Flight_Price_Predictor.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 341,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "sns.set()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "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>Airline</th>\n",
       "      <th>Date_of_Journey</th>\n",
       "      <th>Source</th>\n",
       "      <th>Destination</th>\n",
       "      <th>Route</th>\n",
       "      <th>Dep_Time</th>\n",
       "      <th>Arrival_Time</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Total_Stops</th>\n",
       "      <th>Additional_Info</th>\n",
       "      <th>Price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>24/03/2019</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>New Delhi</td>\n",
       "      <td>BLR → DEL</td>\n",
       "      <td>22:20</td>\n",
       "      <td>01:10 22 Mar</td>\n",
       "      <td>2h 50m</td>\n",
       "      <td>non-stop</td>\n",
       "      <td>No info</td>\n",
       "      <td>3897</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Air India</td>\n",
       "      <td>1/05/2019</td>\n",
       "      <td>Kolkata</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>CCU → IXR → BBI → BLR</td>\n",
       "      <td>05:50</td>\n",
       "      <td>13:15</td>\n",
       "      <td>7h 25m</td>\n",
       "      <td>2 stops</td>\n",
       "      <td>No info</td>\n",
       "      <td>7662</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Jet Airways</td>\n",
       "      <td>9/06/2019</td>\n",
       "      <td>Delhi</td>\n",
       "      <td>Cochin</td>\n",
       "      <td>DEL → LKO → BOM → COK</td>\n",
       "      <td>09:25</td>\n",
       "      <td>04:25 10 Jun</td>\n",
       "      <td>19h</td>\n",
       "      <td>2 stops</td>\n",
       "      <td>No info</td>\n",
       "      <td>13882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>12/05/2019</td>\n",
       "      <td>Kolkata</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>CCU → NAG → BLR</td>\n",
       "      <td>18:05</td>\n",
       "      <td>23:30</td>\n",
       "      <td>5h 25m</td>\n",
       "      <td>1 stop</td>\n",
       "      <td>No info</td>\n",
       "      <td>6218</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>01/03/2019</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>New Delhi</td>\n",
       "      <td>BLR → NAG → DEL</td>\n",
       "      <td>16:50</td>\n",
       "      <td>21:35</td>\n",
       "      <td>4h 45m</td>\n",
       "      <td>1 stop</td>\n",
       "      <td>No info</td>\n",
       "      <td>13302</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Airline Date_of_Journey    Source Destination                  Route  \\\n",
       "0       IndiGo      24/03/2019  Banglore   New Delhi              BLR → DEL   \n",
       "1    Air India       1/05/2019   Kolkata    Banglore  CCU → IXR → BBI → BLR   \n",
       "2  Jet Airways       9/06/2019     Delhi      Cochin  DEL → LKO → BOM → COK   \n",
       "3       IndiGo      12/05/2019   Kolkata    Banglore        CCU → NAG → BLR   \n",
       "4       IndiGo      01/03/2019  Banglore   New Delhi        BLR → NAG → DEL   \n",
       "\n",
       "  Dep_Time  Arrival_Time Duration Total_Stops Additional_Info  Price  \n",
       "0    22:20  01:10 22 Mar   2h 50m    non-stop         No info   3897  \n",
       "1    05:50         13:15   7h 25m     2 stops         No info   7662  \n",
       "2    09:25  04:25 10 Jun      19h     2 stops         No info  13882  \n",
       "3    18:05         23:30   5h 25m      1 stop         No info   6218  \n",
       "4    16:50         21:35   4h 45m      1 stop         No info  13302  "
      ]
     },
     "execution_count": 229,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data=pd.read_excel('Data_Train.xlsx')\n",
    "train_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 230,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 10683 entries, 0 to 10682\n",
      "Data columns (total 11 columns):\n",
      " #   Column           Non-Null Count  Dtype \n",
      "---  ------           --------------  ----- \n",
      " 0   Airline          10683 non-null  object\n",
      " 1   Date_of_Journey  10683 non-null  object\n",
      " 2   Source           10683 non-null  object\n",
      " 3   Destination      10683 non-null  object\n",
      " 4   Route            10682 non-null  object\n",
      " 5   Dep_Time         10683 non-null  object\n",
      " 6   Arrival_Time     10683 non-null  object\n",
      " 7   Duration         10683 non-null  object\n",
      " 8   Total_Stops      10682 non-null  object\n",
      " 9   Additional_Info  10683 non-null  object\n",
      " 10  Price            10683 non-null  int64 \n",
      "dtypes: int64(1), object(10)\n",
      "memory usage: 918.2+ KB\n"
     ]
    }
   ],
   "source": [
    "train_data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "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>Price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>10683.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>9087.064121</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4611.359167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1759.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5277.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>8372.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>12373.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>79512.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Price\n",
       "count  10683.000000\n",
       "mean    9087.064121\n",
       "std     4611.359167\n",
       "min     1759.000000\n",
       "25%     5277.000000\n",
       "50%     8372.000000\n",
       "75%    12373.000000\n",
       "max    79512.000000"
      ]
     },
     "execution_count": 231,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2h 50m     550\n",
       "1h 30m     386\n",
       "2h 55m     337\n",
       "2h 45m     337\n",
       "2h 35m     329\n",
       "          ... \n",
       "27h 30m      1\n",
       "42h 45m      1\n",
       "4h 10m       1\n",
       "33h 45m      1\n",
       "36h 25m      1\n",
       "Name: Duration, Length: 368, dtype: int64"
      ]
     },
     "execution_count": 232,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data['Duration'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 10682 entries, 0 to 10682\n",
      "Data columns (total 11 columns):\n",
      " #   Column           Non-Null Count  Dtype \n",
      "---  ------           --------------  ----- \n",
      " 0   Airline          10682 non-null  object\n",
      " 1   Date_of_Journey  10682 non-null  object\n",
      " 2   Source           10682 non-null  object\n",
      " 3   Destination      10682 non-null  object\n",
      " 4   Route            10682 non-null  object\n",
      " 5   Dep_Time         10682 non-null  object\n",
      " 6   Arrival_Time     10682 non-null  object\n",
      " 7   Duration         10682 non-null  object\n",
      " 8   Total_Stops      10682 non-null  object\n",
      " 9   Additional_Info  10682 non-null  object\n",
      " 10  Price            10682 non-null  int64 \n",
      "dtypes: int64(1), object(10)\n",
      "memory usage: 1001.4+ KB\n"
     ]
    }
   ],
   "source": [
    "train_data.dropna(inplace = True)\n",
    "train_data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 234,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data['Journey_Day']=pd.to_datetime(train_data.Date_of_Journey,format='%d/%m/%Y').dt.day\n",
    "train_data['Journey_Month']=pd.to_datetime(train_data.Date_of_Journey,format='%d/%m/%Y').dt.month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "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>Airline</th>\n",
       "      <th>Date_of_Journey</th>\n",
       "      <th>Source</th>\n",
       "      <th>Destination</th>\n",
       "      <th>Route</th>\n",
       "      <th>Dep_Time</th>\n",
       "      <th>Arrival_Time</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Total_Stops</th>\n",
       "      <th>Additional_Info</th>\n",
       "      <th>Price</th>\n",
       "      <th>Journey_Day</th>\n",
       "      <th>Journey_Month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>24/03/2019</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>New Delhi</td>\n",
       "      <td>BLR → DEL</td>\n",
       "      <td>22:20</td>\n",
       "      <td>01:10 22 Mar</td>\n",
       "      <td>2h 50m</td>\n",
       "      <td>non-stop</td>\n",
       "      <td>No info</td>\n",
       "      <td>3897</td>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Air India</td>\n",
       "      <td>1/05/2019</td>\n",
       "      <td>Kolkata</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>CCU → IXR → BBI → BLR</td>\n",
       "      <td>05:50</td>\n",
       "      <td>13:15</td>\n",
       "      <td>7h 25m</td>\n",
       "      <td>2 stops</td>\n",
       "      <td>No info</td>\n",
       "      <td>7662</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Jet Airways</td>\n",
       "      <td>9/06/2019</td>\n",
       "      <td>Delhi</td>\n",
       "      <td>Cochin</td>\n",
       "      <td>DEL → LKO → BOM → COK</td>\n",
       "      <td>09:25</td>\n",
       "      <td>04:25 10 Jun</td>\n",
       "      <td>19h</td>\n",
       "      <td>2 stops</td>\n",
       "      <td>No info</td>\n",
       "      <td>13882</td>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>12/05/2019</td>\n",
       "      <td>Kolkata</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>CCU → NAG → BLR</td>\n",
       "      <td>18:05</td>\n",
       "      <td>23:30</td>\n",
       "      <td>5h 25m</td>\n",
       "      <td>1 stop</td>\n",
       "      <td>No info</td>\n",
       "      <td>6218</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>01/03/2019</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>New Delhi</td>\n",
       "      <td>BLR → NAG → DEL</td>\n",
       "      <td>16:50</td>\n",
       "      <td>21:35</td>\n",
       "      <td>4h 45m</td>\n",
       "      <td>1 stop</td>\n",
       "      <td>No info</td>\n",
       "      <td>13302</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Airline Date_of_Journey    Source Destination                  Route  \\\n",
       "0       IndiGo      24/03/2019  Banglore   New Delhi              BLR → DEL   \n",
       "1    Air India       1/05/2019   Kolkata    Banglore  CCU → IXR → BBI → BLR   \n",
       "2  Jet Airways       9/06/2019     Delhi      Cochin  DEL → LKO → BOM → COK   \n",
       "3       IndiGo      12/05/2019   Kolkata    Banglore        CCU → NAG → BLR   \n",
       "4       IndiGo      01/03/2019  Banglore   New Delhi        BLR → NAG → DEL   \n",
       "\n",
       "  Dep_Time  Arrival_Time Duration Total_Stops Additional_Info  Price  \\\n",
       "0    22:20  01:10 22 Mar   2h 50m    non-stop         No info   3897   \n",
       "1    05:50         13:15   7h 25m     2 stops         No info   7662   \n",
       "2    09:25  04:25 10 Jun      19h     2 stops         No info  13882   \n",
       "3    18:05         23:30   5h 25m      1 stop         No info   6218   \n",
       "4    16:50         21:35   4h 45m      1 stop         No info  13302   \n",
       "\n",
       "   Journey_Day  Journey_Month  \n",
       "0           24              3  \n",
       "1            1              5  \n",
       "2            9              6  \n",
       "3           12              5  \n",
       "4            1              3  "
      ]
     },
     "execution_count": 235,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2019    10682\n",
       "Name: Date_of_Journey, dtype: int64"
      ]
     },
     "execution_count": 236,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.to_datetime(train_data.Date_of_Journey,format='%d/%m/%Y').dt.year.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data.drop([\"Date_of_Journey\"], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Extracting Hours\n",
    "train_data[\"Dep_hour\"] = pd.to_datetime(train_data[\"Dep_Time\"]).dt.hour\n",
    "\n",
    "# Extracting Minutes\n",
    "train_data[\"Dep_min\"] = pd.to_datetime(train_data[\"Dep_Time\"]).dt.minute"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 239,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data.drop([\"Dep_Time\"], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 240,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data[\"Arrival_hour\"] = pd.to_datetime(train_data.Arrival_Time).dt.hour\n",
    "\n",
    "train_data[\"Arrival_min\"] = pd.to_datetime(train_data.Arrival_Time).dt.minute\n",
    "\n",
    "train_data.drop([\"Arrival_Time\"], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "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>Airline</th>\n",
       "      <th>Source</th>\n",
       "      <th>Destination</th>\n",
       "      <th>Route</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Total_Stops</th>\n",
       "      <th>Additional_Info</th>\n",
       "      <th>Price</th>\n",
       "      <th>Journey_Day</th>\n",
       "      <th>Journey_Month</th>\n",
       "      <th>Dep_hour</th>\n",
       "      <th>Dep_min</th>\n",
       "      <th>Arrival_hour</th>\n",
       "      <th>Arrival_min</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>New Delhi</td>\n",
       "      <td>BLR → DEL</td>\n",
       "      <td>2h 50m</td>\n",
       "      <td>non-stop</td>\n",
       "      <td>No info</td>\n",
       "      <td>3897</td>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>22</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Air India</td>\n",
       "      <td>Kolkata</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>CCU → IXR → BBI → BLR</td>\n",
       "      <td>7h 25m</td>\n",
       "      <td>2 stops</td>\n",
       "      <td>No info</td>\n",
       "      <td>7662</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>50</td>\n",
       "      <td>13</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Jet Airways</td>\n",
       "      <td>Delhi</td>\n",
       "      <td>Cochin</td>\n",
       "      <td>DEL → LKO → BOM → COK</td>\n",
       "      <td>19h</td>\n",
       "      <td>2 stops</td>\n",
       "      <td>No info</td>\n",
       "      <td>13882</td>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "      <td>9</td>\n",
       "      <td>25</td>\n",
       "      <td>4</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>Kolkata</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>CCU → NAG → BLR</td>\n",
       "      <td>5h 25m</td>\n",
       "      <td>1 stop</td>\n",
       "      <td>No info</td>\n",
       "      <td>6218</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "      <td>18</td>\n",
       "      <td>5</td>\n",
       "      <td>23</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>New Delhi</td>\n",
       "      <td>BLR → NAG → DEL</td>\n",
       "      <td>4h 45m</td>\n",
       "      <td>1 stop</td>\n",
       "      <td>No info</td>\n",
       "      <td>13302</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>16</td>\n",
       "      <td>50</td>\n",
       "      <td>21</td>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Airline    Source Destination                  Route Duration  \\\n",
       "0       IndiGo  Banglore   New Delhi              BLR → DEL   2h 50m   \n",
       "1    Air India   Kolkata    Banglore  CCU → IXR → BBI → BLR   7h 25m   \n",
       "2  Jet Airways     Delhi      Cochin  DEL → LKO → BOM → COK      19h   \n",
       "3       IndiGo   Kolkata    Banglore        CCU → NAG → BLR   5h 25m   \n",
       "4       IndiGo  Banglore   New Delhi        BLR → NAG → DEL   4h 45m   \n",
       "\n",
       "  Total_Stops Additional_Info  Price  Journey_Day  Journey_Month  Dep_hour  \\\n",
       "0    non-stop         No info   3897           24              3        22   \n",
       "1     2 stops         No info   7662            1              5         5   \n",
       "2     2 stops         No info  13882            9              6         9   \n",
       "3      1 stop         No info   6218           12              5        18   \n",
       "4      1 stop         No info  13302            1              3        16   \n",
       "\n",
       "   Dep_min  Arrival_hour  Arrival_min  \n",
       "0       20             1           10  \n",
       "1       50            13           15  \n",
       "2       25             4           25  \n",
       "3        5            23           30  \n",
       "4       50            21           35  "
      ]
     },
     "execution_count": 241,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 242,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['2h 50m',\n",
       " '7h 25m',\n",
       " '19h',\n",
       " '5h 25m',\n",
       " '4h 45m',\n",
       " '2h 25m',\n",
       " '15h 30m',\n",
       " '21h 5m',\n",
       " '25h 30m',\n",
       " '7h 50m',\n",
       " '13h 15m',\n",
       " '2h 35m',\n",
       " '2h 15m',\n",
       " '12h 10m',\n",
       " '2h 35m',\n",
       " '26h 35m',\n",
       " '4h 30m',\n",
       " '22h 35m',\n",
       " '23h',\n",
       " '20h 35m',\n",
       " '5h 10m',\n",
       " '15h 20m',\n",
       " '2h 50m',\n",
       " '2h 55m',\n",
       " '13h 20m',\n",
       " '15h 10m',\n",
       " '5h 45m',\n",
       " '5h 55m',\n",
       " '2h 50m',\n",
       " '2h 15m',\n",
       " '2h 15m',\n",
       " '13h 25m',\n",
       " '2h 50m',\n",
       " '22h',\n",
       " '5h 30m',\n",
       " '10h 25m',\n",
       " '5h 15m',\n",
       " '2h 30m',\n",
       " '6h 15m',\n",
       " '11h 55m',\n",
       " '11h 5m',\n",
       " '8h 30m',\n",
       " '22h 5m',\n",
       " '2h 45m',\n",
       " '12h',\n",
       " '2h 50m',\n",
       " '2h 50m',\n",
       " '2h 15m',\n",
       " '16h 5m',\n",
       " '19h 55m',\n",
       " '3h 15m',\n",
       " '25h 20m',\n",
       " '2h 50m',\n",
       " '3h',\n",
       " '2h 50m',\n",
       " '16h 15m',\n",
       " '15h 5m',\n",
       " '2h 15m',\n",
       " '6h 30m',\n",
       " '25h 5m',\n",
       " '12h 25m',\n",
       " '27h 20m',\n",
       " '10h 15m',\n",
       " '10h 30m',\n",
       " '2h 15m',\n",
       " '10h 25m',\n",
       " '2h 50m',\n",
       " '1h 30m',\n",
       " '13h 20m',\n",
       " '2h 15m',\n",
       " '1h 25m',\n",
       " '26h 30m',\n",
       " '7h 20m',\n",
       " '13h 30m',\n",
       " '5h',\n",
       " '2h 45m',\n",
       " '2h 50m',\n",
       " '1h 30m',\n",
       " '19h 5m',\n",
       " '2h 15m',\n",
       " '14h 50m',\n",
       " '2h 40m',\n",
       " '22h 10m',\n",
       " '9h 35m',\n",
       " '10h',\n",
       " '21h 20m',\n",
       " '5h 25m',\n",
       " '18h 45m',\n",
       " '12h 20m',\n",
       " '18h',\n",
       " '9h 15m',\n",
       " '11h 5m',\n",
       " '17h 30m',\n",
       " '13h 20m',\n",
       " '2h 35m',\n",
       " '2h 25m',\n",
       " '2h 40m',\n",
       " '3h',\n",
       " '1h 25m',\n",
       " '16h 35m',\n",
       " '12h 20m',\n",
       " '12h 15m',\n",
       " '7h 30m',\n",
       " '24h',\n",
       " '2h 45m',\n",
       " '8h 55m',\n",
       " '7h 10m',\n",
       " '14h 30m',\n",
       " '30h 20m',\n",
       " '19h 55m',\n",
       " '15h',\n",
       " '2h 55m',\n",
       " '12h 45m',\n",
       " '10h 10m',\n",
       " '5h 55m',\n",
       " '15h 25m',\n",
       " '16h 5m',\n",
       " '10h 10m',\n",
       " '2h 40m',\n",
       " '11h 55m',\n",
       " '2h 40m',\n",
       " '1h 25m',\n",
       " '14h 5m',\n",
       " '20h 15m',\n",
       " '2h 25m',\n",
       " '23h 10m',\n",
       " '18h 10m',\n",
       " '16h',\n",
       " '2h 50m',\n",
       " '2h 20m',\n",
       " '8h',\n",
       " '16h 55m',\n",
       " '3h 10m',\n",
       " '2h 45m',\n",
       " '14h',\n",
       " '23h 50m',\n",
