ML for Titanic survival prediction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 머신러닝을 활용한 타이타닉 생존 예측 프로젝트"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1470,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "zKbTYjWpx03n"
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn import datasets, model_selection, linear_model, ensemble,neighbors\n",
    "from sklearn.metrics import mean_squared_error,accuracy_score\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.ensemble import ExtraTreesClassifier \n",
    "from sklearn.ensemble import RandomForestClassifier \n",
    "from xgboost import plot_importance\n",
    "from xgboost import XGBClassifier\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "\n",
    "## 1. Preparing dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1471,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>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",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  \\\n",
       "0            1         0       3   \n",
       "1            2         1       1   \n",
       "2            3         1       3   \n",
       "\n",
       "                                                Name     Sex   Age  SibSp  \\\n",
       "0                            Braund, Mr. Owen Harris    male  22.0      1   \n",
       "1  Cumings, Mrs. John Bradley (Florence Briggs Th...  female  38.0      1   \n",
       "2                             Heikkinen, Miss. Laina  female  26.0      0   \n",
       "\n",
       "   Parch            Ticket     Fare Cabin Embarked  \n",
       "0      0         A/5 21171   7.2500   NaN        S  \n",
       "1      0          PC 17599  71.2833   C85        C  \n",
       "2      0  STON/O2. 3101282   7.9250   NaN        S  "
      ]
     },
     "execution_count": 1471,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df = pd.read_csv('titanic.csv')\n",
    "data_df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1472,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_df = data_df.drop(['PassengerId','Ticket','Cabin'],axis = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data info\n",
    "\n",
    "- **PassengerId** : Unique ID of passenger\n",
    "- **Survived** : 0 = No, 1 = Yes\n",
    "- **pclass** : Ticket class (1 = 1st, 2 = 2nd, 3 = 3rd)\n",
    "- **sibsp** : # of siblings & spouses aboard the Titanic\n",
    "- **parch** : # of parents / children aboard the Titanic\n",
    "- **ticket** : Ticket number\n",
    "- **cabin** : Cabin number\n",
    "- **embarked** : Port of Embarkation (C = Cherbourg, Q = Queenstown, S = Southampton)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1473,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 127
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 13775,
     "status": "ok",
     "timestamp": 1546723556658,
     "user": {
      "displayName": "Daeyeon Jo",
      "photoUrl": "https://lh4.googleusercontent.com/-9t4wvx9MYls/AAAAAAAAAAI/AAAAAAAAAC0/SPqWlKvQQMk/s64/photo.jpg",
      "userId": "05542487530960824006"
     },
     "user_tz": -540
    },
    "id": "EqR4honW16jn",
    "outputId": "35909889-315d-4a5f-815d-ce1535201ae0",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Survived</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Survived\n",
       "0         0\n",
       "1         1\n",
       "2         1"
      ]
     },
     "execution_count": 1473,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_data = data_df[['Survived']]\n",
    "y_data.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1474,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\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>7.2500</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\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>71.2833</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\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>7.9250</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Pclass                                               Name     Sex   Age  \\\n",
       "0       3                            Braund, Mr. Owen Harris    male  22.0   \n",
       "1       1  Cumings, Mrs. John Bradley (Florence Briggs Th...  female  38.0   \n",
       "2       3                             Heikkinen, Miss. Laina  female  26.0   \n",
       "\n",
       "   SibSp  Parch     Fare Embarked  \n",
       "0      1      0   7.2500        S  \n",
       "1      1      0  71.2833        C  \n",
       "2      0      0   7.9250        S  "
      ]
     },
     "execution_count": 1474,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# del data_df['Survived']\n",
    "x_data = data_df.copy()\n",
    "del x_data['Survived']\n",
    "x_data.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "\n",
    "## 2. Feature engineering & Feature selection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 시도해볼 수 있는 전략들\n",
    "\n",
    "- 불필요한 열이나 예측에 방해가 되는 열은 아예 지우기 (ex. PassengerId)\n",
    "- 결측치 채우기 \n",
    "- Text로 되어있는 Category(Factor)는 숫자로 바꿔주기 (ex. Male/Female -> 0/1)\n",
    "- 실수 범위를 구간 범위로 바꿔주기 \n",
    "- 필요한 경우 기존 열을 바탕으로 새로운 열을 계산해 추가하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1475,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 891 entries, 0 to 890\n",
      "Data columns (total 8 columns):\n",
      "Pclass      891 non-null int64\n",
      "Name        891 non-null object\n",
      "Sex         891 non-null object\n",
      "Age         714 non-null float64\n",
      "SibSp       891 non-null int64\n",
      "Parch       891 non-null int64\n",
      "Fare        891 non-null float64\n",
      "Embarked    889 non-null object\n",
      "dtypes: float64(2), int64(3), object(3)\n",
      "memory usage: 55.8+ KB\n"
     ]
    }
   ],
   "source": [
    "x_data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 요소별 생존율"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 생존율 - Sex\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1476,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sex\n",
       "female    74.203822\n",
       "male      18.890815\n",
       "Name: Survived, dtype: float64"
      ]
     },
     "execution_count": 1476,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df.groupby(['Sex',])['Survived'].mean()*100\n",
    "#groupby ->선택한 column 의 요소 별로 행을 나눔\n",
    "#mean * 100 -> 확률"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1477,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2a593b02518>"
      ]
     },
