4wk_앙상블/앙상블(2)_13기 신민정.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Assignment 2. 캐글 경진대회 참여\n",
    "https://www.kaggle.com/c/house-prices-advanced-regression-techniques"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "_cell_guid": "2dbccbd6-138b-4f1b-9b23-fd60c7525c14",
    "_execution_state": "idle",
    "_uuid": "c9b1d5dff21d39260eb47af6fe7aac4bd03be233"
   },
   "outputs": [],
   "source": [
    "#import some necessary librairies\n",
    "\n",
    "import numpy as np # linear algebra\n",
    "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt  # Matlab-style plotting\n",
    "import seaborn as sns\n",
    "color = sns.color_palette()\n",
    "sns.set_style('darkgrid')\n",
    "import warnings\n",
    "def ignore_warn(*args, **kwargs):\n",
    "    pass\n",
    "warnings.warn = ignore_warn #ignore annoying warning (from sklearn and seaborn)\n",
    "\n",
    "\n",
    "from scipy import stats\n",
    "from scipy.stats import norm, skew "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "_cell_guid": "59617b4b-d797-44ce-9142-05fbfd36aada",
    "_execution_state": "idle",
    "_uuid": "0e694d13459e3e200f6e2c6333c887cbad779ba9"
   },
   "outputs": [],
   "source": [
    "test = pd.read_csv('test_new.csv') \n",
    "train = pd.read_csv('train_new.csv') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "_cell_guid": "3678529f-9d76-4853-88c5-4b2d230a85b6",
    "_execution_state": "idle",
    "_uuid": "3a32f51460a02fbe7a9122db55a740eb378dda97"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Id</th>\n",
       "      <th>MSSubClass</th>\n",
       "      <th>MSZoning</th>\n",
       "      <th>LotFrontage</th>\n",
       "      <th>LotArea</th>\n",
       "      <th>Street</th>\n",
       "      <th>Alley</th>\n",
       "      <th>LotShape</th>\n",
       "      <th>LandContour</th>\n",
       "      <th>Utilities</th>\n",
       "      <th>...</th>\n",
       "      <th>PoolArea</th>\n",
       "      <th>PoolQC</th>\n",
       "      <th>Fence</th>\n",
       "      <th>MiscFeature</th>\n",
       "      <th>MiscVal</th>\n",
       "      <th>MoSold</th>\n",
       "      <th>YrSold</th>\n",
       "      <th>SaleType</th>\n",
       "      <th>SaleCondition</th>\n",
       "      <th>SalePrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>65.0</td>\n",
       "      <td>8450</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>12.247699</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>20</td>\n",
       "      <td>RL</td>\n",
       "      <td>80.0</td>\n",
       "      <td>9600</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>2007</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>12.109016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>68.0</td>\n",
       "      <td>11250</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>12.317171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>70</td>\n",
       "      <td>RL</td>\n",
       "      <td>60.0</td>\n",
       "      <td>9550</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2006</td>\n",
       "      <td>WD</td>\n",
       "      <td>Abnorml</td>\n",
       "      <td>11.849405</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>84.0</td>\n",
       "      <td>14260</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>12.429220</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>855</td>\n",
       "      <td>856</td>\n",
       "      <td>20</td>\n",
       "      <td>RL</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6897</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>2010</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>11.751950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>856</td>\n",
       "      <td>857</td>\n",
       "      <td>80</td>\n",
       "      <td>RL</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10970</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Low</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>MnPrv</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>11.898195</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>857</td>\n",
       "      <td>858</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>65.0</td>\n",
       "      <td>8125</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>2009</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>12.066816</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>858</td>\n",
       "      <td>859</td>\n",
       "      <td>20</td>\n",
       "      <td>RL</td>\n",
       "      <td>80.0</td>\n",
       "      <td>10400</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>MnPrv</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>2009</td>\n",
       "      <td>WD</td>\n",
       "      <td>Family</td>\n",
       "      <td>11.931642</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>859</td>\n",
       "      <td>860</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>NaN</td>\n",
       "      <td>11029</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>IR1</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>2006</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>12.429220</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>860 rows × 81 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Id  MSSubClass MSZoning  LotFrontage  LotArea Street Alley LotShape  \\\n",
       "0      1          60       RL         65.0     8450   Pave   NaN      Reg   \n",
       "1      2          20       RL         80.0     9600   Pave   NaN      Reg   \n",
       "2      3          60       RL         68.0    11250   Pave   NaN      IR1   \n",
       "3      4          70       RL         60.0     9550   Pave   NaN      IR1   \n",
       "4      5          60       RL         84.0    14260   Pave   NaN      IR1   \n",
       "..   ...         ...      ...          ...      ...    ...   ...      ...   \n",
       "855  856          20       RL          NaN     6897   Pave   NaN      IR1   \n",
       "856  857          80       RL          NaN    10970   Pave   NaN      IR1   \n",
       "857  858          60       RL         65.0     8125   Pave   NaN      Reg   \n",
       "858  859          20       RL         80.0    10400   Pave   NaN      Reg   \n",
       "859  860          60       RL          NaN    11029   Pave   NaN      IR1   \n",
       "\n",
       "    LandContour Utilities  ... PoolArea PoolQC  Fence MiscFeature MiscVal  \\\n",
       "0           Lvl    AllPub  ...        0    NaN    NaN         NaN       0   \n",
       "1           Lvl    AllPub  ...        0    NaN    NaN         NaN       0   \n",
       "2           Lvl    AllPub  ...        0    NaN    NaN         NaN       0   \n",
       "3           Lvl    AllPub  ...        0    NaN    NaN         NaN       0   \n",
       "4           Lvl    AllPub  ...        0    NaN    NaN         NaN       0   \n",
       "..          ...       ...  ...      ...    ...    ...         ...     ...   \n",
       "855         Lvl    AllPub  ...        0    NaN    NaN         NaN       0   \n",
       "856         Low    AllPub  ...        0    NaN  MnPrv         NaN       0   \n",
       "857         Lvl    AllPub  ...        0    NaN    NaN         NaN       0   \n",
       "858         Lvl    AllPub  ...        0    NaN  MnPrv         NaN       0   \n",
       "859         Lvl    AllPub  ...        0    NaN    NaN         NaN       0   \n",
       "\n",
       "    MoSold YrSold  SaleType  SaleCondition  SalePrice  \n",
       "0        2   2008        WD         Normal  12.247699  \n",
       "1        5   2007        WD         Normal  12.109016  \n",
       "2        9   2008        WD         Normal  12.317171  \n",
       "3        2   2006        WD        Abnorml  11.849405  \n",
       "4       12   2008        WD         Normal  12.429220  \n",
       "..     ...    ...       ...            ...        ...  \n",
       "855      4   2010        WD         Normal  11.751950  \n",
       "856     10   2008        WD         Normal  11.898195  \n",
       "857      3   2009        WD         Normal  12.066816  \n",
       "858      6   2009        WD         Family  11.931642  \n",
       "859      8   2006        WD         Normal  12.429220  \n",
       "\n",
       "[860 rows x 81 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "_cell_guid": "ff37c1ba-8679-49e0-b3c8-9c53d01b1b04",
    "_execution_state": "idle",
    "_uuid": "816b1463b3dd0daf44949a1fa15ebfbc0e2f1235"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Id</th>\n",
       "      <th>MSSubClass</th>\n",
       "      <th>MSZoning</th>\n",
       "      <th>LotFrontage</th>\n",
       "      <th>LotArea</th>\n",
       "      <th>Street</th>\n",
       "      <th>Alley</th>\n",
       "      <th>LotShape</th>\n",
       "      <th>LandContour</th>\n",
       "      <th>Utilities</th>\n",
       "      <th>...</th>\n",
       "      <th>ScreenPorch</th>\n",
       "      <th>PoolArea</th>\n",
       "      <th>PoolQC</th>\n",
       "      <th>Fence</th>\n",
       "      <th>MiscFeature</th>\n",
       "      <th>MiscVal</th>\n",
       "      <th>MoSold</th>\n",
       "      <th>YrSold</th>\n",
       "      <th>SaleType</th>\n",
       "      <th>SaleCondition</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>861</td>\n",
       "      <td>50</td>\n",
       "      <td>RL</td>\n",
       "      <td>55.0</td>\n",
       "      <td>7642</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>GdPrv</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>2007</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>862</td>\n",
       "      <td>190</td>\n",
       "      <td>RL</td>\n",
       "      <td>75.0</td>\n",
       "      <td>11625</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>2010</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>863</td>\n",
       "      <td>20</td>\n",
       "      <td>RL</td>\n",
       "      <td>81.0</td>\n",
       "      <td>9672</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>GdPrv</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>2010</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>864</td>\n",
       "      <td>20</td>\n",
       "      <td>RL</td>\n",
       "      <td>70.0</td>\n",
       "      <td>7931</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>GdPrv</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>2009</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>865</td>\n",
       "      <td>20</td>\n",
       "      <td>FV</td>\n",
       "      <td>72.0</td>\n",
       "      <td>8640</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>2008</td>\n",
       "      <td>New</td>\n",
       "      <td>Partial</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>595</td>\n",
       "      <td>1456</td>\n",
       "      <td>60</td>\n",
       "      <td>RL</td>\n",
       "      <td>62.0</td>\n",
       "      <td>7917</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>2007</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>596</td>\n",
       "      <td>1457</td>\n",
       "      <td>20</td>\n",
       "      <td>RL</td>\n",
       "      <td>85.0</td>\n",
       "      <td>13175</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>MnPrv</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2010</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>597</td>\n",
       "      <td>1458</td>\n",
       "      <td>70</td>\n",
       "      <td>RL</td>\n",
       "      <td>66.0</td>\n",
       "      <td>9042</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>GdPrv</td>\n",
       "      <td>Shed</td>\n",
       "      <td>2500</td>\n",
       "      <td>5</td>\n",
       "      <td>2010</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>598</td>\n",
       "      <td>1459</td>\n",
       "      <td>20</td>\n",
       "      <td>RL</td>\n",
       "      <td>68.0</td>\n",
       "      <td>9717</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>2010</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>599</td>\n",
       "      <td>1460</td>\n",
       "      <td>20</td>\n",
       "      <td>RL</td>\n",
       "      <td>75.0</td>\n",
       "      <td>9937</td>\n",
       "      <td>Pave</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>600 rows × 80 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Id  MSSubClass MSZoning  LotFrontage  LotArea Street Alley LotShape  \\\n",
       "0     861          50       RL         55.0     7642   Pave   NaN      Reg   \n",
       "1     862         190       RL         75.0    11625   Pave   NaN      Reg   \n",
       "2     863          20       RL         81.0     9672   Pave   NaN      Reg   \n",
       "3     864          20       RL         70.0     7931   Pave   NaN      Reg   \n",
       "4     865          20       FV         72.0     8640   Pave   NaN      Reg   \n",
       "..    ...         ...      ...          ...      ...    ...   ...      ...   \n",
       "595  1456          60       RL         62.0     7917   Pave   NaN      Reg   \n",
       "596  1457          20       RL         85.0    13175   Pave   NaN      Reg   \n",
       "597  1458          70       RL         66.0     9042   Pave   NaN      Reg   \n",
       "598  1459          20       RL         68.0     9717   Pave   NaN      Reg   \n",
       "599  1460          20       RL         75.0     9937   Pave   NaN      Reg   \n",
       "\n",
       "    LandContour Utilities  ... ScreenPorch PoolArea PoolQC  Fence MiscFeature  \\\n",
       "0           Lvl    AllPub  ...           0        0    NaN  GdPrv         NaN   \n",
       "1           Lvl    AllPub  ...           0        0    NaN    NaN         NaN   \n",
       "2           Lvl    AllPub  ...           0        0    NaN  GdPrv         NaN   \n",
       "3           Lvl    AllPub  ...           0        0    NaN  GdPrv         NaN   \n",
       "4           Lvl    AllPub  ...           0        0    NaN    NaN         NaN   \n",
       "..          ...       ...  ...         ...      ...    ...    ...         ...   \n",
       "595         Lvl    AllPub  ...           0        0    NaN    NaN         NaN   \n",
       "596         Lvl    AllPub  ...           0        0    NaN  MnPrv         NaN   \n",
