg1uY1+GlTdyhwLLAVcC/w6cy8qJ9jPAn4N+CtwN3N8VrrP9E896cAC4FPZeYlTd0E4OTm+f8JOK1t2wGfa0S8FPhC0/aHgYsy8yP9vQ5rEhHPAf4ZeHlm/qylqv25bh4R3wf2AG4BDs7M21v+DdfPzBXNPn8MzMnM85vX8QjgZ8C7gX+LiG2Ah4Dt2/fXafvHCnsgY1RETATeAVzXUvwQ5T/TZpRPeUc2b5L9+V9gP+DJwHuAMyJit5b6KcCmwNOBw4FzImLzpu5U4EXAKyifLKcDqyLi6cD3KeGyBSUIvhMRW3VwzI2aYx4LnAcc0hzr1cBnImKHZt0PAPsDewJPowyPnNP2HF8FBPA64NiIeF5m/idwIvDNzJyUmbs2634DWNTs60DgxIh47QCv3TnAI5Re4GHNA4CI2KJ5DWYBW1LC9PsRsWVEbNyUvzEzN2levxsHOMZxlA8CzwLeAExrq7+9eU02BT4LzImIpzZ1R1Be578DXtw8n1Zreq5nAWdl5pObY39rgPatzeuARW3h0Z93Nu3fHLgNOKGDY7wM+D1lWLdvu6Hsb8yxBzL2zI2IFcDGlE/xb+ir6Puk3/hVRHyd8gY7t30nmfn9lsWrIuJyyhvSL5qy5ZSx6hXAvIj4MxAR8TPKG+bumXlns27fp+tDgHmZOa8pvyIibgD+HrhgkMc8ITNXRsQ3gP9LeTN7EPhNRNwC7Er5ZPo+4P2Zuag59gzgDxHxrpZjfDYzHwZuioibmm3/p/21iIhtgVcCb8rMR4AbI+J8Shj/sG3dCcDbgF0y8yHg5oi4gPKJF0pw/y4zL2yWvx4RRwNvBr4NrAJ2jog/ZObdlN5Ff94OHJWZ9wH3RcQsSqgCkJnfbln3mxFxDPBS4D+abc/smxuLiH+l9JYG81yXA8+OiMmZeS+P/YDSiS3X8NxaXdIXMhFxESVwB+uuzOwbwl0REUPd35hjgIw9+2fmlc0b2Vspb8Q7ZebiiHgZcBKwM7ABsCHlTetxIuKNlE+5O1J6shMpQ1F9lvYNHTT+AkyiDCVtRPkE3G474B8j4s0tZesDP+rgmCub3x9uft7TUv9w04a+Y10SEata6ldSPo32WdxP+/vzNOC+Jqj6LKB8em+3FeX/3cK2dVv3tYDHWgA8PTMfioh3UHpmsyPiJ8C/ZOZvB2jTQMcgIt4NfIQyXAN/+7dZ27Zre66HA8cDv42IOyghfGl74yLiB5TwB3hvP8NwSyk9tLUZ7L9Rf/o7eWQo+xtzHMIaozJzZWZeTHnTfFVT/DXgu8C2mbkpZZx7XPu2EbEh8B3KUNTWmbkZMK+/dftxL2X45ln91C0ELszMzVoeG2fmSUM8Zn8WUoaCWo+1UUuvaE3aL2F9F7BFRGzSUvYMoL99LQFWANu2rdu6r+3atvnrvjLzsszch/Lm+lvKMF1/7h7oGBGxXbPd+4Etm9fyZv72Wg64LWt5rpn5u8w8iDK3cjLw783Q22Nk5hubIcBJ/c3hUOaktomI/kJ4MB5qfk5sKZvSto6XIh8ieyBjVHOK5FsoY719wzKbUD5dPtJMhh4MXN7P5n29kyWUrv8bgddT3oTWKDNXRcSXgNOb4aJ7KEMnv6BMgl8fEW8ArqT0PnanjEU/UHvMAXwROCEipmXmgmae5RWZ+R+D2PYeYJ+IGJ+ZqzJzYUT8FPjXiPgopYd0ODC1n+e/MiIuBmZExGGUHsA0YH6zyjzg7Ig4mDJ/8DbKCQWXRsTWlNfjSkpv6s+UIa3+fAs4JiL+H2W48gMtdRtT3jyXwF9PTti5bdujI+JSyhvxJ1rav8bn2gxDXpaZSyLi/mazgdo4oMz8XUR8gTKEdwRlmHM8Zd5q+8w8aS3bL4mIO4FDIuJcymvc34cWDYE9kLHne818xJ8oE4TTMvM3Td1RwPER8SBlvLzfCdBm+OLopv6PlKD5bgdt+Chl6Ol6mrN4gPHNmPtbKWc5LaH0Ej7W1A31mO3Oara/vHm+11EmVQejb1hvaUT0zb8cRAmDu4BLgOMy88oBtn8/ZWhkMfAVWs6QysyllAnsf6EM40wH9mvmE8ZThp3uorxue1LONOvPZylDS3dQPgT0zamQmbdQzqy6lhKGuwA/adn2PMqZYTdRgv3itn2v6bnuS5lv+jPlNX5nM49U42jg85STDu6nDHseAHxvkNsfQfn7WQo8n2auTcNnnDeUkiTVsAciSapigEiSqhggkqQqBogkqcqYOI23+Q7BSyjnt69cy+qSpGIC5TtH12fmo+2VYyJAKOFxda8bIUnrqFcDj7tnzVgJkLsBLrroIqZMaf8yqiSpP4sXL2bq1KkwwHXJxkqArASYMmUK22yzTa/bIknrmn6H/p1ElyRVMUAkSVUMEElSFQNEklTFAJEkVTFAJElVDBBJUhUDpAPLlnsVFD2efxcaq8bKFwmHxQbrT+Dg6f3dvllj2ddOedyda6UxwR6IJKmKASJJqtKVIayI2B6Y21K0GfDkzNwiIk4F3gZsD+ySmTc326wPfBvYAbgdeHtmroiIycDFwN6Zuawb7ZckPV5XeiCZOT8zX9j3oITJ15rqucAewIK2zd4A3JeZuwL3A/s25acAnzI8JKm3uj6JHhEbAFMpAUFmXtOUt6+6HJjY/D4RWBYRewKrMtN7e0hSj/ViDuQtwJ2Z+Yu1rHcF8GBE3AQ8APw38Dlg+gi3T5I0CL04jfcw4EtrWykzVwFH9C1HxLHA+cB2EXFuUzwzM28akVZKktaoqz2QiHg6sCfQ0ZcpIuI5wO6Z+VVgFvAxSk9k1rA3UpI0KN3ugUwDvp+ZSzvc7gzgQ83vGwOrm8ekYWybJKkD3Q6QQ4GjWwsiYhbwD8AU4MqIWJqZz2+pPwS4PjNvbYqOBeY1v39sxFssSepXVwMkM3fsp+xo2kKlrX5O2/KlwKXD3zpJUif8JrokqYoBIkmqYoBIkqoYIJKkKgaIJKmKASJJqmKASJKqGCCSpCoGiCSpigEiSapigEiSqhggkqQqBogkqYoBIkmqYoBIkqoYIJKkKgaIJKmKASJJqmKASJKqGCCSpCoGiCSpigEiSapigEiSqhggkqQqBogkqYoBIkmqYoBIkqoYIJKkKgaIJKmKASJJqmKASJKqrNetA0XEfOCR5gHwceBB4Astqz0FWJyZu0XE5sDFwGTg6sw8qtlPAKdl5n5darokqR9dC5DGgZl5c1vZC/t+iYi5wDXN4lTgR5l5fET8MCJ2brY9Azi6O82VJA1k1AxhRcRTgNcDFzZFy4GJETEe2BBYFhHTgGsz87YeNVOS1Oh2D+SiiBhH6WV8MjPvb6l7N3B5Zt7TLM8BLgB+CcwFlgKHA3t3sb2SpAF0M0BenZkLI2JD4Ezg88AhLfXvAY7pW8jMh4AD+5YjYjbwGWCPiDgSeBQ4JjMXdKPxkqTH6toQVmYubH4+Spk4f2VfXUTsDmwBzOtv24jYo9n2KuBs4FDgPOD4EW20JGlAXemBRMTGwHqZ+UAzhPVO4MaWVQ4DLszMFf1suwEwEzigKZoIrGoek0a04ZKkAXVrCGtr4DsRMQGYANwC9J2W+yTgHcDLBth2OjA7M5c2yzOBG4BllDkRSVIPdCVAMvP3wN8NUPcwsOkatp3ZtnweZfhKktRDo+Y0XknSusUAkSRVMUAkSVUMEElSFQNEklTFAJEkVTFAJElVDBBJUhUDRJJUxQCRJFUxQCRJVQwQSVIVA0SSVMUAkSRVMUAkSVUMEElSFQNEklTFAJEkVTFAJElVDBBJUhUDRJJUxQCRJFUxQCRJVQwQSVIVA0SSVMUAkSRVMUAkSVUMEElSFQNEklTFAJEkVTFAJElV1hvsihHx0cw8tZ/yj2Tm6R3s5zhgBrALcAvwE2BiU3038L7MnB8RmwMXA5OBqzPzqGb7AE7LzP0Ge0xJ0vDrpAdy7ADlnx7sDiJiN2B3YAFAZq4C9s3MXTNzV+AHQF8YTQV+lJm7AM+NiJ2b8jOAD3XQbknSCFhrDyQiXtv8OiEiXgOMa6l+JvDgYA4UERsC5wAHAT/uK8/MB1pWezKwqvl9OTAxIsYDGwLLImIacG1m3jaYY0qSRs5ghrBmNz83Ar7UUr4aWAx8YJDHOh6Y0wxPPaYiIuYBuwH3Aq9viucAFwC/BOYCS4HDgb0HeTxJ0ghaa4Bk5g4AEfHVzHx3zUEi4uXAi4FPDHCMv296GsdQhsSOysyHgANb9jEb+AywR0QcCTwKHJOZC2raJEkamkHPgbSGR0SMb30MYvM9gecBd0TEfGAb4LKI6Ott9M2HzAbe1b5xROzRrHMVcDZwKHAepVcjSeqBTs7C2o0yh/ECynAWlPmQ1cCENW2bmScBJ7Xsaz6wH3BPREzOzHubqn8Eft123A2AmcABTdFEyjzJKmDSYNsvSRpegw4QynzE94DDgL8M0/GnAF+JiPUpYXQHcEjbOtOB2Zm5tFmeCdwALKPMiUiSeqCTANkO+FRmrh7qQTNz+5bFF61l3Zlty+dRhq8kST3UyfdALuFvZ0hJksa4TnogGwGXRMQ1lNN3/6r27CxJ0rqrkwC5pXlIkjT4AMnMz45kQyRJ65ZOTuN97UB1mfnD4WmOJGld0ckQ1uy25a2ADYBFlGtiSZLGkE6GsHZoXY6ICZTLjgzqYoqSpCeW6htKZeZK4ATKF/0kSWPMUO9IuA9/u/y6JGkM6WQSfSHluld9JlK+G3LUcDdKkjT6dTKJ3n6NqoeAWzPzT8PYHknSOqKTSfSroFzKHdgauKe5BLskaQwa9BxIRGwSEV8FHgbuBB6OiAsiYtMRa50kadTqZBL9bGBjYBfgSc3PicCsEWiXJGmU62QOZF/gmZnZdy+QWyPiPcDtw98sSdJo10kP5BHKt89bTabcm1ySNMZ00gM5H7giIk4HFlBuMPVhvLmTJI1JnQTICZTJ86nA04C7gFMys/0aWZKkMaCTIayzgMzMvTNzp8zcG/ifiDhzhNomSRrFOgmQg4Ab2sp+Dhw8fM2RJK0rOgmQ1cCEtrIJHe5DkvQE0cmb/9XA55pvovd9I31GUy5JGmM6mUT/IHApcHdELACeAdwNvHkkGiZJGt06uRbWoojYDXgpsC2wEPiZ18OSpLGpkx4ITVhc1zwkSWOYE+CSpCoGiCSpigEiSapigEiSqhggkqQqBogkqUpHp/HWiogtgQuBZwHLgN8B783MJRFxGOWy8CuBFcCHM/PqiNgcuJhyz5GrM/OoZl8BnJaZ+3Wj7ZKk/nWrB7Kacun3yMxdKHcxPKkJljOBvTPzhcDxwLnNNlOBHzXrPzcidm7KzwA+1KV2S5IG0JUeSGbeB/y4peg64EhgXPPYBLgH2AxY1KyzHJjYXHNrQ2BZREwDrs3M27rRbknSwLoSIK2aQDgS+G5m3hsR7wV+ERH3U3pEezWrzgEuAH4JzAWWAocDe3e7zZKkx+t6gABnA38GPh8RTwbeD7wkMzMi3g5cEhEvyMyHgAP7NoqI2cBngD0i4kjKvdiPycwF3X8KkqSunoUVEacCzwHe0VxX6/XA/ZmZAJn5LcpE++S27fZo6q+iBNChlHuxH9+1xkuSHqNrARIRJwIvAvbPzEeb4juA3SLiKc06rwH+BNzbst0GwExgelM0EVjVPCZ1p/WSpHbdOo33+cAxwK3AT8uZuNyRmQdExCnAVRGxjDIsdWBmrm7ZfDowOzOXNsszKbfWXUaZE5Ek9UC3zsL6DeVsq/7qTgdOX8O2M9uWz6MMX0lqrFqxnPHrrd/rZmiUGem/i15MoksaZuPXW5+fn/J/et0MjTIvmn7+iO7fS5lIkqoYIJKkKgaIJKmKASJJqmKASJKqGCCSpCoGiCSpigEiSapigEiSqhggkqQqBogkqYoBIkmqYoBIkqoYIJKkKgaIJKmKASJJqmKASJKqGCCSpCoGiCSpigEiSapigEiSqhggkqQqBogkqYoBIkmqYoBIkqoYIJKkKgaIJKmKASJJqmKASJKqGCCSpCrrdetAEXEq8DZge2CXzLy5rf44YEZfXURsDlwMTAauzsyjmvUCOC0z9+tW2yVJj9fNHshcYA9gQXtFROwG7N5WNxX4UWbuAjw3InZuys8APjTCbZUkrUXXAiQzr8nMhe3lEbEhcA5wZFvVcmBiRIwHNgSWRcQ04NrMvG3EGyxJWqPRMAdyPDAnM+e3lc8Bng38ErgSWAocDpzc1dZJkvrVtTmQ/kTEy4EXA59or8vMh4ADW9adDXwG2CMijgQeBY7JzMcNiUmSRl6veyB7As8D7oiI+cA2wGUR8frWlSJiD4DMvAo4GzgUOI/Se5Ek9UBPeyCZeRJwUt9yEyL7tZ6hFREbADOBA5qiicCq5jGpW22VJD1WN0/jnQX8AzAFuDIilmbm8wex6XRgdmYubZZnAjcAyyhzIpKkHuhagGTm0cDRa1ln+37KZrYtn0cZvpIk9VCv50AkSesoA0SSVMUAkSRVMUAkSVUMEElSFQNEklTFAJEkVTFAJElVDBBJUhUDRJJUxQCRJFUxQCRJVQwQSVIVA0SSVMUAkSRVMUAkSVUMEElSFQNEklTFAJEkVTFAJElVDBBJUhUDRJJUxQCRJFUxQCRJVQwQSVIVA0SSVMUAkSRVMUAkSVUMEElSFQNEklTFAJEkVVmv1w0AiIgdgQuALYGlwLuB+cC3gR2A24G3Z+aKiJgMXAzsnZnLetNiSdJo6YF8ETgnM3cEzgHOBd4A3JeZuwL3A/s2654CfMrwkKTe6nkPJCKeAuwG7NMUfR34PLAxMLEpmwgsi4g9gVWZeXWHh5kAsHjx4iG399G/3D/kfeiJZdGiRb1uAgBLHnyk103QKDPUv82W98wJ/dX3PECAbYE7M3MlQGaujIi7gN8BD0bETcB1wH