flight_cancellation_final.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os\n",
    "from tqdm import tqdm\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "pd.set_option('display.max_columns',150)\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "flight_path = 'flight_data/'\n",
    "weather_path = 'weather_data/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "airport_codes = ['ORD', 'JFK', 'ATL','LAX','DFW','DEN','SFO','LAS','SEA','CLT',\n",
    "                 'MCO','MIA','PHX','EWR','IAH','IAH','LGA','MSP','DTW','PHL','BOS']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <center>Not require to run"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "#all_flight_files = os.listdir(flight_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['2017_2.csv',\n",
       " '2017_7.csv',\n",
       " '2018_11.csv',\n",
       " '2016_10.csv',\n",
       " '2017_4.csv',\n",
       " '2016_6.csv',\n",
       " '2016_3.csv',\n",
       " '2017_3.csv',\n",
       " '2018_4.csv',\n",
       " '2018_7.csv',\n",
       " '2017_10.csv',\n",
       " '2016_2.csv',\n",
       " '2018_8.csv',\n",
       " '2018_6.csv',\n",
       " '2017_5.csv',\n",
       " '2016_1.csv',\n",
       " '2018_9.csv',\n",
       " '2017_12.csv',\n",
       " '2017_1.csv',\n",
       " '2018_2.csv',\n",
       " '2018_3.csv',\n",
       " '2017_9.csv',\n",
       " '2018_12.csv',\n",
       " '2016_7.csv',\n",
       " '2018_1.csv',\n",
       " '2018_5.csv',\n",
       " '2016_4.csv',\n",
       " '2016_9.csv',\n",
       " '2017_6.csv',\n",
       " '2017_11.csv',\n",
       " '2016_5.csv',\n",
       " '2018_10.csv',\n",
       " '2016_11.csv',\n",
       " '2016_12.csv',\n",
       " '2017_8.csv',\n",
       " '2016_8.csv']"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#all_flight_files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "36"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#len(all_flight_files)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#all_data =[pd.read_csv(os.path.join(flight_path, all_flight_files[i])) for i in range(len(all_flight_files))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\ndf = pd.DataFrame([])\\nfor i in tqdm(range(len(all_data))):\\n    for air in airport_codes:\\n        df = pd.concat([df, all_data[i][all_data[i]['ORIGIN'] == air]])\\n\""
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''\n",
    "df = pd.DataFrame([])\n",
    "for i in tqdm(range(len(all_data))):\n",
    "    for air in airport_codes:\n",
    "        df = pd.concat([df, all_data[i][all_data[i]['ORIGIN'] == air]])\n",
    "'''        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\ndf = df[['FL_DATE',\\n    'ORIGIN','ORIGIN_CITY_NAME','ORIGIN_STATE_NM',\\n    'DEST','DEST_CITY_NAME','DEST_STATE_NM',\\n    'CANCELLED','CANCELLATION_CODE',\\n    'CRS_ELAPSED_TIME','ACTUAL_ELAPSED_TIME','AIR_TIME','DISTANCE',\\n   'CARRIER_DELAY','WEATHER_DELAY','NAS_DELAY','SECURITY_DELAY','LATE_AIRCRAFT_DELAY']]\\n\""
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''\n",
    "df = df[['FL_DATE',\n",
    "    'ORIGIN','ORIGIN_CITY_NAME','ORIGIN_STATE_NM',\n",
    "    'DEST','DEST_CITY_NAME','DEST_STATE_NM',\n",
    "    'CANCELLED','CANCELLATION_CODE',\n",
    "    'CRS_ELAPSED_TIME','ACTUAL_ELAPSED_TIME','AIR_TIME','DISTANCE',\n",
    "   'CARRIER_DELAY','WEATHER_DELAY','NAS_DELAY','SECURITY_DELAY','LATE_AIRCRAFT_DELAY']]\n",
    "'''   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#df.to_csv('features_set_top_20.csv',index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <center>Up to this"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('features_set_top_20.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "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>FL_DATE</th>\n",
