5-sonda/solução/notebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ZPGGGxhcBK9g"
   },
   "source": [
    "# Desafio 5 - Maratona Behind the Code 2021"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "dHDLp1rxBK9m",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "## Bibliotecas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "lxydZp5cBK9n"
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import sys\n",
    "from datetime import timedelta\n",
    "from pathlib import Path\n",
    "from timeit import default_timer as timer\n",
    "from warnings import simplefilter\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "from IPython.display import HTML, display\n",
    "from sklearn import set_config\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.dummy import DummyClassifier\n",
    "from sklearn.ensemble import ExtraTreesClassifier, RandomForestClassifier, StackingClassifier\n",
    "from sklearn.exceptions import ConvergenceWarning, UndefinedMetricWarning\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import (accuracy_score, f1_score, make_scorer,\n",
    "                             plot_confusion_matrix, precision_score,\n",
    "                             recall_score)\n",
    "from sklearn.model_selection import (GridSearchCV, RandomizedSearchCV,\n",
    "                                     train_test_split)\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.preprocessing import (FunctionTransformer, MinMaxScaler,\n",
    "                                   OneHotEncoder, RobustScaler, StandardScaler)\n",
    "from sklearn.svm import SVC, LinearSVC\n",
    "from sklearn.tree import DecisionTreeClassifier"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "P6_NtKfyBK9q",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "## Configurações"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "WY7RrC2dBK9q"
   },
   "outputs": [],
   "source": [
    "seed = 0\n",
    "data_path = Path(\"../desafio/assets/data/\")\n",
    "source_url = \"https://github.com/maratonadev/desafio-5-2021/raw/main/assets/data/\"\n",
    "set_config(display=\"diagram\")\n",
    "\n",
    "# See https://stackoverflow.com/a/55595680/2062663\n",
    "if not sys.warnoptions:\n",
    "    simplefilter(\"ignore\", category=ConvergenceWarning)\n",
    "    simplefilter(\"ignore\", category=UndefinedMetricWarning)\n",
    "    os.environ[\"PYTHONWARNINGS\"] = \"ignore\" # Also affect subprocesses"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zdlPHVPNBK9r",
    "tags": []
   },
   "source": [
    "## Dados"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "QFH1GpA9BK9s",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Download"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "buzbRDQDBK9t",
    "outputId": "3f15674a-1701-4f8b-a341-1bfc92860f69"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ANSWERS.csv  dataset.csv\n"
     ]
    }
   ],
   "source": [
    "!mkdir -p {data_path}\n",
    "!wget -q -N -P {data_path} {source_url}\"dataset.csv\"\n",
    "!wget -q -N -P {data_path} {source_url}\"ANSWERS.csv\"\n",
    "!ls {data_path}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "xngCz4yHBK9u",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Carregamento"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "DICgufEPBK9v",
    "outputId": "2e3987c4-ab11-4f2e-f6b0-8eccfde7309b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['ID', 'GENDER', 'SENIORCITIZEN', 'PARTNER', 'DEPENDENTS', 'TENURE', 'PHONESERVICE', 'MULTIPLELINES', 'INTERNETSERVICE', 'ONLINESECURITY', 'ONLINEBACKUP', 'DEVICEPROTECTION', 'TECHSUPPORT', 'STREAMINGTV', 'STREAMINGMOVIES', 'CONTRACT', 'PAPERLESSBILLING', 'PAYMENTMETHOD', 'MONTHLYCHARGES', 'TOTALCHARGES', 'CHURN']\n"
     ]
    }
   ],
   "source": [
    "df_all = pd.read_csv(data_path / \"dataset.csv\")\n",
    "answers = pd.read_csv(data_path / \"ANSWERS.csv\")\n",
    "\n",
    "assert list(answers.columns) == list(df_all.columns)\n",
    "all_cols = list(df_all.columns)\n",
    "print(all_cols)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LPDbHk0pBK9v",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Informações gerais"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "UZEPUEYaBK9w",
    "outputId": "28ca7128-6d4e-4eab-f515-3d3b8da03727"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 4548 entries, 0 to 4547\n",
      "Data columns (total 21 columns):\n",
      " #   Column            Non-Null Count  Dtype  \n",
      "---  ------            --------------  -----  \n",
      " 0   ID                4548 non-null   int64  \n",
      " 1   GENDER            4520 non-null   object \n",
      " 2   SENIORCITIZEN     4523 non-null   float64\n",
      " 3   PARTNER           4524 non-null   object \n",
      " 4   DEPENDENTS        4526 non-null   object \n",
      " 5   TENURE            4533 non-null   float64\n",
      " 6   PHONESERVICE      4520 non-null   object \n",
      " 7   MULTIPLELINES     4523 non-null   object \n",
      " 8   INTERNETSERVICE   4524 non-null   object \n",
      " 9   ONLINESECURITY    4518 non-null   object \n",
      " 10  ONLINEBACKUP      4518 non-null   object \n",
      " 11  DEVICEPROTECTION  4524 non-null   object \n",
      " 12  TECHSUPPORT       4530 non-null   object \n",
      " 13  STREAMINGTV       4523 non-null   object \n",
      " 14  STREAMINGMOVIES   4523 non-null   object \n",
      " 15  CONTRACT          4525 non-null   object \n",
      " 16  PAPERLESSBILLING  4517 non-null   object \n",
      " 17  PAYMENTMETHOD     4520 non-null   object \n",
      " 18  MONTHLYCHARGES    4527 non-null   float64\n",
      " 19  TOTALCHARGES      4527 non-null   object \n",
      " 20  CHURN             4548 non-null   object \n",
      "dtypes: float64(3), int64(1), object(17)\n",
      "memory usage: 746.3+ KB\n"
     ]
    }
   ],
   "source": [
    "df_all.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "yxVHe072BK9w"
   },
   "source": [
    "- A maioria das variáveis (colunas) parece não ser numérica e precisará ser tratada antes do uso em modelos de machine learning.\n",
    "- Com exceção da variável que identifica cada cliente (`ID`) e da variável de destino (`CHURN`), todas têm dados faltantes (nulos)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "tMAWjiK4BK9x",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Divisão dos dados entre treino e teste"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Iv_0jnyKBK9x"
   },
   "source": [
    "Antes de inspecionar os dados, serão reservados alguns exemplos para teste, que não serão vistos durante a análise ou modelagem dos dados:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "3qlTTAJfBK9y",
    "outputId": "06bfc2f5-6378-4cb5-86cd-d65fad9fbf4d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dimensões dos dados de treino: (4048, 21)\n",
      "Dimensões dos dados de teste: (500, 21)\n"
     ]
    }
   ],
   "source": [
    "test_size = 500\n",
    "target = \"CHURN\"\n",
    "df_train, df_test = train_test_split(\n",
    "    df_all, test_size=test_size, random_state=seed, stratify=df_all[[target]]\n",
    ")\n",
    "print(f\"Dimensões dos dados de treino: {df_train.shape}\")\n",
    "print(f\"Dimensões dos dados de teste: {df_test.shape}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "qET4zpDuBK9y",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Variável destino"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0y0fQwYaBK9y"
   },
   "source": [
    "A variável destino para este desafio é `CHURN`, que indica se o cliente foi perdido ou não. Vamos dar uma olhada em como está a distribuição dessa variável:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 279
    },
    "id": "d07jx2dZBK9z",
    "outputId": "41597eb1-8ee4-4338-f98e-0190f00cd862"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "churn_plot = sns.countplot(\n",
    "    data=df_train, x=target, order=df_train[target].value_counts().index\n",
    ")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 112
    },
    "id": "09iQYH1iBK9z",
    "outputId": "d1f43023-73ac-4b8f-9858-361166bb5e65"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Proporção</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>No</th>\n",
       "      <td>0.733696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Yes</th>\n",
       "      <td>0.266304</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Proporção\n",
       "No    0.733696\n",
       "Yes   0.266304"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train[target].value_counts(normalize=True).rename(\"Proporção\").to_frame()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ui3q268SBK90"
   },
   "source": [
    "- A maioria dos clientes (cerca de 73%) não foi perdido"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "eTQENgInBK90",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Dados de exemplo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 522
    },
    "id": "27VImZQvBK90",
    "outputId": "8f625dac-9ab6-4576-c0dd-89dc6af9b5dc"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</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>263</th>\n",
       "      <td>6155</td>\n",
       "      <td>Male</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>66.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>Two year</td>\n",
       "      <td>No</td>\n",
       "      <td>Credit card (automatic)</td>\n",
       "      <td>34.34</td>\n",
       "      <td>1283.49</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2533</th>\n",
       "      <td>9074</td>\n",
       "      <td>Male</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>DSL</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>12.23</td>\n",
       "      <td>12.23</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>490</th>\n",
       "      <td>1780</td>\n",
       "      <td>Female</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>9.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</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>Mailed check</td>\n",
       "      <td>108.93</td>\n",
       "      <td>554.83</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3655</th>\n",
       "      <td>7462</td>\n",
       "      <td>Female</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>61.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Fiber optic</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>One year</td>\n",
       "      <td>Yes</td>\n",
       "      <td>NaN</td>\n",
       "      <td>118.86</td>\n",
       "      <td>6437.91</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2709</th>\n",
       "      <td>4493</td>\n",
       "      <td>Female</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>DSL</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>Mailed check</td>\n",
       "      <td>116.13</td>\n",
       "      <td>116.13</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2239</th>\n",
       "      <td>8619</td>\n",
