Titanic_Machine_Learning_from_Disaster.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Titanic: Machine Learning from Disaster.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "0TwZavsuOTw5",
        "3Cux9YqROp5w",
        "b0M4PIY1QKtk",
        "jCfjvY-jTjCx",
        "VBWlX_G8Xk7w",
        "HEHqLSkHYQ0w",
        "RsGbsql2baEg",
        "-7YATr2edhes"
      ],
      "toc_visible": true,
      "authorship_tag": "ABX9TyNGUSy+CuZ+tyeTb1a9S0Rn",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/lunayee/Titanic-Machine-Learning-from-Disaster-/blob/main/Titanic_Machine_Learning_from_Disaster.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0TwZavsuOTw5"
      },
      "source": [
        "# 載入套件"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BK04za8NRxeY"
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "#繪圖相關套件\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.gridspec as gridspec\n",
        "import seaborn as sns\n",
        "plt.style.use('ggplot')\n",
        "#標籤編碼\n",
        "from sklearn.preprocessing import LabelEncoder, OneHotEncoder \n",
        "#決策樹\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "#隨機森林(Random Forest)\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from IPython.display import display\n",
        "\n",
        "import warnings\n",
        "warnings.filterwarnings( 'ignore' )"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3Cux9YqROp5w"
      },
      "source": [
        "# 上傳資料\n",
        "透過Kaggle資料競賽，下載鐵達尼號資料集(Link: https://www.kaggle.com/c/titanic/data)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L9ugjbCTOuL2"
      },
      "source": [
        "from google.colab import files\n",
        "uploaded = files.upload"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "b0M4PIY1QKtk"
      },
      "source": [
        "# 匯入資料\n",
        "透過pandas匯入訓練及測試資料，利用shape得知資料維度"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "MQbyOhnlQbpj",
        "outputId": "fe4f14bc-385e-4d38-f2d5-088ffea9a7e0"
      },
      "source": [
        "df_train = pd.read_csv('train.csv')\n",
        "df_test = pd.read_csv('test.csv')\n",
        "submit = pd.read_csv('gender_submission.csv')\n",
        "print(f'train {df_train.shape}')\n",
        "display(df_train.head())\n",
        "print(f'\\ntest{df_test.shape}')\n",
        "display(df_test.head())\n"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "train (891, 12)\n"
          ],
          "name": "stdout"
        },
        {
          "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  ...     Fare Cabin  Embarked\n",
              "0            1         0       3  ...   7.2500   NaN         S\n",
              "1            2         1       1  ...  71.2833   C85         C\n",
              "2            3         1       3  ...   7.9250   NaN         S\n",
              "3            4         1       1  ...  53.1000  C123         S\n",
              "4            5         0       3  ...   8.0500   NaN         S\n",
              "\n",
              "[5 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "test(418, 11)\n"
          ],
          "name": "stdout"
        },
        {
          "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>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>892</td>\n",
              "      <td>3</td>\n",
              "      <td>Kelly, Mr. James</td>\n",
              "      <td>male</td>\n",
              "      <td>34.5</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>330911</td>\n",
              "      <td>7.8292</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Q</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>893</td>\n",
              "      <td>3</td>\n",
              "      <td>Wilkes, Mrs. James (Ellen Needs)</td>\n",
              "      <td>female</td>\n",
              "      <td>47.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>363272</td>\n",
              "      <td>7.0000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>894</td>\n",
              "      <td>2</td>\n",
              "      <td>Myles, Mr. Thomas Francis</td>\n",
              "      <td>male</td>\n",
              "      <td>62.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>240276</td>\n",
              "      <td>9.6875</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Q</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>895</td>\n",
              "      <td>3</td>\n",
              "      <td>Wirz, Mr. Albert</td>\n",
              "      <td>male</td>\n",
              "      <td>27.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>315154</td>\n",
              "      <td>8.6625</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>896</td>\n",
              "      <td>3</td>\n",
              "      <td>Hirvonen, Mrs. Alexander (Helga E Lindqvist)</td>\n",
              "      <td>female</td>\n",
              "      <td>22.0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3101298</td>\n",
              "      <td>12.2875</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   PassengerId  Pclass  ... Cabin Embarked\n",
              "0          892       3  ...   NaN        Q\n",
              "1          893       3  ...   NaN        S\n",
              "2          894       2  ...   NaN        Q\n",
              "3          895       3  ...   NaN        S\n",
              "4          896       3  ...   NaN        S\n",
              "\n",
              "[5 rows x 11 columns]"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jCfjvY-jTjCx"
      },
      "source": [
        "# 欄位型態\n",
        "自定義Col_Types，觀察訓練及測試共同特徵欄位型態"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5sOYU7BOT8zh"
      },
      "source": [
        "#定義判別欄位型態的函數\n",
        "def Col_Types(Data):\n",
        "  Column_Types = Data.dtypes.to_frame().reset_index() #判別每個欄位的型態\n",
        "  Column_Types.columns =['ColumnName','Type']\n",
        "  Column_Types.sort_values(by='Type',inplace=True)\n",
        "  return Column_Types "
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "rNpdKLeLWt0u",
        "outputId": "e0226938-36e6-48e8-9c63-2d5d9fddfd6a"
      },
      "source": [
        "display(Col_Types(df_train))"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "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>ColumnName</th>\n",
              "      <th>Type</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>PassengerId</td>\n",
              "      <td>int64</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Survived</td>\n",
              "      <td>int64</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Pclass</td>\n",
              "      <td>int64</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>SibSp</td>\n",
              "      <td>int64</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>Parch</td>\n",
              "      <td>int64</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>Age</td>\n",
              "      <td>float64</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>Fare</td>\n",
              "      <td>float64</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Name</td>\n",
              "      <td>object</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Sex</td>\n",
              "      <td>object</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>Ticket</td>\n",
              "      <td>object</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>Cabin</td>\n",
              "      <td>object</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>Embarked</td>\n",
              "      <td>object</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     ColumnName     Type\n",
              "0   PassengerId    int64\n",
              "1      Survived    int64\n",
              "2        Pclass    int64\n",
              "6         SibSp    int64\n",
              "7         Parch    int64\n",
              "5           Age  float64\n",
              "9          Fare  float64\n",
              "3          Name   object\n",
              "4           Sex   object\n",
              "8        Ticket   object\n",
              "10        Cabin   object\n",
              "11     Embarked   object"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UyqVH1AqZ5Q2"