8DlwP7VxzjqQBTp04dnhZLLV53xaxeN0Hq37deN1x7eiplKuExRkOADCgzj+j7PSKOBc4HtouIc5vimZl50yB2dT3wauBuYOWwN1SSnpgmUMLj+v4qR0OALASeHhETmt7HBOBpTTkAEfEcYPfMPD4irgbeBYwDvgLsubYDZOajwDUj0XhJeoJ7XM+jT88n0TPzf4EbgYOaooOAX2bmkpbVzgA+1Py+MbAaWAVM6lY7JUmPNRp6IADvAy5ohqn+SDmNF4CIOAS4PjNvbYqOBeY1v3+sq62UJP3VuNWrV/e6DZKkdVDPh7AkSesmA0SSVMUAkSRVMUAkSVUMEK1RRMyPiJsjYnxb2c69bJfGpoj4UkSc3FZ2ZUQc2as2jWUGiAZjEuXLm1KvfRh4e0S8DCAi3kv5XtgXe9qqMcrTeLVGETEfmEH5/s1zM3NZU7Yf8AjlyslbASuAT2bmf/ainRo7ImIf4CzKdfH+C3gVcBrwDOBJwNcz88Sm1/x54LXAo8CfM/OVvWn1E5M9EA3GDcDPgfZhgouAr2XmC4BDgDkRsVW3G6exJTOvAK6iXJ/pOMo18mZl5kuBFwFvbEJmV+A1wE7NbSH261GTn7AMEA3Wp4GPR0Tf5WPGAS8EvgyQmbdQLkmze2+apzHmVMqFUb8J7AXMiogbgZ9RrqX3POD3wPrA7IhwCHYEjJZLmWiUy8yMiHnAR3rdFokSHqsoH4JXAy/JzOXtK0XE8ykBszdwckTslplDvzGQAHsg6swM4J+BTSj/aW8EpgFExPMoQwbX9apxGnsy80HgauATfWURsW1ETGmGUydm5mVN/QPAM3vT0icmA0SDlpmLgAuBLZqiqcAhEfErynzIu9quoix1w1Rgp4j4dUT8mjKstRnlZnVXNjel+xXwA/yAM6w8C0uSVMUeiCSpigEiSapigEiSqhggkqQqBogkqYoBIo2giJgREXN63Q5pJPhNdGkYRMTBlG/pPxd4kPIlyxN62ihphNkDkYYoIj4CnAmcCGxNuSrsF4C3DvNx/MCnUcUvEkpDEBGbAncC78nMb/dTPwPYiXLp+wOAPwDTMvOGpn418JzMvK1Z/gqwKDM/HRF7AXOAsyn3wbgCuH1N+5O6yR6INDQvBzYCLlnDOm8BvkG5vMZ3KfeoGKwplEvHbAf80zDsTxo2Bog0NFsC92bmijWsc01mzsvMlZRrie3awf5XAcdl5qOZ+fAw7E8aNgaINDRLgclrmZ9ovXz4X4CNOpjPWJKZjwzj/qRhY4BIQ3Mt5Xap+1du/xdgYsvylLZ6Jyk1avmpRRqCzHwgIo4FzomIFcDlwHLKDYxeQwmINbkRODgifgPsA+xJuYWwNOrZA5GGKDNPo3wH5NPAEmAh8H5g7iA2/yDwZuB+yn0tBrONNCp4Gq8kqYo9EElSFQNEklTFAJEkVTFAJElVDBBJUhUDRJJUxQCRJFUxQCRJVQwQSVKV/w9ohuM9/6TeEQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "\n",
            "Os clientes Churn representam 26.54% do total\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sBzJtwb9Ivcz"
      },
      "source": [
        "Agora, vamos partir para análise de outras variáveis do conjunto de dados para extrairmos informações relevantes, como por exemplo a correlação entre algumas variáveis ou os clientes mais propensos à cancelarem sua assinatura."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 821
        },
        "id": "NlQ_PM7LI7O0",
        "outputId": "7bb4a313-fbc1-4b0b-95a2-ec5860681c24"
      },
      "source": [
        "def plotar(var):\n",
        " # Função para plotar\n",
        " # gráficos de distribuição entre as variáveis e a Churn sim/não\" \n",
        "\n",
        "  plt.figure(figsize=(8, 4))\n",
        "  p1=sns.kdeplot(df[df['Churn'] == 'Yes'][var], shade=True, color=\"r\", label='Churn = Sim', linewidth=0)\n",
        "  p2=sns.kdeplot(df[df['Churn'] == 'No'][var], shade=True, color=\"#2d5973\", label='Churn = Não', linewidth=0)\n",
        "  plt.legend()\n",
        "plotar('tenure')\n",
        "plotar('MonthlyCharges')\n",
        "plotar('TotalCharges')\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7Nj3feHDJpuh"
      },
      "source": [
        "Através dos gráficos acima vemos que a maioria dos cancelamentos ocorrem nos primeiros meses de contrato do serviço. Além disso, pessoas que gastam mais com as recargas estão mais propensas ao Churn. \n",
        "\n",
        "Veja como após um processamento inicial dos dados já temos informações valiosas para redução do Churn. Nesse exemplo, a empresa poderia implantar benefícios extras para pessoas com gastos maiores e nos primeiros meses de contrato.\n",
        "\n",
        "Após isso, seria necessario acompanhar os indicadores para entender se as medidas foram efetivas e otimizar a estratrégia.\n",
        "\n",
        "Vamos ver o boxplot das variáveis numéricas.\n",
        "\n",
        "* Boxplot é uma ferramenta gráfica para representar a variação de dados observados de uma variável numérica por meio de quartis. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 244
        },
        "id": "-HjgEkgTLRjP",
        "outputId": "71b28219-eafc-4c98-94f5-25aa86e611b5"
      },
      "source": [
        "fig, (ax1,ax2,ax3) = plt.subplots(1,3,figsize=(14,5))\n",
        "\n",
        "sns.boxplot(data=df, x='tenure', color='b', ax=ax1)\n",
        "sns.boxplot(data=df, x='TotalCharges', color='g', ax=ax2)\n",
        "sns.boxplot(data=df, x='MonthlyCharges',color='y' , ax=ax3)\n",
        "\n",
        "plt.tight_layout()\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-1-ef3c759a55b6>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mfig\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0max1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0max2\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0max3\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msubplots\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m3\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mfigsize\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m14\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m5\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0msns\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mboxplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'tenure'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcolor\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'b'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0max\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0max1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0msns\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mboxplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'TotalCharges'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcolor\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'g'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0max\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0max2\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0msns\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mboxplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'MonthlyCharges'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mcolor\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'y'\u001b[0m \u001b[0;34m,\u001b[0m \u001b[0max\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0max3\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'plt' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rjWa24sxOWCC"
      },
      "source": [
        "Através do boxplot não detectamos nenhum outlier evidente. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "p36R9sf5Mvap"
      },
      "source": [
        "def barplot_porcentagens(var, orient='v', axis_name=\"clientes\"): \n",
        "    #função para criar os gráficos com porcentagens\n",
        "    p = df.groupby(var)[\"Churn\"].value_counts().to_frame()/df.shape[0]\n",
        "    p = p.rename({\"Churn\": axis_name}, axis=1).reset_index() \n",
        "\n",
        "    ax = sns.barplot(x = var, y= axis_name, hue='Churn', data=p, orient=orient)\n",
        "    ax.set_yticklabels(['{:.0%}'.format(num) for num in ax.get_yticks()])\n",
        "    ax.plot()\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 285
        },
