Walmart Store Sales Forecasting.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",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datetime import datetime as dt\n",
    "from sklearn import metrics\n",
    "import seaborn as sns\n",
    "import pandas_profiling\n",
    "from sklearn.model_selection import cross_val_score\n",
    "import statsmodels.formula.api as smf\n",
    "from sklearn.ensemble import RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "train=pd.read_csv('data/train.csv',parse_dates=['Date'])\n",
    "features=pd.read_csv('data/features.csv',parse_dates=['Date'])\n",
    "stores=pd.read_csv('data/stores.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Store</th>\n",
       "      <th>Date</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Fuel_Price</th>\n",
       "      <th>MarkDown1</th>\n",
       "      <th>MarkDown2</th>\n",
       "      <th>MarkDown3</th>\n",
       "      <th>MarkDown4</th>\n",
       "      <th>MarkDown5</th>\n",
       "      <th>CPI</th>\n",
       "      <th>Unemployment</th>\n",
       "      <th>IsHoliday</th>\n",
       "      <th>Type</th>\n",
       "      <th>Size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>2010-02-05</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>False</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2010-02-12</td>\n",
       "      <td>38.51</td>\n",
       "      <td>2.548</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>211.242170</td>\n",
       "      <td>8.106</td>\n",
       "      <td>True</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>2010-02-19</td>\n",
       "      <td>39.93</td>\n",
       "      <td>2.514</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>211.289143</td>\n",
       "      <td>8.106</td>\n",
       "      <td>False</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>2010-02-26</td>\n",
       "      <td>46.63</td>\n",
       "      <td>2.561</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>211.319643</td>\n",
       "      <td>8.106</td>\n",
       "      <td>False</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>2010-03-05</td>\n",
       "      <td>46.50</td>\n",
       "      <td>2.625</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>211.350143</td>\n",
       "      <td>8.106</td>\n",
       "      <td>False</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Store       Date  Temperature  Fuel_Price  MarkDown1  MarkDown2  MarkDown3  \\\n",
       "0      1 2010-02-05        42.31       2.572        NaN        NaN        NaN   \n",
       "1      1 2010-02-12        38.51       2.548        NaN        NaN        NaN   \n",
       "2      1 2010-02-19        39.93       2.514        NaN        NaN        NaN   \n",
       "3      1 2010-02-26        46.63       2.561        NaN        NaN        NaN   \n",
       "4      1 2010-03-05        46.50       2.625        NaN        NaN        NaN   \n",
       "\n",
       "   MarkDown4  MarkDown5         CPI  Unemployment  IsHoliday Type    Size  \n",
       "0        NaN        NaN  211.096358         8.106      False    A  151315  \n",
       "1        NaN        NaN  211.242170         8.106       True    A  151315  \n",
       "2        NaN        NaN  211.289143         8.106      False    A  151315  \n",
       "3        NaN        NaN  211.319643         8.106      False    A  151315  \n",
       "4        NaN        NaN  211.350143         8.106      False    A  151315  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1=pd.merge(features,stores,on=['Store'])\n",
    "df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 8190 entries, 0 to 8189\n",
      "Data columns (total 14 columns):\n",
      " #   Column        Non-Null Count  Dtype         \n",
      "---  ------        --------------  -----         \n",
      " 0   Store         8190 non-null   int64         \n",
      " 1   Date          8190 non-null   datetime64[ns]\n",
      " 2   Temperature   8190 non-null   float64       \n",
      " 3   Fuel_Price    8190 non-null   float64       \n",
      " 4   MarkDown1     4032 non-null   float64       \n",
      " 5   MarkDown2     2921 non-null   float64       \n",
      " 6   MarkDown3     3613 non-null   float64       \n",
      " 7   MarkDown4     3464 non-null   float64       \n",
      " 8   MarkDown5     4050 non-null   float64       \n",
      " 9   CPI           7605 non-null   float64       \n",
      " 10  Unemployment  7605 non-null   float64       \n",
      " 11  IsHoliday     8190 non-null   bool          \n",
      " 12  Type          8190 non-null   object        \n",
      " 13  Size          8190 non-null   int64         \n",
      "dtypes: bool(1), datetime64[ns](1), float64(9), int64(2), object(1)\n",
      "memory usage: 903.8+ KB\n"
     ]
    }
   ],
   "source": [
    "df1.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1.drop(['MarkDown1','MarkDown2','MarkDown3','MarkDown4','MarkDown5'],axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>Store</th>\n",
       "      <th>Date</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Fuel_Price</th>\n",
       "      <th>CPI</th>\n",
       "      <th>Unemployment</th>\n",
       "      <th>IsHoliday</th>\n",
       "      <th>Type</th>\n",
       "      <th>Size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>2010-02-05</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>False</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2010-02-12</td>\n",
       "      <td>38.51</td>\n",
       "      <td>2.548</td>\n",
       "      <td>211.242170</td>\n",
       "      <td>8.106</td>\n",
       "      <td>True</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>2010-02-19</td>\n",
       "      <td>39.93</td>\n",
       "      <td>2.514</td>\n",
       "      <td>211.289143</td>\n",
       "      <td>8.106</td>\n",
       "      <td>False</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>2010-02-26</td>\n",
       "      <td>46.63</td>\n",
       "      <td>2.561</td>\n",
       "      <td>211.319643</td>\n",
       "      <td>8.106</td>\n",
       "      <td>False</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>2010-03-05</td>\n",
       "      <td>46.50</td>\n",
       "      <td>2.625</td>\n",
       "      <td>211.350143</td>\n",
       "      <td>8.106</td>\n",
       "      <td>False</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Store       Date  Temperature  Fuel_Price         CPI  Unemployment  \\\n",
       "0      1 2010-02-05        42.31       2.572  211.096358         8.106   \n",
       "1      1 2010-02-12        38.51       2.548  211.242170         8.106   \n",
       "2      1 2010-02-19        39.93       2.514  211.289143         8.106   \n",
       "3      1 2010-02-26        46.63       2.561  211.319643         8.106   \n",
       "4      1 2010-03-05        46.50       2.625  211.350143         8.106   \n",
       "\n",
       "   IsHoliday Type    Size  \n",
       "0      False    A  151315  \n",
       "1       True    A  151315  \n",
       "2      False    A  151315  \n",
       "3      False    A  151315  \n",
       "4      False    A  151315  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1.to_csv('all_features.csv',index=False)"
   ]
  },
  {
   "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>Store</th>\n",
       "      <th>Dept</th>\n",
       "      <th>Date</th>\n",
       "      <th>Weekly_Sales</th>\n",
       "      <th>IsHoliday</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Fuel_Price</th>\n",
       "      <th>CPI</th>\n",
       "      <th>Unemployment</th>\n",
       "      <th>Type</th>\n",
       "      <th>Size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2010-02-05</td>\n",
       "      <td>24924.50</td>\n",
       "      <td>False</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2010-02-05</td>\n",
       "      <td>50605.27</td>\n",
       "      <td>False</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2010-02-05</td>\n",
       "      <td>13740.12</td>\n",
       "      <td>False</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>2010-02-05</td>\n",
       "      <td>39954.04</td>\n",
       "      <td>False</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>2010-02-05</td>\n",
       "      <td>32229.38</td>\n",
       "      <td>False</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Store  Dept       Date  Weekly_Sales  IsHoliday  Temperature  Fuel_Price  \\\n",
       "0      1     1 2010-02-05      24924.50      False        42.31       2.572   \n",
       "1      1     2 2010-02-05      50605.27      False        42.31       2.572   \n",
       "2      1     3 2010-02-05      13740.12      False        42.31       2.572   \n",
       "3      1     4 2010-02-05      39954.04      False        42.31       2.572   \n",
       "4      1     5 2010-02-05      32229.38      False        42.31       2.572   \n",
       "\n",
       "          CPI  Unemployment Type    Size  \n",
       "0  211.096358         8.106    A  151315  \n",
       "1  211.096358         8.106    A  151315  \n",
       "2  211.096358         8.106    A  151315  \n",
       "3  211.096358         8.106    A  151315  \n",
       "4  211.096358         8.106    A  151315  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data=pd.merge(train,df1,on=['Date','Store','IsHoliday'])\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(421570, 11)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "y=[]\n",
    "m=[]\n",
    "d=[]\n",
    "for dt in data['Date']:\n",
    "    y.append(dt.year)\n",
    "    m.append(dt.month)\n",
    "    d.append(dt.day)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['Year'] = pd.Series(y)\n",
    "data['Month'] = pd.Series(m)\n",
    "data['Day'] = pd.Series(d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop('Date',axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>Store</th>\n",
       "      <th>Dept</th>\n",
       "      <th>Weekly_Sales</th>\n",
       "      <th>IsHoliday</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Fuel_Price</th>\n",
       "      <th>CPI</th>\n",
       "      <th>Unemployment</th>\n",
       "      <th>Type</th>\n",
       "      <th>Size</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>24924.50</td>\n",
       "      <td>False</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>50605.27</td>\n",
