Titanic.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Titanic: Machine Learning from Disaster\n",
    "\n",
    "**Para ver o artigo relacionado a este *notebook*, [CLIQUE AQUI](http://sigmoidal.ai/data-science-titanic-python-1/)\n",
    "\n",
    "---\n",
    "\n",
    "O desafio do Titanic é uma das competições mais famosas, e é promovida pelo site [kaggle.com](http://kaggle.com). Tamanha popularidade desta competição fez com que ela fosse considerada como sendo um ***Hello, World!*** para o universo fantástico do Data Science (Ciência de Dados).\n",
    "\n",
    "Com certeza, a maioria de vocês conhece a história por trás do naufrágio deste que seria o *navio mais moderno do mundo* para a época. Apesar da tecnologia disponível, após colidir com um iceberg em 15 de abril de 1912, **seu naufrágio matou 1502 pessoas de um total de 2224 passageiros.**\n",
    "\n",
    "![Data Science, Python e Titanic](img/titanic.jpeg)\n",
    "\n",
    "Como principal fator contribuinte, logo se descobriu que não havia botes salva-vidas suficientes para toda a tripulação e passageiros. \n",
    "\n",
    "Obviamente, aqueles que sobreviveram à tragédia contaram com muita sorte, mas será que alguns grupos de passageiros não tiveram \"mais sorte\" que outros grupos? Será que crianças e mulheres realmente tiveram mais chances de sobreviver? Será que o Jack teve menos chances de escapar do seu destino trágico que a Rose, só pelo fato de ele ter embarcado na 3ª Classe?\n",
    "\n",
    "Com os dados **reais** disponibilizados, agora você pode fazer uma análise exploratória, testar suas hipóteses e até mesmo construir um modelo preditivo (que tal descobrir quais seriam suas chances de sobrevivência na 1ª Classe se você tivesse viajado com sua esposa e filho?!)\n",
    "\n",
    "> **Resumidamente, o objetivo deste desafio é utilizar os dados disponíveis para medir a chance de sobrevivência dos passageiros do Titanic.** \n",
    "\n",
    "## Sobre este notebook\n",
    "\n",
    "Este notebook é parte de uma série de artigos do meu blog, [sigmoidal.ai](http://sigmoidal.ai):\n",
    "\n",
    "* [PARTE I](http://sigmoidal.ai/data-science-titanic-python-1/)\n",
    "* [PARTE II](http://sigmoidal.ai/data-science-titanic-python-2/)\n",
    "\n",
    "A ideia é introduzir o assunto de **DATA SCIENCE** para iniciantes na área, por isso tentei não me aprofundar em nenhum conceito ou análise, de maneira que seja possível reproduzir (e entender) todos passos usados.\n",
    "\n",
    "\n",
    "## Checklist\n",
    "\n",
    "1. Definição do Problema\n",
    "2. Obtenção dos Dados\n",
    "3. Exploração dos Dados\n",
    "4. Preparação dos Dados\n",
    "5. Construção do Modelo\n",
    "6. Avaliação"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Definição do Problema\n",
    "\n",
    "A construção do Titanic levou cerca de 2 anos e custou 7,5 milhões de dólares (valores da época). Com 269 metros de comprimento, 28 metros de largura e 53 metros de altura, operava com uma tripulação de 892 pessoas e poderia levar até 2435 passageiros (espalhados pelas três classes disponíveis).\n",
    "\n",
    "Pensado para ser o mais seguro e luxuoso navio da época, foi lançado ao mar em 1911, ganhando fama de ser \"inafundável\".\n",
    "\n",
    "A descrição completa da competição, assim como o conjunto de dados, está disponível na página do Kaggle [Titanic: Machine Learning from Disaster](https://www.kaggle.com/c/titanic).\n",
    "\n",
    "O naufrágio do Titanic teve como fatores contribuintes causas naturais (como o clima) e causas humanas (negligência e excesso de confiança). Independente das causas, o fato é que seu naufrágio matou 1502 pessoas de um total de 2224 passageiros.\n",
    "\n",
    "Embora aqueles que escaparam com vida tiveram sua boa dose de sorte, alguns grupos de pessoas eram mais propensos a escaparem da morte do que outros. Por exemplo, mulheres, crianças e passageiros da 1ª Classe. Assim, nota-se que existe algum padrão que pode ser extraído dos dados brutos.\n",
    "\n",
    "### Objetivo\n",
    "\n",