        "id": "iw6M56ukNBLw",
        "outputId": "1d2ae28a-2e09-495b-ff7a-9e306359d6e7"
      },
      "source": [
        "barplot_porcentagens('SeniorCitizen')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZAAAAEMCAYAAADqG+D0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAckElEQVR4nO3dfZQedX338fcm2RA2CRhIECWKUM0XbsiNEChYICBFsAWKqZSKPFbUBqpAbyEQkFghVAzRtECURzGWAC0WtCDeVR5Kwo1E4ABHK3x5EGLCg+SBUAwhu8nu/cc1GzZLNtlr2L2u3ez7dc6ea+Y385v57p6c65OZ+c1MQ1tbG5IkVWtQvQuQJPVPBogkqRQDRJJUigEiSSrFAJEklTKk3gXUQkRsAewDvAysrXM5ktRfDAbeBzycmas7LxwQAUIlPObXuwhJ6qcOBB7o3DhQAuRlgLlz57L99tvXuxZJ6hdeeeUVjj/+eCi+QzsbKAGyFmD77bdn7Nix9a5FkvqbDZ769yK6JKkUA0SSVMpAOYUlSaW1trayePFiVq5cWe9SekVjYyPbbbcdW221VVX9DBBJ2oSlS5fS0NBARDBo0OZ14qatrY1Vq1bx4osvAlQVIpvXX0KSesGKFSt473vfu9mFB0BDQwNNTU3ssMMOvPrqq1X13fz+GpLUw9auXUtjY2O9y+hVW265JS0tLVX1MUAkqRsaGhrqXUKvKvP7GSDd0Nzi00868u8hlXfFFVdw9tln17uMHuFF9G4Y2jiYz06ZW+8y+oybZhxf7xKkPu+OO+7ghhtu4Pnnn2f48OHssssuTJ48ud5l9SgDRJJ62A033MA111zD17/+dQ444AAaGxuZP38+99xzD01NTT22nzVr1jBkSP2+xj2FJUk96I033uDyyy9n2rRpHHbYYTQ1NdHY2MghhxzCueeeC0BLSwtTpkxhzz335IgjjuBXv/rVuv4RwcKFC9fNn3feecyaNQuABQsWMHHiRK655hr2339/pk6dyhVXXMGZZ57Z5fZ6kwEiST3oscceY/Xq1XziE5/ocp17772XI444gkceeYRDDjmEiy++uNvbX7p0Ka+//jr33Xffun7vZnvvhgEiST1oxYoVjBo1aqOnliZMmMBBBx3E4MGDOfroo3nqqae6vf1BgwZxxhlnMHToUIYNG/aut/duGCCS1IPe85738Nprr7FmzZou1xk9evS66WHDhrF69eqNrt/RqFGj2GKLLXpse++GASJJPWjPPfdk6NCh3H333aX6b7nllqxatWrd/JIlS9Zb3pfuRzFAJKkHjRw5kjPOOIOLLrqIu+++m1WrVtHS0sL999/PjBkzNtl/l1124c4772Tt2rXMmzePhx9+uAZVl2OASFIP+9znPsd5553Hd77zHT72sY9x8MEHM3fuXA499NBN9r3gggu477772Hvvvbnjjju61adeGtra2mqyo4gYBswCDgXeAn6RmV+MiHHAHGBbYBlwUmY+ExGNwK3ATsBzwLGZuSYiRgO3AYdmZnM39/0h4Pl77rmn9BsJvZHwbd5IqIHmySefZNddd613Gb2u8++5ePFi/vRP/xRgp8x8ofP6tTwCmUElOMZl5njgwqL9KmB2Zo4DZgNXF+2HA8szcw9gBfDJDtu5oLvhIUnqHTW5hTEiRgAnAWMzsw0gM38fEdsBewHtA6ZvBq6MiDFAC9B+y2YT0BwRBwGtmTm/FnVLkrpWq3vg/4jK6amvRcTHgT8AXwVWAS9m5lqAzFwbES8BHwB+DhwTEU8ADwHzgJ8Bn6pRzZKkjahVgAwGdgYey8xzImJf4A7gr7rqkJmtwBfa5yNiGnAdsGNEtJ/mmp6ZT/Re2ZKkrtTqGsjvgDVUTlGRmQuApVSOQHaIiMEAxef7gUUdO0fER4D9MvMHwOXAOcCUYlqSVAc1CZDMXArcR3Gtoxh5tR3wNPA4cFyx6nFUjlKWdNrELOCsYno40Aa0AiN6t3JJUldq+RzgycD3IuJbVC6Qn5iZKyJiMjCnOEX1GpWL7etExAnAw5n5dNE0DbirmD6nNqVLkjqrWYBk5m+BgzfQ/hSw70b63dhp/k7gzp6uT5JUHe9El6Qq9dZrnbu73UMOOYQjjzyS1tbW9dqefvrpjfTqeb6RUJKq1Fuvua7mKQ9vvvkmP/7xj5k0aVKP19FdHoFIUj/0pS99iSuvvJLm5vUfyrFw4UJOPvlkjjrqKCZNmsS8efN6rQYDRJL6od13353ddtuNm2++eb32s88+myOPPJI77riDyy67jHPOOYfly5f3Sg0GiCT1U2eddRbXXnstK1euBKCtrY0nn3yST3/60wB8+MMfZtddd+Xxxx/vlf0bIJLUT+28884cdNBB3HDDDXXZvwEiSf3Yl7/8ZW666SZWrlxJQ0MDu+66K7fffjsAzz33HE899RQf/ehHe2XfjsKSpCo1t6ztlffiNLesZWjj4Kr6bL/99hx99NF873vfA2DmzJlMmzaN73//+wwZMoQZM2awzTbb9HitYIBIUtWq/ZLv6e3ee++9682fe+65nHvuuevm58yZ06N1dcVTWJKkUgwQSVIpBogkqRQDRJJUigEiSSrFAJEklWKASFKVWte01G27U6dO5bLLLluv7ZRTTuGmm27qlZo2xvtAJKlKg4Y08uiMz/f4didMuW6T65x//vkcffTRHHbYYeyxxx7ccsstNDQ0cNxxx22yb08zQCSpHxk5ciQXX3wxU6dOZfbs2Xz3u9/lpptu4swzz+Sll15i9erVHHHEEUyePJnW1lYuuugiHnroIYYOHUpTUxO33HJLj9VigEhSP7P//vuzzz77cMwxxzB16lS++tWvcvrpp7PPPvvQ3NzMKaecwvjx4xk1ahQLFizgrrvuYtCgQbz++us9WocBIkn90KmnnspPf/pT/vzP/5yvfe1r673zY+XKlTz33HNMmjSJNWvWcMEFF7Dvvvvy8Y9/vEdrMEAkqR8aNGgQDQ0NtLa20tDQwA9/+EMaGxvfsd5PfvITFixYwIMPPsjMmTO5/fbbGTNmTM/U0CNbkSTVxYgRI5gwYQLXXHPNuraXX36ZJUuWsHz5clatWsWBBx7I2WefzciRI1m0aFGP7dsjEEnq52bOnMk3vvENjjrqKACGDx/OJZdcwltvvcWFF17ImjVrWLt2LRMnTuzRd4PULEAi4gXgreIH4NzM/M+I2A+4GtgSeAE4ITNfjYhRwG3AaGB+Zp5ebCeAb2XmkbWqXZI6al3T0q0ht2W2O2jIO09DbcjYsWNZsGABAGPGjOHb3/72Bte77bbbeqy+zmp9CuuYzPxo8fOfETEIuBH4u8wcB8wDLi3WPR64LzPHA7tExO5F+yzgrBrXLUnrdPdLvq9st7fU+xrIBOCtzHygmL8KOLaYbgGaipDZAmiOiJOBX2Tms7UvVZLUUa2vgcyNiAbgAeB84IPAwvaFmbk0IgZFxDZUjkzmAI8BPwKWAacCh9a4ZknSBtQyQA7MzEURsQXwT8CVwO1drZyZK4Fj2ucj4nrgQmBiRJwGrAamZubCLjYhST2mra2NhoaGepfRa1pbW6vuU7NTWJm5qPhcDXwH2B/4HbBj+zoRMRpozczlHftGxMSi7/3AFcApwLXARbWoXdLANmzYMJYtW0ZbW1u9S+lxbW1tNDc38+KLLzJ8+PCq+tbkCCQihgNDMvP14hTWZ4DHgUeBLSPigOI6yGTg1k59hwLTgUlFUxPQWvyMqEX9kga2sWPHsnjxYpYsWVLvUnrFkCFD2HrrrRk9enR1/Xqpns7eC/x7RAwGBgO/AU7PzNaIOBG4OiKGUQzj7dR3CnB9Zi4r5qcDjwDNVK6JSFKvamxsZKeddqp3GX1OTQIkM38L7NnFsgeB8RvpO73T/LVUTl9Jkuqo3sN4JUn9lAEiSSrFAJEklWKASJJKMUAkSaUYIJKkUgwQSVIpBogkqRQDRJJUigEiSSrFAJEklWKASJJKMUAkSaUYIJKkUgwQSVIpBogkqRQDRJJUigEiSSrFAJEklWKASJJKMUAkSaUYIJKkUgwQSVIpQ2q9w4j4GvAPwPjM/HVE7AdcDWwJvACckJmvRsQo4DZgNDA/M08v+gfwrcw8sta1S5LeVtMjkIjYC9gPWFjMDwJuBP4uM8cB84BLi9WPB+7LzPHALhGxe9E+CzirlnVLkt6pZgESEVsAs4HTOjRPAN7KzAeK+auAY4vpFqCpCJktgOaIOBn4RWY+W6OyJUldKB0gEbFlEQrddRFwY2a+0KHtgxRHIwCZuRQYFBHbUDky+TDwGHA3sAw4Ffhm2ZolST2n29dAImIm8G+Z+cuIOAL4IdAWEX+dmXdsou/HgL2B87q7v8xcCRzTYRvXAxcCEyPiNGA1MDUzF3axCUlSL6rmCOR44NfF9DTgBOAvgH/sRt+DgF2B5yPiBWAs8J9UjjB2bF8pIkYDrZm5vGPniJgIkJn3A1cApwDXUjmqkSTVQTWjsJoy882I2BbYOTP/HSAidtxEPzLzUt6+OE4RIkcCvwG+GBEHFNdBJgO3duwbEUOB6cCk9jqA1uJnRBX1S5J6UDUB8nREHE/lqOHnsO6IYVXZnWdma0ScCFwdEcMohvF2Wm0KcH1mLivmpwOPAM1UrolIkuqgmgA5HfhnKqOjPle0HQ78rNqdZuaHOkw/CIzfyLrTO81fS+X0lSSpjrodIJn5MPAnndrmAnN7uihJUt9X1Z3oEfEJ4DPAdpl5VETsDWyVmff2SnWSpD6r26OwIuLLwHeBZ4CJRfMqKtckJEkDTDXDeM8CDi1GVLUWbU8B0eNVSZL6vGoCZCSwqJhuKz4bqYyGkiQNMNUEyDzeeSf5GcB9PVeOJKm/qOYi+peBOyLiC8DIiEjgDSo3BEqSBphqhvG+HBH7APtQefzIIuCXmdm68Z6SpM1RNQ9T/HFmHg38svhpb78tM/+yN4qTJPVd1VwD+XgX7Qf3QB2SpH5mk0cgEdH+xNuhHabb7UyH93lIkgaO7pzC+kDxOajDNFSG8i6i8n5zSdIAs8kAycy/AYiIB4sHGUqSVNUorGsjYmsqd56P6LTMZ2FJ0gBTzSisU4DZwB+ANzssaqNyLUSSNIBUcyPhJcAxmfnT3ipGktR/VDOMdwglXh4lSdo8VRMg3wS+GhHV9JEkbaaqOYX198D2wJSIWNZxQWZ+sEerkiT1edUEyAm9VoUkqd+pZhjv/b1ZiCSpf6lmGO8WwDTgOGDbzNw6Ig4DxmXmlb1VoCSpb6rmgvgsYHfgeN5+I+F/A6f1dFGSpL6vmgCZBHw2M39B8U70zHwR2KE3CpMk9W3VXERv7rx+RIwBlm149fVFxI+AnaiEzx+AL2fm4xExDpgDbFts66TMfCYiGoFbiz7PAcdm5pqIGA3cBhyamb6PXZLqpJojkFuBORGxE0BEvA+4Erilm/1Pzsw9MnNPYCbwvaL9KmB2Zo6j8qiUq4v2w4HlmbkHsAL4ZNE+A7jA8JCk+qomQM4Hngd+BbwHeAZ4Cfh6dzpn5usdZrcGWiNiO2Av4Oai/WZgr+LIpgVoKtqbgOaIOAhozcz5VdQtSeoF1QzjbaZyM+HfF1/wSzOzbRPd1hMR1wGHAQ1Ujig+ALyYmWuLfayNiJeK9p8Dx0TEE8BDwDwqj1L5VDX7lCT1jo0GSER8KDNfKKY7P3F3ZEQAkJm/7c7OMvPzxbZOBC4DLtzIuq3AFzrUMg24DtgxItpPc03PzCe6s29JUs/a1CmsX3WYfpbKaatnO/08U+1OM/NfqLxjfTGwQ0QMBig+30/lTYfrRMRHgP0y8wfA5cA5wJRiWpJUBxs9AsnMkR2mSz9EMSJGAKMyc1ExfxSwHHgVeJzKzYk3Fp+PZeaSTpuYBZxVTA+nch9KG51ebCVJqp1qhvG+G8OBWyNiOLCWSngclZltETGZyuiuacBrwEkdO0bECcDDmfl00TQNuKuYPqcm1UuS3mFT10Dm8/Zd513KzImbWP57YL8ulj0F7LuRvjd2mr8TuHNTNUmSetemjkCuq0kVkqR+Z1PXQOa0T0fE5cAtmflgh7Y/AY6lcie5JGkAqebC+HHAI53aHgU+23PlSJL6i2oCpA0Y3KltcJXbkCRtJqr58p8PXNz+TvTi8x+KdknSAFPNMN4zqYx+ejkiFgIfBF4GjuqNwiRJfVs1z8JaHBF7AX9M5VlVi4BfFo8ckSQNMFXdSFiExUPFjyRpAPMCuCSpFANEklSKASJJKsUAkSSVYoBIkkoxQCRJpRggkqRSDBBJUikGiCSpFANEklSKASJJKsUAkSSVYoBIkkoxQCRJpRggkqRSqnofSFkRsS3wL8AfAc3AM8DfZuaSiNgPuBrYEngBOCEzX42IUcBtwGhgfmaeXmwrgG9l5pG1qF2StGG1OgJpA2ZkZmTmeOA54NLiveo3An+XmeOAecClRZ/jgfuK9XeJiN2L9lnAWTWqW5LUhZoESGYuz8z/6tD0ELAjMAF4KzMfKNqvAo4tpluApiJktgCaI+Jk4BeZ+Wwt6pYkda3m10CKQDgN+A/gg8DC9mWZuRQYFBHbUDky+TDwGHA3sAw4FfhmrWuWJL1TTa6BdHIF8AfgSmBSVytl5krgmPb5iLgeuBCYGBGnAauBqZm5sItNqJe0rmlh0JDGepfRZ/j30EBV0wCJiJnAR4CjMrM1In5H5VRW+/LRQGtmLu/UbyJAZt4fEU8CfwzsDVwEnFyr+lUxaEgjj874fL3L6DMmTLmu3iVIdVGzU1gR8Y9Urnl8KjNXF82PAltGxAHF/GTg1k79hgLTgSlFUxPQWvyM6O26JUkbVqthvLsBU4GngQcrI3F5PjMnRcSJwNURMYxiGG+n7lOA6zNzWTE/HXiEynDgU2tQviRpA2oSIJn530BDF8seBMZvpO/0TvPXAtf2aIGSpKp5J7okqRQDRJJUigEiSSrFAJEklWKASJJKMUAkSaUYIJKkUgwQSVIpBogkqRQDRJJUigEiSSrFAJEklWKASJJKMUAkSaUYIJKkUgwQSVIpBogkqRQDRJJUigEiSSrFAJEklWKASJJKMUAkSaUYIJKkUobUYicRMRP4NPAhYHxm/rpoHwfMAbYFlgEnZeYzEdEI3ArsBDwHHJuZayJiNHAbcGhmNteidknShtXqCORHwERgYaf2q4DZmTkOmA1cXbQfDizPzD2AFcAni/YZwAWGhyTVX00CJDMfyMxFHdsiYjtgL+DmoulmYK+IGAO0AE1FexPQHBEHAa2ZOb8WNUuSNq6e10A+ALyYmWsBis+XivafA29ExBPA68A84GJgSp1qlSR1UpNrINXKzFbgC+3zETENuA7YMSLaT3NNz8wn6lGfJKm+RyCLgB0iYjBA8fn+on2diPgIsF9m/gC4HDiHypHI5bUtV5LUUd0CJDNfBR4HjiuajgMey8wlnVadBZxVTA8H2oBWYEQt6pT6i+aWtfUuoc/wb1EbtRrGeznwl8D2wN0RsSwzdwMmA3OKU1SvASd16ncC8HBmPl00TQPuKqbPqUXtUn8xtHEwn50yt95l9Ak3zTi+3iUMCDUJkMw8AzhjA+1PAftupN+NnebvBO7s8QIlSVXzTnRJUikGiCSpFANEklSKASJJKsUAkSSVYoBI2uy0rmmpdwl9Sm/9Pfrko0wk6d0YNKSRR2d8vt5l9BkTplzXK9v1CESSVIoBIkkqxQCRJJVigEiSSjFAJEmlGCCSpFIMEElSKQaIJKkUA0SSVIoBIkkqxQCRJJVigEiSSjFAJEmlGCCSpFIMEElSKQaIJKkUA0SSVEqfeCNhRIwD5gDbAsuAk4AXgFuBnYDngGMzc01EjAZuAw7NzOb6VCxJ6itHIFcBszNzHDAbuBo4HFiemXsAK4BPFuvOAC4wPCSpvup+BBIR2wF7AZ8omm4GrgSGA01FWxPQHBEHAa2ZOb/K3QwGeOWVV0rXufrNFaX7bm4WL17MkjfeqncZfcbixYvrXcI6/jut8N/o+sr+G+3wnTl4Q8sb2traSpbUMyJiAvCDzNytQ9tvgBOA04A/Bh4CzgR+BnwqM5dXuY8DgGpDR5JUcWBmPtC5se5HIBuTmV9on46IacB1wI4RcXXRPD0zn+jGph4GDgReBtb2eKGStHkaDLyPynfoO/SFAFkE7BARgzNzbUQMBt5ftAMQER8B9svMiyJiPnAi0AB8HzhoUzvIzNXAO9JTkrRJz3W1oO4X0TPzVeBx4Lii6Tjgscxc0mG1WcBZxfRwoA1oBUbUqk5J0vr6whEIwGRgTnGa6jUqw3gBiIgTgIcz8+miaRpwVzF9Tk2rlCStU/eL6JKk/qnup7AkSf2TASJJKsUAkSSVYoBIkkrpK6Ow1A9s6KGXmflMfauS3hYRM4FPAx8Cxmfmr+tb0ebNIxBVY0MPvZT6kh8BE4GF9S5kIDBA1C0dHnp5c9F0M7BXRIypX1XS+jLzgcxctOk11RMMEHXXB4AXM3MtQPH5UtEuaQAyQCRJpRgg6q51D70E2NBDLyUNLAaIuqWbD72UNID4LCx1W0TsQmUY7yiKh15mZta3KultEXE58JfA9sBSYFnHl9WpZxkgkqRSPIUlSSrFAJEklWKASJJKMUAkSaUYIJKkUgwQ6V2KiOMj4mf12n5EHBgRDqdWzTmMVwNKRBwAzAB2A9YCTwJnZebDda7rcOACYE/gLeA3wLcy8z82sG4b8JHMfLa2VUrr830gGjAiYivgTuA04N+AocCBwOo61dMANFC58e17wP8BjgLeKOo6AXhHgEh9hUcgGjAiYm/g7sx8TxfLPwecQ+Uu5l8CX8zMhcWyNirB8xVgDDAX+FJmtkXEKcDnM/OAYt0/Af4ZGAc8DZyZmQ8Wy/4L+H/AwVQej/+/gXuAKzLzsi7qWrf9iJhHJVzeBNqAU4HfAzdm5tiI+Gvg+g7dG4FfZObBEbEFcAlwLLAFcDvw95m5KiIOBm4EZgHnUjk6Oz8zb9j0X1YDlddANJA8DayNiDkR8WcRMap9QUQcDZxP5WhgDDCft9990u5IYB8qX/rHAod33kFEbAP8BLicypsbvw38JCK27bDaicAXgZHAMCqPxP9hd36BzJxYTO6RmSMy8187Lf/Xon0ElYdd/rbD73EplVD7KPBhYAdgWofu2wNbF+2nArM7/o2kzgwQDRiZ+T/AAVT+534tsCQi/iMi3gtMBr6RmU9m5hrgH4GPRsSOHTZxaWauyMzfAfdR+SLu7Ajgmcz8l8xck5k3A09ROTXV7vuZ+d/FfrYq2l7uyd81IgYBNwH/lZlXF6fLvkjliGN5Zr5R/I6f6dCtBbgoM1sy8y7gD0D0ZF3avHgNRANKZj4JnALrHg55I/BPwI7AP0fEtzqs3kDlf+Ptr0d9pcOyN4ERG9jF+3nn61QXFttp1/ER+MuKz/cBz3f39+iGS6gc4ZxRzI8BmoBHI9ZlQgMwuGMtRai16+p3lAADRANYZj4VEd8H/pbKl/olmTn3XW72JSph1NEHgf/bYb7jhccs9v1pYOa73DcAEfEZKo/b3yczW4rmpcAqYLfMfLEn9iN5CksDRkTsEhFfiYixxfwHqHzRPgRcBUyNiN2KZVtHxF+V2M1dwLiI+GxEDCkuav8vKqO/3iEz26iMvrowIv4mIraKiEERcUBEXNPFPn4P7NzF77gncAXwqY7vasnMViqn7WYV77cnInYohg9LpRggGkjeAPYFFkTESirB8WvgK5l5O/BN4JaI+J+i/c+q3UFmLqNysf0rVE5PTQGOzMylG+nzQ+Cvgc9ROYL5PTAd+HEXXf4BmBMRKyLi2E7LjqbyvpYHIuIPxc9Pi2XnAs8CDxW/4914jUPvgsN4JUmleAQiSSrFAJEklWKASJJKMUAkSaUYIJKkUgwQSVIpBogkqRQDRJJUigEiSSrl/wM8SqqykNv2CQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 285
        },
        "id": "lXvhx8eENPAJ",
        "outputId": "555352ba-d939-40e2-b621-bb280ebe80b5"
      },
      "source": [
        "barplot_porcentagens(\"gender\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zzhRYCPLOJj5"
      },
      "source": [
        "O gênero dos clientes não possuem influência na sua probabilidade de cancelamento."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 285
        },
        "id": "QKSKVM-DNgUA",
        "outputId": "ab14781d-7957-43bc-96e6-d958f7f240fd"
      },
      "source": [
        "barplot_porcentagens(\"InternetService\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4gkCUNqaN49w"
      },
      "source": [
        "As pessoas que possuem o serviço de internet fibra ótica contrato aparentam ser mais propensos a cancelar os serviços."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 285
        },
        "id": "1phoVP6dNhwA",
        "outputId": "0d82f626-971c-494f-8a96-c2956cb80feb"
      },
      "source": [
        "barplot_porcentagens('PaperlessBilling')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "81X8Yyu5NCez"
      },
      "source": [
        "Pessoas que efetuam através de \"papeis\" são menos provaveis a cancelarem o serviço.\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 285
        },
        "id": "RDTeCzaCOyMy",
        "outputId": "18a04b43-ebc0-4318-cd7c-c4281c4e50be"
      },
      "source": [
        "barplot_porcentagens('Contract')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3OQuR5l8Ox_C"
      },
      "source": [
        "Contratos de maior duração possuem um menor Churn."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 173
        },
        "id": "8UZej53IOfCL",
        "outputId": "ec2dae3a-57c6-4c2b-ecf4-4bd169632949"
      },
      "source": [
        "#selecionar colunas para gráfico Churn para clientes com parceiros ou não\n",
        "partner = df.groupby('Partner')['Churn'].value_counts()\n",
        "partner = partner.to_frame().rename({'Churn': 'clientes'}, axis=1).reset_index()\n",
        "partner"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Partner</th>\n",
              "      <th>Churn</th>\n",
              "      <th>clientes</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>No</td>\n",
              "      <td>No</td>\n",
              "      <td>2441</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>No</td>\n",
              "      <td>Yes</td>\n",
              "      <td>1200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Yes</td>\n",
              "      <td>No</td>\n",
              "      <td>2733</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Yes</td>\n",
              "      <td>Yes</td>\n",
              "      <td>669</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  Partner Churn  clientes\n",
              "0      No    No      2441\n",
              "1      No   Yes      1200\n",
              "2     Yes    No      2733\n",
              "3     Yes   Yes       669"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 149
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "id": "Lw2lD6-hObTx",
        "outputId": "3c6ba56a-554b-4ff4-8144-30558ec45e7f"
      },
      "source": [
        "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14,4))\n",
        "\n",
        "ax1.set_title(\"Parceiros - Sim/Não\")\n",
        "ax2.set_title(\"Dependentes - Sim/Não\")\n",
        "\n",
        "# Plotar Partner column\n",
        "ax = sns.barplot(x='Partner', y= 'clientes', hue='Churn', data=partner, ax=ax1)\n",
        "\n",
        "# Plot Dependents column\n",
        "dep = df.groupby('Dependents')[\"Churn\"].value_counts()\n",