       "      <td>False</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>13740.12</td>\n",
       "      <td>False</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>39954.04</td>\n",
       "      <td>False</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>32229.38</td>\n",
       "      <td>False</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>A</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Store  Dept  Weekly_Sales  IsHoliday  Temperature  Fuel_Price         CPI  \\\n",
       "0      1     1      24924.50      False        42.31       2.572  211.096358   \n",
       "1      1     2      50605.27      False        42.31       2.572  211.096358   \n",
       "2      1     3      13740.12      False        42.31       2.572  211.096358   \n",
       "3      1     4      39954.04      False        42.31       2.572  211.096358   \n",
       "4      1     5      32229.38      False        42.31       2.572  211.096358   \n",
       "\n",
       "   Unemployment Type    Size  Year  Month  Day  \n",
       "0         8.106    A  151315  2010      2    5  \n",
       "1         8.106    A  151315  2010      2    5  \n",
       "2         8.106    A  151315  2010      2    5  \n",
       "3         8.106    A  151315  2010      2    5  \n",
       "4         8.106    A  151315  2010      2    5  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "numeric_var_names=[key for key in dict(data.dtypes) if dict(data.dtypes)[key] in ['float64', 'int64', 'float32', 'int32']]\n",
    "cat_var_names=[key for key in dict(data.dtypes) if dict(data.dtypes)[key] in ['object', 'O','bool']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Store',\n",
       " 'Dept',\n",
       " 'Weekly_Sales',\n",
       " 'Temperature',\n",
       " 'Fuel_Price',\n",
       " 'CPI',\n",
       " 'Unemployment',\n",
       " 'Size',\n",
       " 'Year',\n",
       " 'Month',\n",
       " 'Day']"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "numeric_var_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['IsHoliday', 'Type']"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_var_names  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>Store</th>\n",
       "      <th>Dept</th>\n",
       "      <th>Weekly_Sales</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Fuel_Price</th>\n",
       "      <th>CPI</th>\n",
       "      <th>Unemployment</th>\n",
       "      <th>Size</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>24924.50</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>50605.27</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>13740.12</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>39954.04</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>32229.38</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Store  Dept  Weekly_Sales  Temperature  Fuel_Price         CPI  \\\n",
       "0      1     1      24924.50        42.31       2.572  211.096358   \n",
       "1      1     2      50605.27        42.31       2.572  211.096358   \n",
       "2      1     3      13740.12        42.31       2.572  211.096358   \n",
       "3      1     4      39954.04        42.31       2.572  211.096358   \n",
       "4      1     5      32229.38        42.31       2.572  211.096358   \n",
       "\n",
       "   Unemployment    Size  Year  Month  Day  \n",
       "0         8.106  151315  2010      2    5  \n",
       "1         8.106  151315  2010      2    5  \n",
       "2         8.106  151315  2010      2    5  \n",
       "3         8.106  151315  2010      2    5  \n",
       "4         8.106  151315  2010      2    5  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_num=data[numeric_var_names]\n",
    "data_num.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Replace negative and 0 as missing\n",
    "pd.set_option('mode.chained_assignment', None) # For SettingWithCopyWarning\n",
    "data_num[data_num<=0]=np.NaN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>IsHoliday</th>\n",
       "      <th>Type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>False</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>False</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>False</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>False</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>False</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   IsHoliday Type\n",
       "0      False    A\n",
       "1      False    A\n",
       "2      False    A\n",
       "3      False    A\n",
       "4      False    A"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_cat=data[cat_var_names]\n",
    "data_cat.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data Audit Report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>N</th>\n",
       "      <th>NMISS</th>\n",
       "      <th>SUM</th>\n",
       "      <th>MEAN</th>\n",
       "      <th>MEDIAN</th>\n",
       "      <th>STD</th>\n",
       "      <th>VAR</th>\n",
       "      <th>CV</th>\n",
       "      <th>MIN</th>\n",
       "      <th>P1</th>\n",
       "      <th>P5</th>\n",
       "      <th>P10</th>\n",
       "      <th>P25</th>\n",
       "      <th>P50</th>\n",
       "      <th>P75</th>\n",
       "      <th>P90</th>\n",
       "      <th>P95</th>\n",
       "      <th>P99</th>\n",
       "      <th>MAX</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Store</th>\n",
       "      <td>421570.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.359084e+06</td>\n",
       "      <td>22.200546</td>\n",
       "      <td>22.00000</td>\n",
       "      <td>12.785297</td>\n",
       "      <td>1.634638e+02</td>\n",
       "      <td>0.575900</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>22.00000</td>\n",
       "      <td>33.000000</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>43.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dept</th>\n",
       "      <td>421570.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.865882e+07</td>\n",
       "      <td>44.260317</td>\n",
       "      <td>37.00000</td>\n",
       "      <td>30.492054</td>\n",
       "      <td>9.297654e+02</td>\n",
       "      <td>0.688925</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>37.00000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>92.000000</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>98.000000</td>\n",
       "      <td>99.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Weekly_Sales</th>\n",
       "      <td>420212.0</td>\n",
       "      <td>1358.0</td>\n",
       "      <td>6.737307e+09</td>\n",
       "      <td>16033.114591</td>\n",
       "      <td>7661.70000</td>\n",
       "      <td>22729.492116</td>\n",
       "      <td>5.166298e+08</td>\n",
       "      <td>1.417659</td>\n",
       "      <td>0.010</td>\n",
       "      <td>7.440000</td>\n",
       "      <td>67.551000</td>\n",
       "      <td>311.186000</td>\n",
       "      <td>2120.130000</td>\n",
       "      <td>7661.70000</td>\n",
       "      <td>20271.265000</td>\n",
       "      <td>42922.748000</td>\n",
       "      <td>61275.522000</td>\n",
       "      <td>106566.351400</td>\n",
       "      <td>693099.360000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Temperature</th>\n",
       "      <td>421501.0</td>\n",
       "      <td>69.0</td>\n",
       "      <td>2.533231e+07</td>\n",
       "      <td>60.100233</td>\n",
       "      <td>62.09000</td>\n",
       "      <td>18.432294</td>\n",
       "      <td>3.397495e+02</td>\n",
       "      <td>0.306693</td>\n",
       "      <td>5.540</td>\n",
       "      <td>18.360000</td>\n",
       "      <td>27.310000</td>\n",
       "      <td>33.990000</td>\n",
       "      <td>46.700000</td>\n",
       "      <td>62.09000</td>\n",
       "      <td>74.280000</td>\n",
       "      <td>83.580000</td>\n",
       "      <td>87.270000</td>\n",
       "      <td>92.810000</td>\n",
       "      <td>100.140000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fuel_Price</th>\n",
       "      <td>421570.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.416908e+06</td>\n",
       "      <td>3.361027</td>\n",
       "      <td>3.45200</td>\n",
       "      <td>0.458515</td>\n",
       "      <td>2.102356e-01</td>\n",
       "      <td>0.136421</td>\n",
       "      <td>2.472</td>\n",
       "      <td>2.565000</td>\n",
       "      <td>2.653000</td>\n",
       "      <td>2.720000</td>\n",
       "      <td>2.933000</td>\n",
       "      <td>3.45200</td>\n",
       "      <td>3.738000</td>\n",
       "      <td>3.917000</td>\n",
       "      <td>4.029000</td>\n",
       "      <td>4.202000</td>\n",
       "      <td>4.468000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CPI</th>\n",
       "      <td>421570.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.217360e+07</td>\n",
       "      <td>171.201947</td>\n",
       "      <td>182.31878</td>\n",
       "      <td>39.159276</td>\n",
       "      <td>1.533449e+03</td>\n",
       "      <td>0.228731</td>\n",
       "      <td>126.064</td>\n",
       "      <td>126.111903</td>\n",
       "      <td>126.496258</td>\n",
       "      <td>128.823806</td>\n",
       "      <td>132.022667</td>\n",
       "      <td>182.31878</td>\n",
       "      <td>212.416993</td>\n",
       "      <td>219.444244</td>\n",
       "      <td>221.941558</td>\n",
       "      <td>225.473509</td>\n",
       "      <td>227.232807</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Unemployment</th>\n",
       "      <td>421570.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.355819e+06</td>\n",
       "      <td>7.960289</td>\n",
       "      <td>7.86600</td>\n",
       "      <td>1.863296</td>\n",
       "      <td>3.471872e+00</td>\n",
       "      <td>0.234074</td>\n",
       "      <td>3.879</td>\n",
       "      <td>4.156000</td>\n",
       "      <td>5.326000</td>\n",
       "      <td>5.965000</td>\n",
       "      <td>6.891000</td>\n",
       "      <td>7.86600</td>\n",
       "      <td>8.572000</td>\n",
       "      <td>9.816000</td>\n",
       "      <td>12.187000</td>\n",
       "      <td>14.180000</td>\n",
       "      <td>14.313000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Size</th>\n",
       "      <td>421570.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.764039e+10</td>\n",
       "      <td>136727.915739</td>\n",
       "      <td>140167.00000</td>\n",
       "      <td>60980.583328</td>\n",
