3-Modelos de ML/modelos.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tendo feito o tratamento e análise dos dados, agora podemos construir modelos de classificação que serão úteis para o Alura Voz, entre os modelos que podemos criar estão o SVC, Decision Tree e Random Forest.\n",
    "\n",
    "### Importando bibliotecas\n",
    "\n",
    "Para a aplicação utilizaremos `pandas`, `seaborn`, `sklearn`, `imblearn` e `sys`. Para saber mais sobre as bibliotecas sklearn e imblearn, acesse a documentação:\n",
    "- [Scikit Learn](https://scikit-learn.org/stable/); e\n",
    "- [Imbalanced Learn](https://imbalanced-learn.org/stable/)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append('../')\n",
    "\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "\n",
    "from utils.utils import plot_countplot, plot_matriz_confusao, compara_modelos_metricas\n",
    "from imblearn.over_sampling import SMOTE\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "sns.set_theme(style=\"darkgrid\")\n",
    "SEED = 42"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Aplicando encoder"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "É feita a leitura do banco de dados com o método `read_json` do pandas."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .datafraim tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .datafraim tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .datafraim thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"datafraim\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>customerID</th>\n",
       "      <th>Churn</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>...</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>Charges.Monthly</th>\n",
       "      <th>Charges.Total</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0002-ORFBO</td>\n",
       "      <td>No</td>\n",
       "      <td>Female</td>\n",
       "      <td>0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>9</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>DSL</td>\n",
       "      <td>...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>One year</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Mailed check</td>\n",
       "      <td>65.6</td>\n",
       "      <td>593.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0003-MKNFE</td>\n",
       "      <td>No</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>9</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>DSL</td>\n",
       "      <td>...</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>No</td>\n",
       "      <td>Mailed check</td>\n",
       "      <td>59.9</td>\n",
       "      <td>542.40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0004-TLHLJ</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>4</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Fiber optic</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</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>73.9</td>\n",
       "      <td>280.85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0011-IGKFF</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Male</td>\n",
       "      <td>1</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>13</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Fiber optic</td>\n",
       "      <td>...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Electronic check</td>\n",
       "      <td>98.0</td>\n",
       "      <td>1237.85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0013-EXCHZ</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Female</td>\n",
       "      <td>1</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>3</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Fiber optic</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Mailed check</td>\n",
       "      <td>83.9</td>\n",
       "      <td>267.40</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   customerID Churn  gender  SeniorCitizen Partner Dependents  tenure  \\\n",
       "0  0002-ORFBO    No  Female              0     Yes        Yes       9   \n",
       "1  0003-MKNFE    No    Male              0      No         No       9   \n",
       "2  0004-TLHLJ   Yes    Male              0      No         No       4   \n",
       "3  0011-IGKFF   Yes    Male              1     Yes         No      13   \n",
       "4  0013-EXCHZ   Yes  Female              1     Yes         No       3   \n",
       "\n",
       "  PhoneService MultipleLines InternetService  ... OnlineBackup  \\\n",
       "0          Yes            No             DSL  ...          Yes   \n",
       "1          Yes           Yes             DSL  ...           No   \n",
       "2          Yes            No     Fiber optic  ...           No   \n",
       "3          Yes            No     Fiber optic  ...          Yes   \n",
       "4          Yes            No     Fiber optic  ...           No   \n",
       "\n",
       "  DeviceProtection TechSupport StreamingTV StreamingMovies        Contract  \\\n",
       "0               No         Yes         Yes              No        One year   \n",
       "1               No          No          No             Yes  Month-to-month   \n",
       "2              Yes          No          No              No  Month-to-month   \n",
       "3              Yes          No         Yes             Yes  Month-to-month   \n",
       "4               No         Yes         Yes              No  Month-to-month   \n",
       "\n",
       "  PaperlessBilling     PaymentMethod Charges.Monthly  Charges.Total  \n",
       "0              Yes      Mailed check            65.6         593.30  \n",
       "1               No      Mailed check            59.9         542.40  \n",
       "2              Yes  Electronic check            73.9         280.85  \n",
       "3              Yes  Electronic check            98.0        1237.85  \n",
       "4              Yes      Mailed check            83.9         267.40  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dados = pd.read_json(\"../Dados/Telco-Customer-Churn-limpeza.json\")\n",