        "dep = dep.to_frame().rename({\"Churn\": 'clientes'}, axis=1).reset_index()\n",
        "\n",
        "sns.barplot(x='Dependents', y='clientes', data = dep, hue='Churn');\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 585
        },
        "id": "QgDggykjOtAy",
        "outputId": "5575f165-76ac-425f-be04-f4e09fef0619"
      },
      "source": [
        "ax = sns.catplot(x=\"Churn\", y=\"MonthlyCharges\", row=\"PaymentMethod\", kind=\"bar\", data=df, height=2, aspect=4, orient='v')\n",
        "\n",
        "plt.tight_layout()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x576 with 4 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FsYUXEpHO8G8"
      },
      "source": [
        "## Preparação dos dados\n",
        "Nessa etapa faremos o tratamento dos dados para que possamos realizar o treinamento do nosso modelo de Machine Learning"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZOksvd0eO793"
      },
      "source": [
        "#copiar o dataset original  \n",
        "df_enc = df.copy()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IVr90gHEPTpK"
      },
      "source": [
        "Agora, iremos transformar as variáveis categóricas em numéricas para que possamos treinar nossos modelos. Para isso utilizaremos o LabelEncoder para variáveis binárias e o get_dummy para variáveis categóricas não binárias, disponível na biblioteca sklearn.\n",
        "\n",
        "**O que são variáveis categóricas?**\n",
        "> Variáveis categóricas também são chamadas variáveis quantitativas ou variáveis de atributo. Os valores de uma variável categórica são categorias ou grupos mutuamente exclusivos. Os dados categóricos podem ou não ter alguma ordem lógica.\n",
        "Fonte: [Mintlab](https://support.minitab.com/pt-br/minitab/20/help-and-how-to/statistics/tables/supporting-topics/basics/categorical-and-quantitative-variables/)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "background_save": true,
          "base_uri": "https://localhost:8080/"
        },
        "id": "dLVSdPKMO7mC",
        "outputId": "835caece-8385-43d6-e8f2-fdf9d80e66a9"
      },
      "source": [
        "#importar encoder\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "\n",
        "#parametrizar\n",
        "le = LabelEncoder()\n",
        "\n",
        "for c in categorias_binarias:\n",
        "  df_enc[c] = le.fit_transform(df_enc[c])\n",
        "  print(c, 'passou pelo LabelEncoder')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-1-faf970c2bfd4>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mle\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mLabelEncoder\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      6\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 7\u001b[0;31m \u001b[0;32mfor\u001b[0m \u001b[0mc\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mcategorias_binarias\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      8\u001b[0m   \u001b[0mdf_enc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mc\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mle\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit_transform\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdf_enc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mc\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      9\u001b[0m   \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'passou pelo LabelEncoder'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'categorias_binarias' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_GziU7jMQFfV"
      },
      "source": [
        "Vamos ver como ficaram nossas variáveis categóricas binárias."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "7fKE05etQUKq"
      },
      "source": [
        "df_enc.head()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "F_W2EgD9QbxS"
      },
      "source": [
        "Vamos ver quais eram os valores da variável gender e como eles ficaram após passar pela transformação em variáveis categóricas.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "colPyS25QoSz"
      },
      "source": [
        "print('Antes:', df.gender.unique(), '\\n')\n",
        "print('Depois:', df_enc.gender.unique())"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "011b4u0IREzk"
      },
      "source": [
        "Como podemos ver, os valores categóricos Male e Female, foram transformados em 1 e 0, respectivamente. Isso ajudará no treinamento do nosso modelo de machine learning.\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vx70bbt5RVOT"
      },
      "source": [
        "Vamos verificar se há alguma correlação evidente entre as variáveis."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 815
        },
        "id": "KovvgKaKRL7U",
        "outputId": "6189f72d-5416-444b-fedc-2fa95b82fe6b"
      },
      "source": [
        "corr = df_enc.corr()\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(14,14))\n",
        "sns.heatmap(corr, square=True, annot=True, linewidths=.1, cmap = 'coolwarm');"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x1008 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "i7wWBsATRfIl"
      },
      "source": [
        "Agora, também vamos transformar as variáveis categóricas não binárias, utilizando a função get_dummy"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ECR_x74oReA1",
        "outputId": "14c759ba-5d87-4a5b-833b-d5b84ba7696d"
      },
      "source": [
        "#converter variaveis dummy\n",
        "\n",
        "df_enc = pd.get_dummies(df_enc, columns=categorias_n_binarias)\n",
        "\n",
        "for c in categorias_n_binarias:\n",
        "  print('A variável {} foi transformada'.format(c))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "A variável MultipleLines foi transformada\n",
            "A variável InternetService foi transformada\n",
            "A variável OnlineSecurity foi transformada\n",
            "A variável OnlineBackup foi transformada\n",
            "A variável DeviceProtection foi transformada\n",
            "A variável TechSupport foi transformada\n",
            "A variável StreamingTV foi transformada\n",
            "A variável StreamingMovies foi transformada\n",
            "A variável Contract foi transformada\n",
            "A variável PaymentMethod foi transformada\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 258
        },
        "id": "97Wip0irRnQ4",
        "outputId": "4d822c09-7d7d-4f80-8354-4484c9a3ec43"
      },
      "source": [
        "df_enc.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>gender</th>\n",
              "      <th>SeniorCitizen</th>\n",
              "      <th>Partner</th>\n",
              "      <th>Dependents</th>\n",
              "      <th>tenure</th>\n",
              "      <th>PhoneService</th>\n",
              "      <th>PaperlessBilling</th>\n",