       "      <td>3.718632e+09</td>\n",
       "      <td>0.446000</td>\n",
       "      <td>34875.000</td>\n",
       "      <td>34875.000000</td>\n",
       "      <td>39690.000000</td>\n",
       "      <td>39910.000000</td>\n",
       "      <td>93638.000000</td>\n",
       "      <td>140167.00000</td>\n",
       "      <td>202505.000000</td>\n",
       "      <td>204184.000000</td>\n",
       "      <td>206302.000000</td>\n",
       "      <td>219622.000000</td>\n",
       "      <td>219622.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Year</th>\n",
       "      <td>421570.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8.477640e+08</td>\n",
       "      <td>2010.968591</td>\n",
       "      <td>2011.00000</td>\n",
       "      <td>0.796876</td>\n",
       "      <td>6.350114e-01</td>\n",
       "      <td>0.000396</td>\n",
       "      <td>2010.000</td>\n",
       "      <td>2010.000000</td>\n",
       "      <td>2010.000000</td>\n",
       "      <td>2010.000000</td>\n",
       "      <td>2010.000000</td>\n",
       "      <td>2011.00000</td>\n",
       "      <td>2012.000000</td>\n",
       "      <td>2012.000000</td>\n",
       "      <td>2012.000000</td>\n",
       "      <td>2012.000000</td>\n",
       "      <td>2012.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Month</th>\n",
       "      <td>421570.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.718920e+06</td>\n",
       "      <td>6.449510</td>\n",
       "      <td>6.00000</td>\n",
       "      <td>3.243217</td>\n",
       "      <td>1.051846e+01</td>\n",
       "      <td>0.502863</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>6.00000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>12.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Day</th>\n",
       "      <td>421570.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.607322e+06</td>\n",
       "      <td>15.673131</td>\n",
       "      <td>16.00000</td>\n",
       "      <td>8.753549</td>\n",
       "      <td>7.662462e+01</td>\n",
       "      <td>0.558507</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>16.00000</td>\n",
       "      <td>23.000000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>31.000000</td>\n",
       "      <td>31.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     N   NMISS           SUM           MEAN        MEDIAN  \\\n",
       "Store         421570.0     0.0  9.359084e+06      22.200546      22.00000   \n",
       "Dept          421570.0     0.0  1.865882e+07      44.260317      37.00000   \n",
       "Weekly_Sales  420212.0  1358.0  6.737307e+09   16033.114591    7661.70000   \n",
       "Temperature   421501.0    69.0  2.533231e+07      60.100233      62.09000   \n",
       "Fuel_Price    421570.0     0.0  1.416908e+06       3.361027       3.45200   \n",
       "CPI           421570.0     0.0  7.217360e+07     171.201947     182.31878   \n",
       "Unemployment  421570.0     0.0  3.355819e+06       7.960289       7.86600   \n",
       "Size          421570.0     0.0  5.764039e+10  136727.915739  140167.00000   \n",
       "Year          421570.0     0.0  8.477640e+08    2010.968591    2011.00000   \n",
       "Month         421570.0     0.0  2.718920e+06       6.449510       6.00000   \n",
       "Day           421570.0     0.0  6.607322e+06      15.673131      16.00000   \n",
       "\n",
       "                       STD           VAR        CV        MIN            P1  \\\n",
       "Store            12.785297  1.634638e+02  0.575900      1.000      1.000000   \n",
       "Dept             30.492054  9.297654e+02  0.688925      1.000      1.000000   \n",
       "Weekly_Sales  22729.492116  5.166298e+08  1.417659      0.010      7.440000   \n",
       "Temperature      18.432294  3.397495e+02  0.306693      5.540     18.360000   \n",
       "Fuel_Price        0.458515  2.102356e-01  0.136421      2.472      2.565000   \n",
       "CPI              39.159276  1.533449e+03  0.228731    126.064    126.111903   \n",
       "Unemployment      1.863296  3.471872e+00  0.234074      3.879      4.156000   \n",
       "Size          60980.583328  3.718632e+09  0.446000  34875.000  34875.000000   \n",
       "Year              0.796876  6.350114e-01  0.000396   2010.000   2010.000000   \n",
       "Month             3.243217  1.051846e+01  0.502863      1.000      1.000000   \n",
       "Day               8.753549  7.662462e+01  0.558507      1.000      1.000000   \n",
       "\n",
       "                        P5           P10           P25           P50  \\\n",
       "Store             3.000000      5.000000     11.000000      22.00000   \n",
       "Dept              4.000000      7.000000     18.000000      37.00000   \n",
       "Weekly_Sales     67.551000    311.186000   2120.130000    7661.70000   \n",
       "Temperature      27.310000     33.990000     46.700000      62.09000   \n",
       "Fuel_Price        2.653000      2.720000      2.933000       3.45200   \n",
       "CPI             126.496258    128.823806    132.022667     182.31878   \n",
       "Unemployment      5.326000      5.965000      6.891000       7.86600   \n",
       "Size          39690.000000  39910.000000  93638.000000  140167.00000   \n",
       "Year           2010.000000   2010.000000   2010.000000    2011.00000   \n",
       "Month             1.000000      2.000000      4.000000       6.00000   \n",
       "Day               2.000000      4.000000      8.000000      16.00000   \n",
       "\n",
       "                        P75            P90            P95            P99  \\\n",
       "Store             33.000000      40.000000      43.000000      45.000000   \n",
       "Dept              74.000000      92.000000      95.000000      98.000000   \n",
       "Weekly_Sales   20271.265000   42922.748000   61275.522000  106566.351400   \n",
       "Temperature       74.280000      83.580000      87.270000      92.810000   \n",
       "Fuel_Price         3.738000       3.917000       4.029000       4.202000   \n",
       "CPI              212.416993     219.444244     221.941558     225.473509   \n",
       "Unemployment       8.572000       9.816000      12.187000      14.180000   \n",
       "Size          202505.000000  204184.000000  206302.000000  219622.000000   \n",
       "Year            2012.000000    2012.000000    2012.000000    2012.000000   \n",
       "Month              9.000000      11.000000      12.000000      12.000000   \n",
       "Day               23.000000      28.000000      29.000000      31.000000   \n",
       "\n",
       "                        MAX  \n",
       "Store             45.000000  \n",
       "Dept              99.000000  \n",
       "Weekly_Sales  693099.360000  \n",
       "Temperature      100.140000  \n",
       "Fuel_Price         4.468000  \n",
       "CPI              227.232807  \n",
       "Unemployment      14.313000  \n",
       "Size          219622.000000  \n",
       "Year            2012.000000  \n",
       "Month             12.000000  \n",
       "Day               31.000000  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating Data audit Report\n",
    "def var_summary(x):\n",
    "    return pd.Series([x.count(), x.isnull().sum(), x.sum(), x.mean(), x.median(),  x.std(), x.var(), x.std()/x.mean(), x.min(), x.dropna().quantile(0.01), x.dropna().quantile(0.05),x.dropna().quantile(0.10),x.dropna().quantile(0.25),x.dropna().quantile(0.50),x.dropna().quantile(0.75), x.dropna().quantile(0.90),x.dropna().quantile(0.95), x.dropna().quantile(0.99),x.max()], \n",
    "                  index=['N', 'NMISS', 'SUM', 'MEAN','MEDIAN', 'STD', 'VAR', 'CV','MIN', 'P1' , 'P5' ,'P10' ,'P25' ,'P50' ,'P75' ,'P90' ,'P95' ,'P99' ,'MAX'])\n",
    "\n",
    "data_num.apply(var_summary).T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_summary=data_num.apply(var_summary).T\n",
    "num_summary.to_csv('num_summary.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def cat_summary(x):\n",
    "    return pd.Series([x.count(), x.isnull().sum(), x.value_counts()], \n",
    "                  index=['N', 'NMISS', 'ColumnsNames'])\n",
    "\n",
    "cat_summary=data_cat.apply(cat_summary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>IsHoliday</th>\n",
       "      <th>Type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>N</th>\n",
       "      <td>421570</td>\n",
       "      <td>421570</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NMISS</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ColumnsNames</th>\n",
       "      <td>False    391909\n",
       "True      29661\n",
       "Name: IsHolida...</td>\n",
       "      <td>A    215478\n",
       "B    163495\n",
       "C     42597\n",
       "Name: Type...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                      IsHoliday  \\\n",
       "N                                                        421570   \n",
       "NMISS                                                         0   \n",
       "ColumnsNames  False    391909\n",
       "True      29661\n",
       "Name: IsHolida...   \n",
       "\n",
       "                                                           Type  \n",
       "N                                                        421570  \n",
       "NMISS                                                         0  \n",
       "ColumnsNames  A    215478\n",
       "B    163495\n",
       "C     42597\n",
       "Name: Type...  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_summary"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Handling Outlier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Handling Outliers\n",
    "def outlier_capping(x):\n",
    "    x = x.clip(upper=x.quantile(0.99))\n",
    "    x = x.clip(lower=x.quantile(0.01))\n",
    "    return x\n",
    "\n",
    "data_num=data_num.apply(outlier_capping)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Handling Missing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Handling missings\n",
    "def Missing_imputation(x):\n",
    "    x = x.fillna(x.median())\n",
    "    return x\n",
    "\n",
    "data_num=data_num.apply(Missing_imputation)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dummy Creation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# An utility function to create dummy variable\n",
    "def create_dummies( df, colname ):\n",
    "    col_dummies = pd.get_dummies(df[colname], prefix=colname, drop_first=True)\n",
