Day_010_HW.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# 作業 : (Kaggle)房價預測"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# [作業目標]\n",
    "- 試著模仿範例寫法, 在房價預測中, 觀察去除離群值的影響"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# [作業重點]\n",
    "- 觀察將極端值以上下限值取代, 對於分布與迴歸分數的影響 (In[5], Out[5])\n",
    "- 觀察將極端值資料直接刪除, 對於分布與迴歸分數的影響 (In[6], Out[6])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>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>LotConfig</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",
       "      <th>0</th>\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>Inside</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>2</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\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>FR2</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>2007</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\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>Inside</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>9</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\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>Corner</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>2</td>\n",
       "      <td>2006</td>\n",
       "      <td>WD</td>\n",
       "      <td>Abnorml</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\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>FR2</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>12</td>\n",
       "      <td>2008</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 79 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   MSSubClass MSZoning  LotFrontage  LotArea Street Alley LotShape  \\\n",
       "0          60       RL         65.0     8450   Pave   NaN      Reg   \n",
       "1          20       RL         80.0     9600   Pave   NaN      Reg   \n",
       "2          60       RL         68.0    11250   Pave   NaN      IR1   \n",
       "3          70       RL         60.0     9550   Pave   NaN      IR1   \n",
       "4          60       RL         84.0    14260   Pave   NaN      IR1   \n",
       "\n",
       "  LandContour Utilities LotConfig  ... ScreenPorch PoolArea PoolQC Fence  \\\n",
       "0         Lvl    AllPub    Inside  ...           0        0    NaN   NaN   \n",
       "1         Lvl    AllPub       FR2  ...           0        0    NaN   NaN   \n",
       "2         Lvl    AllPub    Inside  ...           0        0    NaN   NaN   \n",
       "3         Lvl    AllPub    Corner  ...           0        0    NaN   NaN   \n",
       "4         Lvl    AllPub       FR2  ...           0        0    NaN   NaN   \n",
       "\n",
       "  MiscFeature MiscVal  MoSold  YrSold  SaleType  SaleCondition  \n",
       "0         NaN       0       2    2008        WD         Normal  \n",
       "1         NaN       0       5    2007        WD         Normal  \n",
       "2         NaN       0       9    2008        WD         Normal  \n",
       "3         NaN       0       2    2006        WD        Abnorml  \n",
       "4         NaN       0      12    2008        WD         Normal  \n",
       "\n",
       "[5 rows x 79 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 做完特徵工程前的所有準備 (與前範例相同)\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "data_path = 'data2/'\n",
    "df_train = pd.read_csv(data_path + 'house_train.csv.gz')\n",
    "\n",
    "train_Y = np.log1p(df_train['SalePrice'])\n",
    "df = df_train.drop(['Id', 'SalePrice'] , axis=1)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "36 Numeric Features : ['MSSubClass', 'LotFrontage', 'LotArea', 'OverallQual', 'OverallCond', 'YearBuilt', 'YearRemodAdd', 'MasVnrArea', 'BsmtFinSF1', 'BsmtFinSF2', 'BsmtUnfSF', 'TotalBsmtSF', '1stFlrSF', '2ndFlrSF', 'LowQualFinSF', 'GrLivArea', 'BsmtFullBath', 'BsmtHalfBath', 'FullBath', 'HalfBath', 'BedroomAbvGr', 'KitchenAbvGr', 'TotRmsAbvGrd', 'Fireplaces', 'GarageYrBlt', 'GarageCars', 'GarageArea', 'WoodDeckSF', 'OpenPorchSF', 'EnclosedPorch', '3SsnPorch', 'ScreenPorch', 'PoolArea', 'MiscVal', 'MoSold', 'YrSold']\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#只取 int64, float64 兩種數值型欄位, 存於 num_features 中\n",
    "num_features = []\n",
    "for dtype, feature in zip(df.dtypes, df.columns):\n",
    "    if dtype == 'float64' or dtype == 'int64':\n",
    "        num_features.append(feature)\n",
    "print(f'{len(num_features)} Numeric Features : {num_features}\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>MSSubClass</th>\n",
       "      <th>LotFrontage</th>\n",
       "      <th>LotArea</th>\n",
       "      <th>OverallQual</th>\n",
       "      <th>OverallCond</th>\n",
       "      <th>YearBuilt</th>\n",
       "      <th>YearRemodAdd</th>\n",
       "      <th>MasVnrArea</th>\n",
       "      <th>BsmtFinSF1</th>\n",
       "      <th>BsmtFinSF2</th>\n",
       "      <th>...</th>\n",
       "      <th>GarageArea</th>\n",
       "      <th>WoodDeckSF</th>\n",
       "      <th>OpenPorchSF</th>\n",
       "      <th>EnclosedPorch</th>\n",
       "      <th>3SsnPorch</th>\n",
       "      <th>ScreenPorch</th>\n",
       "      <th>PoolArea</th>\n",
       "      <th>MiscVal</th>\n",
       "      <th>MoSold</th>\n",
       "      <th>YrSold</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>60</td>\n",
       "      <td>65.0</td>\n",
       "      <td>8450</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>2003</td>\n",
       "      <td>2003</td>\n",
       "      <td>196.0</td>\n",
       "      <td>706</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>548</td>\n",
       "      <td>0</td>\n",
       "      <td>61</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>20</td>\n",
       "      <td>80.0</td>\n",
       "      <td>9600</td>\n",
       "      <td>6</td>\n",
       "      <td>8</td>\n",
       "      <td>1976</td>\n",
       "      <td>1976</td>\n",
       "      <td>0.0</td>\n",
       "      <td>978</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>460</td>\n",