       "      <td>Female</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>64.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>One year</td>\n",
       "      <td>No</td>\n",
       "      <td>Bank transfer (automatic)</td>\n",
       "      <td>20.66</td>\n",
       "      <td>1337.51</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1234</th>\n",
       "      <td>9229</td>\n",
       "      <td>Female</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>9.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>No internet service</td>\n",
       "      <td>Two year</td>\n",
       "      <td>No</td>\n",
       "      <td>Mailed check</td>\n",
       "      <td>32.08</td>\n",
       "      <td>202.83</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>551</th>\n",
       "      <td>5658</td>\n",
       "      <td>Female</td>\n",
       "      <td>1.0</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>1.0</td>\n",
       "      <td>No</td>\n",
       "      <td>No phone service</td>\n",
       "      <td>DSL</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>Mailed check</td>\n",
       "      <td>18.00</td>\n",
       "      <td>18.0</td>\n",
       "      <td>Yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3594</th>\n",
       "      <td>7884</td>\n",
       "      <td>Female</td>\n",
       "      <td>1.0</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>6.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Fiber optic</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>NaN</td>\n",
       "      <td>No</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Electronic check</td>\n",
       "      <td>205.27</td>\n",
       "      <td>582.07</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>266</th>\n",
       "      <td>4843</td>\n",
       "      <td>Female</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>68.0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>DSL</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Two year</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Electronic check</td>\n",
       "      <td>1.29</td>\n",
       "      <td>6061.59</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        ID  GENDER  SENIORCITIZEN PARTNER DEPENDENTS  TENURE PHONESERVICE  \\\n",
       "263   6155    Male            0.0      No        Yes    66.0          Yes   \n",
       "2533  9074    Male            0.0      No        Yes     1.0          Yes   \n",
       "490   1780  Female            0.0     Yes        Yes     9.0          Yes   \n",
       "3655  7462  Female            0.0      No         No    61.0          Yes   \n",
       "2709  4493  Female            0.0      No         No     1.0          Yes   \n",
       "2239  8619  Female            0.0     Yes        Yes    64.0          Yes   \n",
       "1234  9229  Female            0.0     Yes        Yes     9.0          Yes   \n",
       "551   5658  Female            1.0      No         No     1.0           No   \n",
       "3594  7884  Female            1.0      No         No     6.0          Yes   \n",
       "266   4843  Female            0.0      No         No    68.0          Yes   \n",
       "\n",
       "         MULTIPLELINES INTERNETSERVICE       ONLINESECURITY  \\\n",
       "263                 No              No  No internet service   \n",
       "2533               Yes             DSL                   No   \n",
       "490                Yes             DSL                   No   \n",
       "3655               Yes     Fiber optic                   No   \n",
       "2709                No             DSL                   No   \n",
       "2239                No              No  No internet service   \n",
       "1234                No              No  No internet service   \n",
       "551   No phone service             DSL                   No   \n",
       "3594                No     Fiber optic                  Yes   \n",
       "266                Yes             DSL                  Yes   \n",
       "\n",
       "             ONLINEBACKUP     DEVICEPROTECTION          TECHSUPPORT  \\\n",
       "263   No internet service  No internet service  No internet service   \n",
       "2533                   No                   No                   No   \n",
       "490                   Yes                   No                   No   \n",
       "3655                  Yes                  Yes                   No   \n",
       "2709                   No                   No                   No   \n",
       "2239  No internet service  No internet service  No internet service   \n",
       "1234  No internet service  No internet service  No internet service   \n",
       "551                    No                   No                   No   \n",
       "3594                   No                   No                   No   \n",
       "266                   Yes                  Yes                  Yes   \n",
       "\n",
       "              STREAMINGTV      STREAMINGMOVIES        CONTRACT  \\\n",
       "263   No internet service  No internet service        Two year   \n",
       "2533                   No                   No  Month-to-month   \n",
       "490                    No                   No  Month-to-month   \n",
       "3655                  Yes                  Yes        One year   \n",
       "2709                   No                   No  Month-to-month   \n",
       "2239  No internet service  No internet service        One year   \n",
       "1234  No internet service  No internet service        Two year   \n",
       "551                    No                   No  Month-to-month   \n",
       "3594                  NaN                   No  Month-to-month   \n",
       "266                   Yes                  Yes        Two year   \n",
       "\n",
       "     PAPERLESSBILLING              PAYMENTMETHOD  MONTHLYCHARGES TOTALCHARGES  \\\n",
       "263                No    Credit card (automatic)           34.34      1283.49   \n",
       "2533              Yes           Electronic check           12.23        12.23   \n",
       "490               Yes               Mailed check          108.93       554.83   \n",
       "3655              Yes                        NaN          118.86      6437.91   \n",
       "2709              Yes               Mailed check          116.13       116.13   \n",
       "2239               No  Bank transfer (automatic)           20.66      1337.51   \n",
       "1234               No               Mailed check           32.08       202.83   \n",
       "551               Yes               Mailed check           18.00         18.0   \n",
       "3594              Yes           Electronic check          205.27       582.07   \n",
       "266               Yes           Electronic check            1.29      6061.59   \n",
       "\n",
       "     CHURN  \n",
       "263     No  \n",
       "2533    No  \n",
       "490     No  \n",
       "3655    No  \n",
       "2709    No  \n",
       "2239    No  \n",
       "1234    No  \n",
       "551    Yes  \n",
       "3594    No  \n",
       "266     No  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with pd.option_context(\"display.max_columns\", None):\n",
    "    display(df_train.sample(10, random_state=seed))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "mliLDZXMBK91",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Valores distintos por variável"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 708
    },
    "id": "DRKm88pWBK91",
    "outputId": "67e07e1f-2d99-4528-8e7c-b677550a6eb7"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Nº de valores</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ID</th>\n",
       "      <td>4048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GENDER</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SENIORCITIZEN</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PARTNER</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DEPENDENTS</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TENURE</th>\n",
       "      <td>73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PHONESERVICE</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MULTIPLELINES</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>INTERNETSERVICE</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ONLINESECURITY</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ONLINEBACKUP</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DEVICEPROTECTION</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TECHSUPPORT</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STREAMINGTV</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STREAMINGMOVIES</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CONTRACT</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAPERLESSBILLING</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAYMENTMETHOD</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MONTHLYCHARGES</th>\n",
       "      <td>3671</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TOTALCHARGES</th>\n",
       "      <td>4002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CHURN</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  Nº de valores\n",
       "ID                         4048\n",
       "GENDER                        2\n",
       "SENIORCITIZEN                 2\n",
       "PARTNER                       2\n",
       "DEPENDENTS                    2\n",
       "TENURE                       73\n",
       "PHONESERVICE                  2\n",
       "MULTIPLELINES                 3\n",
       "INTERNETSERVICE               3\n",
       "ONLINESECURITY                3\n",
       "ONLINEBACKUP                  3\n",
       "DEVICEPROTECTION              3\n",
       "TECHSUPPORT                   3\n",
       "STREAMINGTV                   3\n",
       "STREAMINGMOVIES               3\n",
       "CONTRACT                      3\n",
       "PAPERLESSBILLING              2\n",
       "PAYMENTMETHOD                 4\n",
       "MONTHLYCHARGES             3671\n",
       "TOTALCHARGES               4002\n",
       "CHURN                         2"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.nunique().rename(\"Nº de valores\").to_frame()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LIzvBhPEBK91"
   },
   "source": [
    "Observações:\n",
    "- As colunas `ID`, `MONTHLYCHARGES`, `TENURE` e `TOTALCHARGES` assumem um grande número de valores distintos, e possivelmente são contínuas\n",
    "- As demais colunas têm poucos valores distintos, e podemos assumir que são categóricas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fTSc1eBJBK91",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Variáveis categóricas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "xTsg-6t0BK92"
   },
   "source": [