     "execution_count": 1477,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_df.groupby(['Sex',])['Survived'].mean().plot(kind = \"bar\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 생존율 -Embarked \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1478,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Embarked\n",
       "C    55.357143\n",
       "Q    38.961039\n",
       "S    33.695652\n",
       "Name: Survived, dtype: float64"
      ]
     },
     "execution_count": 1478,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df.groupby(['Embarked',])['Survived'].mean()*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1479,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2a593f29a58>"
      ]
     },
     "execution_count": 1479,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_df.groupby(['Embarked',])['Survived'].mean().plot(kind = \"bar\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 생존율 - Parch \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1480,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Parch\n",
       "0    34.365782\n",
       "1    55.084746\n",
       "2    50.000000\n",
       "3    60.000000\n",
       "4     0.000000\n",
       "5    20.000000\n",
       "6     0.000000\n",
       "Name: Survived, dtype: float64"
      ]
     },
     "execution_count": 1480,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df.groupby(['Parch',])['Survived'].mean()*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1481,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2a593f7bf60>"
      ]
     },
     "execution_count": 1481,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEGCAYAAACevtWaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAEFJJREFUeJzt3X+s3Xddx/Hnay1FfjlNdkXoj3VKgVTAAZcOgkHQoV1mWn8M6YzKlNGQ0AzFEEogU2s0cxgJgf5BhRE0mWVAhIvUVeWH4YeMdjhgbSlcyrA3dVDGwoIb6zre/nFO4Xi57fne9tydez97PpJm5/s9n/u97y63z3337Tnfk6pCktSW88Y9gCRp9Iy7JDXIuEtSg4y7JDXIuEtSg4y7JDXIuEtSg4y7JDXIuEtSg5aP6xtfcMEFtXbt2nF9e0lakm699dZvVdXEsHVji/vatWvZv3//uL69JC1JSb7eZZ2XZSSpQcZdkhpk3CWpQcZdkhpk3CWpQZ3inmRjksNJppNsP82a305yMMmBJDeOdkxJ0nwMfSlkkmXATuDFwAywL8lUVR0cWLMOeD3w/Kq6O8lPLdTAkqThupy5bwCmq+pIVZ0AdgObZ615BbCzqu4GqKpvjnZMSdJ8dHkT00rg6MD2DHDJrDVPBkjyKWAZ8GdVdfPsAyXZCmwFWLNmzdnMK52Vtds/vGDHvuO6yxfs2NLZ6nLmnjn2zf5U7eXAOuCFwJXAO5L8xI98UdWuqpqsqsmJiaHvnpUknaUucZ8BVg9srwKOzbHmg1X1QFV9DThML/aSpDHoEvd9wLokFyVZAWwBpmat+QDwIoAkF9C7THNklINKkrobGveqOglsA/YCh4CbqupAkh1JNvWX7QXuSnIQ+Bjw2qq6a6GGliSdWae7QlbVHmDPrH3XDjwu4DX9X5KkMfMdqpLUIOMuSQ0y7pLUIOMuSQ0y7pLUIOMuSQ0y7pLUIOMuSQ0y7pLUIOMuSQ0y7pLUoE73ltG5W8gPiwA/MELS/+eZuyQ1yLhLUoOMuyQ1yLhLUoOMuyQ1yLhLUoOMuyQ1yLhLUoOMuyQ1yLhLUoOMuyQ1yLhLUoOMuyQ1yLhLUoM6xT3JxiSHk0wn2T7H81clOZ7ktv6vq0c/qiSpq6H3c0+yDNgJvBiYAfYlmaqqg7OWvqeqti3AjJKkeepy5r4BmK6qI1V1AtgNbF7YsSRJ56LLJzGtBI4ObM8Al8yx7reSvAD4MvDHVXV09oIkW4GtAGvWrJn/tBobP0lKWlq6nLlnjn01a/tDwNqqegbw78C75zpQVe2qqsmqmpyYmJjfpJKkzrrEfQZYPbC9Cjg2uKCq7qqq+/ubfwc8ezTjSZLORpe47wPWJbkoyQpgCzA1uCDJEwY2NwGHRjeiJGm+hl5zr6qTSbYBe4FlwA1VdSDJDmB/VU0B1yTZBJwEvg1ctYAzS5KG6PIXqlTVHmDPrH3XDjx+PfD60Y4mSTpbvkNVkhpk3CWpQcZdkhpk3CWpQcZdkhpk3CWpQcZdkhpk3CWpQcZdkhpk3CWpQcZdkhpk3CWpQcZdkhpk3CWpQcZdkhpk3CWpQcZdkhpk3CWpQcZdkhpk3CWpQcZdkhpk3CWpQcZdkhpk3CWpQcZdkhrUKe5JNiY5nGQ6yfYzrLsiSSWZHN2IkqT5Ghr3JMuAncBlwHrgyiTr51j3OOAa4JZRDylJmp8uZ+4bgOmqOlJVJ4DdwOY51v0FcD3wvRHOJ0k6C13ivhI4OrA909/3A0meCayuqn8+04GSbE2yP8n+48ePz3tYSVI3XeKeOfbVD55MzgPeDPzJsANV1a6qmqyqyYmJie5TSpLmpUvcZ4DVA9urgGMD248DngZ8PMkdwHOBKf9SVZLGp0vc9wHrklyUZAWwBZg69WRVfaeqLqiqtVW1FvgMsKmq9i/IxJKkoYbGvapOAtuAvcAh4KaqOpBkR5JNCz2gJGn+lndZVFV7gD2z9l17mrUvPPexJEnnwneoSlKDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNci4S1KDjLskNajTLX8Xg7XbP7ygx7/jussX9PiS9FDyzF2SGmTcJalBxl2SGmTcJalBxl2SGmTcJalBxl2SGmTcJalBxl2SGmTcJalBxl2SGtQp7kk2JjmcZDrJ9jmef2WSLya5Lcknk6wf/aiSpK6Gxj3JMmAncBmwHrhyjnjfWFVPr6qLgeuBvx35pJKkzrqcuW8ApqvqSFWdAHYDmwcXVNU9A5uPAWp0I0qS5qvLLX9XAkcHtmeAS2YvSvIq4DXACuCX5jpQkq3AVoA1a9bMd1ZJUkddztwzx74fOTOvqp1V9bPA64A3znWgqtpVVZNVNTkxMTG/SSVJnXWJ+wywemB7FXDsDOt3A79+LkNJks5Nl7jvA9YluSjJCmALMDW4IMm6gc3Lga+MbkRJ0nwNveZeVSeTbAP2AsuAG6rqQJIdwP6qmgK2JbkUeAC4G3jZQg4tSTqzTp+hWlV7gD2z9l078PjVI55LknQOfIeqJDXIuEtSg4y7JDXIuEtSg4y7JDXIuEtSg4y7JDXIuEtSg4y7JDXIuEtSg4y7JDXIuEtSg4y7JDXIuEtSg4y7JDXIuEtSg4y7JDXIuEtSg4y7JDXIuEtSg4y7JDXIuEtSg4y7JDXIuEtSg4y7JDWoU9yTbExyOMl0ku1zPP+aJAeTfCHJR5JcOPpRJUldDY17kmXATuAyYD1wZZL1s5b9FzBZVc8A3gdcP+pBJUnddTlz3wBMV9WRqjoB7AY2Dy6oqo9V1b39zc8Aq0Y7piRpPpZ3WLMSODqwPQNccob1Lwf+Za4nkmwFtgKsWbOm44iSlrK12z+8oMe/47rLF/T4S1WXM/fMsa/mXJj8LjAJvGmu56tqV1VNVtXkxMRE9yklSfPS5cx9Blg9sL0KODZ7UZJLgTcAv1hV949mPEnS2ehy5r4PWJfkoiQrgC3A1OCCJM8E3g5sqqpvjn5MSdJ8DI17VZ0EtgF7gUPATVV1IMmOJJv6y94EPBZ4b5Lbkkyd5nCSpIdAl8syVNUeYM+sfdcOPL50xHNJks6B71CVpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqkHGXpAYZd0lqUKe4J9mY5HCS6STb53j+BUk+l+RkkitGP6YkaT6Gxj3JMmAncBmwHrgyyfpZy/4buAq4cdQDSpLmb3mHNRuA6ao6ApBkN7AZOHhqQVXd0X/u+wswoyRpnrpcllkJHB3Ynunvm7ckW5PsT7L/+PHjZ3MISVIHXeKeOfbV2XyzqtpVVZNVNTkxMXE2h5AkddAl7jPA6oHtVcCxhRlHkjQKXeK+D1iX5KIkK4AtwNTCjiVJOhdD415VJ4FtwF7gEHBTVR1IsiPJJoAkz0kyA7wEeHuSAws5tCTpzLq8Woaq2gPsmbXv2oHH++hdrpEkLQK+Q1WSGmTcJalBxl2SGmTcJalBxl2SGmTcJalBxl2SGmTcJalBxl2SGmTcJalBxl2SGmTcJalBxl2SGmTcJalBxl2SGmTcJalBxl2SGmTcJalBxl2SGmTcJalBxl2SGmTcJalBxl2SGmTcJalBxl2SGtQp7kk2JjmcZDrJ9jmef2SS9/SfvyXJ2lEPKknqbmjckywDdgKXAeuBK5Osn7Xs5cDdVfUk4M3AX496UElSd13O3DcA01V1pKpOALuBzbPWbAbe3X/8PuCXk2R0Y0qS5iNVdeYFyRXAxqq6ur/9e8AlVbVtYM3t/TUz/e2v9td8a9axtgJb+5tPAQ6P6jcyhwuAbw1dtXg5//gs5dnB+cdtoee/sKomhi1a3uFAc52Bz/4vQpc1VNUuYFeH73nOkuyvqsmH4nstBOcfn6U8Ozj/uC2W+btclpkBVg9srwKOnW5NkuXA+cC3RzGgJGn+usR9H7AuyUVJVgBbgKlZa6aAl/UfXwF8tIZd75EkLZihl2Wq6mSSbcBeYBlwQ1UdSLID2F9VU8A7gX9IMk3vjH3LQg7d0UNy+WcBOf/4LOXZwfnHbVHMP/QvVCVJS4/vUJWkBhl3SWqQcZekBnV5nfuSkOSp9N4pu5Lea+yPAVNVdWisg2nRS7IBqKra17+1xkbgS1W1Z8yjnZUkf19Vvz/uOTReTfyFapLXAVfSuzXCTH/3Knqv2tldVdeNa7aHg/5/WFcCt1TVdwf2b6yqm8c32XBJ/pTefZOWA/8GXAJ8HLgU2FtVfzm+6YZLMvtlyQFeBHwUoKo2PeRDnYMkv0Dvlie3V9W/jnueYZJcAhyqqnuSPArYDjwLOAj8VVV9Z2yzNRL3LwM/V1UPzNq/AjhQVevGM9m5S/IHVfWucc9xOkmuAV4FHAIuBl5dVR/sP/e5qnrWOOcbJskX6c39SOBOYNXAH9RbquoZYx1wiCSfoxeSd9D7P9YA/0j/5chV9R/jm264JJ+tqg39x6+g97P0T8CvAB9a7CdmSQ4AP99/yfgu4F7699fq7//Ncc3WymWZ7wNPBL4+a/8T+s8tZX8OLNq4A68Anl1V3+3f6vl9SdZW1VuY+7YUi83JqnoQuDfJV6vqHoCqui/JUvjZmQReDbwBeG1V3ZbkvsUe9QGPGHi8FXhxVR1P8jfAZ4BFHXfgvKo62X88OXAy88kkt41rKGgn7n8EfCTJV4Cj/X1rgCcB2077VYtEki+c7ing8Q/lLGdh2alLMVV1R5IX0gv8hSyNuJ9I8uiquhd49qmdSc5nCZwYVNX3gTcneW//n99gaf25Pi/JT9J7cUeq6jhAVf1vkpNn/tJF4faB/7v+fJLJqtqf5MnAA8O+eCEtpR+C06qqm/v/MjfQu/Ybetfe9/XPyha7xwO/Ctw9a3+ATz/048zLnUkurqrbAPpn8L8G3AA8fbyjdfKCqroffhDKUx7BD2+psej178j6kiSXA/eMe555OB+4ld7PeiX56aq6M8ljWRonB1cDb0nyRnp3gvzPJEfpnWRePc7BmrjmvtQleSfwrqr65BzP3VhVvzOGsTpJsorepY0753ju+VX1qTGMpSUuyaOBx1fV18Y9SxdJHgf8DL0T5pmq+saYRzLuktQi38QkSQ0y7pLUIOOupiV5MMltSW5P8t7+tdxzPeZVSd42ivmkhWLc1br7quriqnoacAJ4ZdcvTLJs4caSFpZx18PJJ+i994EkH0hya5ID/Q9up7//u0l2JLkFeF6S5yT5dJLPJ/ls/1URAE9McnOSryS5fgy/F+mMmniduzRM/7N9LwNO3evmD6vq2/3bDOxL8v6qugt4DL37mlzbv33Fl4CX9m8q9uPAff2vvxh4JnA/cDjJW6vqKNIiYdzVukcNvA38E/Q+EhLgmiS/0X+8GlgH3AU8CLy/v/8pwP9U1T6AU7cmSALwkVM3hUpyELiQH747Who7467W3VdVFw/u6N8i4VLgeVV1b5KPAz/Wf/p7A+9qDr2bcc3l/oHHD+KfJS0yXnPXw9H5wN39sD8VeO5p1n2J3rX150DvXYj9yzvSoucPqh6ObgZe2b9h22F6dx/8EVV1IslLgbf2r83fR++MX1r0vP2AJDXIyzKS1CDjLkkNMu6S1CDjLkkNMu6S1CDjLkkNMu6S1KD/A3rcXq1T9ZHiAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_df.groupby(['Parch',])['Survived'].mean().plot(kind = \"bar\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 생존율 - Sibsp \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1482,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SibSp\n",