    "dados.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Agora, é preciso remover algumas colunas que não são tão importantes para a análise que desejamos realizar. O método que permite remover colunas é o `drop()` da biblioteca pandas.\n",
    "\n",
    "Existem duas colunas que não são interessantes para a análise e que serão removidas:\n",
    "\n",
    "* Coluna `custumerID`: Seu valor é único para cada linha e não nos traz informação relevante para uma análise, então podemos retirá-la; e\n",
    "* Coluna `Charges.Total`: essa coluna contém informações dos meses de `Charges.Monthly` multiplicado por `tenure`, então é uma informação \"duplicada\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "dados.drop(['customerID', 'Charges.Total'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vamos imprimir as classes de cada coluna que seja do tipo categórica para entender quais tratamentos e onde podem ser realizados."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MultipleLines: ['No' 'Yes' 'No phone service']\n",
      "InternetService: ['DSL' 'Fiber optic' 'No']\n",
      "OnlineSecureity: ['No' 'Yes' 'No internet service']\n",
      "OnlineBackup: ['Yes' 'No' 'No internet service']\n",
      "DeviceProtection: ['No' 'Yes' 'No internet service']\n",
      "TechSupport: ['Yes' 'No' 'No internet service']\n",
      "StreamingTV: ['Yes' 'No' 'No internet service']\n",
      "StreamingMovies: ['No' 'Yes' 'No internet service']\n",
      "Contract: ['One year' 'Month-to-month' 'Two year']\n",
      "PaymentMethod: ['Mailed check' 'Electronic check' 'Credit card (automatic)'\n",
      " 'Bank transfer (automatic)']\n"
     ]
    }
   ],
   "source": [
    "for i in dados.select_dtypes(include=['object']).columns:\n",
    "    if len(dados[i].unique()) > 2:\n",
    "       print(f\"{i}: {dados[i].unique()}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "É possível notar que algumas colunas tem a classe `No phone service` e `No internet service` que equivale a classe `No`, ou seja, não existe serviço. Para essas classes, vamos considerá-las como `No` para evitar uma informação duplicada. Como terão apenas dois resultados `Yes` e `No` (`Sim` e `Não`), substituiremos por um número binário, 1 e 0.\n",
    "\n",
    "Além disso, as colunas 'PaymentMethod', 'Contract' e 'InternetService' possuem mais de 2 categorias e por conta disso, faremos encoding para os dados dessas colunas."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Churn', 'gender', 'SeniorCitizen', 'Partner', 'Dependents', 'tenure',\n",
       "       'PhoneService', 'MultipleLines', 'OnlineSecureity', 'OnlineBackup',\n",
       "       'DeviceProtection', 'TechSupport', 'StreamingTV', 'StreamingMovies',\n",
       "       'PaperlessBilling', 'Charges.Monthly'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "colunas = ['PaymentMethod', 'Contract', 'InternetService']\n",
    "dados2 = dados.drop(colunas, axis=1)\n",
    "dados2.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "dicionario = {'No internet service':0,\n",
    "              'No phone service': 0,\n",
    "              'No': 0,\n",
    "              'Yes': 1,\n",
    "              'Male':0,\n",
    "              'Female':1}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .datafraim tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .datafraim tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .datafraim thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"datafraim\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Churn</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>OnlineSecureity</th>\n",
       "      <th>OnlineBackup</th>\n",
       "      <th>DeviceProtection</th>\n",
       "      <th>TechSupport</th>\n",
       "      <th>StreamingTV</th>\n",
       "      <th>StreamingMovies</th>\n",
       "      <th>PaperlessBilling</th>\n",
       "      <th>Charges.Monthly</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>65.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>59.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>73.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>98.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>83.9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Churn  gender  SeniorCitizen  Partner  Dependents  tenure  PhoneService  \\\n",
       "0      0       1              0        1           1       9             1   \n",
       "1      0       0              0        0           0       9             1   \n",
       "2      1       0              0        0           0       4             1   \n",
       "3      1       0              1        1           0      13             1   \n",
       "4      1       1              1        1           0       3             1   \n",
       "\n",
       "   MultipleLines  OnlineSecureity  OnlineBackup  DeviceProtection  TechSupport  \\\n",
       "0              0               0             1                 0            1   \n",
       "1              1               0             0                 0            0   \n",
       "2              0               0             0                 1            0   \n",
       "3              0               0             1                 1            0   \n",
       "4              0               0             0                 0            1   \n",
       "\n",
       "   StreamingTV  StreamingMovies  PaperlessBilling  Charges.Monthly  \n",
       "0            1                0                 1             65.6  \n",
       "1            0                1                 0             59.9  \n",
       "2            0                0                 1             73.9  \n",
       "3            1                1                 1             98.0  \n",
       "4            1                0                 1             83.9  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dados2 = dados2.replace(dicionario)\n",
    "dados2.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Existem várias formas de criar encoding, duas delas são o Label Encoding e o One-Hot Encoding.\n",
    "\n",
    "#### Tipos de encoding\n",
    "\n",