    "Já que teremos que lidar tanto com variáveis categóricas quanto numéricas, e cada tipo de variável precisa de tratamentos específicos antes de serem usadas por modelos de _machine learning_, processaremos cada tipo por `Pipeline` específica do `scikit-learn`. Aqui, será construída uma versão preliminar da _pipeline_ que lidará com os dados categóricos, apenas para facilitar a visualização dos dados. Posteriormente ela será aperfeiçoada para uso nos modelos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 212
    },
    "id": "VZhaA537BK92",
    "outputId": "2069d404-8e75-464c-9a9b-b7fc940d6405"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>GENDER</th>\n",
       "      <th>SENIORCITIZEN</th>\n",
       "      <th>PARTNER</th>\n",
       "      <th>DEPENDENTS</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>CHURN</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4022</td>\n",
       "      <td>4025.0</td>\n",
       "      <td>4025</td>\n",
       "      <td>4028</td>\n",
       "      <td>4022</td>\n",
       "      <td>4024</td>\n",
       "      <td>4026</td>\n",
       "      <td>4021</td>\n",
       "      <td>4022</td>\n",
       "      <td>4027</td>\n",
       "      <td>4031</td>\n",
       "      <td>4028</td>\n",
       "      <td>4025</td>\n",
       "      <td>4028</td>\n",
       "      <td>4023</td>\n",
       "      <td>4023</td>\n",
       "      <td>4048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>Female</td>\n",
       "      <td>0.0</td>\n",
       "      <td>No</td>\n",
       "      <td>No</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>Yes</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Electronic check</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>2016</td>\n",
       "      <td>3364.0</td>\n",
       "      <td>2053</td>\n",
       "      <td>2822</td>\n",
       "      <td>3648</td>\n",
       "      <td>1980</td>\n",
       "      <td>1769</td>\n",
       "      <td>2009</td>\n",
       "      <td>1780</td>\n",
       "      <td>1769</td>\n",
       "      <td>1985</td>\n",
       "      <td>1589</td>\n",
       "      <td>1587</td>\n",
       "      <td>2265</td>\n",
       "      <td>2387</td>\n",
       "      <td>1351</td>\n",
       "      <td>2970</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        GENDER  SENIORCITIZEN PARTNER DEPENDENTS PHONESERVICE MULTIPLELINES  \\\n",
       "count     4022         4025.0    4025       4028         4022          4024   \n",
       "unique       2            2.0       2          2            2             3   \n",
       "top     Female            0.0      No         No          Yes            No   \n",
       "freq      2016         3364.0    2053       2822         3648          1980   \n",
       "\n",
       "       INTERNETSERVICE ONLINESECURITY ONLINEBACKUP DEVICEPROTECTION  \\\n",
       "count             4026           4021         4022             4027   \n",
       "unique               3              3            3                3   \n",
       "top        Fiber optic             No           No               No   \n",
       "freq              1769           2009         1780             1769   \n",
       "\n",
       "       TECHSUPPORT STREAMINGTV STREAMINGMOVIES        CONTRACT  \\\n",
       "count         4031        4028            4025            4028   \n",
       "unique           3           3               3               3   \n",
       "top             No          No             Yes  Month-to-month   \n",
       "freq          1985        1589            1587            2265   \n",
       "\n",
       "       PAPERLESSBILLING     PAYMENTMETHOD CHURN  \n",
       "count              4023              4023  4048  \n",
       "unique                2                 4     2  \n",
       "top                 Yes  Electronic check    No  \n",
       "freq               2387              1351  2970  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "categorical_features = [\n",
    "    \"GENDER\",\n",
    "    \"SENIORCITIZEN\",\n",
    "    \"PARTNER\",\n",
    "    \"DEPENDENTS\",\n",
    "    \"PHONESERVICE\",\n",
    "    \"MULTIPLELINES\",\n",
    "    \"INTERNETSERVICE\",\n",
    "    \"ONLINESECURITY\",\n",
    "    \"ONLINEBACKUP\",\n",
    "    \"DEVICEPROTECTION\",\n",
    "    \"TECHSUPPORT\",\n",
    "    \"STREAMINGTV\",\n",
    "    \"STREAMINGMOVIES\",\n",
    "    \"CONTRACT\",\n",
    "    \"PAPERLESSBILLING\",\n",
    "    \"PAYMENTMETHOD\",\n",
    "    # 'CHURN',  # This is the binary target, not a feature\n",
    "]\n",
    "\n",
    "\n",
    "def cat_mapper(df):\n",
    "    return df.astype(\"category\")\n",
    "\n",
    "\n",
    "categorical_transformer = Pipeline(\n",
    "    steps=[\n",
    "        (\"to_categorical\", FunctionTransformer(cat_mapper)),\n",
    "    ]\n",
    ")\n",
    "\n",
    "df_train_cat = categorical_transformer.fit_transform(\n",
    "    df_train[categorical_features + [target]]\n",
    ")\n",
    "\n",
    "with pd.option_context(\"display.max_columns\", None):\n",
    "    display(df_train_cat.describe())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "id": "g5uQwNRcBK93"
   },
   "outputs": [],
   "source": [
    "def display_value_counts(df):\n",
    "    for col in df:\n",
    "        print(f\"{col}:\")\n",
    "        display(\n",
    "            df[col]\n",
    "            .value_counts()\n",
    "            .reset_index()\n",
    "            .sort_values([col, \"index\"], ascending=False)\n",
    "            .rename(columns={col: \"Quantidade\", \"index\": \"Valor\"})\n",
    "            .set_index(\"Valor\")\n",
    "            .T\n",
    "        )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "L2NxSBuUBK93"
   },
   "source": [
    "Estes são os valores que as variáveis categóricas assumem:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "fy_oHA3nBK93",
    "outputId": "048ce7f0-9355-4fa5-c135-cc711a42ac37"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GENDER:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>Female</th>\n",
       "      <th>Male</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>2016</td>\n",
       "      <td>2006</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor       Female  Male\n",
       "Quantidade    2016  2006"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SENIORCITIZEN:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>0.0</th>\n",
       "      <th>1.0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>3364</td>\n",
       "      <td>661</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor        0.0  1.0\n",
       "Quantidade  3364  661"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PARTNER:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>No</th>\n",
       "      <th>Yes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>2053</td>\n",
       "      <td>1972</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor         No   Yes\n",
       "Quantidade  2053  1972"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DEPENDENTS:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>No</th>\n",
       "      <th>Yes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>2822</td>\n",
       "      <td>1206</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor         No   Yes\n",
       "Quantidade  2822  1206"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PHONESERVICE:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>Yes</th>\n",
       "      <th>No</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>3648</td>\n",
       "      <td>374</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor        Yes   No\n",
       "Quantidade  3648  374"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MULTIPLELINES:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>No</th>\n",
       "      <th>Yes</th>\n",
       "      <th>No phone service</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>1980</td>\n",
       "      <td>1672</td>\n",
       "      <td>372</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor         No   Yes  No phone service\n",
       "Quantidade  1980  1672               372"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INTERNETSERVICE:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>Fiber optic</th>\n",
       "      <th>DSL</th>\n",
       "      <th>No</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>1769</td>\n",
       "      <td>1366</td>\n",
       "      <td>891</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor       Fiber optic   DSL   No\n",
       "Quantidade         1769  1366  891"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ONLINESECURITY:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>No</th>\n",
       "      <th>Yes</th>\n",
       "      <th>No internet service</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>2009</td>\n",
       "      <td>1119</td>\n",
       "      <td>893</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor         No   Yes  No internet service\n",
       "Quantidade  2009  1119                  893"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ONLINEBACKUP:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>No</th>\n",
       "      <th>Yes</th>\n",
       "      <th>No internet service</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>1780</td>\n",
       "      <td>1350</td>\n",
       "      <td>892</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor         No   Yes  No internet service\n",
       "Quantidade  1780  1350                  892"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DEVICEPROTECTION:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>No</th>\n",
       "      <th>Yes</th>\n",
       "      <th>No internet service</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>1769</td>\n",
       "      <td>1367</td>\n",
       "      <td>891</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor         No   Yes  No internet service\n",
       "Quantidade  1769  1367                  891"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TECHSUPPORT:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>No</th>\n",
       "      <th>Yes</th>\n",
       "      <th>No internet service</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>1985</td>\n",
       "      <td>1154</td>\n",
       "      <td>892</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor         No   Yes  No internet service\n",
       "Quantidade  1985  1154                  892"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "STREAMINGTV:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>No</th>\n",
       "      <th>Yes</th>\n",
       "      <th>No internet service</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>1589</td>\n",
       "      <td>1545</td>\n",
       "      <td>894</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor         No   Yes  No internet service\n",
       "Quantidade  1589  1545                  894"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "STREAMINGMOVIES:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>Yes</th>\n",
       "      <th>No</th>\n",