       "0    34.539474\n",
       "1    53.588517\n",
       "2    46.428571\n",
       "3    25.000000\n",
       "4    16.666667\n",
       "5     0.000000\n",
       "8     0.000000\n",
       "Name: Survived, dtype: float64"
      ]
     },
     "execution_count": 1482,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df.groupby(['SibSp',])['Survived'].mean()*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1483,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2a593fefdd8>"
      ]
     },
     "execution_count": 1483,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_df.groupby(['SibSp',])['Survived'].mean().plot(kind = \"bar\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 생존율 - Pclass\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1484,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pclass\n",
       "1    62.962963\n",
       "2    47.282609\n",
       "3    24.236253\n",
       "Name: Survived, dtype: float64"
      ]
     },
     "execution_count": 1484,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df.groupby(['Pclass',])['Survived'].mean()*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1485,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2a59429c198>"
      ]
     },
     "execution_count": 1485,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_df.groupby(['Pclass',])['Survived'].mean().plot(kind = \"bar\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 생존율 - Age\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1486,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "630\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(data_df)):\n",
    "    if  data_df[:]['Age'][i] >= 80:\n",
    "        print(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1487,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_df['Age'] = pd.cut(data_df.Age,bins=[10,20,30,40,50,60,70,80],labels=['10','20','30','40','50','60','70'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1488,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Age\n",
       "10    38.260870\n",
       "20    36.521739\n",
       "30    44.516129\n",
       "40    38.372093\n",
       "50    40.476190\n",
       "60    23.529412\n",
       "70    20.000000\n",
       "Name: Survived, dtype: float64"
      ]
     },
     "execution_count": 1488,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df.groupby(['Age',])['Survived'].mean()*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1489,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2a59431f5f8>"
      ]
     },
     "execution_count": 1489,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_df.groupby(['Age',])['Survived'].mean().plot(kind = \"bar\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 생존율 - Fare\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1490,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    891.000000\n",
       "mean      32.204208\n",
       "std       49.693429\n",
       "min        0.000000\n",
       "25%        7.910400\n",
       "50%       14.454200\n",
       "75%       31.000000\n",
       "max      512.329200\n",
       "Name: Fare, dtype: float64"
      ]
     },
     "execution_count": 1490,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df['Fare'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1491,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_df['Fare'] = pd.cut(data_df.Fare,bins=[0,8,14,31,512],labels=['0','1','2','3'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1492,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Fare\n",
       "0    22.566372\n",
       "1    30.102041\n",
       "2    43.965517\n",
       "3    57.534247\n",
       "Name: Survived, dtype: float64"
      ]
     },
     "execution_count": 1492,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df.groupby(['Fare',])['Survived'].mean()*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1493,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2a59436f550>"
      ]
     },
     "execution_count": 1493,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_df.groupby(['Fare',])['Survived'].mean().plot(kind = \"bar\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 생존율 분석 평\n",
    "\n",
    "#### Sex \n",
    "-> 여성이 압도적으로 높다. \n",
    "\n",
    "#### Embarked \n",
    "-> C(2) > Q(1) > S(0) 순으로 차이를 보인다.\n",
    "\n",
    "#### Parch , Sibsp  \n",
    "->0명일 때 보다 1명일때가 두 케이스 모두 생존율이 20% 올라가지만 2명일 때 부터 감소하는 모습을 보여준다.\n",
    "\n",
    "#### Pclass \n",
    "-> 양의 상관관계를 보여준다.(등급이 높을수록 생존율 up)\n",
    "\n",
    "#### Age\n",
    "-> 40대 이후로 생존율이 감소됨을 알 수 있다.\n",
    "\n",
    "#### Fare\n",
    "-> 양의 상관관계를 보여준다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 데이터전처리\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Parch , SibSp -> 가족 유무로 변경"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1494,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_data['member'] = x_data['Parch'] + x_data['SibSp']\n",
    "x_data = x_data.drop(['Parch','SibSp'],axis = 1)\n",
    "# 가족 유무 별 열추가\n",
    "# Parch SibSp 열 삭제"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1495,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(len(x_data)):\n",
    "    if x_data['member'][i] > 1 :\n",
    "        x_data['member'][i] = 1\n",
    "#값이 1 이상이면 1로 변경"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 성별 숫자로 변경"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1496,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_data['Sex']=LabelEncoder().fit(['female','male']).transform(x_data['Sex'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1497,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>member</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3</td>\n",