       "597         Lvl    AllPub  ...           0        0    NaN  GdPrv        Shed   \n",
       "598         Lvl    AllPub  ...           0        0    NaN    NaN         NaN   \n",
       "599         Lvl    AllPub  ...           0        0    NaN    NaN         NaN   \n",
       "\n",
       "    MiscVal MoSold  YrSold  SaleType  SaleCondition  \n",
       "0         0      6    2007        WD         Normal  \n",
       "1         0      4    2010        WD         Normal  \n",
       "2         0      5    2010        WD         Normal  \n",
       "3         0      7    2009        WD         Normal  \n",
       "4         0      5    2008       New        Partial  \n",
       "..      ...    ...     ...       ...            ...  \n",
       "595       0      8    2007        WD         Normal  \n",
       "596       0      2    2010        WD         Normal  \n",
       "597    2500      5    2010        WD         Normal  \n",
       "598       0      4    2010        WD         Normal  \n",
       "599       0      6    2008        WD         Normal  \n",
       "\n",
       "[600 rows x 80 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "_cell_guid": "b24451a1-fb8c-4094-ad0b-0940469d07fc",
    "_execution_state": "idle",
    "_uuid": "687813c270cbfdedccc7a9e4ec9fbb78a99d54ed"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The train data size before dropping Id feature is : (860, 81) \n",
      "The test data size before dropping Id feature is : (600, 80) \n",
      "\n",
      "The train data size after dropping Id feature is : (860, 80) \n",
      "The test data size after dropping Id feature is : (600, 79) \n"
     ]
    }
   ],
   "source": [
    "#check the numbers of samples and features\n",
    "print(\"The train data size before dropping Id feature is : {} \".format(train.shape))\n",
    "print(\"The test data size before dropping Id feature is : {} \".format(test.shape))\n",
    "\n",
    "#Save the 'Id' column\n",
    "train_ID = train['Id']\n",
    "test_ID = test['Id']\n",
    "\n",
    "#Now drop the  'Id' colum since it's unnecessary for  the prediction process.\n",
    "train.drop(\"Id\", axis = 1, inplace = True)\n",
    "test.drop(\"Id\", axis = 1, inplace = True)\n",
    "\n",
    "#check again the data size after dropping the 'Id' variable\n",
    "print(\"\\nThe train data size after dropping Id feature is : {} \".format(train.shape)) \n",
    "print(\"The test data size after dropping Id feature is : {} \".format(test.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "7d5829c4-b2f1-4ef3-8b02-11f02eb7aabf",
    "_execution_state": "idle",
    "_uuid": "228cb602f1c7a47d3c5250514cab57f7e7bc75e5"
   },
   "source": [
    "# <span style=\"color:blue\"> Data Processing 전처리 같이 해봅시다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "_cell_guid": "637bd0fd-7508-41d1-b240-ea0e8598dddf",
    "_execution_state": "idle",
    "_uuid": "8903aa1a4a700aa2160edb3baf806f3800ae7d9a"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAEHCAYAAABFroqmAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO3dfZxcdX3o8c88LLsEN1luu+6CtgkP4dvUSrxVCBgT80KBG7yBCr22eNUqtdFGba0B1ICA2FIoRK+2phQfyoXW9lojBUpWCPiQECQoxSA0fhMeghSyC2LYbJPssjs7948zZ3N25pwzZ2bOPJ7v+/Xixc6ZMzNnTnbP9/yevt9UPp/HGGOM8Uo3+wCMMca0HgsOxhhjSlhwMMYYU8KCgzHGmBIWHIwxxpTINvsA4jI9PZ3P5ZI38yqTSZHE712OnRd/dl78Jfm8dHVlfgH0F2/vmOCQy+V5+eWDzT6Mhuvrm5PI712OnRd/dl78Jfm89Pf3PuO33bqVjDHGlLDgYIwxpoQFB2OMMSUsOBhjjClhwcEYY0wJCw7GGGNKdMxUVmOMaRVDO0fYsHUPI2MTDPR2s2bZAlYuGmj2YVXEgoMxxsRoaOcI19yzm/GpaQCGxya45p7dAG0VIKxbyRhjYrRh656ZwOAan5pmw9Y9zTmgKjWk5SAiS4DrVHWFiPwmcBOQAnYAH1PVXNH+jwCjhYdPq+oHGnGcxhhTq5GxiYq2t6q6BwcRuRR4L3CgsOkaYJ2qbhGRm4Fzgds8+/cAqOqKeh+bMcbEbaC3m2GfQDDQ292Eo6leI1oOTwLnA7cWHl+gqjkROQIYBEaK9l8MzBGRewrHt05VHyz3IZlMir6+OTEednvIZNKJ/N7l2HnxZ+fFX5zn5ZKzhctuf4zxycNdSz1daS45W9rq3KcaUUNaRBYA/6yqpxUezwfuxek6OltVX/Ls+3rgNOCrwEJgCBBVnQr7jMnJXD6JibOSnDAsjJ0Xf3Ze/MV9XtpptlJ/f+/DwJuKtzdltpKqPgMsFJEPAp8H/sDz9C7gCVXNA7tE5CXgGODZxh+pMcZUbuWigZYNBlE1fLaSiNwhIgsLD8eA6aJdLgLWF/Y9FpgL7G3cERpjjGlGy+Fa4GYReQU4CHwQQERuAS4HvlZ4/n4gD1xUrkvJGGNMvBoy5tAINuZgvOy8+LPz4i/J5yVozMEWwRljjClhwcEYY0wJCw7GGGNKWHAwxhhTwoKDMcaYEhYcjDHGlLB6Dsa0mXZKzWDalwUHY9pIpxSSMa3PupWMaSOdUkjGtD5rORhTg0Z38XRKIRnT+qzlYEyV3C6e4bEJ8hzu4hnaWVyiJD5BBWParZCMaX0WHIypUjO6eNYsW0BPdvafbU82zZplC+r2mcWGdo6w6qbtnLp+C6tu2l7XYGiax7qVjKlSM7p43C6rZs1WsgHx5LDgYEyVmlUruJmFZMJaSxYcOot1KxlTpVbo4mk0GxBPDms5GONRyeyjZnfxNEOzWkum8Sw4GFNQTX96J9QKrsSaZQtmnSPo/NZSUllwMKbA+tPLt5yS2FpKKgsOxhQkvT89asspaa2lpLLgYExB0vvTW6HlZEkFW4fNVjKmIImzj7ya3XJqxopzE8yCgzEFKxcNsO6shQz2dpMCBnu7WXfWwsTcuTY7NYclFWwt1q1kjEdxf7qbKiIJ3RzNnonU7JaLmc2CgzEBmpkqohl9782eiZT0MZ9W05DgICJLgOtUdYWI/CZwE5ACdgAfU9WcZ980sAFYDEwAH1TVJxpxnKZ9FV9MLzlbWD6/r6b3bNYArV9QumKTsuO5UT719pMiv0c1F/lmzkRqdsvFzFb3MQcRuRT4KtBT2HQNsE5VlwJzgHOLXvI7QI+qng58Clhf72M07c1vIPOy2x+reSCzWd0cfkEJYOOO4UjfqR4Du43IxJr0MZ9W04iWw5PA+cCthccXqGpORI4ABoHi37K3AN8BUNUHReRNUT4kk0nR1zcnpkNuH5lMOpHf2+vGbc+U3uFPTnPjtme48PTjZrbdseN51m/exd7RcY6Z18PaM0/i3MXHBr7vMfN6eH50vGR7Hjjvqw+VfX21woJP8XcK2sevxXPjtmd4z1tOqPj35Y4dz3PN5t2MT3q61zbv5qg53bF//wtPP67s96sH+zsqVffgoKobRWSB53FOROYD9wKjgBa9ZG5huysnIllVnQr7nFwuz8svH4zpqNtHX9+cRH5vr70+F3B3u3tuirtqnh8dZ+23HuWB3S8EdtV8eOn8km4O1/Oj41z2r49x4OBE7He2QX3vMPs7BQk7H7ncdMW/L9ffrTOBwTU+Oc31d2vNXXetIsl/R/39vb7bmzKVVVWfUdWFwI3A54ue3g94jzZdLjCYZIsyBbOarhpvN4efuKdZul03QYEBog3Oxj0l1WYRJVPDg4OI3CEiCwsPx4Div9htwDmFfU8DftrAwzNtyG/xGjjdH27/eNiFLOwCv3LRAHeuXkIq4Hnv+9bSL+8dJwjSlU5x8JWpsu8f92K+Zq9/MM3RjKms1wI3i8grwEHggwAicgtwOXAbcKaIPIAzo+kDTThG02LCZt+sXDTAjudGue3RYabzs1/nDsbO7ckyOu7fAI1yB1xummWt016DWjbpFOTzMLcny4GJKfZP5Mq+f9xTUm0WUTI1JDio6h7gtMLPDwBLffZ5n+fhhxtxXKYyzcp7U+7CO7RzhLsef6EkMLjGp6Y5IhN07x/tDrjcBbLWaa9BASqfh4fWLmfVTdtLglvY+8c5JbXZ6x9Mc9giOBNJMxeErf/uk6EX3qC7bq/9EzmO7EpzqGhgNeodsN8FcunxR7Nh6x6u3KQExKXI/fJBLZO5PdnQ92lUv79lYk0eCw4mkmYuCAvqDhoem+DU9VsCL8zFigPD3O4MF7/txMjH771AFgfLIFFaJUM7Rzg0mfN97sDEFEM7R2z1sGk4Cw4mknrcuRZ3Uy09/mi2PbWP4bEJ0imYzjt97mGiBgY/c47IVh3YorRWorRKygWZqbzzWdbvbxrNgoOJJO47V79uqo07hmeed8cPgsYR4lBLYAt7bQoi98tHCTIjYxPW728azoKDiSTuO9coF8VqpYjWoqilSyYoWA72dnPn6iWBrytuLYVNXS0+zkb3+1vhnWSz4GAiifvOtV4DqT3ZdKSgU2uXTFiwDLqo+rWW6n2c1WrmBATTGiw4mMjivHONetfscuf7u2MTdz3+gm8QiBIY0qnZq5ur+U5BwRIIvKhW2loabOLdeiuUDDXNZcHBNIXfnXeQnmy6JDvn4tfM44pNxWm5ysumnEFeqOxuOKg1UPy6VTdtD7yohrWWBnu7SwbmR8YmagpgtWj21FnTfFYm1DSFX3rmCxYPzuQxcmcpBaVtXrloIDDn0dzuDF0Bi96migYjouRHKpcC25s2I6g15F74g6xZtoCH1i5nzbIF3PX4C02vo2wpM4y1HEzT1NpNFdTvf/HbTuSoOd1cf7dG6roqdzcc1MVyxSZl/Xef5MDEVEnQKZYHDr4SnD8ybEFfM7pzbOqsseBg2lbYIHlf3xyWz+8rm+UUyt8NhwWPoAV6fty8SGGf0SrdOTZ11lhwME0VZbpkuaR7YResKBdVN3tr0MWv0sHzargBKiyNxqqbtje8prQFg+TKXHXVVc0+hlhMT+evGh+fbPZhNFxPTxft+L2Hdo5w0T8+wuZdv+C/XnHuqP/rlRzf2/0S/+/f/5NX93azsP9VXHvvLjbc/0zoPn7c83LHY8Mzrw3zX6/k+OHT+zhmXul7Hj2nix8+vY+pOq3I68mm+cQZJ7Cw/1W+n9WVTvHK1PRMyyPsWMt+VhW/L0M7R/jEbY/zf77/FHc8NszRc7oq/txW165/R3E46qjuvcBNxdtT+Xwdl6A20ORkLp/ESk7tWMFqaOcIVw9paD99NgVd2dJEeV5+s5hc7nnxS0/hnbFULGgRm9t6iaMFMa8ny5FdGUbGJujtzpBKpdg/PjVrOqz7WW4akUqONUylvy9+5y/svLerdvw7ikt/f+/DQEk5ZputZBpuw9Y9ZQdwp/KlifKKjU9Ns/67T4YW2CmeFTW3OxOakiOoG8ot+nP1ORJ+4BGsPeME7ly9hM+eI7ySyzM6PjVrZhIcLthTzbHGKWyA3HQ2G3MwDRdn//3o+NTMoPDw2ARXDzkziLx34t6+81U3bQ8dGPYW7wla13DDfU+EvkeYud2ZWYO9YRfeODK+1qpVBshN41nLwTRcuUyrtZjKU3In7m1NlLuouekvwtY1XPy2E33LkpbjTrMtdywjYxNlA2ijppXaeofksuBgGq6emVaLFXeBhF3U5vVky641gMNdVZUEuXk92ZJ++qBj6e3OhL5X0MLAeoi7HnUtdbZNY1lwMA0XtLLZlY1w0R3s7WZumYuoy3uHvmbZAt/370qnWHvGCSX7e3nv5lcuGuCqlRKpBXFkNsXo+BRXbFLe/uUHZi6IQRfeVCr4BMzryXLn6iUNGwz2W8lebWAq1yIzrcXGHEzdePvt3Vk5o+NTlLv2zzkiE9qn/6O1y2feP0p+pjyUrGPwjhvM68my9owTZp4LW9cwtHNk1hoLIHAWU7Yw0+iQZ/R9dHyKz31nV8nrvWMbV4bkjNpfwaK7uMS13qFVVn+baGwqa5tr9hQ8vwCwf3yKuT3ZSGklKuVO3/ROLXWne87tznBocprJgH4rvymYfgPPQGBSv7Dpo8XvdWgyF7iCOux9wlZ1VzN91auZvy9BJV1TwEOFgN8szf47aqagqazWcjBVK75z997tV5JWohJrli3g2nt3lVSN8w72Bt3Jj09Nc+Um5cpNytyeLBOTOcZzhy9XbjfHurMWBn5+2IB28R32qeu3BO4btip7zbIFfO47u0qCXDZFW+c2sjrY7cXGHEzV6lnNzY87xuANDC5v98Sdq5cEdl3lC/+Njk/NCgzF7xM0LlLJhazcvkF97isXDfCZ/3ES83oO37vN7c5wxUpp6+6XuAe3TX01pOUgIkuA61R1hYi8AfhrIAdMAO9T1ZGi/R8BRgsPn1bVDzTiOE1lGjnX3W0ZhC2+Gh6bYMnnt/DG184lnQKfa38kI2MTfPYc8R3PWHr80YGvK+5WWnr80dz+6HBo11pQn3sn5jWyZH7tpe7BQUQuBd4LHChs+iLwMVX9iYh8CPgk8AnP/j0Aqrqi3sdmatPbHT5wHIcUzLqIhA3WgtPF9KNn99f0mQO93axcNMCO50ZLWil3/HSEzT97kbGJXNkSoHc9/gLnnTzI5p+9GHqe3C6mJFwwOzHodapGtByeBM4Hbi08/n1V3ev5/PGi/RcDc0TknsLz61T1wQYcp6lQ2JTLOBQPvg7tHCFVKBdaL95ujm1P7St5fnI6z2ThQj88NsHnvrOLHc+NctujwyXrN8anptn21D7u++hSIHyg2d3uvidYrWbTXHUPDqq6UUQWeB7vBRCRNwMfBYqnKRwEbgC+CiwEhkREVDV0hDOTSdHXNyfOQ28LmUy6ad+7ntMquzIpLjlb2PLMy6zfvIvnR4vvIeJ39JwuLj9nEecuPpY7djwfKc3H5HTedwzENTI2MfPvc8nZwmW3P8Z4mZxRk9N5vvD9p7jw9OMq+wIRNPP3pZXZeSnVlNlKIvJ7wGXAO1T1xaKndwFPqGoe2CUiLwHHAM+GvWcul0/kVLS4p+BFqa/gqmedgyOzaQ4cnIhcZ7paxd1WAP/0w6dnEuDVaqC3e+bfZ/n8PtaduXDW+Q06f/sOTtbl9znJUzbDJPm89Pf3+m5veHAQkfcAHwJWqOovfXa5CHg9sEZEjgXmAnt99jMx8+s3dy+SfgFizbIFgesBajU2kav7bChvt5Wb1mFkbIJUSJrsSvjNxCnucz8lZLqrMc3U0OAgIhngS8DPgW+LCMAPVPVKEbkFuBz4GnCziNyPM+vwonJdSqZ2QztHuGpIffvN/WbTuC2Mehno7a7rbKiebJqlxx/tOw4Qx5hGOkWkNBNzQwb1vauxjWk0WyHd5uJoDkdJQzEYMjPH68iu8AI9UR2ZTTE+lfddUVurwd5ufq2vu+ZZTa4UzDrOqMVwrr13F9/eMRz4HWtdDQ2l3YSXnC0sn99X03t2ooR3K1mxH+MvSveNd8FW0P5u+oo4HKpTYEgXVhnHFRgAsukUc7szFSWmc1d5h33HWltOfonuLrv9MUt0ZyKx9BkJVmnpS7eLKeiiVe2ah7BSmHGbzlN2rUSlJqfz/MoR2Zkpq1Hc9mjwDCdXrWklfBPdTYZ3EyZhrYWJpqLgICInAvOBHwBHqepomZeYFhU1o2mxcrNsqtHI+g5AXVokld7ll/vOcaSViFrFrdKJCCYZInUriciAiHwf+ClwO3AisEdETq/jsZk6qnYmUG93hkOT9V0V3Y4qvcsPKxQUVzGfqFXcrE608RN1zGED8CNgHjCpqj8DrgC+UK8DM/UVdqc72NvNBYsHS4ripIBDk9N1y7jarqq5y3/nyYO+2y9YPBhbMR/fRHddpcdqdaKNn6jBYTlwmaq+wuFW+ZeB36jLUZm6i3KnW5weIw+BtRKSptbKaJ96+0lcsHhwpgWRTjmB4VNvPym2Y/Sr4vYX5/1WybFanWjjJ+qYwz7geOBnnm3HAS/EfkSmIdYsWxA45jA8NhGaEsJQ8xRTcAJEnMHAT/GiO78pm36/C5ZK20QNDl8AviMiXwC6ROT9wMXA39TrwEx9lStxaYJFrV3dLiyVtvETKTio6t+KyC9wUlv8HPjfOPUZbg1/pWlFxdMWzWHlFvFlU8xUnOsklkrbFKtkEdwO4F2q+jqcpHkP1eeQTD35LYyKKpMisMJaJ7j6HGHLn7wldJ92r8ZmTFRRp7K+G/gxzjgDwBuBB0TkvHodmKmPWpLZ5fL1WSPQCgYLBX7cn8vtY0ynizrmcDVwhqo+CjPdTA8Dt+CsezBtwqYnlioefF2zbAFXD+ms8p7ZQtqNSiVx5XESv3MnihocXg38pGjbw4D9i7eZnoA+9Xk9WY7syrT04HRcSf2K+U1FTRWVnKum6l0SVx4n8Tt3qqhjDv+OU+vZ62KcAGHaxNDOkcCL6+j4FIcmcyUL31rJRJ1qO2zYumdWMroNW/eUrOeYnM5XvGI4iSuPk/idO1XUlsNHgX8TkT/FKbxzDM7ah1X1OjATv3J/oKPjU3SlU/SkYTzXeqML9Vp/V3x3G9eK4SSuPE7id+5UUaeyPiYiC4GlOF1JzwHbVXWyngdnauft/41ybZ2czpPEf1RvUaOgxIKVTvuN633aSRK/c6cK7VYSkbMK/z8HOBOYA4zhlO48s7DdtKjiaasmnHt365uTqIoVw3G9TztJ4nduFre07anrt7Dqpu2x1+ko13L4PPBbOHmU/ORx0mqYBvK2Bo6Z18OHl873Heyrdw3mTpNKwanrtzDQ2807Xvdqtj21r6YZN0lceZzE79wMjRj4j1QmVETeCtyvqi2bqzkpZUL96jAElaU8df0WazFUKWqpz3aT5HKYYdrtvPjVPofqSsvWWiZ0I9BV0SeauqhkNsjcns4v9JctZDPtCiiQMK8ny7wqzoPNsDGtrBED/1GDw3bgXSJiAaLJKqnudWCi8+suTOVh21P7OPf1A7OCQN+RWa4+R7j3I29m7RknlPSDR2EzbEyrakSa9ai3VMcBNwNfF5F9eLIoqOqrYzsaU1bU2SAbtu6ZtcK3k8zryXJgYmrm+3lTjA8W+rgvPP24mW6CajPQ2gwb06oakWY9anD449g+0dQk6i9Fp971XrB4kG1P7QusRucOzB01p5vl8/tmtnuzjnoH9FOp4PUTNsPGtKpGDPyXHZAWkaNxWg7/oarjsX1yzJIyIA3lZysN7RzhqiGt26KxZkoXslqU+2rHzuvh9g+eWvb9/Ab4If6qbK2i3QZeGyXJ5yVoQDq05SAiy4C7gFcBwyJyrqr+uNIPF5ElOPUfVojIG4C/BnLABPA+VR3x7JvGqVm9uPD8B1X1iUo/sxMVJzT77Dkyq/vE3eeae3Z3ZGAA5y5/MKBrzWvvaLT7GJt6aYy/ct1KfwF8BvgKcAlwDXBWJR8gIpcC7wUOFDZ9EfiYqv5ERD6Ek7PpE56X/A7Qo6qni8hpwHog8anBg+Y1F3ef3HDfEx29tiGdCi9x6jpmXk/k96y10E3ULKSWrdS0k3JTOBar6hdV9SBwA/CGKj7jSeB8z+PfV1U3w2sWKL7FewvwHQBVfRCf5k4SBU1hXb9518zjoZ0j7J9o2aUosXjnyYOsXDTAurMWBtZd6MmmWXumf5dQ3KtK/YonXXPP7pL3jbqfMa0i8gRwVT0gIhVPGFfVjSKywPN4L4CIvBknod/yopfMBUY9j3MiklXV0HmZmUyKvr45lR5eS7tjx/Os37yLvaPjgX3sz4+Oc8r6LaRDBlY7RQro7u6ir28OF55+HBee7tSe8p6nY+b1sPbMk3jnb7+WXG52ML1jx/Ncs3k345Oe1tdmp/V17uJjqzqmG7c94xu0b9z2zMzxVbJfvWUy6Y77O4mDnZdS5S72dUngLCK/h1Nq9B2q+mLR0/uBXs/jdLnAAJDL5TtqQClooDRIpwcGcAahv/HQs0xMTM4aLF4+v4/lRYPPudx0ye/D9XfrTGBwjU9Oc/3dOqtrrhJBYxt7R8dnfX7U/eotyQOvYZJ8Xvr7e323lwsOWRFZyeEgUfwYVd1UyYGIyHuADwErVPWXPrtsw0kF/s3CmMNPK3n/TrH+u0929NhBLW57dLiqmUT1WFUadd2JZSs17aZccHgBZ+aQ66WixxUl3hORDPAl4OfAt0UE4AeqeqWI3AJcDtyGk/H1AZwg9IGo79/u3AHLVq7G1gqqbSXV4wIddd1JIxYtGROn0OCgqgvi+BBV3QOcVnj43wL2eZ/n4Yfj+Nx2MrRzpKRucadLAecvHmTxa+Zxw31PzAymz+vJsvaMEwLXagSkUSqrHhfoqFNhbcqsaTeRB5hFpA/4XeC1ODOXTlHV79XrwJLmhvueSFRgAKfZuXHHMJt/9iKpVIoUzLpo7nhudCYthtc7Tx4Mfd+hnSPcuO0Z9o6Oz3q/el2go06FrXXKrDGNFDVl9ynAJpz+/1OA1wOPAh9X1a/X9QgjavcV0qes39LsQ2gZPdn0TD0FbzdQOuUEhrDxhkpSmidRkgdewyT5vNSasvtLwIdV9QxgqtBNtBL4dGxHmGA213228alpNu4YLhkfKBcYwArcGxOXqN1Kv4EzUAyFtDaquk1ELCNrjdw7XVPexh3DbNwxPJN5FUq7iKzAvTHxiBocdgPvAO50N4jIGcCuwFeYSKyUZ+WGxya4ekhJpVJMFkas3RXHc3uyvhlbvSVAKxlnsJQXJqmiBoeLgTtF5LvAHBG5GWctwrvqdWCdzHvBSdgYdGym8jjpWT3Gp6Y5IpOiJ5suCbjTntoPUWvtNqJOrzGtKtKYg6puwRmEfhD4GvAEsERV76vjsXUkd8rqsAWGuhibyLHurIUcO6+HFP7TXqOOQdj4hUmySnIr/Ry4ro7HkghJnLLaSAO93axcNDCTyvzUgFlgUcYgwsYvrLvJdLpy9RxepExdFSsTWplOz5raKNkUs8YcwH9BWy2rooNe29udse4m0/HKtRx+tyFHYUwEg73ds+7U4XBd6HRqdpePm+m0llXRQa9NpVKMT80O8u5nW3AwnaJc+owfBD1XyJO0KPYj6mC2nqF6g73d3Ll6ie9zYUWQalkVHfTaKzep7/42XdZ0kkhjDiJyLk5pz9cwO433AZz6CyaCThnITBHe1zjY283S44/mtkeHmc47g8LdmRSHqhxs6UqnAu/0w4oguWm8a0lb4ffaoOSIlmHVdJKoA9LXAzcCYziV2r4MfA74tzodV0fqlDtL9xLfk0kxniu94A+PTbDtqX1ctVJmLqyV1qdwuUn4gi7uQec0ag3paliGVZMEUdNnvAa4FmcR3HGquhWnLnTisqfWotPuLMdzebIp5wJerLgMZrnSnn4Ge7u59yNvDr3rDzqnldSQrpT3u6RwjtNyN5lOEzU47AWOAp4FThSRlKo+C9hMpQqsWbaAnmzUU94epvJwZFeGud2Zkue8A8TeqZ+Dvd1csHiwbKCIOmhcfE7DakjHZeWiAe5cvYSH1i7nztVLLDCYjhP1SrUZuB2YB2wH1ovIXwB76nRcHcm946y2HkGrGh6bCJyi664JuOae3TML/4bHJrjr8RdYs2xBaB3aqIPGfnfx1daENsY4oo45fAInhUYeWAP8HU6d59V1Oq6O5V7wrgiY8dJpBnq7Q1caB60lqKT7yeokGBO/si0HEXkncJ6q/nlh/y8AAvy48J+pUJIuZOUypQZ1tQ2PTbDqpu02/deYJgkNDiJyEfBVnPEGgL8Bfg34M5wA8Zm6Hl2bG9o5wqqbtnPq+i2zLnTX3puMZLbuQHUqoO/ITXURNFBdPKhtjGmc0EpwIvII8KequkVE5gC/BM5R1e+KyInAPap6fIOONVSrVYKrdupmp8im4LyTB7nr8Rd8z4FfdbZVN20P7GIKWgAXJKyyV5LzIiW54lmYJJ+XoEpw5cYcji9kZAU4FWfM4X4AVX3Civ0ES3qdhqO6s2x7ap/vOUin8J362YhCPeXScCc5cBjjVS445ETkCFV9BVgBPFT4GRHpx1khbXx0yoK3IOmUU04hqN25f3yK/T5Fd8B5nd8Ft5YkeVGVS8NtCfWMcZQbkN4CXCwiC4D3ABs9z60DAnMvJV2nLXjz6smmuWql8NDa5YGzigZ6uwPPQdD2oDULca48DmudWP0GYw4rFxwuAf4QeBJ4BieFBiLyFPBu4LK6Hl0b68QFb1DaJRR2Qa/0Yt+IlcdhAcvqTxtzWOiANICIpIBfVdUXPdveBdyrqr+M8iEisgS4TlVXeLZ9AVBVvdFn/0eA0cLDp1X1A+U+o9UGpOHwwKdfV4lfPYJ2kAIeWrt81jbv90ynnJKcg0VptRvdhx80wOg3UcAdHA/6t6pmQLxVJXngNUySz0u1A9Koah54sWjbN6N+sIhcipOH6UDhcT9wC3ASTkK/4v17Cp+xIupntCrv4qzigc5DkzlGA/rk/ZTLhNoo7p138RmwsboAAA/XSURBVPdZevzRs2Ymuf31685a2FIX1nIpvC2hnjGOyGVCa/AkcD5wa+Hxq4CrgJUB+y8G5ojIPTjHt05VH6z3QdZb8SreUwLKVwZphcAATjeS34yfjTuGS/Zt1QI4QSuqa6n9YEynqXtwUNWNhQFt9/HTwNMiEhQcDgI34Cy+WwgMiYioauhtdiaToq9vTkxHXX+ZFPhkuw7kdtc024WnH8dbb/h+5Gm6I2MTTfl3yWTSVX3uhacfN1NFrhNVe146nZ2XUo1oOVRqF/BEoTtrl4