      },
      "source": [
        "# 遺漏值\n",
        "分析資料前要先確認是否有漏值，如果有需要用特徵工程來補植"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xYjpaeX2aHaW"
      },
      "source": [
        "#找漏值\n",
        "def Missing_Count(Data):\n",
        "  missing = Data.isnull().sum() #漏值的數量\n",
        "  missing = missing[missing>0]\n",
        "  missing.sort_values(inplace=True)\n",
        "\n",
        "  Missing_Count = pd.DataFrame({'ColumnName':missing.index,'MissingCount':missing.values})\n",
        "  Missing_Count['Percentage(%)']=Missing_Count['MissingCount'].apply(lambda x:round(x/Data.shape[0]*100,2))\n",
        "  return Missing_Count"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 351
        },
        "id": "qOXNKh5iSqtF",
        "outputId": "b9089b62-b19c-4c9f-8914-93449f171394"
      },
      "source": [
        "print('train:')\n",
        "display(Missing_Count(df_train))\n",
        "print('test:')\n",
        "display(Missing_Count(df_test))"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "train:\n"
          ],
          "name": "stdout"
        },
        {
          "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>ColumnName</th>\n",
              "      <th>MissingCount</th>\n",
              "      <th>Percentage(%)</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Embarked</td>\n",
              "      <td>2</td>\n",
              "      <td>0.22</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Age</td>\n",
              "      <td>177</td>\n",
              "      <td>19.87</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Cabin</td>\n",
              "      <td>687</td>\n",
              "      <td>77.10</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  ColumnName  MissingCount  Percentage(%)\n",
              "0   Embarked             2           0.22\n",
              "1        Age           177          19.87\n",
              "2      Cabin           687          77.10"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "test:\n"
          ],
          "name": "stdout"
        },
        {
          "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>ColumnName</th>\n",
              "      <th>MissingCount</th>\n",
              "      <th>Percentage(%)</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Fare</td>\n",
              "      <td>1</td>\n",
              "      <td>0.24</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Age</td>\n",
              "      <td>86</td>\n",
              "      <td>20.57</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Cabin</td>\n",
              "      <td>327</td>\n",
              "      <td>78.23</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  ColumnName  MissingCount  Percentage(%)\n",
              "0       Fare             1           0.24\n",
              "1        Age            86          20.57\n",
              "2      Cabin           327          78.23"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "w6Z0_MJWWg1a"
      },
      "source": [
        "# 統計量\n",
        "利用describe()，可以計算每個數值欄位的統計量，可當作判斷離群值的依據，並決定是否要對資料進行正規化or標準化"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 731
        },
        "id": "-RrFhpKTW74k",
        "outputId": "bde398ad-efc4-480b-d7e8-4e842e822436"
      },
      "source": [
        "print('train:')\n",
        "display(df_train.describe())\n",
        "\n",
        "print('teat:')\n",
        "display(df_test.describe())"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "train:\n"
          ],
          "name": "stdout"
        },
        {
          "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>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  ...       SibSp       Parch        Fare\n",
              "count   891.000000  891.000000  891.000000  ...  891.000000  891.000000  891.000000\n",
              "mean    446.000000    0.383838    2.308642  ...    0.523008    0.381594   32.204208\n",
              "std     257.353842    0.486592    0.836071  ...    1.102743    0.806057   49.693429\n",
              "min       1.000000    0.000000    1.000000  ...    0.000000    0.000000    0.000000\n",
              "25%     223.500000    0.000000    2.000000  ...    0.000000    0.000000    7.910400\n",
              "50%     446.000000    0.000000    3.000000  ...    0.000000    0.000000   14.454200\n",
              "75%     668.500000    1.000000    3.000000  ...    1.000000    0.000000   31.000000\n",
              "max     891.000000    1.000000    3.000000  ...    8.000000    6.000000  512.329200\n",
              "\n",
              "[8 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "teat:\n"
          ],
          "name": "stdout"
        },
        {
          "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>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>418.000000</td>\n",
              "      <td>418.000000</td>\n",
              "      <td>332.000000</td>\n",
              "      <td>418.000000</td>\n",
              "      <td>418.000000</td>\n",
              "      <td>417.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>1100.500000</td>\n",
              "      <td>2.265550</td>\n",
              "      <td>30.272590</td>\n",
              "      <td>0.447368</td>\n",
              "      <td>0.392344</td>\n",
              "      <td>35.627188</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>120.810458</td>\n",
              "      <td>0.841838</td>\n",
              "      <td>14.181209</td>\n",
              "      <td>0.896760</td>\n",
              "      <td>0.981429</td>\n",
              "      <td>55.907576</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>892.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.170000</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>996.250000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>21.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>7.895800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>1100.500000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>27.000000</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>1204.750000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>39.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>31.500000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>1309.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>76.000000</td>\n",
              "      <td>8.000000</td>\n",
              "      <td>9.000000</td>\n",
              "      <td>512.329200</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       PassengerId      Pclass         Age       SibSp       Parch        Fare\n",
              "count   418.000000  418.000000  332.000000  418.000000  418.000000  417.000000\n",
              "mean   1100.500000    2.265550   30.272590    0.447368    0.392344   35.627188\n",
              "std     120.810458    0.841838   14.181209    0.896760    0.981429   55.907576\n",
              "min     892.000000    1.000000    0.170000    0.000000    0.000000    0.000000\n",
              "25%     996.250000    1.000000   21.000000    0.000000    0.000000    7.895800\n",
              "50%    1100.500000    3.000000   27.000000    0.000000    0.000000   14.454200\n",
              "75%    1204.750000    3.000000   39.000000    1.000000    0.000000   31.500000\n",
              "max    1309.000000    3.000000   76.000000    8.000000    9.000000  512.329200"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VBWlX_G8Xk7w"
      },
      "source": [
        "# 資料分析\n",
        "使用append()隊訓練及與測試集的資料合併，方便後續修補或調整資料"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "7FsXz1XVX2xF",
        "outputId": "4334b418-f9ea-440d-d132-73a09c3b292a"
      },
      "source": [
        "df_data = df_train.append(df_test)\n",
        "display(df_data.head())"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "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>0.0</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>1.0</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>1.0</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>1.0</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>0.0</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  ...     Fare Cabin  Embarked\n",
