univariate_automated_feature_selection.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn import linear_model\n",
    "import sklearn.metrics as metrics\n",
    "from sklearn.feature_selection import SelectKBest\n",
    "from sklearn.feature_selection import chi2\n",
    "from sklearn.feature_selection import RFECV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>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>208500</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>181500</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>223500</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>140000</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>250000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 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",
       "  LandContour Utilities  ... PoolArea PoolQC Fence MiscFeature MiscVal MoSold  \\\n",
       "0         Lvl    AllPub  ...        0    NaN   NaN         NaN       0      2   \n",
       "1         Lvl    AllPub  ...        0    NaN   NaN         NaN       0      5   \n",
       "2         Lvl    AllPub  ...        0    NaN   NaN         NaN       0      9   \n",
       "3         Lvl    AllPub  ...        0    NaN   NaN         NaN       0      2   \n",
       "4         Lvl    AllPub  ...        0    NaN   NaN         NaN       0     12   \n",
       "\n",
       "  YrSold  SaleType  SaleCondition  SalePrice  \n",
       "0   2008        WD         Normal     208500  \n",
       "1   2007        WD         Normal     181500  \n",
       "2   2008        WD         Normal     223500  \n",
       "3   2006        WD        Abnorml     140000  \n",
       "4   2008        WD         Normal     250000  \n",
       "\n",
       "[5 rows x 81 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('./kaggle_data/train.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Cleaning Up the Data\n",
    "1. Converting categorical cols, rating cols, and continuous cols to a format an ML model can read\n",
    "2. Splitting this data 80-20 for testing purposes\n",
    "3. Removing NaNs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# These are the manually picked columns that are all categorical\n",
    "total_categorical_cols = ['MSZoning', 'Street', 'Alley', 'LandContour', \n",
    "                    'LotConfig', 'Neighborhood', 'Condition1', 'Condition2', 'BldgType', \n",
    "                    'HouseStyle', 'RoofStyle', 'RoofMatl', 'Exterior1st', 'Exterior2nd', 'MasVnrType', \n",
    "                    'ExterCond', 'Foundation', 'Heating', 'CentralAir', 'PavedDrive',\n",
    "                    'MiscFeature', 'SaleType', 'SaleCondition']\n",
    "\n",
    "# These columns will have to be manually processed\n",
    "total_rating_cols = ['LotShape', 'Utilities', 'LandSlope', 'ExterQual', 'ExterCond', \n",
    "                   'BsmtQual', 'BsmtCond', 'BsmtExposure', 'BsmtFinType1', 'BsmtFinType2',\n",
    "                  'HeatingQC', 'Electrical', 'KitchenQual', 'Functional', 'FireplaceQu',\n",
    "                    'GarageType', 'GarageFinish', 'GarageQual', 'GarageCond', 'PoolQC', 'Fence']\n",
    "\n",
    "#########\n",
    "# Below will be the columns actually used in the model\n",
    "#########\n",
    "\n",
    "exclude_cols = ['Id', 'SalePrice']\n",
    "\n",
    "category_cols = total_categorical_cols.copy()\n",
    "rating_cols = total_rating_cols.copy()\n",
    "continuous_cols = [col for col in df.columns if col not in category_cols and col not in rating_cols and col not in exclude_cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Take rating columns and convert to numerical scale\n",
    "rating_df = df.copy()[rating_cols]\n",
    "\n",
    "replace_dict = {'Ex': 5, 'Gd': 4, 'TA': 3, 'Fa': 2, 'Po': 1,\n",
    "               'Reg': 1, 'IR1': 2, 'IR2': 3, 'IR3': 4,\n",
    "               'AllPub': 4, 'NoSewr': 3, 'NoSeWa': 2, 'ELO': 1,\n",
    "               'Gtl': 1, 'Mod': 2, 'Sev': 3,\n",
    "               'NA': 0,\n",
    "               'Av': 3, 'Mn': 2, 'No': 1,\n",
    "               'GLQ': 6, 'ALQ': 5, 'BLQ': 4, 'Rec': 3, 'LwQ': 2, 'Unf': 1,\n",