i8BByDkxE2UC6Xb6s+w0oCA8QXGFI4NZCrrWW97K++59svH2Sgt7sp/y5J7kMOY+fFX5LPS39/r+/2VpxOcxGwHkBEjsWpNLe3qUdUg6AUGpUklovTQG83F7/tRN9ZRFHSaFcSGKy/3pj21TItBxG5Bbgc+Bpws4jcj9PVflG5LqVWEXWQFvyriXWl6zt7yS236de3vvT4o9n21L6Zegvu46jBoCeb5h2ve/XMe1h/vTHtrexU1nbR7KmsleRScqdG+qVq2PHcqO/gbhyyKbjCU7oz7Njd6Z1XbtLAwfB5PVn2j0+1ZCBIcjdBGDsv/pJ8XqqeymqiqSSXkruoynsHPzw2EXohjsNUHt/ZQ9XUWwCnAty9H3lzrMdouY2MaQ0WHGJS6SAtOKm7va2ERrTh/Fb7hq0M/uw5EliYyPu6OC7q5ZLiGWMapxUHpNtOJfUGutIpDr4yxSnrt1TVfVRrhVG/9BFhKSVWLhrwrQ/tfZ1fGdBq6jBYbiNjWocFhxhEuXilcPro8/l8xdNI0ym4+hzhR2uXM7en+sZe0OyhcjmQLn7biWSLolI2xaz0GHFc1C23kTGtw4JDDKJcvB5au5wjuzJMVdF39M6TB2e6VYLSYEeSz7P+u0+WTKuNkvAuVVTOzfs4rot6pVlcjTH1Y2MOMQgbtPWq9g5421P7In3WYG83o4de4VBABBrP5RnPOcFleGyCq4ecsQQ3nURQv/6GrXtKpthOTudnBrfjqsPgN73X1koY0xzWcohBufTc7j12tV1Cw2MTMwvpDk3mSrp4erJprj5HuHP1Es55XfSB26k83HDfE2X3Cwpqw2MTDO0cia0OQyNSdhtjorGWQwzci1fQrB73nruWNSXunfno+BRd6RRzj0gzNpHjmHk9fHjp/JkSl3c9/kJF7xtl/COsteJmXnVTXtc6BTWsBWOMaRwLDjFZuWggtB4AwFiV+YyKTU7n+ZUjstz30aWzFu/UWrc6aDqqX3ePyx14vnP1EruoG9NBrFspRmuWLQid1VNpH3zQFFKobL1CGPdww6ajut09lRyLMaa9WXCIWdisnnJjE+kUM33tV58j3PfRpaE1mqNsc983iNvRVW466spFAxUdizGmvVlwiFHYrB6g7B14Pu9MefV20VQy2Bu071UrJfDC7m6PMh01roFnY0zrszGHGNU639/vDryS6mS1lMCMMh3VKqUZkxwWHGIU5QIbtmo46A68khk81ZbAjLrGwGYTGZMMFhxiFOUCG9aKqPdFN+zCbq0CY4yXBYcKhWUfjXKBDWpdNKsynJe1CowxLgsOFYiSUrrcBbZc66KW1NdWC8EYExebrVSBOLKPhqWIqCX1dVxps40xBqzlUJG4so8GtS7Cgk+5FkAtrzXGmGLWcqhAvVNK1xJ8rBaCMSZOFhwqUO9FYLUEH6uFYIyJkwWHCtQ7pfSaZQvoKsp10ZVORQo+tnrZGBMnG3OoUL2nexan9Z6cznPFJmXD1j2hs49snYIxJk4WHFrIhq17AsuI+k2bLWbrFIwxcWlIcBCRJcB1qrrCs+0LgKrqjUX7poENwGJgAvigqpYvV9amvGsTypUCqtfsI1sfYYwpVvcxBxG5FPgq0FN43C8iQ8C5AS/5HaBHVU8HPgWsr/cxNkvx2oQo4p59ZOsjjDF+GjEg/SRwvufxq4CrgFsD9n8L8B0AVX0QeFM9D66ZqqncFvfsozgW9hljOk/du5VUdaOILPA8fhp4WkRWBrxkLjDqeZwTkayqToV9TiaToq9vTs3H20iVtgJ6utJccrbM+p6ZTLqm7x22PqLdzqdXreelU9l58WfnpVQrDkjvB3o9j9PlAgNALpefqaXcLsKS8N25eonvWMDy+X2zvmdf3xz+6YdPVz1mEJZmvN3Op5e3trY5zM6LvySfl/7+Xt/trRgctgGrgG+KyGnAT5t8PHVTLglflNlHd+x4viQZ4BWblCs2KYMRAkXUOg7GmGRpmeAgIrcAlwO3AWeKyAM4JZU/0NQDq6M41ias37wrcNwi6vTXWo/BGNN5UsWLrtrV5GQun8Rm4anrt5Sd6eR2UyVJkrsJwth58Zfk89Lf3/swPhN/LH1GmztmXk/ZfSz5njGmUhYcGmho5wirbtrOqeu3sOqm7bGsJVh75kklOZWKWfI9Y0ylWmbModNFqSJXjXMXH8uBgxNs2LrHd9aRDS4bY6phwaFB6lmMxzuryVJhGGPiYMGhQRpVjMeS7xlj4mBjDg1ixXiMMe3EgkODWDEeY0w7sW6lBrHFZsaYdmLBoYFsPMAY0y6sW8kYY0wJCw7GGGNKWHAwxhhTwoKDMcaYEhYcjDHGlLDgYIwxpoQFB2OMMSUsOBhjjClhi+DMLJbV1RgDCQ8OdiGcrV41J4wx7Sex3UruhXB4bII8hy+EcVRna1dhNSeMMcmS2OBgF8JSjao5YYxpfYkNDnYhLGU1J4wxrsQGB7sQlrKaE8YYV2KDg10IS61cNMC6sxYy2NtNChjs7WbdWQttMNqYBGrIbCURWQJcp6orRORE4GYgDzwGfERVpz37poD/BHYXNv1QVT8d9zFZ8R1/VnPCGAMNCA4icinwXuBAYdPngctV9fsiciNwHnCb5yUnAP+uqqvqfWx2ITTGGH+NaDk8CZwP3Fp4/EbgB4Wfh4CzmB0c3gi8RkS+BxwC/kxVtQHHWVe2psIY007qHhxUdaOILPBsSqlqvvDzGDCv6CV7gb9U1X8RkbcA/wCcUu5zMpkUfX1z4jjk2N2x43mu2byb8UnP4rLNuzlqTjfnLj62pvfOZNIt+72byc6LPzsv/uy8lGrGCmnv4oJe4OWi538MTAGo6v0i8hoR8QYUX7lcnpdfPhjvkcbk+rt1JjC4xienuf5uZfn8vpreu69vTst+72ay8+LPzou/JJ+X/v5e3+3NmK30iIisKPy8Etha9PyVwMcBRGQx8PNygaHV2ZoKY0y7aUbLYS3wFRE5AtgJfAtARO4B/idwLfAPIvIOnBbE+5twjLEa6O1m2CcQJHlNhTGmtaXy+ba+KZ8xOZnLt2qzsDihHThrKuJYQ5Dk5nAYOy/+7Lz4S/J56e/vfRh4U/H2RGdlbRRbU2GMaTcWHBrE1lQYY9pJYtNnGGOMCWbBwRhjTAkLDsYYY0pYcDDGGFPCgoMxxpgSFhyMMcaU6JhFcMCLwDPNPghjjGkz84H+4o2dFByMMcbExLqVjDHGlLDgYIwxpoQFB2OMMSUsOBhjjClhwcEYY0wJCw7GGGNKWMruFiYiS4DrVHWFiJwI3AzkgceAj6jqtIhcCbhV8z6uqg8F7duM7xAnEekCvg4sALqBPwf+AzsvGeArgAA54ANAioSfF5eIvBp4GDgT53vfjJ2Xsqzl0KJE5FLgq0BPYdPngctVdRnOH/55IvLbwFuBJcDvA18O2reRx15H7wFeKnyvlcDfYOcFYBWAqi4FrsD5nnZemLmh+DvgUGGTnZeILDi0rieB8z2P3wj8oPDzEPB24C3APaqaV9WfA1kR6Q/YtxP8C/AZz+Mp7Lygqv8KrC48nA+MYOfFdQNwI/B84bGdl4gsOLQoVd0ITHo2pVTVXc4+BswD5gKjnn3c7X77tj1V/S9VHRORXuBbwOXYeQFAVadE5P8Cf41zbhJ/XkTk/cCLqnq3Z3Piz0tUFhzah7evsxd4Gdhf+Ll4u9++HUFEfg34HnCrqn4DOy8zVPUPgJNwxh+O9DyV1PNyEXCmiHwfeANwC/Bqz/NJPS+RWHBoH4+IyIrCzyuBrcA24GwRSYvIrwNpVf1FwL5tT0QGgHuAT6rq1wub7byIvFdEPl14eBDnovbjpJ8XVV2uqm9V1RXAT4D3AUNJPy9R2Wyl9rEW+IqIHAHsBL6lqjkR2Qr8ECfQfyRo32YccB2sA44GPiMi7tjDnwJfSvh5+Tbw9yKyBegCPo7z/ZL+++LH/o4isqysxhhjSli3kjHGmBIWHIwxxpSw4GCMMaaEBQdjjDElLDgYY4wpYVNZTSKJyOuBy3By6vQBvwDuAi5T1Zd89n8/8FFVfZPPc8twFuUtiPjZf4iTN+tdqvov1X4HY+rJWg4mcUTkNOABnIyupwBHAcuBVwH3iEiqkvdT1a1RA0PBauBrwMcq+RxjGslaDiaJNgBfUtWrPdueFpE/Aq4C+kTkEZzV2BcA3wS2B71ZYRXtt3BSMzwDfFhV7yo8dwbwD8BrC6mhTwZOwEkf/XMRWayqOzzv87fA08BpOIkXHwW+CJyNs/r5Rpw07nkR+RXgS8DSwmc/Afyxqm6r6ewYg7UcTMIU0iP8d5xunVlU9ZCqflJV9xU2/TrwWuCTUd67kOv/H3HSPrveDfyjpw7Ah4BbVHU/cCvw0aK3+Q2c7LOvBe4v7JMHjgNW4KQtf39h378q/H8RTtfY/cC1UY7VmHKs5WCS5tjC/59zN4jIXwJ/XHh4BM4FHGCjqh4CDolI1Pe/BXhQRHpwchydj3NRR0SOxAkWpxf2vRF4SEQu9QSkPPANVZ0QkUGcnD79qnoAOCAi1xeO7+9xxkwO4qQuX4CTGO41UQ/UmDAWHEzSvFD4/zE4XUCo6qeBTwOIyI+BTGGf4UrfXFX/Q0R24VQVywHPquqjhad/Dyft8/c9weZI4A9x6g4A7FPVicLPv45TZOZJz/5p4JeFn4/F6XL6TeBnhe3WG2BiYb9IJlFU9Smcko8XRdi92sRjtwL/C3hX4WfXapwuqjd4/vszYI2IuH+L3s/ci9MqGFDVPlXtwynms7zw/D8D/wr8aqEKnM18MrGxloNJoj8C7haRHPB3qjoiIguAPwEWc7h1UaxLRF5btO2XPvt9A6diXRq4GEBEXoczM+o8VX3R3VFEbsYZJ3gHTkGZGar6bCFb6HWFlNxH4gSA54H34hSpOVAYnF6EE3i6Ip0BY8qwloNJHFV9EHgTziDvwyJyAGdq6yBwuqpuCnjpycCzRf+92+f9X8RJ//xjVXXLU64G7vMGhsK+ozh3/8UD064LgQFgD7AbJzC4KaVXA5eIyH6ctN03A/2FWUzG1MRSdhtjjClhLQdjjDElLDgYY4wpYcHBGGNMCQsOxhhjSlhwMMYYU8KCgzHGmBIWHIwxxpSw4GCMMabE/weIBQQCsRNIIQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "ax.scatter(x = train['GrLivArea'], y = train['SalePrice'])\n",
    "plt.ylabel('SalePrice', fontsize=13)\n",
    "plt.xlabel('GrLivArea', fontsize=13)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q1. 위 그림에서 이상치로 판단되는 데이터를 제거하시오.**\n",
    "\n",
    "**사람마다 기준은 다르나 저는 오른쪽 따로노는 아이가 이상하군요....?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "_cell_guid": "6c5780b2-d4a8-42d9-b902-c6a23eef7d99",
    "_execution_state": "idle",
    "_uuid": "583bb417102d7bebb4aaf14bcb1aebcae86443bb"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Deleting outliers\n",
    "train = train.drop(train[(train['GrLivArea']>4000) & (train['SalePrice']<12.5)].index)\n",
    "\n",
    "\n",
    "#Check the graphic again\n",
    "fig, ax = plt.subplots()\n",
    "ax.scatter(train['GrLivArea'], train['SalePrice'])\n",
    "plt.ylabel('SalePrice', fontsize=13)\n",
    "plt.xlabel('GrLivArea', fontsize=13)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "문제 조건에 맞게 오른쪽에 떨어져있는 하나의 값만 drop"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "802df76d-0e0b-4868-ba16-91335568d2d7",
    "_execution_state": "idle",
    "_uuid": "827a86d65c6d176f4af55224b91b44a47966652d"
   },
   "source": [
    "## Features engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "_cell_guid": "1bd3e9b9-2f42-4251-aadd-5ced84eb1a27",
    "_execution_state": "idle",
    "_uuid": "efc576211e4eed962f04cd94d901c667e6912528"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "all_data size is : (1459, 79)\n"
     ]
    }
   ],
   "source": [
    "ntrain = train.shape[0]\n",
    "ntest = test.shape[0]\n",
    "y_train = train.SalePrice.values\n",
    "all_data = pd.concat((train, test)).reset_index(drop=True)\n",
    "all_data.drop(['SalePrice'], axis=1, inplace=True)\n",
    "print(\"all_data size is : {}\".format(all_data.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "9ce95008-a3b9-43fa-bc4e-649ca0f43768",
    "_execution_state": "idle",
    "_uuid": "abe25f3032a0bed179d58d5911cb42d97b35841b"
   },
   "source": [
    "### Missing Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "_cell_guid": "501b465f-8c80-4b93-81d0-a5d41e08d235",
    "_execution_state": "idle",
    "_uuid": "f97d25548ec8f6c02e2d1ee5a6df6c3d107fdf53",
    "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>Missing Ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>PoolQC</td>\n",
       "      <td>99.520219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MiscFeature</td>\n",
       "      <td>96.298835</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Alley</td>\n",
       "      <td>93.762851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Fence</td>\n",
       "      <td>80.740233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>FireplaceQu</td>\n",
       "      <td>47.292666</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>LotFrontage</td>\n",
       "      <td>17.751885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>GarageYrBlt</td>\n",
       "      <td>5.551748</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>GarageType</td>\n",
       "      <td>5.551748</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>GarageQual</td>\n",
       "      <td>5.551748</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>GarageFinish</td>\n",
       "      <td>5.551748</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>GarageCond</td>\n",
       "      <td>5.551748</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>BsmtFinType2</td>\n",
       "      <td>2.604524</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>BsmtExposure</td>\n",
       "      <td>2.604524</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>BsmtQual</td>\n",
       "      <td>2.535984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>BsmtFinType1</td>\n",
       "      <td>2.535984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>BsmtCond</td>\n",
       "      <td>2.535984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MasVnrArea</td>\n",
       "      <td>0.548321</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>MasVnrType</td>\n",
       "      <td>0.548321</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Electrical</td>\n",
       "      <td>0.068540</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Missing Ratio\n",
       "PoolQC            99.520219\n",
       "MiscFeature       96.298835\n",
       "Alley             93.762851\n",
       "Fence             80.740233\n",
       "FireplaceQu       47.292666\n",
       "LotFrontage       17.751885\n",
       "GarageYrBlt        5.551748\n",
       "GarageType         5.551748\n",
       "GarageQual         5.551748\n",
       "GarageFinish       5.551748\n",
       "GarageCond         5.551748\n",
       "BsmtFinType2       2.604524\n",
       "BsmtExposure       2.604524\n",
       "BsmtQual           2.535984\n",
       "BsmtFinType1       2.535984\n",
       "BsmtCond           2.535984\n",
       "MasVnrArea         0.548321\n",
       "MasVnrType         0.548321\n",
       "Electrical         0.068540"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_data_na = (all_data.isnull().sum() / len(all_data)) * 100\n",
    "all_data_na = all_data_na.drop(all_data_na[all_data_na == 0].index).sort_values(ascending=False)[:30]\n",
    "missing_data = pd.DataFrame({'Missing Ratio' :all_data_na})\n",
    "missing_data.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "_cell_guid": "1c80610d-8f19-43c8-bd54-7d786b0dca49",
    "_execution_state": "idle",
    "_uuid": "2eb6e1361884db6a4f65afc3b158fcbe85c2392e",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Percent missing data by feature')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x864 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f, ax = plt.subplots(figsize=(15, 12))\n",
    "plt.xticks(rotation='90')\n",
    "sns.barplot(x=all_data_na.index, y=all_data_na)\n",
    "plt.xlabel('Features', fontsize=15)\n",
    "plt.ylabel('Percent of missing values', fontsize=15)\n",
    "plt.title('Percent missing data by feature', fontsize=15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "5d88502a-2484-45d7-a750-0d32075133dd",
    "_execution_state": "idle",
    "_uuid": "dbf28eddefa2c0825781a31cde823ac5ece5a1ee"
   },
   "source": [
    "**Data Correlation**\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "_cell_guid": "3fc115c7-33bb-456d-b3e5-4dd985bfbc9f",
    "_execution_state": "idle",
    "_uuid": "0f6b9912c752212f3f4bdca0b60f5fd01c12d2a0"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x29e130e5488>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x648 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Correlation map to see how features are correlated with SalePrice\n",
    "corrmat = train.corr()\n",
    "plt.subplots(figsize=(12,9))\n",
    "sns.heatmap(corrmat, vmax=0.9, square=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "cd681698-02d2-473e-bfc8-2d98a1353a18",
    "_execution_state": "idle",
    "_uuid": "41a6b40f8f03212a624f54167cad456a9f193f93"
   },
   "source": [
    "## Imputing missing values "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q3.결측치를 None으로 바꿔봅시다**  \n",
    "**그냥 바꾸지만 마시고 왜 None으로 바꾸는지 한번 생각해보세요**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "6f7d94ef-5ffb-4e1b-b1a2-a917cce3f357",
    "_execution_state": "idle",
    "_uuid": "d6d054aff0098e4e602ef618e0a4d3706d71dd66"
   },
   "source": [
    "- **PoolQC** : data description says NA means \"No  Pool\". That make sense, given the huge ratio of missing value (+99%) and majority of houses have no Pool at all in general. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<bound method IndexOpsMixin.value_counts of 0       NaN\n",
       "1       NaN\n",
       "2       NaN\n",
       "3       NaN\n",
       "4       NaN\n",
       "       ... \n",
       "1454    NaN\n",
       "1455    NaN\n",
       "1456    NaN\n",
       "1457    NaN\n",
       "1458    NaN\n",
       "Name: PoolQC, Length: 1459, dtype: object>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_data[\"PoolQC\"].value_counts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "_cell_guid": "ca8a04eb-f42b-4c26-a690-bb98c95c6118",
    "_execution_state": "idle",
    "_uuid": "1d94b062f7683d711d479e48530009040185fd4c"
   },
   "outputs": [],
   "source": [
    "all_data[\"PoolQC\"] = all_data[\"PoolQC\"].fillna(\"None\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "ea3aecb9-b2e8-4cdc-853d-999f8e2f789c",
    "_execution_state": "idle",
    "_uuid": "0516efe507d7176591f6af3d503312db5038949a"
   },
   "source": [
    "- **MiscFeature** : data description says NA means \"no misc feature\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x=\"MiscFeature\", data=all_data)\n",
    "plt.title(\"MiscFeature \")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "_cell_guid": "689863b8-4e4e-45d7-9972-8894e8defbe2",
    "_execution_state": "idle",
    "_uuid": "c311993a26d66d66ce82584a3e00d7ed56a3ea5a"
   },
   "outputs": [],
   "source": [