              "0            1       0.0       3  ...   7.2500   NaN         S\n",
              "1            2       1.0       1  ...  71.2833   C85         C\n",
              "2            3       1.0       3  ...   7.9250   NaN         S\n",
              "3            4       1.0       1  ...  53.1000  C123         S\n",
              "4            5       0.0       3  ...   8.0500   NaN         S\n",
              "\n",
              "[5 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HEHqLSkHYQ0w"
      },
      "source": [
        "## 生還者比率\n",
        "分析生還者與罹難者比率是否有明顯極大的落差，看有沒有資訊不平衡的問題，方法為:使用value_count()表示各類別中相異的數量"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "Cr3gdMvHYXcM",
        "outputId": "26884a2a-f798-4ac6-ea4b-3b645f1a601b"
      },
      "source": [
        "Survived_Count = df_data['Survived'].value_counts().reset_index()\n",
        "Survived_Count.columns = ['Survived','Count']\n",
        "Survived_Count"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Survived</th>\n",
              "      <th>Count</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.0</td>\n",
              "      <td>549</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1.0</td>\n",
              "      <td>342</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Survived  Count\n",
              "0       0.0    549\n",
              "1       1.0    342"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "MJ169qOdaGof",
        "outputId": "5f661366-b8dc-4310-dec2-5acab4b722d6"
      },
      "source": [
        "plt.figure()\n",
        "df_data['Survived'].value_counts().plot(kind='pie',autopct='%1.2f%%')\n",
        "plt.title('Survival')\n",
        "plt.ylabel('')\n",
        "plt.show"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<function matplotlib.pyplot.show>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RsGbsql2baEg"
      },
      "source": [
        "## 相關係數\n",
        "利用corr()，計算相關係數，看有沒有明顯的線性關係"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "0IGpNG6HbodV",
        "outputId": "53a6310b-4e7b-4653-961f-01c60161f77f"
      },
      "source": [
        "Corr_Matrix = df_train.corr()\n",
        "Corr = Corr_Matrix.loc['Survived'].sort_values()[:-1]\n",
        "Corr = pd.DataFrame({'Survived':Corr})\n",
        "Corr"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Survived</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Pclass</th>\n",
              "      <td>-0.338481</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Age</th>\n",
              "      <td>-0.077221</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>SibSp</th>\n",
              "      <td>-0.035322</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>PassengerId</th>\n",
              "      <td>-0.005007</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Parch</th>\n",
              "      <td>0.081629</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Fare</th>\n",
              "      <td>0.257307</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             Survived\n",
              "Pclass      -0.338481\n",
              "Age         -0.077221\n",
              "SibSp       -0.035322\n",
              "PassengerId -0.005007\n",
              "Parch        0.081629\n",
              "Fare         0.257307"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-7YATr2edhes"
      },
      "source": [
        "## 生存率\n",
        "觀察其他欄位與生存率的關係"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "id": "T1xBORZcdsqW",
        "outputId": "07ab82c4-e717-44a2-cfc9-0e39ab279b6b"
      },
      "source": [
        "selected_cols=['Sex','Pclass','Age','SibSp','Parch','Embarked']\n",
        "plt.figure(figsize=(10,len(selected_cols)*5))\n",
        "gs = gridspec.GridSpec(len(selected_cols),1)\n",
        "for i, col in enumerate( df_data[selected_cols] ) :        \n",
        "    ax = plt.subplot( gs[i] )\n",
        "    sns.countplot( df_data[col], hue=df_data.Survived )\n",
        "    ax.set_ylabel( 'Counts' )\n",
        "    ax.legend( loc=1 )   # 圖示位置 upper right:1 ; upper left:2\n",
        "    for p in ax.patches:\n",
        "        ax.annotate( '{:,}'.format(p.get_height()), (p.get_x(), p.get_height()+1.5) )#顯示數值\n",
        "plt.show()"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x2160 with 6 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nUHofS1zOyVn"
      },
      "source": [
        "# 特徵工程\n",
        "分析資料後，處理一些無法直接餵入模型的欄位"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WKtius-2PG4s"
      },
      "source": [
        "## (1)字串欄位"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8SmR-il2Qbgd"
      },
      "source": [
        "### 姓名\n",
        "發現擷取出稱謂可能可以提升預測準確度，處理字串資料常用函數有split(),repalce()"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rJi_JbGqPEas",
        "outputId": "d938720c-a83b-44e4-b975-43c60846adbc"
      },
      "source": [
        "# Method 1: split()\n",
        "df_data['Title'] = df_data.Name.str.split(', ', expand=True)[1]\n",
        "df_data['Title'] = df_data.Title.str.split('.', expand=True)[0]\n",
        "df_data['Title'].unique()#留下不重複的值"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['Mr', 'Mrs', 'Miss', 'Master', 'Don', 'Rev', 'Dr', 'Mme', 'Ms',\n",
              "       'Major', 'Lady', 'Sir', 'Mlle', 'Col', 'Capt', 'the Countess',\n",
              "       'Jonkheer', 'Dona'], dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4ezxPyrnQNFp",
        "outputId": "cf6f00e3-dfbe-4bfa-f731-920813a8745b"
      },
      "source": [
        "# Method 2: 正規表示法(Regular Expression)\n",
        "import re\n",
        "\n",
        "regex = re.compile( ' ([A-Za-z]+)\\.' )  \n",
        "df_data['Title'] = df_data.Name.map( lambda x:regex.search(x)[0] )\n",
        "# Dropping the first and the last words\n",
        "df_data['Title'] = df_data.Title.map( lambda x:x[1:][:-1] )  \n",
        "df_data['Title'].unique()"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['Mr', 'Mrs', 'Miss', 'Master', 'Don', 'Rev', 'Dr', 'Mme', 'Ms',\n",
              "       'Major', 'Lady', 'Sir', 'Mlle', 'Col', 'Capt', 'Countess',\n",
              "       'Jonkheer', 'Dona'], dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9nGw8xVNluJ9"
      },
      "source": [
        "將同樣性別或關係的稱謂合併，避免產生某些稱謂僅有少數乘客，例如：Rev(Reverend)為牧師的尊稱、Countess為伯爵夫人等，影響模型的預測準確度。這裏我們將利用 replace( ) 函數，將上述擷取出的稱謂總歸為五類：Mr、Mrs、Miss、Master、Rare。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QlJZEvuMlr_2",
        "outputId": "31de58cc-e3d3-46c3-bce5-5f6dd66e10c2"
      },
      "source": [
        "df_data['Title'] = df_data.Title.replace( ['Don','Rev','Dr','Major','Lady','Sir','Col','Capt','Countess','Jonkheer','Dona'], 'Rare' )\n",
        "df_data['Title'] = df_data.Title.replace( ['Ms','Mlle'], 'Miss' )\n",
        "df_data['Title'] = df_data.Title.replace( 'Mme', 'Mrs' )\n",
        "df_data['Title'].unique()"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['Mr', 'Mrs', 'Miss', 'Master', 'Rare'], dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BIk38o5MmPOY"
      },
      "source": [
        "# 刪除原始資料中的 Name 欄位 \n",
        "df_data.drop( 'Name', axis=1, inplace=True )#inplace=True表示修改原本值，默認為flase"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nOULvPjLQ8A9"
      },
      "source": [
        "### 船票號碼\n",
        "假設船票中，前面英文字母表示位於同個區域中，擷取船票字串部分文字訊息，作為位置分區"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bZeqghIQRUGT",
        "outputId": "7986edc1-5e63-473e-e377-39aca79d9c28"
      },
      "source": [
        "df_data['Ticket_info']=df_data.Ticket.apply(lambda x:x.replace('.','').replace('/','').split(' ')[0].strip() if not x.isdigit() else 'X')#strip()刪除前後空格\n",
        "df_data['Ticket_info'].unique()"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['A5', 'PC', 'STONO2', 'X', 'PP', 'CA', 'SCParis', 'SCA4', 'A4',\n",
              "       'SP', 'SOC', 'WC', 'SOTONOQ', 'WEP', 'STONO', 'C', 'SCPARIS',\n",
              "       'SOP', 'Fa', 'LINE', 'FCC', 'SWPP', 'SCOW', 'PPP', 'SC', 'SCAH',\n",
              "       'AS', 'SOPP', 'FC', 'SOTONO2', 'CASOTON', 'SCA3', 'STONOQ', 'AQ4',\n",