       "      <td>Moran, Mr. James</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.4583</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1</td>\n",
       "      <td>McCarthy, Mr. Timothy J</td>\n",
       "      <td>1</td>\n",
       "      <td>54.0</td>\n",
       "      <td>51.8625</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>3</td>\n",
       "      <td>Palsson, Master. Gosta Leonard</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>21.0750</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>3</td>\n",
       "      <td>Johnson, Mrs. Oscar W (Elisabeth Vilhelmina Berg)</td>\n",
       "      <td>0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>11.1333</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2</td>\n",
       "      <td>Nasser, Mrs. Nicholas (Adele Achem)</td>\n",
       "      <td>0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>30.0708</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Pclass                                               Name  Sex   Age  \\\n",
       "0       3                            Braund, Mr. Owen Harris    1  22.0   \n",
       "1       1  Cumings, Mrs. John Bradley (Florence Briggs Th...    0  38.0   \n",
       "2       3                             Heikkinen, Miss. Laina    0  26.0   \n",
       "3       1       Futrelle, Mrs. Jacques Heath (Lily May Peel)    0  35.0   \n",
       "4       3                           Allen, Mr. William Henry    1  35.0   \n",
       "5       3                                   Moran, Mr. James    1   NaN   \n",
       "6       1                            McCarthy, Mr. Timothy J    1  54.0   \n",
       "7       3                     Palsson, Master. Gosta Leonard    1   2.0   \n",
       "8       3  Johnson, Mrs. Oscar W (Elisabeth Vilhelmina Berg)    0  27.0   \n",
       "9       2                Nasser, Mrs. Nicholas (Adele Achem)    0  14.0   \n",
       "\n",
       "      Fare Embarked  member  \n",
       "0   7.2500        S       1  \n",
       "1  71.2833        C       1  \n",
       "2   7.9250        S       0  \n",
       "3  53.1000        S       1  \n",
       "4   8.0500        S       0  \n",
       "5   8.4583        Q       0  \n",
       "6  51.8625        S       0  \n",
       "7  21.0750        S       1  \n",
       "8  11.1333        S       1  \n",
       "9  30.0708        C       1  "
      ]
     },
     "execution_count": 1497,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_data.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Embarked 숫자로 변경, 결측치 채움"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1498,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_data['Embarked'] = x_data['Embarked'].fillna('S')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1499,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>member</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Pclass                                               Name  Sex   Age  \\\n",
       "0       3                            Braund, Mr. Owen Harris    1  22.0   \n",
       "1       1  Cumings, Mrs. John Bradley (Florence Briggs Th...    0  38.0   \n",
       "2       3                             Heikkinen, Miss. Laina    0  26.0   \n",
       "3       1       Futrelle, Mrs. Jacques Heath (Lily May Peel)    0  35.0   \n",
       "4       3                           Allen, Mr. William Henry    1  35.0   \n",
       "\n",
       "      Fare Embarked  member  \n",
       "0   7.2500        S       1  \n",
       "1  71.2833        C       1  \n",
       "2   7.9250        S       0  \n",
       "3  53.1000        S       1  \n",
       "4   8.0500        S       0  "
      ]
     },
     "execution_count": 1499,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1500,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_data['Embarked']=LabelEncoder().fit(['S','Q','C']).transform(x_data['Embarked'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Age 결측치 채움"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1501,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Mr          517\n",
       "Miss        182\n",
       "Mrs         125\n",
       "Master       40\n",
       "Dr            7\n",
       "Rev           6\n",
       "Col           2\n",
       "Mlle          2\n",
       "Major         2\n",
       "Countess      1\n",
       "Capt          1\n",
       "Mme           1\n",
       "Jonkheer      1\n",
       "Sir           1\n",
       "Don           1\n",
       "Lady          1\n",
       "Ms            1\n",
       "Name: Title, dtype: int64"
      ]
     },
     "execution_count": 1501,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_data_test = [x_data] # combining train and test dataset\n",
    "for dataset in x_data_test:\n",
    "    dataset['Title'] = dataset['Name'].str.extract(' ([A-Za-z]+)\\.', expand=False)\n",
    "x_data['Title'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1502,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>member</th>\n",
       "      <th>Title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Pclass                                               Name  Sex   Age  \\\n",
       "0       3                            Braund, Mr. Owen Harris    1  22.0   \n",
       "1       1  Cumings, Mrs. John Bradley (Florence Briggs Th...    0  38.0   \n",
       "2       3                             Heikkinen, Miss. Laina    0  26.0   \n",
       "3       1       Futrelle, Mrs. Jacques Heath (Lily May Peel)    0  35.0   \n",
       "4       3                           Allen, Mr. William Henry    1  35.0   \n",
       "\n",
       "      Fare  Embarked  member  Title  \n",
       "0   7.2500         2       1      0  \n",
       "1  71.2833         0       1      2  \n",
       "2   7.9250         2       0      1  \n",
       "3  53.1000         2       1      2  \n",
       "4   8.0500         2       0      0  "
      ]
     },
     "execution_count": 1502,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "name_mapping = {\"Mr\": 0, \"Miss\": 1, \"Mrs\": 2,\n",
    "                 \"Master\": 3, \"Dr\": 4, \"Rev\": 4, \"Col\": 4, \"Major\": 4, \"Mlle\": 4,\"Countess\": 4,\n",
    "                 \"Ms\": 4, \"Lady\": 4, \"Jonkheer\": 4, \"Don\": 4, \"Dona\" : 4, \"Mme\": 4,\"Capt\": 4,\"Sir\": 4 }\n",
    "for dataset in x_data_test :\n",
    "    dataset['Title'] = dataset['Title'].map(name_mapping)\n",
    "x_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1503,
   "metadata": {},
   "outputs": [],
   "source": [
    "# x_data_test = [x_data] # combining train and test dataset\n",
    "# for dataset in x_data_test:\n",