    "all_data[\"MiscFeature\"] = all_data[\"MiscFeature\"].fillna(\"None\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "None의 의미가  'misc feature없음'이므로 결측치이다. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "d467b9e0-0793-41e5-aaa1-d1686946ba9f",
    "_execution_state": "idle",
    "_uuid": "44cfbd9eecce5e4f2bc91a72ed095270e6010f57"
   },
   "source": [
    "- **Alley** : data description says NA means \"no alley access\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "_cell_guid": "0f822e30-09f3-45f8-b09e-0ffed4ff985c",
    "_execution_state": "idle",
    "_uuid": "35709caf5cb2b0220cd9043f76462a116de15059"
   },
   "outputs": [],
   "source": [
    "all_data[\"Alley\"] = all_data[\"Alley\"].fillna(\"None\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "0bab3349-eddb-4977-acd5-8dad4553ff7f",
    "_execution_state": "idle",
    "_uuid": "3265fc176140dbf1b641e18f2d57b91554c85771"
   },
   "source": [
    "- **Fence** : data description says NA means \"no fence\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "_cell_guid": "c0d6e4f4-df14-467f-bab2-86bfd0493a35",
    "_execution_state": "idle",
    "_uuid": "6d9d6998298a7fea5677f0916bcc9f33dc99e231"
   },
   "outputs": [],
   "source": [
    "all_data[\"Fence\"] = all_data[\"Fence\"].fillna(\"None\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "49747765-81f1-4b23-ad63-534c7fe48b7a",
    "_execution_state": "idle",
    "_uuid": "4869889d4b2b118bc6dcef239cbc27546a7b445e"
   },
   "source": [
    "- **FireplaceQu** : data description says NA means \"no fireplace\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "_cell_guid": "22b88c5b-6419-424f-9805-2c2b11e85dca",
    "_execution_state": "idle",
    "_uuid": "fb8cdb59e4ecba1f7ec78eb3b34c7209c34b1c86"
   },
   "outputs": [],
   "source": [
    "all_data[\"FireplaceQu\"] = all_data[\"FireplaceQu\"].fillna(\"None\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**A3. 위의 data의 결측치를 모두 None으로 치환\n",
    "결측치를 한번에 처리하기 위하여 None으로 통일하였습니다.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "357994d1-974a-49ee-98e4-a19bc524fcb6",
    "_execution_state": "idle",
    "_uuid": "2b00af603b7dbc9114e62c421b2b886154d31959"
   },
   "source": [
    "- **LotFrontage** : "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " **Q4. the area of each street connected to the house property most likely have a similar area to other houses in its neighborhood. 인 변수입니다. 그러면 결측치를 어떻게 바꾸는게 좋을까요?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "_cell_guid": "b8cd4872-67c6-4c82-a1a3-807d08efb658",
    "_execution_state": "idle",
    "_uuid": "56c268a7217e85301d2847ceeca410d62e781b89"
   },
   "outputs": [],
   "source": [
    "all_data[\"LotFrontage\"] = all_data.groupby(\"Neighborhood\")[\"LotFrontage\"].transform(\n",
    "    lambda x: x.mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "Transformation : 그룹 별 계산을 수행하고 생성된 객체를 반환\n",
    "\n",
    "**A4. 집과 연결된 각 거리의 지역은 그 지역의 다른 집과 비슷한 지역 일 가능성이 높기 때문에 data를 그룹화 해줍니다. Q4의 경우, data 내 NaN을 각 그룹에서 파생된 mean 값으로 치환했습니다.**\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "https://pandas.pydata.org/pandas-docs/stable/user_guide/groupby.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "c0432c37-f664-4da1-a4d8-87a7cb450bd1",
    "_execution_state": "idle",
    "_uuid": "53712c1d6c7cf2945c6ba8b537171ea6b84f085e"
   },
   "source": [
    "- **Q5. GarageType, GarageFinish, GarageQual and GarageCond** : Replacing missing data with None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "_cell_guid": "f264979e-f5e2-4aae-b313-7468570c0294",
    "_execution_state": "idle",
    "_uuid": "d556da5418394ed50d3c14bd02668714cee0fb4a"
   },
   "outputs": [],
   "source": [
    "for col in ('GarageType', 'GarageFinish', 'GarageQual', 'GarageCond'):\n",
    "    all_data[col] = all_data[col].fillna('None')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**A5. fillna를 이용해 결측치를 None으로 치환하였습니다.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "92173e82-fd12-4fa0-b7dd-f66fb7b0be3b",
    "_execution_state": "idle",
    "_uuid": "b38ea3ec1c4b8fa4b31e8f86f378798319c05a71"
   },
   "source": [
    "- **Q6. GarageYrBlt, GarageArea and GarageCars** : Replacing missing data with 0 (Since No garage = no cars in such garage.)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "_cell_guid": "ada7acfa-8bc3-4d52-9f42-28a78cefc950",
    "_execution_state": "idle",
    "_uuid": "2726cce29df2a2621c54c4c8a06eb570a7191552"
   },
   "outputs": [],
   "source": [
    "for col in ('GarageYrBlt', 'GarageArea', 'GarageCars'):\n",
    "    all_data[col] = all_data[col].fillna(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**A5. fillna를 이용해 결측치를 0로 치환하였습니다.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "b9f955c5-4e03-4835-8df7-1284120453ff",
    "_execution_state": "idle",
    "_uuid": "1729c36d95461d03679eb95e18eaac3d37aeb076"
   },
   "source": [
    "- **Q7. BsmtFinSF1, BsmtFinSF2, BsmtUnfSF, TotalBsmtSF, BsmtFullBath and BsmtHalfBath** : missing values are likely zero for having no basement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "_cell_guid": "dd9e7ef7-3cee-437a-9a33-20987e238425",
    "_execution_state": "idle",
    "_uuid": "3008a6bd07ed883ff92a83ebe390940a24010fcb"
   },
   "outputs": [],
   "source": [
    "for col in ('BsmtFinSF1', 'BsmtFinSF2', 'BsmtUnfSF','TotalBsmtSF', 'BsmtFullBath', 'BsmtHalfBath'):\n",
    "    all_data[col] = all_data[col].fillna(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "441b397e-c527-4bd5-8d3c-df6aeba72192",
    "_execution_state": "idle",
    "_uuid": "d4ed0fcea9f8bb309118ba2fb703431e386c6cbd"
   },
   "source": [
    "- **Q8. BsmtQual, BsmtCond, BsmtExposure, BsmtFinType1 and BsmtFinType2** : For all these categorical basement-related features, NaN means that there is no  basement.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "_cell_guid": "6d7aa4f6-41b9-4f27-acb0-8af6aca2c8ff",
    "_execution_state": "idle",
    "_uuid": "ee281d3c691a50795e93bb797603c697d320b286"
   },
   "outputs": [],
   "source": [
    "for col in ('BsmtQual', 'BsmtCond', 'BsmtExposure', 'BsmtFinType1', 'BsmtFinType2'):\n",
    "    all_data[col] = all_data[col].fillna(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "e7f2515f-8438-4075-9d22-0b37c1a2ef01",
    "_execution_state": "idle",
    "_uuid": "556165bc959cd6ff2354cb774341ac947d052f6a"
   },
   "source": [
    "- **Q9. MasVnrArea and MasVnrType** : NA most likely means no masonry veneer for these houses. We can fill 0 for the area and None for the type. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "_cell_guid": "63a3a030-482f-4292-a887-16880f7a2882",
    "_execution_state": "idle",
    "_uuid": "800ff604bb70023baf1d403b53b674b765d9047c"
   },
   "outputs": [],
   "source": [
    "all_data[\"MasVnrType\"] = all_data[\"MasVnrType\"].fillna(\"None\")\n",
    "all_data[\"MasVnrArea\"] = all_data[\"MasVnrArea\"].fillna(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "f2f7f337-de24-44ec-93a6-09428ec9a252",
    "_execution_state": "idle",
    "_uuid": "ded1f7ea0443654bd033eeed1361056f820a925b"
   },
   "source": [
    "- **MSZoning (The general zoning classification)** :  'RL' is by far  the most common value.  So we can fill in missing values with 'RL'  \n",
    "\n",
    "**Q10. MSZoning에서 가장 많이 존재하는 값으로 결측치를 대체하세요**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RL         1150\n",
       "RM          218\n",
       "FV           65\n",
       "RH           16\n",
       "C (all)      10\n",
       "Name: MSZoning, dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_data['MSZoning'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "RL이 최빈값입니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'RL'"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#최빈값만 return\n",
    "all_data['MSZoning'].value_counts(dropna=True).idxmax()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Int64Index([], dtype='int64')"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_data.index[all_data['MSZoning'].isnull()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "???결측치가 없낭?? 우선 최빈값으로 대치하장"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "_cell_guid": "0e4f28b5-784c-4369-92f4-bda15fa55172",
    "_execution_state": "idle",
    "_uuid": "ca93ce9b34c853c1bb11d837eaff6c2986234757"
   },
   "outputs": [],
   "source": [
    "\n",
    "most_freq = all_data['MSZoning'].value_counts(dropna=True).idxmax()\n",
    "\n",
    "all_data['MSZoning'].fillna(most_freq, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**A10. 최빈값을 return하는 함수(value_counts(dropna=True).idxmax())를 사용하여 결측치를 처리했습니다.**\n",
    "\n",
    "https://yganalyst.github.io/study/Pd_6/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "99310a4b-9dbf-4273-81b5-94bf0adf338c",
    "_execution_state": "idle",
    "_uuid": "de39073aef3db7d13894fa82285eff803d6e5c1b"
   },
   "source": [
    "- **Utilities** : For this categorical feature all records are \"AllPub\", except for one \"NoSeWa\"  and 2 NA . Since the house with 'NoSewa' is in the training set, **this feature won't help in predictive modelling**. We can then safely  remove it.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "_cell_guid": "facec65d-945f-4be1-86c8-1be011cc2bd0",
    "_execution_state": "idle",
    "_uuid": "de8ee2aca9eb3eeed7442cd9d12843cc64746708"
   },
   "outputs": [],
   "source": [
    "all_data = all_data.drop(['Utilities'], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "c6aabfcd-4a1d-4845-88a1-b2f5cb7d8901",
    "_execution_state": "idle",
    "_uuid": "3450d048089ba8ff98807c6ae3b745b0a34165f9"
   },
   "source": [
    "- **Functional** : data description says NA means typical"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "_cell_guid": "c968aa5e-c34f-442e-9388-69e074a2c84e",
    "_execution_state": "idle",
    "_uuid": "e2d331e991fa868320b5d089f61c58b54ce7cb37"
   },
   "outputs": [],
   "source": [
    "all_data[\"Functional\"] = all_data[\"Functional\"].fillna(\"Typ\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "4f2bad4d-8db8-4ac3-8991-ffa2c775a72d",
    "_execution_state": "idle",
    "_uuid": "c122f5b5596bc4ce615c6b620b7e1824a54a47b8"
   },
   "source": [
    "- **Electrical** : It has one NA value. Since this feature has mostly 'SBrkr', we can set that for the missing value.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q11.  가장 많이 존재하는 값으로 결측치를 대체하세요**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "_cell_guid": "d93eab84-7759-4201-bd7d-d450399478f7",
    "_execution_state": "idle",
    "_uuid": "d31761ee8878ce4143cf21adde149cd622c5a039"
   },
   "outputs": [],
   "source": [
    "all_data['Electrical'].fillna(all_data['Electrical'].value_counts(dropna=True).idxmax() , inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "3d089b8e-3392-4067-a5a1-4f6d4e3e8fb5",
    "_execution_state": "idle",
    "_uuid": "8e0fd54a802f1eddfec0aa4104412c19e1c266d0"
   },
   "source": [
    "- **KitchenQual**: Only one NA value, and same as Electrical, we set 'TA' (which is the most frequent)  for the missing value in KitchenQual.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "_cell_guid": "9f68e47a-5b50-436f-9e31-ff65a1cb1687",
    "_execution_state": "idle",
    "_uuid": "23071ddb69b67b1d742bb15d260421d57aed5583"
   },
   "outputs": [],
   "source": [
    "all_data['KitchenQual'].fillna(all_data['KitchenQual'].value_counts(dropna=True).idxmax() , inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "e32aec7d-2841-4bfc-a6d2-d181000af4b8",
    "_execution_state": "idle",
    "_uuid": "e8aca612daf293e7228eed9647370036d003e0fb"
   },
   "source": [
    "- **Exterior1st and Exterior2nd** : Again Both Exterior 1 & 2 have only one missing value. We will just substitute in the most common string\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "_cell_guid": "85903feb-7fdb-4911-9a6a-f29629bbf308",
    "_execution_state": "idle",
    "_uuid": "6375e9c7c145e0f2832cbe03addbce3d90cb44c4"
   },
   "outputs": [],
   "source": [
    "all_data['Exterior1st'].fillna(all_data['Exterior1st'].value_counts(dropna=True).idxmax() , inplace = True)\n",
    "all_data['Exterior2nd'].fillna(all_data['Exterior2nd'].value_counts(dropna=True).idxmax() , inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "b87d629d-6820-4f3e-80f3-efc16436b982",
    "_execution_state": "idle",
    "_uuid": "67f41f0ad648d536d70e2ccb617a4b495fb7f782"
   },
   "source": [
    "- **SaleType** : Fill in again with most frequent which is \"WD\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "_cell_guid": "aaa407ac-ad4a-4313-8530-ac96826be6e5",
    "_execution_state": "idle",
    "_uuid": "06c104efedaf57a4052c8f61c502cb6313e0beea"
   },
   "outputs": [],
   "source": [
    "all_data['SaleType'].fillna(all_data['SaleType'].value_counts(dropna=True).idxmax() , inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**A11. Q10과 동일하게 처리했습니다.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "8ddca94e-7cfa-4f25-9333-29e013c9c30d",
    "_execution_state": "idle",
    "_uuid": "8a3a35f5dc00732929143480a8a0b34a2a502c27"
   },
   "source": [
    "- **MSSubClass** : Na most likely means No building class. We can replace missing values with None\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "_cell_guid": "23dedd0c-7d5e-4870-8588-8bc4890f627b",
    "_execution_state": "idle",
    "_uuid": "9608a2eada6013e7f14e9341862cca0dca4621e9"
   },
   "outputs": [],
   "source": [
    "all_data['MSSubClass'] = all_data['MSSubClass'].fillna(\"None\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "9193270b-8e7b-4a24-b769-d6eb661a1ae7",
    "_execution_state": "idle",
    "_uuid": "465570fcce8944ee31b1709ec6954d03cbf32ff8"
   },
   "source": [
    "Is there any remaining missing value ? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "_cell_guid": "0adf05cf-ce60-4169-805c-ca776e60e85a",
    "_execution_state": "idle",
    "_uuid": "b091fa2ebef19425019e2e550410d0376b9e9fac"
   },
   "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>Missing Ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [Missing Ratio]\n",
       "Index: []"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Check remaining missing values if any \n",
    "all_data_na = (all_data.isnull().sum() / len(all_data)) * 100\n",
    "all_data_na = all_data_na.drop(all_data_na[all_data_na == 0].index).sort_values(ascending=False)\n",
    "missing_data = pd.DataFrame({'Missing Ratio' :all_data_na})\n",