              "       'A', 'LP', 'AQ3'], dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gMlxvs2rVtB6"
      },
      "source": [
        "#刪除本來的Ticket資料\n",
        "df_data.drop('Ticket',axis=1,inplace=True)"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fqQAg64pWZxW"
      },
      "source": [
        "## (2)填補缺漏值\n",
        "運用平均、中位數、眾數、線性回歸、決策數等補植，並利用fillna函數填補"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 198
        },
        "id": "-wJ85wfrW2Rv",
        "outputId": "3b208fbc-2fd2-4a7c-cc40-29f13fa2c002"
      },
      "source": [
        "Missing_Count(df_data.drop('Survived',axis=1))\n"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>ColumnName</th>\n",
              "      <th>MissingCount</th>\n",
              "      <th>Percentage(%)</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Fare</td>\n",
              "      <td>1</td>\n",
              "      <td>0.08</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Embarked</td>\n",
              "      <td>2</td>\n",
              "      <td>0.15</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Age</td>\n",
              "      <td>263</td>\n",
              "      <td>20.09</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Cabin</td>\n",
              "      <td>1014</td>\n",
              "      <td>77.46</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  ColumnName  MissingCount  Percentage(%)\n",
              "0       Fare             1           0.08\n",
              "1   Embarked             2           0.15\n",
              "2        Age           263          20.09\n",
              "3      Cabin          1014          77.46"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "snA5dqrVaq64"
      },
      "source": [
        "### 票價:用中位數填補"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vRNmbGaea5F4"
      },
      "source": [
        "df_data['Fare'].fillna(df_data.Fare.median(),inplace=True)"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cNrMaXn8bT2O"
      },
      "source": [
        "### 登船港口:用眾數填補"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 105
        },
        "id": "Oz6YAcSMbgQu",
        "outputId": "71abd8ec-0679-42f5-8966-c7856faa06c1"
      },
      "source": [
        "display(df_data['Embarked'].value_counts())\n",
        "df_data['Embarked'].fillna('S',inplace=True)"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "S    914\n",
              "C    270\n",
              "Q    123\n",
              "Name: Embarked, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4KScuLLbcBZe"
      },
      "source": [
        "### 年齡\n",
        "因為它缺漏太多，可能會影響預測結果，此時用age搭配Sex,Pclass欄位觀察"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 379
        },
        "id": "VNaFyP0gcYqx",
        "outputId": "213e5714-7250-4f15-ddb0-594a370670ea"
      },
      "source": [
        "df_data['isAge']=df_data['Age'].isnull().apply(lambda x:0 if x==True else 1)\n",
        "display(pd.crosstab(df_data.isAge,df_data.Sex,margins=True))\n",
        "display(pd.crosstab(df_data.isAge,df_data.Pclass,margins=True))"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "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>Sex</th>\n",
              "      <th>female</th>\n",
              "      <th>male</th>\n",
              "      <th>All</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>isAge</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>78</td>\n",
              "      <td>185</td>\n",
              "      <td>263</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>388</td>\n",
              "      <td>658</td>\n",
              "      <td>1046</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>All</th>\n",
              "      <td>466</td>\n",
              "      <td>843</td>\n",
              "      <td>1309</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "Sex    female  male   All\n",
              "isAge                    \n",
              "0          78   185   263\n",
              "1         388   658  1046\n",
              "All       466   843  1309"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "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>Pclass</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>All</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>isAge</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>39</td>\n",
              "      <td>16</td>\n",
              "      <td>208</td>\n",
              "      <td>263</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>284</td>\n",
              "      <td>261</td>\n",
              "      <td>501</td>\n",
              "      <td>1046</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>All</th>\n",
              "      <td>323</td>\n",
              "      <td>277</td>\n",
              "      <td>709</td>\n",
              "      <td>1309</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "Pclass    1    2    3   All\n",
              "isAge                      \n",
              "0        39   16  208   263\n",
              "1       284  261  501  1046\n",
              "All     323  277  709  1309"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uDM3f2xPesvr"
      },
      "source": [
        "Age 欄位的缺漏值明顯分佈在 Pclass 3 中，約佔 29.34%(=208/709)，進一步觀察在 Pclass 1 及 Pclass 2 的乘客中，Age 欄位對於存活率是否有明顯的影響？"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 321
        },
        "id": "AFrMXytkeynQ",
        "outputId": "2daa1f10-5499-4664-be11-b77735849d18"
      },
      "source": [
        "index_S=(df_data.isAge==1)&(df_data.Pclass!=3)&(df_data.Survived==1)\n",
        "index_D=(df_data.isAge==1)&(df_data.Pclass!=3)&(df_data.Survived==0)\n",
        "plt.figure()\n",
        "sns.distplot(df_data.loc[index_S,'Age'],label='Survived',bins=20)\n",
        "sns.distplot(df_data.loc[index_D,'Age'],label='Dead',bins=20)\n",
        "plt.title('Pclass=1or2')\n",
        "plt.legend()"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f3b71664898>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yNhUSGEZhRYF"
      },
      "source": [
        "從上圖可觀察到，在 Pclass 1 及 Pclass 2 的乘客中，年齡約小於 17 歲的乘客中有明顯較高的生存率，若年齡超過 17 歲則生存率並無明顯的差別，所以我們可大膽假設 Age 欄位與生存率具有關聯性，另外，因為年齡與稱謂(Title)也有相當程度的關聯，因此，我們將先計算每個稱謂的年齡平均值，再用此平均值來填補每個 Title 中所對應 Age 欄位的缺漏值"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 451
        },
        "id": "MC3-LAmwhTUH",
        "outputId": "6ac35685-ee6d-4d61-fc66-00ca65ced0f5"
      },
      "source": [
        "# 計算每個 Title 的年齡平均值及中位數\n",
        "Age_Mean=df_data[['Title','Age']].groupby(by=['Title']).mean()\n",
        "Age_Median=df_data[['Title','Age']].groupby(by=['Title']).median()\n",
        "\n",
        "Age_Mean.columns =['Age Mean']\n",
        "Age_Median.columns =['Age Median']\n",
        "Age_Mean.reset_index( inplace=True )#整理表格\n",
        "Age_Median.reset_index( inplace=True )\n",
        "\n",
        "display(Age_Mean)\n",
        "display(Age_Median)"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "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>Title</th>\n",
              "      <th>Age Mean</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Master</td>\n",
              "      <td>5.482642</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Miss</td>\n",
              "      <td>21.824366</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Mr</td>\n",
              "      <td>32.252151</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Mrs</td>\n",
              "      <td>36.918129</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Rare</td>\n",
              "      <td>45.178571</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    Title   Age Mean\n",
              "0  Master   5.482642\n",
              "1    Miss  21.824366\n",
              "2      Mr  32.252151\n",
              "3     Mrs  36.918129\n",
              "4    Rare  45.178571"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "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>Title</th>\n",
              "      <th>Age Median</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Master</td>\n",
              "      <td>4.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Miss</td>\n",