    "    df = pd.concat([df, col_dummies], axis=1)\n",
    "    df.drop( colname, axis = 1, inplace = True )\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "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>IsHoliday_True</th>\n",
       "      <th>Type_B</th>\n",
       "      <th>Type_C</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   IsHoliday_True  Type_B  Type_C\n",
       "0               0       0       0\n",
       "1               0       0       0\n",
       "2               0       0       0\n",
       "3               0       0       0\n",
       "4               0       0       0"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for c_feature in ['IsHoliday', 'Type']:\n",
    "    data_cat.loc[:,c_feature] = data_cat[c_feature].astype('category')\n",
    "    data_cat = create_dummies(data_cat , c_feature )\n",
    "    \n",
    "data_cat.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Final file for analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Store</th>\n",
       "      <th>Dept</th>\n",
       "      <th>Weekly_Sales</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Fuel_Price</th>\n",
       "      <th>CPI</th>\n",
       "      <th>Unemployment</th>\n",
       "      <th>Size</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Day</th>\n",
       "      <th>IsHoliday_True</th>\n",
       "      <th>Type_B</th>\n",
       "      <th>Type_C</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>24924.50</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>50605.27</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>13740.12</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>39954.04</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>32229.38</td>\n",
       "      <td>42.31</td>\n",
       "      <td>2.572</td>\n",
       "      <td>211.096358</td>\n",
       "      <td>8.106</td>\n",
       "      <td>151315</td>\n",
       "      <td>2010</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Store  Dept  Weekly_Sales  Temperature  Fuel_Price         CPI  \\\n",
       "0      1     1      24924.50        42.31       2.572  211.096358   \n",
       "1      1     2      50605.27        42.31       2.572  211.096358   \n",
       "2      1     3      13740.12        42.31       2.572  211.096358   \n",
       "3      1     4      39954.04        42.31       2.572  211.096358   \n",
       "4      1     5      32229.38        42.31       2.572  211.096358   \n",
       "\n",
       "   Unemployment    Size  Year  Month  Day  IsHoliday_True  Type_B  Type_C  \n",
       "0         8.106  151315  2010      2    5               0       0       0  \n",
       "1         8.106  151315  2010      2    5               0       0       0  \n",
       "2         8.106  151315  2010      2    5               0       0       0  \n",
       "3         8.106  151315  2010      2    5               0       0       0  \n",
       "4         8.106  151315  2010      2    5               0       0       0  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.concat([data_num, data_cat], axis=1)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>Store</th>\n",
       "      <th>Dept</th>\n",
       "      <th>Weekly_Sales</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Fuel_Price</th>\n",
       "      <th>CPI</th>\n",
       "      <th>Unemployment</th>\n",
       "      <th>Size</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Day</th>\n",
       "      <th>IsHoliday_True</th>\n",
       "      <th>Type_B</th>\n",
       "      <th>Type_C</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Store</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.024011</td>\n",
       "      <td>-0.083732</td>\n",
       "      <td>-0.052251</td>\n",
       "      <td>0.064984</td>\n",
       "      <td>-0.210949</td>\n",
       "      <td>0.208671</td>\n",
       "      <td>-0.182881</td>\n",
       "      <td>0.002997</td>\n",
       "      <td>0.001011</td>\n",
       "      <td>-0.000015</td>\n",
       "      <td>-0.000548</td>\n",
       "      <td>-0.233461</td>\n",
       "      <td>0.439004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dept</th>\n",
       "      <td>0.024011</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.143319</td>\n",
       "      <td>0.004383</td>\n",
       "      <td>0.003514</td>\n",
       "      <td>-0.007483</td>\n",
       "      <td>0.007874</td>\n",
       "      <td>-0.003022</td>\n",
       "      <td>0.003688</td>\n",
       "      <td>0.000894</td>\n",
       "      <td>-0.000664</td>\n",
       "      <td>0.000906</td>\n",
       "      <td>-0.029920</td>\n",
       "      <td>0.028344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Weekly_Sales</th>\n",
       "      <td>-0.083732</td>\n",
       "      <td>0.143319</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000576</td>\n",
       "      <td>0.001502</td>\n",
       "      <td>-0.022807</td>\n",
       "      <td>-0.025323</td>\n",
       "      <td>0.252590</td>\n",
       "      <td>-0.009287</td>\n",
       "      <td>0.025578</td>\n",
       "      <td>-0.007982</td>\n",
       "      <td>0.007948</td>\n",
       "      <td>-0.134705</td>\n",
       "      <td>-0.099054</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Temperature</th>\n",
       "      <td>-0.052251</td>\n",
       "      <td>0.004383</td>\n",
       "      <td>0.000576</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.142011</td>\n",
       "      <td>0.184608</td>\n",
       "      <td>0.095511</td>\n",
       "      <td>-0.059287</td>\n",
       "      <td>0.066476</td>\n",
       "      <td>0.234499</td>\n",
       "      <td>0.025848</td>\n",
       "      <td>-0.156730</td>\n",
       "      <td>-0.107740</td>\n",
       "      <td>0.135709</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fuel_Price</th>\n",
       "      <td>0.064984</td>\n",
       "      <td>0.003514</td>\n",
       "      <td>0.001502</td>\n",
       "      <td>0.142011</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.162609</td>\n",
       "      <td>-0.034938</td>\n",
       "      <td>0.004468</td>\n",
       "      <td>0.780315</td>\n",
       "      <td>-0.042186</td>\n",
       "      <td>0.028013</td>\n",
       "      <td>-0.077839</td>\n",
       "      <td>0.037372</td>\n",
       "      <td>0.001369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CPI</th>\n",
       "      <td>-0.210949</td>\n",
       "      <td>-0.007483</td>\n",
       "      <td>-0.022807</td>\n",
       "      <td>0.184608</td>\n",
       "      <td>-0.162609</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.300190</td>\n",
       "      <td>-0.003307</td>\n",
       "      <td>0.074335</td>\n",
       "      <td>0.005130</td>\n",
       "      <td>0.002701</td>\n",
       "      <td>-0.001934</td>\n",
       "      <td>-0.081883</td>\n",
       "      <td>-0.006557</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Unemployment</th>\n",
       "      <td>0.208671</td>\n",
       "      <td>0.007874</td>\n",
       "      <td>-0.025323</td>\n",
       "      <td>0.095511</td>\n",
       "      <td>-0.034938</td>\n",
       "      <td>-0.300190</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.068099</td>\n",
       "      <td>-0.236754</td>\n",
       "      <td>-0.012963</td>\n",
       "      <td>-0.003802</td>\n",
       "      <td>0.010305</td>\n",
       "      <td>-0.013574</td>\n",
       "      <td>0.175291</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Size</th>\n",
       "      <td>-0.182881</td>\n",
       "      <td>-0.003022</td>\n",
       "      <td>0.252590</td>\n",
       "      <td>-0.059287</td>\n",
       "      <td>0.004468</td>\n",
       "      <td>-0.003307</td>\n",
       "      <td>-0.068099</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.004975</td>\n",
       "      <td>-0.001210</td>\n",
       "      <td>-0.000386</td>\n",
       "      <td>0.000593</td>\n",
       "      <td>-0.455646</td>\n",
       "      <td>-0.528852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Year</th>\n",
       "      <td>0.002997</td>\n",
       "      <td>0.003688</td>\n",
       "      <td>-0.009287</td>\n",
       "      <td>0.066476</td>\n",
       "      <td>0.780315</td>\n",
       "      <td>0.074335</td>\n",
       "      <td>-0.236754</td>\n",
       "      <td>-0.004975</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.194288</td>\n",
       "      <td>0.005835</td>\n",
       "      <td>-0.056746</td>\n",
       "      <td>-0.000732</td>\n",
       "      <td>0.005244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Month</th>\n",
       "      <td>0.001011</td>\n",
       "      <td>0.000894</td>\n",
       "      <td>0.025578</td>\n",
       "      <td>0.234499</td>\n",
       "      <td>-0.042186</td>\n",
       "      <td>0.005130</td>\n",
       "      <td>-0.012963</td>\n",
       "      <td>-0.001210</td>\n",
       "      <td>-0.194288</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.016363</td>\n",
       "      <td>0.123376</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>-0.000092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Day</th>\n",
       "      <td>-0.000015</td>\n",
       "      <td>-0.000664</td>\n",
       "      <td>-0.007982</td>\n",
       "      <td>0.025848</td>\n",
       "      <td>0.028013</td>\n",
       "      <td>0.002701</td>\n",
       "      <td>-0.003802</td>\n",
       "      <td>-0.000386</td>\n",
       "      <td>0.005835</td>\n",
       "      <td>0.016363</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.045465</td>\n",
       "      <td>0.000503</td>\n",
       "      <td>-0.000098</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IsHoliday_True</th>\n",
       "      <td>-0.000548</td>\n",
       "      <td>0.000906</td>\n",
       "      <td>0.007948</td>\n",
       "      <td>-0.156730</td>\n",
       "      <td>-0.077839</td>\n",
       "      <td>-0.001934</td>\n",
       "      <td>0.010305</td>\n",
       "      <td>0.000593</td>\n",
       "      <td>-0.056746</td>\n",
       "      <td>0.123376</td>\n",
       "      <td>0.045465</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000167</td>\n",
       "      <td>-0.001017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Type_B</th>\n",
       "      <td>-0.233461</td>\n",
       "      <td>-0.029920</td>\n",