    "missing_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "결측치 처리 성공쓰~"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "7996debf-c724-4365-9d63-f6104bae6140",
    "_execution_state": "idle",
    "_uuid": "915505b9c21b4dd84cc466660a68fb0d6b84cfae"
   },
   "source": [
    "### More features engeneering"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "0d5076fa-b4ab-4787-bebb-8fd445b8815a",
    "_execution_state": "idle",
    "_uuid": "1cfcb671a97068569efb4e7855da91aa30c5bbde"
   },
   "source": [
    "**Transforming some numerical variables that are really categorical**   \n",
    "\n",
    "숫자형태를 한 값일 뿐 실제 연속형도 숫자 데이터도 아니죠? 따라서 문자형식으로 바꾼후 라벨링 인코더 하겠습니다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q12. 데이터를 문자형으로 바꾸세요.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('int64')"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_data['MSSubClass'].dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "#MSSubClass=The building class\n",
    "all_data['MSSubClass'] = all_data['MSSubClass'].astype('str')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('O')"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_data['MSSubClass'].dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "str"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(all_data['MSSubClass'][0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "string으로 잘 변환되는것을 확인\\\n",
    "\n",
    "\n",
    "dtype('0') :\\\n",
    "Pandas dtype : object  \\\n",
    "Python type : str\\\n",
    "NumPy type :  string_, unicode_ \\\n",
    "Usage : Text\n",
    "                         \n",
    "\n",
    "https://stackoverflow.com/questions/37561991/what-is-dtypeo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "_cell_guid": "a52dc2f9-ca02-4024-987a-165ce630b356",
    "_execution_state": "idle",
    "_uuid": "cc7557817a4442e799e4e4c84dd1efd8bd08867a"
   },
   "outputs": [],
   "source": [
    "\n",
    "#Changing OverallCond into a categorical variable\n",
    "all_data['OverallCond'] = all_data['OverallCond'].astype('str')\n",
    "\n",
    "\n",
    "#Year and month sold are transformed into categorical features.\n",
    "all_data['YrSold'] = all_data['YrSold'].astype('str')\n",
    "all_data['MoSold'] = all_data['MoSold'].astype('str')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**A12. dtpye으로 series전체를 문자형으로 변환**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "9f80c0e7-3f3f-45c5-b111-e36f4e31e814",
    "_execution_state": "idle",
    "_uuid": "c4743ffb7fbb050edca7c77dc7cb6520577c1398"
   },
   "source": [
    "**Label Encoding some categorical variables that may contain information in their ordering set** "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q13. LabelEncoder로 해당 범주형 데이터를 바꿔보세요..**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "_cell_guid": "81c97efb-4f76-4e87-861a-10a60ab5c84b",
    "_execution_state": "idle",
    "_uuid": "fdb5ddf0a49a3c6df303c569c9f3509c79ac8b61"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape all_data: (1459, 78)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import LabelEncoder\n",
    "cols = ('FireplaceQu', 'BsmtQual', 'BsmtCond', 'GarageQual', 'GarageCond', \n",
    "        'ExterQual', 'ExterCond','HeatingQC', 'PoolQC', 'KitchenQual', 'BsmtFinType1', \n",
    "        'BsmtFinType2', 'Functional', 'Fence', 'BsmtExposure', 'GarageFinish', 'LandSlope',\n",
    "        'LotShape', 'PavedDrive', 'Street', 'Alley', 'CentralAir', 'MSSubClass', 'OverallCond', \n",
    "        'YrSold', 'MoSold')\n",
    "\n",
    "for col in cols:\n",
    "    lbl = LabelEncoder() \n",
    "    lbl.fit(list(all_data[col].values)) \n",
    "    all_data[col] = lbl.transform(list(all_data[col].values))\n",
    "\n",
    "\n",
    "# shape        \n",
    "print('Shape all_data: {}'.format(all_data.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "a4879ef7-ab0d-4955-bc48-7ebcfa04b3bd",
    "_execution_state": "idle",
    "_uuid": "9976d6288bc183d443fbccc2bde439d5bc3a87b1"
   },
   "source": [
    "**Adding one more important feature**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "_cell_guid": "fc1a8f1a-f003-4538-8e60-d819f46362a3",
    "_execution_state": "idle",
    "_uuid": "208f8d22188786227fff4a978dc3b11b4e1ffd90"
   },
   "outputs": [],
   "source": [
    "# Adding total sqfootage feature \n",
    "all_data['TotalSF'] = all_data['TotalBsmtSF'] + all_data['1stFlrSF'] + all_data['2ndFlrSF']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "91c73aad-82d1-4301-b540-b2f69dc13902",
    "_execution_state": "idle",
    "_uuid": "aa36d6e3253e354b46d9c9c6f2e8a4089c76be16"
   },
   "source": [
    "**Skewed features**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "_cell_guid": "c5972a73-7e86-4164-a9d6-58432dae1933",
    "_execution_state": "idle",
    "_uuid": "53c471c7008c66590f257e70866f8a3037813f13"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Skew in numerical features: \n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Skew</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>MiscVal</td>\n",
       "      <td>24.443278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>PoolArea</td>\n",
       "      <td>14.807992</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>LotArea</td>\n",
       "      <td>12.290047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3SsnPorch</td>\n",
       "      <td>10.290132</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>LowQualFinSF</td>\n",
       "      <td>8.998885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>LandSlope</td>\n",
       "      <td>4.806884</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>KitchenAbvGr</td>\n",
       "      <td>4.482026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>BsmtFinSF2</td>\n",
       "      <td>4.249219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>ScreenPorch</td>\n",
       "      <td>4.116334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>BsmtHalfBath</td>\n",
       "      <td>4.097541</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   Skew\n",
       "MiscVal       24.443278\n",
       "PoolArea      14.807992\n",
       "LotArea       12.290047\n",
       "3SsnPorch     10.290132\n",
       "LowQualFinSF   8.998885\n",
       "LandSlope      4.806884\n",
       "KitchenAbvGr   4.482026\n",
       "BsmtFinSF2     4.249219\n",
       "ScreenPorch    4.116334\n",
       "BsmtHalfBath   4.097541"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "numeric_feats = all_data.dtypes[all_data.dtypes != \"object\"].index\n",
    "\n",
    "skewed_feats = all_data[numeric_feats].apply(lambda x: skew(x.dropna())).sort_values(ascending=False)\n",
    "print(\"\\nSkew in numerical features: \\n\")\n",
    "skewness = pd.DataFrame({'Skew' :skewed_feats})\n",
    "skewness.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "9f110087-b707-4073-a1df-0a0a9d6ccbd3",
    "_execution_state": "idle",
    "_uuid": "cf63bdc9f4f80d81f1bfa14f89d65ff104d45e5b"
   },
   "source": [
    "**Box Cox Transformation of (highly) skewed features**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "d1d18243-42d8-4a21-808d-784c21e53973",
    "_execution_state": "idle",
    "_uuid": "eab0b4c0a85ae2fbe1bdeea0eedd113904ef3eb1"
   },
   "source": [
    "We use the scipy  function boxcox1p which computes the Box-Cox transformation of **\\\\(1 + x\\\\)**. \n",
    "\n",
    "Note that setting \\\\( \\lambda = 0 \\\\) is equivalent to log1p used above for the target variable.  \n",
    "\n",
    "See [this page][1] for more details on Box Cox Transformation as well as [the scipy function's page][2]\n",
    "[1]: http://onlinestatbook.com/2/transformations/box-cox.html\n",
    "[2]: https://docs.scipy.org/doc/scipy-0.19.0/reference/generated/scipy.special.boxcox1p.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "_cell_guid": "d8ebce87-c55d-46c6-8f06-8b34116d7370",
    "_execution_state": "idle",
    "_uuid": "969fdff338ef46f064d8f855782c96d322a264b1"
   },
   "outputs": [],
   "source": [
    "skewness = skewness[abs(skewness) > 0.75]\n",
    "\n",
    "\n",
    "from scipy.special import boxcox1p\n",
    "skewed_features = skewness.index\n",
    "lam = 0.15\n",
    "for feat in skewed_features:\n",
    "    all_data[feat] = boxcox1p(all_data[feat], lam)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "39639caf-31a4-4401-a663-0ba9536b39bf",
    "_execution_state": "idle",
    "_uuid": "5a13a6e2a3e48975de9129d1593bd38df44a1069"
   },
   "source": [
    "**Getting dummy categorical features**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q14. 더미화하세요.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "_cell_guid": "c8e63516-e4e2-4f36-a60e-1c8316392c60",
    "_execution_state": "idle",
    "_uuid": "acd44e283867425257ffd1fb2f4893cdbff43f67"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1459, 221)\n"
     ]
    }
   ],
   "source": [
    "all_data = pd.get_dummies(all_data)\n",
    "print(all_data.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**A14. get_dummies를 사용하여 one_hot encoding된 dummy변수로 변환"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "243cf047-c2ba-4ae5-a531-22ef9b7cfbfe",
    "_execution_state": "idle",
    "_uuid": "fe9d78c7e37142ee8089826eca3065e0fa5803c1"
   },
   "source": [
    "Getting the new train and test sets. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "_cell_guid": "0a75646f-1974-40ad-a085-ff7bc08454a5",
    "_execution_state": "idle",
    "_uuid": "89e464095544a53177d5a009b914ba4c660072a7"
   },
   "outputs": [],
   "source": [
    "train = all_data[:ntrain]\n",
    "test = all_data[ntrain:]   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "461af83d-a928-4645-8512-5e4dbcaf7be0",
    "_execution_state": "idle",
    "_uuid": "10aab4cee97832560e2627a490e01e80c0ffb814"
   },
   "source": [
    "#Modelling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <span style=\"color:blue\"> 이 아래부터의 과정은 앙상블을 정말 모르겠다.(코드) 그런 분들을 위해 참조용으로 있을 뿐. 알아서 하셔도 무방합니다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "811925a6-341f-4cae-89c9-00983868a6b2",
    "_execution_state": "idle",
    "_uuid": "be4e4b315682b26359eba1ba3d65022aca9501e1"
   },
   "source": [
    "**Import librairies**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ! pip install xgboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ! pip install lightgbm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "_cell_guid": "135e8ac5-ce46-4a5f-b205-13f827ef33b8",
    "_execution_state": "idle",
    "_uuid": "fc664fbe27561a3697d0210921107b0e14b7d211"
   },
   "outputs": [],
   "source": [
    "from sklearn.linear_model import ElasticNet, Lasso,  BayesianRidge, LassoLarsIC\n",
    "from sklearn.ensemble import RandomForestRegressor,  GradientBoostingRegressor\n",
    "from sklearn.kernel_ridge import KernelRidge\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.preprocessing import RobustScaler\n",
    "from sklearn.base import BaseEstimator, TransformerMixin, RegressorMixin, clone\n",
    "from sklearn.model_selection import KFold, cross_val_score, train_test_split\n",
    "from sklearn.metrics import mean_squared_error\n",
    "import xgboost as xgb\n",
    "import lightgbm as lgb"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "7aa6ebb9-27a7-4bd3-a7b2-4ddc6a0abbed",
    "_execution_state": "idle",
    "_uuid": "056b657c8f0de30d4708c600eabbb33684c64479"
   },
   "source": [
    "**Define a cross validation strategy**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "4a2b5181-44f2-4c74-b482-aae0f5afc25a",
    "_execution_state": "idle",
    "_uuid": "dc0d7a3013f349988b3f2c84a6c130d6ad350170"
   },
   "source": [
    "We use the **cross_val_score** function of Sklearn. However this function has not a shuffle attribut, we add then one line of code,  in order to shuffle the dataset  prior to cross-validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "_cell_guid": "f396260b-e182-4a87-9a2a-b92b9375ea6f",
    "_execution_state": "idle",
    "_uuid": "5c12551d092a6c5cf32d86398b054da7af3047b8"
   },
   "outputs": [],
   "source": [
    "#Validation function\n",
    "n_folds = 5\n",
    "\n",
    "def rmsle_cv(model):\n",
    "    kf = KFold(n_folds, shuffle=True, random_state=42).get_n_splits(train.values)\n",
    "    rmse= np.sqrt(-cross_val_score(model, train.values, y_train, scoring=\"neg_mean_squared_error\", cv = kf))\n",
    "    return(rmse)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "42e1565e-77a1-41a7-ac31-893e405d34ad",
    "_execution_state": "busy",
    "_uuid": "643ae2c4a88576ebbd55824ce8e654486087a6e0"
   },
   "source": [
    "##Base models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "578f088d-1a84-41cb-b945-ec64800f2308",
    "_execution_state": "idle",
    "_uuid": "370125198a0cbbc9336cbf179f00a2ebb02cb063"
   },
   "source": [
    "-  **LASSO  Regression**  : \n",
    "\n",
    "This model may be very sensitive to outliers. So we need to made it more robust on them. For that we use the sklearn's  **Robustscaler()**  method on pipeline "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "_cell_guid": "03f45cb7-0a40-45ea-94e8-64fd7ff1e8f6",
    "_execution_state": "idle",
    "_uuid": "2a50c954cb771d350c3092c3658486ba4d22aba5"
   },
   "outputs": [],
   "source": [
    "lasso = make_pipeline(RobustScaler(), Lasso(alpha =0.0005, random_state=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "2c826f7b-ac66-421c-a7ae-29dfdd765bdb",
    "_execution_state": "idle",
    "_uuid": "30e9756cf63991715b48e8c53bc57906fc76f380"
   },
   "source": [
    "- **Elastic Net Regression** :\n",
    "\n",
    "again made robust to outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "_cell_guid": "e635cc7e-caeb-4f8b-ae78-c41f8eb0be59",
    "_execution_state": "idle",
    "_uuid": "b614cf1bdee86a3b1cbdde05298f9f7ae023799b"
   },
   "outputs": [],
   "source": [
    "ENet = make_pipeline(RobustScaler(), ElasticNet(alpha=0.0005, l1_ratio=.9, random_state=3))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "7aae5316-4e32-4203-bff5-3b38c1f657c3",
    "_execution_state": "idle",
    "_uuid": "0775061bb477242f1332a048778e879ca540a216"
   },
   "source": [
    "- **Kernel Ridge Regression** :"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "_cell_guid": "805343d9-0af6-43a2-a351-c0b25c62fcf0",