       "      <td>298</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>2007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>60</td>\n",
       "      <td>68.0</td>\n",
       "      <td>11250</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>2001</td>\n",
       "      <td>2002</td>\n",
       "      <td>162.0</td>\n",
       "      <td>486</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>608</td>\n",
       "      <td>0</td>\n",
       "      <td>42</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>2008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>70</td>\n",
       "      <td>60.0</td>\n",
       "      <td>9550</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>1915</td>\n",
       "      <td>1970</td>\n",
       "      <td>0.0</td>\n",
       "      <td>216</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>642</td>\n",
       "      <td>0</td>\n",
       "      <td>35</td>\n",
       "      <td>272</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>60</td>\n",
       "      <td>84.0</td>\n",
       "      <td>14260</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>2000</td>\n",
       "      <td>2000</td>\n",
       "      <td>350.0</td>\n",
       "      <td>655</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>836</td>\n",
       "      <td>192</td>\n",
       "      <td>84</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>2008</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 36 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   MSSubClass  LotFrontage  LotArea  OverallQual  OverallCond  YearBuilt  \\\n",
       "0          60         65.0     8450            7            5       2003   \n",
       "1          20         80.0     9600            6            8       1976   \n",
       "2          60         68.0    11250            7            5       2001   \n",
       "3          70         60.0     9550            7            5       1915   \n",
       "4          60         84.0    14260            8            5       2000   \n",
       "\n",
       "   YearRemodAdd  MasVnrArea  BsmtFinSF1  BsmtFinSF2  ...  GarageArea  \\\n",
       "0          2003       196.0         706           0  ...         548   \n",
       "1          1976         0.0         978           0  ...         460   \n",
       "2          2002       162.0         486           0  ...         608   \n",
       "3          1970         0.0         216           0  ...         642   \n",
       "4          2000       350.0         655           0  ...         836   \n",
       "\n",
       "   WoodDeckSF  OpenPorchSF  EnclosedPorch  3SsnPorch  ScreenPorch  PoolArea  \\\n",
       "0           0           61              0          0            0         0   \n",
       "1         298            0              0          0            0         0   \n",
       "2           0           42              0          0            0         0   \n",
       "3           0           35            272          0            0         0   \n",
       "4         192           84              0          0            0         0   \n",
       "\n",
       "   MiscVal  MoSold  YrSold  \n",
       "0        0       2    2008  \n",
       "1        0       5    2007  \n",
       "2        0       9    2008  \n",
       "3        0       2    2006  \n",
       "4        0      12    2008  \n",
       "\n",
       "[5 rows x 36 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 削減文字型欄位, 只剩數值型欄位\n",
    "df = df[num_features]\n",
    "df = df.fillna(-1)\n",
    "MMEncoder = MinMaxScaler()\n",
    "train_num = train_Y.shape[0]\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# 作業1\n",
    "* 試著限制 '1樓地板面積(平方英尺)' (1stFlrSF) 欄位的上下限, 看看能否再進一步提高分數?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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"
    },
    {
     "data": {
      "text/plain": [
       "0.846643681071632"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 顯示 1stFlrSF 與目標值的散佈圖\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "sns.regplot(x = df['1stFlrSF'][:train_num], y=train_Y)\n",
    "plt.show()\n",
    "\n",
    "# 做線性迴歸, 觀察分數\n",
    "train_X = MMEncoder.fit_transform(df)\n",
    "estimator = LinearRegression()\n",
    "cross_val_score(estimator, train_X, train_Y, cv=5).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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"
    },
    {
     "data": {
      "text/plain": [
       "0.8868529812963668"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 將 1stFlrSF 限制在你覺得適合的範圍內, 調整離群值\n",
    "\"\"\"\n",
    "Your Code Here\n",
    "\"\"\"\n",
    "df['1stFlrSF'] = df['1stFlrSF'].clip(500, 2500)\n",
    "sns.regplot(x = df['1stFlrSF'], y = train_Y)\n",
    "plt.show()\n",
    "\n",
    "# 做線性迴歸, 觀察分數\n",
    "train_X = MMEncoder.fit_transform(df)\n",
    "estimator = LinearRegression()\n",
    "cross_val_score(estimator, train_X, train_Y, cv=5).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# 作業2\n",
    "* 續前題, 去除離群值有兩類方式 :  捨棄離群值(刪除離群的資料) 以及調整離群值,  \n",
    "請試著用同樣的上下限, 改為 '捨棄離群值' 的方法, 看看結果會變好還是變差? 並試著解釋原因。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": true
   },
   "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"
    },
    {
     "data": {
      "text/plain": [
       "0.8934564523411981"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 將 1stFlrSF 限制在你覺得適合的範圍內, 捨棄離群值\n",
    "valid_index = (df['1stFlrSF'] >= 500) & (df['1stFlrSF'] <= 2500)\n",
    "df = df[valid_index]\n",
    "train_Y = train_Y[valid_index]\n",
    "sns.regplot(x = df['1stFlrSF'], y = train_Y)\n",
    "plt.show()\n",
    "\n",
    "# 做線性迴歸, 觀察分數\n",
    "train_X = MMEncoder.fit_transform(df)\n",
    "estimator = LinearRegression()\n",
    "cross_val_score(estimator, train_X, train_Y, cv=5).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "捨棄離群值比直接將過高或是過低的調整成就近的數值來的好，因為只是將周邊的離群值調整成鎖定範圍的結果下，X值不變會將整體趨勢線修改了\n",
    "不會是漂亮的線性，所以透過剔除離群值的方式來看，會發現結果會變好很多，會剔除比較偏遠的X值。"
   ]
  }
 ],
 "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": 2
}