              "      <th>MonthlyCharges</th>\n",
              "      <th>TotalCharges</th>\n",
              "      <th>Churn</th>\n",
              "      <th>MultipleLines_No</th>\n",
              "      <th>MultipleLines_No phone service</th>\n",
              "      <th>MultipleLines_Yes</th>\n",
              "      <th>InternetService_DSL</th>\n",
              "      <th>InternetService_Fiber optic</th>\n",
              "      <th>InternetService_No</th>\n",
              "      <th>OnlineSecurity_No</th>\n",
              "      <th>OnlineSecurity_No internet service</th>\n",
              "      <th>OnlineSecurity_Yes</th>\n",
              "      <th>OnlineBackup_No</th>\n",
              "      <th>OnlineBackup_No internet service</th>\n",
              "      <th>OnlineBackup_Yes</th>\n",
              "      <th>DeviceProtection_No</th>\n",
              "      <th>DeviceProtection_No internet service</th>\n",
              "      <th>DeviceProtection_Yes</th>\n",
              "      <th>TechSupport_No</th>\n",
              "      <th>TechSupport_No internet service</th>\n",
              "      <th>TechSupport_Yes</th>\n",
              "      <th>StreamingTV_No</th>\n",
              "      <th>StreamingTV_No internet service</th>\n",
              "      <th>StreamingTV_Yes</th>\n",
              "      <th>StreamingMovies_No</th>\n",
              "      <th>StreamingMovies_No internet service</th>\n",
              "      <th>StreamingMovies_Yes</th>\n",
              "      <th>Contract_Month-to-month</th>\n",
              "      <th>Contract_One year</th>\n",
              "      <th>Contract_Two year</th>\n",
              "      <th>PaymentMethod_Bank transfer (automatic)</th>\n",
              "      <th>PaymentMethod_Credit card (automatic)</th>\n",
              "      <th>PaymentMethod_Electronic check</th>\n",
              "      <th>PaymentMethod_Mailed check</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>29.85</td>\n",
              "      <td>29.85</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>34</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>56.95</td>\n",
              "      <td>1889.50</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>53.85</td>\n",
              "      <td>108.15</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>45</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>42.30</td>\n",
              "      <td>1840.75</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>70.70</td>\n",
              "      <td>151.65</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   gender  ...  PaymentMethod_Mailed check\n",
              "0       0  ...                           0\n",
              "1       1  ...                           1\n",
              "2       1  ...                           1\n",
              "3       1  ...                           0\n",
              "4       0  ...                           0\n",
              "\n",
              "[5 rows x 41 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 158
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "v5SjQ_5mSQkK"
      },
      "source": [
        "### Balancear os dados"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kPevoo_iSQg0"
      },
      "source": [
        "Como vimos acima, os dados da variável Churn estão desbalanceados, possuindo uma quantiade muito maior de uma categoria do que outra. Chegou a hora de balancear o conjunto de dados para que não haja viés nas previsões. Para isso, vamos utilizar o SMOTE.\n",
        "\n",
        "\n",
        ">A SMOTE (técnica de sobreamostragem minoritária sintética) é uma técnica estatística para aumentar o número de casos em seu conjunto de um modo equilibrado. O módulo funciona gerando novas instâncias de casos minoritários existentes que você fornece como entrada. Essa implementação de SMOTE não altera o número de casos de maioria.\n",
        "\n",
        "\n",
        "Fonte: [Microsoft Machine Learning](https://docs.microsoft.com/pt-br/azure/machine-learning/algorithm-module-reference/smote#:~:text=A%20SMOTE%20(t%C3%A9cnica%20de%20sobreamostragem,dataset%20in%20a%20balanced%20way.)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gFYsHnzhSQcc"
      },
      "source": [
        "!pip install imbalanced-learn -q\n",
        "#Importing SMOTE\n",
        "from imblearn.over_sampling import SMOTE"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9X9wdAUITRC5"
      },
      "source": [
        "Como é a distribuição atual da variável Churn\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 355
        },
        "id": "JrpbPc2ZSO1s",
        "outputId": "794e6a14-bb67-4b99-d415-0dd5aa82b2ea"
      },
      "source": [
        "ax = sns.countplot(x=\"Churn\", data=df_enc)\n",
        "plt.title('Churn')\n",
        "plt.show()\n",
        "\n",
        "churn_yes = df_enc.Churn.value_counts()[1]/df.shape[0]\n",
        "churn_no = df_enc.Churn.value_counts()[0]/df.shape[0]\n",
        "\n",
        "print('\\n')\n",
        "print('Os clientes Churn representam {:,.2%} do total'.format(churn_yes))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "\n",
            "Os clientes Churn representam 26.54% do total\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_LpLfDDsTYhZ"
      },
      "source": [
        "#1 parametrizar \n",
        "smt = SMOTE()\n",
        "\n",
        "#2 separar o modelo entre features e target\n",
        "X = df_enc.drop('Churn', axis=1)\n",
        "y = df_enc.Churn\n",
        "\n",
        "#aplicar smote nos dados\n",
        "X, y = smt.fit_sample(X, y)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Sw3fHkWwTkBK"
      },
      "source": [
        "Como ficaram os dados da variável Churn após a aplicação do SMOTE"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 374
        },
        "id": "tfzEDxIFTef4",
        "outputId": "f53c68ed-145b-45df-8a87-fb4e90ad3270"
      },
      "source": [
        "ax = sns.countplot(x = y)\n",
        "plt.title('Churn')\n",
        "plt.show()\n",
        "\n",
        "print('\\n')\n",
        "print(np.bincount(y))\n",
        "print('Agora temos {} entradas'.format(y.shape[0]))\n",
        "print('sendo 50% para Churn Sim e 50% para Churn Não')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "\n",
            "[5174 5174]\n",
            "Agora temos 10348 entradas\n",
            "sendo 50% para Churn Sim e 50% para Churn Não\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Jzod3Z3eUBoH"
      },
      "source": [
        "## Construção do modelo\n",
        "\n",
        "Agora que nossos dados já estão balanceados e devidamente preparados, vamos dar início ao processo de construção do nosso modelo que será responsável por prever se um cliente cancelará o serviço ou não.\n",
        "\n",