    "Espera-se que você consiga fazer uma **análise sobre quais variáveis tiveram maior influência na probabilidade de sobrevivência** (ou seja, que tipo de pessoa teve mais chance de escapar com vida).\n",
    "\n",
    "Após analisar os dados, espera-se também que você seja capaz de **construir um modelo que dê a previsão de sobrevivência** para um passageiro qualquer que seja fornecido como *input*.\n",
    "\n",
    "\n",
    "### Métrica de Desempenho\n",
    "\n",
    "O *score* é calculado em relação ao número de previsões corretas que seu modelo fez. Ou seja, é considerada apenas a **acurácia** do modelo.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Obtenção dos Dados\n",
    "\n",
    "[Os dados disponibilizados pelo Kaggle](https://www.kaggle.com/c/titanic/data) foram divididos em dois grupos:\n",
    "\n",
    "* Dataset de treino (```train.csv```)\n",
    "    * Deve ser usado para construir o modelo de Machine Learning. Neste conjunto de dados, é informado se o passageiro sobreviveu ou não. \n",
    "    * Composto por diversas *features* como gênero do passageiro e classe do embarque.\n",
    "* Dataset de Teste (```test.csv```)\n",
    "    * Deve ser usado como dados que nunca forma vistos pelo modelo. Neste conjunto de dados, não é informado se o passageiro sobreviveu ou não.\n",
    "    \n",
    "Também é disponibilizado um modelo (gabarito) de como as previsões devem ser enviadas para a plataforma do Kaggle (```gender_submission```).\n",
    "\n",
    "Para conseguir baixar os arquivos, é precisar criar um login no Kaggle antes. Para facilitar, os arquivos estão junto com este notebook (pasta ```./data/...```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ".\n",
      "├── Titanic-Copy1.ipynb\n",
      "├── Titanic.ipynb\n",
      "├── data\n",
      "│   ├── gender_submission.csv\n",
      "│   ├── test.csv\n",
      "│   └── train.csv\n",
      "├── img\n",
      "│   ├── kaggle_lr.png\n",
      "│   ├── kaggle_tree.png\n",
      "│   └── titanic.jpeg\n",
      "├── submission.csv\n",
      "├── submission_lr.csv\n",
      "└── submission_tree.csv\n",
      "\n",
      "2 directories, 11 files\n"
     ]
    }
   ],
   "source": [
    "# listar os arquivos no diretório\n",
    "!tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# importar os pacotes necessários\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# permitir visualizar todas as colunas\n",
    "pd.options.display.max_columns = None\n",
    "\n",
    "# renderizar os gráficos diretamento no notebook\n",
    "# %matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = pd.read_csv('data/train.csv')\n",
    "test = pd.read_csv('data/test.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q1.** Quantas variáveis (colunas) e quantas entradas (linhas) possui o arquivo?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Variáveis:\t12\n",
      "Entradas:\t891\n"
     ]
    }
   ],
   "source": [
    "# verificando as dimensões do DataFrame\n",
    "print(\"Variáveis:\\t{}\\nEntradas:\\t{}\".format(train.shape[1], train.shape[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Exploração dos Dados\n",
    "\n",
    "Em uma análise inicial, serão estudadas cada variável. É importante verificar seu nome, tipo, porcentagem de valores faltantes, presença de outliers e tipo de distribuição.\n",
    "\n",
    "Como se trata de um Aprendizado Supervisionado (*Supervised Learning*), a variável ```Survived``` é definida como sendo a nossa variável alvo (*target*)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dicionário de Dados\n",
    "\n",
    "* **PassengerId:** Número de identificação do passageiro\n",
    "* **Survived:** Informa se o passageiro sobreviveu ao desastre\n",
    "    * 0 = Não\n",
    "    * 1 = Sim\n",
    "* **Pclass:** Classe do bilhete\n",
    "    * 1 = 1ª Classe\n",
    "    * 2 = 2ª Classe\n",
    "    * 3 = 3ª Classe\n",
    "* **Name:** Nome do passageiro\n",
    "* **Sex:** Sexo do passageiro\n",
    "* **Age:** Idade do passageiro\n",
    "* **SibSp:** Quantidade de cônjuges e irmãos a bordo\n",
    "* **Parch:** Quantidade de pais e filhos a bordo\n",