       "      <th>No internet service</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>1587</td>\n",
       "      <td>1548</td>\n",
       "      <td>890</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor        Yes    No  No internet service\n",
       "Quantidade  1587  1548                  890"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CONTRACT:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>Month-to-month</th>\n",
       "      <th>Two year</th>\n",
       "      <th>One year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>2265</td>\n",
       "      <td>950</td>\n",
       "      <td>813</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor       Month-to-month  Two year  One year\n",
       "Quantidade            2265       950       813"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PAPERLESSBILLING:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>Yes</th>\n",
       "      <th>No</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>2387</td>\n",
       "      <td>1636</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor        Yes    No\n",
       "Quantidade  2387  1636"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PAYMENTMETHOD:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>Electronic check</th>\n",
       "      <th>Bank transfer (automatic)</th>\n",
       "      <th>Mailed check</th>\n",
       "      <th>Credit card (automatic)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>1351</td>\n",
       "      <td>902</td>\n",
       "      <td>901</td>\n",
       "      <td>869</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor       Electronic check  Bank transfer (automatic)  Mailed check  \\\n",
       "Quantidade              1351                        902           901   \n",
       "\n",
       "Valor       Credit card (automatic)  \n",
       "Quantidade                      869  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display_value_counts(df_train[categorical_features])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zwkzmbXtBK94"
   },
   "source": [
    "Observações:\n",
    "- Antes de serem utilizadas por um modelo, será conveniente converter essas variáveis em valores numéricos usando \"One-Hot-Encoding\".\n",
    "- Como há valores faltantes para essas variáveis (vide [informações gerais](#Informações-gerais)), será importante imputar algum valor (por exemplo, o que for mais frequente em cada coluna)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "32_K4JjjBK94",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Variáveis numéricas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "GJBa4vbKBK94"
   },
   "source": [
    "Por hora, apenas para visualizar estatísticas sobre os valores numéricos, valores textuais nessas colunas serão trocados por `NaN` usando um *transformer* personalizado:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 300
    },
    "id": "wXu9WeY-BK94",
    "outputId": "e1d00849-1b73-4e9d-c629-b57a755e3c4e"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>MONTHLYCHARGES</th>\n",
       "      <th>TENURE</th>\n",
       "      <th>TOTALCHARGES</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4048.000000</td>\n",
       "      <td>4030.000000</td>\n",
       "      <td>4034.000000</td>\n",
       "      <td>4025.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5458.245306</td>\n",
       "      <td>95.130290</td>\n",
       "      <td>31.853991</td>\n",
       "      <td>2265.379839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2606.169444</td>\n",
       "      <td>75.697319</td>\n",
       "      <td>24.447564</td>\n",
       "      <td>2258.428774</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1004.000000</td>\n",
       "      <td>0.090000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.170000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>3199.750000</td>\n",
       "      <td>33.837500</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>415.740000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5435.500000</td>\n",
       "      <td>71.935000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>1400.900000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>7705.750000</td>\n",
       "      <td>142.977500</td>\n",
       "      <td>54.000000</td>\n",
       "      <td>3652.090000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>9998.000000</td>\n",
       "      <td>345.240000</td>\n",
       "      <td>72.000000</td>\n",
       "      <td>8734.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                ID  MONTHLYCHARGES       TENURE  TOTALCHARGES\n",
       "count  4048.000000     4030.000000  4034.000000   4025.000000\n",
       "mean   5458.245306       95.130290    31.853991   2265.379839\n",
       "std    2606.169444       75.697319    24.447564   2258.428774\n",
       "min    1004.000000        0.090000     0.000000      0.170000\n",
       "25%    3199.750000       33.837500     8.000000    415.740000\n",
       "50%    5435.500000       71.935000    28.000000   1400.900000\n",
       "75%    7705.750000      142.977500    54.000000   3652.090000\n",
       "max    9998.000000      345.240000    72.000000   8734.000000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "numeric_features = [\n",
    "    # 'ID',  # This is just an identifier, not a feature\n",
    "    \"MONTHLYCHARGES\",\n",
    "    \"TENURE\",\n",
    "    \"TOTALCHARGES\",\n",
    "]\n",
    "\n",
    "\n",
    "def num_mapper(df):\n",
    "    def col_mapper(col):\n",
    "        return pd.to_numeric(col, errors=\"coerce\")\n",
    "\n",
    "    return df.apply(col_mapper)\n",
    "\n",
    "\n",
    "numeric_transformer = Pipeline(\n",
    "    steps=[\n",
    "        (\"to_numeric\", FunctionTransformer(num_mapper)),\n",
    "    ]\n",
    ")\n",
    "\n",
    "df_train_num = numeric_transformer.fit_transform(df_train[[\"ID\"] + numeric_features])\n",
    "\n",
    "df_train_clean = pd.concat([df_train_cat, df_train_num], axis=1)[\n",
    "    df_train.columns\n",
    "].copy()\n",
    "\n",
    "with pd.option_context(\"display.max_columns\", None):\n",
    "    display(df_train_num.describe())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "oXJc6cw6BK95"
   },
   "source": [
    "Estes são os valores que as variáveis numéricas assumem:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 459
    },
    "id": "OoNeRL8JBK95",
    "outputId": "81940ecb-54a2-40fb-abbc-780d3518d916"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ID:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>9998</th>\n",
       "      <th>9997</th>\n",
       "      <th>9995</th>\n",
       "      <th>...</th>\n",
       "      <th>1008</th>\n",
       "      <th>1006</th>\n",
       "      <th>1004</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 4048 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor       9998  9997  9995  ...  1008  1006  1004\n",
       "Quantidade     1     1     1  ...     1     1     1\n",
       "\n",
       "[1 rows x 4048 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MONTHLYCHARGES:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>33.22</th>\n",
       "      <th>24.03</th>\n",
       "      <th>13.21</th>\n",
       "      <th>...</th>\n",
       "      <th>0.17</th>\n",
       "      <th>0.16</th>\n",
       "      <th>0.09</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 3671 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor       33.22  24.03  13.21  ...  0.17   0.16   0.09 \n",
       "Quantidade      4      4      4  ...      1      1      1\n",
       "\n",
       "[1 rows x 3671 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TENURE:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>1.0</th>\n",
       "      <th>72.0</th>\n",
       "      <th>2.0</th>\n",
       "      <th>...</th>\n",
       "      <th>59.0</th>\n",
       "      <th>38.0</th>\n",
       "      <th>0.0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>370</td>\n",
       "      <td>192</td>\n",
       "      <td>125</td>\n",
       "      <td>...</td>\n",
       "      <td>28</td>\n",
       "      <td>25</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 73 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor       1.0   72.0  2.0   ...  59.0  38.0  0.0 \n",
       "Quantidade   370   192   125  ...    28    25     3\n",
       "\n",
       "[1 rows x 73 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TOTALCHARGES:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Valor</th>\n",
       "      <th>4452.81</th>\n",
       "      <th>1475.30</th>\n",
       "      <th>1264.95</th>\n",
       "      <th>...</th>\n",
       "      <th>0.36</th>\n",
       "      <th>0.30</th>\n",
       "      <th>0.17</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Quantidade</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 4001 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Valor       4452.81  1475.30  1264.95  ...  0.36     0.30     0.17   \n",
       "Quantidade        2        2        2  ...        1        1        1\n",
       "\n",
       "[1 rows x 4001 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with pd.option_context(\"display.max_columns\", 6):\n",
    "    display_value_counts(df_train_num)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "xqz3lg_pBK95",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Correlações"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 550
    },
    "id": "NmsK9UOsBK95",
    "outputId": "ffa18c1d-8111-46c3-e995-be422c741cc3"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 592x540 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(\n",
    "    df_train_clean[numeric_features + [target]], hue=target, palette=\"Set1\", corner=True\n",
    ")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "9oaOm08FBK95",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Variáveis categóricas (gráficos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "F2QBH9NPBK96",
    "outputId": "b87685dc-e5c3-49cc-a499-027445dd9a5d"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x2160 with 16 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(nrows=len(categorical_features), sharex=True, figsize=(10, 30))\n",
    "for i, col in enumerate(categorical_features):\n",
    "    group_means = (\n",
    "        (df_train_clean.groupby([col])[target].count() / len(df_train_clean))\n",
    "        .rename(f\"{target}_RATE\")\n",
    "        .sort_values(ascending=False)\n",
    "        .to_frame()\n",
    "    )\n",
    "    sns.barplot(x=f\"{target}_RATE\", y=group_means.index, data=group_means, ax=axes[i])\n",
    "    fig.subplots_adjust(top=0.97)\n",
    "    fig.suptitle(\"Proporção de CHURN entre os clientes de cada grupo\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vranf1u5BK96",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### Pré-processamento dos dados"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "uqhlQCzfBK96"