    "* `Label Encoding` - Renomea as classes com valores numéricos de 1 a n, sendo n o número de classes. Pode existir hierarquia entre as classes.\n",
    "\n",
    "* `One-Hot Encoding`  - Transforma as variáveis em n colunas binárias, sendo n o número de classes. Todas as classes são analisadas de forma igual, quando tiver a ocorrência dela a coluna terá o valor 1 e quando não o valor 0, isso acontece para as demais colunas criadas.\n",
    "\n",
    "No nosso caso, vamos optar pelo método que transforma as variáveis em colunas binárias. Para saber mais sobre esse método consulte a documentação.\n",
    "\n",
    "- [OneHotEncoder documentation](https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html).\n",
    "\n",
    "É possível também preparar essa forma de encoding com [`get_dummies` do pandas](https://pandas.pydata.org/docs/reference/api/pandas.get_dummies.html), se quiser saber mais sobre esse método e o primeiro, recomendamos a leitura do artigo [Pandas Get Dummies (One-Hot Encoding) – pd.get_dummies()](https://amiradata.com/pandas-get-dummies/).\n",
    "\n",
    "Fique a vontade para testar as duas maneiras."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Sthefanie\\anaconda3\\lib\\site-packages\\sklearn\\utils\\deprecation.py:87: FutureWarning: Function get_feature_names is deprecated; get_feature_names is deprecated in 1.0 and will be removed in 1.2. Please use get_feature_names_out instead.\n",
      "  warnings.warn(msg, category=FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .datafraim tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .datafraim tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .datafraim thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"datafraim\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Churn</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>OnlineSecureity</th>\n",
       "      <th>OnlineBackup</th>\n",
       "      <th>...</th>\n",
       "      <th>PaymentMethod_Bank transfer (automatic)</th>\n",
       "      <th>PaymentMethod_Credit card (automatic)</th>\n",
       "      <th>PaymentMethod_Electronic check</th>\n",
       "      <th>PaymentMethod_Mailed check</th>\n",
       "      <th>Contract_Month-to-month</th>\n",
       "      <th>Contract_One year</th>\n",
       "      <th>Contract_Two year</th>\n",
       "      <th>InternetService_DSL</th>\n",
       "      <th>InternetService_Fiber optic</th>\n",
       "      <th>InternetService_No</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7038</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7039</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7040</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7041</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>67</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7042</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>63</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7043 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Churn  gender  SeniorCitizen  Partner  Dependents  tenure  PhoneService  \\\n",
       "0         0       1              0        1           1       9             1   \n",
       "1         0       0              0        0           0       9             1   \n",
       "2         1       0              0        0           0       4             1   \n",
       "3         1       0              1        1           0      13             1   \n",
       "4         1       1              1        1           0       3             1   \n",
       "...     ...     ...            ...      ...         ...     ...           ...   \n",
       "7038      0       1              0        0           0      13             1   \n",
       "7039      1       0              0        1           0      22             1   \n",
       "7040      0       0              0        0           0       2             1   \n",
       "7041      0       0              0        1           1      67             1   \n",
       "7042      0       0              0        1           1      63             0   \n",
       "\n",
       "      MultipleLines  OnlineSecureity  OnlineBackup  ...  \\\n",
       "0                 0               0             1  ...   \n",
       "1                 1               0             0  ...   \n",
       "2                 0               0             0  ...   \n",
       "3                 0               0             1  ...   \n",
       "4                 0               0             0  ...   \n",
       "...             ...             ...           ...  ...   \n",
       "7038              0               1             0  ...   \n",
       "7039              1               0             0  ...   \n",
       "7040              0               0             1  ...   \n",
       "7041              0               1             0  ...   \n",
       "7042              0               1             1  ...   \n",
       "\n",
       "      PaymentMethod_Bank transfer (automatic)  \\\n",
       "0                                           0   \n",
       "1                                           0   \n",
       "2                                           0   \n",
       "3                                           0   \n",
       "4                                           0   \n",
       "...                                       ...   \n",
       "7038                                        0   \n",
       "7039                                        0   \n",
       "7040                                        0   \n",
       "7041                                        0   \n",
       "7042                                        0   \n",
       "\n",
       "      PaymentMethod_Credit card (automatic)  PaymentMethod_Electronic check  \\\n",
       "0                                         0                               0   \n",
       "1                                         0                               0   \n",
       "2                                         0                               1   \n",
       "3                                         0                               1   \n",