    "               'SBrkr': 5, 'FuseA': 4, 'FuseF': 3, 'FuseP': 2, 'Mix': 1,\n",
    "               'Typ': 8, 'Min1': 7, 'Min2': 6, 'Mod': 5, 'Maj1': 4, 'Maj2': 3, 'Sev': 2, 'Sal': 1,\n",
    "               '2Types': 6, 'Attchd': 5, 'Basment': 4, 'BuiltIn': 3, 'CarPort': 2, 'Detchd': 1,\n",
    "               'Fin': 3, 'RFn': 2, 'Unf': 1,\n",
    "               'GdPrv': 4, 'MnPrv': 3, 'GdWo': 2, 'MnWw': 1}\n",
    "\n",
    "rating_df.replace(replace_dict, inplace=True)\n",
    "\n",
    "na_to_zero_cols = ['LotShape', 'BsmtQual', 'BsmtCond', 'BsmtExposure', 'BsmtFinType1', 'BsmtFinType2',\n",
    "                  'FireplaceQu', 'GarageType', 'GarageFinish', 'GarageCond', 'GarageQual',\n",
    "                   'PoolQC', 'Fence', 'Electrical']\n",
    "\n",
    "na_to_zero_cols = [col for col in na_to_zero_cols if col in rating_df]\n",
    "\n",
    "rating_df[na_to_zero_cols] = rating_df.loc[:, na_to_zero_cols].fillna(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Select categorical columns and get dummy column values (different approach from what we did before with category type)\n",
    "# Preferably pick columns that are not like rating columns (Bad, Okay, Great). We should handle those separately\n",
    "dummy_df = pd.get_dummies(df[category_cols])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Num rows after filtering out nans: 1121\n",
      "Percent remaining after filtering: 77%\n",
      "Num columns: 220\n"
     ]
    }
   ],
   "source": [
    "YVar = df[['SalePrice']]\n",
    "XVar = pd.concat([df[continuous_cols], dummy_df, rating_df], axis=1)\n",
    "\n",
    "# Filter out nan values\n",
    "YVar = YVar[XVar.notna().all(axis='columns')]\n",
    "XVar = XVar[XVar.notna().all(axis='columns')]\n",
    "\n",
    "# TODO GO BACK AND FIGURE OUT WHERE ALL THESE NANS ARE COMING FROM\n",
    "print('Num rows after filtering out nans: ' + str(len(XVar)))\n",
    "print('Percent remaining after filtering: ' + str(round(len(XVar) / len(df) * 100)) + '%')\n",
    "print('Num columns: ' + str(len(XVar.columns)))\n",
    "\n",
    "XVar = XVar.apply(pd.to_numeric)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(893, 220)\n"
     ]
    }
   ],
   "source": [
    "# Split data 80-20 so we can test how well the model works\n",
    "msk = np.random.rand(len(XVar)) < 0.8\n",
    "\n",
    "train_x = XVar[msk]\n",
    "train_y = YVar[msk]\n",
    "\n",
    "test_x = XVar[~msk]\n",
    "test_y = YVar[~msk]\n",
    "\n",
    "print(train_x.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Univariate Feature Selection\n",
    "1. Find x best columns to use based on univariate feature selection\n",
    "2. Produce univariate_train_x which selects those columns from train_x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['MSSubClass', 'LotFrontage', 'LotArea', 'MasVnrArea', 'BsmtFinSF1', 'BsmtFinSF2', 'BsmtUnfSF', 'TotalBsmtSF', '1stFlrSF', '2ndFlrSF', 'LowQualFinSF', 'GrLivArea', 'BsmtHalfBath', 'GarageArea', 'WoodDeckSF', 'OpenPorchSF', 'EnclosedPorch', '3SsnPorch', 'ScreenPorch', 'PoolArea', 'MiscVal', 'MoSold', 'MSZoning_C (all)', 'MSZoning_FV', 'Street_Grvl', 'LandContour_HLS', 'LandContour_Low', 'LotConfig_CulDSac', 'LotConfig_FR2', 'LotConfig_FR3', 'Neighborhood_Blmngtn', 'Neighborhood_BrDale', 'Neighborhood_CollgCr', 'Neighborhood_Crawfor', 'Neighborhood_Gilbert', 'Neighborhood_IDOTRR', 'Neighborhood_MeadowV', 'Neighborhood_NPkVill', 'Neighborhood_NoRidge', 'Neighborhood_NridgHt', 'Neighborhood_SawyerW', 'Neighborhood_Somerst', 'Neighborhood_StoneBr', 'Neighborhood_Timber', 'Neighborhood_Veenker', 'Condition1_RRAn', 'Condition2_PosN', 'Condition2_RRNn', 'BldgType_Twnhs', 'HouseStyle_2.5Fin', 'RoofStyle_Gambrel', 'RoofMatl_WdShngl', 'Exterior1st_BrkFace', 'Exterior1st_CemntBd', 'Exterior1st_Stucco', 'Exterior2nd_Brk Cmn', 'Exterior2nd_BrkFace', 'Exterior2nd_CmentBd', 'Exterior2nd_ImStucc', 'Exterior2nd_Stucco', 'MasVnrType_BrkCmn', 'MasVnrType_Stone', 'ExterCond_Ex', 'Foundation_Slab', 'Foundation_Stone', 'Heating_Grav', 'Heating_Wall', 'CentralAir_N', 'SaleType_CWD', 'SaleType_Con', 'SaleType_ConLD', 'SaleType_New', 'SaleType_Oth', 'SaleCondition_Alloca', 'SaleCondition_Partial', 'BsmtFinType1', 'FireplaceQu', 'GarageType', 'PoolQC', 'Fence']\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>MSSubClass</th>\n",