    "* **Ticket:** Número da passagem\n",
    "* **Fare:** Preço da Passagem\n",
    "* **Cabin:** Número da cabine do passageiro\n",
    "* **Embarked:** Porto no qual o passageiro embarcou\n",
    "    * C = Cherbourg\n",
    "    * Q = Queenstown\n",
    "    * S = Southampton"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q1. Quais os tipos de cada variável?**\n",
    "\n",
    "* Variáveis Numéricas\n",
    "    * Discretas\n",
    "        * Pclass\n",
    "        * SibSp\n",
    "        * Parch\n",
    "    * Contínuas\n",
    "        * Age\n",
    "        * Fare\n",
    "    * Alvo\n",
    "        * Survived\n",
    "    * Chave Primária\n",
    "        * PassengerId\n",
    "* Variáveis Categóricas\n",
    "    * Name\n",
    "    * Sex\n",
    "    * Ticket\n",
    "    * Cabin\n",
    "    * Embarked"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PassengerId      int64\n",
       "Survived         int64\n",
       "Pclass           int64\n",
       "Name            object\n",
       "Sex             object\n",
       "Age            float64\n",
       "SibSp            int64\n",
       "Parch            int64\n",
       "Ticket          object\n",
       "Fare           float64\n",
       "Cabin           object\n",
       "Embarked        object\n",
       "dtype: object"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  \\\n",
       "0            1         0       3   \n",
       "1            2         1       1   \n",
       "2            3         1       3   \n",
       "3            4         1       1   \n",
       "4            5         0       3   \n",
       "\n",
       "                                                Name     Sex   Age  SibSp  \\\n",
       "0                            Braund, Mr. Owen Harris    male  22.0      1   \n",
       "1  Cumings, Mrs. John Bradley (Florence Briggs Th...  female  38.0      1   \n",
       "2                             Heikkinen, Miss. Laina  female  26.0      0   \n",
       "3       Futrelle, Mrs. Jacques Heath (Lily May Peel)  female  35.0      1   \n",
       "4                           Allen, Mr. William Henry    male  35.0      0   \n",
       "\n",
       "   Parch            Ticket     Fare Cabin Embarked  \n",
       "0      0         A/5 21171   7.2500   NaN        S  \n",
       "1      0          PC 17599  71.2833   C85        C  \n",
       "2      0  STON/O2. 3101282   7.9250   NaN        S  \n",
       "3      0            113803  53.1000  C123        S  \n",
       "4      0            373450   8.0500   NaN        S  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# identificar o tipo de cada variável\n",
    "display(train.dtypes)\n",
    "\n",
    "# ver as 5 primeiras entradas do conjunto de treino\n",
    "display(train.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q2. Qual a porcentagem de valores faltantes?**\n",
    "\n",
    "A informação sobre a **Cabin** é a que possui o maior número de informações faltantes, com mais de 77%. Após, a coluna **Age** não possui valor em quase 20% dos passageiros. Por fim, a coluna informando o Porto de embarque (**Embarked**) possui apenas 2% de informações faltantes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Cabin          0.771044\n",
       "Age            0.198653\n",
       "Embarked       0.002245\n",
       "Fare           0.000000\n",
       "Ticket         0.000000\n",
       "Parch          0.000000\n",
       "SibSp          0.000000\n",
       "Sex            0.000000\n",
       "Name           0.000000\n",
       "Pclass         0.000000\n",
       "Survived       0.000000\n",
       "PassengerId    0.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# ver a porcentagem valores faltantes\n",
    "(train.isnull().sum() / train.shape[0]).sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q3. Como é a distribuição estatística dos dados?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "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>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>714.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>446.000000</td>\n",