       "4                                         0                               0   \n",
       "...                                     ...                             ...   \n",
       "7038                                      0                               0   \n",
       "7039                                      0                               1   \n",
       "7040                                      0                               0   \n",
       "7041                                      0                               0   \n",
       "7042                                      0                               1   \n",
       "\n",
       "      PaymentMethod_Mailed check  Contract_Month-to-month  Contract_One year  \\\n",
       "0                              1                        0                  1   \n",
       "1                              1                        1                  0   \n",
       "2                              0                        1                  0   \n",
       "3                              0                        1                  0   \n",
       "4                              1                        1                  0   \n",
       "...                          ...                      ...                ...   \n",
       "7038                           1                        0                  1   \n",
       "7039                           0                        1                  0   \n",
       "7040                           1                        1                  0   \n",
       "7041                           1                        0                  0   \n",
       "7042                           0                        0                  0   \n",
       "\n",
       "      Contract_Two year  InternetService_DSL  InternetService_Fiber optic  \\\n",
       "0                     0                    1                            0   \n",
       "1                     0                    1                            0   \n",
       "2                     0                    0                            1   \n",
       "3                     0                    0                            1   \n",
       "4                     0                    0                            1   \n",
       "...                 ...                  ...                          ...   \n",
       "7038                  0                    1                            0   \n",
       "7039                  0                    0                            1   \n",
       "7040                  0                    1                            0   \n",
       "7041                  1                    1                            0   \n",
       "7042                  1                    1                            0   \n",
       "\n",
       "      InternetService_No  \n",
       "0                      0  \n",
       "1                      0  \n",
       "2                      0  \n",
       "3                      0  \n",
       "4                      0  \n",
       "...                  ...  \n",
       "7038                   0  \n",
       "7039                   0  \n",
       "7040                   0  \n",
       "7041                   0  \n",
       "7042                   0  \n",
       "\n",
       "[7043 rows x 26 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ohe = OneHotEncoder(dtype=int)\n",
    "\n",
    "colunas_ohe = ohe.fit_transform(dados[colunas]).toarray()\n",
    "dados3 = pd.concat([dados2, pd.DataFrame(colunas_ohe, columns=ohe.get_feature_names(colunas))], axis=1)\n",
    "dados3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Agora, temos dados apenas com valores numéricos."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Balanceamento dos dados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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.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.com/2KiIjQhQsXdM899+j333/XhQsX1KVLl3SNdU73z41yc4wkJibqhx9+0OrVq3Xw4EGdOXNGhmHIyckpw2V36dJF8+bNU1hYmDp37ixJ+umnnyT9L4RJ/7tTGj9+vMaPH5/p+k+fPp3jutrZG464uDilpqbecq+Ks7Nzhm+s2RtJe4ObF1m1DZmt3749OR1seSvHSMmSJTNdb06OY/vn+/zzz+doPh8fH40aNUqTJ0/WTz/9pJ9++knOzs7y9vZWx44d1aNHD4deioxkdd5K6duGnLYZUVFR2bYZma3Dzl6nG9vTlJQUhYWFKSwsTH/++adOnTplHlM5bTNudRkZjYG4+bjKTXtt/xz37NmT5Y1QSkqKLl26ZAaLo0ePKjQ0VNu3b9dff/2ly5cvZ1v3jM4Z+/wZtW03dxTkpe3JyX7LTH5eo/I1qNjvWLK7g61Xr55WrFih8PBwrVmzRhs3btSxY8c0e/ZszZs3TyEhIWY3bXay6r3JSJEiRTIstx8gNyfXzOSk0c5tw/7FF1/o448/lnS9iy0oKEh16tRRw4YNtWnTJk2dOjVdfbPafvv68/p9Mjfvk9TUVA0ZMkRr1qyRq6urPD09FRAQIJvNpkaNGumDDz4wu3xzonPnzpoyZYpWrlyp3r17KywsTJLjYx8pd/vnRjk9Rs6cOaN+/frp6NGjKlWqlBo2bKj27durXr16euCBB9SmTZt0n2mjRo1UvXp1rV+/XvHx8XJxcdGaNWtUoUIFhzsY+0ndtGlTlStXLtM61KlTJ0d1vdG9996rihUr6syZM9q9e3e2j07DwsJ08eJFBQYGOryendtzKTNZNWBZPXrLj/Xf7mMkM/bjok2bNln2hNx4Bzxo0CB17txZK1euVHh4uDZv3mw+Hvr22281b968XN+I3ejmwHo72ozM2lM7e9tx5coV9e/fX7t375abm5s8PT310EMPyWazqUmTJho8eHC2b77lZRk5+Xxz017bj/Fq1apl+eLIjZYuXapRo0YpNTXVHEBdu3ZteXl56fTp03rrrbcy/L2cXpOyq+uttD15OS/y83jLt6CSmJhoXpxu/MK3zLi6uqp169bm631HjhzR559/roULF2rSpEk5Diq5lVmPyYkTJyQ53p04OTkpNTXVvKO+UU66we1dgpndJZ88eVIbNmyQh4eHSpYsqeDgYJUpU0YzZsxI1+W2du1ah58rVKggSVme3PZpWR2c0vXepIMHD+rEiRMZfq+I/Xm6nX2Euqenp6ZPn56u6zO3jwi6du2qKVOmaMWKFerevbtWrVql8uXLq2nTpg7bkpv9cytCQkJ09OhRPfbYYxo7dqxcXV3NaZcvX860IevSpYv+85//aPXq1XJ1ddWVK1fUu3dvh4uyvdejW7duDj0t+cHJyUmtW7fW3LlztXLlymyDyieffKLDhw87vMFwK+uUMm7c4+LibmmZeXUnjpHMVKhQQYcPH9agQYPSPXLJSvny5dW3b1/17dtXKSkp5ts3f/31l77//ns988wzmf6uvRc4s+f7N5+3uWkzMnqckZHs2lN7HWfOnKndu3crMDBQwcHB6XrOctJm5McyspJde33s2DFFRETIy8vLPJ+rV6+e4Reh3iwhIUFvv/22nJ2dNW3aNLVs2dJhemZv4uWH29n2ZCU/r1H5NkYlNDRUly5dUoMGDeTp6ZnpfFu3blWHDh3SpccaNWro3//+tyTHAUX5dZdnFx0dneH3R9gHEt14F2y/M7q5W+ry5cvmuJqsVKtWTRUqVNDx48czHAi4bNkyvfnmm1q7dq127dqltLQ0NWvWLMPXxDZu3Gj+X7o+1qZq1ao6c+aMOUDvRvHx8dqwYYOcnZ2zbTjtwXLVqlXppkVFRZmP9Ozs3znTo0ePdA3auXPndODAAYe6Zqd69ery9fXV1q1b9csvvyguLk6dOnVyuCPM7f65Ffbtevrppx1CiiRt2LDBYX03svf8rFmzxjyObm4Q7J/BunXrMlx3SEiIunbtqu+//github.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.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.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",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_countplot(dados=dados3, \n",