   },
   "source": [
    "Será criada uma _pipeline_ de pré-processamento com transformações específicas para cada tipo de variável:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "id": "g2c9zUMRBK96"
   },
   "outputs": [],
   "source": [
    "categorical_transformer = Pipeline(\n",
    "    steps=[\n",
    "        (\"imputer\", SimpleImputer(strategy=\"most_frequent\")),\n",
    "        (\"onehot\", OneHotEncoder(handle_unknown=\"ignore\")),\n",
    "    ]\n",
    ")\n",
    "\n",
    "numeric_transformer = Pipeline(\n",
    "    steps=[\n",
    "        (\"to_numeric\", FunctionTransformer(num_mapper)),\n",
    "        (\"imputer\", SimpleImputer(strategy=\"constant\", fill_value=0)),\n",
    "        (\"scaler\", StandardScaler()),\n",
    "    ]\n",
    ")\n",
    "\n",
    "preprocessor = ColumnTransformer(\n",
    "    transformers=[\n",
    "        (\"num\", numeric_transformer, numeric_features),\n",
    "        (\"cat\", categorical_transformer, categorical_features),\n",
    "    ]\n",
    ")\n",
    "\n",
    "# preprocessor.fit(df_train)\n",
    "# onehot_features = list(preprocessor.named_transformers_[\"cat\"][\"onehot\"].get_feature_names())\n",
    "# df_train_preprocessed = pd.DataFrame(\n",
    "#     preprocessor.transform(df_train),\n",
    "#     columns=numeric_features + onehot_features,\n",
    "#     index=df_train.index\n",
    "# )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "## Criação de um modelo"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nosso objetivo é obter um algoritmo de Machine Learning que identifique se um cliente será perdido ou não, e que tenha o maior score F1 possível. Para começar, vamos comparar uma simples árvore de decisão com um modelo fictício, que apenas chuta suas predições (sem levar em conta nenhum feature), para ter valores de referência ao buscar uma solução ótima em termos da métrica F1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "id": "tb7qaXCNBK97"
   },
   "outputs": [],
   "source": [
    "pipe = Pipeline(\n",
    "    [(\"preprocessor\", preprocessor), (\"clf\", DecisionTreeClassifier(random_state=seed))]\n",
    ")\n",
    "# display(pipe)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "oq7YoWGPBK97"
   },
   "source": [
    "Abaixo, separamos os dados que queremos predizer dos dados que utilizamos como informações para a predição."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "WH5WQT5mBK97",
    "outputId": "4486b338-afcc-4d92-812a-4b92d44b2c9d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(4048, 21)\n",
      "(4048,)\n",
      "(500, 21)\n",
      "(500,)\n"
     ]
    }
   ],
   "source": [
    "X_train = df_train[all_cols]\n",
    "y_train = df_train[target]\n",
    "X_test = df_test[all_cols]\n",
    "y_test = df_test[target]\n",
    "print(X_train.shape)\n",
    "print(y_train.shape)\n",
    "print(X_test.shape)\n",
    "print(y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "id": "vzQYXh_oBK98"
   },
   "outputs": [],
   "source": [
    "param_grid = [\n",
    "    {\n",
    "        \"clf\": [\n",
    "            SVC(\n",
    "                kernel=\"rbf\",\n",
    "                C=1289890.261253308,\n",
    "                gamma=2.381685551976161e-10,\n",
    "                max_iter=1000,\n",
    "                random_state=seed,\n",
    "            )\n",
    "        ],\n",
    "        \"preprocessor__num__imputer__strategy\": [\"constant\"],\n",
    "        \"preprocessor__num__imputer__fill_value\": [-1],\n",
    "        \"preprocessor__num__scaler\": [RobustScaler()],\n",
    "    },\n",
    "    {\n",
    "        \"clf\": [GaussianNB()],\n",
    "        \"preprocessor__num__imputer__strategy\": [\"mean\"],\n",
    "        \"preprocessor__num__scaler\": [\"passthrough\"],\n",
    "    },\n",
    "    {\n",
    "        \"clf\": [\n",
    "            MLPClassifier(\n",
    "                activation=\"logistic\",\n",
    "                alpha=1e-09,\n",
    "                hidden_layer_sizes=(39, 36),\n",
    "                learning_rate_init=0.00031622776601683794,\n",
    "                learning_rate=\"constant\",\n",
    "                max_iter=10000,\n",
    "                random_state=seed,\n",
    "            )\n",
    "        ],\n",
    "        \"preprocessor__num__imputer__strategy\": [\"mean\"],\n",
    "        \"preprocessor__num__scaler\": [StandardScaler()],\n",
    "    },\n",
    "    {\n",
    "        \"clf\": [KNeighborsClassifier(n_neighbors=115, weights=\"distance\")],\n",
    "        \"preprocessor__num__imputer__strategy\": [\"most_frequent\"],\n",
    "        \"preprocessor__num__scaler\": [StandardScaler()],\n",
    "    },\n",
    "    {\n",
    "        \"clf\": [\n",
    "            LinearSVC(\n",
    "                C=231.0129700083158, loss=\"hinge\", max_iter=10000, random_state=seed\n",
    "            )\n",
    "        ],\n",
    "        \"preprocessor__num__imputer__strategy\": [\"constant\"],\n",
    "        \"preprocessor__num__imputer__fill_value\": [0],\n",
    "        \"preprocessor__num__scaler\": [RobustScaler()],\n",
    "    },\n",
    "    {\n",
    "        \"clf\": [\n",
    "            LogisticRegression(\n",
    "                C=0.2848035868435802,\n",
    "                max_iter=100000,\n",
    "                penalty=\"l1\",\n",
    "                solver=\"liblinear\",\n",
    "                random_state=seed,\n",
    "            )\n",
    "        ],\n",
    "        \"preprocessor__num__imputer__strategy\": [\"most_frequent\"],\n",
    "        \"preprocessor__num__scaler\": [StandardScaler()],\n",
    "    },\n",
    "    {\n",
    "        \"clf\": [\n",
    "            DecisionTreeClassifier(\n",
    "                criterion=\"entropy\",\n",
    "                max_depth=5,\n",
    "                max_features=0.8,\n",
    "                min_samples_leaf=90,\n",
    "                min_samples_split=55,\n",
    "                random_state=seed,\n",
    "            )\n",
    "        ],\n",
    "        \"preprocessor__num__imputer__strategy\": [\"mean\"],\n",
    "        \"preprocessor__num__scaler\": [\"passthrough\"],\n",
    "    },\n",
    "    {\n",
    "        \"clf\": [\n",
    "            RandomForestClassifier(\n",
    "                max_depth=8,\n",
    "                n_estimators=50,\n",
    "                max_features=0.825,\n",
    "                min_samples_split=5,\n",
    "                min_samples_leaf=5,\n",
    "                criterion=\"entropy\",\n",
    "                random_state=seed,\n",
    "            )\n",
    "        ],\n",
    "        \"preprocessor__num__scaler\": [MinMaxScaler()],\n",
    "        \"preprocessor__num__imputer__strategy\": [\"median\"],\n",
    "    },\n",
    "    {\n",
    "        \"clf\": [StackingClassifier(estimators=[\n",
    "            ('svc', SVC(\n",
    "                kernel=\"rbf\",\n",
    "                C=1289890.261253308,\n",
    "                gamma=2.381685551976161e-10,\n",
    "                max_iter=1000,\n",
    "                random_state=seed,\n",
    "            )),\n",
    "            ('gnb', GaussianNB()),\n",
    "            ('mlp', MLPClassifier(\n",
    "                activation=\"logistic\",\n",
    "                alpha=1e-09,\n",
    "                hidden_layer_sizes=(39, 36),\n",
    "                learning_rate_init=0.00031622776601683794,\n",
    "                learning_rate=\"constant\",\n",
    "                max_iter=10000,\n",
    "                random_state=seed,\n",
    "            )),\n",
    "            ('knn', KNeighborsClassifier(n_neighbors=115, weights=\"distance\")),\n",
    "            ('lsvc', LinearSVC(\n",
    "                C=231.0129700083158, loss=\"hinge\", max_iter=10000, random_state=seed\n",
    "            )),\n",
    "            ('lr', LogisticRegression(\n",
    "                C=0.2848035868435802,\n",
    "                max_iter=100000,\n",
    "                penalty=\"l1\",\n",
    "                solver=\"liblinear\",\n",
    "                random_state=seed,\n",
    "            )),\n",
    "            ('dtc', DecisionTreeClassifier(\n",
    "                criterion=\"entropy\",\n",
    "                max_depth=5,\n",
    "                max_features=0.8,\n",
    "                min_samples_leaf=90,\n",
    "                min_samples_split=55,\n",
    "                random_state=seed,\n",
    "            )),\n",
    "            ('rfc', RandomForestClassifier(\n",
    "                max_depth=8,\n",
    "                n_estimators=50,\n",
    "                max_features=0.825,\n",
    "                min_samples_split=5,\n",
    "                min_samples_leaf=5,\n",
    "                criterion=\"entropy\",\n",
    "                random_state=seed,\n",
    "            )),\n",
    "            ('etc', ExtraTreesClassifier(n_estimators=113, random_state=seed)),\n",
    "        ],\n",
    "        final_estimator=SVC())],\n",
    "        \"preprocessor__num__imputer__strategy\": [\"median\"],\n",
    "        \"preprocessor__num__scaler\": [StandardScaler()],\n",
    "    },\n",
    "    {\n",
    "        \"clf\": [ExtraTreesClassifier(n_estimators=113, random_state=seed)],\n",
    "        \"preprocessor__num__imputer__strategy\": [\"most_frequent\"],\n",
    "        \"preprocessor__num__scaler\": [\"passthrough\"],\n",
    "    },\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "KLA_gUpjBK98",
    "outputId": "dfc84459-bbba-4e1d-b3a5-30bac4f6ec70"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 10 folds for each of 10 candidates, totalling 100 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  34 tasks      | elapsed:    9.9s\n",
      "[Parallel(n_jobs=-1)]: Done 100 out of 100 | elapsed:  2.1min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tempo gasto para treinar o modelo 10 vezes com cada combinação de parâmetros para encontrar a melhor: 0:02:06.038013\n"
     ]
    }
   ],
   "source": [
    "cv_num = 10\n",
    "verbose = 1\n",
    "\n",
    "f1_scorer = make_scorer(f1_score, pos_label=\"Yes\")\n",
    "accuracy_scorer = make_scorer(accuracy_score)\n",
    "precision_scorer = make_scorer(precision_score, pos_label=\"Yes\")\n",
    "recall_scorer = make_scorer(recall_score, pos_label=\"Yes\")\n",
    "scoring = {\n",
    "    \"f1\": f1_scorer,\n",
    "    \"accuracy\": accuracy_scorer,\n",
    "    \"precision\": precision_scorer,\n",
    "    \"recall\": recall_scorer,\n",
    "}\n",
    "scoring_name = \"f1\"\n",
    "\n",
    "cv_train_size = int(((cv_num - 1) / cv_num) * len(X_train))\n",
    "cv_test_size = len(X_train) - cv_train_size\n",
    "\n",
    "search_args = {\n",
    "    \"estimator\": pipe,\n",
    "    \"cv\": cv_num,\n",
    "    \"n_jobs\": -1,\n",
    "    \"verbose\": verbose,\n",
    "    \"scoring\": scoring,\n",
    "    \"refit\": scoring_name,\n",
    "}\n",
    "\n",
    "# cv = RandomizedSearchCV(\n",