              "      <td>22.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Mr</td>\n",
              "      <td>29.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Mrs</td>\n",
              "      <td>35.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Rare</td>\n",
              "      <td>47.5</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    Title  Age Median\n",
              "0  Master         4.0\n",
              "1    Miss        22.0\n",
              "2      Mr        29.0\n",
              "3     Mrs        35.0\n",
              "4    Rare        47.5"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0t4H7m1UkpMi"
      },
      "source": [
        "# 利用每個 Title 的年齡平均數，填補每個 Title 所對應 Age 的缺漏值\n",
        "df_data.loc[(df_data.Age.isnull())&(df_data.Title=='Master'),'Age'] = Age_Mean.loc[Age_Mean.Title=='Master','Age Mean'][0]\n",
        "df_data.loc[(df_data.Age.isnull())&(df_data.Title=='Miss'),'Age'] = Age_Mean.loc[Age_Mean.Title=='Miss','Age Mean'][1]\n",
        "df_data.loc[(df_data.Age.isnull())&(df_data.Title=='Mr'),'Age'] = Age_Mean.loc[Age_Mean.Title=='Mr','Age Mean'][2]\n",
        "df_data.loc[(df_data.Age.isnull())&(df_data.Title=='Mrs'),'Age'] = Age_Mean.loc[Age_Mean.Title=='Mrs','Age Mean'][3]\n",
        "df_data.loc[(df_data.Age.isnull())&(df_data.Title=='Rare'),'Age'] = Age_Mean.loc[Age_Mean.Title=='Rare','Age Mean'][4]\n",
        "\n",
        "# 刪除 isAge 欄位 \n",
        "df_data.drop( 'isAge', axis=1, inplace=True )"
      ],
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SnBgm6r6pDde"
      },
      "source": [
        "填滿 Age 欄位的缺漏值後，我們將年齡特徵再區分成未滿17歲及17歲以上，未滿17歲的乘客標為1，17歲以上的乘客標為0"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 504
        },
        "id": "7eUOepgYpFzV",
        "outputId": "76b2c575-e216-4edb-bb8a-8ff03cc2ba42"
      },
      "source": [
        "# 乘客年齡是否未滿17歲(是：1；否：0)\n",
        "df_data['is_Age_17'] = (df_data.Age<17)*1\n",
        "df_data"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>PassengerId</th>\n",
              "      <th>Survived</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>Cabin</th>\n",
              "      <th>Embarked</th>\n",
              "      <th>Title</th>\n",
              "      <th>Ticket_info</th>\n",
              "      <th>is_Age_17</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>male</td>\n",
              "      <td>22.000000</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>7.2500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "      <td>Mr</td>\n",
              "      <td>A5</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>female</td>\n",
              "      <td>38.000000</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>71.2833</td>\n",
              "      <td>C85</td>\n",
              "      <td>C</td>\n",
              "      <td>Mrs</td>\n",
              "      <td>PC</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>1.0</td>\n",
              "      <td>3</td>\n",
              "      <td>female</td>\n",
              "      <td>26.000000</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>7.9250</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "      <td>Miss</td>\n",
              "      <td>STONO2</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>female</td>\n",
              "      <td>35.000000</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>53.1000</td>\n",
              "      <td>C123</td>\n",
              "      <td>S</td>\n",
              "      <td>Mrs</td>\n",
              "      <td>X</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>male</td>\n",
              "      <td>35.000000</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>8.0500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "      <td>Mr</td>\n",
              "      <td>X</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",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>413</th>\n",
              "      <td>1305</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3</td>\n",
              "      <td>male</td>\n",
              "      <td>32.252151</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>8.0500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "      <td>Mr</td>\n",
              "      <td>A5</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>414</th>\n",
              "      <td>1306</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1</td>\n",
              "      <td>female</td>\n",
              "      <td>39.000000</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>108.9000</td>\n",
              "      <td>C105</td>\n",
              "      <td>C</td>\n",
              "      <td>Rare</td>\n",
              "      <td>PC</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>415</th>\n",
              "      <td>1307</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3</td>\n",
              "      <td>male</td>\n",
              "      <td>38.500000</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>7.2500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "      <td>Mr</td>\n",
              "      <td>SOTONOQ</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>416</th>\n",
              "      <td>1308</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3</td>\n",
              "      <td>male</td>\n",
              "      <td>32.252151</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>8.0500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "      <td>Mr</td>\n",
              "      <td>X</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>417</th>\n",
              "      <td>1309</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3</td>\n",
              "      <td>male</td>\n",
              "      <td>5.482642</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>22.3583</td>\n",
              "      <td>NaN</td>\n",
              "      <td>C</td>\n",
              "      <td>Master</td>\n",
              "      <td>X</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1309 rows × 13 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     PassengerId  Survived  Pclass  ...   Title  Ticket_info  is_Age_17\n",
              "0              1       0.0       3  ...      Mr           A5          0\n",
              "1              2       1.0       1  ...     Mrs           PC          0\n",
              "2              3       1.0       3  ...    Miss       STONO2          0\n",
              "3              4       1.0       1  ...     Mrs            X          0\n",
              "4              5       0.0       3  ...      Mr            X          0\n",
              "..           ...       ...     ...  ...     ...          ...        ...\n",
              "413         1305       NaN       3  ...      Mr           A5          0\n",
              "414         1306       NaN       1  ...    Rare           PC          0\n",
              "415         1307       NaN       3  ...      Mr      SOTONOQ          0\n",
              "416         1308       NaN       3  ...      Mr            X          0\n",
              "417         1309       NaN       3  ...  Master            X          1\n",
              "\n",
              "[1309 rows x 13 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MunCM2mSpuhs"
      },
      "source": [
        "### 船艙號碼\n",
        "缺漏值過多，僅取出頭一個英文字母作為分類標籤其他缺漏值以NoCabin表示"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c921EN4sqIb1",
        "outputId": "ad4af1f3-97af-4437-86af-6fc5d588405a"
      },
      "source": [
        "#觀察船艙的種類\n",
        "df_data['Cabin'].unique()"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([nan, 'C85', 'C123', 'E46', 'G6', 'C103', 'D56', 'A6',\n",
              "       'C23 C25 C27', 'B78', 'D33', 'B30', 'C52', 'B28', 'C83', 'F33',\n",
              "       'F G73', 'E31', 'A5', 'D10 D12', 'D26', 'C110', 'B58 B60', 'E101',\n",
              "       'F E69', 'D47', 'B86', 'F2', 'C2', 'E33', 'B19', 'A7', 'C49', 'F4',\n",
              "       'A32', 'B4', 'B80', 'A31', 'D36', 'D15', 'C93', 'C78', 'D35',\n",
              "       'C87', 'B77', 'E67', 'B94', 'C125', 'C99', 'C118', 'D7', 'A19',\n",