    "_execution_state": "idle",
    "_uuid": "3199c83513d93407c818ce1ed43c6c52e7f5a8c6"
   },
   "outputs": [],
   "source": [
    "KRR = KernelRidge(alpha=0.6, kernel='polynomial', degree=2, coef0=2.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "5a66c27c-be80-4ec0-8953-eaeb2a7dd2e7",
    "_execution_state": "idle",
    "_uuid": "14b60a7e4296cccb39042c9c625a1480d59a01c1"
   },
   "source": [
    "- **Gradient Boosting Regression** :\n",
    "\n",
    "With **huber**  loss that makes it robust to outliers\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "_cell_guid": "af13332c-fd37-40bb-a078-6bad6caaa2ab",
    "_execution_state": "idle",
    "_uuid": "9a983f0f62a0dde7689b20a8e52022bb189478b4"
   },
   "outputs": [],
   "source": [
    "GBoost = GradientBoostingRegressor(n_estimators=3000, learning_rate=0.05,\n",
    "                                   max_depth=4, max_features='sqrt',\n",
    "                                   min_samples_leaf=15, min_samples_split=10, \n",
    "                                   loss='huber', random_state =5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "d44ac87e-bf01-440b-ab22-b2868eb6ae48",
    "_execution_state": "idle",
    "_uuid": "53d7991f7dd03fcd7fb5ab1ec26fcd0614d002d3"
   },
   "source": [
    "- **XGBoost** :"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "_cell_guid": "ed738a4c-c246-443c-a3c1-39df25f988b7",
    "_execution_state": "idle",
    "_uuid": "57c24b596ceb46d6f32ebf9501d672d7e469c15b"
   },
   "outputs": [],
   "source": [
    "model_xgb = xgb.XGBRegressor(colsample_bytree=0.4603, gamma=0.0468, \n",
    "                             learning_rate=0.05, max_depth=3, \n",
    "                             min_child_weight=1.7817, n_estimators=2200,\n",
    "                             reg_alpha=0.4640, reg_lambda=0.8571,\n",
    "                             subsample=0.5213, silent=1,\n",
    "                             random_state =7, nthread = -1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "a43ca74d-093c-4a56-a76c-b3223bf82fbc",
    "_execution_state": "idle",
    "_uuid": "460f3ccf7d5c33ea9f8a826bbf056d759e7b5119"
   },
   "source": [
    "- **LightGBM** :"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "_cell_guid": "dd84d7db-3f83-4e4e-b02f-7632ca5ee4ac",
    "_execution_state": "idle",
    "_uuid": "4c94cf90f0ef0d350c5e66f3bd397865bfcc61ae"
   },
   "outputs": [],
   "source": [
    "model_lgb = lgb.LGBMRegressor(objective='regression',num_leaves=5,\n",
    "                              learning_rate=0.05, n_estimators=720,\n",
    "                              max_bin = 55, bagging_fraction = 0.8,\n",
    "                              bagging_freq = 5, feature_fraction = 0.2319,\n",
    "                              feature_fraction_seed=9, bagging_seed=9,\n",
    "                              min_data_in_leaf =6, min_sum_hessian_in_leaf = 11)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "9e1eff6a-e937-45e4-96ef-41593e31e1bb",
    "_execution_state": "idle",
    "_uuid": "71bce529300e2f3d9f9f475d01bd7001258dbede"
   },
   "source": [
    "###Base models scores"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "84ddecce-7671-44e5-919d-97348bf413f4",
    "_execution_state": "idle",
    "_uuid": "cae4987b8ec89e90a90d7826c4ec98d315cac00b"
   },
   "source": [
    "cross-validation rmsle error로 모델 성능을 확인해 봅시다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "_cell_guid": "2d0cc958-1654-425c-90ed-1ceb9edd7186",
    "_execution_state": "idle",
    "_uuid": "7d994349237b9304b0d17719e1af077e69288229"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Lasso score: 0.1171 (0.0123)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "score = rmsle_cv(lasso)\n",
    "print(\"\\nLasso score: {:.4f} ({:.4f})\\n\".format(score.mean(), score.std()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "_cell_guid": "7cf6faaf-d69a-4268-b192-a9e60d207c28",
    "_execution_state": "idle",
    "_uuid": "b6d299b9d4a0cdb23ddd8459b3935da2948016d6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ElasticNet score: 0.1175 (0.0124)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "score = rmsle_cv(ENet)\n",
    "print(\"ElasticNet score: {:.4f} ({:.4f})\\n\".format(score.mean(), score.std()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "_cell_guid": "a1195106-2170-47f2-86a7-c4f3be683aa8",
    "_execution_state": "idle",
    "_uuid": "437dc093e88d661a369539520af1b4c37d1a0c1a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Kernel Ridge score: 0.1250 (0.0140)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "score = rmsle_cv(KRR)\n",
    "print(\"Kernel Ridge score: {:.4f} ({:.4f})\\n\".format(score.mean(), score.std()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "_cell_guid": "43dd152f-7c49-41b6-8f8e-a5864b1e2a71",
    "_execution_state": "idle",
    "_uuid": "e9d8c4bd191f77d8d275f53c0c1a6cf344151294"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gradient Boosting score: 0.1262 (0.0119)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "score = rmsle_cv(GBoost)\n",
    "print(\"Gradient Boosting score: {:.4f} ({:.4f})\\n\".format(score.mean(), score.std()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "_cell_guid": "30738ecc-39f8-44ed-9f42-68518beb7e6a",
    "_execution_state": "idle",
    "_uuid": "5f52ccf39d01165e61a7c6be8b788be4e58e286b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Xgboost score: 0.1246 (0.0085)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "score = rmsle_cv(model_xgb)\n",
    "print(\"Xgboost score: {:.4f} ({:.4f})\\n\".format(score.mean(), score.std()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "_cell_guid": "41e0eab9-630d-48d3-905b-e4663aad2262",
    "_execution_state": "idle",
    "_uuid": "5cd5377ee097fbc6fd14b42b4ea654221b097e59"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LGBM score: 0.1225 (0.0120)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "score = rmsle_cv(model_lgb)\n",
    "print(\"LGBM score: {:.4f} ({:.4f})\\n\" .format(score.mean(), score.std()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "1114bc71-7eb5-4a7c-97a1-42a69cc21130",
    "_execution_state": "idle",
    "_uuid": "06d3adc16585b54a85113882975297c67672ea07"
   },
   "source": [
    "##Stacking  models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "56746043-0d77-4687-a8f2-ae494efae3a8",
    "_execution_state": "idle",
    "_uuid": "2410d2172ddc108475db49214c52c21e66aeee59"
   },
   "source": [
    "###Simplest Stacking approach : Averaging base models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "d0145496-896a-44e3-b01b-e12546328f06",
    "_execution_state": "idle",
    "_uuid": "5ecc887f1ab4001c872862cecf3a0b350ac51a23"
   },
   "source": [
    "**Averaged base models class**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "_cell_guid": "49e44ad6-8dc4-4a67-8079-adbac934fec4",
    "_execution_state": "idle",
    "_uuid": "ff3ee5889bcac40847909c3a71285d2b8f9d431f"
   },
   "outputs": [],
   "source": [
    "class AveragingModels(BaseEstimator, RegressorMixin, TransformerMixin):\n",
    "    def __init__(self, models):\n",
    "        self.models = models\n",
    "        \n",
    "    # we define clones of the original models to fit the data in\n",
    "    def fit(self, X, y):\n",
    "        self.models_ = [clone(x) for x in self.models]\n",
    "        \n",
    "        # Train cloned base models\n",
    "        for model in self.models_:\n",
    "            model.fit(X, y)\n",
    "\n",
    "        return self\n",
    "    \n",
    "    #Now we do the predictions for cloned models and average them\n",
    "    def predict(self, X):\n",
    "        predictions = np.column_stack([\n",
    "            model.predict(X) for model in self.models_\n",
    "        ])\n",
    "        return np.mean(predictions, axis=1)   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "825eb99f-d509-4203-b0c6-4ff77f696322",
    "_execution_state": "idle",
    "_uuid": "f05bf966ea7a7b5e6f8ca5d641ebd11281d54d0d"
   },
   "source": [
    "**Averaged base models score**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "_cell_guid": "d480916f-89e7-4bcc-9b9d-b54492591654",
    "_execution_state": "idle",
    "_uuid": "81ce9e148b7e735f465b4b6508511dea44fbf791"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Averaged base models score: 0.1146 (0.0123)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "averaged_models = AveragingModels(models = (ENet, GBoost, KRR, lasso))\n",
    "score = rmsle_cv(averaged_models)\n",
    "print(\" Averaged base models score: {:.4f} ({:.4f})\\n\".format(score.mean(), score.std()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "387761c7-9dc5-41aa-8cda-5315b6a72fbf",
    "_execution_state": "idle",
    "_uuid": "01b68302f0ec3af42a70794bc339bf5956ab2569"
   },
   "source": [
    "###Less simple Stacking : Adding a Meta-model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "cb18e314-968d-4765-942a-5706d0f4f815",
    "_execution_state": "idle",
    "_uuid": "69f216ce13eb61f0d07403986a2d38b11e18ae6a"
   },
   "source": [
    "In this approach, we add a meta-model on averaged base models and use the out-of-folds predictions of these base models to train our meta-model. \n",
    "\n",
    "The procedure, for the training part, may be described as follows:\n",
    "\n",
    "\n",
    "1. Split the total training set into two disjoint sets (here **train** and .**holdout** )\n",
    "\n",
    "2. Train several base models on the first part (**train**)\n",
    "\n",
    "3. Test these base models on the second part (**holdout**)\n",
    "\n",
    "4. Use the predictions from 3)  (called  out-of-folds predictions) as the inputs, and the correct responses (target variable) as the outputs  to train a higher level learner called **meta-model**.\n",
    "\n",
    "The first three steps are done iteratively . If we take for example a 5-fold stacking , we first split the training data into 5 folds. Then we will do 5 iterations. In each iteration,  we train every base model on 4 folds and predict on the remaining fold (holdout fold). \n",
    "\n",
    "So, we will be sure, after 5 iterations , that the entire data is used to get out-of-folds predictions that we will then use as \n",
    "new feature to train our meta-model in the step 4.\n",
    "\n",
    "For the prediction part , We average the predictions of  all base models on the test data  and used them as **meta-features**  on which, the final prediction is done with the meta-model.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "7a1d4b95-ad19-4522-8459-99f0839a49bb",
    "_uuid": "5e232cc0edea67af1fc672ee07a93c435e7e2f98"
   },
   "source": [
    "![Faron](http://i.imgur.com/QBuDOjs.jpg)\n",
    "\n",
    "(Image taken from [Faron](https://www.kaggle.com/getting-started/18153#post103381))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "b9ca03ee-0377-4313-a236-59d3d972cac3",
    "_uuid": "a46c5bab3855a4728c52765379c1674de3123a25"
   },
   "source": [
    "On this gif, the base models are algorithms 0, 1, 2 and the meta-model is algorithm 3. The entire training dataset is \n",
    "A+B (target variable y known) that we can split into train part (A) and holdout part (B). And the test dataset is C. \n",
    "\n",
    "B1 (which is the prediction from the holdout part)  is the new feature used to train the meta-model 3 and C1 (which\n",
    "is the prediction  from the test dataset) is the meta-feature on which the final prediction is done. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "bd10661e-6eec-4789-83fa-d55b77619252",
    "_execution_state": "idle",
    "_uuid": "5374a729325ac38423ff82891f1cc887f14ba317"
   },
   "source": [
    "**Stacking averaged Models Class**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "_cell_guid": "03326750-2442-4e14-8774-6e2ce9330173",
    "_execution_state": "idle",
    "_uuid": "9115cf7180ba9491bd0a2c5bd566e18238c9de80"
   },
   "outputs": [],
   "source": [
    "class StackingAveragedModels(BaseEstimator, RegressorMixin, TransformerMixin):\n",
    "    def __init__(self, base_models, meta_model, n_folds=5):\n",
    "        self.base_models = base_models\n",
    "        self.meta_model = meta_model\n",
    "        self.n_folds = n_folds\n",
    "   \n",
    "    # We again fit the data on clones of the original models\n",
    "    def fit(self, X, y):\n",
    "        self.base_models_ = [list() for x in self.base_models]\n",
    "        self.meta_model_ = clone(self.meta_model)\n",
    "        kfold = KFold(n_splits=self.n_folds, shuffle=True, random_state=156)\n",
    "        \n",
    "        # Train cloned base models then create out-of-fold predictions\n",
    "        # that are needed to train the cloned meta-model\n",
    "        out_of_fold_predictions = np.zeros((X.shape[0], len(self.base_models)))\n",
    "        for i, model in enumerate(self.base_models):\n",
    "            for train_index, holdout_index in kfold.split(X, y):\n",
    "                instance = clone(model)\n",
    "                self.base_models_[i].append(instance)\n",
    "                instance.fit(X[train_index], y[train_index])\n",
    "                y_pred = instance.predict(X[holdout_index])\n",
    "                out_of_fold_predictions[holdout_index, i] = y_pred\n",
    "                \n",
    "        # Now train the cloned  meta-model using the out-of-fold predictions as new feature\n",
    "        self.meta_model_.fit(out_of_fold_predictions, y)\n",
    "        return self\n",
    "   \n",
    "    #Do the predictions of all base models on the test data and use the averaged predictions as \n",
    "    #meta-features for the final prediction which is done by the meta-model \n",
    "    def predict(self, X):\n",
    "        meta_features = np.column_stack([\n",
    "            np.column_stack([model.predict(X) for model in base_models]).mean(axis=1)\n",
    "            for base_models in self.base_models_ ])\n",
    "        return self.meta_model_.predict(meta_features)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "da4c9354-b5c2-4994-8ffd-550416a5c4db",
    "_execution_state": "idle",
    "_uuid": "5fdbcf5c678b260adf80cf39b0f3bb63a26213e1"
   },
   "source": [