       "      <th>LotFrontage</th>\n",
       "      <th>LotArea</th>\n",
       "      <th>MasVnrArea</th>\n",
       "      <th>BsmtFinSF1</th>\n",
       "      <th>BsmtFinSF2</th>\n",
       "      <th>BsmtUnfSF</th>\n",
       "      <th>TotalBsmtSF</th>\n",
       "      <th>1stFlrSF</th>\n",
       "      <th>2ndFlrSF</th>\n",
       "      <th>...</th>\n",
       "      <th>SaleType_ConLD</th>\n",
       "      <th>SaleType_New</th>\n",
       "      <th>SaleType_Oth</th>\n",
       "      <th>SaleCondition_Alloca</th>\n",
       "      <th>SaleCondition_Partial</th>\n",
       "      <th>BsmtFinType1</th>\n",
       "      <th>FireplaceQu</th>\n",
       "      <th>GarageType</th>\n",
       "      <th>PoolQC</th>\n",
       "      <th>Fence</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>70</td>\n",
       "      <td>60.0</td>\n",
       "      <td>9550</td>\n",
       "      <td>0.0</td>\n",
       "      <td>216</td>\n",
       "      <td>0</td>\n",
       "      <td>540</td>\n",
       "      <td>756</td>\n",
       "      <td>961</td>\n",
       "      <td>756</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>50</td>\n",
       "      <td>85.0</td>\n",
       "      <td>14115</td>\n",
       "      <td>0.0</td>\n",
       "      <td>732</td>\n",
       "      <td>0</td>\n",
       "      <td>64</td>\n",
       "      <td>796</td>\n",
       "      <td>796</td>\n",
       "      <td>566</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>20</td>\n",
       "      <td>75.0</td>\n",
       "      <td>10084</td>\n",
       "      <td>186.0</td>\n",
       "      <td>1369</td>\n",
       "      <td>0</td>\n",
       "      <td>317</td>\n",
       "      <td>1686</td>\n",
       "      <td>1694</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>50</td>\n",
       "      <td>51.0</td>\n",
       "      <td>6120</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>952</td>\n",
       "      <td>952</td>\n",
       "      <td>1022</td>\n",
       "      <td>752</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>190</td>\n",
       "      <td>50.0</td>\n",
       "      <td>7420</td>\n",
       "      <td>0.0</td>\n",
       "      <td>851</td>\n",
       "      <td>0</td>\n",
       "      <td>140</td>\n",
       "      <td>991</td>\n",
       "      <td>1077</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</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>1451</td>\n",
       "      <td>20</td>\n",
       "      <td>78.0</td>\n",
       "      <td>9262</td>\n",
       "      <td>194.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1573</td>\n",
       "      <td>1573</td>\n",
       "      <td>1578</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1452</td>\n",
       "      <td>180</td>\n",
       "      <td>35.0</td>\n",
       "      <td>3675</td>\n",
       "      <td>80.0</td>\n",
       "      <td>547</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>547</td>\n",
       "      <td>1072</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1454</td>\n",
       "      <td>20</td>\n",
       "      <td>62.0</td>\n",
       "      <td>7500</td>\n",
       "      <td>0.0</td>\n",
       "      <td>410</td>\n",
       "      <td>0</td>\n",
       "      <td>811</td>\n",
       "      <td>1221</td>\n",
       "      <td>1221</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1458</td>\n",
       "      <td>20</td>\n",
       "      <td>68.0</td>\n",
       "      <td>9717</td>\n",
       "      <td>0.0</td>\n",
       "      <td>49</td>\n",
       "      <td>1029</td>\n",
       "      <td>0</td>\n",
       "      <td>1078</td>\n",
       "      <td>1078</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1459</td>\n",
       "      <td>20</td>\n",
       "      <td>75.0</td>\n",
       "      <td>9937</td>\n",
       "      <td>0.0</td>\n",
       "      <td>830</td>\n",
       "      <td>290</td>\n",
       "      <td>136</td>\n",