       "      <td>0.383838</td>\n",
       "      <td>2.308642</td>\n",
       "      <td>29.699118</td>\n",
       "      <td>0.523008</td>\n",
       "      <td>0.381594</td>\n",
       "      <td>32.204208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>257.353842</td>\n",
       "      <td>0.486592</td>\n",
       "      <td>0.836071</td>\n",
       "      <td>14.526497</td>\n",
       "      <td>1.102743</td>\n",
       "      <td>0.806057</td>\n",
       "      <td>49.693429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.420000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>223.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>20.125000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.910400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>446.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>14.454200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>668.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>31.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>891.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>80.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>512.329200</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       PassengerId    Survived      Pclass         Age       SibSp  \\\n",
       "count   891.000000  891.000000  891.000000  714.000000  891.000000   \n",
       "mean    446.000000    0.383838    2.308642   29.699118    0.523008   \n",
       "std     257.353842    0.486592    0.836071   14.526497    1.102743   \n",
       "min       1.000000    0.000000    1.000000    0.420000    0.000000   \n",
       "25%     223.500000    0.000000    2.000000   20.125000    0.000000   \n",
       "50%     446.000000    0.000000    3.000000   28.000000    0.000000   \n",
       "75%     668.500000    1.000000    3.000000   38.000000    1.000000   \n",
       "max     891.000000    1.000000    3.000000   80.000000    8.000000   \n",
       "\n",
       "            Parch        Fare  \n",
       "count  891.000000  891.000000  \n",
       "mean     0.381594   32.204208  \n",
       "std      0.806057   49.693429  \n",
       "min      0.000000    0.000000  \n",
       "25%      0.000000    7.910400  \n",
       "50%      0.000000   14.454200  \n",
       "75%      0.000000   31.000000  \n",
       "max      6.000000  512.329200  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ver histograma das variáveis numéricas\n",
    "train.hist(figsize=(10,8));"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q4. Quais grupos de pessoas tinham mais chances de sobrevivência??**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "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>Survived</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sex</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>female</th>\n",
       "      <td>0.742038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>male</th>\n",
       "      <td>0.188908</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Survived\n",
       "Sex             \n",
       "female  0.742038\n",
       "male    0.188908"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Analisar a probabilidade de sobrevivência pelo Sexo\n",
    "train[['Sex', 'Survived']].groupby(['Sex']).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotar os gráficos para Survived vs. Sex, Pclass e Embarked\n",
    "fig, (axis1, axis2, axis3) = plt.subplots(1,3, figsize=(12,4))\n",
    "\n",
    "sns.barplot(x='Sex', y='Survived', data=train, ax=axis1)\n",
    "sns.barplot(x='Pclass', y='Survived', data=train, ax=axis2)\n",
    "sns.barplot(x='Embarked', y='Survived', data=train, ax=axis3);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x12b315d30>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x216 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ver influência da idade na probabilidade de sobrevivência\n",
    "age_survived = sns.FacetGrid(train, col='Survived')\n",
    "age_survived.map(sns.distplot, 'Age')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 16 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotar uma scatter matrix\n",