              "       'B49', 'D', 'C22 C26', 'C106', 'C65', 'E36', 'C54',\n",
              "       'B57 B59 B63 B66', 'C7', 'E34', 'C32', 'B18', 'C124', 'C91', 'E40',\n",
              "       'T', 'C128', 'D37', 'B35', 'E50', 'C82', 'B96 B98', 'E10', 'E44',\n",
              "       'A34', 'C104', 'C111', 'C92', 'E38', 'D21', 'E12', 'E63', 'A14',\n",
              "       'B37', 'C30', 'D20', 'B79', 'E25', 'D46', 'B73', 'C95', 'B38',\n",
              "       'B39', 'B22', 'C86', 'C70', 'A16', 'C101', 'C68', 'A10', 'E68',\n",
              "       'B41', 'A20', 'D19', 'D50', 'D9', 'A23', 'B50', 'A26', 'D48',\n",
              "       'E58', 'C126', 'B71', 'B51 B53 B55', 'D49', 'B5', 'B20', 'F G63',\n",
              "       'C62 C64', 'E24', 'C90', 'C45', 'E8', 'B101', 'D45', 'C46', 'D30',\n",
              "       'E121', 'D11', 'E77', 'F38', 'B3', 'D6', 'B82 B84', 'D17', 'A36',\n",
              "       'B102', 'B69', 'E49', 'C47', 'D28', 'E17', 'A24', 'C50', 'B42',\n",
              "       'C148', 'B45', 'B36', 'A21', 'D34', 'A9', 'C31', 'B61', 'C53',\n",
              "       'D43', 'C130', 'C132', 'C55 C57', 'C116', 'F', 'A29', 'C6', 'C28',\n",
              "       'C51', 'C97', 'D22', 'B10', 'E45', 'E52', 'A11', 'B11', 'C80',\n",
              "       'C89', 'F E46', 'B26', 'F E57', 'A18', 'E60', 'E39 E41',\n",
              "       'B52 B54 B56', 'C39', 'B24', 'D40', 'D38', 'C105'], dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TimN24SsqVAl",
        "outputId": "b01d0490-84c1-46c5-8c8c-d0b1bc04cb9d"
      },
      "source": [
        "#取出第一個字母\n",
        "df_data['Cabin']=df_data['Cabin'].apply(lambda x:str(x)[0] if not pd.isnull(x) else 'NoCabin')\n",
        "df_data['Cabin'].unique()"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['NoCabin', 'C', 'E', 'G', 'D', 'A', 'B', 'F', 'T'], dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 29
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0vlUx28Er0gt"
      },
      "source": [
        "最後檢查有沒有沒補到的值"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 90
        },
        "id": "Y3vnZaTCr0CR",
        "outputId": "025c9bff-0025-4224-de3b-34dc2a485821"
      },
      "source": [
        "Missing_Count(df_data)"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>ColumnName</th>\n",
              "      <th>MissingCount</th>\n",
              "      <th>Percentage(%)</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Survived</td>\n",
              "      <td>418</td>\n",
              "      <td>31.93</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  ColumnName  MissingCount  Percentage(%)\n",
              "0   Survived           418          31.93"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jj1ZcI_pu9wp"
      },
      "source": [
        "## (3)離群值\n",
        "利用直方圖或盒鬚圖判斷"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Spj-NzFRvTis"
      },
      "source": [
        "### 票價"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 304
        },
        "id": "YC3f4_2WvcLU",
        "outputId": "f9d4dfa4-25ff-4ff3-bbf7-5c8736c7a339"
      },
      "source": [
        "# 對 Fare 欄位取對數\n",
        "df_data['LogFare'] = np.log1p( df_data.Fare )\n",
        "# 直方圖(Histogram)\n",
        "fig, axs = plt.subplots(1,2)\n",
        "\n",
        "plt.subplot(1,2,1)\n",
        "sns.distplot(df_data.Fare,bins=45,label='bins=45')\n",
        "plt.xlabel('Fare')\n",
        "plt.ylabel('Count')\n",
        "plt.legend()\n",
        "\n",
        "plt.subplot( 1,2,2 )\n",
        "sns.distplot( df_data.LogFare, kde=True, bins=45, color='skyblue', label='bins = 45' )\n",
        "plt.xlabel( 'Log Fare' ) \n",
        "plt.ylabel( '' ) \n",
        "plt.legend( )\n",
        "\n"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f3b6fc4b080>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 31
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 304
        },
        "id": "Mzu_14rwxrln",
        "outputId": "65518cb3-df3a-40aa-a3ae-fa0f7d6dec86"
      },
      "source": [
        "#盒鬚圖\n",
        "fig,axs=plt.subplots(1,2)\n",
        "\n",
        "plt.subplot(1,2,1)\n",
        "sns.boxplot(df_data.Fare)\n",
        "plt.xlabel('Fare')\n",
        "\n",
        "plt.subplot( 1,2,2 )\n",
        "sns.boxplot( df_data.LogFare, orient='h', color='skyblue' )\n",
        "plt.xlabel( 'Log Fare' ) "
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'Log Fare')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YdgtEhyXycGf"
      },
      "source": [
        "從上面直方圖及盒鬚圖可知，初步判斷 Fare 欄位可能存有離群值，計算 Fare 欄位的四分位數(Quartile)，且以四分位數對 Fare 欄位進行分組，觀察生存率是否會因票價組別的高低有所差異，"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 234
        },
        "id": "nK3kIvz3y29t",
        "outputId": "3c44ce67-4804-45be-a8bb-c7866f18c50b"
      },
      "source": [
        "#計算各個百分位數\n",
        "Q1 =np.percentile( df_data.Fare,q=25)\n",
        "Q2 =np.percentile( df_data.Fare,q=50)\n",
        "Q3 =np.percentile( df_data.Fare,q=75)\n",
        "IQR=Q3-Q1\n",
        "\n",
        "# 依照四分位數，對 Fare 欄位進行分組\n",
        "Fare_bin = [ 0, Q1, Q2, Q3, Q3+1.5*IQR, df_data.Fare.max() ]\n",
        "df_data[ 'Fare_Group' ] = pd.cut( df_data.Fare.values, Fare_bin )\n",
        "Group_Count=df_data[ 'Fare_Group' ].value_counts().reset_index()\n",
        "Group_Count.columns=['Fare_Group', 'Counts']\n",
        "Group_Count.sort_values(by='Fare_Group')"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Fare_Group</th>\n",
              "      <th>Counts</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>(0.0, 7.896]</td>\n",
              "      <td>320</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>(7.896, 14.454]</td>\n",
              "      <td>321</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>(14.454, 31.275]</td>\n",
              "      <td>328</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>(31.275, 66.344]</td>\n",
              "      <td>152</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>(66.344, 512.329]</td>\n",
              "      <td>171</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          Fare_Group  Counts\n",
              "2       (0.0, 7.896]     320\n",
              "1    (7.896, 14.454]     321\n",
              "0   (14.454, 31.275]     328\n",
              "4   (31.275, 66.344]     152\n",
              "3  (66.344, 512.329]     171"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4Q8F77_R2RoT"
      },
      "source": [
        "再次藉由長條圖及盒鬚圖，觀察不同的票價組別，彼此間生存率的差異性"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 359
        },
        "id": "CUmWy1cn2UAP",
        "outputId": "785d6470-5ee5-4b39-aa17-d9b5a9c86e36"
      },
      "source": [
        "plt.subplots( figsize=(12,5) )\n",
        "sns.countplot(df_data.Fare_Group,hue=df_data.Survived)\n",
        "plt.ylabel('Count')\n"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0, 0.5, 'Count')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 34
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 566
        },
        "id": "uZLm7rVK3h6M",
        "outputId": "54aaac26-2957-4175-e840-e5dab7dab264"
      },
      "source": [
        "plt.subplots( figsize=(15,12) )\n",
        "sns.boxplot( x='LogFare', y='Fare_Group', data=df_data, hue='Survived' )\n"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f3b6fa1ec18>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 35
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x864 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "O7N26RhU4-Pa"
      },
      "source": [
        "從上圖可明顯觀察出，票價組別越高生存率相對越大，且每一票價組別中，生還者的票價中位數皆高於罹難者的票價中位數；所以，我們可大膽推測 Fare 與生存率有相關性，Fare 越高生存機率越大，且對 Fare 欄位僅做對數轉換並不做離群值的處理。"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "v-AS_VLu6JZA"
      },
      "source": [
        "## (4)讀熟編碼、標籤編碼\n",
        "對於定性且無序的特徵欄位可以利用get_dummies()將其轉換成定量資料，這裏我們對 Embarked 欄位做獨熱編碼；另外，因為女性的生存率明顯高於男性，故我們對 Sex 欄位做標籤編碼。\n",
        "\n",
        "\n",
        "*   熟讀編碼(One hot encoding) : 為每個類別新增一個欄位，用 0/1 表示是否\n",
        "*   標籤編碼(Label encoding) : 把每個類別 mapping 到某個整數，不會增加新欄位\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QL73oQyK9C_m"
      },
      "source": [
        "### 登船港口"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hF761RlZ6kSF"