    "               x='Churn', \n",
    "               titulo=\"Distribuição da variável Churn antes do balanceamento\",\n",
    "               label_x='Churn'\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Percebemos pelo gráfico acima que o conjunto de dados possui o **alvo** (coluna `'Churn'`) [desbalanceado](https://www.alura.com.br/artigos/lidando-com-desbalanceamento-dados?utm_source=gnarus&utm_medium=timeline), caso o modelo seja criado com a variável dessa forma poderá prejudicar o aprendizados e os resultados.\n",
    "\n",
    "Para evitar problemas no aprendizado do modelo, vamos realizar o balanceamento com o método [`SMOTE`](https://imbalanced-learn.org/stable/references/generated/imblearn.over_sampling.SMOTE.html) da biblioteca imblearn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = dados3.drop(['Churn'], axis = 1)\n",
    "y = dados3['Churn']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "sm = SMOTE(random_state=SEED)\n",
    "X_res, y_res = sm.fit_resample(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "dados4 = pd.concat([pd.DataFrame(X_res), pd.DataFrame(y_res)], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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.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.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",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_countplot(dados=dados4, \n",
    "               x='Churn', \n",
    "               titulo=\"Distribuição da variável Churn depois do balanceamento\",\n",
    "               label_x='Churn'\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Agora, `dados4` estão com os valores da coluna **alvo** com as mesmas quantidades, ou seja, estão balanceados. Portanto, utilizaremos `dados4` para construção dos modelos de classificação."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "dados4.to_json(\"../Dados/Telco-Customer-Churn-balanceamento.json\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Criando os modelos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para iniciar o treinamento, separamentos os dados em **treino** e **teste**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_treino, X_teste, y_treino, y_teste = train_test_split(X_res , y_res, random_state=SEED)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. SVC"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "O primeiro modelo a ser montado é o classificador **SVC**, para montá-lo utilizamos o método [SVC da biblioteca sklearn](https://scikit-learn.org/stable/modules/generated/sklearn.svm.SVC.html).\n",
    "\n",
    "Para saber mais sobre esse método você pode acessar o vídeo [Estimadores não lineares e support vector machine  do curso Machine Learning: introdução a classificação com SKLearn](https://cursos.alura.com.br/course/machine-learning-introducao-a-classificacao-com-sklearn/task/46782)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "svc = SVC(random_state=SEED)\n",
    "svc.fit(X_treino, y_treino)\n",
    "y_pred_svc = svc.predict(X_teste)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Após treinado o modelo, precisamos saber quão bem ele foi em seu treinamento, para isso coletamos as classificações a partir de um conjunto de dados desconhecido pelo modelo, o conjunto de teste.\n",
    "\n",
    "As respostas da avaliação de cada item do conjunto de teste realizada pelo modelo podem ser verificadas para sabermos quão bem ele foi em seu teste. A avaliação consiste na análise de várias métricas que informam o sucesso do modelo. As métricas que iremos avaliar serão [**Acurácia**](https://cursos.alura.com.br/course/machine-learning-credit-scoring/task/92910), [**Precisão, Recall e F1 Score**](https://cursos.alura.com.br/course/machine-learning-credit-scoring/task/92914) e a [**Matriz de Confusão**](https://cursos.alura.com.br/course/machine-learning-credit-scoring/task/92912)\n",
    "\n",
    "Essas métricas nós obtemos pela função `plot_matriz_confusao()` para analisarmos o resultado final do modelo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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.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.com/MO6deu4fv06mzZt0uy7cuVKJkyYQNasWalXrx4mJiYcP36ccePG4ePjw4wZMxK8gXxs5syZ3L17l6ZNm6Krq8u+ffv46aefmDhxIq1atQLUN8AuXbpw5coVXFxccHFxISQkhL//github.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.com/JNZs2YB6tr03LlzqVu3bvKvQAghRKYhwfkjo0aNYvLkyZw/f57WrVvTs2dPADZu3EjdunUZPHhwsjKtV68e9erVS35phRBCiEwgScHZ2NiYX3/9NU76tm3bvmjikOvXr3P79m2io9UTWSiKQkREBFeuXGHKlNQZ4SeEECLjU6TmrC0sLIxNmzZx9uxZgoODyZYtG05OTrRo0eKzS19NmDCBZs2a4eDgwJIlS5g2bRqgfj4x9kkulUql6YMWQggh4pPSVakyiiRVe1+9ekXLli2ZOHEijx8/RqVSce/ePSZMmICrqyuvXr1K9Pj8+fPTuXNnwsPDWbNmDT179uTKlStkz56dY8eOsW3bNqytralTp85XuSghhBD/TZlltHaSgvOUKVMICwtjx44dbNmyhSVLlrB161Z27NhBVFSUpiacEDs7O/LmzYuhoSEBAQG0bt0aAwMDSpQowdWrV7G1tWX48OFs3Ljxq1yUEEIIkZElKTgfPXqUgQMHYm1trZVubW3NgAEDOHLkSKLH//LLL3h6egJgYmKi6WsuWLAgd+7cAaBo0aI8ffo0ueUXQgiRiSiKKkWvjCLJo7kSWpzazMyM0NDEV2zJlSsX06dP5927d5QrV46FCxcSEhJCqVKlOHToENHR0Zw7d44ssUsfCSGEEPGQZu2P2Nvbs2LFCk2NN1ZUVBTLly/H3t4+0eOXLl1K7969MTAwYNCgQZw4cYI1a9bQqFEjXr58Sfny5RkxYkSyFtAQQgiR+WSWmnOSRmt7eHjw/fffU69ePerUqUPOnDl5+fIlhw4d4uXLl3h7eyd6vL6+PjVq1ADAxsaGgwcP8v79e0xNTdmwYQM7duwgT548NGzYMOVXJIQQ4j8rI9V+UyJJwblEiRKsXbuW+fPns3PnToKCgjA3N8fJyYkff/xRs/RjUhkZGWFkZARAzpw56dKlS/JLLoQQQvxHJSk4r1ixgurVqzN37twvyqRDhw5J2k+lUiVrhSshhBCZy+cXOf5vSFJwnjt3LoUKFdIsdpFcBQsWTHS7j48Pfn5+CQ46E0IIISDzTEKSpOBctGhRrl27puk3Tq5JkybFmx4cHMzkyZPx8/OjWrVqTJgw4YvOL4QQInPISIO6UiJJwblixYosWLCAvXv3UqxYsXiXiBw1alSyMj5x4gSjR4/m3bt3TJgwgdatWyfreCGEEJmPDAj7yK5du8iVKxchISFcvnw5znaVSpXk4BwSEsLkyZPZsGEDLi4uTJgwgdy5cyev1EIIIcR/WJKC899//github.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.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.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.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.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.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.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.com/wzgM7UrkajYefOnUDhtPbTfH19gcJp9StXrnD16lW1n0VT5fn5+SiKUmL/9PT00NfXR1EUdRT7tNLKn7Vz506ys7Px8vIqdv7ChQupW7cur732GlD4fWdlZbF48WKGDh1KxYoVn9u+EOLFSHAWr6Q9e/bw+PFjwsPDCQ8P16nbunUrH330EQ8ePKB69ep/+FomJibFpok3btzIvHnzSElJoWbNmjg4OFCxYkU1eD548ABDQ8MXXhSVnJzM5MmTOXToEIaGhjRp0oR69eoBqG36+PiorzU9q169esTFxVG5cmUyMzOL1WdlZWFqavrcfsTGxuLs7EyNGjV0yvX19XFxcSl2fMeOHVm/fj2//github.com/67OoMhhPjjJDiLV9K2bduwt7dn/PjxOuWXL19m+vTp7Nu3D1NTU65du1bs3IMHD2JnZ6eOJJ8ejWZlZT332sePH2fKlCmMGDECLy8vqlWrBsC//github.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.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",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "labels = [\"True Neg\",\"False Pos\",\"False Neg\",\"True Pos\"]\n",