    "**Stacking Averaged models Score**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {
    "_cell_guid": "4db03a27-e9fb-484d-bbfe-2058f16dce77",
    "_execution_state": "idle",
    "_uuid": "f2c78b5950097660d3f8b84bade8d8dbdc3964f2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stacking Averaged models score: 0.1128 (0.0127)\n"
     ]
    }
   ],
   "source": [
    "stacked_averaged_models_l = StackingAveragedModels(base_models = (ENet, GBoost, KRR),\n",
    "                                                 meta_model = lasso)\n",
    "\n",
    "\n",
    "score = rmsle_cv(stacked_averaged_models_l)                 \n",
    "print(\"Stacking Averaged models score: {:.4f} ({:.4f})\".format(score.mean(), score.std()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stacking Averaged models score: 0.1128 (0.0127)\n"
     ]
    }
   ],
   "source": [
    "stacked_averaged_models_E = StackingAveragedModels(base_models = (lasso, GBoost, KRR),\n",
    "                                                 meta_model = ENet)\n",
    "\n",
    "\n",
    "score = rmsle_cv(stacked_averaged_models_E)                 \n",
    "print(\"Stacking Averaged models score: {:.4f} ({:.4f})\".format(score.mean(), score.std()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stacking Averaged models score: 0.1290 (0.0171)\n"
     ]
    }
   ],
   "source": [
    "stacked_averaged_models_G = StackingAveragedModels(base_models = (ENet, lasso, KRR),\n",
    "                                                 meta_model = GBoost)\n",
    "\n",
    "\n",
    "score = rmsle_cv(stacked_averaged_models_G)                 \n",
    "print(\"Stacking Averaged models score: {:.4f} ({:.4f})\".format(score.mean(), score.std()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stacking Averaged models score: 0.1126 (0.0125)\n"
     ]
    }
   ],
   "source": [
    "stacked_averaged_models_K = StackingAveragedModels(base_models = (ENet, GBoost, lasso),\n",
    "                                                 meta_model = KRR)\n",
    "\n",
    "\n",
    "score = rmsle_cv(stacked_averaged_models_K)                 \n",
    "print(\"Stacking Averaged models score: {:.4f} ({:.4f})\".format(score.mean(), score.std()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**score가 가장 낮은 \n",
    "base_models = (ENet, GBoost, KRR),meta_model = lasso 선택**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "_cell_guid": "232c3959-c6e1-4535-8ad4-62892edc3f06",
    "_execution_state": "idle",
    "_uuid": "07f9ef433905b61a08a36790254d6a34661f0653"
   },
   "outputs": [],
   "source": [
    "def rmsle(y, y_pred):\n",
    "    return np.sqrt(mean_squared_error(y, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "999a8cc6-5083-4fca-bc90-616ac2f3ef8b",
    "_execution_state": "idle",
    "_uuid": "b7b74b70e6514b7623bc67cfec2b4f5d37c98707"
   },
   "source": [
    "### Final Training and Prediction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "717b4b02-8bcf-4df3-8994-f6a113110115",
    "_execution_state": "idle",
    "_uuid": "115d9e90a84c33213f0f0de7d86b6098f29ca7d8"
   },
   "source": [
    "**StackedRegressor:**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#같은 방식으로 여러 모델을 학습하고 예측값을 뺴보세요."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0724130536281536\n"
     ]
    }
   ],
   "source": [
    "stacked_averaged_models_l.fit(train.values, y_train)\n",
    "stacked_train_pred = stacked_averaged_models_l.predict(train.values)\n",
    "stacked_pred = stacked_averaged_models_l.predict(test.values)\n",
    "print(rmsle(y_train, stacked_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.09903577722644424\n"
     ]
    }
   ],
   "source": [
    "lasso.fit(train, y_train)\n",
    "lasso_train_pred = lasso.predict(train)\n",
    "lasso_pred = lasso.predict(test)\n",
    "print(rmsle(y_train, lasso_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0982620594465728\n"
     ]
    }
   ],
   "source": [
    "ENet.fit(train, y_train)\n",
    "EN_train_pred = ENet.predict(train)\n",
    "EN_pred = ENet.predict(test)\n",
    "print(rmsle(y_train, EN_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.08092327032291538\n"
     ]
    }
   ],
   "source": [
    "KRR.fit(train, y_train)\n",
    "KRR_train_pred = KRR.predict(train)\n",
    "KRR_pred = KRR.predict(test)\n",
    "print(rmsle(y_train, KRR_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.08007218728783401\n"
     ]
    }
   ],
   "source": [
    "model_xgb.fit(train, y_train)\n",
    "xgb_train_pred = model_xgb.predict(train)\n",
    "# xgb_pred = np.expm1(model_xgb.predict(test))\n",
    "xgb_pred = model_xgb.predict(test)\n",
    "print(rmsle(y_train, xgb_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.061940192068503146\n"
     ]
    }
   ],
   "source": [
    "model_lgb.fit(train, y_train)\n",
    "lgb_train_pred = model_lgb.predict(train)\n",
    "lgb_pred = model_lgb.predict(test.values)\n",
    "print(rmsle(y_train, lgb_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.048980281442959586\n"
     ]
    }
   ],
   "source": [
    "GBoost.fit(train, y_train)\n",
    "GBoost_train_pred = GBoost.predict(train)\n",
    "GBoost_pred = GBoost.predict(test.values)\n",
    "print(rmsle(y_train, GBoost_train_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "rmse가 낮은 순서 :\n",
    "1. GBoost\n",
    "2. model_lgb\n",
    "3. stacked_train_pred\n",
    "\n",
    "이 세가지 model들의 조함으로 mse값이 가장 작은 parameter를 선택한다.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 278,
   "metadata": {},
   "outputs": [],
   "source": [
    "rmse_list = []\n",
    "idx_i = []\n",
    "idx_j = []\n",
    "idx_k = []\n",
    "\n",
    "def myrange(start, end, step):\n",
    "    r = start\n",
    "    while(r<end):\n",
    "        yield r\n",
    "        r += step\n",
    "\n",
    "        \n",
    "\n",
    "for i in myrange(0, 1, 0.01):\n",
    "    for j in myrange(0, 1 - i, 0.01):\n",
    "        k = 1 - i -j\n",
    "        rmse_list.append(rmsle(y_train,stacked_train_pred*i + xgb_train_pred*j + lgb_train_pred*k))\n",
    "        idx_i.append(i)\n",
    "        idx_j.append(j)\n",
    "        idx_k.append(k)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 279,
   "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>rmse</th>\n",
       "      <th>i</th>\n",
       "      <th>j</th>\n",
       "      <th>k</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.061940</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.062038</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.062138</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.062240</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.062345</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5045</td>\n",
       "      <td>0.072014</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5046</td>\n",
       "      <td>0.072165</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5047</td>\n",
       "      <td>0.072047</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5048</td>\n",
       "      <td>0.072196</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5049</td>\n",
       "      <td>0.072229</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5050 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          rmse     i     j     k\n",
       "0     0.061940  0.00  0.00  1.00\n",
       "1     0.062038  0.00  0.01  0.99\n",
       "2     0.062138  0.00  0.02  0.98\n",
       "3     0.062240  0.00  0.03  0.97\n",
       "4     0.062345  0.00  0.04  0.96\n",
       "...        ...   ...   ...   ...\n",
       "5045  0.072014  0.97  0.01  0.02\n",
       "5046  0.072165  0.97  0.02  0.01\n",
       "5047  0.072047  0.98  0.00  0.02\n",
       "5048  0.072196  0.98  0.01  0.01\n",
       "5049  0.072229  0.99  0.00  0.01\n",
       "\n",
       "[5050 rows x 4 columns]"
      ]
     },
     "execution_count": 279,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dic = {'rmse' : rmse_list, 'i' : idx_i, 'j' : idx_j, 'k' : idx_k}\n",
    "df_idx =pd.DataFrame(dic)\n",
    "df_idx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.061940192068503146"
      ]
     },
     "execution_count": 280,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_idx['rmse'].min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Int64Index([], dtype='int64')"
      ]
     },
     "execution_count": 281,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_idx.index[df_idx['rmse'] == 0.048436886206649106 ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "rmse    0.071976\n",
       "i       0.830000\n",
       "j       0.160000\n",
       "k       0.010000\n",
       "Name: 4913, dtype: float64"
      ]
     },
     "execution_count": 282,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_idx.loc[4913]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "최소의 rmse는 0.4633이고, 그때의 weight 비율은\\\n",
    "\n",
    "**GBoost_train_pred x 0.83\\\n",
    "lgb_train_pred x 0.16\\\n",
    "stacked_train_pred x 0.01**\n",
    "이다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "metadata": {
    "_cell_guid": "619452b2-c395-48fe-81ab-d6b1d355236b",
    "_execution_state": "idle",
    "_uuid": "07500cf506f6a90c6439c2dabf81ab966cf1c792"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSLE score on train data:\n",
      "0.04843688620664907\n"
     ]
    }
   ],
   "source": [
    "print('RMSLE score on train data:')\n",
    "print(rmsle(y_train,GBoost_train_pred*0.83 +\n",
    "               stacked_train_pred*0.01 + lgb_train_pred*0.16 ))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**rmse가 최소인 두개의 model로 해보자**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 208,
   "metadata": {},
   "outputs": [],
   "source": [
    "rmse_list = []\n",
    "idx_i = []\n",
    "idx_j = []\n",
    "\n",
    "def myrange(start, end, step):\n",
    "    r = start\n",
    "    while(r<end):\n",
    "        yield r\n",
    "        r += step\n",
    "\n",
    "        \n",
    "\n",
    "for i in myrange(0, 1, 0.01):\n",
    "    j = 1 - i\n",
    "    rmse_list.append(rmsle(y_train,GBoost_train_pred*i + lgb_train_pred*j))\n",
    "    idx_i.append(i)\n",
    "    idx_j.append(j)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 209,
   "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>rmse</th>\n",
       "      <th>i</th>\n",
       "      <th>j</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.061940</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.061649</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.061359</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.061072</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.060787</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>95</td>\n",
       "      <td>0.048653</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>96</td>\n",
       "      <td>0.048709</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>97</td>\n",
       "      <td>0.048770</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>98</td>\n",
       "      <td>0.048836</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>99</td>\n",
       "      <td>0.048906</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        rmse     i     j\n",
       "0   0.061940  0.00  1.00\n",
       "1   0.061649  0.01  0.99\n",
       "2   0.061359  0.02  0.98\n",
       "3   0.061072  0.03  0.97\n",
       "4   0.060787  0.04  0.96\n",
       "..       ...   ...   ...\n",
       "95  0.048653  0.95  0.05\n",
       "96  0.048709  0.96  0.04\n",
       "97  0.048770  0.97  0.03\n",
       "98  0.048836  0.98  0.02\n",
       "99  0.048906  0.99  0.01\n",
       "\n",
       "[100 rows x 3 columns]"
      ]
     },
     "execution_count": 209,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dic = {'rmse' : rmse_list, 'i' : idx_i, 'j' : idx_j}\n",
    "df_idx =pd.DataFrame(dic)\n",
    "df_idx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.048320739496795044"
      ]
     },
     "execution_count": 210,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_idx['rmse'].min()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "3개의 model로 했을 때 보다 rmse가 아주 조금 낮다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Int64Index([83], dtype='int64')"
      ]
     },
     "execution_count": 206,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_idx.index[df_idx['rmse'] == 0.048320739496795044 ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "rmse    0.048321\n",
       "i       0.830000\n",
       "j       0.170000\n",
       "Name: 83, dtype: float64"
      ]
     },
     "execution_count": 207,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_idx.loc[83]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "stacked_averaged_model이 포함되지 않는다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**rmse가 최소인 네개의 model로 해보자**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 201,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_xgb\n",
    "\n",
    "rmse_list = []\n",
    "idx_i = []\n",
    "idx_j = []\n",
    "idx_k = []\n",
    "\n",
    "def myrange(start, end, step):\n",
    "    r = start\n",
    "    while(r<end):\n",
    "        yield r\n",
    "        r += step\n",
    "\n",
    "        \n",
    "\n",
    "for i in myrange(0, 1, 0.01):\n",
    "    for j in myrange(0, 1-i, 0.01):\n",
    "        for k in myrange(0, 1-i-j, 0.01):\n",
    "            l = 1 - i - j - k\n",
    "       \n",
    "            rmse_list.append(rmsle(y_train,GBoost_train_pred*i + lgb_train_pred*j + stacked_train_pred*k + xgb_train_pred*l))\n",
    "            idx_i.append(i)\n",
    "            idx_j.append(j)\n",
    "            idx_k.append(k)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 202,
   "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>rmse</th>\n",
       "      <th>i</th>\n",
       "      <th>j</th>\n",
       "      <th>k</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.080072</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.079895</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.079720</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.079546</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.079374</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>171695</td>\n",
       "      <td>0.048845</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>171696</td>\n",