       "      <td>1256</td>\n",
       "      <td>1256</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>893 rows × 80 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      MSSubClass  LotFrontage  LotArea  MasVnrArea  BsmtFinSF1  BsmtFinSF2  \\\n",
       "3             70         60.0     9550         0.0         216           0   \n",
       "5             50         85.0    14115         0.0         732           0   \n",
       "6             20         75.0    10084       186.0        1369           0   \n",
       "8             50         51.0     6120         0.0           0           0   \n",
       "9            190         50.0     7420         0.0         851           0   \n",
       "...          ...          ...      ...         ...         ...         ...   \n",
       "1451          20         78.0     9262       194.0           0           0   \n",
       "1452         180         35.0     3675        80.0         547           0   \n",
       "1454          20         62.0     7500         0.0         410           0   \n",
       "1458          20         68.0     9717         0.0          49        1029   \n",
       "1459          20         75.0     9937         0.0         830         290   \n",
       "\n",
       "      BsmtUnfSF  TotalBsmtSF  1stFlrSF  2ndFlrSF  ...  SaleType_ConLD  \\\n",
       "3           540          756       961       756  ...               0   \n",
       "5            64          796       796       566  ...               0   \n",
       "6           317         1686      1694         0  ...               0   \n",
       "8           952          952      1022       752  ...               0   \n",
       "9           140          991      1077         0  ...               0   \n",
       "...         ...          ...       ...       ...  ...             ...   \n",
       "1451       1573         1573      1578         0  ...               0   \n",
       "1452          0          547      1072         0  ...               0   \n",
       "1454        811         1221      1221         0  ...               0   \n",
       "1458          0         1078      1078         0  ...               0   \n",
       "1459        136         1256      1256         0  ...               0   \n",
       "\n",
       "      SaleType_New  SaleType_Oth  SaleCondition_Alloca  SaleCondition_Partial  \\\n",
       "3                0             0                     0                      0   \n",
       "5                0             0                     0                      0   \n",
       "6                0             0                     0                      0   \n",
       "8                0             0                     0                      0   \n",
       "9                0             0                     0                      0   \n",
       "...            ...           ...                   ...                    ...   \n",
       "1451             1             0                     0                      1   \n",
       "1452             0             0                     0                      0   \n",
       "1454             0             0                     0                      0   \n",
       "1458             0             0                     0                      0   \n",
       "1459             0             0                     0                      0   \n",
       "\n",
       "      BsmtFinType1  FireplaceQu  GarageType  PoolQC  Fence  \n",
       "3              5.0          4.0         1.0     0.0    0.0  \n",
       "5              6.0          0.0         5.0     0.0    3.0  \n",
       "6              6.0          4.0         5.0     0.0    0.0  \n",
       "8              1.0          3.0         1.0     0.0    0.0  \n",
       "9              6.0          3.0         5.0     0.0    0.0  \n",
       "...            ...          ...         ...     ...    ...  \n",
       "1451           1.0          4.0         5.0     0.0    0.0  \n",
       "1452           6.0          0.0         4.0     0.0    0.0  \n",
       "1454           6.0          0.0         5.0     0.0    0.0  \n",