    "#     dataset['Title'] = dataset['Name'].str.extract(' ([A-Za-z]+)\\.', expand=False)\n",
    "# x_data['Title'].value_counts()\n",
    "\n",
    "# name_mapping = {\"Mr\": 0, \"Miss\": 1, \"Mrs\": 2,\n",
    "#                  \"Master\": 3, \"Dr\": 4, \"Rev\": 4, \"Col\": 4, \"Major\": 4, \"Mlle\": 4,\"Countess\": 4,\n",
    "#                  \"Ms\": 4, \"Lady\": 4, \"Jonkheer\": 4, \"Don\": 4, \"Dona\" : 4, \"Mme\": 4,\"Capt\": 4,\"Sir\": 4 }\n",
    "# for dataset in x_data_test :\n",
    "#     dataset['Title'] = dataset['Title'].map(name_mapping)\n",
    "# x_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1504,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 891 entries, 0 to 890\n",
      "Data columns (total 8 columns):\n",
      "Pclass      891 non-null int64\n",
      "Name        891 non-null object\n",
      "Sex         891 non-null int32\n",
      "Age         714 non-null float64\n",
      "Fare        891 non-null float64\n",
      "Embarked    891 non-null int32\n",
      "member      891 non-null int64\n",
      "Title       891 non-null int64\n",
      "dtypes: float64(2), int32(2), int64(3), object(1)\n",
      "memory usage: 48.8+ KB\n"
     ]
    }
   ],
   "source": [
    "x_data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1505,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>member</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Title</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2.410058</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>32.368090</td>\n",
       "      <td>24.441560</td>\n",
       "      <td>1.603482</td>\n",
       "      <td>0.232108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.307692</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>21.773973</td>\n",
       "      <td>43.797873</td>\n",
       "      <td>1.401099</td>\n",
       "      <td>0.450549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>35.898148</td>\n",
       "      <td>45.138533</td>\n",
       "      <td>1.480000</td>\n",
       "      <td>0.840000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2.625000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.574167</td>\n",
       "      <td>34.703125</td>\n",
       "      <td>1.650000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.333333</td>\n",
       "      <td>0.740741</td>\n",
       "      <td>42.384615</td>\n",
       "      <td>39.111422</td>\n",
       "      <td>1.259259</td>\n",
       "      <td>0.259259</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Pclass       Sex        Age       Fare  Embarked    member\n",
       "Title                                                              \n",
       "0      2.410058  1.000000  32.368090  24.441560  1.603482  0.232108\n",
       "1      2.307692  0.000000  21.773973  43.797873  1.401099  0.450549\n",
       "2      2.000000  0.000000  35.898148  45.138533  1.480000  0.840000\n",
       "3      2.625000  1.000000   4.574167  34.703125  1.650000  1.000000\n",
       "4      1.333333  0.740741  42.384615  39.111422  1.259259  0.259259"
      ]
     },
     "execution_count": 1505,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_data.groupby(['Title']).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1506,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_data.loc[(x_data.Age.isnull())&(x_data.Title==0),'Age']=32\n",
    "x_data.loc[(x_data.Age.isnull())&(x_data.Title==1),'Age']=22\n",
    "x_data.loc[(x_data.Age.isnull())&(x_data.Title==2),'Age']=36\n",
    "x_data.loc[(x_data.Age.isnull())&(x_data.Title==3),'Age']=5\n",
    "x_data.loc[(x_data.Age.isnull())&(x_data.Title==4),'Age']=42"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1507,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "x_data = x_data.drop(['Name'],axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1508,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>member</th>\n",
       "      <th>Title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>32.0</td>\n",
       "      <td>8.4583</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>54.0</td>\n",
       "      <td>51.8625</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>21.0750</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>11.1333</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>30.0708</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Pclass  Sex   Age     Fare  Embarked  member  Title\n",
       "0       3    1  22.0   7.2500         2       1      0\n",
       "1       1    0  38.0  71.2833         0       1      2\n",
       "2       3    0  26.0   7.9250         2       0      1\n",
       "3       1    0  35.0  53.1000         2       1      2\n",
       "4       3    1  35.0   8.0500         2       0      0\n",
       "5       3    1  32.0   8.4583         1       0      0\n",
       "6       1    1  54.0  51.8625         2       0      0\n",
       "7       3    1   2.0  21.0750         2       1      3\n",
       "8       3    0  27.0  11.1333         2       1      2\n",
       "9       2    0  14.0  30.0708         0       1      2"
      ]
     },
     "execution_count": 1508,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_data.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1509,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.preprocessing import RobustScaler\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "ct = ColumnTransformer([\n",
    "        ('age', StandardScaler(), ['Age']),\n",
    "        ('fare', RobustScaler(), ['Fare'])\n",
    "    ], remainder='passthrough')\n",
    "#age,fare 값 스캐일러 적용\n",
    "#RobustScaler-> outlier 최적화, 25% 와 75% 사이의 값을 다룬다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1510,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_data = ct.fit_transform(x_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1511,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.582415</td>\n",
       "      <td>-0.312011</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.624899</td>\n",
       "      <td>2.461242</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.280586</td>\n",
       "      <td>-0.282777</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.398528</td>\n",
       "      <td>1.673732</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.398528</td>\n",
       "      <td>-0.277363</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          0         1    2    3    4    5    6\n",
       "0 -0.582415 -0.312011  3.0  1.0  2.0  1.0  0.0\n",
       "1  0.624899  2.461242  1.0  0.0  0.0  1.0  2.0\n",