       " '21h 40m',\n",
       " '21h 15m',\n",
       " '2h 45m',\n",
       " '10h 50m',\n",
       " '8h',\n",
       " '8h 15m',\n",
       " '5h 25m',\n",
       " '8h 35m',\n",
       " '11h 50m',\n",
       " '27h 35m',\n",
       " '8h 25m',\n",
       " '20h 55m',\n",
       " '4h 50m',\n",
       " '8h 10m',\n",
       " '2h 20m',\n",
       " '24h 25m',\n",
       " '2h 50m',\n",
       " '12h 20m',\n",
       " '12h 15m',\n",
       " '23h 35m',\n",
       " '16h 35m',\n",
       " '2h 50m',\n",
       " '25h 45m',\n",
       " '26h 10m',\n",
       " '28h 50m',\n",
       " '2h 45m',\n",
       " '25h 15m',\n",
       " '9h 20m',\n",
       " '4h 30m',\n",
       " '15h 25m',\n",
       " '1h 30m',\n",
       " '2h 40m',\n",
       " '9h 10m',\n",
       " '2h 45m',\n",
       " '22h 35m',\n",
       " '3h 5m',\n",
       " '11h 30m',\n",
       " '9h 30m',\n",
       " '12h 10m',\n",
       " '17h 35m',\n",
       " '5h 5m',\n",
       " '25h 50m',\n",
       " '20h',\n",
       " '13h',\n",
       " '18h 25m',\n",
       " '1h 30m',\n",
       " '9h 30m',\n",
       " '16h',\n",
       " '2h 20m',\n",
       " '4h 30m',\n",
       " '5h 45m',\n",
       " '24h 10m',\n",
       " '2h 35m',\n",
       " '2h 45m',\n",
       " '4h 55m',\n",
       " '25h 35m',\n",
       " '6h 20m',\n",
       " '3h 5m',\n",
       " '18h 40m',\n",
       " '2h 15m',\n",
       " '12h 10m',\n",
       " '19h 25m',\n",
       " '2h 50m',\n",
       " '8h 25m',\n",
       " '9h 15m',\n",
       " '2h 50m',\n",
       " '3h 15m',\n",
       " '2h 50m',\n",
       " '12h 20m',\n",
       " '2h 55m',\n",
       " '9h 35m',\n",
       " '2h 35m',\n",
       " '9h 30m',\n",
       " '29h 20m',\n",
       " '2h 55m',\n",
       " '5h 15m',\n",
       " '9h 5m',\n",
       " '10h 45m',\n",
       " '4h 55m',\n",
       " '1h 30m',\n",
       " '11h 40m',\n",
       " '2h 45m',\n",
       " '2h 55m',\n",
       " '10h',\n",
       " '2h 45m',\n",
       " '10h 15m',\n",
       " '1h 25m',\n",
       " '9h 15m',\n",
       " '22h 55m',\n",
       " '28h 50m',\n",
       " '37h 25m',\n",
       " '2h 50m',\n",
       " '1h 30m',\n",
       " '1h 30m',\n",
       " '25h 40m',\n",
       " '2h 35m',\n",
       " '6h 15m',\n",
       " '17h 30m',\n",
       " '5h 45m',\n",
       " '2h 35m',\n",
       " '25h 30m',\n",
       " '13h 55m',\n",
       " '2h 55m',\n",
       " '10h 15m',\n",
       " '8h 40m',\n",
       " '5h 30m',\n",
       " '12h',\n",
       " '2h 15m',\n",
       " '2h 15m',\n",
       " '2h 50m',\n",
       " '9h 10m',\n",
       " '10h 25m',\n",
       " '2h 20m',\n",
       " '4h 30m',\n",
       " '3h 10m',\n",
       " '23h 30m',\n",
       " '27h 20m',\n",
       " '12h 35m',\n",
       " '24h 15m',\n",
       " '1h 20m',\n",
       " '2h 45m',\n",
       " '3h',\n",
       " '11h',\n",
       " '13h 30m',\n",
       " '2h 50m',\n",
       " '2h 55m',\n",
       " '2h 45m',\n",
       " '11h 15m',\n",
       " '14h 35m',\n",
       " '13h 55m',\n",
       " '14h 5m',\n",
       " '15h 5m',\n",
       " '2h 15m',\n",
       " '15h',\n",
       " '12h 55m',\n",
       " '9h',\n",
       " '12h 15m',\n",
       " '7h 40m',\n",
       " '27h 20m',\n",
       " '1h 25m',\n",
       " '1h 20m',\n",
       " '5h 30m',\n",
       " '11h 45m',\n",
       " '3h 15m',\n",
       " '24h 55m',\n",
       " '27h 20m',\n",
       " '2h 55m',\n",
       " '17h 5m',\n",
       " '2h 40m',\n",
       " '8h',\n",
       " '3h',\n",
       " '27h 35m',\n",
       " '7h 30m',\n",
       " '22h 55m',\n",
       " '8h 35m',\n",
       " '29h 55m',\n",
       " '22h 15m',\n",
       " '1h 20m',\n",
       " '2h 40m',\n",
       " '3h',\n",
       " '12h 10m',\n",
       " '2h 35m',\n",
       " '11h 45m',\n",
       " '2h 35m',\n",
       " '1h 30m',\n",
       " '2h 40m',\n",
       " '2h 35m',\n",
       " '22h 55m',\n",
       " '14h 40m',\n",
       " '15h 25m',\n",
       " '7h 15m',\n",
       " '20h 10m',\n",
       " '20h 45m',\n",
       " '2h 30m',\n",
       " '23h 50m',\n",
       " '27h',\n",
       " '5h 30m',\n",
       " '11h 50m',\n",
       " '2h 15m',\n",
       " '24h 30m',\n",
       " '2h 20m',\n",
       " '6h 30m',\n",
       " '20h 25m',\n",
       " '2h 35m',\n",
       " '2h 55m',\n",
       " '5h 35m',\n",
       " '2h 30m',\n",
       " '14h 45m',\n",
       " '12h 20m',\n",
       " '25h 30m',\n",
       " '5h 35m',\n",
       " '11h 45m',\n",
       " '2h 55m',\n",
       " '5h 40m',\n",
       " '4h 5m',\n",
       " '8h',\n",
       " '2h 45m',\n",
       " '2h 40m',\n",
       " '2h 55m',\n",
       " '15h 55m',\n",
       " '23h 10m',\n",
       " '7h 20m',\n",
       " '2h 50m',\n",
       " '7h 30m',\n",
       " '3h',\n",
       " '7h 45m',\n",
       " '25h 20m',\n",
       " '10h 25m',\n",
       " '30h 20m',\n",
       " '2h 35m',\n",
       " '9h',\n",
       " '15h 10m',\n",
       " '28h 20m',\n",
       " '3h',\n",
       " '2h 45m',\n",
       " '4h 55m',\n",
       " '2h 40m',\n",
       " '22h 10m',\n",
       " '10h 10m',\n",
       " '2h 50m',\n",
       " '4h 20m',\n",
       " '11h',\n",
       " '3h',\n",
       " '22h 35m',\n",
       " '3h 5m',\n",
       " '2h 40m',\n",
       " '2h 50m',\n",
       " '21h 5m',\n",
       " '11h 50m',\n",
       " '25h 35m',\n",
       " '15h 25m',\n",
       " '2h 25m',\n",
       " '3h 40m',\n",
       " '2h 55m',\n",
       " '3h 10m',\n",
       " '11h 50m',\n",
       " '2h 50m',\n",
       " '2h 50m',\n",
       " '1h 30m',\n",
       " '11h 30m',\n",
       " '8h 50m',\n",
       " '20h 15m',\n",
       " '20h 10m',\n",
       " '16h 5m',\n",
       " '12h 35m',\n",
       " '17h 5m',\n",
       " '3h',\n",
       " '15h 30m',\n",
       " '11h 30m',\n",
       " '2h 20m',\n",
       " '23h 45m',\n",
       " '1h 30m',\n",
       " '2h 35m',\n",
       " '24h 45m',\n",
       " '21h 35m',\n",
       " '5h 30m',\n",
       " '12h 15m',\n",
       " '8h 5m',\n",
       " '5h',\n",
       " '2h 50m',\n",
       " '6h 25m',\n",
       " '2h 20m',\n",
       " '2h 45m',\n",
       " '12h 45m',\n",
       " '15h 50m',\n",
       " '27h',\n",
       " '26h 25m',\n",
       " '2h 55m',\n",
       " '24h 50m',\n",
       " '8h 40m',\n",
       " '25h 45m',\n",
       " '2h 50m',\n",
       " '6h 15m',\n",
       " '10h 30m',\n",
       " '26h',\n",
       " '17h 5m',\n",
       " '21h 20m',\n",
       " '5h 35m',\n",
       " '2h 50m',\n",
       " '25h 50m',\n",
       " '12h 10m',\n",
       " '17h 5m',\n",
       " '8h 30m',\n",
       " '23h 5m',\n",
       " '9h 20m',\n",
       " '7h 55m',\n",
       " '1h 30m',\n",
       " '2h 50m',\n",
       " '2h 15m',\n",
       " '10h',\n",
       " '2h 35m',\n",
       " '7h 45m',\n",
       " '2h 55m',\n",
       " '26h 20m',\n",
       " '2h 40m',\n",
       " '11h',\n",
       " '19h 25m',\n",
       " '2h 55m',\n",
       " '2h 45m',\n",
       " '23h 15m',\n",
       " '9h',\n",
       " '3h 10m',\n",
       " '6h 25m',\n",
       " '8h 40m',\n",
       " '2h 50m',\n",
       " '12h 20m',\n",
       " '5h 20m',\n",
       " '4h',\n",
       " '9h 45m',\n",
       " '2h 55m',\n",
       " '26h 20m',\n",
       " '2h 15m',\n",
       " '9h 45m',\n",
       " '24h 55m',\n",
       " '11h 55m',\n",
       " '8h 20m',\n",
       " '17h 25m',\n",
       " '2h 50m',\n",
       " '7h 5m',\n",
       " '21h 15m',\n",
       " '5h',\n",
       " '2h 25m',\n",
       " '5h 15m',\n",
       " '15h 10m',\n",
       " '11h 45m',\n",
       " '3h',\n",
       " '4h 55m',\n",
       " '2h 45m',\n",
       " '2h 30m',\n",
       " '9h 45m',\n",
       " '2h 40m',\n",
       " '6h 30m',\n",
       " '5h 30m',\n",
       " '2h 40m',\n",
       " '34h 5m',\n",
       " '15h 25m',\n",
       " '9h 30m',\n",
       " '6h 5m',\n",
       " '2h 40m',\n",
       " '5h 50m',\n",
       " '2h 45m',\n",
       " '27h 20m',\n",
       " '3h',\n",
       " '5h 15m',\n",
       " '16h',\n",
       " '7h',\n",
       " '8h 40m',\n",
       " '9h',\n",
       " '2h 20m',\n",
       " '2h 20m',\n",
       " '6h 20m',\n",
       " '4h 25m',\n",
       " '2h 20m',\n",
       " '11h 5m',\n",
       " '16h 55m',\n",
       " '13h 45m',\n",
       " '25h 45m',\n",
       " '13h 55m',\n",
       " '5h 15m',\n",
       " '19h 15m',\n",
       " '7h 30m',\n",
       " '9h',\n",
       " '24h',\n",
       " '18h 10m',\n",
       " '2h 55m',\n",
       " '1h 30m',\n",
       " '21h 20m',\n",
       " '6h 15m',\n",
       " '14h 30m',\n",
       " '1h 20m',\n",
       " '12h 55m',\n",
       " '2h 25m',\n",
       " '22h 30m',\n",
       " '22h 55m',\n",
       " '5h 25m',\n",
       " '1h 25m',\n",
       " '7h 15m',\n",
       " '12h 20m',\n",
       " '20h 10m',\n",
       " '21h 15m',\n",
       " '2h 15m',\n",
       " '10h 30m',\n",
       " '14h 5m',\n",
       " '2h 50m',\n",
       " '16h 25m',\n",
       " '5h 15m',\n",
       " '13h 50m',\n",
       " '2h 15m',\n",
       " '7h 15m',\n",
       " '27h 5m',\n",
       " '27h 35m',\n",
       " '5h',\n",
       " '2h 50m',\n",
       " '2h 50m',\n",
       " '15h 20m',\n",
       " '5h 20m',\n",
       " '10h 10m',\n",
       " '27h 35m',\n",
       " '9h 35m',\n",
       " '28h 10m',\n",
       " '14h',\n",
       " '2h 35m',\n",
       " '2h 55m',\n",
       " '2h 35m',\n",
       " '4h 40m',\n",
       " '2h 30m',\n",
       " '11h 15m',\n",
       " '7h 55m',\n",
       " '8h 50m',\n",
       " '1h 25m',\n",
       " '15h 40m',\n",
       " '2h 35m',\n",
       " '2h 55m',\n",
       " '13h 25m',\n",
       " '4h 35m',\n",
       " '18h 30m',\n",
       " '2h 15m',\n",
       " '38h 15m',\n",
       " '3h 5m',\n",
       " '25h 50m',\n",
       " '3h 10m',\n",
       " '2h 50m',\n",
       " '11h 30m',\n",
       " '6h 35m',\n",
       " '27h 35m',\n",
       " '2h 35m',\n",
       " '7h 30m',\n",
       " '3h 15m',\n",
       " '9h 35m',\n",
       " '12h 30m',\n",
       " '11h 20m',\n",
       " '3h',\n",
       " '2h 35m',\n",
       " '1h 30m',\n",
       " '26h 10m',\n",
       " '4h 55m',\n",
       " '2h 45m',\n",
       " '8h 40m',\n",
       " '7h 5m',\n",
       " '2h 45m',\n",
       " '24h 50m',\n",
       " '2h 20m',\n",
       " '11h',\n",
       " '9h',\n",
       " '2h 55m',\n",
       " '3h 5m',\n",
       " '2h 35m',\n",
       " '13h',\n",
       " '1h 25m',\n",
       " '7h 35m',\n",
       " '2h 40m',\n",
       " '12h 55m',\n",
       " '29h 20m',\n",
       " '5h',\n",
       " '8h 40m',\n",
       " '3h',\n",
       " '3h',\n",
       " '2h 45m',\n",
       " '12h 10m',\n",
       " '2h 20m',\n",
       " '7h 55m',\n",
       " '29h 35m',\n",
       " '26h 55m',\n",
       " '12h 45m',\n",
       " '7h 35m',\n",
       " '2h 30m',\n",
       " '11h 45m',\n",
       " '23h 40m',\n",
       " '13h 50m',\n",
       " '12h 50m',\n",
       " '9h 50m',\n",
       " '2h 30m',\n",
       " '21h 55m',\n",
       " '26h 10m',\n",
       " '2h 55m',\n",
       " '20h 15m',\n",
       " '20h 10m',\n",
       " '10h 55m',\n",
       " '2h 30m',\n",
       " '21h 10m',\n",
       " '2h 35m',\n",
       " '20h 40m',\n",
       " '2h 50m',\n",
       " '6h 30m',\n",
       " '30h',\n",
       " '13h 10m',\n",
       " '2h 55m',\n",
       " '18h',\n",
       " '8h',\n",
       " '8h 45m',\n",
       " '12h 30m',\n",
       " '13h 20m',\n",
       " '6h 10m',\n",
       " '7h 30m',\n",
       " '19h 15m',\n",
       " '2h 35m',\n",
       " '12h',\n",
       " '3h',\n",
       " '2h 50m',\n",
       " '1h 25m',\n",
       " '23h 5m',\n",
       " '2h 50m',\n",
       " '22h 5m',\n",
       " '2h 15m',\n",
       " '5h 45m',\n",
       " '2h 50m',\n",
       " '9h 30m',\n",
       " '8h 55m',\n",
       " '17h 45m',\n",
       " '2h 50m',\n",
       " '21h 45m',\n",
       " '12h 35m',\n",
       " '1h 30m',\n",
       " '2h 15m',\n",
       " '3h 55m',\n",
       " '2h 45m',\n",
       " '7h 30m',\n",
       " '17h 20m',\n",
       " '1h 30m',\n",
       " '1h 30m',\n",
       " '30h 30m',\n",
       " '5h',\n",
       " '11h 15m',\n",
       " '2h 50m',\n",
       " '19h 5m',\n",
       " '6h 35m',\n",
       " '7h 30m',\n",
       " '2h 15m',\n",
       " '2h 45m',\n",
       " '12h 45m',\n",
       " '2h 50m',\n",
       " '26h 35m',\n",
       " '12h 55m',\n",
       " '1h 30m',\n",
       " '8h 40m',\n",
       " '14h',\n",
       " '23h 30m',\n",
       " '29h 55m',\n",
       " '22h 10m',\n",
       " '4h 50m',\n",
       " '2h 45m',\n",
       " '13h 30m',\n",
       " '13h 20m',\n",
       " '2h 20m',\n",
       " '11h 55m',\n",
       " '5h 10m',\n",
       " '11h 45m',\n",
       " '13h 50m',\n",
       " '3h',\n",
       " '6h 10m',\n",
       " '2h 20m',\n",
       " '10h 45m',\n",
       " '2h 15m',\n",
       " '2h 55m',\n",
       " '1h 30m',\n",
       " '7h 35m',\n",
       " '26h 35m',\n",
       " '4h 45m',\n",
       " '21h 25m',\n",
       " '8h 35m',\n",
       " '8h 30m',\n",
       " '2h 15m',\n",
       " '12h 40m',\n",
       " '8h 25m',\n",
       " '25h 40m',\n",
       " '2h 45m',\n",
       " '29h 20m',\n",
       " '2h 40m',\n",
       " '2h 50m',\n",
       " '2h 20m',\n",
       " '7h 15m',\n",
       " '20h 15m',\n",
       " '11h 45m',\n",
       " '2h 35m',\n",
       " '13h 30m',\n",
       " '1h 30m',\n",
       " '24h 35m',\n",
       " '20h 35m',\n",
       " '2h 50m',\n",
       " '4h 35m',\n",
       " '2h 20m',\n",
       " '11h 45m',\n",
       " '1h 30m',\n",
       " '12h 50m',\n",
       " '24h 30m',\n",
       " '11h 30m',\n",
       " '4h',\n",
       " '11h',\n",
       " '5h 25m',\n",
       " '2h 35m',\n",
       " '5h 10m',\n",
       " '10h 10m',\n",
       " '26h 55m',\n",
       " '7h 15m',\n",
       " '8h 10m',\n",
       " '2h 55m',\n",
       " '5h 15m',\n",
       " '2h 20m',\n",
       " '1h 25m',\n",
       " '4h 50m',\n",
       " '2h 20m',\n",
       " '2h 20m',\n",
       " '13h 15m',\n",
       " '19h 10m',\n",
       " '22h 40m',\n",
       " '14h 50m',\n",
       " '15h 10m',\n",
       " '7h 5m',\n",
       " '1h 25m',\n",
       " '2h 55m',\n",
       " '8h',\n",
       " '2h 45m',\n",
       " '14h 55m',\n",
       " '1h 30m',\n",
       " '16h 15m',\n",
       " '2h 50m',\n",
       " '5h',\n",
       " '22h 15m',\n",
       " '5h 30m',\n",
       " '8h 35m',\n",
       " '15h 25m',\n",
       " '21h',\n",
       " '13h 30m',\n",
       " '2h 50m',\n",
       " '6h 45m',\n",
       " '1h 30m',\n",
       " '8h 40m',\n",
       " '1h 30m',\n",
       " '2h 50m',\n",
       " '2h 35m',\n",
       " '16h 5m',\n",
       " '23h 50m',\n",
       " '27h 5m',\n",
       " '16h 15m',\n",
       " '15h 25m',\n",
       " '2h 50m',\n",
       " '8h 35m',\n",
       " '13h 50m',\n",
       " '2h 40m',\n",
       " '2h 25m',\n",
       " '3h 10m',\n",
       " '2h 55m',\n",
       " '28h 40m',\n",
       " '2h 50m',\n",
       " '2h 35m',\n",
       " '9h 40m',\n",
       " '16h 40m',\n",
       " '6h 20m',\n",
       " '27h',\n",
       " '14h 35m',\n",
       " '12h 10m',\n",
       " '7h 35m',\n",
       " '2h 20m',\n",
       " '19h 10m',\n",
       " '20h',\n",
       " '3h',\n",
       " '5h 10m',\n",
       " '2h 40m',\n",
       " '29h 35m',\n",
       " '4h 55m',\n",
       " '3h',\n",
       " '12h 30m',\n",
       " '12h 35m',\n",
       " '26h 20m',\n",
       " '13h 15m',\n",
       " '2h 50m',\n",
       " '12h 45m',\n",
       " '16h 20m',\n",
       " '2h 40m',\n",
       " '8h 15m',\n",
       " '12h',\n",
       " '2h 50m',\n",
       " '11h',\n",
       " '2h 20m',\n",
       " '13h 50m',\n",
       " '8h 10m',\n",
       " '12h 10m',\n",
       " '2h 40m',\n",
       " '2h 45m',\n",
       " '2h 50m',\n",
       " '24h',\n",
       " '3h',\n",
       " '12h 35m',\n",
       " '25h 50m',\n",
       " '2h 55m',\n",
       " '17h 5m',\n",
       " '29h 20m',\n",
       " '8h 55m',\n",
       " '3h 5m',\n",
       " '9h',\n",
       " '10h 25m',\n",
       " '16h 45m',\n",
       " '2h 45m',\n",
       " '26h 20m',\n",
       " '13h 15m',\n",
       " '10h 30m',\n",
       " '7h 35m',\n",
       " '11h',\n",
       " '2h 50m',\n",
       " '25h 45m',\n",
       " '6h 25m',\n",
       " '1h 25m',\n",
       " '20h 10m',\n",
       " '11h 55m',\n",
       " '2h 50m',\n",
       " '5h',\n",
       " '27h',\n",
       " '2h 50m',\n",
       " '1h 15m',\n",
       " '10h 15m',\n",
       " '24h 50m',\n",
       " '2h 50m',\n",
       " '20h',\n",
       " '5h 30m',\n",
       " '2h 45m',\n",
       " '6h 55m',\n",
       " '2h 45m',\n",
       " '5h 15m',\n",
       " '6h 15m',\n",
       " '2h 50m',\n",
       " '19h 25m',\n",
       " '2h 55m',\n",
       " '11h 25m',\n",
       " '14h 5m',\n",
       " '2h 45m',\n",
       " '3h 15m',\n",
       " '14h 20m',\n",
       " '12h 30m',\n",
       " '5h 20m',\n",
       " '26h 20m',\n",
       " '3h 5m',\n",
       " '1h 30m',\n",
       " '7h 5m',\n",
       " '6h 10m',\n",
       " '8h',\n",
       " '12h 5m',\n",
       " '12h 55m',\n",
       " '9h 35m',\n",
       " '15h 10m',\n",
       " '11h 40m',\n",
       " '2h 55m',\n",
       " '4h 30m',\n",
       " '2h 40m',\n",
       " '2h 30m',\n",
       " '24h 10m',\n",
       " '28h 20m',\n",
       " '1h 25m',\n",
       " '10h 10m',\n",
       " '11h 30m',\n",
       " '2h 50m',\n",
       " '13h 45m',\n",
       " '6h 25m',\n",
       " '12h 55m',\n",
       " '4h 25m',\n",
       " '13h 30m',\n",
       " '24h 5m',\n",
       " '16h 5m',\n",
       " '8h 20m',\n",
       " '3h 40m',\n",
       " '9h 30m',\n",
       " '28h 15m',\n",
       " '13h 20m',\n",
       " '17h 50m',\n",
       " '20h 20m',\n",
       " '28h 5m',\n",
       " '15h 25m',\n",
       " '12h 5m',\n",
       " '10h 20m',\n",
       " '12h 25m',\n",
       " '7h 35m',\n",
       " '2h 50m',\n",
       " '4h 35m',\n",
       " '11h 15m',\n",
       " '2h 35m',\n",
       " '26h',\n",
       " '2h 25m',\n",
       " '2h 35m',\n",
       " '14h 15m',\n",
       " '2h 35m',\n",
       " '22h 5m',\n",
       " '6h 35m',\n",
       " '8h 30m',\n",
       " '18h 45m',\n",
       " '35h 15m',\n",
       " '2h 15m',\n",
       " '2h 30m',\n",
       " '2h 20m',\n",
       " '2h 30m',\n",
       " '7h 35m',\n",
       " '35h 35m',\n",
       " '26h 30m',\n",
       " '11h 30m',\n",
       " '26h 40m',\n",
       " '3h',\n",
       " '1h 30m',\n",
       " '2h 55m',\n",
       " '14h 15m',\n",
       " '16h 55m',\n",
       " '12h 20m',\n",
       " '29h 35m',\n",
       " '20h 20m',\n",
       " '8h 25m',\n",
       " '13h 50m',\n",
       " '10h',\n",
       " '19h 25m',\n",
       " '28h',\n",
       " '2h 50m',\n",
       " '17h 20m',\n",
       " '2h 45m',\n",
       " '2h 55m',\n",
       " '13h 25m',\n",
       " '12h 55m',\n",
       " '9h 45m',\n",
       " '21h 25m',\n",
       " '15h 40m',\n",
       " '2h 20m',\n",
       " '14h 25m',\n",
       " '2h 50m',\n",
       " '6h 35m',\n",
       " '22h 10m',\n",
       " '14h 25m',\n",
       " '2h 30m',\n",
       " '23h 35m',\n",
       " '13h 5m',\n",
       " '1h 30m',\n",
       " '5h 5m',\n",
       " '1h 20m',\n",
       " '22h 55m',\n",
       " '1h 30m',\n",
       " '2h 35m',\n",
       " '37h 20m',\n",
       " '2h 50m',\n",
       " '2h 50m',\n",
       " '36h 10m',\n",
       " '9h 15m',\n",
       " '1h 25m',\n",
       " '5h 5m',\n",
       " '2h 15m',\n",
       " '15h 10m',\n",
       " '25h 55m',\n",
       " '5h 40m',\n",
       " '7h 30m',\n",
       " '17h 45m',\n",
       " '2h 20m',\n",
       " '8h 15m',\n",
       " '3h 40m',\n",
       " '5h 55m',\n",
       " '2h 30m',\n",
       " '17h 30m',\n",
       " '13h 50m',\n",
       " '2h 15m',\n",
       " '21h 35m',\n",
       " '9h 30m',\n",
       " '8h 40m',\n",
       " '10h 25m',\n",
       " '35h 5m',\n",
       " '14h 15m',\n",
       " '3h',\n",
       " '25h 50m',\n",
       " '2h 30m',\n",
       " '8h 40m',\n",
       " '10h',\n",
       " '11h 30m',\n",
       " '7h 30m',\n",
       " ...]"
      ]
     },