        "Estamos lidando com um problema de classificação, então utilizaremos modelos que condizem com a nosso necessidade de classficar \n",
        "\n",
        "* Churn \n",
        "   * Sim \n",
        "   * Não\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iZ5hD5pIU2j2"
      },
      "source": [
        "!pip install scikit-plot -q"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zvyelDDbU35O"
      },
      "source": [
        "#importando as bibliotecas que serão utilizadas\n",
        "\n",
        "#métricas \n",
        "from sklearn.metrics import roc_auc_score\n",
        "from sklearn.metrics import confusion_matrix\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.metrics import f1_score\n",
        "from sklearn.metrics import recall_score\n",
        "from sklearn.metrics import classification_report\n",
        "import scikitplot as skplt\n",
        "from sklearn.model_selection import cross_val_score\n",
        "from sklearn.metrics import make_scorer\n",
        "from sklearn.model_selection import cross_validate\n",
        "from sklearn.model_selection import GridSearchCV\n",
        "from matplotlib import pyplot\n",
        "\n",
        "#modelos\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.preprocessing import StandardScaler"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sJxzUfN5VERA"
      },
      "source": [
        "Vamos utilizar 4 modelos de classificação para comparação e definir qual será o que possui o melhor desempenho dentre os 4 escolhidos.\n",
        "\n",
        "Os modelos são:\n",
        "* Árvore de decisão\n",
        "* Regressão Logística\n",
        "* SVC (Support Vector Machine)\n",
        "* Floresta de decisão aleatória"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "B4CTbboiV7bY"
      },
      "source": [
        "A seguir, faremos as seguintes etapas:\n",
        "\n",
        "\n",
        "*   Dividir nossos dados em treino e teste para que não ocorra overfitting\n",
        "*   Normalizar os dados\n",
        "*   Treinar os modelos\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jLiUTXMLVD7A"
      },
      "source": [
        "#1 importar modelos\n",
        "dt = DecisionTreeClassifier()\n",
        "lr = LogisticRegression()\n",
        "svc = SVC()\n",
        "rf = RandomForestClassifier()\n",
        "\n",
        "names = ['Decision Tree', 'Logistic Regression', 'SVC', 'Random Forest']\n",
        "classifiers = [dt, lr, svc, rf]\n",
        "\n",
        "#4 separar entre treino e teste\n",
        "X_train, X_test, y_train, y_test = train_test_split(X,y)\n",
        "\n",
        "\n",
        "#4.1 normalizar os dados\n",
        "scaler = StandardScaler()\n",
        "scaler = scaler.fit(X_train)\n",
        "X_train = scaler.transform(X_train)\n",
        "X_test = scaler.transform(X_test)\n",
        "\n",
        "\n",
        "#5 treinar modelo\n",
        "model_dt = dt.fit(X_train, y_train)\n",
        "model_lr = lr.fit(X_train, y_train)\n",
        "model_svc = svc.fit(X_train, y_train)\n",
        "model_rf = rf.fit(X_train, y_train)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I8jv7YxBV2ud"
      },
      "source": [
        "Após treinarmos os modelos nos dados de treino, vamos verificar seu desempenho através de Cross-validation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jvPZiI1xVwYa",
        "outputId": "0ef85766-6cd3-4740-eabd-cab0811d7bce"
      },
      "source": [
        "print('Cross Validation dos Modelos: recall\\n')\n",
        "models = (model_dt, model_lr, model_svc, model_rf)\n",
        "for n, m in zip(names, models):\n",
        "    scores = cross_val_score(m, X_train, y_train, cv=5, scoring='recall')\n",
        "    print(n, \"tem recall de : %0.3f (+/- %0.2f)\\n\" % (scores.mean(), scores.std() * 2))\n",
        "\n",
        "print('')\n",
        "print('Cross Validation dos Modelos: Acurácia\\n')\n",
        "for n, m in zip(names, models):\n",
        "    scores_2 = cross_val_score(m, X_train, y_train, cv=5, scoring='accuracy')\n",
        "    print(n, \"tem acurácia de: %0.3f (+/- %0.2f)\\n\" % (scores_2.mean(), scores_2.std() * 2))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Cross Validation dos Modelos: recall\n",
            "\n",
            "Decision Tree tem recall de : 0.808 (+/- 0.02)\n",
            "\n",
            "Logistic Regression tem recall de : 0.826 (+/- 0.03)\n",
            "\n",
            "SVC tem recall de : 0.832 (+/- 0.03)\n",
            "\n",
            "Random Forest tem recall de : 0.827 (+/- 0.03)\n",
            "\n",
            "\n",
            "Cross Validation dos Modelos: Acurácia\n",
            "\n",
            "Decision Tree tem acurácia de: 0.804 (+/- 0.01)\n",
            "\n",
            "Logistic Regression tem acurácia de: 0.776 (+/- 0.01)\n",
            "\n",
            "SVC tem acurácia de: 0.828 (+/- 0.01)\n",
            "\n",
            "Random Forest tem acurácia de: 0.855 (+/- 0.01)\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "t0LPKnowWUB6"
      },
      "source": [
        "Vemos que os modelos Random Forest e SVC foram os que tiveram um melhor desempenho.\n",
        " \n",
        "Agora, daremos início a parametrização para otimizar o modelo de Floresta de Decisão. "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hI_HomWmW61w"
      },
      "source": [
        "#### Tunning dos parâmetros"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SyrZlFwAXMsh",
        "outputId": "0c518a36-fce9-4ce5-d115-489b446bfc7e"
      },
      "source": [
        "parametros = {'n_estimators': [10,15,20,30,50,75,100]}\n",
        "parametros_d = {'max_depth': [5,7,10,15,20,30]}\n",
        "\n",
        "grid_search = GridSearchCV(rf, parametros, scoring=\"recall\")\n",
        "clf = GridSearchCV(rf, parametros)\n",
        "clf.fit(X_train, y_train)\n",
        "\n",
        "# ver melhor parâmetro\n",
        "print(clf.best_params_)\n",
        "# ver todos parâmetros testados\n",
        "means = clf.cv_results_['mean_test_score']\n",
        "stds = clf.cv_results_['std_test_score']\n",
        "\n",
        "for mean, std, params in zip(means, stds, clf.cv_results_['params']):\n",
        "        print(\"%0.3f (+/-%0.03f) for %r\"\n",
        "              % (mean, std * 2, params))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "{'n_estimators': 50}\n",
            "0.846 (+/-0.012) for {'n_estimators': 10}\n",
            "0.849 (+/-0.013) for {'n_estimators': 15}\n",
            "0.852 (+/-0.020) for {'n_estimators': 20}\n",
            "0.852 (+/-0.011) for {'n_estimators': 30}\n",
            "0.858 (+/-0.009) for {'n_estimators': 50}\n",
            "0.856 (+/-0.009) for {'n_estimators': 75}\n",
            "0.855 (+/-0.014) for {'n_estimators': 100}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hArYsYz3XQSj",
        "outputId": "b0f15a8c-cd77-4b12-b336-3ae42b440930"