       "2 -0.280586 -0.282777  3.0  0.0  2.0  0.0  1.0\n",
       "3  0.398528  1.673732  1.0  0.0  2.0  1.0  2.0\n",
       "4  0.398528 -0.277363  3.0  1.0  2.0  0.0  0.0"
      ]
     },
     "execution_count": 1511,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(x_data).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "\n",
    "# 2. Train - Test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1512,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train, x_test, y_train, y_test = model_selection.train_test_split(x_data, y_data, test_size=0.3, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1513,
   "metadata": {},
   "outputs": [],
   "source": [
    "params = {'n_estimators': 1000,\n",
    "          'max_depth': 5,\n",
    "          'min_samples_split': 5,\n",
    "          'learning_rate': 0.01}\n",
    "#하이퍼 파라미터"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1514,
   "metadata": {},
   "outputs": [],
   "source": [
    "svc_params = {'C': 1, 'gamma': 0.1, 'kernel': 'rbf'}\n",
    "\n",
    "#SVM 하이퍼파라미터"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1515,
   "metadata": {},
   "outputs": [],
   "source": [
    "from vecstack import StackingTransformer\n",
    "#Stacking (ExtraTreesClassifier, RandomForestClassifier, XGBClassifier)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "\n",
    "# 3. Create model instance variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1516,
   "metadata": {},
   "outputs": [],
   "source": [
    "logisticModel = linear_model.LogisticRegression() # 로지스틱회귀"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1517,
   "metadata": {},
   "outputs": [],
   "source": [
    "KnnModel = neighbors.KNeighborsClassifier() # KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1518,
   "metadata": {},
   "outputs": [],
   "source": [
    "GBModel = ensemble.GradientBoostingClassifier(**params) # GB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1519,
   "metadata": {},
   "outputs": [],
   "source": [
    "XGBModel = XGBClassifier(**params) # XGB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1520,
   "metadata": {},
   "outputs": [],
   "source": [
    "RanModel = RandomForestClassifier(random_state = 0, n_jobs = -1, n_estimators = 100, max_depth = 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1521,
   "metadata": {},
   "outputs": [],
   "source": [
    "ETModel =ExtraTreesClassifier(random_state = 0, n_jobs = -1, n_estimators = 100, max_depth = 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1405,
   "metadata": {},
   "outputs": [],
   "source": [
    "svc = SVC(**svc_params) #SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1406,
   "metadata": {},
   "outputs": [],
   "source": [
    "estimators = [ \n",
    "    ('ExtraTrees', ExtraTreesClassifier(random_state = 0, n_jobs = -1, n_estimators = 100, max_depth = 3)),\n",
    "    ('RandomForest', RandomForestClassifier(random_state = 0, n_jobs = -1, n_estimators = 100, max_depth = 3)),\n",
    "    ('XGB', XGBClassifier(seed = 0, n_jobs = -1, learning_rate = 0.01, n_estimators = 1000, max_depth = 5))]\n",
    "#Stacking (ExtraTreesClassifier, RandomForestClassifier, XGBClassifier)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1407,
   "metadata": {},
   "outputs": [],
   "source": [
    "stack = StackingTransformer(estimators, \n",
    "                            regression = False, \n",
    "                            metric = accuracy_score, \n",
    "                            n_folds = 4, stratified = True, shuffle = True, \n",
    "                            random_state = 0, verbose = 2) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "\n",
    "# 4. Train the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1408,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "logisticModel 검증 점수 :  [0.784      0.768      0.816      0.84677419 0.78225806]\n",
      "logisticModel 검증 평균 점수 : 0.7994064516129031 \n",
      "\n",
      "KnnModel 검증 점수 :  [0.76       0.784      0.808      0.79032258 0.81451613]\n",
      "KnnModel 검증 평균 점수 :  0.7913677419354839 \n",
      "\n",
      "GBModel 검증 점수 :  [0.84       0.744      0.808      0.80645161 0.79032258]\n",
      "GBModel 검증 평균 점수 : 0.7977548387096774 \n",
      "\n",
      "XGBModel교차 검증 점수 :  [0.808      0.8        0.784      0.80645161 0.7983871 ]\n",
      "XGBModel교차 검증 평균 점수 : 0.799367741935484 \n",
      "\n",
      "RanModel 검증 점수 :  [0.784      0.792      0.792      0.84677419 0.81451613]\n",
      "RanModel 검증 평균 점수 : 0.8058580645161291 \n",
      "\n",
      "svc_score 검증 점수 :  [0.776      0.816      0.808      0.85483871 0.80645161]\n",
      "svc_score 검증 평균 점수 : 0.812258064516129 \n",
      "\n",
      "ETModel_score 검증 점수 :  [0.792      0.784      0.808      0.83870968 0.7983871 ]\n",
      "ETModel_score 검증 평균 점수 : 0.8042193548387097 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.model_selection import KFold\n",
    "\n",
    "kfold = KFold(n_splits = 5,shuffle = True,random_state=0)\n",
    "\n",
    "XGBModel_score = cross_val_score(XGBModel, x_train, y_train, cv=kfold)\n",
    "logisticModel_score = cross_val_score(logisticModel, x_train, y_train, cv=kfold)\n",
    "KnnModel_score = cross_val_score(KnnModel, x_train, y_train, cv=kfold)\n",
    "GBModel_score = cross_val_score(GBModel, x_train, y_train, cv =kfold)\n",
    "RanModel_score = cross_val_score(RanModel, x_train, y_train, cv = kfold)\n",
    "svc_score = cross_val_score(svc, x_train, y_train, cv = kfold)\n",
    "ETModel_score = cross_val_score(ETModel, x_train, y_train, cv = kfold)\n",
    "\n",
    "print('logisticModel 검증 점수 : ', logisticModel_score)\n",
    "print(\"logisticModel 검증 평균 점수 :\",logisticModel_score.mean(),\"\\n\")\n",
    "\n",
    "print('KnnModel 검증 점수 : ', KnnModel_score)\n",
    "print(\"KnnModel 검증 평균 점수 : \",KnnModel_score.mean(),\"\\n\")\n",
    "\n",
    "print('GBModel 검증 점수 : ', GBModel_score)\n",
    "print(\"GBModel 검증 평균 점수 :\",GBModel_score.mean(),\"\\n\")\n",
    "\n",
    "print('XGBModel교차 검증 점수 : ', XGBModel_score)\n",
    "print(\"XGBModel교차 검증 평균 점수 :\",XGBModel_score.mean(),\"\\n\")\n",
    "\n",
    "print('RanModel 검증 점수 : ', RanModel_score)\n",
    "print(\"RanModel 검증 평균 점수 :\",RanModel_score.mean(),\"\\n\")\n",
    "\n",
    "print('svc_score 검증 점수 : ', svc_score)\n",
    "print(\"svc_score 검증 평균 점수 :\",svc_score.mean(),\"\\n\")\n",
    "\n",
    "print('ETModel_score 검증 점수 : ', ETModel_score)\n",
    "print(\"ETModel_score 검증 평균 점수 :\",ETModel_score.mean(),\"\\n\")\n",
    "\n",
    "#모델 별 교차검증\n",
    "#Fold를 5개로 나누어서 검증.\n",