       "      <td>-0.134705</td>\n",
       "      <td>-0.107740</td>\n",
       "      <td>0.037372</td>\n",
       "      <td>-0.081883</td>\n",
       "      <td>-0.013574</td>\n",
       "      <td>-0.455646</td>\n",
       "      <td>-0.000732</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>0.000503</td>\n",
       "      <td>0.000167</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.266848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Type_C</th>\n",
       "      <td>0.439004</td>\n",
       "      <td>0.028344</td>\n",
       "      <td>-0.099054</td>\n",
       "      <td>0.135709</td>\n",
       "      <td>0.001369</td>\n",
       "      <td>-0.006557</td>\n",
       "      <td>0.175291</td>\n",
       "      <td>-0.528852</td>\n",
       "      <td>0.005244</td>\n",
       "      <td>-0.000092</td>\n",
       "      <td>-0.000098</td>\n",
       "      <td>-0.001017</td>\n",
       "      <td>-0.266848</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   Store      Dept  Weekly_Sales  Temperature  Fuel_Price  \\\n",
       "Store           1.000000  0.024011     -0.083732    -0.052251    0.064984   \n",
       "Dept            0.024011  1.000000      0.143319     0.004383    0.003514   \n",
       "Weekly_Sales   -0.083732  0.143319      1.000000     0.000576    0.001502   \n",
       "Temperature    -0.052251  0.004383      0.000576     1.000000    0.142011   \n",
       "Fuel_Price      0.064984  0.003514      0.001502     0.142011    1.000000   \n",
       "CPI            -0.210949 -0.007483     -0.022807     0.184608   -0.162609   \n",
       "Unemployment    0.208671  0.007874     -0.025323     0.095511   -0.034938   \n",
       "Size           -0.182881 -0.003022      0.252590    -0.059287    0.004468   \n",
       "Year            0.002997  0.003688     -0.009287     0.066476    0.780315   \n",
       "Month           0.001011  0.000894      0.025578     0.234499   -0.042186   \n",
       "Day            -0.000015 -0.000664     -0.007982     0.025848    0.028013   \n",
       "IsHoliday_True -0.000548  0.000906      0.007948    -0.156730   -0.077839   \n",
       "Type_B         -0.233461 -0.029920     -0.134705    -0.107740    0.037372   \n",
       "Type_C          0.439004  0.028344     -0.099054     0.135709    0.001369   \n",
       "\n",
       "                     CPI  Unemployment      Size      Year     Month  \\\n",
       "Store          -0.210949      0.208671 -0.182881  0.002997  0.001011   \n",
       "Dept           -0.007483      0.007874 -0.003022  0.003688  0.000894   \n",
       "Weekly_Sales   -0.022807     -0.025323  0.252590 -0.009287  0.025578   \n",
       "Temperature     0.184608      0.095511 -0.059287  0.066476  0.234499   \n",
       "Fuel_Price     -0.162609     -0.034938  0.004468  0.780315 -0.042186   \n",
       "CPI             1.000000     -0.300190 -0.003307  0.074335  0.005130   \n",
       "Unemployment   -0.300190      1.000000 -0.068099 -0.236754 -0.012963   \n",
       "Size           -0.003307     -0.068099  1.000000 -0.004975 -0.001210   \n",
       "Year            0.074335     -0.236754 -0.004975  1.000000 -0.194288   \n",
       "Month           0.005130     -0.012963 -0.001210 -0.194288  1.000000   \n",
       "Day             0.002701     -0.003802 -0.000386  0.005835  0.016363   \n",
       "IsHoliday_True -0.001934      0.010305  0.000593 -0.056746  0.123376   \n",
       "Type_B         -0.081883     -0.013574 -0.455646 -0.000732  0.000019   \n",
       "Type_C         -0.006557      0.175291 -0.528852  0.005244 -0.000092   \n",
       "\n",
       "                     Day  IsHoliday_True    Type_B    Type_C  \n",
       "Store          -0.000015       -0.000548 -0.233461  0.439004  \n",
       "Dept           -0.000664        0.000906 -0.029920  0.028344  \n",
       "Weekly_Sales   -0.007982        0.007948 -0.134705 -0.099054  \n",
       "Temperature     0.025848       -0.156730 -0.107740  0.135709  \n",
       "Fuel_Price      0.028013       -0.077839  0.037372  0.001369  \n",
       "CPI             0.002701       -0.001934 -0.081883 -0.006557  \n",
       "Unemployment   -0.003802        0.010305 -0.013574  0.175291  \n",
       "Size           -0.000386        0.000593 -0.455646 -0.528852  \n",
       "Year            0.005835       -0.056746 -0.000732  0.005244  \n",
       "Month           0.016363        0.123376  0.000019 -0.000092  \n",
       "Day             1.000000        0.045465  0.000503 -0.000098  \n",
       "IsHoliday_True  0.045465        1.000000  0.000167 -0.001017  \n",
       "Type_B          0.000503        0.000167  1.000000 -0.266848  \n",
       "Type_C         -0.000098       -0.001017 -0.266848  1.000000  "
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# correlation matrix (ranges from 1 to -1)\n",
    "corrm=data.corr()\n",
    "corrm.to_csv('corrm.csv')\n",
    "corrm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x24aaeb8a648>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualize correlation matrix in Seaborn using a heatmap\n",
    "sns.heatmap(data.corr())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Store', 'Dept', 'Weekly_Sales', 'Temperature', 'Fuel_Price', 'CPI',\n",
       "       'Unemployment', 'Size', 'Year', 'Month', 'Day', 'IsHoliday_True',\n",
       "       'Type_B', 'Type_C'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Building"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:           Weekly_Sales   R-squared:                       0.092\n",
      "Model:                            OLS   Adj. R-squared:                  0.092\n",
      "Method:                 Least Squares   F-statistic:                     3565.\n",
      "Date:                Mon, 15 Jun 2020   Prob (F-statistic):               0.00\n",
      "Time:                        23:01:18   Log-Likelihood:            -4.7688e+06\n",
      "No. Observations:              421570   AIC:                         9.538e+06\n",
      "Df Residuals:                  421557   BIC:                         9.538e+06\n",
      "Df Model:                          12                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==================================================================================\n",
      "                     coef    std err          t      P>|t|      [0.025      0.975]\n",
      "----------------------------------------------------------------------------------\n",
      "Intercept       6.153e+05   1.41e+05      4.364      0.000    3.39e+05    8.92e+05\n",
      "Store           -123.3926      2.765    -44.626      0.000    -128.812    -117.973\n",
      "Dept              97.7619      1.000     97.722      0.000      95.801      99.723\n",
      "Temperature        8.5571      1.874      4.566      0.000       4.884      12.230\n",
      "Fuel_Price       302.0885    120.409      2.509      0.012      66.090     538.087\n",
      "CPI              -22.4223      0.902    -24.857      0.000     -24.190     -20.654\n",
      "Unemployment    -240.0170     18.626    -12.886      0.000    -276.523    -203.511\n",
      "Size               0.0957      0.001    162.021      0.000       0.095       0.097\n",
      "Year            -303.9263     70.265     -4.325      0.000    -441.644    -166.209\n",
      "Month            136.9573     10.266     13.340      0.000     116.835     157.079\n",
      "Day              -20.6505      3.491     -5.916      0.000     -27.492     -13.809\n",
      "IsHoliday_True   538.1171    122.994      4.375      0.000     297.053     779.182\n",
      "Type_C          5603.5887    133.762     41.892      0.000    5341.419    5865.758\n",
      "==============================================================================\n",
      "Omnibus:                   160041.822   Durbin-Watson:                   1.275\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):           583787.922\n",
      "Skew:                           1.931   Prob(JB):                         0.00\n",
      "Kurtosis:                       7.280   Cond. No.                     6.93e+08\n",
      "==============================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "[2] The condition number is large, 6.93e+08. This might indicate that there are\n",
      "strong multicollinearity or other numerical problems.\n"
     ]
    }
   ],
   "source": [
    "lm=smf.ols('Weekly_Sales~Store+Dept+Temperature+Fuel_Price+CPI+Unemployment+Size+Year+Month+Day+IsHoliday_True++Type_C', data).fit()\n",
    "print(lm.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop('Type_B',axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "X=data[data.columns.difference(['Weekly_Sales'])]\n",
    "y=data['Weekly_Sales']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeRegressor  \n",
    "from sklearn.model_selection import cross_val_predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor(ccp_alpha=0.0, criterion='mse', max_depth=5,\n",
       "                      max_features=None, max_leaf_nodes=None,\n",
       "                      min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                      min_samples_leaf=1, min_samples_split=2,\n",
       "                      min_weight_fraction_leaf=0.0, presort='deprecated',\n",
       "                      random_state=0, splitter='best')"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regressor = DecisionTreeRegressor(max_depth=5,random_state=0)  \n",
    "regressor.fit(X_train, y_train) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred=regressor.predict(X_train)\n",
    "y_pred=regressor.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train : \n",
      "Mean Absolute Error: 9113.460578665334\n",
      "Mean Squared Error: 191174213.57522035\n",
      "Root Mean Squared Error: 13826.57635046436\n"
     ]
    }
   ],
   "source": [
    "print('Train : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_train, pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_train, pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_train, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test : \n",
      "Mean Absolute Error: 9186.83618067357\n",
      "Mean Squared Error: 194877398.09940347\n",
      "Root Mean Squared Error: 13959.84950131639\n"
     ]
    }
   ],
   "source": [
    "print('Test : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared for Train: 0.5546226722716019\n",