       "      <th>ORIGIN</th>\n",
       "      <th>ORIGIN_CITY_NAME</th>\n",
       "      <th>ORIGIN_STATE_NM</th>\n",
       "      <th>DEST</th>\n",
       "      <th>DEST_CITY_NAME</th>\n",
       "      <th>DEST_STATE_NM</th>\n",
       "      <th>CANCELLED</th>\n",
       "      <th>CANCELLATION_CODE</th>\n",
       "      <th>CRS_ELAPSED_TIME</th>\n",
       "      <th>ACTUAL_ELAPSED_TIME</th>\n",
       "      <th>AIR_TIME</th>\n",
       "      <th>DISTANCE</th>\n",
       "      <th>CARRIER_DELAY</th>\n",
       "      <th>WEATHER_DELAY</th>\n",
       "      <th>NAS_DELAY</th>\n",
       "      <th>SECURITY_DELAY</th>\n",
       "      <th>LATE_AIRCRAFT_DELAY</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-02-01</td>\n",
       "      <td>ORD</td>\n",
       "      <td>Chicago, IL</td>\n",
       "      <td>Illinois</td>\n",
       "      <td>SFO</td>\n",
       "      <td>San Francisco, CA</td>\n",
       "      <td>California</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>298.0</td>\n",
       "      <td>336.0</td>\n",
       "      <td>280.0</td>\n",
       "      <td>1846.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-02-01</td>\n",
       "      <td>ORD</td>\n",
       "      <td>Chicago, IL</td>\n",
       "      <td>Illinois</td>\n",
       "      <td>DEN</td>\n",
       "      <td>Denver, CO</td>\n",
       "      <td>Colorado</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>170.0</td>\n",
       "      <td>186.0</td>\n",
       "      <td>148.0</td>\n",
       "      <td>888.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-02-01</td>\n",
       "      <td>ORD</td>\n",
       "      <td>Chicago, IL</td>\n",
       "      <td>Illinois</td>\n",
       "      <td>TPA</td>\n",
       "      <td>Tampa, FL</td>\n",
       "      <td>Florida</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>172.0</td>\n",
       "      <td>165.0</td>\n",
       "      <td>123.0</td>\n",
       "      <td>1012.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-02-01</td>\n",
       "      <td>ORD</td>\n",
       "      <td>Chicago, IL</td>\n",
       "      <td>Illinois</td>\n",
       "      <td>PHX</td>\n",
       "      <td>Phoenix, AZ</td>\n",
       "      <td>Arizona</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>235.0</td>\n",
       "      <td>259.0</td>\n",
       "      <td>215.0</td>\n",
       "      <td>1440.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-02-01</td>\n",
       "      <td>ORD</td>\n",
       "      <td>Chicago, IL</td>\n",
       "      <td>Illinois</td>\n",
       "      <td>SLC</td>\n",
       "      <td>Salt Lake City, UT</td>\n",
       "      <td>Utah</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>214.0</td>\n",
       "      <td>233.0</td>\n",
       "      <td>204.0</td>\n",
       "      <td>1250.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      FL_DATE ORIGIN ORIGIN_CITY_NAME ORIGIN_STATE_NM DEST  \\\n",
       "0  2017-02-01    ORD      Chicago, IL        Illinois  SFO   \n",
       "1  2017-02-01    ORD      Chicago, IL        Illinois  DEN   \n",
       "2  2017-02-01    ORD      Chicago, IL        Illinois  TPA   \n",
       "3  2017-02-01    ORD      Chicago, IL        Illinois  PHX   \n",
       "4  2017-02-01    ORD      Chicago, IL        Illinois  SLC   \n",
       "\n",
       "       DEST_CITY_NAME DEST_STATE_NM  CANCELLED CANCELLATION_CODE  \\\n",
       "0   San Francisco, CA    California        0.0               NaN   \n",