     "execution_count": 242,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "duration=list(train_data['Duration'])\n",
    "duration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 243,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "25"
      ]
     },
     "execution_count": 243,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x=\"21h 25m\"\n",
    "int(x.split(sep = \"m\")[0].split(sep=\"h\")[-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 244,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(len(duration)):\n",
    "    if(len(duration[i].split())!=2):\n",
    "        if \"h\" in duration[i]:\n",
    "            duration[i]=duration[i].strip()+\"0m\"\n",
    "        else:\n",
    "            duration[i]=\"0h\"+duration[i]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 246,
   "metadata": {},
   "outputs": [],
   "source": [
    "duration_hours=[]\n",
    "duration_mins=[]\n",
    "for i in range(len(duration)):\n",
    "    duration_hours.append(int(duration[i].split(sep=\"h\")[0]))   \n",
    "    duration_mins.append(int(duration[i].split(sep = \"m\")[0].split(sep=\"h\")[-1]))   #"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 247,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data[\"Duration_hours\"] = duration_hours\n",
    "train_data[\"Duration_mins\"] = duration_mins"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 248,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data.drop('Duration',axis = 1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 249,
   "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>Airline</th>\n",
       "      <th>Source</th>\n",
       "      <th>Destination</th>\n",
       "      <th>Route</th>\n",
       "      <th>Total_Stops</th>\n",
       "      <th>Additional_Info</th>\n",
       "      <th>Price</th>\n",
       "      <th>Journey_Day</th>\n",
       "      <th>Journey_Month</th>\n",
       "      <th>Dep_hour</th>\n",
       "      <th>Dep_min</th>\n",
       "      <th>Arrival_hour</th>\n",
       "      <th>Arrival_min</th>\n",
       "      <th>Duration_hours</th>\n",
       "      <th>Duration_mins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>New Delhi</td>\n",
       "      <td>BLR → DEL</td>\n",
       "      <td>non-stop</td>\n",
       "      <td>No info</td>\n",
       "      <td>3897</td>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>22</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Air India</td>\n",
       "      <td>Kolkata</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>CCU → IXR → BBI → BLR</td>\n",
       "      <td>2 stops</td>\n",
       "      <td>No info</td>\n",
       "      <td>7662</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>50</td>\n",
       "      <td>13</td>\n",
       "      <td>15</td>\n",
       "      <td>7</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Jet Airways</td>\n",
       "      <td>Delhi</td>\n",
       "      <td>Cochin</td>\n",
       "      <td>DEL → LKO → BOM → COK</td>\n",
       "      <td>2 stops</td>\n",
       "      <td>No info</td>\n",
       "      <td>13882</td>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "      <td>9</td>\n",
       "      <td>25</td>\n",
       "      <td>4</td>\n",
       "      <td>25</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>Kolkata</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>CCU → NAG → BLR</td>\n",
       "      <td>1 stop</td>\n",
       "      <td>No info</td>\n",
       "      <td>6218</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "      <td>18</td>\n",
       "      <td>5</td>\n",
       "      <td>23</td>\n",
       "      <td>30</td>\n",
       "      <td>5</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>New Delhi</td>\n",
       "      <td>BLR → NAG → DEL</td>\n",
       "      <td>1 stop</td>\n",
       "      <td>No info</td>\n",
       "      <td>13302</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>16</td>\n",
       "      <td>50</td>\n",
       "      <td>21</td>\n",
       "      <td>35</td>\n",
       "      <td>4</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Airline    Source Destination                  Route Total_Stops  \\\n",
       "0       IndiGo  Banglore   New Delhi              BLR → DEL    non-stop   \n",
       "1    Air India   Kolkata    Banglore  CCU → IXR → BBI → BLR     2 stops   \n",
       "2  Jet Airways     Delhi      Cochin  DEL → LKO → BOM → COK     2 stops   \n",
       "3       IndiGo   Kolkata    Banglore        CCU → NAG → BLR      1 stop   \n",
       "4       IndiGo  Banglore   New Delhi        BLR → NAG → DEL      1 stop   \n",
       "\n",
       "  Additional_Info  Price  Journey_Day  Journey_Month  Dep_hour  Dep_min  \\\n",
       "0         No info   3897           24              3        22       20   \n",
       "1         No info   7662            1              5         5       50   \n",
       "2         No info  13882            9              6         9       25   \n",
       "3         No info   6218           12              5        18        5   \n",
       "4         No info  13302            1              3        16       50   \n",
       "\n",
       "   Arrival_hour  Arrival_min  Duration_hours  Duration_mins  \n",
       "0             1           10               2             50  \n",
       "1            13           15               7             25  \n",
       "2             4           25              19              0  \n",
       "3            23           30               5             25  \n",
       "4            21           35               4             45  "
      ]
     },
     "execution_count": 249,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 250,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Jet Airways                          3849\n",
       "IndiGo                               2053\n",
       "Air India                            1751\n",
       "Multiple carriers                    1196\n",
       "SpiceJet                              818\n",
       "Vistara                               479\n",
       "Air Asia                              319\n",
       "GoAir                                 194\n",
       "Multiple carriers Premium economy      13\n",
       "Jet Airways Business                    6\n",
       "Vistara Premium economy                 3\n",
       "Trujet                                  1\n",
       "Name: Airline, dtype: int64"
      ]
     },
     "execution_count": 250,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data[\"Airline\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 251,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABQQAAAGkCAYAAABnzShmAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdfZSVZ2Eu/Gt/DBAYCplxBpBQgh9NNLFJ2qmttR1qakOQUFvanHpI9Zye16ppj3rSFVoEmpS+ZRF9WQn1WLQ9tnoabWusCj10mJg2TaISI0ZrGiWJNhARPDAMJHwkDMPe+/0DoZkkkEmczZ5h/35rZc3sez/7methZ/i45r6fu1Cr1WoBAAAAAJpCsdEBAAAAAIAzRyEIAAAAAE1EIQgAAAAATUQhCAAAAABNRCEIAAAAAE1EIQgAAAAATaTc6ACN0t9/KNVqrdExAAAAAGDEdXRMPuVzZggCAAAAQBNRCAIAAABAE1EIAgAAAEATUQgCAAAAQBNRCAIAAABAE1EIAgAAAEATUQgCAAAAQBNRCAIAAABAE1EIAgAAAEATUQgCAAAAQBOpayG4YcOGLFiwIAsWLMj73//+JMnmzZuzcOHCXHHFFbnllltOHrt169YsWrQo8+bNy/Lly3Ps2LEkya5du3LNNdfkyiuvzLXXXpvDhw8nSQ4cOJB3vOMdmT9/fq655pr09fXV81IAAAAA4KxQt0LwqaeeyqpVq3Lrrbdmw4YN+epXv5o777wzy5Yty7p169LT05MHH3wwd999d5JkyZIlueGGG3L77benVqvltttuS5KsXLkyixcvTm9vby6++OKsW7cuSbJ27dp0dXVl06ZNufrqq7Nq1ap6Xcro9tRTjU4AAAAAwBhSt0KwUqmkWq3mqaeeyrFjx3Ls2LG0trZm9uzZmTVrVsrlchYuXJje3t7s3LkzR44cyaWXXpokWbRoUXp7ezM4OJgtW7Zk3rx5Q8aT5K677srChQuTJFdddVXuueeeDA4O1utyRqe+/uQfP5/s7W90EgAAAADGiLoVgq2trXnve9+b+fPnZ+7cuZk5c2b27NmTjo6Ok8d0dnZm9+7dzxrv6OjI7t27s3///rS2tqZcLg8ZTzLkNeVyOa2trdm3b1+9Lmf0qVaTr37t+Odbvnb8MQAAAAA8j3K9TvzQQw/lM5/5TP7lX/4lkydPzvXXX5/t27enUCicPKZWq6VQKKRarT7n+ImPT/fMx09/TbE4/H6zvb31BV7R6PLkN7bm8MDA8QcDRzNp965M/PFXNTYUAAAAAKNe3QrBL37xi3nd616X9vb2JMeX+/7lX/5lSqXSyWP6+vrS2dmZ6dOnD9kUZO/evens7ExbW1sOHjyYSqWSUql08vjk+OzCvXv3Zvr06Tl27FgOHz6cqVOnDjtff/+hVKu1EbraM+zIkeS+f00qleOPjx3L4S//aw6f+5JkwoTGZgMAAACg4To6Jp/yubotGb7wwguzefPmPPnkk6nVarnzzjtzySWXZNu2bXnsscdSqVSycePGdHd3Z+bMmRk/fnzuv//+JMd3J+7u7k5LS0u6urrS09OTJFm/fn26u7uTJHPnzs369euTJD09Penq6kpLS0u9Lmd0+e7OpHa8zJzY9ePHx2q1ZMfOBoYCAAAAYCwo1Gq1uk2T+4u/+It89rOfTUtLS17zmtfkxhtvzNe+9rWsXr06AwMDmTt3bt73vvelUCjkoYceyooVK3Lo0KFcdNFFWb16dcaNG5edO3dm6dKl6e/vz4wZM3LzzTdnypQpefzxx7N06dLs2LEjkydPzpo1a3LeeecNO9uYnyG46Z/S/l9/LcVx41I9ejT9H/9MMv+NZggCAAAAcNoZgnUtBEezMV0IJskj30nH6y9L5YmDKU2ZnL7N/5q88uWNTgUAAADAKNCQJcPU2StellqtltKUyanVasnL5zQ6EQAAAABjgEJwrCoWUygUcvS7u47vvPwCdlgGAAAAoHlZMjwGtbdPSvE5CsBqtZr+/sMNSAQAAADAaHK6JcPlM5iDEVIsFvPUvz38rPFzXnNBA9IAAAAAMJZYZwoAAAAATUQhCAAAAABNRCEIAAAAAE1EIQgAAAAATcQuw2NIe9ukFEvP3+FWK9X077PbMAAAAECzssvwWaJYKubQF7c873GtP/dTZyANAAAAAGORJcMAAAAA0EQUggAAAADQRBSCAAAAANBEFIIAAAAA0EQUggAAAADQRBSCAAAAANBEFIIAAAAA0EQUggAAAADQRBSCAAAAANBEFIIAAAAA0EQKtVqt1ugQjdDffyjV6ui99Pa2SSmWRq6vrVaq6d93eMTOBwAAAMDo1dEx+ZTPlc9gDl6AYqmYJ/7xzhE735QFl4/YuQAAAAAYuywZBgAAAIAmohAEAAAAgCaiEAQAAACAJqIQBAAAAIAmohAEAAAAgCaiEAQAAACAJqIQBAAAAIAmohAEAAAAgCaiEAQAAACAJqIQBAAAAIAmohAEAAAAgCaiEAQAAACAJqIQBAAAAIAmUq7XiT/96U/nE5/4xMnH3/ve9/LmN785b3zjG7N69eoMDAxk/vz5ue6665IkW7duzfLly3P48OF0dXVl5cqVKZfL2bVrV5YsWZL+/v7MmTMna9asyaRJk3LgwIFcf/312bFjR9ra2rJ27dp0dHTU63IAAAAA4KxQtxmCV199dTZs2JANGzZkzZo1aW9vz2//9m9n2bJlWbduXXp6evLggw/m7rvvTpIsWbIkN9xwQ26//fbUarXcdtttSZKVK1dm8eLF6e3tzcUXX5x169YlSdauXZuurq5s2rQpV199dVatWlWvSwEAAACAs8YZWTL8R3/0R7nuuuuyY8eOzJ49O7NmzUq5XM7ChQvT29ubnTt35siRI7n00kuTJIsWLUpvb28GBwezZcuWzJs3b8h4ktx1111ZuHBhkuSqq67KPffck8HBwTNxOQAAAAAwZtVtyfAJmzdvzpEjRzJ//vxs3LhxyLLezs7O7N69O3v27Bky3tHRkd27d2f//v1pbW1NuVweMp5kyGvK5XJaW1uzb9++TJs2bVi52ttbR+oSx4yOjsmNjgAAAABAg9W9EPy7v/u7/NZv/VaSpFqtplAonHyuVqulUCiccvzEx6d75uOnv6ZYHP6Ex/7+Q6lWay/kUs6oepR3fX0HR/ycAAAAAIw+p+uW6rpk+OjRo9myZUsuv/zyJMn06dPT19d38vm+vr50dnY+a3zv3r3p7OxMW1tbDh48mEqlMuT45Pjswr179yZJjh07lsOHD2fq1Kn1vBwAAAAAGPPqWgg+/PDDOf/88zNx4sQkySWXXJJt27blscceS6VSycaNG9Pd3Z2ZM2dm/Pjxuf/++5MkGzZsSHd3d1paWtLV1ZWenp4kyfr169Pd3Z0kmTt3btavX58k6enpSVdXV1paWup5OQAAAAAw5tV1yfCOHTsyffr0k4/Hjx+fm266Ke9+97szMDCQuXPn5sorr0ySrFmzJitWrMihQ4dy0UUX5W1ve1uS5MYbb8zSpUvz4Q9/ODNmzMjNN9+cJHnve9+bpUuXZsGCBZk8eXLWrFlTz0sBAAAAgLNCoVarjd4b6dVRo+4h2N42McVS6Yx/3eGqVirp3/dko2MAAAAA8EM43T0E676pCEMVS6Xs+5sNjY5xSm2L39zoCAAAAADUUV3vIQgAAAAAjC4KQQAAAABoIgpBAAAAAGgiCkEAAAAAaCIKQQAAAABoIgpBAAAAAGgiCkEAAAAAaCIKQQAAAABoIgpBAAAAAGgiCkEAAAAAaCIKQQAAAABoIgpBAAAAAGgiCkEAAAAAaCIKQQAAAABoIgpBAAAAAGgiCkEAAAAAaCIKQQAAAABoIgpBAAAAAGgiCkEAAAAAaCIKQQAAAABoIgpBAAAAAGgiCkEAAAAAaCIKQQAAAABoIgpBAAAAAGgiCkEAAAAAaCIKQQAAAABoIgpBAAAAAGgiCkEAAAAAaCIKQQAAAABoIgpBAAAAAGgiCkEAAAAAaCIKQQAAAABoIoVarVZrdIhG6O8/lGr19Jfefu7EFMulM5To7FU9Vkn//icbHQMAAACgaXR0TD7lc+V6fuE777wzH/rQh/LUU0/l9a9/fVasWJHNmzdn9erVGRgYyPz583PdddclSbZu3Zrly5fn8OHD6erqysqVK1Mul7Nr164sWbIk/f39mTNnTtasWZNJkyblwIEDuf7667Njx460tbVl7dq16ejoGNH8xXIpfR/+xIiesxl1XPubjY4AAAAAwA/Ubcnwjh07cuONN2bdunX5h3/4h3zrW9/K3XffnWXLlmXdunXp6enJgw8+mLvvvjtJsmTJktxwww25/fbbU6vVcttttyVJVq5cmcWLF6e3tzcXX3xx1q1blyRZu3Zturq6smnTplx99dVZtWpVvS4FAAAAAM4adSsE77jjjrzpTW/K9OnT09LSkltuuSXnnHNOZs+enVmzZqVcLmfhwoXp7e3Nzp07c+TIkVx66aVJkkWLFqW3tzeDg4PZsmVL5s2bN2Q8Se66664sXLgwSXLVVVflnnvuyeDgYL0uBwAAAADOCnVbMvzYY4+lpaUl73rXu/L9738/v/ALv5BXvvKVQ5b1dnZ2Zvfu3dmzZ8+Q8Y6OjuzevTv79+9Pa2tryuXykPEkQ15TLpfT2tqaffv2Zdq0afW6JAAAAAAY8+pWCFYqlXz1q1/NrbfemokTJ+baa6/NhAkTUigUTh5Tq9VSKBRSrVafc/zEx6d75uOnv6ZYHP6Ex/b21hd4RfwwTncjSwAAAADOnLoVgi95yUvyute9Lm1tbUmSN77xjent7U2p9B+79vb19aWzszPTp09PX1/fyfG9e/ems7MzbW1tOXjwYCqVSkql0snjk+OzC/fu3Zvp06fn2LFjOXz4cKZOnTrsfMPZZViJNXL6+g42OgIAAABA0zhdr1W3ewi+4Q1vyBe/+MUcOHAglUolX/jCF3LllVdm27Zteeyxx1KpVLJx48Z0d3dn5syZGT9+fO6///4kyYYNG9Ld3Z2WlpZ0dXWlp6cnSbJ+/fp0d3cnSebOnZv169cnSXp6etLV1ZWWlpZ6XQ4AAAAAnBUKtVrt9NPkfgh///d/n49//OMZHBzM61//+qxYsSL33XdfVq9enYGBgcydOzfve9/7UigU8tBDD2XFihU5dOhQLrrooqxevTrjxo3Lzp07s3Tp0vT392fGjBm5+eabM2XKlDz++ONZunRpduzYkcmTJ2fNmjU577zzhp1tuDME+z78iR/2l6HpdVz7m2YIAgAAAJxBp5shWNdCcDRTCJ45CkEAAACAM6shS4YBAAAAgNFHIQgAAAAATUQhCAAAAABNRCEIAAAAAE1EIQgAAAAATUQhCAAAAABNRCEIAAAAAE1EIQgAAAAATUQhCAAAAABNRCEIAAAAAE1EIQgAAAAATUQhCAAAAABNRCEII6yy8xuNjgAAAABwSgpBGEFH7//bHL1nbY5+7W8bHQUAAADgOSkEYYRUjj6VyiOfP/75w59P5ehTDU4EAAAA8GwKQRghR+/4k2c8XtWgJAAAAACnphCEEVD5v99KDuwaOnhgZyr/96HGBAIAAAA4hXKjA4xW27Y9mi1bdubAw99sdJQxr+s738mUKdMaHaOuBr/2N6cY/0RKb/qT53wOAAAAoBHMEIQR0PITv/mCxgEAAAAaxQzBU5gz52V57WsvSd9j+xodZczreMUr0td3sNEx6qo0/cLkR146dNnwj8w8Pg4AAAAwipghCCNk3C+teMbj5Q1KAgAAAHBqCkEYIaVx56T0Y1cc//yCK1Iad06DEwEAAAA8myXDMILG/eR/TmX6q1OaeUmjowAAAAA8JzMEYYQpAwEAAIDRTCEIAAAAAE1EIQgAAAAATUQhCAAAAABNRCEIAAAAAE2kUKvVao0O0Qj9/YdSrZ7+0tvPnZhiuXSGEp29qscq6d//ZKNjAAAAADSNjo7Jp3yufAZzjDn1KLE6OiZn399sGPHzjpS2xW9OX9/BRscAAAAAoE4sGQYAAACAJqIQBAAAAIAmohAEAAAAgCaiEAQAAACAJqIQBAAAAIAmUtddht/61rdm3759KZePf5k//uM/zuHDh7N69eoMDAxk/vz5ue6665IkW7duzfLly3P48OF0dXVl5cqVKZfL2bVrV5YsWZL+/v7MmTMna9asyaRJk3LgwIFcf/312bFjR9ra2rJ27dp0dHTU83IAAAAAYMyr2wzBWq2W7du3Z8OGDSf/u+CCC7Js2bKsW7cuPT09efDBB3P33XcnSZYsWZIbbrght99+e2q1Wm677bYkycqVK7N48eL09vbm4osvzrp165Ika9euTVdXVzZt2pSrr746q1atqtelAAAAAMBZo26F4KOPPpok+W//7b/ll3/5l/OJT3wiDzzwQGbPnp1Zs2alXC5n4cKF6e3tzc6dO3PkyJFceumlSZJFixalt7c3g4OD2bJlS+bNmzdkPEnuuuuuLFy4MEly1VVX5Z577sng4GC9LgcAAAAAzgp1WzJ84MCBvO51r8sf/uEfZnBwMG9729vy9re/fciy3s7OzuzevTt79uwZMt7R0ZHdu3dn//79aW1tPbnk+MR4kiGvKZfLaW1tzb59+zJt2rRh5Wtvbx2pS31BatVq2ha/uSFfezhq1Wo6OiY3OgYAAAAAdVK3QvCyyy7LZZdddvLxr//6r+eDH/xgfvInf/LkWK1WS6FQSLVaTaFQeNb4iY9P98zHT39NsTj8CY/9/YdSrdaGffyZ1tExOU/8450jdr4pCy5PX9/BETsfAAAAAKPX6SZ81W3J8Fe/+tXce++9Jx/XarXMnDkzfX19J8f6+vrS2dmZ6dOnDxnfu3dvOjs709bWloMHD6ZSqQw5Pjk+u3Dv3r1JkmPHjuXw4cOZOnVqvS4HAAAAAM4KdSsEDx48mA984AMZGBjIoUOH8rnPfS6/93u/l23btuWxxx5LpVLJxo0b093dnZkzZ2b8+PG5//77kyQbNmxId3d3Wlpa0tXVlZ6eniTJ+vXr093dnSSZO3du1q9fnyTp6elJV1dXWlpa6nU5AAAAAHBWKNRqtbqtm127dm1uv/32VKvVLF68OP/lv/yX3HvvvVm9enUGBgYyd+7cvO9970uhUMhDDz2UFStW5NChQ7nooouyevXqjBs3Ljt37szSpUvT39+fGTNm5Oabb86UKVPy+OOPZ+nSpdmxY0cmT56cNWvW5Lzzzht2NkuGAQAAADhbnW7JcF0LwdFMIQgAAADA2aoh9xAEAAAAAEYfhSAAAAAANJFyowPw3KqVaqYsuHxEzwcAAAAACsFRqn/f4WeNdXRMzqEvbnne17b+3E+5XyAAAAAAz8mSYQAAAABoIgpBAAAAAGgilgyPIdVKNa0/91PDOg4AAAAAnotCcAw5cV/Bjo7JeerfHn7W8+e85gL3DgQAAADgtBSCY1C1Us05r7ngOccBAAAA4HTcQ3AM6v/yA+n7879JrVZLktRqtfT9+d+k/74HGpwMAAAAgNHODMGx6EdnJt/cmr0f+eR/jJVKyayZjcsEAAAAwJhghuBYNGFCctGFx0vA5PjHi191fBwAAAAATkMhOFa94mXJhPHHP58wPnn5nMbmAQAAAGBMUAiOVcVi0vUTxz//qZ84/hgAAAAAnkehdmJniibT338o1epZcOlPPZWcc06jUwAAAAAwinR0TD7lc6aVjXXKQAAAAABeAIUgAAAAADQRhSAAAAAANBGFIAAAAAA0EYUgAAAAADQRhSAAAAAANBGFIAAAAAA0EYUgAAAAADQRhSAAAAAANJFhFYLVajUf/ehH8wd/8Ac5dOhQ/vzP/zyVSqXe2QAAAACAETasQvADH/hAHnnkkTzwwANJki984QtZvXp1XYMBAAAAACNvWIXgvffem5tuuinjx49Pa2tr/uqv/ipf+tKX6p0NAAAAABhhwyoEy+VyisX/OHTcuHEpl8t1CwUAAAAA1MewWr0f+7Efyyc/+clUKpU8+uij+fjHP54LL7yw3tkAAAAAgBE2rBmCy5cvzze/+c309/dn8eLFefLJJ7Ns2bJ6ZwMAAAAARlihVqvVhnPgjh07MmvWrBw6dCjf/e538+pXv7re2eqqv/9QqtVhXToAAAAAjCkdHZNP+dywZgjeeuut+Z3f+Z0kyf79+/Pud787n/70p0cmHQAAAABwxgyrEPzUpz6Vv/3bv02SzJo1K+vXr89f//Vf1zUYAAAAADDyhlUIViqVtLa2nnw8efLkFAqFuoUCAAAAAOpjWIXgy172sqxZsyY7duzIjh078qd/+qc5//zzh/UF3v/+92fp0qVJks2bN2fhwoW54oorcsstt5w8ZuvWrVm0aFHmzZuX5cuX59ixY0mSXbt25ZprrsmVV16Za6+9NocPH06SHDhwIO94xzsyf/78XHPNNenr63sh1wwAAAAATWtYheDKlSuzffv2/Mqv/Ep+/dd/Pdu3b88f/dEfPe/r7r333nzuc59Lkhw5ciTLli3LunXr0tPTkwcffDB33313kmTJkiW54YYbcvvtt6dWq+W22247+XUXL16c3t7eXHzxxVm3bl2SZO3atenq6sqmTZty9dVXZ9WqVS/m2gEAAACg6QyrEHzJS16SD33oQ7n//vtz33335ZZbbklbW9tpX/P444/nlltuybve9a4kyQMPPJDZs2dn1qxZKZfLWbhwYXp7e7Nz584cOXIkl156aZJk0aJF6e3tzeDgYLZs2ZJ58+YNGU+Su+66KwsXLkySXHXVVbnnnnsyODj44n4FAAAAAKCJlE/35KpVq7J8+fKTpd4zfeQjHznla2+44YZcd911+f73v58k2bNnTzo6Ok4+39nZmd27dz9rvKOjI7t3787+/fvT2tqacrk8ZPyZ5yqXy2ltbc2+ffsybdq04VwzAAAAADSt0xaCr3vd65Lk5Cy94fr0pz+dGTNm5HWve10++9nPJkmq1eqQjUhqtVoKhcIpx098fLpTbWRSq9VSLA5rsuNJ7e2tz38QAAAAAJxlTlsIXn755UmS9evX53//7/897JP29PSkr68vb37zm/PEE0/kySefzM6dO1MqlU4e09fXl87OzkyfPn3IpiB79+5NZ2dn2tracvDgwVQqlZRKpZPHJ8dnF+7duzfTp0/PsWPHcvjw4UydOvUFXXh//6FUq7UX9BoAAAAAGAs6Oiaf8rlhTas7ePBgnnzyyWF/wY997GPZuHFjNmzYkPe85z25/PLL89GPfjTbtm3LY489lkqlko0bN6a7uzszZ87M+PHjc//99ydJNmzYkO7u7rS0tKSrqys9PT1JjpeS3d3dSZK5c+dm/fr1SY6Xj11dXWlpaRl2PgAAAABoVqedIXjCOeeckze84Q254IILMnHixJPjp7uH4DONHz8+N910U9797ndnYGAgc+fOzZVXXpkkWbNmTVasWJFDhw7loosuytve9rYkyY033pilS5fmwx/+cGbMmJGbb745SfLe9743S5cuzYIFCzJ58uSsWbNm2DkAAAAAoJkVarXaadfNPvLII/niF7+YgYGBTJ8+fchzv/qrv1rXcPVkyTAAAAAAZ6vTLRk+7QzBz3zmM3n/+9+f2bNn57vf/W7WrFmTn//5nx/xgAAAAADAmXHaQvDWW2/N//k//yfTpk3L17/+9dxyyy0KQQAAAAAYw553U5Fp06YlSS677LLs37+/7oEAAAAAgPo5bSFYKBSGPC6VSnUNA1AvR5/a1+gIAAAAMCo87wzBp3tmQQgwFhzq35oH77g2h/ofanQUAAAAaLjT7jL86le/OhMmTDj5+MiRI5kwYUJqtVoKhUK+9rWvnZGQ9WCXYWgOtWol37zzf+Tok3sybuK0XHT5LSkUzXYGAADg7Paidxm+4447RjwMwJm0Z1tvjg08kSQ5NvB4+rbdns6Xv6nBqQAAAKBxTlsIzpw580zlABhxg0cez/cf+lSqlYEkSbUykF0P/V3OnfmzaZkwtcHpAAAAoDFe0D0EAcaSfTu/lFqtOmSsVqtm/84vNSgRAAAANJ5CEDhrtc18fQqFob/NFQrFnDvz9Q1KBAAAAI2nEATOWi0TpmbGhb+RYml8kqRYGp+XXvgWy4UBAABoagpBGGETJ45rdASepnPOlSmPP14AlsdPTceceQ1OBAAAAI2lEIQRNmnS+EZH4GkKxVLOv+zaJMn5l/1OCsVSgxMBAABAYxVqtVqt0SEaob//UKrVprx06qyjY3L6+g42OgbPcHj/o5l07ssaHQMAAADOiI6Oyad8zgxB4Kx3qH9rHv7C+3Ko/6FGRwEAAICGUwgCZ7VatZLtX1+XJNn+9XWpVSsNTgQAAACNpRAEzmp7tvXm2MATSZJjA4+nb9vtDU4EAAAAjaUQBM5ag0cez/cf+lSqlYEkSbUykF0P/V0Gjzze4GQAAADQOApB4Ky1b+eXUqtV84tv/fv84lv/PklSq1Wzf+eXGpwMAAAAGkchCJy12ma+PoVCMaWWc1JqOSdJUigUc+7M1zc4GQAAADROudEBAOqlZcLUzLjwN1IoFJIkxdL4vPTCt6RlwtQGJwMAAIDGMUMQOKt1zrny5Ofl8VPTMWdeA9MAAABA4ykEgbNaoVg6+fn5l/3OkMcAAADQjBSCQNNobb+w0REAAACg4RSCAAAAANBEFIIANMRTR/obHQEAAKApKQQBOOP27vtWNt31zuzdv7XRUQAAAJqOQhCAM6pareS+f705SfKVr9+carXS4EQAAADNRSEIwBn1ne3/mIGjjydJjhzdn+889o8NTgQAANBcFIIAnDFHBvbnm9/+5JCxbz7yyRwZeLxBiQAAAJqPQhBGWK1WS3v7pEbHgFHp0R13pFYbukS4Vqtk2447GpQIAACg+ZQbHQDONoVCIYVCodExYFTynQEAANB4ZggCcMbMmXVFCoWhP4sqFMqZM+uXGpQIAACg+SgEAThjJoyfmoteufhpI4Vc9MprMmH81IZlAgAAaDYKQQDOqFecvyATxp+bJJkw/ty84vw3NTgRAABAc6lrIfinf/qnedOb3pQFCxbkYx/7WJJk8+bNWbhwYa644orcchhnarkAACAASURBVMstJ4/dunVrFi1alHnz5mX58uU5duxYkmTXrl255pprcuWVV+baa6/N4cOHkyQHDhzIO97xjsyfPz/XXHNN+vr66nkpAIyQYrGU115yXZLktZdel2Kx1OBEAAAAzaVuheBXvvKVfPnLX84//MM/5DOf+UxuvfXWPPTQQ1m2bFnWrVuXnp6ePPjgg7n77ruTJEuWLMkNN9yQ22+/PbVaLbfddluSZOXKlVm8eHF6e3tz8cUXZ926dUmStWvXpqurK5s2bcrVV1+dVatW1etSABhhL2l7deb/wp/nJee+qtFRAAAAmk7dCsHXvva1+eu//uuUy+X09/enUqnkwIEDmT17dmbNmpVyuZyFCxemt7c3O3fuzJEjR3LppZcmSRYtWpTe3t4MDg5my5YtmTdv3pDxJLnrrruycOHCJMlVV12Ve+65J4ODg/W6HABG2DkT2hsdAQAAoCmVn/+QF6+lpSUf/OAH81d/9Ve58sors2fPnnR0dJx8vrOzM7t3737WeEdHR3bv3p39+/entbU15XJ5yHiSIa8pl8tpbW3Nvn37Mm3atGFla29vHanLhOfU0TG50RF4Bu8JAAAA1LkQTJL3vOc9+e3f/u28613vyvbt21MoFE4+V6vVUigUUq1Wn3P8xMene+bjp7+mWBz+hMf+/kOpVmsv8Grg+Z0onfr6DjY4CSd4TwAAAGg2p5sUU7clw//+7/+erVu3JknOOeecXHHFFbnvvvuGbP7R19eXzs7OTJ8+fcj43r1709nZmba2thw8eDCVSmXI8cnx2YV79+5Nkhw7diyHDx/O1KlT63U5AAAAAHBWqFsh+L3vfS8rVqzI0aNHc/To0fzzP/9z3vKWt2Tbtm157LHHUqlUsnHjxnR3d2fmzJkZP3587r///iTJhg0b0t3dnZaWlnR1daWnpydJsn79+nR3dydJ5s6dm/Xr1ydJenp60tXVlZaWlnpdDgAjbOLEcY2OAAAA0JQKtVqtbutm/+f//J/ZtGlTSqVSrrjiirz73e/Ovffem9WrV2dgYCBz587N+973vhQKhTz00ENZsWJFDh06lIsuuiirV6/OuHHjsnPnzixdujT9/f2ZMWNGbr755kyZMiWPP/54li5dmh07dmTy5MlZs2ZNzjvvvGFns2SYerE8dfTxnoxOHR2TvScAAAB1crolw3UtBEczhSD1onwafbwno5NCEAAAoH4acg9BAAAAAGD0UQgCAAAAQBNRCAIAAABAEyk3OgCcbZr0tpwAAADAGKEQhBFWKBQaHQEAAADglCwZBgAAAIAmohAEAAAAgCaiEAQAAACAJqIQhBHU3j7pOT8HAAAAGC1sKgIjqFgs5vDWf0mSTHrVGxqcBgAAAODZzBAEAAAAgCaiEATOerVaLbVardExeJoTS+otrQcAADjzFILAWa9QKKRQKDQ6Bk9TLBaza/cDKRb9MQQAAHCm+ZcYAAAAADQRhSAAAAAANBGFIAAAAAA0EYUgAAAAADQRhSAAAAAANBGFIAAAAAA0EYUgAJAkeWKgv9ERAACAM0AhCABk2xNbs2rLu7Ltia2NjgIAANSZQhBGSHvbxCTJpFe9IZNe9YYhY8BQtVo1L53246nVqo2OQpJKrZJPPfKhJMmnvv1nqdQqDU7E0+0fONToCAAAnGUUgjBCiqVSnrjvtiH/FUulRseCUalQKObhf/98CgV/DI0GX9q1KYcGn0iSHDr6eDbv6m1wIk54+Inv57otn8gjT3y/0VEAADiL+JcYADSxg0f35/OP/V2OVgeSJEerA7n9sb/NwaOPNzgZlVo1H33kziTJ//r2v6RiRu2osX/gqUZHAAD4oSgEAaCJfb3vS6k+o2iq1qr5174vNSgRJ/zTrgdzYPB48XTg6JP5510PNjgRSfLI4/35vS/fkUeesAkPADB2KQQBoIld1vFzKT5j6XaxUMylHa9vUCKS5PGjT+azj30lA9VjSZKB6rF85rGv5ImjTzY4WXOr1Kr56MNfT5J89KGvm7UJAIxZCkHgrNbePuk5PweOmzxuaq6Y/ZaMK45Pkowrjs+82f85k8dNbXCy5nZf37dTqdWGjFVrtXy57zsNSkSS/NPObTlw9Pjy+gNHB/LPO7c1OBEAwIujEATOasViMTu/fUd2fvuOFIt+y4Pn8vqXzk/rDwrAyeOm5mdfemWDE/EzHa9MqVAYMlYsFPIzHa9oUCKeOHokn9v2UAaqx3fhHqhW8tltD+WJo0canAwA4IXzr2MAaHKlQim/8crfTZL8p1f+bkoFO6Q32pRxE7No9mvz6avem09f9Z6ML5bza7NfmynjJjY6WtP68p6dzzlr8749OxuUCADgxVMIAgCZM+VVWf5TH8mcKa9qdBR+4I0vvTgTW8ZlYsv4TBk3Mb/40osbHamp/UznzOectfnTnTMblAgA4MVTCAIASZIp49sbHYGnKT1ts5e3v/INQx5z5k0ZNyG/OufCjC8en0E7rljKojkXZsq4CQ1ORpK85CWtaW9vbXQMABgz/M0SAGCUqtVqqdVq+bEpMxodhSRvnDknPzLu+AY8U8aNzy/OnNPgRDzdMyZwAgCnoRAEABilCoVCClqOUaNUKObtF1yWJHn7hZeZtTmK+F4BgBem3OgAAAAwVvzY1Pbc/DO/lHPHn9PoKAAAL1pdf6z5oQ99KAsWLMiCBQvygQ98IEmyefPmLFy4MFdccUVuueWWk8du3bo1ixYtyrx587J8+fIcO3YsSbJr165cc801ufLKK3Pttdfm8OHDSZIDBw7kHe94R+bPn59rrrkmfX199bwUAAAAADgr1K0Q3Lx5c774xS/mc5/7XNavX59vfvOb2bhxY5YtW5Z169alp6cnDz74YO6+++4kyZIlS3LDDTfk9ttvT61Wy2233ZYkWblyZRYvXpze3t5cfPHFWbduXZJk7dq16erqyqZNm3L11Vdn1apV9boUAABIkjzyeH9+78t35JEn+hsdBQDgRatbIdjR0ZGlS5dm3LhxaWlpyctf/vJs3749s2fPzqxZs1Iul7Nw4cL09vZm586dOXLkSC699NIkyaJFi9Lb25vBwcFs2bIl8+bNGzKeJHfddVcWLlyYJLnqqqtyzz33ZHBwsF6XAwBAk6vUqvnow19Pknz0oa+nUqs2OBEAwItTt0Lwla985cmCb/v27dm0aVMKhUI6OjpOHtPZ2Zndu3dnz549Q8Y7Ojqye/fu7N+/P62trSmXy0PGkwx5TblcTmtra/bt21evywEAOONO7DLM6PBPO7flwNGBJMmBowP5553bGpwIAODFqfumIt/+9rfzzne+M7//+7+fUqmU7du3n3yuVqulUCikWq0O2RXsxPiJj093qt3DarVaisXh95vt7a0v7ELgReromNzoCDyN92P08Z7A8/N90nj7jzyV9dsfzkC1kiQZqFbyue0P502vuiDnTrDByGjhewUAhqeuheD999+f97znPVm2bFkWLFiQr3zlK0M2/+jr60tnZ2emT58+ZHzv3r3p7OxMW1tbDh48mEqlklKpdPL45Pjswr1792b69Ok5duxYDh8+nKlTpw47W3//oVSrfuLOyDnVX0D7+g6e4SQ83TPfF+/H6PD098V7Aqd24nvF90nj3f69f8+x6vElwotf/Zr8zbf+LZVqNT1bH84V5728wenwvQIAz3a6H5TVbcnw97///fzu7/5u1qxZkwULFiRJLrnkkmzbti2PPfZYKpVKNm7cmO7u7sycOTPjx4/P/fffnyTZsGFDuru709LSkq6urvT09CRJ1q9fn+7u7iTJ3Llzs379+iRJT09Purq60tLSUq/LAQCgif1M58yUfrBSZfGrfzxJUiwU8tOdMxsZCwDgRanbDMG//Mu/zMDAQG666aaTY295y1ty00035d3vfncGBgYyd+7cXHnllUmSNWvWZMWKFTl06FAuuuiivO1tb0uS3HjjjVm6dGk+/OEPZ8aMGbn55puTJO9973uzdOnSLFiwIJMnT86aNWvqdSlwSu1tE1MslU4+nvLT/+lZx3R0TE61Ukn/vifPZDQAYARNGTchvzrnwnxu20NJknHFUhbNuTBTxk1ocDIAgBeuUGvSO1VbMsxI6OiYnP47PvS8x7X/0n+3hKVBOjomZ+e370iSzHzlL3kfRomOjsl5+N8/nwtefoX3BE6hrX1SSj+4P3KlWs2+/sMNTkSlVs37vnJnPnbVr+a3Nn4uq197eUqFui244QWwZBgAnu10S4brvqkIAJzQ1j4xpeLxWbUXvPyKJMf/kKpUK9nXbxYtPF2pWMw/fffBJMkbf/TiBqchSUqFYt5+wWVJkrdfeJkyEAAYsxSCwFmrvW1ikuMzA58+Zvl245SKpXz1G7c+a7zrkrc2IA3AC/djU9uPf5zS3uAkAAAvnkIQOGsVS6V852ufHDL2ip+4pkFpAAAAYHSwzgEAAAAAmohCEAAAAACaiEIQAAAAAJqIewgCZ532tokplo7vZPtc9ww8sfV6tVKxwQgAAABNRyEInHWKpVL+7Z6bn/e413T/3hlIAwDUU1v7pCGf7+s/3MA0ADA2KAQBAIAxq1Qs5s7t30+SXH7+jAanAYCxQSEIw9Tedk6KpWd/y7T/0n8f1utPLFM9oVo5lv59T41INgAAAIDhUgjCMBVL5ez+zA0jdr5pv/bHI3YuAAAAgOGyyzAAAAAANBEzBIExra3tnJSeYyn3cDcMeeZS7krlWPZZyj3i2tompvSDnZ+7Lnnrcx7T0TE5lUol++z8DAAAUFcKQWBMK5XK2fKPfzBi5/upBe8fsXPxH0qlUv5l85rnPe4NP3v9GUgDAADQ3CwZBgCSJBMnjmt0BAAA4AxQCAIASZJJk8Y3OgIAAHAGWDJM02tvOyfF57gH3XMZ6Z2Bn3n/ulOpVo6lv4nua3eq+wKeykgv8x3O++JegwAAAIxVCkGaXrFUzq6PXdvoGKf10t/6cKMjnFGlUjn3fOq/NjrGaXX/xscbHWHUOlWhO9z7A9roBRiN2tonpVT8j8U1J36vqlSr2dd/uFGxAF6QowO1jBtfaHQMYBRQCAIwokqlcv7h80tG7Hy/fMX/N2LnAnixSsViPvqN+581/vZLfrIBaaBx2tompVQa2TtPVSrV7NunWK+3g49X8/DXBnPBT7Rk8lR3D4NmpxAEAABgWEqlYj6yuW9Ez/mun+0Y0fPxbLVqLdu+NZgk2f6twVz8M+NSKJopCM1MIciY037uOSmWR/Z/3bGwJHe49xscjuqxY+nfbwkmAAA0gz3fq+TY0eOfDx49/njaj6oDoJn5HYAxp1gu57sf/PVGxxjTfvQ9f9/oCAAAjCLntk1KeZhLgesxo284P/w+Vqlmv6XFL9jgQC07H62kWj3+uFpNdj5aSdv0UlrGmSUIzUohCAAA0OTKpWL+389/v9ExTusPr5jR6AhjUv/uSmq145+/4qKJ+c43n0wtyb7dlUybpRKAZuW7H/ihnWpX2R/GWNjFdySXcY+FnXRfyPs80huBDOfXeiz8GsJwtbVPSpK88UcvHjJmN1sAXqj26aXserSSX/zVtrS0FHP+BRNy5/p9aZtWanQ0oIEUgsAPrVQq5/a/fFOjY4xp8/6fnkZHeF6lUjl/89m3NjrGKS1edGujI8CIKRWL+eTWLw0Zu+ZVr29QGhi9TpTnl58/Y8iY8hz+Q8u4Qma+rJSWlmL27RlMW2fL8ceWC0NTUwgCAABjUqlYzN9969EhY2959csalAZGr87zSqd9DDSf4d01FgA469VqtbT/YLYNAHD2KBQLp31M41SfrDY6Ak1KIQgAJEkKhUKKRX81AAA4Eyp7Knnqc0+lsqfS6Cg0IUuGAQCAUenctkkpl07/g4rnWiJ8qs2ojlWq2b/P/QVpTrVaLW2dLamd2HKYhqpVaxm4dyBJMnDvQM5ZeI6Zm5xRCkGAs1A9dn5ORv/GHSO583PSfDsX+wcCMNqUS8Usv/trI3a+VXN/YsTOBWNNoVDIzm1HMnPOhEZHIcngw4OpPXX87161p2oZfGQw4y4c1+BUNBOFIMBZqFQq589vndfoGGPeO996e6MjnFGFgp9KwzM9fYba2y/5yec8pqNjsplnL9BwZv6dMNIl3nB/eOQ9Beql+lQ1gw8MJidWCleSwW8MpmV2Swrn+PsYZ4ZCkDFl27ZHs2XLzuz99uONjjKm/ex3vpMpU6Y1OgYAjHrlUjE3ffkLz3vc0p/5+TOQ5uxRLhXz2z1fanSM0/pfb3p9oyMAZ5n2ttYUSz8o/P77lFMeV63U0r/v0BlKRbNSCAIAaXva7sJt7ZOyr9+sGIDRatu2R/Poo98Z0XP+yI+ck+3f2D+i5xxpPcfOzYEDI3crj5e97BWZM+fZ96A8Wwwpn37gxHLhp8+UVT7VR3vbpBSHORP6mYqlwrNmM1cr1fSbtcwIUggypsyZ87K89rWX5Lvf+atGRxnTfvQVr0hf38FGxwBGkVKxmC9+944kyc/96C81OA0Az2dS6/iURnhn+F+75NwRPV89/MiPnDNy55oycucajYqlQh647/n/zv/jPz2y92DmuGKpmN3rR65kn/Yro//7k7FFIQgAADCGnPgh+Xs+t6PRUca0D14wyw/JecFeyMy/kS7xhnsPVLMJGY66FoKHDh3KW97ylnzkIx/Jeeedl82bN2f16tUZGBjI/Pnzc9111yVJtm7dmuXLl+fw4cPp6urKypUrUy6Xs2vXrixZsiT9/f2ZM2dO1qxZk0mTJuXAgQO5/vrrs2PHjrS1tWXt2rXp6Oio56UAAAAATa5YKmbnx/saHeO0Zv5X/QjPr26F4De+8Y2sWLEi27dvT5IcOXIky5Yty6233poZM2bkne98Z+6+++7MnTs3S5YsyZ/8yZ/k0ksvzbJly3Lbbbdl8eLFWblyZRYvXpwFCxbkz/7sz7Ju3bosWbIka9euTVdXV/7iL/4i69evz6pVq7J27dp6XQoAwBlxbtvElEulJMk1r3r2hgYnZgYcq1Syf9+TZzQbAIxF7edOSrE8ssvrx0LhNtzZhMNRPVZN/34zDs82dSsEb7vtttx44435/d///STJAw88kNmzZ2fWrFlJkoULF6a3tzeveMUrcuTIkVx66aVJkkWLFuWDH/xgrr766mzZsiV/9md/dnL8N3/zN7NkyZLcdddd+eQnP5kkueqqq/LHf/zHGRwcTEtLS70uBziFEzs/f2OrGxH/MF5u52cgSblUys33b3re437vJ+efgTQAMPYVy8VsX/t/Gx1jTDv/f0xvdATqoG6F4KpVq4Y83rNnz5BlvZ2dndm9e/ezxjs6OrJ79+7s378/ra2tKZfLQ8afea5yuZzW1tbs27cv06YN/x/T7e2tL/ra4GwwUj8xeuKJSdm7d0RO1fRG8qd4jJxmeF+qtWqSoZuJtL9kUoqFkf1pOkNVqtXn3BBguGXfM//fPNX5OHOa4feLZjOa39NK9f9v784DY7r6PoB/JzNJJCQRjGiUlmhjLS/VRJGNmiSyJwia4BFL25Q+rSVPqVKqWlSpNqpae8UWIbETSxRBUbR24UEVSSyRWDLL+0eeuTUyk0zkJpPJfD//kJl7z/xm7r3nnvu7556jwZzwRqYOw6yp1JoqvY0rE38HKg33keqn0iYVUavVkEj+mfJco9FAIpEYfF3779Oe/fvpdazK2ADOyXkAtVpTpnWoamBFJA6xBlB2cnJBYGAzSG/MFaU8S9VM5JmfeZyIxxIGG5fLHZByepnOa2Et3raI725KcrkDEjKSRCtvWtdobrMKUJb6lL+/8czlPGVJ29S5Tk3IjJwowVSUKjXucKIEo7H+KmIu9U1VV533keqspP2/0hKCDRo0wO3b/wy8efv2bdSvX7/Y69nZ2ahfvz7q1KmDvLw8qFQqSKVSYXmgqHdhdnY2GjRoAKVSifz8fNSuXbuyvgoRUZWmfYz77B9sMJfXBT7KTUREFoKJNiIiy1JpCcG2bdsiKysLV65cwYsvvoi0tDRERkaiYcOGsLW1xW+//YYOHTpg/fr18PLygrW1NV5//XVs2rQJwcHBSElJgZeXFwDA29sbKSkpGD58ODZt2oTXX3+d4wcSEREREVWQ2nVqwlrk3mM/BhafOKeqEbNnUaFKjbtMuhFVKu2N8lsX7po6FLPW5cIbvEleDVVaQtDW1hbTpk3D+++/j8ePH8Pb2xv+/v4AgBkzZmD8+PF48OABWrVqhdjYWADAp59+ioSEBCQmJuKFF17A119/DQAYOXIkEhIS0LNnTzg4OGDGjBmV9TWoClArlWg8Yo2pwzBraqXS1CFQBWrSpCneeKMtrubMNnUoZk/sR7mJiMyVtdQKYWt2mjoMs5YS1c3UIRAREQkqPCGYnp4u/L9Tp07YsGFDsWWaN2+ONWuKJ3gaNmyIpUuXFnu9du3amDdvnriBktnIufNQ1PLkcgf8tfAdUcsUm+ugRCYliIiIKpihMdQSPLsatf6zvck43hkRkWlpb5RfPsdZhsvj5WYNeD1aDVVaD0EiIiKqOurUsYdUKhX+DmvxdrFl5HIHqFQq5OYWVGZoRCYjk1phVPpW0cqb4acQrSwiopLUqVMLUmnxSThf8zDusfdiM9erNMjNfSBKbERUNTEhSEREZIGkUikW/Daz1OXiOnxUCdEQERFReUilEmSm3xOtPA8/J9HKIqKqiQlBIiIiIpE417GH7Kmel6WZ1jVa1M83ZgIEpUqFO+z1SWWgHZT//omTpg7FrF1o9xIH5ScioiqDCUEiIiIikcikUgza+oOpwyjRQsUwU4dARERERCbGhCARlZtKpYRi8CZTh2HWVCrO/ExERFWTdlD++QW2pg7FrHHmeiIiqkqYECSicsvNFX/m570rB4papti8+ixio56IiIiIKlSdOjUh1TP7uT5ij/tnzDAUKpUauVV8NnW1Uo2XP2hg6jDMmlqpNnUIVAGYECSLp1Yp4Too0dRhlEjN3mNEVA7OdewgkxY/5Rs7YcizFwRKlRJ3RL4RQFSRyjK2o9gzA3NcRyIqD6nUCrvTck0dhkE+QXVMHUKpcu6Im7CUyx1wfdFtUcsUW8OBcnZeoFIxIUgWL8fIi1q53AE3104Q7XNdIj9jJW2ASqWEV59Fpg6jRHzEl8yJTCrDtIzRopWX0HW6aGURVQaZVIqBG9eZOgyDFvUMN3UIREREZGGYECSiKqcsjyDXqWMHqZ6eT89LpVKK/gi0KahUSgyL2WrqMMweE7/VX+069rAuw6zAxjCHSTuM6bVmrEKVCnfZu42IiIjIrDAhSERmTV/yTi53wMm9X5e6bhuvD6ttL82KSGrK5Q74JTlG9HLF0i9iabXdnlRxrKVS9FzHHo/lsTFcvN6nREREVZ1apUbDgXJTh1EitYpj/lHpmBAkompHrVKhjdeHRi1HRERVS0X02gSq/mO57LVJZP7KMgFIWVT1cfrErL/MYZKSHCPjq1unJqxE3B/UKrXRn01kDCYEiajayfnfRVDdOvaweuaiUq1SCe8TEVHVYy2VImjNclOHYdbSovqbOgQiiySVWmHzymxTh2HWAvrUM3UIotGXvKtbpxaspJJS11WrNMjJfVARYREJmBAkomorJ7cAdevWhJVV0Z05tVrNZCBVK4ZmD9ZH7IlAjO0NwBmJiYiIiIpok3yFpwvRwKM2rGx1exCqC9XIuctegFQ5mBAkomotJycf9erVEv5PVJ3IpDKM2hpr6jBKNEOxxNQhEBEREVUpspdl+O+Pt4Cn56+TAfYh9pDYld6DkEgMTAgSGUmtUsIl8jNRy6PKkZ3N7vZERETlUahSIyWqm6nDMGuFHOS/WsvKuoTDh6/j2Mn7pg7FrL3S4XU4ObmYOowKJ7GTwPo1axT+XgioAEgB67bWTAZSpWJCkMhIOQZms83ZPrfUdeu+Fc/ZT8nsqVRK9ItYauowDFIxyU5lpL14y/v9oqlDMWsX2lwQ7eJN2CYn/hClPEt1oZ1420TrrsgD2cvlDhiy6VdRyxTbj4Gd2X4jogpj7W4N5TklNA80RQnCV61NHRJZGCYEiYjIKLlGjgMnlztgw7bRon1uSI/pvCAjIiKiKq1Jk6Z44422kORxUpHyaNasnsW0+yRWEth2ssWj7Y9g28kWEiv2DqTKxYQgERFRJSnLJCDGMocx+oydgMQYYk5Sor14++4xe6OVR7NmzUS7eNNuk+8L7ohSnqUSc5sQEVHFkdaXwi7cDlb2VqUvTCQyJgSJykGtUqHuW/FGLUdEJJPKMGidv6nDMGsLw7eYOgQiEoFSpcaPgZ1NHUaJlBzzj4gqAZOBZCpMCBKVQ05ugfB/udwB9zJX6bzv5NGbd+jJ4qhUSoT0mC5qeUREVL3cKceYhHK5A+Yc/rPU5UZ0bMl2GFUqlUqNgD71TB2GWVMxEV/tZWTsxp496aYOw+x5e/uha1efcpXBhCCRSNQqFZw8ehd7jcjS6BtrUC53wK79M0pd1/fNUdX24k07WUL275z1ujzEnMCCqqZClQppUf1NHYZZK2T7g8gkckWefAcoakPtTssVvVyx+ATVqbZtN6LqjglBIpHk5BZALndA/uldAICaLXx1ehASEVHVU6hSYWO4eJPgWCKxk093K+Dc6VzHHjKpVPRyxaJUqXCHbQYi0kOlUsMnqI6pwzCIPfqorLp29Sl3zzYSBxOCRERElaBJk6Zo36EVAgMDTR2KWRNzUhFA/OSTXO6AQVt/ELVMsS1UDLO43hzGJtvkcgeMSt8q2ufO8FNY3G9dGZQqNUZ0bGnUckTmztheh3K5AzLT74n2uR5+Tqy/iKo5JgSJiIgqiZiJLOB/yYutsaKWKbYZiiUWdUGhVKmwUDHM1GGUSMnHSQ1SqtSY4acQtTwSX3nGHySqrlQqDTz8nEQtj4iqNyYEiUSkVqtRs4Wv8H8iKqJSqeD75iijliMygCF/qgAAIABJREFUZ3zs07zpSzTJ5Q6YdjCj1HUTPLtaVPKbiKqW3FyOUUxEZcOEIJGIcnLyIZc7CP8noiK5zyRJ5HIHnL24De5uPXgBTUREREREVMmYECQSkerWOUDewdRhEJGFUKqUmKFYYuowSqRUKU0dAlG5KFVqJHh2NWo5IiIiInPBhCCRSDRqFQoPLgBadRD+llhV3RkNiUxJo1HD3a0HNBpeQJeHsWMSyuUOmJYh3ky6CV2ns2cnWQztY8RyuQMW/P5bsffj2nbg8UBERERmhwlBIpEoz+2A5tF9aDQaABooz+2EdfMepg6LqEqSSKzw180TcHV5zdShEBEZRaVWI65t8acAVBwzmIiIiMwQE4JEItA8vAfliXWA6jHuH1qNe5mrAJktZC95QGIn3mxfRNWFWq2Gq8trnHynkihVSiR0nS5qeUSWJpdjAxMREVE1woQgkQhUVw4CmqLZUe9lrip6Ua2G6komZOwlSFQMJ92pXMY+WlyvXi0AQHY2ZyokIiIiIqrOrEwdAFF1IH3JE5A8M16glRWkL3mYJiAioueQnf2AyUAiIiIiIgvAhCCRCCR2TpC9Fg5IbYtekNpA1iaCjwsTERERERERUZXDhCCRSGSvdofEzhHA/xKEr3YzcURERERERERERMWZdUIwNTUVgYGB6NGjB5YvX27qcMjCSayksPaIAwBYe8ZBYiUtZQ0iIiIiIiIioson0Wg0GlMH8Txu3ryJvn37Ijk5GTY2NoiOjsbXX3+NZs2aGbV+Ts4DqNVm+dWpitMU3IHE3tnUYRARERERERGRBZPLHQy+Z7Y9BPfv3w9PT0/Url0b9vb2UCgU2LJli6nDImIykIiIiIiIiIiqNLNNCN66dQtyuVz4u379+rh586YJIyIiIiIiIiIiIqr6ZKYO4Hmp1WpIJBLhb41Go/N3aerWrVURYREREREREREREVVpZpsQbNCgAY4cOSL8ffv2bdSvX9/o9TmGIBERERERERERVVfVcgzBN998EwcOHEBubi4ePnyIbdu2wcvLy9RhERERERERERERVWlm20PQxcUF//73vxEbG4vCwkJERUXhtddeM3VYREREREREREREVZpEo9FY5HOzfGSYiIiIiIiIiIiqq2r5yDARERERERERERGVHROCREREREREREREFoQJQSIiIiIiIiIiIgtitpOKlJeVlcTUIRAREREREREREVU6i51UhIiIiIiIiIiIyBLxkWEiIiIiIiIiIiILwoQgERERERERERGRBWFCkIiIiIiIiIiIyIIwIUhERERERERERGRBmBAkIiIiIiIiIiKyIEwIEhERERERERERWRAmBImIiIiIiIiIiCwIE4JEREREREREREQWhAlBIiIiIiIiIiIiC2LRCcHk5GQkJCSUuMyuXbuwcOFCve/duXMHbdq0wc8//6zz+rhx43Dy5EnR4nxe3377LTp37ozQ0FCEhIQgODgYBw8efK6yZs+ejZ07d4ocYeWr7tvc1K5duwZ3d3dMmDBB5/XTp0/D3d0dycnJJa7v5+eHa9eu4erVq/j4448BACdPnsS4ceNKXC8hIaHUsivKkCFDcPPmTZN8timdO3cO7u7u2Lp1q/CaMb+FMcfgs7T7xc6dOzF79uznirc627JlCyIiIoR6fsGCBSUuL0Z9FRMTg8zMzBKX+c9//oPr16+X63OqI6VSicTERAQEBCAwMBAKhQLz5s2DRqMpdd2lS5eidevWuH37tvDazZs3MWTIkIoMudpyd3cv0/Lffvstvv32WwBAaGio8Hpubi7GjRuHt956CwEBAQgPD68WbabK1LdvX2zcuFHntYKCAri7u5e4f+fl5eG9994r9+ez/SIOd3d3hIaGIiwsDMHBwYiOjsbZs2dF/Qyg+lyXPK/nbYMZWteQFStWYMWKFeWK1ViG6gAPDw/89ddflVIP6MN9unqYNGkSQkNDERgYiNatWyM0NBShoaFYu3atUes/fc7X5+m639zITB1AVXfq1CmD76WmpsLPzw8rV67EoEGDIJFIAACff/55ZYVXqujoaLz//vsAiho1gwcPxv79+8tczsiRI8UOrcoy921uarVr10ZGRgZUKhWkUikAYNOmTahTp47RZfz111+4evUqAKBNmzZo06ZNhcQqhh9//NHUIZjE2rVr4e/vj5UrV0KhUACo+N+iW7du6NatW4V+hrm5efMmvvzySyQnJ8PZ2Rn5+fmIiYlBkyZNDP5WlVVfZWZmVlgD3ZxNmjQJ2dnZWLlyJRwdHfHgwQO89957cHBwQP/+/UtcNzk5Gd26dcPatWsxfPhwAICLi4vF1kOmtH79egDAkydPMGDAACgUCmzZsgVSqRSXLl3C4MGD0bBhQzRv3tzEkZqHyMhIpKamomfPnsJr27Ztg0KhwJw5cwyud+/ePZw+fVqUGNh+EYf22ACKbmJMmDABK1euFPUzLOm6RJ/ytMH0rWtI3759yx2rsQzVAR4eHnB1dS3x+4lZD+jDfdr8ffrppwCKbv7ExsbqbFNjlLb803W/ubHoHoJPu3LlCgYNGoTw8HD07dsXf/75Jy5cuICkpCQkJSXpzR4nJyejX79+sLGx0el5p+05kZmZiaioKERERGDo0KHo2rWrsEzXrl2xadMmAMAPP/yABQsW4ObNmxg8eDB69+4NHx8foSdMv3798OuvvwIANBoNevToIVwEhoSEICwsDHPnzi31O+bl5aFu3boAig4GPz8/4T3tXe/CwkKMHj0aYWFhCAsLw6pVqwD8cwfz2rVrCAsLw+jRoxEUFIQBAwbg7t27AIC9e/ciKioKYWFhiI+Px507dwBAb5wHDhxAREQEIiIiMGjQIOTm5hq5pcRjCdvcFGrWrIkWLVrg8OHDwmu//vor3nzzTeHvp3tl6OsxNmXKFJw6dQqTJk1CZmYmYmJiABT9zp9//jnCw8MRGBiIffv2Ffv8lJQUhIeHIzQ0FB9//DEeP35cbJnU1FQEBgaiZ8+eSEhIQGFhocFtkZycjJiYGAQHB+Prr79GQkIChg8fjoCAAKSnpwu9AlQqFb744guEh4cjJCQEixYtAgCdfWLs2LFVYt8vr8LCQqSmpuKDDz7AH3/8gf/+978A/ukh8exvZkhCQgKmTJmCvn37ws/PTzjm7t69iyFDhiA4OBgffPCBsA2f3lc2b96M3r17IyQkBP7+/jh69GgFf+uq6c6dOygsLMSjR48AFB1/06ZNQ7NmzeDn54evvvpKqM///PNPAP/UVxqNBtOnT4dCoUBgYCAWL14MQH/dWBJ9x9z8+fNx69YtDB06VDgXEPD3339jw4YNmDZtGhwdHQEAtWrVwoQJE1CvXj1kZ2dj2LBhCA4ORnh4OPbu3Suse+bMGdy7dw9DhgzBqlWroFarAeiez5+tn8g4mZmZ+Ne//oV3330XCoUCI0aMwJMnTwAACxYsQI8ePdCnTx+cOHFCWEd7Htu6dStsbW0RHx8vJJGaNm2KiRMnQqVSAQCOHz+OXr16ISQkBAMGDMCVK1cq+RtWfQEBATh69KjQpgSADRs2wNPTU9i/U1NTERoaioiICIwYMQKPHz/GlClTcOvWLeHmw6xZs9C7d28oFArExMQgOzsbAODp6Ym4uDiEhoaisLAQ48ePR58+fdCtWze8++67ePz4MdsvFdB+8fDwwLlz54TfID4+HgqFAqdPnzZ43eDn54eZM2ciIiICvXv3xu7duxEbGwtvb2+hHf30dYm+6xkA6Ny5MyZMmICwsDDExcVh8+bN6NevH/z8/HDo0KFisRo6912/fh2xsbEICgpCVFQUzpw5A6AosRYUFITg4GAkJCQgPz8fANClSxdMnjwZYWFhiIyMFJIEhuqBmJgYfPHFF4iKikJwcDD27NmDwYMHw8fHB4sWLYJarYafnx+ysrIAFCW/lixZgnfffbfMbTBD7TdA//Xa07/nsmXL0KtXLwQFBSE8PByXLl0qdfuXhaE6IDIyUmc7V2Q9oG1LlYT7tPj7dEFBAby9vfXWeRXNz88PH3zwARQKBU6cOGHwt9fW/fn5+Rg7diwiIiIQGhqKtLQ0ALp1v7lhQvB/xo4di9GjR2PdunWYPHky/v3vf6NZs2aIjo5GdHQ0IiMjdZY/c+YMsrOz8frrryMgIMDgXYLLly9j8eLFmD9/Pl544QWcO3cOFy9ehEqlEg7cjIwM+Pr6Ii0tDUFBQVi1ahVSU1OxePFi5ObmIjIyUshKHzlyBI0bN4ZSqcTevXuxYcMGrFixAhcuXNB7ECUlJSE0NBQBAQEYOHAgBgwYUOLvcOzYMdy7dw8pKSn44YcfcOTIkWLLnDlzBoMGDUJaWhocHR2RmpqK3NxczJw5Ez/99BNSUlLQpUsXzJgxA9evX9cb5/fff4+JEyciOTkZb775ZqkXnBWhum7zqiAgIEB4FOHEiRNwd3eHtbW10euPHz8erVu3Fu7mPO3BgwdYt24dZs6ciYSEBOGiDQDOnz+PVatWISkpCevXr0fdunXx008/6ax/8+ZNfPHFF/j555+xceNGqFQq7Nmzx+C20K6zbt06fPjhhwCKehFs3rxZ56ShTZ6vW7cOa9aswc6dO4XjR7tPfPnll1Vi3y+vPXv2wNXVFU2aNEH37t31HgvP/maG/P333/jll1+QmJiIr776CgAwZ84ctGzZEqmpqejfv7/QkNNSq9VISkrCvHnzsGHDBsTFxWH+/PnifUEz0rx5c3Tr1g3du3dHVFQUpk+fDrVajZdeegkAYG9vj5SUFIwYMQJjx47VWXfLli04evQoUlNTsXr1aiQnJ+P27dt660ZDDB1zQ4cORf369TF//nw4OztX6G9gTk6cOAE3Nzc4OTnpvO7m5gaFQoHJkyfD09MTqampmDNnDj7++GNh/9f26mjdujVkMhkyMjL0foa++olKd+zYMUyYMAGbN2/GX3/9hX379uHkyZNYu3Yt1q1bh4ULF+Lvv/8utt7vv/+Ojh07Fnvd29sbrVq1wpMnT/Dhhx/ik08+wYYNGxAdHV1qvWiJatasiW7dumHLli0Ais4hWVlZ6NKli7DMN998g59//hnJyclo2LAhLl26hPHjx6N+/fr47rvvcOXKFVy6dAlJSUnYunUrXnjhBWzYsAFA0c2TIUOGYP369Th+/Disra2xcuVKbN++HXl5ecIQCGy/iNd+0Wg02LhxI/7v//5PeE37qKqLi4ve6watevXqITk5GW5ubpg/fz5+/vlnTJ8+vUzn+uzsbHh5eSElJQWPHz/Gjh078Msvv+D9998XboA9zdC5b9KkSVAoFEhLS8P777+PxMREnD17FvPmzcPSpUuRmpoKOzs7IYl2+/ZtdOrUCSkpKejYsSOWL19eaj2g0WiwZs0aKBQKTJkyBXPnzsXy5cvx3XffwcrKCmFhYcK+PHfuXNSuXRvu7u5lboMZar8Zul7TevDgAXbs2IGlS5ciLS0NPj4+WL58udHbwhjG1AFAxdYDe/bsKTFG7tMVs09v27YNPj4+sLW1Nfq3EJOXlxe2bt1qVG/wxMREtGrVCsnJyVi+fDnmzZuHq1evllj3V3V8ZBhFmd5Tp07hP//5j/BaQUFBib0a1qxZA39/f0ilUgQGBuL7779HdnY26tWrp7NckyZN4ODgAKBoZztw4ABkMhliY2OxceNG5OXlITs7G25ubnBzc8PBgwfx008/4fz58ygsLMTDhw8REBCAWbNmoaCgAOvWrUNERARcXFxga2uL6Oho+Pr6YtSoUXoPoqcfGb506RL69++PJk2awMXFRe/3euWVV5CVlYXBgwfDy8sLY8aMKbZM3bp10bJlS2H5e/fu4ffff8eNGzcQGxsLoOiC3cnJyWCc3bp1Q3x8PLp3745u3bqhc+fOJW0i0VXnbV4V+Pn54ZtvvoFarcbmzZsREBAg3AErr969ewMAWrRoAblcrjOOR2ZmJq5cuSIsU1hYKOyrWseOHUP79u3RoEEDAMD06dOF9/RtCwBo2bIlZLJ/qsvXXnutWFwHDhzA6dOnhZ6jBQUFOHv2LJo1a6azT5h63xeD9g4iAAQGBmLUqFHFHnV49jczpHPnzpBIJHj11VeFu8KHDh3CzJkzAQAdO3ZEo0aNdNaxsrLCd999h/T0dGRlZeHQoUOwsrLc+1uTJk3Cu+++i3379mHfvn3o3bu30AjVHgt+fn5ISEjQ6dFx+PBhBAQEwMbGBjY2Nli/fn2JdaO+xJ4xxxzp0g41ARQlZRMTE6FWq2FjY4Nr165hypQpAIBGjRqhbdu2+P333+Hl5YXU1FRh/NqAgAAkJSXB29u7WPn66icq3SuvvCKcF9zc3HDv3j1kZWXB29sbNWvWBAD4+/sLPTMNmTFjBjIyMvDo0SN07doVvXv3hqOjo7BdAgICMGHCBOTl5QnnBSoSERGB2bNnIzo6GqmpqQgJCdGp2319fdG3b190794dCoUCLVq0wLVr14T3X3rpJYwdOxarV69GVlYWjh8/jsaNGwvvt23bFkDReaV27dpYvnw5Ll26hMuXLwvne7Zfyt9+0Y619eTJE7i5ueGzzz4r9vmGrhu0vLy8AACurq6oX78+ZDIZXF1dcf/+fb2faYi2nIYNG6JDhw5Cmc+WU9K57/Dhw0JPO29vb3h7e2PZsmXw9fUVzot9+vTRWVf7hNArr7yCI0eO4PLlywbrgWe/b9u2bWFnZ4eGDRsKcWp7ZY4cORKpqakIDg4GUPY2mKH2W2nXGLVq1cLMmTOxceNGXL58GRkZGWjRooWRW8F4+uoAbc9rrYqsBwoKCvTGxX26YvdpYzoQVCTtPmGM/fv349GjR8JTTQUFBTh//rzQTjBHFpcQPHLkCBo1agQXFxdoNBpIpVKhIf70s+F///03ateurbeMJ0+eIC0tDTKZTOeRnOTkZAwdOlRn2Ro1agj/9/Hxwdy5c2FjY4ORI0di8+bNSE1NFe58TJs2DVevXkVQUBC6d++O/fv3Q6PRwN7eXshcHzx4EJ9++ilkMhlWr16NQ4cOYe/evYiOjsbSpUvRpEkTg9+9adOmaN++PY4fPw5/f3+dAcyVSiVkMhmcnZ2xceNG/Prrr9izZw/Cw8OLDfD69AlCIpFAo9FApVKhffv2mDdvHgDg8ePHyM/PNxjnwIED4evri127dmH69Ok4ceIE3nnnHYOxl4clb3NTqVmzJpo3b47ffvsNBw8exEcffVSsQa3RaCCRSKBUKstU9tMNA7VardPgUalUCAgIwPjx4wEUnQy1j2xpyWQynQtybYJk/vz5ercFoLtN9f2t/ezRo0ejR48eQrk1a9bE8ePHdZavzH2/IuTk5CAjIwN//PEHlixZAo1Gg/v372P79u06y+n7jfTR1idPbxNtvaL1bGMwPz8fUVFRCAkJQceOHeHu7i76nWpzsXv3bhQUFCAwMBCRkZGIjIzEqlWrsGbNGgDQOT7UarXOb/nssXDt2jU4OTnprRtr1qyJHTt2oHv37gAg1KXGHHP0j9atW+PixYt48OABatWqBX9/f/j7+wtj2jw7sYj2/Lpr1y7k5eUhPj4eQFGyICcnR2+PNWOPPdKlr23zbF0kk8l0enUBRds0KSlJ+HvUqFEYNWoUkpOTcejQIb0JRO12JV0dO3bE7du3cePGDWzYsKHY0Cjjx4/HmTNnsGfPHowePRrx8fHCBTFQNAb0Rx99hIEDB0KhUMDKykpn+2mPjZ07d2LOnDmIjY1FRESEzs1gtl/K334paawtbXmGrhu0nu6VWdLNxWePUe31jJaNjY3w/2fbEk8r6brg6fI0Gg0uXrxY7LjWaDQ6+8PTbRuNRlNqPVDa933xxRfh6uqKtWvX4s6dO9i6dSt27txZpjZYSe23nj176r3G0Lpx4wZiYmLw9ttvw8vLC/Xq1auQMftKqwOAiq0HDE3uxX264vbpbdu2IScnp0xJObE9+920nv3tgaLfdfr06WjVqhWAol6bTk5OZj10kcV1qVi7di127NgBADh79iwaNWoEBwcHvPzyy8IB8+uvvwoDe0ul0mIn/F27dsHZ2Rn79u1Deno60tPT8dlnn2HlypUlzhLYqlUrZGVl4fLly3Bzc4OHhwcSExPh6+srfO7gwYMREBCArKws3Lx5UzjYIiMjMWvWLHTt2hW2trb4888/8fbbb6Njx44YO3Ys3NzchOfwDbl//z7+/PNPtGzZEo6Ojrh79y5yc3Px5MkT4dGjnTt3YvTo0fDx8cH48eNhb2+PGzdulPq7tm3bFsePHxdi+P777/HVV18ZjLNXr17Iz8/HwIEDMXDgwAp9bNKSt7kpBQQEYObMmcLjbU9zdnbG+fPnodFo9I5zpW8baGkb5idPnsT9+/fx6quvCu95eHhg+/btyMnJgUajwcSJE4t1oW/Tpg2OHz8uzNI5depU7Ny5s8RtYQxPT0+sWrUKhYWFyM/PR79+/XD8+PFiy1Xmvl8R1q9fD09PT+zduxfp6enYtWsXhg8frnNBXF6dOnUSjs0TJ07ojHEDFD3CJJFIMHz4cGGbW+rFdY0aNTBz5kzh7rhGo8Hp06eFO/faGzrbt28v9qhqx44dsW3bNqE3SVxcHLKzsw3WjWPHjsXdu3dRUFCAq1evolGjRiUec9qEIf3D1dUVISEhGDt2rHCXXKlUYvfu3bCysoKnp6eQzL169SqOHj2Kdu3aITk5GSNHjhTOPxkZGejQoQNWr15tyq9T7XXq1ElIxj5+/LjYRTdQ1Mvm4cOHSExMRGFhIQAIj6BaWVmhadOmuHv3rjD+4KZNm+Dq6mrwBqSlCwsLQ2JiIpycnHR69SiVSvTo0QPOzs4YNmwYQkNDcfr0achkMqG9cPjwYbzxxhvo27cvXn75ZezevVtvHXTgwAEEBAQgMjISjo6OyMzM1FmO7ZeKb78Yum4oK0PXM2VV0nXB66+/LpxL9+/fj08++QRvvPEG0tPThScbVq1aBQ8PD4Pli1EPREZGYurUqXj55Zefqw1WUvuttGuMkydP4qWXXsLAgQPRpk0b7Nixo8LO74bqAKBy64Gy4j79fPv0lClTEBIS8lzfUWzG/Paenp7CzNu3bt1CSEgIbty4UWLdX9VZXA/BoUOHYsyYMVi2bBkaNGiAb775BkBRt/uJEydiwYIFsLa2xqxZsyCRSISKsV69esLAwNqJJZ4WFBSEr7/+usSDViKRoEOHDkI3fk9PT6xevVoYe2bYsGEYM2YMatSogQYNGqB169a4du0aGjdujA4dOkAikQjj2rVs2RLt2rVDUFAQ7Ozs0L59e6Fr7tOSkpKwY8cOWFlZ4fHjx+jVqxc6deoEAIiLi0NUVBQaNGggzILm5eWFbdu2oWfPnrC1tUVISIjOAMqGyOVyTJ06FR988AHUajVcXFwwffp0ODs7643Tzs4OCQkJkMlksLe3Fx6RqgiWts2rCl9fX4wbN07vrFkfffQRhg8fjnr16qFDhw7FHtV2c3NDXl4eRo8ejaioKJ33rl69ivDwcABFgwY/fXesefPmiI+Px4ABA6BWq9GiRYtiPThdXFwwbtw4DB48GGq1Gu3atUNERATs7Oz0bgtjRUdH48qVKwgPD4dSqURERAQ8PDyEcYm0Pvzww0rb9yvCunXrio0p179/fyxYsAC1atUS5TNGjBiBhIQE9OzZE02bNi32yHDz5s3RokULBAQEQCKRoEuXLvjtt99E+Wxz4+npifj4eAwfPlxIRnTt2hXvvfceUlNTcfToUaxZswZ2dnaYNm2azrpvvfUWTp06hYiICKjVasTGxqJJkyZ660YbGxuMGTMGffv2RWFhIfr16wcXFxe4uLgYPOZ8fHwwdOhQLFiwoNg2tGQTJ07EwoULERsbC5VKhfz8fHh4eODHH3+Evb09JkyYgOTkZABFg1RbWVkhMzMTU6dO1Sln0KBBmDhxovAoE4mvRYsWGDBgAKKiouDo6AhXV9diy9jY2GDJkiX45ptvEBYWBqCop4hCoUBcXBxsbGwwa9YsTJ48GQ8fPoSTkxNmzZpV2V/FbERERMDPz6/YbOgymQwjRozAv/71L9ja2qJu3brC5Dyurq6IiYnBjBkzEB8fLzxOaeg83qtXL4waNQobN26EtbU12rdvr3Pzm+2Xim+/GLpuKCsHBwe91zPPw9B1wYQJEzB+/Hj88ssvsLOzw5QpU9CsWTMMGzYMMTExKCwsRKtWrUqcTECMeqBHjx4YM2YMxo0bp/O6sW2wktpvtra2eq8x/vjjDwBFw7usWLECgYGB0Gg06NixI86fP1+m+I1lqA4AKr4eKMtx8yzu08+3T3/yySdVph1jzG8fHx+PiRMnIigoSOhd3bhxYzg4OAh1//Nsd1OSaErq3kRVgkajwblz5zB27FikpKSYOhyqBNzm+mln8yrpjhURFfHz88OSJUvw4osvmjoUIiKLxvYLlYdGo8HevXuxYsUK4ZFUInNmTvt0Xl4evLy8cOzYMVOHUiEsroegOVq8eDEWLFiA2bNnmzoUqiTc5kRERERENHXqVOzatQs//vijqUMhEoW57NPXr19Hnz59hCf2qiP2ECQiIiIiIiIiIrIgFjepCBERERERERERkSVjQpCIiIiIiIiIiMiCMCFIRERERERERERkQZgQJCIiIqIyKSwsRJcuXRAXFye8dvLkSYwYMULv8gkJCfjpp58AAKGhobh//36lxElERERE+jEhSERERERlsn37djRv3hynTp3CxYsXAQBt2rTBnDlzSl13/fr1cHR0rOgQiYiIiKgEMlMHQERERETmZcWKFQgMDETjxo2xePFifPbZZ8jMzMTkyZORlpaGhIQE3L17F1evXoWPj4/Ouu7u7jhw4AB2796N7du3w8rKCleuXEGNGjXw5Zdfws3NDXl5efj8889x7tw5FBYWolOnThgzZgxkMjZdiYiIiMTAHoJEREREZLQLFy5pB5j6AAAB9klEQVTg2LFj8Pf3R1hYGNavX487d+4UW+7Ro0fYuHEjRo8ebbCsw4cP45NPPkFaWhratm2L+fPnAwCmTp2KVq1aITk5GSkpKbhz5w4WLlxYYd+JiIiIyNLwNisRERERGW3FihXw9fWFs7MznJ2d8eKLL2LVqlVo166dznIdOnQotaxWrVqhQYMGAICWLVti+/btAIDdu3fj5MmTWLNmDYCi5CIRERERiYcJQSIiIiIySkFBAdavXw8bGxv4+fkBAB48eIBly5ahdevWOsva29uXWl6NGjWE/0skEmg0GgCAWq3G7Nmz4ebmBgC4f/8+JBKJWF+DiIiIyOLxkWEiIiIiMkpqaipq166NjIwMpKenIz09HTt27EBBQQFyc3NF+5wuXbpg0aJF0Gg0ePLkCd555x0sW7ZMtPKJiIiILB0TgkRERERklBUrVmDQoEGQSqXCa46OjoiJicGiRYtE+5xx48ahoKAAwcHBCA4Oxquvvoq4uDjRyiciIiKydBKN9tkMIiIiIiIiIiIiqvbYQ5CIiIiIiIiIiMiCMCFIRERERERERERkQZgQJCIiIiIiIiIisiBMCBIREREREREREVkQJgSJiIiIiIiIiIgsCBOCREREREREREREFoQJQSIiIiIiIiIiIgvChCAREREREREREZEF+X94Ir5cqI9gvwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1296x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# From graph we can see that Jet Airways Business have the highest Price.\n",