      "R-squared for Test: 0.5536357417116555\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import r2_score\n",
    "\n",
    "print(\"R-squared for Train:\",r2_score(y_train, pred))\n",
    "print(\"R-squared for Test:\",r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tuning a regression tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "# list of values to try\n",
    "max_depth_range = range(5, 15)\n",
    "\n",
    "# list to store the average RMSE for each value of max_depth\n",
    "RMSE_Scores = []\n",
    "MSE_Scores = []\n",
    "\n",
    "# use LOOCV with each value of max_depth\n",
    "for depth in max_depth_range:\n",
    "    treereg = DecisionTreeRegressor(max_depth=depth, random_state=345)\n",
    "    \n",
    "    MSE_scores = cross_val_score(treereg, X_train, y_train, scoring='neg_mean_squared_error')\n",
    "    \n",
    "    RMSE_Scores.append(np.mean(np.sqrt(-MSE_scores)))\n",
    "    MSE_Scores.append(MSE_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[13833.326858074766, 11024.756643306917, 10023.294758833663, 9007.377498713522, 8124.771136829191, 7238.566247975117, 6411.657165032743, 5726.776764045036, 5123.310696214251, 4637.977222783054]\n"
     ]
    }
   ],
   "source": [
    "print (RMSE_Scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'RMSE (lower is better)')"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot max_depth (x-axis) versus RMSE (y-axis)\n",
    "plt.plot(max_depth_range, RMSE_Scores)\n",
    "plt.xlabel('max_depth')\n",
    "plt.ylabel('RMSE (lower is better)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Building Final Decision Tree Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor(ccp_alpha=0.0, criterion='mse', max_depth=14,\n",
       "                      max_features=None, max_leaf_nodes=None,\n",
       "                      min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                      min_samples_leaf=1, min_samples_split=2,\n",
       "                      min_weight_fraction_leaf=0.0, presort='deprecated',\n",
       "                      random_state=345, splitter='best')"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# max_depth=11 was best, so fit a tree using that parameter\n",
    "treereg = DecisionTreeRegressor(max_depth=14, random_state=345)\n",
    "treereg.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2.14902285e-02, 7.00694314e-03, 6.83762529e-01, 1.35995066e-03,\n",
       "       1.43007854e-03, 1.61351466e-02, 1.90447029e-01, 5.43791520e-02,\n",
       "       5.53736048e-03, 1.20822802e-02, 6.03788651e-03, 3.31415733e-04])"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "treereg.feature_importances_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "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>feature</th>\n",
       "      <th>importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CPI</td>\n",
       "      <td>0.021490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Day</td>\n",
       "      <td>0.007007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Dept</td>\n",
       "      <td>0.683763</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Fuel_Price</td>\n",
       "      <td>0.001360</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>IsHoliday_True</td>\n",
       "      <td>0.001430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Month</td>\n",
       "      <td>0.016135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Size</td>\n",
       "      <td>0.190447</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Store</td>\n",
       "      <td>0.054379</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Temperature</td>\n",
       "      <td>0.005537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Type_C</td>\n",
       "      <td>0.012082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Unemployment</td>\n",
       "      <td>0.006038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Year</td>\n",
       "      <td>0.000331</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           feature  importance\n",
       "0              CPI    0.021490\n",
       "1              Day    0.007007\n",
       "2             Dept    0.683763\n",
       "3       Fuel_Price    0.001360\n",
       "4   IsHoliday_True    0.001430\n",
       "5            Month    0.016135\n",
       "6             Size    0.190447\n",
       "7            Store    0.054379\n",
       "8      Temperature    0.005537\n",
       "9           Type_C    0.012082\n",
       "10    Unemployment    0.006038\n",
       "11            Year    0.000331"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# \"Gini importance\" of each feature: the (normalized) total reduction of error brought by that feature\n",
    "pd.DataFrame({'feature':data.columns.difference(['Weekly_Sales']), 'importance':treereg.feature_importances_})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Creating a tree diagram"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create a Graphviz file\n",
    "from sklearn.tree import export_graphviz\n",
    "export_graphviz(treereg, out_file='tree_vehicles.png', feature_names=data.columns.difference(['Weekly_Sales']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Making predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred=treereg.predict(X_train)\n",
    "y_pred=treereg.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train : \n",
      "Mean Absolute Error: 2069.9751911511285\n",
      "Mean Squared Error: 16407973.300713757\n",
      "Root Mean Squared Error: 4050.6756597774843\n"
     ]
    }
   ],
   "source": [
    "print('Train : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_train, pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_train, pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_train, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test : \n",
      "Mean Absolute Error: 2371.930765936819\n",
      "Mean Squared Error: 21575823.535832647\n",
      "Root Mean Squared Error: 4644.978313817262\n"
     ]
    }
   ],
   "source": [
    "print('Test : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared for Train: 0.9617744508244808\n",
      "R-squared for Test: 0.9505808443490213\n"
     ]
    }
   ],
   "source": [
    "print(\"R-squared for Train:\",r2_score(y_train, pred))\n",
    "print(\"R-squared for Test:\",r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random Forest (RF)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "rfr = RandomForestRegressor(max_depth=5,n_estimators=20, random_state=0)  \n",
    "rfr.fit(X_train, y_train)  \n",
    "pred = rfr.predict(X_train) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred= rfr.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train : \n",
      "Mean Absolute Error: 9112.008923160754\n",
      "Mean Squared Error: 191091445.083154\n",
      "Root Mean Squared Error: 13823.582932190699\n"
     ]
    }
   ],
   "source": [
    "print('Train : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_train, pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_train, pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_train, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test : \n",
      "Mean Absolute Error: 9185.46851970489\n",
      "Mean Squared Error: 194781616.3495224\n",
      "Root Mean Squared Error: 13956.418464259461\n"
     ]
    }
   ],
   "source": [
    "print('Test : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared for Train: 0.5548154974917361\n",
      "R-squared for Test: 0.5538551286193225\n"
     ]
    }
   ],
   "source": [
    "print(\"R-squared for Train:\",r2_score(y_train, pred))\n",
    "print(\"R-squared for Test:\",r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tuning a Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "param_grid={'max_depth': range(8,15),\n",
    "            'n_estimators': (10, 50)}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Perform Grid-Search\n",
    "gsc = GridSearchCV(estimator=RandomForestRegressor(),param_grid=param_grid, verbose=0, n_jobs=-1)\n",
    "    \n",
    "grid_result = gsc.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9603400848923205"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_result.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'max_depth': 14, 'n_estimators': 50}"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_result.best_params_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Building Final Random Forest Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(bootstrap=True, ccp_alpha=0.0, criterion='mse',\n",
       "                      max_depth=14, max_features='auto', max_leaf_nodes=None,\n",
       "                      max_samples=None, min_impurity_decrease=0.0,\n",
       "                      min_impurity_split=None, min_samples_leaf=1,\n",
       "                      min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "                      n_estimators=50, n_jobs=None, oob_score=False,\n",
       "                      random_state=0, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfr = RandomForestRegressor(max_depth=14,n_estimators=50, random_state=0)\n",
    "rfr.fit(X_train, y_train) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred = rfr.predict(X_train)\n",
    "y_pred = rfr.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train : \n",
      "Mean Absolute Error: 1914.5229463008654\n",
      "Mean Squared Error: 13192842.094737627\n",
      "Root Mean Squared Error: 3632.1952170467966\n"
     ]
    }
   ],
   "source": [
    "print('Train : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_train, pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_train, pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_train, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test : \n",
      "Mean Absolute Error: 2140.0953291057695\n",
      "Mean Squared Error: 16860115.683780324\n",
      "Root Mean Squared Error: 4106.107120348947\n"
     ]
    }
   ],
   "source": [
    "print('Test : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared for Train: 0.9692647211806887\n",
      "R-squared for Test: 0.961382114574377\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import r2_score\n",
    "\n",
    "print(\"R-squared for Train:\",r2_score(y_train, pred))\n",