    "#확률이 일관되는 걸 확인함으로써 오버피팅을 확인할 수 있음.\n",
    "# n_splits 는 데이터 분할 수\n",
    "# shuffle은  매번 데이터를 분할하기전 섞을지 말지 여부"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1409,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "          intercept_scaling=1, max_iter=100, multi_class='warn',\n",
       "          n_jobs=None, penalty='l2', random_state=None, solver='warn',\n",
       "          tol=0.0001, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 1409,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logisticModel.fit(x_train, y_train) #로지스틱 train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1410,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "           metric_params=None, n_jobs=None, n_neighbors=5, p=2,\n",
       "           weights='uniform')"
      ]
     },
     "execution_count": 1410,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "KnnModel.fit(x_train, y_train) #KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1411,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingClassifier(criterion='friedman_mse', init=None,\n",
       "              learning_rate=0.01, loss='deviance', max_depth=5,\n",
       "              max_features=None, max_leaf_nodes=None,\n",
       "              min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "              min_samples_leaf=1, min_samples_split=5,\n",
       "              min_weight_fraction_leaf=0.0, n_estimators=1000,\n",
       "              n_iter_no_change=None, presort='auto', random_state=None,\n",
       "              subsample=1.0, tol=0.0001, validation_fraction=0.1,\n",
       "              verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 1411,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "GBModel.fit(x_train, y_train)#GB train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1412,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "       colsample_bytree=1, gamma=0, learning_rate=0.01, max_delta_step=0,\n",
       "       max_depth=5, min_child_weight=1, min_samples_split=5, missing=None,\n",
       "       n_estimators=1000, n_jobs=1, nthread=None,\n",
       "       objective='binary:logistic', random_state=0, reg_alpha=0,\n",
       "       reg_lambda=1, scale_pos_weight=1, seed=None, silent=True,\n",
       "       subsample=1)"
      ]
     },
     "execution_count": 1412,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "XGBModel.fit(x_train, y_train)#XGB train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1413,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "            max_depth=3, max_features='auto', max_leaf_nodes=None,\n",
       "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "            min_samples_leaf=1, min_samples_split=2,\n",
       "            min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=-1,\n",
       "            oob_score=False, random_state=0, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 1413,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "RanModel.fit(x_train, y_train)#Random Forest train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1414,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ExtraTreesClassifier(bootstrap=False, class_weight=None, criterion='gini',\n",
       "           max_depth=3, max_features='auto', max_leaf_nodes=None,\n",
       "           min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "           min_samples_leaf=1, min_samples_split=2,\n",
       "           min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=-1,\n",
       "           oob_score=False, random_state=0, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 1414,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ETModel.fit(x_train, y_train)#Extra trees train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1415,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=1, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma=0.1, kernel='rbf',\n",
       "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
       "  tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 1415,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc.fit(x_train, y_train) #SVM train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1416,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "task:         [classification]\n",
      "n_classes:    [2]\n",
      "metric:       [accuracy_score]\n",
      "variant:      [A]\n",
      "n_estimators: [3]\n",
      "\n",
      "estimator  0: [ExtraTrees: ExtraTreesClassifier]\n",
      "    fold  0:  [0.78980892]\n",
      "    fold  1:  [0.79487179]\n",
      "    fold  2:  [0.81290323]\n",
      "    fold  3:  [0.80645161]\n",
      "    ----\n",
      "    MEAN:     [0.80100889] + [0.00914059]\n",
      "\n",
      "estimator  1: [RandomForest: RandomForestClassifier]\n",
      "    fold  0:  [0.79617834]\n",
      "    fold  1:  [0.80769231]\n",
      "    fold  2:  [0.83225806]\n",
      "    fold  3:  [0.81290323]\n",
      "    ----\n",
      "    MEAN:     [0.81225799] + [0.01303668]\n",
      "\n",
      "estimator  2: [XGB: XGBClassifier]\n",
      "    fold  0:  [0.82165605]\n",
      "    fold  1:  [0.79487179]\n",
      "    fold  2:  [0.80000000]\n",
      "    fold  3:  [0.80645161]\n",
      "    ----\n",
      "    MEAN:     [0.80574486] + [0.01006097]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "stack = stack.fit(x_train, y_train) #Stacking"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1417,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train set was detected.\n",
      "Transforming...\n",
      "\n",
      "estimator  0: [ExtraTrees: ExtraTreesClassifier]\n",
      "    model from fold  0: done\n",
      "    model from fold  1: done\n",
      "    model from fold  2: done\n",
      "    model from fold  3: done\n",
      "    ----\n",
      "    DONE\n",
      "\n",
      "estimator  1: [RandomForest: RandomForestClassifier]\n",
      "    model from fold  0: done\n",
      "    model from fold  1: done\n",
      "    model from fold  2: done\n",
      "    model from fold  3: done\n",
      "    ----\n",
      "    DONE\n",
      "\n",
      "estimator  2: [XGB: XGBClassifier]\n",
      "    model from fold  0: done\n",
      "    model from fold  1: done\n",
      "    model from fold  2: done\n",
      "    model from fold  3: done\n",
      "    ----\n",
      "    DONE\n",
      "\n",
      "Transforming...\n",
      "\n",
      "estimator  0: [ExtraTrees: ExtraTreesClassifier]\n",
      "    model from fold  0: done\n",
      "    model from fold  1: done\n",
      "    model from fold  2: done\n",