    "columns=['Parch', 'SibSp', 'Age', 'Pclass']\n",
    "pd.plotting.scatter_matrix(train[columns], figsize=(15, 10));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "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": [
    "# plotar o heatmap para as variáveis numéricas\n",
    "sns.heatmap(train.corr(), cmap='coolwarm', fmt='.2f', linewidths=0.1,\n",
    "            vmax=1.0, square=True, linecolor='white', annot=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pouca gente sabe que o método ```df.describe()``` pode ser usado também para ver algo sobre a estatística descritiva em colunas que não são números. Para isso, é só passar o argumento ```include=['O']```. \n",
    "\n",
    "**ATENÇÃO, é a letra 'O', e não o número 0!**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "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>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>891</td>\n",
       "      <td>891</td>\n",
       "      <td>891</td>\n",
       "      <td>204</td>\n",
       "      <td>889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>891</td>\n",
       "      <td>2</td>\n",
       "      <td>681</td>\n",
       "      <td>147</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>Lefebre, Miss. Jeannie</td>\n",
       "      <td>male</td>\n",
       "      <td>CA. 2343</td>\n",
       "      <td>C23 C25 C27</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>1</td>\n",
       "      <td>577</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>644</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          Name   Sex    Ticket        Cabin Embarked\n",
       "count                      891   891       891          204      889\n",
       "unique                     891     2       681          147        3\n",
       "top     Lefebre, Miss. Jeannie  male  CA. 2343  C23 C25 C27        S\n",
       "freq                         1   577         7            4      644"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.describe(include=['O'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Olhando esses dados a gente observa que:\n",
    "* Os valores para ```Name``` são únicos, um para cada pessoa;\n",
    "* Não há *missing values* para ```Name``` e ```Sex```;\n",
    "* Existem apenas os sexos Masculino e Feminino, sendo que o com maior frequência é o Masculino (577/891 = 65%);\n",
    "* O Porto no qual embarcaram mais passageiros é o de S = Southampton."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4 Preparação dos Dados\n",
    "\n",
    "Até o presente momento, tudo o que fizemos foi importar os dados para estruturas *DataFrame*, formular hipóteses, iniciar uma análise exploratória dos dados e visualizar graficos e correlações entre aquelas variáveis que julgamos pertinente.\n",
    "\n",
    "Um projeto de data science não é amarrado e rigoroso, onde a gente segue um passo-a-passo linear, mas sim um processo iterativo, onde a gente vai e volta sempre que for necessário.\n",
    "\n",
    "Quando a gente pega o *notebook* de alguém, fica sempre a impressão que está bem estruturado e que foi \"direto ao ponto*. No entanto, antes de trazer uma versão bonitinha, eu vou e volto ao começo inúmeras vezes!\n",
    "\n",
    "Lembre-se disso: uma versão final é uma versão que foi reescrita várias vezes. Vá e volte ao começo sempre que precisar para adicionar informações novas que fizeram sentido :)\n",
    "\n",
    "### Juntando os *datasets* de treino e teste\n",
    "\n",
    "Uma dica para quando você vai preparar os dados para um modelo de *Machine Learning* é juntar os *datasets* de treino e teste em um único, e separar novamente no final.\n",
    "\n",
    "Muitas vezes a gente vai fazer um *feature engineering*, criar variáveis *dummy* ou codificar as variáveis. Daí, nosso modelo vai ser treinado em cima dessa arquitetura, e os dados de teste vão ter que seguir essa mesma estrutura.\n",
    "\n",
    "Por isso, é muito mais fácil fazer todas as etapas para um único *DataFrame* e dividir novamente entre treino e teste."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "df_merged.shape: (1309 x 11)\n"