      },
      "source": [
        "OneHot_Embarked = pd.get_dummies( df_data.Embarked,prefix='Embarked' )\n",
        "df_data=pd.concat([df_data,OneHot_Embarked],axis=1)\n",
        "df_data.drop( 'Embarked', axis=1, inplace=True )\n"
      ],
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vuH1TWYg9Ian"
      },
      "source": [
        "### 性別"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Wt3qUXHh9L6N"
      },
      "source": [
        "# Label Encoding\n",
        "Sex_mapping = { 'male':0, 'female':1 }\n",
        "df_data[ 'Sex' ] = df_data.Sex.map( Sex_mapping )\n"
      ],
      "execution_count": 37,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-xs3H3S4BFqY"
      },
      "source": [
        "# 訓練集與測試集\n",
        "先確認是否每個欄位都皆為數值狀態，並且將資料分割回訓練及測試"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 342
        },
        "id": "eK-04BQeBYa2",
        "outputId": "4870ac66-5fde-48b7-97a2-eb9050057805"
      },
      "source": [
        "df_data.head()"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>PassengerId</th>\n",
              "      <th>Survived</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>Cabin</th>\n",
              "      <th>Title</th>\n",
              "      <th>Ticket_info</th>\n",
              "      <th>is_Age_17</th>\n",
              "      <th>LogFare</th>\n",
              "      <th>Fare_Group</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>1</td>\n",
              "      <td>0.0</td>\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>NoCabin</td>\n",
              "      <td>Mr</td>\n",
              "      <td>A5</td>\n",
              "      <td>0</td>\n",
              "      <td>2.110213</td>\n",
              "      <td>(0.0, 7.896]</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\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>C</td>\n",
              "      <td>Mrs</td>\n",
              "      <td>PC</td>\n",
              "      <td>0</td>\n",
              "      <td>4.280593</td>\n",
              "      <td>(66.344, 512.329]</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>1.0</td>\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>NoCabin</td>\n",
              "      <td>Miss</td>\n",
              "      <td>STONO2</td>\n",
              "      <td>0</td>\n",
              "      <td>2.188856</td>\n",
              "      <td>(7.896, 14.454]</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>1.0</td>\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>C</td>\n",
              "      <td>Mrs</td>\n",
              "      <td>X</td>\n",
              "      <td>0</td>\n",
              "      <td>3.990834</td>\n",
              "      <td>(31.275, 66.344]</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>0.0</td>\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>NoCabin</td>\n",
              "      <td>Mr</td>\n",
              "      <td>X</td>\n",
              "      <td>0</td>\n",
              "      <td>2.202765</td>\n",
              "      <td>(7.896, 14.454]</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   PassengerId  Survived  Pclass  ...  Embarked_C  Embarked_Q  Embarked_S\n",
              "0            1       0.0       3  ...           0           0           1\n",
              "1            2       1.0       1  ...           1           0           0\n",
              "2            3       1.0       3  ...           0           0           1\n",
              "3            4       1.0       1  ...           0           0           1\n",
              "4            5       0.0       3  ...           0           0           1\n",
              "\n",
              "[5 rows x 17 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7EITZ_ujCRyT"
      },
      "source": [
        "## 類別型態轉為整數\n",
        "檢視後發現，因為我們之前有處理過字串欄位，所以資料中還存有 4 欄類別型態的欄位：Cabin、Embarked、Title、Ticket_info，因此，我們也需要將它們轉換成數值型態欄位："
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 342
        },
        "id": "U2rZ0LVMCZ3K",
        "outputId": "37a0fc2b-38ca-47e2-bf3a-64769b2df33f"
      },
      "source": [
        "for col in ['Title','Ticket_info','Cabin']:\n",
        "  df_data[col]= df_data[col].astype('category').cat.codes\n",
        "\n",
        "df_data.head()"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>PassengerId</th>\n",
              "      <th>Survived</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>Cabin</th>\n",
              "      <th>Title</th>\n",
              "      <th>Ticket_info</th>\n",
              "      <th>is_Age_17</th>\n",
              "      <th>LogFare</th>\n",
              "      <th>Fare_Group</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>1</td>\n",
              "      <td>0.0</td>\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>7</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>2.110213</td>\n",
              "      <td>(0.0, 7.896]</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\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>2</td>\n",
              "      <td>3</td>\n",
              "      <td>14</td>\n",
              "      <td>0</td>\n",
              "      <td>4.280593</td>\n",
              "      <td>(66.344, 512.329]</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>1.0</td>\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>7</td>\n",
              "      <td>1</td>\n",
              "      <td>31</td>\n",
              "      <td>0</td>\n",
              "      <td>2.188856</td>\n",
              "      <td>(7.896, 14.454]</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>1.0</td>\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>2</td>\n",
              "      <td>3</td>\n",
              "      <td>36</td>\n",
              "      <td>0</td>\n",
              "      <td>3.990834</td>\n",
              "      <td>(31.275, 66.344]</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>0.0</td>\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>7</td>\n",
              "      <td>2</td>\n",
              "      <td>36</td>\n",
              "      <td>0</td>\n",
              "      <td>2.202765</td>\n",
              "      <td>(7.896, 14.454]</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   PassengerId  Survived  Pclass  ...  Embarked_C  Embarked_Q  Embarked_S\n",
              "0            1       0.0       3  ...           0           0           1\n",
              "1            2       1.0       1  ...           1           0           0\n",
              "2            3       1.0       3  ...           0           0           1\n",
              "3            4       1.0       1  ...           0           0           1\n",
              "4            5       0.0       3  ...           0           0           1\n",
              "\n",
              "[5 rows x 17 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9aDYViFxHoTR"
      },
      "source": [
        "## 訓練集&測試集\n",
        "最後，我們將整體資料分割回訓練集與測試集兩部份，訓練集用來餵入模型進行訓練"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KXPas4x4KMiA"
      },
      "source": [
        "#刪除毫無關聯的欄位\n",
        "df_data.drop(['PassengerId','Fare','Fare_Group'],inplace=True,axis=1)"
      ],
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qmEB1gUHH2fb",
        "outputId": "61fe7691-3fb8-4f40-d867-a3d9fcbb4a5d"
      },
      "source": [
        "# 產生訓練集和測試集\n",
        "Train = df_data[pd.notnull(df_data.Survived)]\n",
        "Test = df_data[pd.isnull(df_data.Survived)]\n",
        "Test.drop(['Survived'],axis=1,inplace=True)\n",
        "\n",
        "#將測試的Survived單獨拆出\n",
        "Y_Train=Train.Survived\n",
        "X_Train=Train.drop(['Survived'],axis=1)\n",
        "\n",
        "print( f'Shape of X_Train = {X_Train.shape}' )\n",
        "print( f'Shape of Y_Train = {Y_Train.shape}' )\n",
        "print( f'Shape of Test = {Test.shape}' )"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Shape of X_Train = (891, 13)\n",
            "Shape of Y_Train = (891,)\n",
            "Shape of Test = (418, 13)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DSQ4p2ZgMMn9"
      },
      "source": [
        "# 特徵重要性\n",