      "    model from fold  3: done\n",
      "    ----\n",
      "    DONE\n",
      "\n",
      "estimator  1: [RandomForest: RandomForestClassifier]\n",
      "    model from fold  0: done\n",
      "    model from fold  1: done\n",
      "    model from fold  2: done\n",
      "    model from fold  3: done\n",
      "    ----\n",
      "    DONE\n",
      "\n",
      "estimator  2: [XGB: XGBClassifier]\n",
      "    model from fold  0: done\n",
      "    model from fold  1: done\n",
      "    model from fold  2: done\n",
      "    model from fold  3: done\n",
      "    ----\n",
      "    DONE\n",
      "\n"
     ]
    }
   ],
   "source": [
    "S_train = stack.transform(x_train)\n",
    "S_test = stack.transform(x_test) #Stacking"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1418,
   "metadata": {},
   "outputs": [],
   "source": [
    "StackModel = XGBClassifier(seed = 0, n_jobs = -1, learning_rate = 0.01, n_estimators = 1000, max_depth = 5) \n",
    "StackModel = StackModel.fit(S_train, y_train) \n",
    "y_pred = StackModel.predict(S_test)  #Stacking"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "\n",
    "# 5. Predict on test data & Check the result with metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1448,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  78.36\n"
     ]
    }
   ],
   "source": [
    "# 로지스틱 회귀\n",
    "logisticModel_acc = round(accuracy_score(logisticModel.predict(x_test), y_test) * 100, 2)\n",
    "print('Accuracy: ',logisticModel_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1450,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  79.48\n"
     ]
    }
   ],
   "source": [
    "# KNN\n",
    "KnnModel_acc = round(accuracy_score(KnnModel.predict(x_test), y_test) * 100, 2)\n",
    "print('Accuracy: ',KnnModel_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1449,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  85.07\n"
     ]
    }
   ],
   "source": [
    "#Gradient Boosting \n",
    "GBModel_acc =round(accuracy_score(GBModel.predict(x_test), y_test) * 100, 2)\n",
    "print('Accuracy: ',GBModel_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1453,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  84.33\n"
     ]
    }
   ],
   "source": [
    "#Extreme Gradient Boosting\n",
    "XGBModel_acc = round(accuracy_score(XGBModel.predict(x_test), y_test) * 100, 2)\n",
    "print('Accuracy: ',XGBModel_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1455,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  79.1\n"
     ]
    }
   ],
   "source": [
    "#Extra trees\n",
    "ETModel_acc = round(accuracy_score(ETModel.predict(x_test), y_test) * 100, 2)\n",
    "print('Accuracy: ',ETModel_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1457,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  81.34\n"
     ]
    }
   ],
   "source": [
    "#Random Forest\n",
    "RanModel_acc = round(accuracy_score(RanModel.predict(x_test), y_test) * 100, 2)\n",
    "print('Accuracy: ',RanModel_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1459,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  81.72\n"
     ]
    }
   ],
   "source": [
    "#Support Vector Machine\n",
    "svc_acc = round(accuracy_score(svc.predict(x_test), y_test) * 100, 2)\n",
    "print('Accuracy: ',svc_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1462,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  82.84\n"
     ]
    }
   ],
   "source": [
    "#Stacking\n",
    "Stack_acc = round(accuracy_score(y_test, y_pred) * 100, 2)\n",
    "print('Accuracy: ',Stack_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1467,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Model</th>\n",
       "      <th>Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Gradient Boosting</td>\n",
       "      <td>85.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>XGB</td>\n",
       "      <td>84.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Stacking</td>\n",
       "      <td>82.84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>SVC</td>\n",
       "      <td>81.72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Random forest</td>\n",
       "      <td>81.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>KNN</td>\n",
       "      <td>79.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Extra tree</td>\n",
       "      <td>79.10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Logistic</td>\n",
       "      <td>78.36</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Model  Score\n",
       "2  Gradient Boosting  85.07\n",
       "3                XGB  84.33\n",
       "7           Stacking  82.84\n",
       "6                SVC  81.72\n",
       "5      Random forest  81.34\n",
       "1                KNN  79.48\n",
       "4         Extra tree  79.10\n",
       "0           Logistic  78.36"
      ]
     },
     "execution_count": 1467,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "models = pd.DataFrame({\n",
    "    'Model': ['Logistic', 'KNN', 'Gradient Boosting', \n",
    "              'XGB', 'Extra tree', 'Random forest', \n",
    "              'SVC', 'Stacking', \n",
    "              ],\n",
    "    'Score': [\n",
    "logisticModel_acc,\n",
    "KnnModel_acc,\n",
    "GBModel_acc,\n",
    "XGBModel_acc,\n",
    "ETModel_acc,\n",
    "RanModel_acc,\n",
    "svc_acc,\n",
    "Stack_acc]})\n",
    "models.sort_values(by='Score', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1427,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve, auc\n",
    "models = [XGBModel,GBModel,logisticModel,RanModel]\n",
    "labels =['XGBModel','GBModel','logisticModel','RanModel']\n",
    "colors = ['blue','darkorange','red','green']\n",
    "\n",
    "fpr = dict()\n",
    "tpr = dict()\n",
    "roc_auc = dict()\n",
    "for i,model in zip(range(4),models):\n",
    "    fpr[i], tpr[i], _ =roc_curve(y_true=y_test, y_score=model.predict_proba(x_test)[:,1])\n",
    "    roc_auc[i] = auc(fpr[i], tpr[i])\n",
    "\n",
    "plt.figure(figsize=(10, 10))\n",
    "for i,color,label in zip(range(4),colors,labels):\n",
    "    plt.plot(fpr[i], tpr[i], color=color, lw=2, label='{0}(area = {1})'.format(label,roc_auc[i]))\n",
    "\n",
    "plt.plot([0, 1], [0, 1], color='navy', lw=2, linestyle='--')\n",
    "\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.title(\"ROC curve\")\n",
    "plt.show()\n"
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "name": "Titanic with XGBoost.ipynb",
   "provenance": [],
   "version": "0.3.2"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}