    "# Apart from the first Airline almost all are having similar median\n",
    "\n",
    "# Airline vs Price\n",
    "sns.catplot(y = \"Price\", x = \"Airline\", data = train_data.sort_values(\"Price\", ascending = False), kind=\"boxen\", height = 6, aspect = 3)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 252,
   "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>Airline_Air India</th>\n",
       "      <th>Airline_GoAir</th>\n",
       "      <th>Airline_IndiGo</th>\n",
       "      <th>Airline_Jet Airways</th>\n",
       "      <th>Airline_Jet Airways Business</th>\n",
       "      <th>Airline_Multiple carriers</th>\n",
       "      <th>Airline_Multiple carriers Premium economy</th>\n",
       "      <th>Airline_SpiceJet</th>\n",
       "      <th>Airline_Trujet</th>\n",
       "      <th>Airline_Vistara</th>\n",
       "      <th>Airline_Vistara Premium economy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\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",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Airline_Air India  Airline_GoAir  Airline_IndiGo  Airline_Jet Airways  \\\n",
       "0                  0              0               1                    0   \n",
       "1                  1              0               0                    0   \n",
       "2                  0              0               0                    1   \n",
       "3                  0              0               1                    0   \n",
       "4                  0              0               1                    0   \n",
       "\n",
       "   Airline_Jet Airways Business  Airline_Multiple carriers  \\\n",
       "0                             0                          0   \n",
       "1                             0                          0   \n",
       "2                             0                          0   \n",
       "3                             0                          0   \n",
       "4                             0                          0   \n",
       "\n",
       "   Airline_Multiple carriers Premium economy  Airline_SpiceJet  \\\n",
       "0                                          0                 0   \n",
       "1                                          0                 0   \n",
       "2                                          0                 0   \n",
       "3                                          0                 0   \n",
       "4                                          0                 0   \n",
       "\n",
       "   Airline_Trujet  Airline_Vistara  Airline_Vistara Premium economy  \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": 252,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# As Airline is Nominal Categorical data we will perform OneHotEncoding\n",
    "\n",
    "Airline = train_data[[\"Airline\"]]\n",
    "\n",
    "Airline = pd.get_dummies(Airline, drop_first= True)\n",
    "\n",
    "Airline.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Delhi       4536\n",
       "Kolkata     2871\n",
       "Banglore    2197\n",
       "Mumbai       697\n",
       "Chennai      381\n",
       "Name: Source, dtype: int64"
      ]
     },
     "execution_count": 253,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data[\"Source\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1QAAAEUCAYAAAAspncYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeXxU1f3/8dcsSYQkEhImBAMiCgULyJbKIiSVWkIIEU3FBVza71cpaJHamhoIBYNSEFOIW1x+Vq24omKiOEn0K7IIKhhaEQtuBIRAQzbMQraZub8/KClhCSFkcrO8n4+HjzBn7p37OTjczHvOuedaDMMwEBERERERkbNmNbsAERERERGRtkqBSkREREREpIkUqERERERERJpIgUpERERERKSJFKhERERERESaSIFKRERERESkiexmF2CWoqJyPB6tGC8iIiIiIg1zOAJP+5xGqERERERERJpIgUpERERERKSJFKhERERERESaSIFKRERERESkibwaqDIyMoiNjSU2NpaHHnoIgM2bNxMXF8eECRNYsWJF3bY7d+4kPj6e6OhokpKScLlcABw4cIDp06czceJEZs2aRUVFBQClpaXMmDGDmJgYpk+fTkFBgTe7Iv9xuKzS7BJERERERFoNrwWqyspKFi9ezMqVK8nIyODzzz9n7dq1zJs3j7S0NJxOJzt27GD9+vUAJCQksGDBArKzszEMg1WrVgGQnJzMtGnTyMrKYtCgQaSlpQGQmppKREQEmZmZTJ06lcWLF3urK/If3+8vYsFTH/D9/iKzSxERERERaRW8Fqjcbjcej4fKykpcLhcul4uAgAB69+5Nr169sNvtxMXFkZWVRV5eHlVVVQwdOhSA+Ph4srKyqK2tZevWrURHR9drB1i3bh1xcXEATJ48mQ0bNlBbW+ut7nR4bo+Hl5z/AOClzH/g9nhMrkhERERExHxeC1QBAQHMmTOHmJgYoqKiCA8P59ChQzgcjrptQkNDyc/PP6nd4XCQn59PSUkJAQEB2O32eu1AvX3sdjsBAQEUFxd7qzsd3oZtuZQdqQagrKKaDdtyTa5IRERERMR8Xrux765du3jrrbf46KOPCAwM5N5772XPnj1YLJa6bQzDwGKx4PF4Ttl+7OfxTnx8/D5Wa+PzYUhIwFn2qOM6XFaJc9PX1NS6AaipdePc9DUTxvYnKLCTydWJiIiIiJjHa4Hq448/ZvTo0YSEhABHp+v97W9/w2az1W1TUFBAaGgoYWFh9RaVKCwsJDQ0lODgYMrKynC73dhstrrt4ejoVmFhIWFhYbhcLioqKggKCmp0fUVF5Xg8RjP1tn376PPvcbvrT/Fzuz28//HXXBlxiUlViYiIiIi0DIcj8LTPeW3K34ABA9i8eTNHjhzBMAzWrl3LkCFDyM3NZe/evbjdbtasWUNkZCTh4eH4+fmRk5MDHF0dMDIyEh8fHyIiInA6nQCkp6cTGRkJQFRUFOnp6QA4nU4iIiLw8fHxVnc6tBGXhmO11h8ZtFotjLg03KSKRERERERaB4thGF4bpnnmmWdYvXo1Pj4+DB48mIULF7Jt2zaWLFlCdXU1UVFRzJ07F4vFwq5du5g/fz7l5eUMHDiQJUuW4OvrS15eHomJiRQVFdGjRw+WL19Oly5dOHz4MImJiezbt4/AwEBSUlLo2bNno2vTCNXZ+ejz73nv413U1Lrx9bERO3aARqdEREREpENoaITKq4GqNVOgOjtuj4cH/7aWosNH6BbUmaT/HY/tLK5ZExERERFpq0yZ8ifti81q5eaYYQBMjxmmMCUiIiIigkaozC6jzTlcVqmV/URERESkQ9EIlTQbhSkRERERkf9SoBIREREREWkiBSoREREREZEmUqASERERERFpIgUqERERERGRJlKgEhERERERaSIFKhERERERkSZSoBIREREREWkiBSoREREREZEmUqASERERERFpIgUqERERERGRJlKgEhERERERaSIFKhERERERkSZSoBIREREREWkiBSoREREREZEmUqASERERERFpIru3XviNN97gpZdeqnu8f/9+pkyZwlVXXcWSJUuorq4mJiaGe+65B4CdO3eSlJRERUUFERERJCcnY7fbOXDgAAkJCRQVFdGnTx9SUlLw9/entLSUe++9l3379hEcHExqaioOh8Nb3RERERERETmJ10aopk6dSkZGBhkZGaSkpBASEsIdd9zBvHnzSEtLw+l0smPHDtavXw9AQkICCxYsIDs7G8MwWLVqFQDJyclMmzaNrKwsBg0aRFpaGgCpqalERESQmZnJ1KlTWbx4sbe6IiIiIiIickotMuXv/vvv55577mHfvn307t2bXr16YbfbiYuLIysri7y8PKqqqhg6dCgA8fHxZGVlUVtby9atW4mOjq7XDrBu3Tri4uIAmDx5Mhs2bKC2trYluiMiIiIiIgK0QKDavHkzVVVVxMTEcOjQoXrT8kJDQ8nPzz+p3eFwkJ+fT0lJCQEBAdjt9nrtQL197HY7AQEBFBcXe7s7IiIiIiIidbx2DdUxr732Gr/5zW8A8Hg8WCyWuucMw8BisZy2/djP4534+Ph9rNbG58OQkICz6YaIiIiIiMhJvBqoampq2Lp1K0uXLgUgLCyMgoKCuucLCgoIDQ09qb2wsJDQ0FCCg4MpKyvD7XZjs9nqtoejo1uFhYWEhYXhcrmoqKggKCio0bUVFZXj8RjN1FMREREREWmvHI7A0z7n1Sl/X3/9NRdddBGdO3cGYMiQIeTm5rJ3717cbjdr1qwhMjKS8PBw/Pz8yMnJASAjI4PIyEh8fHyIiIjA6XQCkJ6eTmRkJABRUVGkp6cD4HQ6iYiIwMfHx5vdERERERERqcdiGIbXhmmcTicffPABK1asqGv75JNP6pZNj4qKYu7cuVgsFnbt2sX8+fMpLy9n4MCBLFmyBF9fX/Ly8khMTKSoqIgePXqwfPlyunTpwuHDh0lMTGTfvn0EBgaSkpJCz549G12bRqhERERERKQxGhqh8mqgas0UqEREREREpDFMm/InIiIiIiLSnilQiYiIiIiINJEClYiIiIiISBMpUImIiIiIiDSRApWIiIiIiEgTKVCJiIiIiIg0kQKViIiIiIhIEylQiYiIiIiINJEClYiIiIiISBMpUImIiIiIiDSRApWclcNllWaXICIiIiLSaihQSaN9v7+IBU99wPf7i8wuRURERESkVVCgkkZxezy85PwHAC9l/gO3x2NyRSIiIiIi5lOgkkbZsC2XsiPVAJRVVLNhW67JFYmIiIiImE+BSs6otLyK9z7eRU2tG4CaWjfvfbyL0ooqkysTERERETGXApWcUc6uPDweo16bx2OQszPPpIpERERERFoHBSo5oxGXhmO1WgD41S8HA2C1WhhxabiZZYmIiIiImE6BSs7ofP/ziB07AF8fG9f98jJ8fWzEjh3A+f7nmV2aiIiIiIipvBqo1q5dS3x8PDExMTz44IMAbN68mbi4OCZMmMCKFSvqtt25cyfx8fFER0eTlJSEy+UC4MCBA0yfPp2JEycya9YsKioqACgtLWXGjBnExMQwffp0CgoKvNmVDi9yeB8C/f0AON/fj8jhfUyuSERERETEfF4LVPv27WPhwoWkpaXxzjvv8K9//Yv169czb9480tLScDqd7Nixg/Xr1wOQkJDAggULyM7OxjAMVq1aBUBycjLTpk0jKyuLQYMGkZaWBkBqaioRERFkZmYydepUFi9e7K2uCGCzWrk5ZhgA02OGYbNqcFNERERExGufij/44AMmTZpEWFgYPj4+rFixgk6dOtG7d2969eqF3W4nLi6OrKws8vLyqKqqYujQoQDEx8eTlZVFbW0tW7duJTo6ul47wLp164iLiwNg8uTJbNiwgdraWm91R4BLeobU+ykiIiIi0tHZvfXCe/fuxcfHh5kzZ3Lw4EF+/vOf069fPxwOR902oaGh5Ofnc+jQoXrtDoeD/Px8SkpKCAgIwG6312sH6u1jt9sJCAiguLiY7t27N6q+kJCA5upqh+NwBJpdgoiIiIhIq+C1QOV2u/n8889ZuXIlnTt3ZtasWZx33nlYLJa6bQzDwGKx4PF4Ttl+7OfxTnx8/D7Ws5iGVlRUftJS4HJmDkcgBQVlZpchIiIiItJiGhpQ8NqUv27dujF69GiCg4M577zzuOqqq9i8eXO9xSMKCgoIDQ0lLCysXnthYSGhoaEEBwdTVlaG2+2utz0cHd0qLCwEwOVyUVFRQVBQkLe6IyIiIiIichKvBaorr7ySjz/+mNLSUtxuNxs3bmTixInk5uayd+9e3G43a9asITIykvDwcPz8/MjJyQEgIyODyMhIfHx8iIiIwOl0ApCenk5kZCQAUVFRpKenA+B0OomIiMDHx8db3RERERERETmJxTAMr817e/PNN3nhhReora3liiuuYP78+Xz22WcsWbKE6upqoqKimDt3LhaLhV27djF//nzKy8sZOHAgS5YswdfXl7y8PBITEykqKqJHjx4sX76cLl26cPjwYRITE9m3bx+BgYGkpKTQs2fPRtemKX9Noyl/IiIiItLRNDTlz6uBqjVToGoaBSoRERER6WhMuYZKRERERESkvVOgEhERERERaSIFKhERERERkSZSoJJGCwkJoINeciciIiIickpeu7GvtD9W66lvqiwiIiIi0lFphEpERERERKSJFKhERERERESaSIFKRERERESkiXQNlTSaFqQQEREREalPgUoazWLRohQiIiIiIsfTlD8RadVcFYfNLkFERETktBSoRKTVqvr39xxYlUxV/m6zSxERERE5JQUqEWmVDI+boo2vAlC88RUMj9vkikREREROpkAlIq1S2c6NeKrKAHBXllG282OTKxIRERE5mQKViLQ67iOl/LgtE8NVA4DhquHHbU7clWUmVyYiIiJSnwKViLQ6FbnbMAxPvTbD8FCxe5tJFYmIiIicmgKViLQ6/hePwGKpf3qyWKz4XzzcpIpERERETs2r96G65ZZbKC4uxm4/ephFixZRUVHBkiVLqK6uJiYmhnvuuQeAnTt3kpSUREVFBRERESQnJ2O32zlw4AAJCQkUFRXRp08fUlJS8Pf3p7S0lHvvvZd9+/YRHBxMamoqDofDm90RkRZi6xRIl+ExddP+LHZfugyfhK1ToNmliYiIiNTjtREqwzDYs2cPGRkZdf/179+fefPmkZaWhtPpZMeOHaxfvx6AhIQEFixYQHZ2NoZhsGrVKgCSk5OZNm0aWVlZDBo0iLS0NABSU1OJiIggMzOTqVOnsnjxYm91RURMEHjpuLoAZesUSOClY02uSERERORkXgtUu3cfvW/M//zP/3D11Vfz0ksvsX37dnr37k2vXr2w2+3ExcWRlZVFXl4eVVVVDB06FID4+HiysrKora1l69atREdH12sHWLduHXFxcQBMnjyZDRs2UFtb663uiEgLs1htBI+9CYDgcdOwWG0mVyQiIiJyMq8FqtLSUkaPHs0TTzzBCy+8wGuvvcaBAwfqTcsLDQ0lPz+fQ4cO1Wt3OBzk5+dTUlJCQEBA3ZTBY+1AvX3sdjsBAQEUFxd7qzsiYoLzwi7hgusXcl73i80uRUREROSUvHYN1bBhwxg2bFjd4+uuu45HH32UESNG1LUZhoHFYsHj8WCxWE5qP/bzeCc+Pn4fq7Xx+TAkJKDR20p9DoeuY5EWpPebiIiItGJeC1Sff/45tbW1jB49GjgaeMLDwykoKKjbpqCggNDQUMLCwuq1FxYWEhoaSnBwMGVlZbjdbmw2W932cHR0q7CwkLCwMFwuFxUVFQQFBTW6vqKicjweo5l62zEcC1IFBboXkIiIiIh0HA0NKHhtyl9ZWRnLli2jurqa8vJy3n77bf7whz+Qm5vL3r17cbvdrFmzhsjISMLDw/Hz8yMnJweAjIwMIiMj8fHxISIiAqfTCUB6ejqRkZEAREVFkZ6eDoDT6SQiIgIfHx9vdUdEREREROQkFsMwvDZMk5qaSnZ2Nh6Ph2nTpnHbbbfxySef1C2bHhUVxdy5c7FYLOzatYv58+dTXl7OwIEDWbJkCb6+vuTl5ZGYmEhRURE9evRg+fLldOnShcOHD5OYmMi+ffsIDAwkJSWFnj17Nro2jVCdPY1QiYiIiEhH1NAIlVcDVWumQHX2FKjEDK6Kw9j9Gz+dV0RERKS5mTLlT0TkXFX9+3sOrEqmKn+32aWIiIiInJIClYi0SobHTdHGVwEo3vgKhsdtckUiIiIiJ1OgEpFWqWznRjxVR6eXuivLKNv5sckViYiIiJxMgUpEWh33kVJ+3JaJ4aoBwHDV8OM2J+5KXb8nIiIirYsClYi0OhW52zAMT702w/BQsXubSRWJiIiInJoClYi0Ov4Xj8BiqX96slis+F883KSKRERERE5NgUpEWh1bp0C6DI/BYvcFwGL3pcvwSdg6nX7JUhEREREzKFCJSKsUeOm4ugBl6xRI4KVjTa5IRERE5GQKVCLSKlmsNoLH3gRA8LhpWKw2kysSEREROVmjApXH4+HZZ5/lvvvuo7y8nKeffhq3W/eEERHvOi/sEi64fiHndb/Y7FJERERETqlRgWrZsmV88803bN++HYCNGzeyZMkSrxYmIgJg9w8yuwQRERGR02pUoPrkk09YunQpfn5+BAQE8Nxzz7Fp0yZv1yYiIiIiItKqNSpQ2e12rNb/burr64vdbvdaUSIiIiIiIm1Bo1LRT37yE15++WXcbje7d+/mhRdeYMCAAd6uTUREREREpFVr1AhVUlISX331FUVFRUybNo0jR44wb948b9cmIiIiIiLSqlkMwzAas+G+ffvo1asX5eXl/PDDD/z0pz/1dm1eVVRUjsfTqK7LfzgcR+8JVFBQZnIlIiIiIiIt59jn4FNp1AjVypUrufPOOwEoKSlh9uzZvPHGG81TnYiIiIiISBvVqED1+uuv8+qrrwLQq1cv0tPTefHFF71amIiIiIiISGvXqEDldrsJCAioexwYGIjFYmnUAR566CESExMB2Lx5M3FxcUyYMIEVK1bUbbNz507i4+OJjo4mKSkJl8sFwIEDB5g+fToTJ05k1qxZVFRUAFBaWsqMGTOIiYlh+vTpFBQUNK63IiIiIiIizahRgeriiy8mJSWFffv2sW/fPh555BEuuuiiM+73ySef8PbbbwNQVVXFvHnzSEtLw+l0smPHDtavXw9AQkICCxYsIDs7G8MwWLVqFQDJyclMmzaNrKwsBg0aRFpaGgCpqalERESQmZnJ1KlTWbx4cVP6LiIiIiIick4aFaiSk5PZs2cP11xzDddddx179uzh/vvvb3Cfw4cPs2LFCmbOnAnA9u3b6d27N7169cJutxMXF0dWVhZ5eXlUVVUxdOhQAOLj48nKyqK2tpatW7cSHR1drx1g3bp1xMXFATB58mQ2bNhAbW1tk/4CREREREREmqpR96Hq1q0bjz/++Fm98IIFC7jnnns4ePAgAIcOHcLhcNQ9HxoaSn5+/kntDoeD/Px8SkpKCAgIqLuB8LH2E1/LbrcTEBBAcXEx3bt3P6saRUREREREzkWDgWrx4sUkJSXVjTKd6Kmnnjpl+xtvvEGPHj0YPXo0q1evBsDj8dS77sowDCwWy2nbj/083umu2zIMA6u1UYNtdUJCAs68kZxSQ8tGioiIiIh0JA0GqtGjRwPUTbtrLKfTSUFBAVOmTOHHH3/kyJEj5OXlYbPZ6rYpKCggNDSUsLCweotKFBYWEhoaSnBwMGVlZbjdbmw2W932cHR0q7CwkLCwMFwuFxUVFQQFBZ1VjboP1dnTfahEREREpCNqaEChwUA1fvx4ANLT0/n73//e6AM+//zzdX9evXo1W7ZsITk5mQkTJrB371569uzJmjVr+NWvfkV4eDh+fn7k5OQwYsQIMjIyiIyMxMfHh4iICJxOJ3FxcaSnpxMZGQlAVFQU6enpzJw5E6fTSUREBD4+Po2uT0REREREpDk06hqqsrIyjhw5QufOnZt8ID8/P5YuXcrs2bOprq4mKiqKiRMnApCSksL8+fMpLy9n4MCB3HrrrQAsXLiQxMREnnzySXr06MHy5csBmDNnDomJicTGxhIYGEhKSkqT6xKR1s1VcRi7/9mNQIuIiIi0FIthGGec9zZ9+nS+++47+vfvXy9Une4aqrZAU/7Onqb8SUur+vf3HMp8nNBJszmv+8VmlyMiIiIdVJOn/AF88803/OIXv2Ds2LGEhYU1a2EiIqdjeNwUbXwVgOKNr9Ajfi4Wq+0Me4mIiIi0rAYD1VtvvcVDDz1E7969+eGHH0hJSWHcuHEtVZuIdGBlOzfiqTo6GuquLKNs58ecPzDK5KpERERE6mswUK1cuZJ3332X7t27849//IMVK1YoUImI17mPlPLjtkwMVw0AhquGH7c58b94OLZOWrZfREREWo8z3rzp2M1yhw0bRklJidcLEhGpyN2GYXjqtRmGh4rd20yqSEREROTUGgxUJ95I9/j7SImIeIv/xSOwWKwMnfMYQ+c8BoDFYsX/4uEmVyYiIiJSX6OWTT/mxIAlIuINtk6BdBkeg82vEwAWuy9dhk/SdD8RERFpdRoMVF9//TXDh//3G+GqqiqGDx+OYRhYLBa2bdP0m47EYxhYLRaCQ/wpLqowuxxp5wIv/e/1mrZOgQReOtbEakREREROrcFA9cEHH7RUHdIGWC0WPt3+A6Muu9DsUqQDOH6J9OBx07RkuoiIiLRKDQaq8PDwlqpDROS0dFNfERERaa3OuMqfiIiIiIiInNpZLUohItKSDMMwuwQRERGRBmmESkRaLYvFotVFpUX9WFlqdgkiItLGKFCJiIgAu4v28OCHy8gt3mN2KSIi0oYoUImISIfn9rh5/Yu3AHj9n6txe9wmVyQiIm2FApWIiHR4m/Z8Snl1OQBl1eVs2vOpyRWJiEhboUAlIiIdWmlVGdlf/x817loAatw1ZH/9f5T9J2CJiIg0RIFKREQ6tH8e2I7nhBUlPYbBP/O2m1SRiIi0JQpUIiLSoQ0LH4L1hNUkrRYLQ8MvM6kiERFpS7waqB555BEmTZpEbGwszz//PACbN28mLi6OCRMmsGLFirptd+7cSXx8PNHR0SQlJeFyuQA4cOAA06dPZ+LEicyaNYuKigoASktLmTFjBjExMUyfPp2CggJvdkVERNqpQL8Aovtfha/NBwBfmw/R/a8i0C/A5MpERKQt8Fqg2rJlC59++invvPMOb731FitXrmTXrl3MmzePtLQ0nE4nO3bsYP369QAkJCSwYMECsrOzMQyDVatWAZCcnMy0adPIyspi0KBBpKWlAZCamkpERASZmZlMnTqVxYsXe6srIiLSzl1x0SgC/QIBCPQL5IqLRplckYiItBVeC1SXX345L774Ina7naKiItxuN6WlpfTu3ZtevXpht9uJi4sjKyuLvLw8qqqqGDp0KADx8fFkZWVRW1vL1q1biY6OrtcOsG7dOuLi4gCYPHkyGzZsoLa21lvdERGRdsxmtXH9kHgAbhgaj81qM7kiERFpK+zefHEfHx8effRRnnvuOSZOnMihQ4dwOBx1z4eGhpKfn39Su8PhID8/n5KSEgICArDb7fXagXr72O12AgICKC4upnv37o2qLSREUznOhcMRaHYJ0oHo/SYtweEYzE96JdPVP8jsUkREpA3xaqACuPvuu7njjjuYOXMme/bswXLchb+GYWCxWPB4PKdsP/bzeCc+Pn4fq7XxA25FReV4PMaZN5Q6x3+oLSgoM7ES6SiOvef0fpOWY6PgiN5vIiJSX0Nf7nptyt/333/Pzp07AejUqRMTJkzgs88+q7d4REFBAaGhoYSFhdVrLywsJDQ0lODgYMrKynC73fW2h6OjW4WFhQC4XC4qKioICtK3iiIiIiIi0nK8Fqj279/P/Pnzqampoaamhg8//JAbb7yR3Nxc9u7di9vtZs2aNURGRhIeHo6fnx85OTkAZGRkEBkZiY+PDxERETidTgDS09OJjIwEICoqivT0dACcTicRERH4+Ph4qzsiIiIiIiIn8dqUv6ioKLZv384111yDzWZjwoQJxMbGEhwczOzZs6muriYqKoqJEycCkJKSwvz58ykvL2fgwIHceuutACxcuJDExESefPJJevTowfLlywGYM2cOiYmJxMbGEhgYSEpKire6IiIiIiIickoWwzA65IVEuobq7DkcgXy6/QdGXXahrmmRFqFrqERERKQ1MOUaKmlfgoP9ARh12YX1HouIiIiIdGReX+VP2gebzYpz4666x5PGDTCxGhERERGR1kEjVCLSahmGQQedlSwiHUTNj4fNLkFEzpGuoZIGBQf7Y7OdPne73R6KiytasCLpSHQNlYi0Z2W7v2fXE48w4K45BF58idnliEgDGrqGSlP+pEE2m5WV7+ac9vlb4ka0YDUiIiLtg+F2s/vVlwDIfe0lBt83H4vNZnJVItIUmvInIiIi0sLyP96Aq/zo6HttWRn5mzaYXJGINJUClYiIiEgLqi0tJS/rPTw1NQB4amrIy3yP2rJSkysTkabQlD+pp2uwP/YTrpk607S+4+eUutweSnRNlYiIyGkV/SMHw+Op12Z4PBT9I4ewyCtNqkpEmkqBSuqx26w88tLGJu8/5+ZxzViNiIhI+xMyfAR5We9x/NJYFquVkGG6LlmkLdKUPxEREZEW5BN4PuETY7H6+tLv2ilYfX0Jj4nFJ/B8s0sTkSbQCJWIiIhIC+s+NpJDmzbQ95op9ImZSPHhKrNLEpEm0giViIiISAuz2Gz8/OGlWCwWbH5+WjJdpA1ToBIRERExgc1up2jnLiwWi9mliMg5UKASERERERFpIgUqERERkRYWEux/9OelA+o9FpG2R4tSiEirZRjGmTcSEWmDrDYr+zd+XPe457ixJlYjIudCgUpEWi1dVyAiIiKtnab8iYiIiIh0EBXlWqK/uXk1UD3++OPExsYSGxvLsmXLANi8eTNxcXFMmDCBFStW1G27c+dO4uPjiY6OJikpCZfLBcCBAweYPn06EydOZNasWVRUVABQWlrKjBkziImJYfr06RQUFHizKyIiIiIibdrBvIaqMCEAAB/HSURBVBJeeXY9/84rMbuUdsVrgWrz5s18/PHHvP3226Snp/PVV1+xZs0a5s2bR1paGk6nkx07drB+/XoAEhISWLBgAdnZ2RiGwapVqwBITk5m2rRpZGVlMWjQINLS0gBITU0lIiKCzMxMpk6dyuLFi73VFRERERGRNs3j8bA++0sA1r3/JR6Px+SK2g+vBSqHw0FiYiK+vr74+PhwySWXsGfPHnr37k2vXr2w2+3ExcWRlZVFXl4eVVVVDB06FID4+HiysrKora1l69atREdH12sHWLduHXFxcQBMnjyZDRs2UFtb663uiEgLO37FK61+JSIicm6++ucPVB6pAaCyooavvvjB5IraD68tStGvX7+6P+/Zs4fMzExuvvlmHA5HXXtoaCj5+fkcOnSoXrvD4SA/P5+SkhICAgKw2+312oF6+9jtdgICAiguLqZ79+6Nqi8kJOCc+yin5nAEml2CtBP5n38AQPeIX+p9JSLtns5z4i3lZVXkfPIdLpcbAJfLTc7m77h8TD8CAs4zubq2z+ur/H377bf89re/5U9/+hM2m409e/bUPWcYBhaLBY/HU281r2Ptx34e73SrfhmGgdXa+AG3oqJyPB4tyXyi5jiZFxSUNUMl0tGd+F7U+0pawo+VpXTpdL7ZZUg7FhLcGavNBpx6qXSP201R8ZGWLkvauS+37Tnpc6/HY7Bl07cMHn6ROUW1MQ19RvZqoMrJyeHuu+9m3rx5xMbGsmXLlnqLRxQUFBAaGkpYWFi99sLCQkJDQwkODqasrAy3243NZqvbHo6ObhUWFhIWFobL5aKiooKgoCBvdkdERNqx3UV7ePKTZ7lzzO30Cb7I7HKknbLabHz7dvppn+937TUtWI10FJf078Hnm7+r12axWrikfw+TKmpfvHYN1cGDB7nrrrtISUkhNjYWgCFDhpCbm8vevXtxu92sWbOGyMhIwsPD8fPzIycnB4CMjAwiIyPx8fEhIiICp9MJQHp6OpGRkQBERUWRnn70hOR0OomIiMDHx8db3RERkXbM7XHz+hdvAfD6P1fj9rhNrkhEpPl09vcjYkxf7Pajo6N2u42IMX3p7O9ncmXtg9dGqP72t79RXV3N0qVL69puvPFGli5dyuzZs6muriYqKoqJEycCkJKSwvz58ykvL2fgwIHceuutACxcuJDExESefPJJevTowfLlywGYM2cOiYmJxMbGEhgYSEpKire6IiIt6PjpMN0jflnX7nAEaiqMeM2mPZ9SWlUKwI9VpWza8ymRF19hclUiIs1n4NAL+eqLHyj7sZJO/r4MHHKh2SW1GxbDMDrkhUS6hurUHI5AHnlpY5P3n3PzOF3rIufE4Qhk/9rXT/lcz/E36P0lza60qoyla/9KrcdV1+ZjtTP3F/cS6KcFjKR5ORyBZ5zyp/OceMvBvBLWvLGFuKmXExbe1exy2pSGrqHy6o19RUREWrt/HtiO64Qpfi6Pm3/mbTepIhER7+gR3pVpt0cpTDUzr6/yJ61L167+2O0N5+g5N487p2OcLsG7XB5KSirO6bVFRJpbv26XYFB/xoKBQd9ul5hUkYiI9/hrmfRmp0DVwdjtVhY8nm3KsRf9LtqU40rrdvw1U8f0HH/Dabc/PrDrmippDt8Wfo/VYsVjeHhy+kMA3PXKXL4r/J4e5zfu3oYiItJxKVCJiKmsNhu7M55q0r4XT5nZzNVIRzQsfAjZX/8fNW4PnXyPfnNrt9oYGn6ZyZVJe3CqL43OtDS6vjgSb+rc2ZcjR2rMLqNdUaASEZEOLdAvgOj+V5G56wOOrdMU3f8qLUghzcJqs7H9mWebvP9lM25vxmpEwN/fT4GqmSlQiYhXnerb2ROdy0hTQ6vu6JtdaawrLhrFpGE/x2KxYBgGV1w0yuySRESkjVCgEhGvstps7HrpL6Yce8DN80w5rrQ9NquNTr7nsevf3zEgrC82a8NfAoiIiByjQCUiDQrp2gmr/dxOFWYGm4ZGsM7E43JRVFLZjNVIaxUc4g/AgLC+dY+Li7QqqYiInJkClYg0yGq38+WTCWaXYYrBsx42uwRpZsHBnbGdYQoqgM1qPWUYd7vdFGsaqYiIHEeBSkRE2pWuwZ2xNyI0NYXNZmtw1NPldlOiwCUi0qEoUImISKvizUDkbfYzBK4zUSATEW8KCQmoW81Umo8ClYiItCp2m40/vpFsdhmm+OvUhWaXICLtmNVqMbuEdkmBqo3p2rUzdvu5fXO76HfRzVTN2Tunb25dbkpKOuY3t8FdO2E7x4UhzkVHvpboXN6z58LtclGsBTFERERaPQWqNsZut3HX4rfNLsMUTyRda3YJprHZ7eQs080dO5IRf2r6jUBFRESk5ShQiYjISboGd8JuM+9XREee+mbWqKjL7aKkWKOiZyu4a2dsZ5g5ctmMc/tCrKH3hNvlpriDzt4QaS0UqERE5CR2m51fPz/H7DKkBb3wm0fMLqFNstltfPqXpaYdf9S8RNOOLSJHKVCJiIiIiLQTwcH+2GzWBrdpcNTT7aG4WDc2PxteDVTl5eXceOONPPXUU/Ts2ZPNmzezZMkSqquriYmJ4Z577gFg586dJCUlUVFRQUREBMnJydjtdg4cOEBCQgJFRUX06dOHlJQU/P39KS0t5d5772Xfvn0EBweTmpqKw+HwZldERERERFo9m83Ku29tbfL+cb/6WTNW0zE0HF/PwRdffMFNN93Enj17AKiqqmLevHmkpaXhdDrZsWMH69evByAhIYEFCxaQnZ2NYRisWrUKgOTkZKZNm0ZWVhaDBg0iLS0NgNTUVCIiIsjMzGTq1KksXrzYW90QERERERE5La+NUK1atYqFCxfypz/9CYDt27fTu3dvevXqBUBcXBxZWVn07duXqqoqhg4dCkB8fDyPPvooU6dOZevWrTzxxBN17TfffDMJCQmsW7eOl19+GYDJkyezaNEiamtr8fHx8VZ3REyTm7ubrVvz2L27wOxSpAV1+e47unTpbnYZIiIicgZeC1QnjhodOnSo3rS80NBQ8vPzT2p3OBzk5+dTUlJCQEAA9v/ce+dY+4mvZbfbCQgIoLi4mO7d9eFDRERERERaTostSuHxeLBY/nt3ZsMwsFgsp20/9vN4Jz4+fh+r9exmL4aEBJzV9tI6mLWcsJkcjiHAEHJ2rDa7FGlBffv2NbsE6YA64jm2PdD/N2luek+dnRYLVGFhYRQU/HfKUkFBAaGhoSe1FxYWEhoaSnBwMGVlZbjdbmw2W932cHR0q7CwkLCwMFwuFxUVFQQFBZ1VPUVF5Xg8RvN0rgV19Dd4QUGZ2SWYoqP/f++ozHq/5+buprAwj8KdB005vpjjO00zbZLWcH7uqL8b5dSa4z2p99TJGvp79dqiFCcaMmQIubm57N27F7fbzZo1a4iMjCQ8PBw/Pz9ycnIAyMjIIDIyEh8fHyIiInA6nQCkp6cTGRkJQFRUFOnp6QA4nU4iIiJ0/ZSIiIiIiLS4Fhuh8vPzY+nSpcyePZvq6mqioqKYOHEiACkpKcyfP5/y8nIGDhzIrbfeCsDChQtJTEzkySefpEePHixfvhyAOXPmkJiYSGxsLIGBgaSkpLRUNwAI6toZnzPcFd2bnki61rRjm82sbwJrXW4O60700kH06XMxl18+hFX52WaXIi2ob9+++lZaRKQJvB6o1q5dW/fn0aNH884775y0zYABA3jzzTdPag8PD2flypUntQcFBfHUU081b6Fnwcdu46Y/vWza8aXlvbpsutkliIiIiEgr1GIjVCIiIiKtUXDXTtjsTf9INGpeYjNWc/aaOnvD7XJRXFLZzNWIdDwKVCIiItKh2ex2PvrDvWaX0eKuXN6yl0xI8wgO9sdma3gZhLhf/eycjtFQSHe7PRQXV5zT67c3ClQibYDb5WLEn541uwxpQW6Xy+wSRESkFbLZrLz07HrTjn/z7VGmHbu1UqASaQPMnJLhcATy5ZMJph3fTINnPayL9EVEpNl17doZ+zkscGZ2qGnqNFOXy01JO1zkS4HqLOXm7mbr1jyK874yuxRpQbo/i4iIiDQXu93GEw87zS6jxd2VMMnsErxCgUpEGuRxuRg862GzyzCFR9PuRERE5AwUqM7SsfuzrFxXYnYp0oI68v1Zis5xumFIcGesNnPu2+Zxuykqbn9TC0Tam3NdZa85dNQFGsy6vyNolUFpPxSoRMSrzhRoHI5Admc07b5yF0+Z2WGDrkh7YrPbcd76G7PLkBY26cXnzS5BpFkoUImIqTxuNxdPmdnkfcU7XG4XL/zmEbPLkBbkcmuKq3Qc57ooRHNor9cTnYmZo6LeWhRDgaoJal1uXl023ewypAXVuvTB3VtON4IVfL4dm18nAFxVR/jqxb9wwbX3tWRpHVpJsbkrS/7xjWTTjm+mv05dqFFXkRZgt9v4S9KbZpchLWze4uu88roKVE1w2MTlHh2OQO5a/LZpxzfTE0nX6oNGB1GZ9w3/fP5Jhif8PwC+ePTuo+0Hv6NTj75mliYtwOV289epC80uwxQujbqKeN2xFZtz9203uxRpYd99N9QrqzYrULUxLpebJ5KuNbsMU7g0StRhlGw59ZcGJZ++RSeNUrV7Jee4kEjX4M7YTVoIxeV2n3P9IiLStlgMwzDMLsIMRUXleDwdr+tdu/pjt1tNObbL5aGkpMKUY0vbUnnwOwqynjip3THxLo1QSbNLWfco+eWHSJu+lM6+nTAMg9+88HvCArrzx5/PNru8di83dzeFhXn844knzS5FWtgNj6ww7R6PreEaKml553INVUPXfmmEqoM520ATHOyPzXb6AOZ2eyguVkiS5tWpR19sQd1xH86va7MHhSlMiVdcMyiOpz/9G3e+nFivfcqgySZVJCLe5o2FCVqLM312OxN9tjt7ClTSIP2DErOExc4h7+V5dY+7x95tYjXSnvXt1ofuAaHklx+qa+seEErfbn1MrKrjOHZ/R157w+xSpIV15Hs8etOpPrsdqajm9ec31rt8wu5j44Zfj6Ozv19LltcuacqfiLRaxZ++TfnODQT8NIrgkdeYXY60Y5U1VSx4/8G6x4smzKeT73kmVtSxtIYb+0rL0419W9aX2/bw+ebvcLnc2O02Iq7oy+BhF5ldVpuhKX8i0iYFj7qW8y7oT+cLf2p2KdLOdfI9j7EXjebjPZ8wrs8YhakWZvaH6o4a6BRoOpaBQy/kqy9+oOzHSjr5+zJwyIVml9RuaIRKRETkP3b++2suDetvdhnSQYSE+GO1WjEMg8LCcrPLkQ7gYF4Ja97YQtzUywkL72p2OW1KQyNUbTpQvfvuuzz55JO4XC5uu+02pk9v/M12FahEREREpKOpKK/CP0Cj8GerXU75y8/PZ8WKFaxevRpfX19uvPFGRo4cSd++WgVMRERERORUFKaanzk3JGoGmzdvZtSoUQQFBdG5c2eio6PJysoyuywREREREelA2mygOnToEA6Ho+5xaGgo+fn5DewhIiIiIiLSvNrslD+Px4PFYql7bBhGvcdnEhIS4I2yRERERESkA2mzgSosLIzPP/+87nFBQQGhoaGN3l+LUoiIiIiISGM0tChFm53yN2bMGD755BOKi4uprKzk/fffJzIy0uyyRERERESkA2mzI1Tdu3fnnnvu4dZbb6W2tpbrrruOyy67zOyyRERERESkA2nT96E6FyUlFZryJyIiIiIiZ9TQ+gsdNlCJiIiIiIicqzZ7DZWIiIiIiIjZFKhERERERESaSIFKRERERESkiRSoREREREREmkiBSkREREREpIkUqERERERERJpIgUpERERERKSJFKhERERERESaSIFKRERERESkiRSo2qnPPvuMYcOGMWXKFK6++mpiYmL4+9//3qzHeOyxx3jsscea9TWl/du/fz+DBg1iypQpTJkyhejoaObOnUthYWGD+4wfPx6AxMREVq9efdI2r776Kq+++qrX6pbW77PPPuOWW26pe1xeXs7111/P0qVLT7vP+PHj2b9/P6tXryYxMbHRx1q1ahVr1qw5p3ql7dq/fz/9+/dnwYIF9dp37txJ//79T3mOOhunO8+dzpdffklSUtI5HVNat/LycpKTk5k8eTJTpkzhlltu4auvvjrpvNeSHnnkET788ENTjt3a2M0uQLxn0KBBrFy5Ejj6DzE2NpYrrriCvn37mlyZdHShoaFkZGQAYBgGy5cv5+677+aVV15p8mvedNNNzVWetAMVFRXcfvvtXH755dx7773N/vrbtm3j8ssvb/bXlbYjKCiIjRs34na7sdlsADidToKDg1u8lsGDBzN48OAWP660DI/Hwx133MHIkSNJT0/Hbrfz6aefcscdd7Bw4ULT6pozZ45px25tFKg6iOrqamw2G4GBgWRmZvL8889TVVVFTU0Nf/nLXxg+fDi33HILgwcPJicnh+LiYubPn09UVBT//ve/uffee/nxxx/5yU9+wtatW9mwYUO91//oo49ITU3F4/HQq1cvFi1aRLdu3Rg/fjyXXXYZO3fu5JVXXmHjxo38/e9/x+PxMHDgQBYuXIifn59JfyvSGlgsFmbPns0VV1zBrl272LBhA5mZmbjdbsaOHUtCQsJJ+6xbt45XXnmFoqIiZs6cyQ033FA3Wjp79uyW7oK0MkeOHGHGjBmMGjWK3//+98Dpz1GnsnjxYoqKinj44Yd5//33TzpfVlVVsXbtWj799FMcDgfdu3fngQce4MiRIxQXFzNjxgwF/A7A39+fAQMGsHXrVkaNGgXApk2bGDNmDAD9+/fn66+/BmD16tVs2bKFpUuXMn78eGJjY9m0aRN2u50777yT5557jr1793LfffcxadIk4Oh57qWXXqK2tpZZs2YxadIkysvLmTdvHvn5+Rw6dIjRo0ezePFitmzZwuOPP173Jaq0L5999hkHDx7k7rvvxmo9Orls1KhRLFmyhIqKCoqLi7njjjv44Ycf6NOnD48++ii+vr6kp6ef8jPX2LFjiY6OJicnB5vNRmpqKr169WL8+PFcffXVfPzxx1RWVvLQQw8xaNAgtmzZwooVK6iqqqK0tJS5c+dy1VVXkZiYyOWXX058fLzJf0Pm05S/dmzHjh1MmTKFuLg4xo8fz+WXX063bt147bXXeOqpp3jnnXe4/fbbeeaZZ+r2qa2t5fXXX2fu3Lk88sgjwNEPFzExMbz77rtMnDiR/Pz8escpKipiwYIFPPHEE7z77rsMHz6cRYsW1T0fGRlJdnY2xcXFrFq1itdee42MjAxCQkL429/+1jJ/GdKq+fr60rt3b3bt2sWOHTt48803SU9PJz8/n3feeeek7WtqanjjjTd4+umnWbFihQkVS2tVWVnJb3/7W7755ht+/etfA2c+Rx3vscceIz8/n2XLlmGxWE55vhwzZgzjx4/n7rvvZty4cbzxxhvceeedvPXWW7z44ossW7asBXssZoqJiSE7OxuA7du3079/f3x8fM64X7du3Vi9ejWXXHIJzzzzDM899xwPP/xwvd/HlZWVrFq1imeffZa//OUvFBQUsG7dOi699FJef/11srOz2bp1K1999ZXX+ietw7/+9S8GDBhQF6aOiYqKIiQkhAMHDrBgwQIyMzMpLCxk8+bNfPvtt6f9zFVQUMDo0aNJT0/nZz/7GS+//HLdawYFBfHmm29y44038vTTTwPw0ksv8eCDD/L222/z4IMP1n0+lP/SCFU7duKUv9tvv51nn32WJ554grVr15Kbm8uWLVvq/QMdN24cAP369ePw4cPA0W/clixZAsAvf/lLzj///HrH2b59O5dddhk9e/YE4IYbbqj3S2HIkCHA0W9Y9u7dy/XXXw8cDW8//elPvdF1aYMsFgsvvvgixcXFdd92VVVVccEFFzBixIh62/7iF7/AYrHQr18/SkpKzChXWqkvv/ySOXPmcPHFFzN//nwef/zxM56jjtmwYQPFxcW8+eab2O1Hfz02dL48JjExkY0bN/L000/zzTffcOTIEe92UlqN8ePH1418ZmZmEhMTg9PpPON+kZGRAFxwwQWEhoZit9u54IILKC0trdvm2muvxW630717d4YOHcoXX3zB5MmT2b59Oy+88AK7d+/m8OHDer91AFartcHZPAMGDKBXr14AXHLJJZSUlLB///4GP3Md/3nv888/P2X7+++/D8DDDz/MRx99RFZWFl988QUVFRXN28F2QIGqgwgICCAmJoaPPvqIt99+m6uvvpqf/exn9O/fv943E8f+wVoslro2m82GYRinfW2Px1PvsWEYuFyuk17T7XYTExPD/PnzgaPXOLjd7nPvnLR5NTU15ObmMnLkSOLi4vjNb34DQGlpKTab7aTQdOx6hePfpyIAw4YN484776SyspJrrrmG1157DYfDUW+bE89Rx4SHh3PPPfewaNEiXnvtNSorK7nuuutOe7485ve//z3nn38+V155JZMmTdJiFR3IsWl/OTk5fPrpp/zxj3+sF6gMw8BisZz0fjt+FOtYeD/RsfMcHP096+Pjw8qVK8nOzub6669nzJgxfPPNNw3+fpb2YdCgQbzyyit176djli9fzpgxY+q9hywWC4ZhnPEz1/Gf945/D53qc+C0adMYOXIkI0eOZPTo0V65LrWt05S/DsLtdrNlyxbOO+88LBYLM2fOZOTIkXzwwQdnDDWjR4/m3XffBWD9+vX1vkGDoyNQX3zxBfv37wfg9ddfZ+TIkSe9zrHjFRUVYRgG999/f7OvPChtj8fj4bHHHmPIkCH86le/IiMjg4qKClwuF3fddVfddBqRxjj2QbVTp04sW7aMZcuWERgY2Khz1CWXXMLUqVPp1KkTL7/8Mnv27Dnt+dJms9X9edOmTdx9991cddVVddeX6suijiMmJoa//vWvDBo0qN4H265du/Ltt99iGAZr164969d97733MAyDvLw8duzYweDBg9m0aRM33HADV199NdXV1ezateukLzWl/YmIiCAkJITHH3+87tyyceNGVq9eTXFx8Sn3aa7PXIcPH2bPnj3MmTOHyMhIPvzwQ53fTkEjVO3YsWuojn071r9/fx544AHmz59PTEwMFouFsWPHkpOT0+DrJCUlcd9997Fq1SoGDBhw0pS/bt26sWjRIn73u99RW1vLBRdcwOLFi096nQEDBvC73/2O2267DY/Hw6WXXsqMGTOatc/SNhw6dIgpU6YA1L0Xli9fTpcuXdi1axfXX389brebcePGce2115KXl2dyxdIWDRkyhF//+tc88MADjTpHHXP//fdz00038dZbb3HppZee8nw5ZswYli9fTmBgILNnz2batGn4+fkxYMAAwsPD2b9/P717926proqJrrzySpKSkk5a8eyPf/wjM2fOpFu3bowYMeKspyd37tyZ+Ph4XC4XixYtIjg4mNtuu43777+fZ555hoCAAIYNG8b+/fu58MILm7NL0spYLBbS0tJYsmQJkydPxm6307VrV5555hnKyspOuU9zfeYKCgriuuuuIzY2FrvdzqhRo6iqqtJU0xNYDI0Vyxm8+OKLjBkzhr59+/LVV1/x5z//+ZzvsSEiIiIi0h5ohErOqHfv3vzhD3+ouyjygQceMLskEREREZFWQSNUIiIiIiIiTaRFKURERERERJpIgUpERERERKSJFKhERERERESaSItSiIhIu/DPf/6Tv/71rxw+fBjDMAgLC+O+++6jX79+ZpcmIiLtmBalEBGRNq+mpoZx48bx3HPPMXDgQAAyMjJYsWIFH374ITabzeQKRUSkvdIIlYiItHmVlZWUlZXVu9nk1VdfTUBAAG63mzfffJOVK1ditVrp1q0bf/7zn+nTpw+JiYn069eP//3f/wWo93j8+PFcdtllfP311/zhD3+gb9++LFiwgOLiYqxWK7NmzWLSpEnk5+ezaNEiDh48SG1tLbGxscycOdOsvwoREWlhClQiItLmdenShYSEBG6//Xa6devG8OHDGTlyJLGxseTk5PDss8/y+uuvExwczOrVq7nrrrt47733zvi6/fr1IzU1FYBrr72W6667junTp3Pw4EFuueUWIiMjSUhI4Ne//jXjx4+nurqaO+64gwv/fzv37tJIFIZh/JkgBEVBB6YQRLG2sbCwElIJMV4QDKQQEfIXiKWNiFaKWIiIhYWNjWBlpaay8YKNRSorIQYhiFiIt1gsyAq7LAxkNy7Pr5w5nOGthpdv5nR2kk6nax1bklQHPJRCkvRfmJ6e5uTkhLm5OaIoYmtri7GxMQqFAul0mjAMARgfH6dcLnNzc/PHPfv6+gC4v7+nWCwyMTEBQHt7O4eHhyQSCc7OzlhbW2N0dJRsNkupVKJYLNYuqCSprjihkiR9excXF1xeXpLP50mlUqRSKWZmZshkMhwfHzM0NPRlfbVa5fX1lSAI+PlX4peXly/rmpqaAGho+PG6DILg89719TVRFFGtVtnd3aWxsRGASqVCMpmsSU5JUv1xQiVJ+vbCMGRjY4Pz8/PPa3d3dzw+PpLP5zk4OKBSqQCwt7dHa2srXV1dtLW1cXV1BUC5XOb09PSX+zc3N9PT08P+/j4ApVKJXC7H09MTvb29bG9vA/Dw8EAul+Po6KiWcSVJdcQJlSTp2+vu7mZ9fZ3V1VVub29JJpO0tLSwtLTEwMAAb29vTE1N8f7+ThiGbG5ukkgkmJycZHZ2lsHBQTo6Oujv7//tM1ZWVpifn2dnZ4cgCFhcXCSKIpaXl1lYWGB4eJjn52cymQwjIyN/Mb0k6V/y2HRJkiRJislP/iRJkiQpJguVJEmSJMVkoZIkSZKkmCxUkiRJkhSThUqSJEmSYrJQSZIkSVJMFipJkiRJislCJUmSJEkxfQDIfu4jRlnJXAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 864x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Source vs Price\n",
    "\n",
    "sns.catplot(y = \"Price\", x = \"Source\", data = train_data.sort_values(\"Price\", ascending = False), kind=\"boxen\", height = 4, aspect = 3)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 255,
   "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>Source_Chennai</th>\n",
       "      <th>Source_Delhi</th>\n",
       "      <th>Source_Kolkata</th>\n",
       "      <th>Source_Mumbai</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",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>1</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>1</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",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Source_Chennai  Source_Delhi  Source_Kolkata  Source_Mumbai\n",
       "0               0             0               0              0\n",
       "1               0             0               1              0\n",
       "2               0             1               0              0\n",
       "3               0             0               1              0\n",
       "4               0             0               0              0"
      ]
     },
     "execution_count": 255,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# As Source is Nominal Categorical data we will perform OneHotEncoding\n",
    "\n",
    "Source = train_data[[\"Source\"]]\n",
    "\n",
    "Source = pd.get_dummies(Source, drop_first= True)\n",
    "\n",
    "Source.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Cochin       4536\n",
       "Banglore     2871\n",
       "Delhi        1265\n",
       "New Delhi     932\n",
       "Hyderabad     697\n",
       "Kolkata       381\n",
       "Name: Destination, dtype: int64"
      ]
     },
     "execution_count": 256,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data[\"Destination\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 257,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Destination vs Price\n",
    "\n",
    "sns.catplot(y = \"Price\", x = \"Destination\", data = train_data.sort_values(\"Price\", ascending = False), kind=\"boxen\", height = 4, aspect = 3)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 258,
   "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>Destination_Cochin</th>\n",
       "      <th>Destination_Delhi</th>\n",
       "      <th>Destination_Hyderabad</th>\n",
       "      <th>Destination_Kolkata</th>\n",
       "      <th>Destination_New Delhi</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",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</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>0</td>\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",
       "      <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",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Destination_Cochin  Destination_Delhi  Destination_Hyderabad  \\\n",
       "0                   0                  0                      0   \n",
       "1                   0                  0                      0   \n",
       "2                   1                  0                      0   \n",
       "3                   0                  0                      0   \n",
       "4                   0                  0                      0   \n",
       "\n",
       "   Destination_Kolkata  Destination_New Delhi  \n",
       "0                    0                      1  \n",
       "1                    0                      0  \n",
       "2                    0                      0  \n",
       "3                    0                      0  \n",
       "4                    0                      1  "
      ]
     },
     "execution_count": 258,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Destination = train_data[[\"Destination\"]]\n",
    "\n",
    "Destination = pd.get_dummies(Destination, drop_first = True)\n",
    "\n",
    "Destination.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 259,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DEL → BOM → COK                2376\n",
       "BLR → DEL                      1552\n",
       "CCU → BOM → BLR                 979\n",
       "CCU → BLR                       724\n",
       "BOM → HYD                       621\n",
       "                               ... \n",
       "BLR → HBX → BOM → BHO → DEL       1\n",
       "BOM → RPR → VTZ → HYD             1\n",
       "BLR → BOM → IXC → DEL             1\n",
       "BLR → HBX → BOM → AMD → DEL       1\n",
       "BOM → COK → MAA → HYD             1\n",
       "Name: Route, Length: 128, dtype: int64"
      ]
     },
     "execution_count": 259,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data[\"Route\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 260,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Additional_Info contains almost 80% no_info\n",
    "# Route and Total_Stops are related to each other\n",
    "\n",
    "train_data.drop([\"Route\", \"Additional_Info\"], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 261,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1 stop      5625\n",
       "non-stop    3491\n",
       "2 stops     1520\n",
       "3 stops       45\n",
       "4 stops        1\n",
       "Name: Total_Stops, dtype: int64"
      ]
     },
     "execution_count": 261,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data[\"Total_Stops\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 262,
   "metadata": {},
   "outputs": [],
   "source": [
    "# As this is case of Ordinal Categorical type we perform LabelEncoder\n",
    "# Here Values are assigned with corresponding keys\n",
    "\n",
    "train_data.replace({\"non-stop\": 0, \"1 stop\": 1, \"2 stops\": 2, \"3 stops\": 3, \"4 stops\": 4}, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 263,
   "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>Airline</th>\n",
       "      <th>Source</th>\n",
       "      <th>Destination</th>\n",
       "      <th>Total_Stops</th>\n",
       "      <th>Price</th>\n",
       "      <th>Journey_Day</th>\n",
       "      <th>Journey_Month</th>\n",
       "      <th>Dep_hour</th>\n",
       "      <th>Dep_min</th>\n",
       "      <th>Arrival_hour</th>\n",
       "      <th>Arrival_min</th>\n",
       "      <th>Duration_hours</th>\n",
       "      <th>Duration_mins</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>New Delhi</td>\n",
       "      <td>0</td>\n",
       "      <td>3897</td>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>22</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Air India</td>\n",
       "      <td>Kolkata</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>2</td>\n",
       "      <td>7662</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>50</td>\n",
       "      <td>13</td>\n",
       "      <td>15</td>\n",
       "      <td>7</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Jet Airways</td>\n",
       "      <td>Delhi</td>\n",
       "      <td>Cochin</td>\n",
       "      <td>2</td>\n",
       "      <td>13882</td>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "      <td>9</td>\n",
       "      <td>25</td>\n",
       "      <td>4</td>\n",
       "      <td>25</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>Kolkata</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>1</td>\n",
       "      <td>6218</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "      <td>18</td>\n",
       "      <td>5</td>\n",
       "      <td>23</td>\n",
       "      <td>30</td>\n",
       "      <td>5</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>New Delhi</td>\n",
       "      <td>1</td>\n",
       "      <td>13302</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>16</td>\n",
       "      <td>50</td>\n",
       "      <td>21</td>\n",
       "      <td>35</td>\n",
       "      <td>4</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Airline    Source Destination  Total_Stops  Price  Journey_Day  \\\n",
       "0       IndiGo  Banglore   New Delhi            0   3897           24   \n",
       "1    Air India   Kolkata    Banglore            2   7662            1   \n",
       "2  Jet Airways     Delhi      Cochin            2  13882            9   \n",
       "3       IndiGo   Kolkata    Banglore            1   6218           12   \n",
       "4       IndiGo  Banglore   New Delhi            1  13302            1   \n",
       "\n",
       "   Journey_Month  Dep_hour  Dep_min  Arrival_hour  Arrival_min  \\\n",
       "0              3        22       20             1           10   \n",
       "1              5         5       50            13           15   \n",
       "2              6         9       25             4           25   \n",
       "3              5        18        5            23           30   \n",
       "4              3        16       50            21           35   \n",
       "\n",
       "   Duration_hours  Duration_mins  \n",
       "0               2             50  \n",
       "1               7             25  \n",
       "2              19              0  \n",
       "3               5             25  \n",
       "4               4             45  "
      ]
     },
     "execution_count": 263,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 264,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_train=pd.concat([train_data,Airline,Source,Destination],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 265,
   "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>Airline</th>\n",
       "      <th>Source</th>\n",
       "      <th>Destination</th>\n",
       "      <th>Total_Stops</th>\n",
       "      <th>Price</th>\n",
       "      <th>Journey_Day</th>\n",
       "      <th>Journey_Month</th>\n",
       "      <th>Dep_hour</th>\n",
       "      <th>Dep_min</th>\n",
       "      <th>Arrival_hour</th>\n",
       "      <th>...</th>\n",
       "      <th>Airline_Vistara Premium economy</th>\n",
       "      <th>Source_Chennai</th>\n",
       "      <th>Source_Delhi</th>\n",
       "      <th>Source_Kolkata</th>\n",
       "      <th>Source_Mumbai</th>\n",
       "      <th>Destination_Cochin</th>\n",
       "      <th>Destination_Delhi</th>\n",
       "      <th>Destination_Hyderabad</th>\n",
       "      <th>Destination_Kolkata</th>\n",
       "      <th>Destination_New Delhi</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>New Delhi</td>\n",
       "      <td>0</td>\n",
       "      <td>3897</td>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>22</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Air India</td>\n",
       "      <td>Kolkata</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>2</td>\n",
       "      <td>7662</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>50</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Jet Airways</td>\n",
       "      <td>Delhi</td>\n",
       "      <td>Cochin</td>\n",
       "      <td>2</td>\n",
       "      <td>13882</td>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "      <td>9</td>\n",
       "      <td>25</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>Kolkata</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>1</td>\n",
       "      <td>6218</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "      <td>18</td>\n",
       "      <td>5</td>\n",
       "      <td>23</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>IndiGo</td>\n",
       "      <td>Banglore</td>\n",
       "      <td>New Delhi</td>\n",
       "      <td>1</td>\n",
       "      <td>13302</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>16</td>\n",
       "      <td>50</td>\n",
       "      <td>21</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Airline    Source Destination  Total_Stops  Price  Journey_Day  \\\n",
       "0       IndiGo  Banglore   New Delhi            0   3897           24   \n",
       "1    Air India   Kolkata    Banglore            2   7662            1   \n",
       "2  Jet Airways     Delhi      Cochin            2  13882            9   \n",
       "3       IndiGo   Kolkata    Banglore            1   6218           12   \n",
       "4       IndiGo  Banglore   New Delhi            1  13302            1   \n",
       "\n",
       "   Journey_Month  Dep_hour  Dep_min  Arrival_hour  ...  \\\n",
       "0              3        22       20             1  ...   \n",
       "1              5         5       50            13  ...   \n",
       "2              6         9       25             4  ...   \n",
       "3              5        18        5            23  ...   \n",
       "4              3        16       50            21  ...   \n",
       "\n",
       "   Airline_Vistara Premium economy  Source_Chennai  Source_Delhi  \\\n",
       "0                                0               0             0   \n",
       "1                                0               0             0   \n",
       "2                                0               0             1   \n",
       "3                                0               0             0   \n",
       "4                                0               0             0   \n",
       "\n",
       "   Source_Kolkata  Source_Mumbai  Destination_Cochin  Destination_Delhi  \\\n",
       "0               0              0                   0                  0   \n",
       "1               1              0                   0                  0   \n",
       "2               0              0                   1                  0   \n",
       "3               1              0                   0                  0   \n",
       "4               0              0                   0                  0   \n",
       "\n",
       "   Destination_Hyderabad  Destination_Kolkata  Destination_New Delhi  \n",
       "0                      0                    0                      1  \n",
       "1                      0                    0                      0  \n",
       "2                      0                    0                      0  \n",