       "      <td>0.048987</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>171697</td>\n",
       "      <td>0.049044</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>171698</td>\n",
       "      <td>0.048910</td>\n",
       "      <td>0.98</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>171699</td>\n",
       "      <td>0.048980</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>171700 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            rmse     i     j     k\n",
       "0       0.080072  0.00  0.00  0.00\n",
       "1       0.079895  0.00  0.00  0.01\n",
       "2       0.079720  0.00  0.00  0.02\n",
       "3       0.079546  0.00  0.00  0.03\n",
       "4       0.079374  0.00  0.00  0.04\n",
       "...          ...   ...   ...   ...\n",
       "171695  0.048845  0.97  0.02  0.00\n",
       "171696  0.048987  0.98  0.00  0.00\n",
       "171697  0.049044  0.98  0.00  0.01\n",
       "171698  0.048910  0.98  0.01  0.00\n",
       "171699  0.048980  0.99  0.00  0.00\n",
       "\n",
       "[171700 rows x 4 columns]"
      ]
     },
     "execution_count": 202,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dic = {'rmse' : rmse_list, 'i' : idx_i, 'j' : idx_j, 'k' : idx_k}\n",
    "df_idx =pd.DataFrame(dic)\n",
    "df_idx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 203,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.04840357930082236"
      ]
     },
     "execution_count": 203,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_idx['rmse'].min()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "두개의 model로 앙상블 하였을때가 가장 낮은 score였지만 stacking model이 안들어가있기 때문에 차이가 많이 없는 3개 model조합으로 하였다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "844b5e21-7bd2-4a2b-9f7a-2e755ed06ecb",
    "_execution_state": "idle",
    "_uuid": "59443e95f66cb9e595cff9a3666824299239126b"
   },
   "source": [
    "**Ensemble prediction:**\n",
    "GBoost_train_pred x 0.83\n",
    "lgb_train_pred x 0.16\n",
    "stacked_train_pred x 0.01\n",
    "사용"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 283,
   "metadata": {
    "_cell_guid": "3ec2c58f-6bee-46a6-a263-1fe2cf3569cb",
    "_execution_state": "idle",
    "_uuid": "18996472b775bd9114fea7f08c8a554d4dafe774"
   },
   "outputs": [],
   "source": [
    "ensemble =  stacked_pred*0.83 + xgb_pred*0.16+ lgb_pred*0.01\n",
    "# 앙상블 결과 저장 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 284,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([12.02765015, 11.77694103, 11.89592668, 11.88219443, 12.24601123,\n",
       "       11.79626611, 12.46294686, 11.77518242, 12.07731349, 12.36550123,\n",
       "       11.51223767, 12.12992577, 11.65920437, 11.79690918, 11.45453833,\n",
       "       12.70436925, 11.86480371, 12.70466911, 11.92371946, 11.81176153,\n",
       "       11.94301193, 12.29316435, 12.0526508 , 11.77215766, 11.65377926,\n",
       "       12.42952145, 12.00186087, 11.84132291, 12.58627693, 11.97025025,\n",
       "       11.66842855, 12.04560427, 11.87770494, 11.95699283, 11.71273132,\n",
       "       12.01215254, 11.50333561, 11.86982359, 13.07269614, 11.8445689 ,\n",
       "       11.61927175, 11.91038735, 12.11630209, 12.43373133, 11.66626232,\n",
       "       11.73516052, 12.49390269, 12.33906644, 11.79275036, 12.07424955,\n",
       "       11.94312557, 11.93348238, 11.48760183, 11.96671661, 12.01678331,\n",
       "       11.40354321, 10.79668456, 11.80712693, 12.43681392, 12.1319725 ,\n",
       "       12.171493  , 11.98996577, 12.0940123 , 12.19776005, 12.24357823,\n",
       "       11.93219657, 12.64394239, 12.12516222, 12.45619873, 12.30478961,\n",
       "       12.29775719, 11.7583002 , 12.67428702, 12.20790571, 12.36088301,\n",
       "       11.43891249, 12.14606691, 12.40016423, 12.40490711, 12.30021356,\n",
       "       12.06034248, 12.3495595 , 11.67168954, 11.88654963, 11.92007832,\n",
       "       11.98174849, 11.89527844, 12.52551548, 12.27319843, 12.06686898,\n",
       "       11.84701365, 11.651606  , 11.75040987, 12.00636277, 11.67423671,\n",
       "       11.98514011, 11.74729689, 11.7501026 , 12.17521789, 12.01260538,\n",
       "       11.6896778 , 12.49856667, 11.98491505, 12.53721317, 12.32522509,\n",
       "       12.0777924 , 11.86516875, 11.80075786, 11.20311407, 11.74158998,\n",
       "       11.3737634 , 12.06651725, 11.6920678 , 12.24599909, 12.13182766,\n",
       "       12.02904749, 11.35467262, 12.1730474 , 11.53484544, 11.82154723,\n",
       "       12.14583569, 12.67957279, 12.06979358, 12.38934045, 11.81584624,\n",
       "       11.7245648 , 11.76419338, 12.8885859 , 12.13324183, 12.18454318,\n",
       "       12.6454665 , 12.15073749, 12.22220637, 12.05146667, 12.73480557,\n",
       "       11.62884421, 11.75035398, 11.95341675, 11.41859551, 12.16030844,\n",
       "       11.13469812, 11.2578024 , 12.40767863, 11.90415095, 12.16991717,\n",
       "       11.82081671, 12.01140447, 11.43460476, 12.43904766, 11.61470058,\n",
       "       11.73774784, 11.64690354, 11.98975006, 11.47485551, 11.74943   ,\n",
       "       12.22458167, 12.30634575, 12.13867726, 12.0362452 , 12.18557443,\n",
       "       11.95135996, 12.21320438, 11.6718345 , 12.16909403, 12.61753024,\n",
       "       11.6923402 , 11.87269828, 12.59983674, 11.72369383, 11.69808293,\n",
       "       11.84163621, 12.3026168 , 12.65519012, 12.35448372, 11.73570731,\n",
       "       11.20578485, 12.62930913, 12.46067908, 11.51090872, 11.25157256,\n",
       "       11.94583337, 12.02391304, 12.20152868, 12.48431099, 12.65620613,\n",
       "       11.79988586, 13.04643802, 11.86076479, 11.87174963, 11.4993894 ,\n",
       "       12.13299196, 12.16718178, 12.06817848, 11.8925401 , 12.4726257 ,\n",
       "       11.98021305, 12.28970878, 12.44028757, 12.76301455, 12.17906806,\n",
       "       12.22342884, 11.37298513, 11.79267742, 11.80268021, 11.92982004,\n",
       "       12.50072768, 12.06516509, 12.07024154, 12.07825757, 11.78087225,\n",
       "       11.78823396, 11.93859992, 11.52897642, 11.86514811, 12.14591478,\n",
       "       12.13747593, 11.99482656, 11.78937334, 11.85859567, 11.80057364,\n",
       "       11.96875367, 11.74396936, 12.17155216, 11.93943524, 12.09492001,\n",
       "       11.95754224, 11.48436025, 12.47849853, 11.83597404, 12.20668816,\n",
       "       11.41243016, 11.96195791, 12.04196809, 11.96731595, 11.77541654,\n",
       "       12.06001494, 11.70260724, 11.97511284, 11.62728233, 11.97098999,\n",
       "       10.9787123 , 11.6876878 , 11.7763056 , 11.94701908, 11.61406317,\n",
       "       12.69536129, 12.14987845, 12.44644837, 12.01996581, 12.62579517,\n",
       "       12.10579749, 12.17874565, 11.65658926, 11.80648616, 11.68150229,\n",
       "       12.66665401, 12.16931419, 11.81412754, 11.82148056, 11.76337695,\n",
       "       11.62531296, 12.31557352, 11.28279795, 11.44262582, 12.03067627,\n",
       "       11.63203903, 12.23976557, 12.39308026, 12.30591521, 11.76738075,\n",
       "       11.88116386, 11.64834408, 11.79889413, 12.3600841 , 12.08563925,\n",
       "       11.54825026, 11.70258812, 11.45678666, 12.23071582, 11.6916817 ,\n",
       "       11.70474878, 12.1207924 , 12.8799764 , 11.48171913, 11.41766101,\n",
       "       11.79607793, 12.0899629 , 12.09063652, 11.63815615, 11.97531783,\n",
       "       11.70026941, 11.95570819, 12.25338557, 11.71289582, 12.29931497,\n",
       "       12.22005119, 12.09513656, 12.25491672, 12.38677094, 12.13789424,\n",
       "       11.88513934, 12.28842659, 11.76051451, 11.79934754, 12.20891698,\n",
       "       12.31660879, 12.42748818, 12.17445547, 12.21279852, 13.14547063,\n",
       "       11.90217137, 11.98381548, 11.99541985, 12.35020762, 12.30938407,\n",
       "       12.68993998, 11.73002713, 11.62957141, 11.76667719, 11.38617651,\n",
       "       12.42959414, 12.53884966, 13.10909511, 11.90552419, 12.24912508,\n",
       "       11.714797  , 11.68986555, 12.5407488 , 12.12961571, 12.17162946,\n",
       "       12.16024434, 12.04515398, 11.80565788, 11.98432715, 11.87667039,\n",
       "       12.01043714, 12.36637578, 11.89779275, 12.06210642, 11.85841853,\n",
       "       11.46570839, 12.18623575, 11.75716955, 12.22637585, 11.92223302,\n",
       "       12.57068636, 11.62180039, 12.29118885, 11.86672475, 12.56790644,\n",
       "       12.16862022, 12.24697889, 11.32149363, 11.80498161, 11.77348135,\n",
       "       11.63514334, 11.80744422, 12.37639567, 11.19383855, 11.37843374,\n",
       "       11.56161888, 11.77544709, 11.98263392, 11.88871378, 12.22329905,\n",
       "       11.87131912, 12.36318002, 11.9015178 , 12.79083817, 11.86991283,\n",
       "       12.19176524, 11.89692047, 11.56384985, 11.78604963, 11.66848997,\n",
       "       11.84279597, 12.04396785, 12.28921205, 11.84885624, 12.37746814,\n",
       "       12.26942161, 12.44967772, 11.99589008, 12.96509692, 12.18035191,\n",
       "       12.17937916, 12.18155366, 11.85641228, 11.94732746, 11.70838163,\n",
       "       12.49304302, 12.14780655, 11.65662816, 12.49182249, 12.118278  ,\n",
       "       11.80193419, 12.68358865, 11.41786822, 12.12490786, 11.95196567,\n",
       "       12.1890429 , 11.7362389 , 11.8556704 , 11.98726043, 11.96956566,\n",
       "       12.07270171, 11.72880749, 12.76229886, 12.82485497, 11.84572381,\n",
       "       12.3852328 , 12.04075513, 11.7935229 , 12.0799772 , 11.83190326,\n",
       "       11.87813546, 11.934349  , 12.19117361, 12.42092048, 11.25059834,\n",
       "       12.25769084, 12.16772617, 11.89785062, 11.86653627, 12.14137467,\n",
       "       11.73244563, 11.9349072 , 12.11196677, 12.44594705, 12.57632352,\n",
       "       11.98336816, 11.66985743, 11.74731176, 12.12086533, 11.63275193,\n",
       "       11.8026964 , 11.91988056, 11.8277738 , 13.43530575, 12.04646032,\n",
       "       12.33561202, 11.9347146 , 12.64793987, 12.35535355, 11.87366557,\n",
       "       12.68404789, 12.22121147, 11.87181312, 11.90302138, 12.07263095,\n",
       "       12.65431751, 12.21621346, 12.68708546, 12.69087418, 11.61556316,\n",
       "       12.14483207, 12.50103116, 12.1661574 , 12.51737106, 11.59346891,\n",
       "       11.97240833, 11.24643758, 12.34236682, 11.32992335, 12.58959405,\n",
       "       11.08626597, 11.34255665, 11.78693905, 12.38321355, 12.13559174,\n",
       "       12.45394957, 11.80137839, 11.58371189, 11.76434997, 11.65808306,\n",
       "       11.98279989, 11.83868188, 10.97703179, 12.3478916 , 11.71800646,\n",
       "       11.58197945, 11.99296811, 12.39548816, 12.07489469, 12.11962654,\n",
       "       11.5046903 , 12.49445182, 12.59110386, 12.4091193 , 11.91011001,\n",
       "       12.15990796, 12.04587167, 11.74482107, 12.89212189, 12.36468979,\n",
       "       12.1575864 , 11.63183095, 11.96503661, 12.01729359, 12.84636311,\n",
       "       12.16604841, 12.5281681 , 11.74590463, 12.01903169, 11.85691447,\n",
       "       12.2404353 , 12.19299805, 11.83334008, 11.83548312, 12.4374478 ,\n",
       "       11.67463102, 12.13755752, 12.46015859, 13.00624155, 12.41924769,\n",
       "       12.42284977, 11.48625251, 11.82930078, 11.51307706, 11.9419494 ,\n",
       "       11.22927998, 12.33685959, 11.93949296, 11.96344009, 11.58587633,\n",
       "       11.63140083, 12.54226344, 11.99504505, 12.75699866, 11.80663932,\n",
       "       12.34526845, 11.80435389, 11.74724757, 11.93939481, 12.35272108,\n",
       "       12.57639578, 11.9933252 , 11.84035824, 11.82115692, 12.00523268,\n",
       "       11.68223638, 12.12981542, 12.16634668, 12.47361209, 11.49711565,\n",
       "       12.46479405, 11.76906679, 11.64573713, 11.67302759, 12.27174653,\n",
       "       12.31158803, 11.95430635, 11.48599951, 12.45728181, 12.13746973,\n",
       "       12.32419192, 11.80119899, 12.69092284, 11.78173324, 12.35506166,\n",
       "       12.03402163, 11.86082708, 11.82618672, 12.14540814, 11.86811544,\n",
       "       11.9192187 , 12.45549003, 11.86075062, 11.60973851, 12.17605351,\n",
       "       12.18039264, 11.76882177, 11.56557887, 12.15846423, 12.04102592,\n",
       "       12.01032576, 11.67403423, 12.86108099, 11.85531561, 12.09650996,\n",
       "       12.22105052, 11.8536749 , 12.76440233, 11.59369109, 12.18273278,\n",
       "       11.71650683, 11.90697943, 12.45799611, 11.62890033, 11.33780653,\n",
       "       11.83032843, 12.46567363, 11.75689759, 11.81214507, 12.15107613,\n",
       "       12.08741564, 12.19949391, 12.39995168, 11.82627741, 11.91511944])"
      ]
     },
     "execution_count": 284,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ensemble"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "434ca649-2fa0-46a5-ab29-7f403448ddf7",
    "_execution_state": "idle",
    "_uuid": "c9f02561da543f4901dcd2051acbd6c197108dd5"
   },
   "source": [
    "**Submission**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 285,
   "metadata": {
    "_cell_guid": "3db46af9-e18a-43bb-9699-45b851f835e5",
    "_execution_state": "idle",
    "_uuid": "93f6915cf25c7bb6b6fa6e74ad7b853387ac1db5"
   },
   "outputs": [],
   "source": [
    "sub = pd.DataFrame()\n",
    "sub['Id'] = test_ID\n",
    "sub['SalePrice'] = ensemble\n",
    "sub.to_csv('submission.csv',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 286,
   "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>Id</th>\n",
       "      <th>SalePrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>861</td>\n",
       "      <td>12.027650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>862</td>\n",
       "      <td>11.776941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>863</td>\n",
       "      <td>11.895927</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>864</td>\n",
       "      <td>11.882194</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>865</td>\n",
       "      <td>12.246011</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Id  SalePrice\n",
       "0  861  12.027650\n",
       "1  862  11.776941\n",
       "2  863  11.895927\n",
       "3  864  11.882194\n",
       "4  865  12.246011"
      ]
     },
     "execution_count": 286,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sub.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}