    "#     n_iter=3000, param_distributions=param_grid, random_state=seed, **search_args\n",
    "# )\n",
    "cv = GridSearchCV(param_grid=param_grid, **search_args)\n",
    "\n",
    "start = timer()\n",
    "cv.fit(X_train, y_train)\n",
    "end = timer()\n",
    "print(\n",
    "    f\"Tempo gasto para treinar o modelo {cv_num} vezes com cada combinação de parâmetros para encontrar a melhor: {timedelta(seconds=end-start)}\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "exPcenhtBK99",
    "outputId": "b9abc971-be33-41a3-933d-5e4b63f0f91f"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<h4>Dados brutos da validação cruzada</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cada combinação de parâmetros foi utilizada em 10 rodadas de treino com 3643 exemplos e teste com 405.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean_test_f1</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "      <th>mean_fit_time</th>\n",
       "      <th>params</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rank_test_f1</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.615632</td>\n",
       "      <td>0.755679</td>\n",
       "      <td>0.529485</td>\n",
       "      <td>0.736492</td>\n",
       "      <td>1.148629</td>\n",
       "      <td>{'clf': SVC(C=1289890.261253308, gamma=2.381685551976161e-10, max_iter=1000,\n",
       "    random_state=0), 'preprocessor__num__imputer__fill_value': -1, 'preprocessor__num__imputer__strategy': 'constant', 'preprocessor__num__scaler': RobustScaler()}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.598983</td>\n",
       "      <td>0.697388</td>\n",
       "      <td>0.462943</td>\n",
       "      <td>0.848780</td>\n",
       "      <td>0.161320</td>\n",
       "      <td>{'clf': GaussianNB(), 'preprocessor__num__imputer__strategy': 'mean', 'preprocessor__num__scaler': 'passthrough'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.598262</td>\n",
       "      <td>0.796195</td>\n",
       "      <td>0.630311</td>\n",
       "      <td>0.570492</td>\n",
       "      <td>5.382881</td>\n",
       "      <td>{'clf': MLPClassifier(activation='logistic', alpha=1e-09, hidden_layer_sizes=(39, 36),\n",
       "              learning_rate_init=0.00031622776601683794, max_iter=10000,\n",
       "              random_state=0), 'preprocessor__num__imputer__strategy': 'mean', 'preprocessor__num__scaler': StandardScaler()}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.596686</td>\n",
       "      <td>0.791994</td>\n",
       "      <td>0.616304</td>\n",
       "      <td>0.578859</td>\n",
       "      <td>0.239787</td>\n",
       "      <td>{'clf': KNeighborsClassifier(n_neighbors=115, weights='distance'), 'preprocessor__num__imputer__strategy': 'most_frequent', 'preprocessor__num__scaler': StandardScaler()}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.596011</td>\n",
       "      <td>0.797427</td>\n",
       "      <td>0.635717</td>\n",
       "      <td>0.564997</td>\n",
       "      <td>4.094500</td>\n",
       "      <td>{'clf': LinearSVC(C=231.0129700083158, loss='hinge', max_iter=10000, random_state=0), 'preprocessor__num__imputer__fill_value': 0, 'preprocessor__num__imputer__strategy': 'constant', 'preprocessor__num__scaler': RobustScaler()}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.595555</td>\n",
       "      <td>0.803355</td>\n",
       "      <td>0.658888</td>\n",
       "      <td>0.544505</td>\n",
       "      <td>0.354686</td>\n",
       "      <td>{'clf': LogisticRegression(C=0.2848035868435802, max_iter=100000, penalty='l1',\n",
       "                   random_state=0, solver='liblinear'), 'preprocessor__num__imputer__strategy': 'most_frequent', 'preprocessor__num__scaler': StandardScaler()}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.595271</td>\n",
       "      <td>0.791007</td>\n",
       "      <td>0.614106</td>\n",
       "      <td>0.580616</td>\n",
       "      <td>0.204471</td>\n",
       "      <td>{'clf': DecisionTreeClassifier(criterion='entropy', max_depth=5, max_features=0.8,\n",
       "                       min_samples_leaf=90, min_samples_split=55,\n",
       "                       random_state=0), 'preprocessor__num__imputer__strategy': 'mean', 'preprocessor__num__scaler': 'passthrough'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.589631</td>\n",
       "      <td>0.801374</td>\n",
       "      <td>0.655255</td>\n",
       "      <td>0.537063</td>\n",
       "      <td>1.118791</td>\n",
       "      <td>{'clf': RandomForestClassifier(criterion='entropy', max_depth=8, max_features=0.825,\n",
       "                       min_samples_leaf=5, min_samples_split=5, n_estimators=50,\n",
       "                       random_state=0), 'preprocessor__num__imputer__strategy': 'median', 'preprocessor__num__scaler': MinMaxScaler()}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.587559</td>\n",
       "      <td>0.800144</td>\n",
       "      <td>0.652631</td>\n",
       "      <td>0.536120</td>\n",
       "      <td>64.682951</td>\n",
       "      <td>{'clf': StackingClassifier(estimators=[('svc',\n",
       "                                SVC(C=1289890.261253308,\n",
       "                                    gamma=2.381685551976161e-10, max_iter=1000,\n",
       "                                    random_state=0)),\n",
       "                               ('gnb', GaussianNB()),\n",
       "                               ('mlp',\n",
       "                                MLPClassifier(activation='logistic',\n",
       "                                              alpha=1e-09,\n",
       "                                              hidden_layer_sizes=(39, 36),\n",
       "                                              learning_rate_init=0.00031622776601683794,\n",
       "                                              max_iter=10000, random_state=0)),\n",
       "                               ('knn',\n",
       "                                KNeighborsClassifier(n_neighbors=115,\n",
       "                                                     weights=...\n",
       "                                DecisionTreeClassifier(criterion='entropy',\n",
       "                                                       max_depth=5,\n",
       "                                                       max_features=0.8,\n",
       "                                                       min_samples_leaf=90,\n",
       "                                                       min_samples_split=55,\n",
       "                                                       random_state=0)),\n",
       "                               ('rfc',\n",
       "                                RandomForestClassifier(criterion='entropy',\n",
       "                                                       max_depth=8,\n",
       "                                                       max_features=0.825,\n",
       "                                                       min_samples_leaf=5,\n",
       "                                                       min_samples_split=5,\n",
       "                                                       n_estimators=50,\n",
       "                                                       random_state=0)),\n",
       "                               ('etc',\n",
       "                                ExtraTreesClassifier(n_estimators=113,\n",
       "                                                     random_state=0))],\n",
       "                   final_estimator=SVC()), 'preprocessor__num__imputer__strategy': 'median', 'preprocessor__num__scaler': StandardScaler()}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.523538</td>\n",
       "      <td>0.770007</td>\n",
       "      <td>0.584089</td>\n",
       "      <td>0.474939</td>\n",
       "      <td>1.152382</td>\n",
       "      <td>{'clf': ExtraTreesClassifier(n_estimators=113, random_state=0), 'preprocessor__num__imputer__strategy': 'most_frequent', 'preprocessor__num__scaler': 'passthrough'}</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              mean_test_f1  mean_test_accuracy  mean_test_precision  \\\n",
       "rank_test_f1                                                          \n",
       "1                 0.615632            0.755679             0.529485   \n",
       "2                 0.598983            0.697388             0.462943   \n",
       "3                 0.598262            0.796195             0.630311   \n",
       "4                 0.596686            0.791994             0.616304   \n",
       "5                 0.596011            0.797427             0.635717   \n",
       "6                 0.595555            0.803355             0.658888   \n",
       "7                 0.595271            0.791007             0.614106   \n",
       "8                 0.589631            0.801374             0.655255   \n",
       "9                 0.587559            0.800144             0.652631   \n",
       "10                0.523538            0.770007             0.584089   \n",
       "\n",
       "              mean_test_recall  mean_fit_time  \\\n",
       "rank_test_f1                                    \n",
       "1                     0.736492       1.148629   \n",
       "2                     0.848780       0.161320   \n",
       "3                     0.570492       5.382881   \n",
       "4                     0.578859       0.239787   \n",
       "5                     0.564997       4.094500   \n",
       "6                     0.544505       0.354686   \n",
       "7                     0.580616       0.204471   \n",
       "8                     0.537063       1.118791   \n",
       "9                     0.536120      64.682951   \n",
       "10                    0.474939       1.152382   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           params  \n",
       "rank_test_f1                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       \n",
       "1                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                {'clf': SVC(C=1289890.261253308, gamma=2.381685551976161e-10, max_iter=1000,\n",
       "    random_state=0), 'preprocessor__num__imputer__fill_value': -1, 'preprocessor__num__imputer__strategy': 'constant', 'preprocessor__num__scaler': RobustScaler()}  \n",
       "2                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               {'clf': GaussianNB(), 'preprocessor__num__imputer__strategy': 'mean', 'preprocessor__num__scaler': 'passthrough'}  \n",
       "3                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   {'clf': MLPClassifier(activation='logistic', alpha=1e-09, hidden_layer_sizes=(39, 36),\n",
       "              learning_rate_init=0.00031622776601683794, max_iter=10000,\n",
       "              random_state=0), 'preprocessor__num__imputer__strategy': 'mean', 'preprocessor__num__scaler': StandardScaler()}  \n",