       "3                      0                    0                      0  \n",
       "4                      0                    0                      1  \n",
       "\n",
       "[5 rows x 33 columns]"
      ]
     },
     "execution_count": 265,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 266,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10682, 33)"
      ]
     },
     "execution_count": 266,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 267,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_train.drop([\"Airline\", \"Source\", \"Destination\"], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 268,
   "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>Total_Stops</th>\n",
       "      <th>Price</th>\n",
       "      <th>Journey_Day</th>\n",
       "      <th>Journey_Month</th>\n",
       "      <th>Dep_hour</th>\n",
       "      <th>Dep_min</th>\n",
       "      <th>Arrival_hour</th>\n",
       "      <th>Arrival_min</th>\n",
       "      <th>Duration_hours</th>\n",
       "      <th>Duration_mins</th>\n",
       "      <th>...</th>\n",
       "      <th>Airline_Vistara Premium economy</th>\n",
       "      <th>Source_Chennai</th>\n",
       "      <th>Source_Delhi</th>\n",
       "      <th>Source_Kolkata</th>\n",
       "      <th>Source_Mumbai</th>\n",
       "      <th>Destination_Cochin</th>\n",
       "      <th>Destination_Delhi</th>\n",
       "      <th>Destination_Hyderabad</th>\n",
       "      <th>Destination_Kolkata</th>\n",
       "      <th>Destination_New Delhi</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>3897</td>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>22</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>50</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>7662</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>50</td>\n",
       "      <td>13</td>\n",
       "      <td>15</td>\n",
       "      <td>7</td>\n",
       "      <td>25</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>13882</td>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "      <td>9</td>\n",
       "      <td>25</td>\n",
       "      <td>4</td>\n",
       "      <td>25</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</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>6218</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "      <td>18</td>\n",
       "      <td>5</td>\n",
       "      <td>23</td>\n",
       "      <td>30</td>\n",
       "      <td>5</td>\n",
       "      <td>25</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>13302</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>16</td>\n",
       "      <td>50</td>\n",
       "      <td>21</td>\n",
       "      <td>35</td>\n",
       "      <td>4</td>\n",
       "      <td>45</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 30 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Total_Stops  Price  Journey_Day  Journey_Month  Dep_hour  Dep_min  \\\n",
       "0            0   3897           24              3        22       20   \n",
       "1            2   7662            1              5         5       50   \n",
       "2            2  13882            9              6         9       25   \n",
       "3            1   6218           12              5        18        5   \n",
       "4            1  13302            1              3        16       50   \n",
       "\n",
       "   Arrival_hour  Arrival_min  Duration_hours  Duration_mins  ...  \\\n",
       "0             1           10               2             50  ...   \n",
       "1            13           15               7             25  ...   \n",
       "2             4           25              19              0  ...   \n",
       "3            23           30               5             25  ...   \n",
       "4            21           35               4             45  ...   \n",
       "\n",
       "   Airline_Vistara Premium economy  Source_Chennai  Source_Delhi  \\\n",
       "0                                0               0             0   \n",
       "1                                0               0             0   \n",
       "2                                0               0             1   \n",
       "3                                0               0             0   \n",
       "4                                0               0             0   \n",
       "\n",
       "   Source_Kolkata  Source_Mumbai  Destination_Cochin  Destination_Delhi  \\\n",
       "0               0              0                   0                  0   \n",
       "1               1              0                   0                  0   \n",
       "2               0              0                   1                  0   \n",
       "3               1              0                   0                  0   \n",
       "4               0              0                   0                  0   \n",
       "\n",
       "   Destination_Hyderabad  Destination_Kolkata  Destination_New Delhi  \n",
       "0                      0                    0                      1  \n",
       "1                      0                    0                      0  \n",
       "2                      0                    0                      0  \n",
       "3                      0                    0                      0  \n",
       "4                      0                    0                      1  \n",
       "\n",
       "[5 rows x 30 columns]"
      ]
     },
     "execution_count": 268,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 269,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10682, 30)"
      ]
     },
     "execution_count": 269,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 270,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Total_Stops', 'Price', 'Journey_Day', 'Journey_Month', 'Dep_hour',\n",
       "       'Dep_min', 'Arrival_hour', 'Arrival_min', 'Duration_hours',\n",
       "       'Duration_mins', 'Airline_Air India', 'Airline_GoAir', 'Airline_IndiGo',\n",
       "       'Airline_Jet Airways', 'Airline_Jet Airways Business',\n",
       "       'Airline_Multiple carriers',\n",
       "       'Airline_Multiple carriers Premium economy', 'Airline_SpiceJet',\n",
       "       'Airline_Trujet', 'Airline_Vistara', 'Airline_Vistara Premium economy',\n",
       "       'Source_Chennai', 'Source_Delhi', 'Source_Kolkata', 'Source_Mumbai',\n",
       "       'Destination_Cochin', 'Destination_Delhi', 'Destination_Hyderabad',\n",
       "       'Destination_Kolkata', 'Destination_New Delhi'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 270,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_train.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_col=data_train.columns.drop('Price')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 276,
   "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>Total_Stops</th>\n",
       "      <th>Journey_Day</th>\n",
       "      <th>Journey_Month</th>\n",
       "      <th>Dep_hour</th>\n",
       "      <th>Dep_min</th>\n",
       "      <th>Arrival_hour</th>\n",
       "      <th>Arrival_min</th>\n",
       "      <th>Duration_hours</th>\n",
       "      <th>Duration_mins</th>\n",
       "      <th>Airline_Air India</th>\n",
       "      <th>...</th>\n",
       "      <th>Airline_Vistara Premium economy</th>\n",
       "      <th>Source_Chennai</th>\n",
       "      <th>Source_Delhi</th>\n",
       "      <th>Source_Kolkata</th>\n",
       "      <th>Source_Mumbai</th>\n",
       "      <th>Destination_Cochin</th>\n",
       "      <th>Destination_Delhi</th>\n",
       "      <th>Destination_Hyderabad</th>\n",
       "      <th>Destination_Kolkata</th>\n",
       "      <th>Destination_New Delhi</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>22</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>50</td>\n",
       "      <td>13</td>\n",
       "      <td>15</td>\n",
       "      <td>7</td>\n",
       "      <td>25</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "      <td>9</td>\n",
       "      <td>25</td>\n",
       "      <td>4</td>\n",
       "      <td>25</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</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>12</td>\n",
       "      <td>5</td>\n",
       "      <td>18</td>\n",
       "      <td>5</td>\n",
       "      <td>23</td>\n",
       "      <td>30</td>\n",
       "      <td>5</td>\n",
       "      <td>25</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>16</td>\n",
       "      <td>50</td>\n",
       "      <td>21</td>\n",
       "      <td>35</td>\n",
       "      <td>4</td>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10678</th>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>4</td>\n",
       "      <td>19</td>\n",
       "      <td>55</td>\n",
       "      <td>22</td>\n",
       "      <td>25</td>\n",
       "      <td>2</td>\n",
       "      <td>30</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10679</th>\n",
       "      <td>0</td>\n",
       "      <td>27</td>\n",
       "      <td>4</td>\n",
       "      <td>20</td>\n",
       "      <td>45</td>\n",
       "      <td>23</td>\n",
       "      <td>20</td>\n",
       "      <td>2</td>\n",
       "      <td>35</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10680</th>\n",
       "      <td>0</td>\n",
       "      <td>27</td>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>20</td>\n",
       "      <td>11</td>\n",
       "      <td>20</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10681</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>11</td>\n",
       "      <td>30</td>\n",
       "      <td>14</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10682</th>\n",
       "      <td>2</td>\n",
       "      <td>9</td>\n",
       "      <td>5</td>\n",
       "      <td>10</td>\n",
       "      <td>55</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>8</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10682 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Total_Stops  Journey_Day  Journey_Month  Dep_hour  Dep_min  \\\n",
       "0                0           24              3        22       20   \n",
       "1                2            1              5         5       50   \n",
       "2                2            9              6         9       25   \n",
       "3                1           12              5        18        5   \n",
       "4                1            1              3        16       50   \n",
       "...            ...          ...            ...       ...      ...   \n",
       "10678            0            9              4        19       55   \n",
       "10679            0           27              4        20       45   \n",
       "10680            0           27              4         8       20   \n",
       "10681            0            1              3        11       30   \n",
       "10682            2            9              5        10       55   \n",
       "\n",
       "       Arrival_hour  Arrival_min  Duration_hours  Duration_mins  \\\n",
       "0                 1           10               2             50   \n",
       "1                13           15               7             25   \n",
       "2                 4           25              19              0   \n",
       "3                23           30               5             25   \n",
       "4                21           35               4             45   \n",
       "...             ...          ...             ...            ...   \n",
       "10678            22           25               2             30   \n",
       "10679            23           20               2             35   \n",
       "10680            11           20               3              0   \n",
       "10681            14           10               2             40   \n",
       "10682            19           15               8             20   \n",
       "\n",
       "       Airline_Air India  ...  Airline_Vistara Premium economy  \\\n",
       "0                      0  ...                                0   \n",
       "1                      1  ...                                0   \n",
       "2                      0  ...                                0   \n",
       "3                      0  ...                                0   \n",
       "4                      0  ...                                0   \n",
       "...                  ...  ...                              ...   \n",
       "10678                  0  ...                                0   \n",
       "10679                  1  ...                                0   \n",
       "10680                  0  ...                                0   \n",
       "10681                  0  ...                                0   \n",
       "10682                  1  ...                                0   \n",
       "\n",
       "       Source_Chennai  Source_Delhi  Source_Kolkata  Source_Mumbai  \\\n",
       "0                   0             0               0              0   \n",
       "1                   0             0               1              0   \n",
       "2                   0             1               0              0   \n",
       "3                   0             0               1              0   \n",
       "4                   0             0               0              0   \n",
       "...               ...           ...             ...            ...   \n",
       "10678               0             0               1              0   \n",
       "10679               0             0               1              0   \n",
       "10680               0             0               0              0   \n",
       "10681               0             0               0              0   \n",
       "10682               0             1               0              0   \n",
       "\n",
       "       Destination_Cochin  Destination_Delhi  Destination_Hyderabad  \\\n",
       "0                       0                  0                      0   \n",
       "1                       0                  0                      0   \n",
       "2                       1                  0                      0   \n",
       "3                       0                  0                      0   \n",
       "4                       0                  0                      0   \n",
       "...                   ...                ...                    ...   \n",
       "10678                   0                  0                      0   \n",
       "10679                   0                  0                      0   \n",
       "10680                   0                  1                      0   \n",
       "10681                   0                  0                      0   \n",
       "10682                   1                  0                      0   \n",
       "\n",
       "       Destination_Kolkata  Destination_New Delhi  \n",
       "0                        0                      1  \n",
       "1                        0                      0  \n",
       "2                        0                      0  \n",
       "3                        0                      0  \n",
       "4                        0                      1  \n",
       "...                    ...                    ...  \n",
       "10678                    0                      0  \n",
       "10679                    0                      0  \n",
       "10680                    0                      0  \n",
       "10681                    0                      1  \n",
       "10682                    0                      0  \n",
       "\n",
       "[10682 rows x 29 columns]"
      ]
     },
     "execution_count": 276,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X=data_train.loc[:,feature_col]\n",
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 277,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     3897\n",
       "1     7662\n",
       "2    13882\n",
       "3     6218\n",
       "4    13302\n",
       "Name: Price, dtype: int64"
      ]
     },
     "execution_count": 277,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y=data_train.loc[:,'Price']\n",
    "y.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 278,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x1296 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Finds correlation between Independent and dependent attributes\n",
    "\n",
    "plt.figure(figsize = (18,18))\n",
    "sns.heatmap(train_data.corr(), annot = True, cmap = \"RdYlGn\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 279,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ExtraTreesRegressor()"
      ]
     },
     "execution_count": 279,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Important feature using ExtraTreesRegressor\n",
    "\n",
    "from sklearn.ensemble import ExtraTreesRegressor\n",
    "selection = ExtraTreesRegressor()\n",
    "selection.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2.39698320e-01 1.44063740e-01 5.43710633e-02 2.43736101e-02\n",
      " 2.12096743e-02 2.70678604e-02 1.89802755e-02 1.02221339e-01\n",
      " 1.76791160e-02 1.01473849e-02 1.81134079e-03 1.79869145e-02\n",
      " 1.45752865e-01 6.73389504e-02 1.84417790e-02 8.77628693e-04\n",
      " 3.06281919e-03 1.14524640e-04 5.25530165e-03 8.70300497e-05\n",
      " 3.84750364e-04 7.55458508e-03 3.24879672e-03 8.90538521e-03\n",
      " 1.27758714e-02 1.53574105e-02 5.92119438e-03 3.65537334e-04\n",
      " 2.49449315e-02]\n"
     ]
    }
   ],
   "source": [
    "print(selection.feature_importances_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (12,8))\n",
    "feat_importances = pd.Series(selection.feature_importances_, index=X.columns)\n",
    "feat_importances.nlargest(20).plot(kind='barh')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor()"
      ]
     },
     "execution_count": 282,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "reg_rf = RandomForestRegressor()\n",
    "reg_rf.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 287,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 288,
   "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>Total_Stops</th>\n",
       "      <th>Journey_Day</th>\n",
       "      <th>Journey_Month</th>\n",
       "      <th>Dep_hour</th>\n",
       "      <th>Dep_min</th>\n",
       "      <th>Arrival_hour</th>\n",
       "      <th>Arrival_min</th>\n",
       "      <th>Duration_hours</th>\n",
       "      <th>Duration_mins</th>\n",
       "      <th>Airline_Air India</th>\n",
       "      <th>...</th>\n",
       "      <th>Airline_Vistara Premium economy</th>\n",
       "      <th>Source_Chennai</th>\n",
       "      <th>Source_Delhi</th>\n",
       "      <th>Source_Kolkata</th>\n",
       "      <th>Source_Mumbai</th>\n",
       "      <th>Destination_Cochin</th>\n",
       "      <th>Destination_Delhi</th>\n",
       "      <th>Destination_Hyderabad</th>\n",
       "      <th>Destination_Kolkata</th>\n",
       "      <th>Destination_New Delhi</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10005</th>\n",
       "      <td>1</td>\n",
       "      <td>27</td>\n",
       "      <td>5</td>\n",
       "      <td>8</td>\n",
       "      <td>30</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>10</td>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3684</th>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>5</td>\n",
       "      <td>11</td>\n",
       "      <td>30</td>\n",
       "      <td>12</td>\n",
       "      <td>35</td>\n",
       "      <td>25</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1034</th>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>4</td>\n",
       "      <td>15</td>\n",
       "      <td>45</td>\n",