     ]
    }
   ],
   "source": [
    "# salvar os índices dos datasets para recuperação posterior\n",
    "train_idx = train.shape[0]\n",
    "test_idx = test.shape[0]\n",
    "\n",
    "# salvar PassengerId para submissao ao Kaggle\n",
    "passengerId = test['PassengerId']\n",
    "\n",
    "# extrair coluna 'Survived' e excluir ela do dataset treino\n",
    "target = train.Survived.copy()\n",
    "train.drop(['Survived'], axis=1, inplace=True)\n",
    "\n",
    "# concatenar treino e teste em um único DataFrame\n",
    "df_merged = pd.concat(objs=[train, test], axis=0).reset_index(drop=True)\n",
    "\n",
    "print(\"df_merged.shape: ({} x {})\".format(df_merged.shape[0], df_merged.shape[1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Selecionar as *features*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Como qualquer conjunto de dados do mundo real, você vai se deparar sempre com dados que não servem para nada e outros que não tem peso ou significância nenhuma no seu modelo.\n",
    "\n",
    "Muitas vezes nosso julgamento pode ser equivocado, mas infelizmente é papel seu, como cientista de dados, escolher quais *features* serão usadas para o modelo de *Machine Learning*.\n",
    "\n",
    "No nosso caso, vamos desconsiderar as variáveis ```['PassengerId', 'Name', 'Ticket', 'Cabin']```, pois **aparentemente** não parecem relevantes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_merged.drop(['PassengerId', 'Name', 'Ticket', 'Cabin'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Assim, ficamos com as seguintes variáveis a serem tratadas e preparadas: ```['Pclass', 'Sex', 'Age', 'SibSp', 'Parch', 'Fare', 'Embarked']```."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Valores faltante"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vamos dar uma olhada nos valores que faltam em cada columna e tratar esses campos vazios.\n",
    "\n",
    "Normalmente, há duas abordagens mais utilizadas quando a gente encontra *missing values*:\n",
    "* Preencher esses valores arbitrariamente (média, mediana, valor mais frequente); e\n",
    "* Excluir a linha inteira.\n",
    "\n",
    "Cada caso é um caso e novamente você, cientista de dados, é quem vai tomar a decisão sobre qual passo seguir.\n",
    "\n",
    "Na maioria das vezes não é desejável jogar informação de uma linha inteira só por causa de um campo faltando. Sempre que possível é melhor você preencher o campo, e é isso que vamos fazer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pclass        0\n",
       "Sex           0\n",
       "Age         263\n",
       "SibSp         0\n",
       "Parch         0\n",
       "Fare          1\n",
       "Embarked      2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# completar ou apagar valores faltantes nos datasets de treino e teste\n",
    "df_merged.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para as variáveis idade e tarifa, vou colocar o valor da mediana, e para a variável do porto de embarque, vou colocar o valor com maior frequência."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# age\n",
    "age_median = df_merged['Age'].median()\n",
    "df_merged['Age'].fillna(age_median, inplace=True)\n",
    "\n",
    "# fare\n",
    "fare_median = df_merged['Fare'].median()\n",
    "df_merged['Fare'].fillna(fare_median, inplace=True)\n",
    "\n",
    "# embarked\n",
    "embarked_top = df_merged['Embarked'].value_counts()[0]\n",
    "df_merged['Embarked'].fillna(embarked_top, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Preparar as variáveis pro modelo\n",
    "\n",
    "O modelo matemático exige que trabalhemos com variáveis numéricas, ou seja, temos transformar os dados de entrada que estão em formato categoria para números.\n",
    "\n",
    "Como você vê abaixo, eu convertir os valores possíveis de ```Sex```para ```{'male': 0, 'female': 1}```. Já em relação à variável ```Embarked```, apliquei o conceito de variáveis dummies.\n",
    "\n",