       "1458           6.0          0.0         5.0     0.0    0.0  \n",
       "1459           4.0          0.0         5.0     0.0    0.0  \n",
       "\n",
       "[893 rows x 80 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Try running Univariate feature selection with chi^2 (supposed to work well for sparse data)\n",
    "selector = SelectKBest(chi2, k=80)\n",
    "selector.fit_transform(train_x, train_y)\n",
    "\n",
    "univ_feature_mask = selector.get_support()\n",
    "univ_features = []\n",
    "\n",
    "for b, feature in zip(univ_feature_mask, train_x.columns):\n",
    "    if b: \n",
    "        univ_features.append(feature)\n",
    "        \n",
    "univariate_train_x = train_x[univ_features]\n",
    "univariate_test_x = test_x[univ_features]\n",
    "\n",
    "print(univ_features)\n",
    "univariate_train_x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[col for col in univ_features if col in total_categorical_cols]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Run Linear Regression and Perform Analysis\n",
    "1. Create model on univariate/rfecv data\n",
    "2. Test model on test_x dataset\n",
    "3. View results, average difference, residuals"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r2:  0.7723\n"
     ]
    }
   ],
   "source": [
    "# Run linear regression with sklearn\n",
    "regr = linear_model.LinearRegression()\n",
    "regr.fit(univariate_train_x, train_y)\n",
    "\n",
    "# Make predictions using the testing set, then convert to pandas series\n",
    "saleprice_preds = regr.predict(univariate_test_x)\n",
    "saleprice_preds = pd.Series([x for lst in saleprice_preds for x in lst])\n",
    "\n",
    "# Print out coefficient information\n",
    "#print('Coefficients: \\n', regr.coef_)\n",
    "r2 = metrics.r2_score(test_y, saleprice_preds)\n",
    "print('r2: ', round(r2,4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>actual_price</th>\n",
       "      <th>predicted_price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>208500</td>\n",
       "      <td>210804.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>181500</td>\n",
       "      <td>198451.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>223500</td>\n",
       "      <td>220067.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>250000</td>\n",
       "      <td>311919.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>279500</td>\n",
       "      <td>230839.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>223</td>\n",
       "      <td>160000</td>\n",
       "      <td>163907.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>224</td>\n",
       "      <td>129000</td>\n",
       "      <td>88674.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>225</td>\n",
       "      <td>175000</td>\n",
       "      <td>176590.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>226</td>\n",
       "      <td>210000</td>\n",
       "      <td>233429.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>227</td>\n",
       "      <td>266500</td>\n",
       "      <td>265809.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>228 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     actual_price  predicted_price\n",
       "0          208500         210804.0\n",
       "1          181500         198451.0\n",
       "2          223500         220067.0\n",
       "3          250000         311919.0\n",
       "4          279500         230839.0\n",
       "..            ...              ...\n",
       "223        160000         163907.0\n",
       "224        129000          88674.0\n",
       "225        175000         176590.0\n",
       "226        210000         233429.0\n",
       "227        266500         265809.0\n",
       "\n",
       "[228 rows x 2 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Construct the results dataframe\n",
    "actual_price = test_y.reset_index()['SalePrice']\n",
    "pred_price = saleprice_preds.round()\n",
    "\n",
    "res = pd.DataFrame({'actual_price': actual_price, 'predicted_price': pred_price})\n",
    "res"
   ]
  },
  {
   "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>actual_price</th>\n",