    "print(\"R-squared for Test:\",r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Gradient Boosting (GBM)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "from sklearn.model_selection import ShuffleSplit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingRegressor(alpha=0.9, ccp_alpha=0.0, criterion='friedman_mse',\n",
       "                          init=None, learning_rate=0.1, loss='ls', max_depth=3,\n",
       "                          max_features=None, max_leaf_nodes=None,\n",
       "                          min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                          min_samples_leaf=1, min_samples_split=2,\n",
       "                          min_weight_fraction_leaf=0.0, n_estimators=100,\n",
       "                          n_iter_no_change=None, presort='deprecated',\n",
       "                          random_state=None, subsample=1.0, tol=0.0001,\n",
       "                          validation_fraction=0.1, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gbrt=GradientBoostingRegressor()\n",
    "gbrt.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred=gbrt.predict(X_train)\n",
    "y_pred=gbrt.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1.09856812e-02 7.43814121e-04 7.51437878e-01 1.04415965e-06\n",
      " 0.00000000e+00 2.38718421e-03 1.99210102e-01 2.90507155e-02\n",
      " 4.23591299e-04 4.90473107e-03 8.26020113e-04 2.92381134e-05]\n"
     ]
    }
   ],
   "source": [
    "#One of the benefits of growing trees is that we can understand how important each of the features are\n",
    "print(gbrt.feature_importances_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train : \n",
      "Mean Absolute Error: 6551.714582016237\n",
      "Mean Squared Error: 100815080.85690081\n",
      "Root Mean Squared Error: 10040.671334970626\n"
     ]
    }
   ],
   "source": [
    "print('Train : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_train, pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_train, pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_train, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test : \n",
      "Mean Absolute Error: 6622.518007051183\n",
      "Mean Squared Error: 104317195.93169282\n",
      "Root Mean Squared Error: 10213.578997182762\n"
     ]
    }
   ],
   "source": [
    "print('Test : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared for Train: 0.7651317587918205\n",
      "R-squared for Test: 0.7610627592378938\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import r2_score\n",
    "\n",
    "print(\"R-squared for Train:\",r2_score(y_train, pred))\n",
    "print(\"R-squared for Test:\",r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tuning a Gradient Boosting Model (GBM)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "param_grid={'learning_rate' : [0.1,0.01],# 0.02,0.05,0.1],\n",
    "            'max_depth' : [4,6]}# 5,6]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Perform Grid-Search\n",
    "gbr = GridSearchCV(estimator=GradientBoostingRegressor(),param_grid=param_grid)\n",
    "    \n",
    "grid_result = gbr.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9114086574838197"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_result.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'learning_rate': 0.1, 'max_depth': 6}"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_result.best_params_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Building Final Gradient Boosting Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingRegressor(alpha=0.9, ccp_alpha=0.0, criterion='friedman_mse',\n",
       "                          init=None, learning_rate=0.01, loss='ls', max_depth=6,\n",
       "                          max_features=None, max_leaf_nodes=None,\n",
       "                          min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                          min_samples_leaf=1, min_samples_split=2,\n",
       "                          min_weight_fraction_leaf=0.0, n_estimators=100,\n",
       "                          n_iter_no_change=None, presort='deprecated',\n",
       "                          random_state=None, subsample=1.0, tol=0.0001,\n",
       "                          validation_fraction=0.1, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gbrt=GradientBoostingRegressor(learning_rate=0.01, max_depth=6)\n",
    "gbrt.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred=gbrt.predict(X_train)\n",
    "y_pred=gbrt.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train : \n",
      "Mean Absolute Error: 8757.334746550901\n",
      "Mean Squared Error: 156533706.61385474\n",
      "Root Mean Squared Error: 12511.34311790124\n"
     ]
    }
   ],
   "source": [
    "print('Train : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_train, pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_train, pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_train, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test : \n",
      "Mean Absolute Error: 8840.985051279671\n",
      "Mean Squared Error: 161696310.42731223\n",
      "Root Mean Squared Error: 12715.986411887685\n"
     ]
    }
   ],
   "source": [
    "print('Test : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared for Train: 0.635324437081214\n",
      "R-squared for Test: 0.6296366106292444\n"
     ]
    }
   ],
   "source": [
    "print(\"R-squared for Train:\",r2_score(y_train, pred))\n",
    "print(\"R-squared for Test:\",r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# XGBOOST"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "import xgboost as xgb\n",
    "import warnings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBRegressor(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "             colsample_bynode=1, colsample_bytree=1, gamma=0,\n",
       "             importance_type='gain', learning_rate=0.1, max_delta_step=0,\n",
       "             max_depth=3, min_child_weight=1, missing=None, n_estimators=100,\n",
       "             n_jobs=1, nthread=None, objective='reg:squarederror',\n",
       "             random_state=0, reg_alpha=0, reg_lambda=1, scale_pos_weight=1,\n",
       "             seed=None, silent=None, subsample=1, verbosity=1)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "warnings.filterwarnings(\"ignore\", category=FutureWarning)\n",
    "xg_reg = xgb.XGBRegressor(objective='reg:squarederror')\n",
    "xg_reg.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred=xg_reg.predict(X_train)\n",
    "y_pred=xg_reg.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train : \n",
      "Mean Absolute Error: 6594.8074345317855\n",
      "Mean Squared Error: 101694321.61318773\n",
      "Root Mean Squared Error: 10084.360248086525\n"
     ]
    }
   ],
   "source": [
    "print('Train : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_train, pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_train, pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_train, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test : \n",
      "Mean Absolute Error: 6660.103601740729\n",
      "Mean Squared Error: 105085339.31117935\n",
      "Root Mean Squared Error: 10251.114052198393\n"
     ]
    }
   ],
   "source": [
    "print('Test : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared for Train: 0.7630833972939929\n",
      "R-squared for Test: 0.7593033363741468\n"
     ]
    }
   ],
   "source": [
    "print(\"R-squared for Train:\",r2_score(y_train, pred))\n",
    "print(\"R-squared for Test:\",r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tuning a XGBoost Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb1 = xgb.XGBRegressor(objective='reg:squarederror')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "parameters = {'learning_rate': [0.01,0.1], #so called `eta` value\n",
    "              'max_depth': [4,5,6]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 6 candidates, totalling 30 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=5)]: Using backend LokyBackend with 5 concurrent workers.\n",
      "[Parallel(n_jobs=5)]: Done  30 out of  30 | elapsed: 20.9min finished\n"
     ]
    }
   ],
   "source": [
    "xgb_grid = GridSearchCV(xgb1,parameters,n_jobs = 5,verbose=True)\n",
    "\n",
    "grid_result = xgb_grid.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9108918449651803\n"
     ]
    }
   ],
   "source": [
    "print(xgb_grid.best_score_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'learning_rate': 0.1, 'max_depth': 6}\n"
     ]
    }
   ],
   "source": [
    "print(xgb_grid.best_params_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Building Final XGBoost Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "xg_reg = xgb.XGBRegressor(learning_rate=0.1, max_depth=9,objective='reg:squarederror')\n",
    "xg_reg.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred=xg_reg.predict(X_train)\n",
    "y_pred=xg_reg.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train : \n",
      "Mean Absolute Error: 2117.702650236579\n",
      "Mean Squared Error: 12360320.23925157\n",
      "Root Mean Squared Error: 3515.7247103906716\n"
     ]
    }
   ],
   "source": [
    "print('Train : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_train, pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_train, pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_train, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test : \n",
      "Mean Absolute Error: 2243.382353355045\n",
      "Mean Squared Error: 14381383.631431144\n",
      "Root Mean Squared Error: 3792.2794769677967\n"
     ]
    }
   ],
   "source": [
    "print('Test : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared for Train: 0.9712042419577729\n",
      "R-squared for Test: 0.9670596195330486\n"
     ]
    }
   ],
   "source": [
    "print(\"R-squared for Train:\",r2_score(y_train, pred))\n",
    "print(\"R-squared for Test:\",r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import neighbors\n",
    "from sklearn.metrics import mean_squared_error \n",
    "from math import sqrt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsRegressor(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "                    metric_params=None, n_jobs=None, n_neighbors=5, p=2,\n",
       "                    weights='uniform')"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn = neighbors.KNeighborsRegressor(n_neighbors=5)\n",
    "knn.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred=knn.predict(X_train)\n",