       "4                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      {'clf': KNeighborsClassifier(n_neighbors=115, weights='distance'), 'preprocessor__num__imputer__strategy': 'most_frequent', 'preprocessor__num__scaler': StandardScaler()}  \n",
       "5                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                             {'clf': LinearSVC(C=231.0129700083158, loss='hinge', max_iter=10000, random_state=0), 'preprocessor__num__imputer__fill_value': 0, 'preprocessor__num__imputer__strategy': 'constant', 'preprocessor__num__scaler': RobustScaler()}  \n",
       "6                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 {'clf': LogisticRegression(C=0.2848035868435802, max_iter=100000, penalty='l1',\n",
       "                   random_state=0, solver='liblinear'), 'preprocessor__num__imputer__strategy': 'most_frequent', 'preprocessor__num__scaler': StandardScaler()}  \n",
       "7                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                        {'clf': DecisionTreeClassifier(criterion='entropy', max_depth=5, max_features=0.8,\n",
       "                       min_samples_leaf=90, min_samples_split=55,\n",
       "                       random_state=0), 'preprocessor__num__imputer__strategy': 'mean', 'preprocessor__num__scaler': 'passthrough'}  \n",
       "8                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    {'clf': RandomForestClassifier(criterion='entropy', max_depth=8, max_features=0.825,\n",
       "                       min_samples_leaf=5, min_samples_split=5, n_estimators=50,\n",
       "                       random_state=0), 'preprocessor__num__imputer__strategy': 'median', 'preprocessor__num__scaler': MinMaxScaler()}  \n",
       "9             {'clf': StackingClassifier(estimators=[('svc',\n",
       "                                SVC(C=1289890.261253308,\n",
       "                                    gamma=2.381685551976161e-10, max_iter=1000,\n",
       "                                    random_state=0)),\n",
       "                               ('gnb', GaussianNB()),\n",
       "                               ('mlp',\n",
       "                                MLPClassifier(activation='logistic',\n",
       "                                              alpha=1e-09,\n",
       "                                              hidden_layer_sizes=(39, 36),\n",
       "                                              learning_rate_init=0.00031622776601683794,\n",
       "                                              max_iter=10000, random_state=0)),\n",
       "                               ('knn',\n",
       "                                KNeighborsClassifier(n_neighbors=115,\n",
       "                                                     weights=...\n",
       "                                DecisionTreeClassifier(criterion='entropy',\n",
       "                                                       max_depth=5,\n",
       "                                                       max_features=0.8,\n",
       "                                                       min_samples_leaf=90,\n",
       "                                                       min_samples_split=55,\n",
       "                                                       random_state=0)),\n",
       "                               ('rfc',\n",
       "                                RandomForestClassifier(criterion='entropy',\n",
       "                                                       max_depth=8,\n",
       "                                                       max_features=0.825,\n",
       "                                                       min_samples_leaf=5,\n",
       "                                                       min_samples_split=5,\n",
       "                                                       n_estimators=50,\n",
       "                                                       random_state=0)),\n",
       "                               ('etc',\n",
       "                                ExtraTreesClassifier(n_estimators=113,\n",
       "                                                     random_state=0))],\n",
       "                   final_estimator=SVC()), 'preprocessor__num__imputer__strategy': 'median', 'preprocessor__num__scaler': StandardScaler()}  \n",
       "10                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           {'clf': ExtraTreesClassifier(n_estimators=113, random_state=0), 'preprocessor__num__imputer__strategy': 'most_frequent', 'preprocessor__num__scaler': 'passthrough'}  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display(HTML(\"<h4>Dados brutos da validação cruzada</h4>\"))\n",
    "print(\n",
    "    f\"Cada combinação de parâmetros foi utilizada em {cv_num} rodadas de treino com {cv_train_size} exemplos e teste com {cv_test_size}.\"\n",
    ")\n",
    "\n",
    "rank_col = f\"rank_test_{scoring_name}\"\n",
    "cv_df = pd.DataFrame(cv.cv_results_).sort_values(rank_col).set_index(rank_col)\n",
    "\n",
    "cols = [\n",
    "    f\"mean_test_{scoring_name}\",\n",
    "    \"mean_test_accuracy\",\n",
    "    \"mean_test_precision\",\n",
    "    \"mean_test_recall\",\n",
    "    \"mean_fit_time\",\n",
    "    \"params\",\n",
    "]\n",
    "with pd.option_context(\"display.max_colwidth\", None):\n",
    "    display(cv_df[cols].head(20))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 169
    },
    "id": "89VtUbP7BK99",
    "outputId": "57b1fb62-3717-4bdc-d896-3c4966f3094d"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<h4>Maior média de f1</h4>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<ul>\n",
       "<li>Maior média de f1: 0.61563 (desvio padrão: 0.02955)</li>\n",
       "<li>Parâmetros que maximizaram a média de f1:<br/><code>{'clf': SVC(C=1289890.261253308, gamma=2.381685551976161e-10, max_iter=1000,\n",
       "    random_state=0), 'preprocessor__num__imputer__fill_value': -1, 'preprocessor__num__imputer__strategy': 'constant', 'preprocessor__num__scaler': RobustScaler()}</code></li>\n",
       "<li>Modelo com a maior média de f1:</li>\n",
       "</ul>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = cv.best_estimator_\n",
    "\n",
    "display(HTML(f\"<h4>Maior média de {scoring_name}</h4>\"))\n",
    "display(\n",
    "    HTML(\n",
    "        f\"\"\"<ul>\n",
    "<li>Maior média de {scoring_name}: {cv.best_score_:.5f} (desvio padrão: {cv.cv_results_[\"std_test_\" + scoring_name][cv.best_index_]:.5f})</li>\n",
    "<li>Parâmetros que maximizaram a média de {scoring_name}:<br/><code>{cv.best_params_}</code></li>\n",
    "<li>Modelo com a maior média de {scoring_name}:</li>\n",
    "</ul>\"\"\"\n",
    "    )\n",
    ")\n",
    "# display(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "iZHgK-ctBK99",
    "outputId": "17f49273-5dd1-42ce-e597-068ab3c97d6b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1-score do modelo no conjunto de testes: 0.54185\n"
     ]
    }
   ],
   "source": [
    "y_pred = model.predict(X_test)\n",
    "score = f1_score(y_test, y_pred, pos_label=\"Yes\")\n",
    "print(f\"F1-score do modelo no conjunto de testes: {score:.5f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 334
    },
    "id": "J5fsuQclBK99",
    "outputId": "7b47fdb1-6b83-4b9c-9b62-568ad5b0e6ef"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cmd = plot_confusion_matrix(model, X_test, y_test, display_labels=[\"Não\", \"Sim\"])\n",
    "cmd.ax_.set(xlabel=\"Previsão\", ylabel=\"Realidade\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Re-treinamento com todos os dados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(4548, 21)\n",
      "(4548,)\n"
     ]
    }
   ],
   "source": [
    "X = df_all[all_cols]\n",
    "y = df_all[target]\n",
    "print(X.shape)\n",
    "print(y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>div.sk-top-container {color: black;background-color: white;}div.sk-toggleable {background-color: white;}label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.2em 0.3em;box-sizing: border-box;text-align: center;}div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}div.sk-estimator {font-family: monospace;background-color: #f0f8ff;margin: 0.25em 0.25em;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;}div.sk-estimator:hover {background-color: #d4ebff;}div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 2em;bottom: 0;left: 50%;}div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;}div.sk-item {z-index: 1;}div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;}div.sk-parallel-item {display: flex;flex-direction: column;position: relative;background-color: white;}div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}div.sk-parallel-item:only-child::after {width: 0;}div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0.2em;box-sizing: border-box;padding-bottom: 0.1em;background-color: white;position: relative;}div.sk-label label {font-family: monospace;font-weight: bold;background-color: white;display: inline-block;line-height: 1.2em;}div.sk-label-container {position: relative;z-index: 2;text-align: center;}div.sk-container {display: inline-block;position: relative;}</style><div class=\"sk-top-container\"><div class=\"sk-container\"><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"f02f386a-df51-4bb0-a048-ff4c02c6fe81\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"f02f386a-df51-4bb0-a048-ff4c02c6fe81\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[('preprocessor',\n",
       "                 ColumnTransformer(transformers=[('num',\n",
       "                                                  Pipeline(steps=[('to_numeric',\n",
       "                                                                   FunctionTransformer(func=<function num_mapper at 0x7f2b975e2dc0>)),\n",
       "                                                                  ('imputer',\n",
       "                                                                   SimpleImputer(fill_value=-1,\n",
       "                                                                                 strategy='constant')),\n",
       "                                                                  ('scaler',\n",
       "                                                                   RobustScaler())]),\n",
       "                                                  ['MONTHLYCHARGES', 'TENURE',\n",
       "                                                   'TOTALCHARGES']),\n",
       "                                                 ('cat',\n",
       "                                                  Pipeline(steps=[('imputer',\n",
       "                                                                   SimpleImputer(stra...\n",
       "                                                                   OneHotEncoder(handle_unknown='ignore'))]),\n",
       "                                                  ['GENDER', 'SENIORCITIZEN',\n",
       "                                                   'PARTNER', 'DEPENDENTS',\n",