       "      <th>predicted_price</th>\n",
       "      <th>residual</th>\n",
       "      <th>percent_diff</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>208500</td>\n",
       "      <td>210804.0</td>\n",
       "      <td>-2304.0</td>\n",
       "      <td>-1.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>181500</td>\n",
       "      <td>198451.0</td>\n",
       "      <td>-16951.0</td>\n",
       "      <td>-9.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>223500</td>\n",
       "      <td>220067.0</td>\n",
       "      <td>3433.0</td>\n",
       "      <td>1.54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>250000</td>\n",
       "      <td>311919.0</td>\n",
       "      <td>-61919.0</td>\n",
       "      <td>-24.77</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>279500</td>\n",
       "      <td>230839.0</td>\n",
       "      <td>48661.0</td>\n",
       "      <td>17.41</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>223</td>\n",
       "      <td>160000</td>\n",
       "      <td>163907.0</td>\n",
       "      <td>-3907.0</td>\n",
       "      <td>-2.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>224</td>\n",
       "      <td>129000</td>\n",
       "      <td>88674.0</td>\n",
       "      <td>40326.0</td>\n",
       "      <td>31.26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>225</td>\n",
       "      <td>175000</td>\n",
       "      <td>176590.0</td>\n",
       "      <td>-1590.0</td>\n",
       "      <td>-0.91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>226</td>\n",
       "      <td>210000</td>\n",
       "      <td>233429.0</td>\n",
       "      <td>-23429.0</td>\n",
       "      <td>-11.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>227</td>\n",
       "      <td>266500</td>\n",
       "      <td>265809.0</td>\n",
       "      <td>691.0</td>\n",
       "      <td>0.26</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>228 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     actual_price  predicted_price  residual  percent_diff\n",
       "0          208500         210804.0   -2304.0         -1.11\n",
       "1          181500         198451.0  -16951.0         -9.34\n",
       "2          223500         220067.0    3433.0          1.54\n",
       "3          250000         311919.0  -61919.0        -24.77\n",
       "4          279500         230839.0   48661.0         17.41\n",
       "..            ...              ...       ...           ...\n",
       "223        160000         163907.0   -3907.0         -2.44\n",
       "224        129000          88674.0   40326.0         31.26\n",
       "225        175000         176590.0   -1590.0         -0.91\n",
       "226        210000         233429.0  -23429.0        -11.16\n",
       "227        266500         265809.0     691.0          0.26\n",
       "\n",
       "[228 rows x 4 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res['residual'] = res.actual_price - res.predicted_price\n",
    "res['percent_diff'] = (res['residual'] / res.actual_price * 100).round(decimals=2)\n",
    "res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average percent_difference: 13.84\n"
     ]
    }
   ],
   "source": [
    "print('Average percent_difference: %.2f' % (res.percent_diff.abs().mean()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x268b3633748>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "res.percent_diff.plot.kde()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x268b3520988>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Predicted price is actually really close to actual, but not when actual price is very very high\n",
    "res.plot(kind='line', y=['actual_price', 'predicted_price'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x268b3b13e48>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "res.plot(kind='line', y='residual')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.7.4 64-bit ('base': conda)",
   "language": "python",
   "name": "python37464bitbaseconda536c566210f0426bb8cfea15c7ee20a0"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}