    "As variáveis dummies (*dummy variables*) assumem aqui apenas valores 0 e 1, criando uma nova coluna para cada valor possível da variável categórica. Para ficar mais fácil entender, veja como fica o *DataFrame* após todos esses tratamentos:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "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>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Embarked_914</th>\n",
       "      <th>Embarked_C</th>\n",
       "      <th>Embarked_Q</th>\n",
       "      <th>Embarked_S</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>71.2833</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>3</td>\n",
       "      <td>1</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Pclass  Sex   Age  SibSp  Parch     Fare  Embarked_914  Embarked_C  \\\n",
       "0       3    0  22.0      1      0   7.2500             0           0   \n",
       "1       1    1  38.0      1      0  71.2833             0           1   \n",
       "2       3    1  26.0      0      0   7.9250             0           0   \n",
       "3       1    1  35.0      1      0  53.1000             0           0   \n",
       "4       3    0  35.0      0      0   8.0500             0           0   \n",
       "\n",
       "   Embarked_Q  Embarked_S  \n",
       "0           0           1  \n",
       "1           0           0  \n",
       "2           0           1  \n",
       "3           0           1  \n",
       "4           0           1  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# converter 'Sex' em 0 e 1\n",
    "df_merged['Sex'] = df_merged['Sex'].map({'male': 0, 'female': 1})\n",
    "\n",
    "# dummie variables para 'Embaked'\n",
    "embarked_dummies = pd.get_dummies(df_merged['Embarked'], prefix='Embarked')\n",
    "df_merged = pd.concat([df_merged, embarked_dummies], axis=1)\n",
    "df_merged.drop('Embarked', axis=1, inplace=True)\n",
    "\n",
    "display(df_merged.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Recuperando os datasets de treino e teste\n",
    "\n",
    "Aqui a gente acabou de preparar o *dataset* para o modelo e vou dividir ```df_merged``` em ```train``` e ```test```, exatamente como era no começo.\n",
    "\n",
    "Agora você consegue entender a razão da concatenação lá em cima. Teríamos trabalho dobrado e ainda teria o risco de errarmos em alguma etapa."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# recuperar datasets de treino e teste\n",
    "train = df_merged.iloc[:train_idx]\n",
    "test = df_merged.iloc[train_idx:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5 Construção do Modelo e Avaliação"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Eu vivo falando no meu [Instagram](http://instagram.com/carlos_melo.py) que a etapa mais demorada de qualquer projeto de *Data Science* é a Análise Exploratória de Dados e a preparação/tratamento deles.\n",
    "\n",
    "Todo cuidado que tivemos serviu para entender o problema, testar hipóteses e descartar dados desnecessários/redundantes.\n",
    "\n",
    "Com poucas linhas de código a gente consegue criar e otimizar modelos de *Machine Learning* básicos. Aqui, vamos construir dois:\n",
    "\n",
    "* Modelo de Regressão Logística\n",
    "* Modelo de Árvore de Decisão"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# importar bibliotecas do modelo de Machine Learning\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.tree import DecisionTreeClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Acurácia do modelo de Regressão Logística: 80.13\n"
     ]
    }
   ],
   "source": [
    "# criar um modelo de Regressão Logística\n",
    "lr_model = LogisticRegression(solver='liblinear')\n",
    "lr_model.fit(train, target)\n",
    "\n",
    "# verificar a acurácia do modelo\n",
    "acc_logReg = round(lr_model.score(train, target) * 100, 2)\n",
    "print(\"Acurácia do modelo de Regressão Logística: {}\".format(acc_logReg))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Com praticamente duas linhas, construímos um modelo de Regressão Logística com acurácia de 80% em cima do conjunto de dados de treino.\n",
    "\n",