       "                                                   'PHONESERVICE',\n",
       "                                                   'MULTIPLELINES',\n",
       "                                                   'INTERNETSERVICE',\n",
       "                                                   'ONLINESECURITY',\n",
       "                                                   'ONLINEBACKUP',\n",
       "                                                   'DEVICEPROTECTION',\n",
       "                                                   'TECHSUPPORT', 'STREAMINGTV',\n",
       "                                                   'STREAMINGMOVIES',\n",
       "                                                   'CONTRACT',\n",
       "                                                   'PAPERLESSBILLING',\n",
       "                                                   'PAYMENTMETHOD'])])),\n",
       "                ('clf',\n",
       "                 SVC(C=1289890.261253308, gamma=2.381685551976161e-10,\n",
       "                     max_iter=1000, random_state=0))])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"4e1b607e-c15f-4265-baf3-7330f376c9fe\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"4e1b607e-c15f-4265-baf3-7330f376c9fe\">preprocessor: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(transformers=[('num',\n",
       "                                 Pipeline(steps=[('to_numeric',\n",
       "                                                  FunctionTransformer(func=<function num_mapper at 0x7f2b975e2dc0>)),\n",
       "                                                 ('imputer',\n",
       "                                                  SimpleImputer(fill_value=-1,\n",
       "                                                                strategy='constant')),\n",
       "                                                 ('scaler', RobustScaler())]),\n",
       "                                 ['MONTHLYCHARGES', 'TENURE', 'TOTALCHARGES']),\n",
       "                                ('cat',\n",
       "                                 Pipeline(steps=[('imputer',\n",
       "                                                  SimpleImputer(strategy='most_frequent')),\n",
       "                                                 ('onehot',\n",
       "                                                  OneHotEncoder(handle_unknown='ignore'))]),\n",
       "                                 ['GENDER', 'SENIORCITIZEN', 'PARTNER',\n",
       "                                  'DEPENDENTS', 'PHONESERVICE', 'MULTIPLELINES',\n",
       "                                  'INTERNETSERVICE', 'ONLINESECURITY',\n",
       "                                  'ONLINEBACKUP', 'DEVICEPROTECTION',\n",
       "                                  'TECHSUPPORT', 'STREAMINGTV',\n",
       "                                  'STREAMINGMOVIES', 'CONTRACT',\n",
       "                                  'PAPERLESSBILLING', 'PAYMENTMETHOD'])])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"8e230249-95fe-4158-a22a-c9f8d995ac43\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"8e230249-95fe-4158-a22a-c9f8d995ac43\">num</label><div class=\"sk-toggleable__content\"><pre>['MONTHLYCHARGES', 'TENURE', 'TOTALCHARGES']</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"d8672515-dd50-4010-970f-b2cbc74d5b96\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"d8672515-dd50-4010-970f-b2cbc74d5b96\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=<function num_mapper at 0x7f2b975e2dc0>)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"0d177b02-5dec-4a79-9414-75ad06dd55b5\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"0d177b02-5dec-4a79-9414-75ad06dd55b5\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(fill_value=-1, strategy='constant')</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"96a0c707-d672-46ff-adec-729ae632c020\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"96a0c707-d672-46ff-adec-729ae632c020\">RobustScaler</label><div class=\"sk-toggleable__content\"><pre>RobustScaler()</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"1cec00e2-e8c1-4928-8b55-db1ab12c4b3b\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"1cec00e2-e8c1-4928-8b55-db1ab12c4b3b\">cat</label><div class=\"sk-toggleable__content\"><pre>['GENDER', 'SENIORCITIZEN', 'PARTNER', 'DEPENDENTS', 'PHONESERVICE', 'MULTIPLELINES', 'INTERNETSERVICE', 'ONLINESECURITY', 'ONLINEBACKUP', 'DEVICEPROTECTION', 'TECHSUPPORT', 'STREAMINGTV', 'STREAMINGMOVIES', 'CONTRACT', 'PAPERLESSBILLING', 'PAYMENTMETHOD']</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"bd6bf637-2b0a-43ae-b362-a2c29394c4ed\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"bd6bf637-2b0a-43ae-b362-a2c29394c4ed\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy='most_frequent')</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"401abe88-bca5-49f8-a9f8-4d84c8bab6c0\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"401abe88-bca5-49f8-a9f8-4d84c8bab6c0\">OneHotEncoder</label><div class=\"sk-toggleable__content\"><pre>OneHotEncoder(handle_unknown='ignore')</pre></div></div></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"1baf8550-b3c3-49f0-a78a-481093d907e5\" type=\"checkbox\" ><label class=\"sk-toggleable__label\" for=\"1baf8550-b3c3-49f0-a78a-481093d907e5\">SVC</label><div class=\"sk-toggleable__content\"><pre>SVC(C=1289890.261253308, gamma=2.381685551976161e-10, max_iter=1000,\n",
       "    random_state=0)</pre></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "Pipeline(steps=[('preprocessor',\n",
       "                 ColumnTransformer(transformers=[('num',\n",
       "                                                  Pipeline(steps=[('to_numeric',\n",
       "                                                                   FunctionTransformer(func=<function num_mapper at 0x7f2b975e2dc0>)),\n",
       "                                                                  ('imputer',\n",
       "                                                                   SimpleImputer(fill_value=-1,\n",
       "                                                                                 strategy='constant')),\n",
       "                                                                  ('scaler',\n",
       "                                                                   RobustScaler())]),\n",
       "                                                  ['MONTHLYCHARGES', 'TENURE',\n",
       "                                                   'TOTALCHARGES']),\n",
       "                                                 ('cat',\n",
       "                                                  Pipeline(steps=[('imputer',\n",
       "                                                                   SimpleImputer(stra...\n",
       "                                                                   OneHotEncoder(handle_unknown='ignore'))]),\n",
       "                                                  ['GENDER', 'SENIORCITIZEN',\n",
       "                                                   'PARTNER', 'DEPENDENTS',\n",
       "                                                   'PHONESERVICE',\n",
       "                                                   'MULTIPLELINES',\n",
       "                                                   'INTERNETSERVICE',\n",
       "                                                   'ONLINESECURITY',\n",
       "                                                   'ONLINEBACKUP',\n",
       "                                                   'DEVICEPROTECTION',\n",
       "                                                   'TECHSUPPORT', 'STREAMINGTV',\n",
       "                                                   'STREAMINGMOVIES',\n",
       "                                                   'CONTRACT',\n",
       "                                                   'PAPERLESSBILLING',\n",
       "                                                   'PAYMENTMETHOD'])])),\n",
       "                ('clf',\n",
       "                 SVC(C=1289890.261253308, gamma=2.381685551976161e-10,\n",
       "                     max_iter=1000, random_state=0))])"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X, y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "DaaULG8RBK9-"
   },
   "source": [
    "## Preenchimento do arquivo de respostas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 141
    },
    "id": "MUALvWJZBK9-",
    "outputId": "7b23cff8-8437-4d23-821a-ce27577c4221"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>GENDER</th>\n",
       "      <th>SENIORCITIZEN</th>\n",
       "      <th>...</th>\n",
       "      <th>MONTHLYCHARGES</th>\n",
       "      <th>TOTALCHARGES</th>\n",
       "      <th>CHURN</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1294</th>\n",
       "      <td>3100</td>\n",
       "      <td>Female</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>123.64</td>\n",
       "      <td>972.34</td>\n",
       "      <td>Yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>406</th>\n",
       "      <td>4362</td>\n",
       "      <td>Female</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>7.00</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Yes</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        ID  GENDER  SENIORCITIZEN  ... MONTHLYCHARGES TOTALCHARGES  CHURN\n",
       "1294  3100  Female              0  ...         123.64       972.34    Yes\n",
       "406   4362  Female              1  ...           7.00          7.0    Yes\n",
       "\n",
       "[2 rows x 21 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "answers[\"CHURN\"] = model.predict(answers)\n",
    "answers.to_csv(\"ANSWERS.csv\", index=False)\n",
    "\n",
    "with pd.option_context(\"display.max_columns\", 6):\n",
    "    display(answers.sample(2, random_state=seed))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 112
    },
    "id": "IH2dejIeBK9-",
    "outputId": "1f800bed-69f4-4f4c-b08b-db93b81bd2d2"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Proporção</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Yes</th>\n",
       "      <td>0.57855</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>No</th>\n",
       "      <td>0.42145</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Proporção\n",
       "Yes    0.57855\n",
       "No     0.42145"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "answers[target].value_counts(normalize=True).rename(\"Proporção\").to_frame()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "YTspWZrWBK9-",
    "outputId": "f0829aea-321d-4ae7-c711-f2c688a031c7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "código.zip\n"
     ]
    }
   ],
   "source": [
    "!rm -rf código.zip\n",
    "!zip -q código.zip notebook.ipynb\n",
    "!ls código.zip"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "notebook.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}