    "y_pred=knn.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train : \n",
      "Mean Absolute Error: 8648.869870672417\n",
      "Mean Squared Error: 176199297.09501198\n",
      "Root Mean Squared Error: 13274.008328120484\n"
     ]
    }
   ],
   "source": [
    "print('Train : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_train, pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_train, pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_train, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test : \n",
      "Mean Absolute Error: 10565.921852348762\n",
      "Mean Squared Error: 262544677.29306823\n",
      "Root Mean Squared Error: 16203.230458555732\n"
     ]
    }
   ],
   "source": [
    "print('Test : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared for Train: 0.5895096382498192\n",
      "R-squared for Test: 0.3986446797298866\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import r2_score\n",
    "\n",
    "print(\"R-squared for Train:\",r2_score(y_train, pred))\n",
    "print(\"R-squared for Test:\",r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tuning a KNN Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE value for k=  1 is: 22495.640140571264\n",
      "RMSE value for k=  2 is: 18717.297663330246\n",
      "RMSE value for k=  3 is: 17376.9892066167\n",
      "RMSE value for k=  4 is: 16682.66991467301\n",
      "RMSE value for k=  5 is: 16203.230458555732\n",
      "RMSE value for k=  6 is: 15929.847023300377\n",
      "RMSE value for k=  7 is: 15732.660540643019\n",
      "RMSE value for k=  8 is: 15625.594942327576\n",
      "RMSE value for k=  9 is: 15544.1131094506\n",
      "RMSE value for k=  10 is: 15470.957801628083\n",
      "RMSE value for k=  11 is: 15436.68262909747\n",
      "RMSE value for k=  12 is: 15412.068815983162\n",
      "RMSE value for k=  13 is: 15390.570608379076\n",
      "RMSE value for k=  14 is: 15381.792247639041\n",
      "RMSE value for k=  15 is: 15368.375764961836\n",
      "RMSE value for k=  16 is: 15360.071382395065\n",
      "RMSE value for k=  17 is: 15354.474171983227\n",
      "RMSE value for k=  18 is: 15359.969667928903\n",
      "RMSE value for k=  19 is: 15361.536267631978\n",
      "RMSE value for k=  20 is: 15364.461759931062\n"
     ]
    }
   ],
   "source": [
    "rmse_val = [] #to store rmse values for different k\n",
    "for K in range(20):\n",
    "    K = K+1\n",
    "    model = neighbors.KNeighborsRegressor(n_neighbors = K)\n",
    "\n",
    "    model.fit(X_train, y_train)  #fit the model\n",
    "    y_pred=model.predict(X_test) #make prediction on test set\n",
    "    error = sqrt(mean_squared_error(y_test,y_pred)) #calculate rmse\n",
    "    rmse_val.append(error) #store rmse values\n",
    "    print('RMSE value for k= ' , K , 'is:', error)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x21880a38608>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting the rmse values against k values\n",
    "curve = pd.DataFrame(rmse_val) #elbow curve \n",
    "curve.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Building Final KNN Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsRegressor(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "                    metric_params=None, n_jobs=None, n_neighbors=11, p=2,\n",
       "                    weights='uniform')"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn = neighbors.KNeighborsRegressor(n_neighbors=11)\n",
    "knn.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred=knn.predict(X_train)\n",
    "y_pred=knn.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train : \n",
      "Mean Absolute Error: 9193.475024224781\n",
      "Mean Squared Error: 195534278.7311073\n",
      "Root Mean Squared Error: 13983.357205303286\n"
     ]
    }
   ],
   "source": [
    "print('Train : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_train, pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_train, pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_train, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test : \n",
      "Mean Absolute Error: 10133.786803743722\n",
      "Mean Squared Error: 238291170.59147954\n",
      "Root Mean Squared Error: 15436.68262909747\n"
     ]
    }
   ],
   "source": [
    "print('Test : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared for Train: 0.544465056704445\n",
      "R-squared for Test: 0.4541970353921062\n"
     ]
    }
   ],
   "source": [
    "print(\"R-squared for Train:\",r2_score(y_train, pred))\n",
    "print(\"R-squared for Test:\",r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Neural Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.wrappers.scikit_learn import KerasRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_regressor():\n",
    "    regressor = Sequential()\n",
    "    regressor.add(Dense(units=1, input_dim=12))\n",
    "    regressor.add(Dense(units=1))\n",
    "    regressor.compile(optimizer='adam', loss='mean_squared_error',  metrics=['mae'])\n",
    "    return regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "regressor = KerasRegressor(build_fn=build_regressor, batch_size=256,epochs=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "295099/295099 [==============================] - 10s 34us/step - loss: 404500523.0518 - mae: 13865.5596 - accuracy: 0.0000e+00\n",
      "Epoch 2/10\n",
      "295099/295099 [==============================] - 10s 34us/step - loss: 403970222.4783 - mae: 13877.8975 - accuracy: 0.0000e+00\n",
      "Epoch 3/10\n",
      "295099/295099 [==============================] - 11s 38us/step - loss: 403487369.0797 - mae: 13893.9863 - accuracy: 0.0000e+00\n",
      "Epoch 4/10\n",
      "295099/295099 [==============================] - 10s 36us/step - loss: 403074496.1023 - mae: 13906.4248 - accuracy: 0.0000e+00\n",
      "Epoch 5/10\n",
      "295099/295099 [==============================] - 11s 37us/step - loss: 402665645.2541 - mae: 13930.8408 - accuracy: 0.0000e+00\n",
      "Epoch 6/10\n",
      "295099/295099 [==============================] - 11s 36us/step - loss: 402359236.6439 - mae: 13941.4443 - accuracy: 0.0000e+00\n",
      "Epoch 7/10\n",
      "295099/295099 [==============================] - 10s 34us/step - loss: 402046216.2664 - mae: 13956.8320 - accuracy: 0.0000e+00\n",
      "Epoch 8/10\n",
      "295099/295099 [==============================] - 10s 35us/step - loss: 401821130.0352 - mae: 13984.2754 - accuracy: 0.0000e+00\n",
      "Epoch 9/10\n",
      "295099/295099 [==============================] - 11s 38us/step - loss: 401578487.7839 - mae: 13994.7520 - accuracy: 0.0000e+00\n",
      "Epoch 10/10\n",
      "295099/295099 [==============================] - 11s 38us/step - loss: 401397019.9419 - mae: 14009.4834 - accuracy: 0.0000e+00\n"
     ]
    }
   ],
   "source": [
    "results=regressor.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred= regressor.predict(X_train)\n",
    "y_pred= regressor.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train : \n",
      "Mean Absolute Error: 14088.42659670954\n",
      "Mean Squared Error: 401303510.91938746\n",
      "Root Mean Squared Error: 20032.561267081837\n"
     ]
    }
   ],
   "source": [
    "print('Train : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_train, pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_train, pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_train, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test : \n",
      "Mean Absolute Error: 14127.499989383492\n",
      "Mean Squared Error: 407555117.61347127\n",
      "Root Mean Squared Error: 20187.99439304141\n"
     ]
    }
   ],
   "source": [
    "print('Test : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared for Train: 0.06508580859951463\n",
      "R-squared for Test: 0.06650006845656364\n"
     ]
    }
   ],
   "source": [
    "print(\"R-squared for Train:\",r2_score(y_train, pred))\n",
    "print(\"R-squared for Test:\",r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Voting Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.ensemble import VotingRegressor\n",
    "import warnings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "warnings.filterwarnings(\"ignore\", category=FutureWarning)  # Ignore FutureWarning\n",
    "reg1 = GradientBoostingRegressor(learning_rate=0.01, max_depth=6)\n",
    "reg2 = RandomForestRegressor(max_depth=14,n_estimators=50, random_state=0)\n",
    "reg3 = xgb.XGBRegressor(learning_rate=0.1, max_depth=4,objective='reg:squarederror')\n",
    "ereg = VotingRegressor(estimators=[('gb', reg1), ('rf', reg2), ('xg', reg3)])\n",
    "ereg = ereg.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred= ereg.predict(X_train)\n",
    "y_pred= ereg.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train : \n",
      "Mean Absolute Error: 4892.469171544005\n",
      "Mean Squared Error: 53956103.154803604\n",
      "Root Mean Squared Error: 7345.481819105103\n"
     ]
    }
   ],
   "source": [
    "print('Train : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_train, pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_train, pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_train, pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test : \n",
      "Mean Absolute Error: 4986.839395443702\n",
      "Mean Squared Error: 57516767.77942866\n",
      "Root Mean Squared Error: 7583.9809980925365\n"
     ]
    }
   ],
   "source": [
    "print('Test : ')\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))  \n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))  \n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "R-squared for Train: 0.8742988157852736\n",
      "R-squared for Test: 0.8682585582556266\n"
     ]
    }
   ],
   "source": [
    "print(\"R-squared for Train:\",r2_score(y_train, pred))\n",
    "print(\"R-squared for Test:\",r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Save Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "# Saving model to disk of random forest\n",
    "pickle.dump(rfr, open('model.pkl','wb'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load Model and Predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([24195.97538878])"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pickle\n",
    "model=pickle.load(open('model.pkl', 'rb'))\n",
    "model.predict([[211.0963582,5,1,2.572,0,2,151315,1,42.31,0,8.106,2010]])"
   ]
  },
  {
   "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}