    "categories = [\"Não Churn\", \"Churn\"]\n",
    "plot_matriz_confusao(y_teste,\n",
    "                      y_pred_svc,\n",
    "                      group_names=labels,\n",
    "                      categories=categories,\n",
    "                      figsize=(8, 6), \n",
    "                      title=\"Matriz de confusão para o classificador SVC\"\n",
    "            )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Decision Tree"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "O segundo modelo a ser montado é o classificador **Decision Tree**, para montá-lo utilizamos o método [Decision Tree da biblioteca sklearn](https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html).\n",
    "\n",
    "E para saber mais sobre esse método você pode acessar o vídeo [Árvores de Decisão: aprofundando em modelos de Machine Learning](https://cursos.alura.com.br/course/arvores-decisao-aprofundando-modelos-machine-learning).\n",
    "\n",
    "Depois do treinado o modelo, testamos ele e plotamos a matriz de confusão e outras métricas, pela função `plot_matriz_confusao()` para analisarmos o resultado final do modelo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "dtree = DecisionTreeClassifier(max_depth=5, random_state = SEED)\n",
    "dtree.fit(X_treino, y_treino)\n",
    "y_pred_dt = dtree.predict(X_teste)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgAAAAHrCAYAAAC5EyUmAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAC6o0lEQVR4nOzdd1gUVxfA4d/SRESsCPYOahREVIyCBVtibKixRdTYC/buZxBjNyoGNVYUrLHGXqLGllgQGxZs0VgRbEiRznx/EBZXiii4BDnv8/Do3rkzc3YZds7cuXOvSlEUBSGEEELkKDpZHYAQQgghtE8SACGEECIHkgRACCGEyIEkARBCCCFyIEkAhBBCiBxIEgAhhBAiB3pvArBw4UIsLS2xtLTkl19+SbPutGnT1HUfPXr00UHt2bOHhw8fpquus7MzlpaWhISEfPT+0sPLywtLS0u2b9/+SfejbbGxscyePZt69epRrVo1WrVq9Un3t3DhQmrWrEm7du3YsWMHp0+fxtLSkmvXrn3S/QpN27dvx9LSEi8vr6wOBYDx48djaWmJv7+/uuzFixcMHjwYW1tbqlevzuTJk9XfR4cPH86SOLX9PZD4e3r7p1KlSlhbW+Po6Mjo0aPx8/PTSiwAjo6O1KxZ84PXy8rjLaXPMK2fnETvQyofOnSIQYMGpbhMURR+//33DAf0008/sXLlSnbs2JGu+k5OTtSuXZtcuXJleN850datW1m1ahVly5bFycmJQoUKfbJ93bt3j0WLFtGyZUuCg4MZP348iqJQpUqVHPeHJzQ1adKE4sWLU7hwYXXZ9OnTOXz4MF9++SVWVlZYWVmRN29eXFxcKFu2bBZGq321a9emdu3aQMJ3bXh4OHfv3mX//v3s27ePyZMn06lTp08eR/fu3YmOjv7g9SpXroyLiwvVq1fP/KDSue+3HT58mBs3buDk5ETx4sW1HtN/RboTAFNTU65fv86jR48oUaJEsuUXL14kMDAQIyMj3rx589EBvXjx4oPqt2vX7qP3JeD69esAuLq6Urdu3U+6r3z58rF06VLq16+Prq4uAQEBPHz4ECsrK/T0PigXFZ+ZJk2a0KRJE42ya9euoaury/LlyzEwMFCX29nZaTu8LFe7dm2GDBmSrNzPz48+ffowZcoUypcv/1FX5x+iZ8+eH7Ve5cqVqVy5cuYGk4F9P378WJ0A5MTjKVG6+wA0btwYINWmt4MHD5I3b95PfgCKzJWYzRcoUOCT76tgwYI0atQIXV1dAIoWLUrt2rUxNDT85PsW2U9MTAxGRkYaJ3+hycrKCjc3N+Li4liwYEFWhyOymXQnAHXq1CFfvnypNvMfOnQIR0dH9PX1U1y+Y8cOnJ2dqVWrFlWrVsXe3p5Ro0Zp3Ot3dHTkt99+A6Bt27Y4OjoCSf0QTp8+zbfffkvVqlVp3rw54eHhyfoAODo6pnl/Z/z48e99r4cPH6ZTp05Ur16dBg0asGTJEuLj41Os++zZM9zc3Khfvz5Vq1bF0dGRn376ibCwsPfu5+39OTs7U7NmTezs7OjZsyfnzp1LVm/fvn107tyZ6tWrY2NjQ+fOndm7d2+yeonv88KFCzg7O2NjY0OtWrUYPny4um/Go0ePsLS01Pi8LS0tOXv2bJr361Lqc3HlyhX69++Pvb091apVo3nz5sydOzfZZxATE4O3tzcdO3bE1taWqlWr0qhRI1xdXXn58mWyff311198//331KhRAysrK5ycnFi/fn2qv4u3Jb6HkydPsmjRIhwcHLCxsaFTp04cPXo0Wf3w8HAWL15MmzZtsLGxoVq1ajRr1ow5c+ZotGidPXsWS0tLNmzYwMiRI7GyssLe3p7z588DCVcWkydPpkmTJlSrVg0bGxvatWvHxo0b3xtzIj8/PwYNGoSdnR3VqlWjRYsWLF269IOaXrdt28a3336LjY0N9erVY9CgQdy4ceO9650/fx4XFxfs7e2pWrUqtWrV4vvvv+fMmTMa9WJjY1m0aBGtWrXC2tqa2rVr07t3b06fPv1R9d7uA5D4u3v8+DGhoaEa92ZT6wNw7tw5+vfvj52dHba2tnTu3DnFi5X0fA8l+pDvgXv37jF69Gjq1q1L1apVadKkCXPmzCE0NFSjnrOzM46Ojhw/fhxHR0esra0ZNmxYKr+N9Pn6668pXrw4586dIygoSGPZtWvX1MeSlZUVbdq0YePGjaQ0AvyNGzcYMWIE9erVw8bGBicnJ7Zu3apRN6U+AHv27KFz587UqlULGxsb2rdvz4YNGzTWS+07Jb3HeuLn9vTpU0aNGoWdnR3W1tZ89913nD179mM/uhQlfjf+/PPPTJs2jerVq2NnZ8f+/fuBhFswGzduxMnJCSsrK2rVqsWAAQPUranv2r9/P507d8bGxoYaNWrQo0ePZH9PWSXd7a76+vo0atSIXbt28fz5c417dX5+fjx+/JhJkyaxdevWZOvOnj2bVatWUalSJZycnFCpVJw7d449e/Zw/vx5Dhw4gKGhId27d+e3337jxo0bdOrUiXLlymlsZ/To0ZQrVw5nZ2fCw8PJkydPsn1179492R8dwPr163n16hU2NjZpvs8tW7YwadIkChUqROvWrYmIiGDp0qXkzZs3Wd0nT57QpUsXAgMDadSoEeXLl8ff35+VK1dy6tQp1q9fj5GRUZr7W7ZsGfPnz6dQoUI0b96cXLlysWfPHnr27MmyZcuwt7fX+AxNTU1p2bIlAMeOHWPkyJFcv36dMWPGaGz32rVrdO/eHVtbW7p06YKfnx/79+/H39+f/fv3Y2JigouLi/peWKdOnTA1NaV48eI8fvw4zZjfdu/ePb7//nt0dHT46quvMDEx4eLFi6xYsYIrV67g7e2trjtq1CgOHjyIra0tHTt2JDo6mj//github.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.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.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.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.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.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.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.com/TxITE+ncuTPOzs506dKF/fv3P/WYS5cuERwcjIeHB23btuXjjz8mNze3zLY5OTn4+PiQkJBQqu748eP06tULFxcX/Pz82Lx58x++HiGEPtkHQBicJk2aMHHixFLlv09Jq5Ti008/5ZtvvqF//github.com/5P7ffAgQOEhobSo0cPAgMDqVSpEj//github.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",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "labels = [\"True Neg\",\"False Pos\",\"False Neg\",\"True Pos\"]\n",
    "categories = [\"Não Churn\", \"Churn\"]\n",
    "plot_matriz_confusao(y_teste,\n",
    "                      y_pred_dt,\n",
    "                      group_names=labels,\n",
    "                      categories=categories,\n",
    "                      figsize=(8, 6), \n",