        "使用最基本的樹狀模型 - 決策樹(Decision Tree)，來評估特徵欄位對於標籤欄位(Survived)的重要性"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 522
        },
        "id": "MbckcpbsMUXa",
        "outputId": "f4b0216b-af81-4c2c-dcfe-fb1a8982e3e0"
      },
      "source": [
        "#建立決策樹\n",
        "DTC = RandomForestClassifier()\n",
        "DTC.fit(X_Train,Y_Train)\n",
        "#排序特徵重要性\n",
        "col_names=X_Train.columns\n",
        "importances=DTC.feature_importances_\n",
        "Feature_Rank=pd.DataFrame({'Feature_Name':col_names,'Importance':importances})\n",
        "Feature_Rank.sort_values(by='Importance',inplace=True,ascending=False)\n",
        "Feature_Rank"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Feature_Name</th>\n",
              "      <th>Importance</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Age</td>\n",
              "      <td>0.210215</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>LogFare</td>\n",
              "      <td>0.202834</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Sex</td>\n",
              "      <td>0.171163</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>Title</td>\n",
              "      <td>0.107771</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Pclass</td>\n",
              "      <td>0.068162</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>Cabin</td>\n",
              "      <td>0.066783</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>SibSp</td>\n",
              "      <td>0.046110</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>Ticket_info</td>\n",
              "      <td>0.043400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Parch</td>\n",
              "      <td>0.033389</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>Embarked_C</td>\n",
              "      <td>0.016254</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>Embarked_S</td>\n",
              "      <td>0.013895</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>is_Age_17</td>\n",
              "      <td>0.010681</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>Embarked_Q</td>\n",
              "      <td>0.009343</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Feature_Name  Importance\n",
              "2           Age    0.210215\n",
              "9       LogFare    0.202834\n",
              "1           Sex    0.171163\n",
              "6         Title    0.107771\n",
              "0        Pclass    0.068162\n",
              "5         Cabin    0.066783\n",
              "3         SibSp    0.046110\n",
              "7   Ticket_info    0.043400\n",
              "4         Parch    0.033389\n",
              "10   Embarked_C    0.016254\n",
              "12   Embarked_S    0.013895\n",
              "8     is_Age_17    0.010681\n",
              "11   Embarked_Q    0.009343"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 53
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "t0Ywp6HvPAIF"
      },
      "source": [
        "# 建立模型\n",
        "在完成評估特徵重要性後，我們選擇使用隨機森林(Random Forest)來預測資料"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IkQCywWoPEo-"
      },
      "source": [
        "# 隨機森林參數：\n",
        "# n_estimators: 樹的數量(default=10)。\n",
        "# min_samples_leaf: 最終葉節點最少樣本數(default=1)；\n",
        "#                   當樣本不大時，可不設定使用預設，若樣本數量非常大時，則推薦增加此參數值。\n",
        "# min_samples_split:節點再劃分時所需的最小樣本數(default=2)；\n",
        "#                   當樣本不大時，可不設定使用預設，若樣本數量非常大時，則推薦增加此參數值。\n",
        "# oob_score: 是否採用袋外樣本(out-of-bag samples)(未使用的樣本)來評估模型的準確度(default=False)。\n",
        "\n",
        "RFC = RandomForestClassifier( n_estimators = 1000,min_samples_split = 20,min_samples_leaf = 1,oob_score = True,random_state = 1,n_jobs = -1 ) "
      ],
      "execution_count": 54,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lg_j7b5iPtld"
      },
      "source": [
        "# 訓練模型\n",
        "餵入訓練集資料(X_Train,Y_Train)訓練隨機森林(Random Forest)模型。一般而言，為了衡量模型的預測能力，在訓練模型之前，我們還會從訓練集中抽取出一小部份作為驗證集(Validation)，用來評估模型的準確度，而剩下沒被抽取到的資料則用來訓練模型，另外，為避免只參考某一特定的抽樣驗證結果，導致模型產生偏差，我們會藉由交叉驗證(Cross Validation)，來評估模型的準確度是否能達到我們要求的水準。\n",
        "因為隨機森林可藉由前述所提的袋外樣本評估準確度，因此這裏我們便不需進行交叉驗證；然而，我們還想知道特徵欄位的多寡對於模型準確度的影響，所以，先把所有的欄位餵入模型進行訓練：\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jM-UqcZhPyy5",
        "outputId": "d9aa218e-96b6-485c-e71d-33d7e9c7c6f3"
      },
      "source": [
        "# 訓練模型\n",
        "RFC.fit( X_Train, Y_Train )\n",
        "print( 'oob_score = {:.6f}'.format(RFC.oob_score_) )"
      ],
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "oob_score = 0.827160\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xt2NTpayQC-h"
      },
      "source": [
        "將所有的特徵欄位餵入模型進行訓練，訓練完成的模型準確率約 82.7%；接下來，我們試著參考上述特徵重要性的結果，只選擇部份特徵欄位來訓練模型，觀察這樣的模型準確度是否能有所提升？"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pwTD1EwpQG2C",
        "outputId": "1c2695bc-d340-4932-d41a-37ca45cebd21"
      },
      "source": [
        "RFC_2 = RandomForestClassifier( n_estimators = 1000,min_samples_split = 20,min_samples_leaf = 1,oob_score = True,random_state = 1,n_jobs = -1 ) \n",
        "\n",
        "# 篩選部份特徵欄位餵入模型進行訓練\n",
        "SelectedFeatures = ['Age','Sex','LogFare','Title','Pclass','Cabin']\n",
        "RFC_2.fit( X_Train[SelectedFeatures], Y_Train )\n",
        "\n",
        "print( f'Selected Features :\\n   {SelectedFeatures}' )\n",
        "print( 'oob_score = {:.6f}'.format(RFC_2.oob_score_) )"
      ],
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Selected Features :\n",
            "   ['Age', 'Sex', 'LogFare', 'Title', 'Pclass', 'Cabin']\n",
            "oob_score = 0.841751\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hMwnW-XfQ5cj"
      },
      "source": [
        "經觀察，只篩選部份特徵欄位所訓練出的模型準確率約 84.1%，略高於所有特徵欄位所訓練出的模型，從這我們可得到一個寶貴的經驗，就是適當的挑選出關鍵性的特徵欄位進行訓練，對於模型的準確度將會有所提升"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3JhBB9PBQ-uj"
      },
      "source": [
        "# 測試集預測"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 506
        },
        "id": "_YXbE4eWRVlt",
        "outputId": "41a4d0d8-ec28-46e0-e481-22b4ad3db788"
      },
      "source": [
        "# 預測測試集資料\n",
        "Test_pred = RFC_2.predict( Test[SelectedFeatures] )\n",
        "\n",
        "# 檢視預測罹難或生還的機率\n",
        "#print( RFC_2.predict_proba(Test[SelectedFeatures]) )\n",
        "\n",
        "# 提交檔案\n",
        "submit['Survived'] = Test_pred.astype(int)\n",
        "submit.to_csv( 'Titanic_RandomForest.csv', index=False )\n",
        "print( f'預測結果：' )\n",
        "submit "
      ],
      "execution_count": 61,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "預測結果：\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>PassengerId</th>\n",
              "      <th>Survived</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>892</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>893</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>894</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>895</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>896</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>413</th>\n",
              "      <td>1305</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>414</th>\n",
              "      <td>1306</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>415</th>\n",
              "      <td>1307</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>416</th>\n",
              "      <td>1308</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>417</th>\n",
              "      <td>1309</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>418 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     PassengerId  Survived\n",
              "0            892         0\n",
              "1            893         0\n",
              "2            894         0\n",
              "3            895         0\n",
              "4            896         1\n",
              "..           ...       ...\n",
              "413         1305         0\n",
              "414         1306         1\n",
              "415         1307         0\n",
              "416         1308         0\n",
              "417         1309         1\n",
              "\n",
              "[418 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 61
        }
      ]
    }
  ]
}