      },
      "source": [
        "grid_search = GridSearchCV(rf, parametros_d, scoring=\"recall\")\n",
        "clf = GridSearchCV(rf, parametros_d)\n",
        "clf.fit(X_train, y_train)\n",
        "\n",
        "# ver melhor parâmetro\n",
        "print(clf.best_params_)\n",
        "# ver todos parâmetros testados\n",
        "means = clf.cv_results_['mean_test_score']\n",
        "stds = clf.cv_results_['std_test_score']\n",
        "\n",
        "for mean, std, params in zip(means, stds, clf.cv_results_['params']):\n",
        "        print(\"%0.3f (+/-%0.03f) for %r\"\n",
        "              % (mean, std * 2, params))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "{'max_depth': 20}\n",
            "0.827 (+/-0.008) for {'max_depth': 5}\n",
            "0.842 (+/-0.008) for {'max_depth': 7}\n",
            "0.854 (+/-0.004) for {'max_depth': 10}\n",
            "0.857 (+/-0.008) for {'max_depth': 15}\n",
            "0.858 (+/-0.009) for {'max_depth': 20}\n",
            "0.856 (+/-0.012) for {'max_depth': 30}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1FtlULUkb6wR"
      },
      "source": [
        "> Quais as features foram as mais importantes para a criação do modelo de Random Forest?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XbmXoFLqasko",
        "outputId": "ae463738-1099-492f-cafa-7cd6321a3d63"
      },
      "source": [
        "importance = rf.feature_importances_\n",
        "\n",
        "for i,v in enumerate(importance):\n",
        "  print('Feature: %0d, Score: %.4f' % (i,v))\n",
        "plt.bar([x for x in range(len(importance))], importance)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Feature: 0, Score: 0.0236\n",
            "Feature: 1, Score: 0.0155\n",
            "Feature: 2, Score: 0.0267\n",
            "Feature: 3, Score: 0.0194\n",
            "Feature: 4, Score: 0.0941\n",
            "Feature: 5, Score: 0.0032\n",
            "Feature: 6, Score: 0.0315\n",
            "Feature: 7, Score: 0.0859\n",
            "Feature: 8, Score: 0.1066\n",
            "Feature: 9, Score: 0.0121\n",
            "Feature: 10, Score: 0.0032\n",
            "Feature: 11, Score: 0.0112\n",
            "Feature: 12, Score: 0.0096\n",
            "Feature: 13, Score: 0.0277\n",
            "Feature: 14, Score: 0.0014\n",
            "Feature: 15, Score: 0.0626\n",
            "Feature: 16, Score: 0.0009\n",
            "Feature: 17, Score: 0.0267\n",
            "Feature: 18, Score: 0.0296\n",
            "Feature: 19, Score: 0.0013\n",
            "Feature: 20, Score: 0.0144\n",
            "Feature: 21, Score: 0.0221\n",
            "Feature: 22, Score: 0.0016\n",
            "Feature: 23, Score: 0.0105\n",
            "Feature: 24, Score: 0.0621\n",
            "Feature: 25, Score: 0.0031\n",
            "Feature: 26, Score: 0.0269\n",
            "Feature: 27, Score: 0.0096\n",
            "Feature: 28, Score: 0.0055\n",
            "Feature: 29, Score: 0.0091\n",
            "Feature: 30, Score: 0.0103\n",
            "Feature: 31, Score: 0.0040\n",
            "Feature: 32, Score: 0.0098\n",
            "Feature: 33, Score: 0.0890\n",
            "Feature: 34, Score: 0.0165\n",
            "Feature: 35, Score: 0.0207\n",
            "Feature: 36, Score: 0.0105\n",
            "Feature: 37, Score: 0.0134\n",
            "Feature: 38, Score: 0.0578\n",
            "Feature: 39, Score: 0.0104\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "V9gWuRA3Xpa6"
      },
      "source": [
        "### Verificar o modelo parametrizado\n",
        "Temos nosso modelo escolhido e validado através dos dados de treino, agora chegou a hora de testar com nossos dados de teste e ver como ele vai se sair!"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0RBkjwhLXuAy",
        "outputId": "506dab15-647d-4c54-8dff-7c900481439c"
      },
      "source": [
        "#1 importar modelos\n",
        "rf = RandomForestClassifier(max_depth=10, n_estimators=100)\n",
        "\n",
        "#2 treinar modelo\n",
        "model_rf = rf.fit(X_train, y_train)\n",
        "\n",
        "#3 ML nos dados de teste\n",
        "y_pred_rf = model_rf.predict(X_test)\n",
        "\n",
        "#4 Classification Report\n",
        "print(classification_report(y_test, y_pred_rf))\n",
        "print(\"Recall: {:.3f}\\n\".format(recall_score(y_test, y_pred_rf)))\n",
        "print(\"Acurácia: {:.3f}\\n\".format(accuracy_score(y_test, y_pred_rf)))\n",
        "print(\"AUC: {:.3}\\n\".format(roc_auc_score(y_test, y_pred_rf)))\n",
        "\n",
        "# plotar matriz de confusão\n",
        "skplt.metrics.plot_confusion_matrix(y_test, y_pred_rf, normalize=True,\n",
        "                                    title='Random Forest',\n",
        "                                    title_fontsize=16,\n",
        "                                    text_fontsize=14);\n",
        "                                  "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.87      0.84      0.86      1311\n",
            "           1       0.84      0.88      0.86      1276\n",
            "\n",
            "    accuracy                           0.86      2587\n",
            "   macro avg       0.86      0.86      0.86      2587\n",
            "weighted avg       0.86      0.86      0.86      2587\n",
            "\n",
            "Recall: 0.875\n",
            "\n",
            "Acurácia: 0.857\n",
            "\n",
            "AUC: 0.858\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KbKEMq-hYECd"
      },
      "source": [
        "Pronto! Nosso modelo está treinado e pronto para ser implementado na previsão de Churns! Na base de teste o modelo conseguiu prever corretamente 86% dos clientes que viriam a cancelar o serviço. "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "P7pLs1wDYWtZ"
      },
      "source": [
        "## Conclusão \n",
        "Além de aprender sobre Churn e o poder da retenção de clientes em uma empresa, conseguimos extrair informações valiosas do conjunto de dados.\n",
        "\n",
        "Foi realizada a análise e o tratamento dos dados, bem como o ajuste das variáveis categóricas para que fosse possível treinar e implementar o modelo de machine learning.\n",
        "\n",
        "Após o treinamento dos modelos, optamos utilizar o modelo de Random Forest, então, foi feito a otimização dos parâmetros do modelo.\n",
        "\n",
        "Em seguida usamos o modelo treinado para fazer as previsões nos dados de teste, obtendo uma pocentagem de 86% para clientes que se tornariam Churn. Ainda há muitas otimizações que podem ser feitas, mas através desse projeto foi alcançado um excelente resultado, capaz de oferecer informações valiosas para a empresa, ajudado a mesma a atingir seu objetivo de reduzir a Churn Rate e aumentar a receita.\n",
        "\n",
        "\n"
      ]
    }
  ]
}