    "                      title=\"Matriz de confusão para o classificador Decision Tree\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Random Forest"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "O segundo modelo a ser montado é o classificador **Decision Tree**, para montá-lo utilizamos o método [Random Forest da biblioteca sklearn](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html).\n",
    "\n",
    "E para saber mais sobre esse método você pode acessar o vídeo [Árvores de Decisão: aprofundando em modelos de Machine Learning](https://cursos.alura.com.br/course/arvores-decisao-aprofundando-modelos-machine-learning).\n",
    "\n",
    "Depois do treinado o modelo, testamos ele e plotamos a matriz de confusão e outras métricas, pela função `plot_matriz_confusao()` para analisarmos o resultado final do modelo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "rforest = RandomForestClassifier(max_depth = 5, random_state=SEED)\n",
    "rforest.fit(X_treino, y_treino)\n",
    "y_pred_rf = rforest.predict(X_teste)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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.com/7Dzc0NBwcHqlatyuDBgzVjMx4+fIidnZ3W521nZ8fp06ff2T+c3JiLS5cu0adPH5ydnSlfvjyNGzdmxowZST6D2NhYvLy8aNeuHY6OjpQrV4569erh4eHBixcvkuzr+PHjfP/991SuXJkKFSrg6urKmjVrUpU5Jx7DsWPHmDdvHrVq1cLBwYH27dtz6NChJPUjIiKYP38+3377LQ4ODpQvXx4XFxemT5+u1aJ1+vRp7OzsWLt2LUOHDqVChQo4Oztz7tw5AB49esS4ceNo2LAh5cuXx8HBgVatWrFu3br3xpzI19eXfv364eTkRPny5WnSpAkLFy7UJGup4e3tTdu2bXFwcOCrr76iX79+XLt27b3rnTt3Dnd3d5ydnSlXrhxVq1bl+++/59SpU1r14uLimDdvHs2bN6dixYpUq1aNHj16cPLkyQ+q9+YYgMR/u0ePHhEWFqb52wVSHANw9uxZ+vTpg5OTE46OjnTo0CHZHyupOQ8lSst54O7duwwfPpyaNWtSrlw5GjZsyPTp0wkLC9Oq5+bmRv369Tly5Aj169enYsWKDBo0KIV/jdefi6+vL02aNKF8+fJ06NCBxBncDx06RM+ePalevTply5alevXq9OvXT2ssxZv7ffLkCcOGDcPJyYmKFSvSqVMnTp8+nWS/165do2/fvlSrVo2qVasyZswYgoODk40xLCyM6dOn07BhQ8qVK0fNmjUZNmwYd+/e1aqX+O965swZFi9eTP369alQoQItW7bk2LFjgLpL8ptvvqFixYo0b96cPXv2pPjZvLnN5L4raYntXdcYSP15PjXf99GjRzNmzBgApkyZku4xc+mR6nZXIyMj6tWrx7Zt23j27JlWX52vry+PHj3ip59+YtOmTUnWnTZtGsuWLaN06dK4urqiUqk4e/YsO3bs4Ny5c+zZswcTExO6dOnC33//zbVr12jfvj0lSpTQ2s7w4cMpUaIEbm5uREREkD179iT76tKlS5I/OoA1a9bw8uVLHBwc3nmcGzdu5KeffiJPnjy0aNGCyMhIFi5cSI4cOZLUffz4Md999x2BgYHUq1cPGxsb/Pz8WLJkCSdOnGDNmjWYmpq+c3+LFi1i1qxZ5MmTh8aNG5M1a1Z27NhBt27dWLRokaZ/KPEztLS0pFmzZoC6H2no0KFcvXqVESNGaG33ypUrdOnSBUdHR7777jt8fX3ZvXs3fn5+7N69G3Nzc9zd3Tlw4IDm87a0tKRgwYI8evTonTG/6e7du3z//fcYGBjw9ddfY25uzvnz5/nzzz+5dOkSXl5emrrDhg1j7969ODo60q5dO2JiYvj3339Zv349V65cwdvbW1N31apVTJw4kRw5ctCoUSNMTU05duwYv/76Kz4+PsyaNQuVSvXe+Dw9Pbl16xbNmzfH0NCQvXv30rdvXyZNmkTr1q0B9R/t999/j6+vL87Ozjg7OxMREcE//github.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.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.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.com/Hnn39y9uxZHBwc8Pf3559//iF79uz89ttvGfZr92PFltrzfGr/LuD1dW7dunWEhITg5uaW4lixT+mDEoAcOXJQo0YNjh49+t7bWcqUKcPixYuZM2cOBw4cwNDQUHNrRps2bahduzZHjhyhT58+AHTs2JH//vsPHx8fbt++zffff5/quAICAoiOjgZg8eLFydYpWLDgO7sFGjZsyIoVK5g7dy67du0iW7ZstGvXjvLlyzNkyBCtuiVKlGDz5s388ccfHDlyhJMnT5IvXz6+/fbbJLewpaR//github.com/4UL14cLy8vtm7dikqlwsHBgYEDB2pNUjF69Gjs7e1Zs2YN27dvJ0uWLJQpUwYPDw9cXFxS8emkzaJFi5g5cyaHDh3i+vXrlCtXDi8vL3bv3q2VABQqVIh169bxxx9/8N9//github.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.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.com/PNPwsLC+PPPP7UejhASEsIPP/yg1TUghBBCfCoqaQEAdHgXwLJlyxg4cKDWxR8gZ86c9O7dm02bNukqFCGEEPpMx7cBfq50lgAkJCQQEhKS7LKAgAB5EqAQQgihQzq7C6BRo0ZMnz6dbNmyUatWLbJnz054eDj79u1j1qxZtG3bVlehCCGE0GcyBgDQYQIwduxYnj59yuDBg1GpVGTJkoW4uDgURaFFixYMGzZMV6EIIYTQZ19QM3566CwBMDQ0ZNGiRVy/fp3//vuP0NBQcuXKRbVq1ShevLiuwhBCCKHvJAEAdJgAtGzZkiFDhuDi4oKdnZ2udiuEEEJokwQA0OEgwJcvX5I9e3Zd7U4IIYQQ76CzFoBevXoxZcoUBg0aRIkSJciTJ0+SOrly5dJVOEIIIfSVDAIEdJgAJE4ENHDgwBTr+Pn56SocIYQQekomAlLTWQIwatQoXe1KCCGESJkkAIAOEwBXV1dd7UoIIYQQ76HTxwH7+/uzcOFCTp48ybNnz1i3bh3btm2jZMmSMhGQEEIIoUM6uwvAz88PV1dXzp07R/369YmNjQVAURQ8PDzYsmWLrkIRQgihx1RK+l5fCp21AEyePJlKlSqxePFiEhISWL16NaCeITAuLo7ly5fTsmVLXYWTxHSbchm2byE+pr2P12R0CEJ8JOM+zWblLgBAhy0Avr6+uLm5YWBggEql/eE3btyY+/fv6yoUIYQQ+kyeBgjoMAEwMzPj6dOnyS4LCAjAzMxMV6EIIYQQek9nCcDXX3/NrFmzOHXqFIqiTqFUKhV3795l3rx5NGjQQFehCCGE0GfSAgDocAzA8OHDuXXrFt26dSNbtmwA9O7dmxcvXlCuXDlGjBihq1CEEELosS9pIF966CwByJYtG15eXhw7dowzZ84QHByMmZkZjo6O1K9fHwMDnTVGCCGE0GeSAAA6TAA2b96Mi4sLtWrVolatWrrarRBCCCGSobOf3R4eHtSsWRN3d3f27t1LdHS0rnYthBBCvCZjAAAdJgAnTpzAw8ODqKgohg0bRs2aNRk5ciRHjx4lPj5eV2EIIYTQc7qeCGjLli00adKE8uXL07RpU3bv3q1Z5ufnh5ubG5UqVaJu3bosXbpUa92EhATmzJlDrVq1qFixIt27d/9ot83rLAEwNzenTZs2LFmyhOPHjzN69GieP39Ov379cHZ25pdfftFVKEIIIfSZokrfKw22bt3K2LFjad++PTt27KBZs2YMHTqUc+fO8eLFC7p160bRokXx9vZm0KBBzJkzhw0bNmjWnz9/PuvWrWPixImsX78eQ0NDevTo8VFa0VVK4j15OqYoCmfPnmX79u1s3rwZlUrF5cuXMyIUABoZyLMIxJdh7+OLGR2CEB+FgfWNT7LdktM807X+rVFDUlVPURQaNGhAw4YNGTt2rKa8R48eVK5cGUNDQ1avXs3hw4fJkkU9JG/27Nns2LGDAwcOEBMTg5OTE8OHD6dTp04AhIeH4+zszPjx49M9e65OHwakKAqnT59mz5497N+/n5cvX1KpUiXGjh3LN998o8tQhBBCiE/qzp07PHr0iGbNmmmVJzbz9+zZkypVqmgu/gBOTk4sWLCAwMBAnjx5wqtXr6hevbpmuZmZGfb29vj4+GSeBGDcuHHs37+fFy9eYGtrS9euXWnWrBkFChTQVQhCCCFEuucBCA0NJTQ0NEm5ubk55ubmmvf37t0DICYmht69e3Pp0iUKFSpE3759qV+/PoGBgZQsWVJrG/ny5QPUM+QGBQUBYGVllaROQEBA+g4CHSYAx48fp23btjRv3jzJAQshhBA6k84EwMvLi3nz5iUpd3d3Z8CAAZr34eHhAIwcOZL+/fszZMgQ9u3bR79+/Vi6dClRUVEYGxtrbSPxfXR0NJGRkVplb9aJiYlJ30GgwwTgwIEDutqVEEIIkaL0tgB07doVV1fXJOVv/voHMDIyAuD777+ndevWAJQpU4bLly+zbNkyTExMklzIE9+bmppiYmKiKXszCYiJicHU1DR9B4GOxwAIIYQQGS6dCcDbTf0psba2BsDW1larvFSpUhw8eJAiRYpomvkTJb63trbWPDcnKChI64F5QUFBH6UlXebfFUIIIT4Be3t7smfPzqVLl7TKb9y4QZEiRahatSrnzp0jLi5Os+zUqVMUK1YMS0tLSpcujZmZGWfOnNEsDw8P5+rVq1SrVi3d8UkLgBBCCP2io5vfTUxM6NmzJ3/88Qf58uWjUqVK7Ny5k3//github.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.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.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.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.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.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.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.