    "Esse resultado está excelente para um primeiro modelo. Vou aproveitar e fazer a previsão sobre o *dataset* de teste para submetermos ao [Kaggle](http://www.kaggle.com) e comparar o desempenho."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_lr = lr_model.predict(test)\n",
    "\n",
    "submission = pd.DataFrame({\n",
    "    \"PassengerId\": passengerId,\n",
    "    \"Survived\": y_pred_lr\n",
    "})\n",
    "\n",
    "# gerar arquivo csv\n",
    "submission.to_csv('./submission_lr.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Como você vê na figura abaixo, nosso modelo conseguiu atingir um *score* de 75%. \n",
    "\n",
    "<img src=\"img/kaggle_lr.png\" width=\"80%\">\n",
    "\n",
    "Não é incomum termos desempenho inferior quando submetemos as previsões, pois nosso modelo foi treinado 100% em cima do conjunto de treino: isso significa que ele tem um *fit* bem melhor ao *dataset train*.\n",
    "\n",
    "Vamos ver como se sai um modelo baseado em Árvores de Decisão, seguindo os mesmo passos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Acurácia do modelo de Árvore de Decisão: 82.72\n"
     ]
    }
   ],
   "source": [
    "# criar um modelo de árvore de decisão\n",
    "tree_model = DecisionTreeClassifier(max_depth=3)\n",
    "tree_model.fit(train, target)\n",
    "\n",
    "# verificar a acurácia do modelo\n",
    "acc_tree = round(tree_model.score(train, target) * 100, 2)\n",
    "print(\"Acurácia do modelo de Árvore de Decisão: {}\".format(acc_tree))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Um pequeno ganho na acurácia, chegando agora a quase 83%. Novamente vou gerar um arquivo e submeter o resultado ao [Kaggle](http://www.kaggle.com)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_tree = tree_model.predict(test)\n",
    "\n",
    "submission = pd.DataFrame({\n",
    "    \"PassengerId\": passengerId,\n",
    "    \"Survived\": y_pred_tree\n",
    "})\n",
    "\n",
    "# gerar arquivo csv\n",
    "submission.to_csv('./submission_tree.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Como você vê na figura abaixo, nosso modelo conseguiu atingir um *score* de 75%. \n",
    "\n",
    "<img src=\"img/kaggle_tree.png\" width=\"80%\">\n",
    "\n",
    "Seguindo a previsão, melhoramos a nossa performance em cima do conjunto de testes também, obtendo um *score* de 78%.\n",
    "\n",
    "### Eu sobreviveria ao Titanic?\n",
    "\n",
    "Por fim, já que nosso modelo está pronto, vamos ver se eu e minha esposa sobreviveríamos ao naufrágio do Titanic.\n",
    "\n",
    "Vou considerar que viajaríamos na 2ª Classe, eu com 35 anos e ela com 30, levando nosso bebê Theo, tendo pago o preço médio do Ticket e embarcados no porto de Southampton (Reino Unido)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Carlos Melo:\t0\n",
      "Fernanda:\t1\n"
     ]
    }
   ],
   "source": [
    "# declarar os valores das variáveis para mim e minha esposa\n",
    "carlos_melo = np.array([2, 0, 35, 1, 1, 32.2, 0, 0, 0, 1]).reshape((1, -1))\n",
    "fernanda = np.array([2, 1, 30, 1, 1, 32.2, 0, 0, 0, 1]).reshape((1, -1))\n",
    "\n",
    "# verificar se nós teríamos sobrevivido\n",
    "print(\"Carlos Melo:\\t{}\".format(tree_model.predict(carlos_melo)[0]))\n",
    "print(\"Fernanda:\\t{}\".format(tree_model.predict(fernanda)[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Infelizmente, de acordo com meu próprio modelo, eu não teria sobrevivido ao desastre do Titanic. Já minha esposa foi classificada como sobrevivente. Parece que a realidade retratada no filme de James Cameron traz algumas verdades.\n",
    "\n",
    "Espero que tenha aproveitado este artigo e consiga replicar no seu computador.\n",
    "\n",
    "Com a sua evolução natural, você consegue cada vez mais trabalhar dados e parâmetros, além de aplicar conceitos como *feature engineering*, para obter resultados cada vez melhores.\n",
    "\n",
    "Fato é, o desafio do Titanic é a melhor escolha para quem está começando no Data Science, e já abre sua mente para outros tipos de problemas :)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}