       "      <td>22</td>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3909</th>\n",
       "      <td>1</td>\n",
       "      <td>21</td>\n",
       "      <td>3</td>\n",
       "      <td>12</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>12</td>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3088</th>\n",
       "      <td>2</td>\n",
       "      <td>24</td>\n",
       "      <td>6</td>\n",
       "      <td>17</td>\n",
       "      <td>15</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>26</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5734</th>\n",
       "      <td>1</td>\n",
       "      <td>27</td>\n",
       "      <td>3</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>25</td>\n",
       "      <td>19</td>\n",
       "      <td>25</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5191</th>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>5</td>\n",
       "      <td>14</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>45</td>\n",
       "      <td>6</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5390</th>\n",
       "      <td>1</td>\n",
       "      <td>15</td>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>12</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>860</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>3</td>\n",
       "      <td>25</td>\n",
       "      <td>2</td>\n",
       "      <td>45</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7270</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>25</td>\n",
       "      <td>15</td>\n",
       "      <td>25</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8545 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Total_Stops  Journey_Day  Journey_Month  Dep_hour  Dep_min  \\\n",
       "10005            1           27              5         8       30   \n",
       "3684             1            9              5        11       30   \n",
       "1034             1           24              4        15       45   \n",
       "3909             1           21              3        12       50   \n",
       "3088             2           24              6        17       15   \n",
       "...            ...          ...            ...       ...      ...   \n",
       "5734             1           27              3         9        0   \n",
       "5191             1            9              5        14        5   \n",
       "5390             1           15              5        12       50   \n",
       "860              0            3              3         0       40   \n",
       "7270             1            1              6        13        0   \n",
       "\n",
       "       Arrival_hour  Arrival_min  Duration_hours  Duration_mins  \\\n",
       "10005            19           15              10             45   \n",
       "3684             12           35              25              5   \n",
       "1034             22            5               6             20   \n",
       "3909              1           35              12             45   \n",
       "3088             19           15              26              0   \n",
       "...             ...          ...             ...            ...   \n",
       "5734              4           25              19             25   \n",
       "5191             20           45               6             40   \n",
       "5390              1           30              12             40   \n",
       "860               3           25               2             45   \n",
       "7270              4           25              15             25   \n",
       "\n",
       "       Airline_Air India  ...  Airline_Vistara Premium economy  \\\n",
       "10005                  0  ...                                0   \n",
       "3684                   0  ...                                0   \n",
       "1034                   0  ...                                0   \n",
       "3909                   0  ...                                0   \n",
       "3088                   1  ...                                0   \n",
       "...                  ...  ...                              ...   \n",
       "5734                   0  ...                                0   \n",
       "5191                   0  ...                                0   \n",
       "5390                   0  ...                                0   \n",
       "860                    0  ...                                0   \n",
       "7270                   0  ...                                0   \n",
       "\n",
       "       Source_Chennai  Source_Delhi  Source_Kolkata  Source_Mumbai  \\\n",
       "10005               0             1               0              0   \n",
       "3684                0             1               0              0   \n",
       "1034                0             1               0              0   \n",
       "3909                0             1               0              0   \n",
       "3088                0             1               0              0   \n",
       "...               ...           ...             ...            ...   \n",
       "5734                0             1               0              0   \n",
       "5191                0             0               1              0   \n",
       "5390                0             1               0              0   \n",
       "860                 0             0               0              0   \n",
       "7270                0             1               0              0   \n",
       "\n",
       "       Destination_Cochin  Destination_Delhi  Destination_Hyderabad  \\\n",
       "10005                   1                  0                      0   \n",
       "3684                    1                  0                      0   \n",
       "1034                    1                  0                      0   \n",
       "3909                    1                  0                      0   \n",
       "3088                    1                  0                      0   \n",
       "...                   ...                ...                    ...   \n",
       "5734                    1                  0                      0   \n",
       "5191                    0                  0                      0   \n",
       "5390                    1                  0                      0   \n",
       "860                     0                  0                      0   \n",
       "7270                    1                  0                      0   \n",
       "\n",
       "       Destination_Kolkata  Destination_New Delhi  \n",
       "10005                    0                      0  \n",
       "3684                     0                      0  \n",
       "1034                     0                      0  \n",
       "3909                     0                      0  \n",
       "3088                     0                      0  \n",
       "...                    ...                    ...  \n",
       "5734                     0                      0  \n",
       "5191                     0                      0  \n",
       "5390                     0                      0  \n",
       "860                      0                      1  \n",
       "7270                     0                      0  \n",
       "\n",
       "[8545 rows x 29 columns]"
      ]
     },
     "execution_count": 288,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 289,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor()"
      ]
     },
     "execution_count": 289,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "reg_rf = RandomForestRegressor()\n",
    "reg_rf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 290,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred=reg_rf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 291,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9535125131973249"
      ]
     },
     "execution_count": 291,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reg_rf.score(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 292,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.797414733850812"
      ]
     },
     "execution_count": 292,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reg_rf.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 293,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1dc5b129c10>"
      ]
     },
     "execution_count": 293,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(y_test-y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(y_test,y_pred,alpha=0.5)\n",
    "plt.xlabel(\"y_test\")\n",
    "plt.ylabel(\"y_pred\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 295,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 296,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAE: 1175.8928950768336\n",
      "MSE: 4368154.919154097\n",
      "RMSE: 2090.013138512315\n"
     ]
    }
   ],
   "source": [
    "print('MAE:', metrics.mean_absolute_error(y_test, y_pred))\n",
    "print('MSE:', metrics.mean_squared_error(y_test, y_pred))\n",
    "print('RMSE:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 297,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.797414733850812"
      ]
     },
     "execution_count": 297,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.r2_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 303,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from scipy.stats import chi2_contingency \n",
    "from sklearn import preprocessing \n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.model_selection import GridSearchCV, cross_val_score\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "from xgboost import XGBRegressor\n",
    "from sklearn.model_selection import RandomizedSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 304,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross validation score: 81.49% (+/- 3.37%)\n"
     ]
    }
   ],
   "source": [
    "#Ramdom Forest\n",
    "estimator = RandomForestRegressor()\n",
    "param_grid = { \n",
    "            \"n_estimators\"      : [100,300,500],\n",
    "            \"max_features\"      : [\"auto\", \"sqrt\", \"log2\"],\n",
    "            \"min_samples_split\" : [2,4,8],\n",
    "            \"bootstrap\": [True,False],\n",
    "            }\n",
    "rfmodel = RandomizedSearchCV(estimator, param_grid, n_jobs=-1, cv=5)\n",
    "rfmodel.fit(X_train, y_train)\n",
    "scores = cross_val_score(rfmodel, X_train, y_train, cv=5)\n",
    "print(\"Cross validation score: {0:.2%} (+/- {1:.2%})\".format(np.mean(scores), np.std(scores)*2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 305,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred=rfmodel.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 306,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8301192327585967"
      ]
     },
     "execution_count": 306,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.r2_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 307,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, estimator=RandomForestRegressor(), n_jobs=-1,\n",
       "                   param_distributions={'bootstrap': [True, False],\n",
       "                                        'max_features': ['auto', 'sqrt',\n",
       "                                                         'log2'],\n",
       "                                        'min_samples_split': [2, 4, 8],\n",
       "                                        'n_estimators': [100, 300, 500]})"
      ]
     },
     "execution_count": 307,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfmodel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 310,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\tanis\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:278: UserWarning: The total space of parameters 9 is smaller than n_iter=10. Running 9 iterations. For exhaustive searches, use GridSearchCV.\n",
      "  warnings.warn(\n",
      "C:\\Users\\tanis\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:278: UserWarning: The total space of parameters 9 is smaller than n_iter=10. Running 9 iterations. For exhaustive searches, use GridSearchCV.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GBM Cross validation score: 82.35% (+/- 3.69%)\n",
      "XGBoost Cross validation score: 82.88% (+/- 3.62%)\n"
     ]
    }
   ],
   "source": [
    "#Gradient Boosting and XGBoost\n",
    "params={'learning_rate':[0.01,0.05,0.1],'max_depth':[3,5,7],'n_estimators':[500]}\n",
    "gbm = GradientBoostingRegressor()\n",
    "xgb = XGBRegressor()\n",
    "best_gbm = RandomizedSearchCV(gbm, params, cv=5, n_jobs=-1)\n",
    "best_xgb = RandomizedSearchCV(xgb, params, cv=5, n_jobs=-1)\n",
    "best_gbm.fit(X_train, y_train)\n",
    "best_xgb.fit(X_train, y_train)\n",
    "scores = cross_val_score(best_gbm.best_estimator_,X_train, y_train, cv=5)\n",
    "print(\"GBM Cross validation score: {0:.2%} (+/- {1:.2%})\".format(np.mean(scores), np.std(scores)*2))\n",
    "scores = cross_val_score(best_xgb.best_estimator_,X_train, y_train, cv=5)\n",
    "print(\"XGBoost Cross validation score: {0:.2%} (+/- {1:.2%})\".format(np.mean(scores), np.std(scores)*2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 311,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred=best_gbm.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 312,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8510251532560799"
      ]
     },
     "execution_count": 312,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.r2_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 313,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred=best_xgb.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 314,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8636492300969105"
      ]
     },
     "execution_count": 314,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.r2_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 328,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'n_estimators': 500, 'max_depth': 5, 'learning_rate': 0.1}"
      ]
     },
     "execution_count": 328,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_xgb.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 316,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, estimator=GradientBoostingRegressor(), n_jobs=-1,\n",
       "                   param_distributions={'learning_rate': [0.01, 0.05, 0.1],\n",
       "                                        'max_depth': [3, 5, 7],\n",
       "                                        'n_estimators': [500]})"
      ]
     },
     "execution_count": 316,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_gbm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 348,
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb = XGBRegressor(n_estimators= 500, max_depth= 5, learning_rate= 0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 349,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1, 27,  5, ...,  0,  0,  0],\n",
       "       [ 1,  9,  5, ...,  0,  0,  0],\n",
       "       [ 1, 24,  4, ...,  0,  0,  0],\n",
       "       ...,\n",
       "       [ 1, 15,  5, ...,  0,  0,  0],\n",
       "       [ 0,  3,  3, ...,  0,  0,  1],\n",
       "       [ 1,  1,  6, ...,  0,  0,  0]], dtype=int64)"
      ]
     },
     "execution_count": 349,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 350,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBRegressor(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "             colsample_bynode=1, colsample_bytree=1, gamma=0, gpu_id=-1,\n",
       "             importance_type='gain', interaction_constraints='',\n",
       "             learning_rate=0.1, max_delta_step=0, max_depth=5,\n",
       "             min_child_weight=1, missing=nan, monotone_constraints='()',\n",
       "             n_estimators=500, n_jobs=4, num_parallel_tree=1, random_state=0,\n",
       "             reg_alpha=0, reg_lambda=1, scale_pos_weight=1, subsample=1,\n",
       "             tree_method='exact', validate_parameters=1, verbosity=None)"
      ]
     },
     "execution_count": 350,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb.fit(np.array(X_train), np.array(y_train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 352,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8636492300969105"
      ]
     },
     "execution_count": 352,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.r2_score(np.array(y_test),xgb.predict(np.array(X_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 353,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBRegressor(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "             colsample_bynode=1, colsample_bytree=1, gamma=0, gpu_id=-1,\n",
       "             importance_type='gain', interaction_constraints='',\n",
       "             learning_rate=0.1, max_delta_step=0, max_depth=5,\n",
       "             min_child_weight=1, missing=nan, monotone_constraints='()',\n",
       "             n_estimators=500, n_jobs=4, num_parallel_tree=1, random_state=0,\n",
       "             reg_alpha=0, reg_lambda=1, scale_pos_weight=1, subsample=1,\n",
       "             tree_method='exact', validate_parameters=1, verbosity=None)"
      ]
     },
     "execution_count": 353,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 356,
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(xgb,open('xgb_flight2.pkl','wb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 357,
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb1=pickle.load(open('xgb_flight2.pkl','rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 358,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBRegressor(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "             colsample_bynode=1, colsample_bytree=1, gamma=0, gpu_id=-1,\n",
       "             importance_type='gain', interaction_constraints='',\n",
       "             learning_rate=0.1, max_delta_step=0, max_depth=5,\n",
       "             min_child_weight=1, missing=nan, monotone_constraints='()',\n",
       "             n_estimators=500, n_jobs=4, num_parallel_tree=1, random_state=0,\n",
       "             reg_alpha=0, reg_lambda=1, scale_pos_weight=1, subsample=1,\n",
       "             tree_method='exact', validate_parameters=1, verbosity=None)"
      ]
     },
     "execution_count": 358,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 359,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "29"
      ]
     },
     "execution_count": 359,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb1.n_features_in_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 360,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.03757793, 0.0199332 , 0.0298377 , 0.00578415, 0.00741072,\n",
       "       0.01069133, 0.00663488, 0.14191565, 0.00501736, 0.01221145,\n",
       "       0.00950581, 0.10152066, 0.10046367, 0.34758794, 0.02532265,\n",
       "       0.01222642, 0.02678936, 0.00079546, 0.01112413, 0.        ,\n",
       "       0.00251365, 0.01803814, 0.01788999, 0.01557613, 0.        ,\n",
       "       0.00420649, 0.        , 0.        , 0.02942519], dtype=float32)"
      ]
     },
     "execution_count": 360,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb1.feature_importances_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 364,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6075     16655\n",
       "3544      4959\n",
       "9291      9187\n",
       "5032      3858\n",
       "2483     12898\n",
       "         ...  \n",
       "9797      7408\n",
       "9871      4622\n",
       "10063     7452\n",
       "8802      8824\n",
       "8617     14151\n",
       "Name: Price, Length: 2137, dtype: int64"
      ]
     },
     "execution_count": 364,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test"
   ]
  }
 ],
 "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}