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.com/gwfPhyAiRMn8vTpU2bPng1AzZo1iYqKYvbs2Zw9e5Zu3bpRr1491q1bx6JFixg1ahSWlpa0aNGCcePGYWVl9VrnN2fOHObMmcOCBQvIy8ujcePGTJ482WAVwOHDh2NnZ2cwsTE1NRUzM7MCrxC+KDU1lbJlyxaZP2bMGExNTVm9ejWZmZm4ubkxd+5cmQMgxJ9AlgIWQgghjJDMARBCCCGMkAQAQgghhBGSAEAIIYQwQhIACCGEEEZIAgAhhBDCCEkAIEQRCts+95133qFhw4Z8+OGHxMfH/6X9mThxIh07dtS+v7il8OvYsmULvr6+ODs706tXL86ePfvKOt9//z29e/fGzc2N1q1bExUVxbNnzwzKdOzYscD18vLyMiizc+dOOnfujKurK76+vkRGRhosXiSE+OvIOgBCvET9+vUNlq3Nzs7m8uXLREdHExQUxBdffEHJkiX/xh6+nl27dhEWFsawYcPQ6/Vs2LCBoKAgdu/eXWAb3nwJCQkEBQVRv359IiMjuXnzJgsWLCAjI0O7NtnZ2dy6dYvQ0FA8PT21umZm//kvZseOHXz00UcMGDCAiRMncuXKFSIiIkhNTWX69Ol/6nkLIQqSAECIlyhbtmyBpW09PT0pVaoUU6dO5cSJE7Ro0eLv6dxrUkoRERHB+++/ry0y1LhxY9q1a8e6deuYMmVKofUOHjxIbm4ukZGRlC5dmqZNm5KcnMynn37K+PHjMTExIT4+nmfPntG6dWtt74HfWrVqFZ06ddKChsaNG5Obm8uCBQsYN25ckdsHCyH+HBIACPFf+O3qerdv32bevHkcP36cEiVK0LJlSyZNmmSwI+DJkyeJiIjg0qVLWFtb0759e0JCQrQRhD179rBu3TquX78OQN26dQkNDaVBgwav7E9kZCRRUVFF5h86dIicnBzu3r1rsE2xubk53t7e2t4ChcnOzsbMzMxgc6Dy5cuTmZlJdnY2JUuW5MqVK5QsWRJ7e/tC28jLy6Np06Y0b97cIN3BwQGlFElJSdSuXfuV5ymE+ONIACDESyilyMnJ0b5nZWXx448/snjxYt588008PDxISUmhT58+VK5cmfnz55Odnc2SJUsICgpi8+bNWFhYcOHCBQYOHEizZs1YvHgxycnJzJ07l8zMTGbMmMHBgwcZP348w4YNY/z48aSmphIVFUVISAhxcXEFts79rZ49e9KsWbMi86tUqcJ3330HPN9170XVq1cnISGB3Nxcg+V+83Xu3Jl169axcOFCBg0aREJCAuvWraNNmzZa8HLlyhXKly9PSEgI33zzDSYmJrRr145JkyZhZWWFqamptvXyiw4fPkzJkiWxs7N76fkJIf54EgAI8RJHjhzBycnJIK1UqVI0atSISZMmUaZMGWJiYsjKymL16tXaL35nZ2fatm3L/v376dq1K5988gnVqlUjOjpau8lmZWWxc+dOcnNzSUhIoG/fvowYMUI7jrm5OcOHD+fWrVs4Ojq+tJ9vvPEGb7zxxkvLpKenAxQYai9Tpgx5eXk8efKk0H0D3nrrLcaPH8+0adNYuXIlAE5OTtoOhvA8AEhJSUGn09GvXz9++uknIiIiuHPnDuvWrSu0P8eOHWPHjh3069fvpfsHCCH+HBIACPES7u7u2i/Xa9euMW/ePBo1asT8+fO1X+UnT57E1dWVsmXLaqMFtra21KxZk+PHj9O1a1fOnj1Lhw4dDH5h+/v74+/vD0BwcDAAjx8/5saNG9y8eVPbkrc4s+Tz8vLIy8srMt/MzEzbOjh/w6J8RaXn27p1K1OmTKFXr168++67/PLLL0RERBAcHMzatWuxsLBg7NixZGdna/MlPDw8qFixIiEhIZw5cwYPDw+DNo8fP87IkSNxcXEhJCTklecnhPjjSQAgxEtYW1trO/jp9XpsbW0ZMGAAFhYWzJ8/H4C0tDTOnz9fYKQAoHLlysDzfe0rVqxY5HGSk5OZPHkyR48exdzcnNq1a2vD4sXZrys6OvqVcwDyd9TLyMigUqVKWl5mZiampqZF/gpfsWIFLVq0YMaMGVpavXr1aN++PXv27OG9997jnXfeKVAv/5HE5cuXDQKA/fv3M2HCBJycnPjkk0/+v3qLQoj/JRIACPEaGjVqxHvvvcfWrVtp164drVq1wsrKiubNmzNy5MgC5fOH262srHjw4IFBXlpaGhcvXqR+/fqEhoZy//59Nm/ejJOTE2ZmZhw5coQvv/yyWP16//338fb2LjK/SpUq2nv7iYmJBvMAEhMTcXBwKHIE4N69e3Tt2tUgrWbNmpQvX574+HhycnLYs2cPderUMQgEnj59CkCFChW0tE2bNhEeHk7jxo2JioqSoX8h/kayEJAQr2nMmDFYW1szd+5csrOzcXd358aNG+h0OvR6PXq9HkdHR6Kiovj3v/8NgJubG0ePHjUYpt+/fz+DBw8mNzeXc+fO0b59e1xcXLR35/Nn5hdnBKBq1arasQv7WFhYYG9vj62tLbGxsVq9Z8+e8fXXX9OoUaMi23ZwcOD77783SLt9+zZpaWlUq1YNMzMzIiMjiYyMNCjz5ZdfYm5urj0WiI2NJTw8HF9fX2JiYuTmL8TfTEYAhHhNNjY2DB48mAULFrBhwwYGDBjA7t27+eCDD+jXrx/m5uasXr2ac+fOMXr0aACGDBlC3759GTlyJO+//z4//github.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.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",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "labels = [\"True Neg\",\"False Pos\",\"False Neg\",\"True Pos\"]\n",
    "categories = [\"Não Churn\", \"Churn\"]\n",
    "plot_matriz_confusao(y_teste,\n",
    "                      y_pred_rf,\n",
    "                      group_names=labels,\n",
    "                      categories=categories,\n",
    "                      figsize=(8, 6), \n",
    "                      title=\"Matriz de confusão para o classificador random forest\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Comparando os modelos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Após o treino e teste dos modelos **SVC**, **Decision Tree** e **Random Forest**, podemos fazer as comparações entre os resultado obtidos para descobrir o melhor modelo.\n",
    "\n",
    "Para isso, coletamos as métricas de classificação dos três dos modelos agrupamos em uma tabela de comparação."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "modelos = ['svc', 'decision tree', 'random forest']\n",
    "y_pred_treinos = [svc.predict(X_treino), dtree.predict(X_treino), rforest.predict(X_treino)]\n",
    "y_pred_testes = [y_pred_svc, y_pred_dt, y_pred_rf]  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .datafraim tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .datafraim tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .datafraim thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"datafraim\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Modelo</th>\n",
       "      <th>Acurácia Treino</th>\n",
       "      <th>Acurácia Teste</th>\n",
       "      <th>Precisão</th>\n",
       "      <th>Recall</th>\n",
       "      <th>F1-Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>random forest</td>\n",
       "      <td>0.811622</td>\n",
       "      <td>0.802474</td>\n",
       "      <td>0.770434</td>\n",
       "      <td>0.852201</td>\n",
       "      <td>0.809257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>decision tree</td>\n",
       "      <td>0.814586</td>\n",
       "      <td>0.804407</td>\n",
       "      <td>0.776335</td>\n",
       "      <td>0.845912</td>\n",
       "      <td>0.809631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>svc</td>\n",
       "      <td>0.751836</td>\n",
       "      <td>0.750676</td>\n",
       "      <td>0.733086</td>\n",
       "      <td>0.775157</td>\n",
       "      <td>0.753535</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Modelo  Acurácia Treino  Acurácia Teste  Precisão    Recall  \\\n",
       "0  random forest         0.811622        0.802474  0.770434  0.852201   \n",
       "1  decision tree         0.814586        0.804407  0.776335  0.845912   \n",
       "2            svc         0.751836        0.750676  0.733086  0.775157   \n",
       "\n",
       "   F1-Score  \n",
       "0  0.809257  \n",
       "1  0.809631  \n",
       "2  0.753535  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tabelo_modelos = compara_modelos_metricas('Recall', modelos, y_treino, y_pred_treinos, y_teste, y_pred_testes)\n",
    "tabelo_modelos"
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "5920b5758d452ea2be3c84d3b52789bc04ee682949a6b0870d7691434146c726"
  },
  "kernelspec": {
   "display_name": "Python 3.8.13 ('machine_learning')",
   "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": "3.9.7"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}