       "1          Denver, CO      Colorado        0.0               NaN   \n",
       "2           Tampa, FL       Florida        0.0               NaN   \n",
       "3         Phoenix, AZ       Arizona        0.0               NaN   \n",
       "4  Salt Lake City, UT          Utah        0.0               NaN   \n",
       "\n",
       "   CRS_ELAPSED_TIME  ACTUAL_ELAPSED_TIME  AIR_TIME  DISTANCE  CARRIER_DELAY  \\\n",
       "0             298.0                336.0     280.0    1846.0            0.0   \n",
       "1             170.0                186.0     148.0     888.0           10.0   \n",
       "2             172.0                165.0     123.0    1012.0            NaN   \n",
       "3             235.0                259.0     215.0    1440.0            0.0   \n",
       "4             214.0                233.0     204.0    1250.0            0.0   \n",
       "\n",
       "   WEATHER_DELAY  NAS_DELAY  SECURITY_DELAY  LATE_AIRCRAFT_DELAY  \n",
       "0            0.0       31.0             0.0                  0.0  \n",
       "1            0.0       16.0             0.0                  0.0  \n",
       "2            NaN        NaN             NaN                  NaN  \n",
       "3            0.0       15.0             0.0                  0.0  \n",
       "4            0.0       17.0             0.0                  0.0  "
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0    10218590\n",
       "1.0      145283\n",
       "Name: CANCELLED, dtype: int64"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['CANCELLED'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "B    80312\n",
       "A    34750\n",
       "C    30049\n",
       "D      172\n",
       "Name: CANCELLATION_CODE, dtype: int64"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['CANCELLATION_CODE'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "CANCELLATION_CODE_dict  = {\"A\":\"Carrier\",\"B\":\"Weather\",\"C\":\"National Air System\",\"D\":\"Security\"}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_cncl_reason = [CANCELLATION_CODE_dict.get(item,item)  for item in df['CANCELLATION_CODE'].value_counts().index]\n",
    "cncl_count = df['CANCELLATION_CODE'].value_counts().values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaYAAAEWCAYAAAAtuzN2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzt3Xm4HVWZ7/Hvj4QwkwQICAkQGuIACAgRothOYAhc26CCgEqigmlskFbb240jNMpV2oEWB+y0RBLbJgyCRBsIIUCjzGEKBAQCAolMgRAGmeG9f6z3kMrOPufsnHN2Th34fZ5nP7tq1apVq4Zdbw1rVykiMDMzq4s1+rsCZmZmVQ5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MSdLakkLSqP6uS1+RtKOk+ZKeljSlv+vTFUkTJC2s9D8k6V09LGtvSTf3Xe1WKPtjkv6Sy/QtbSj/u5J+kd1vlLSsMmykpCslPSXpBElrSPq1pGWSLu/rurxeSfqCpEdyHa/XB+X1eFt+vap1YMoNo+PziqRnK/2f6GbcFXZ0A5GkqyV9shdFfAX4n4hYPyKm9lW96qTZAUVEXBwRO7dpkicBn8llenubpgFARNwZEcMqSf8A3BsRG0TE14C9gHcAm0fEu9tZl2ZeizvcDEQnAn+b6/iv/V0nAEkzJX29i+FrSrpM0hJJT0i6UdJ+q7OOfWlwf1egKxGxfke3pHuBwyPi4v6r0YCzNXBBf1fitULSmsAWwIIejj8oIl7uRRW2Bm5r6L8nIp7tQV0GR8RLvajLKuuPafbA5sCgiLhjVUeUtAZARLzS57Xq3svAUcCfIuKlPGC4UNLWEfFYP9QH6MU6j4gB8QHuBfZuSFsH+CnwILAY+B6wJrAx8CzwCvB0fjYG9gSuAZ4AHqAc/Q7OstYGAhjVyfQ3AWYADwGPA2dUhh0J3A08BpwDbJbpbwZeaijnauCT2X0EMBc4GViWZeydw35A2diey/r/oJN6fZSys1oGXAyMyfQrG8bfqtV5AkZQAtoSYClwHuWovDoPx+b3k8D5wPDK8PfmsCeA+4GPV9bXvwOLcpo/BtbKYROAhZUyHgLeld1drbdrc739Nedz/yZlvRX4Qy6j+cC+lWEzs06zgaeAK4CtmyyroVl+x7QWtFj2ycBFOc67mpS7XU7zqVzm/wH8onH7AU4HXgSez3ocm+v2pez/aub7cNZjWdZr+4Zl+mVKYH0m07bM9fsocA9wRCX/d4Ff57SfynJ3yWFnUX5fz+T0j24ybxOAhcA3gIeB/2yhjt8E/pzTuxX4P5Vhbwb+mNvBEmBGZdh7gBty2NXA21vdXhu2k7/mOn4auKDFso+nbJ/P0WT/Qevb8iDgJzlvTwA3A28Cjm5Y92d1s68U8M4cZ6cmw7fK+dywkrZn1meN7P974A7K7/9/gJGVvKdQ9rdPUn5/4xq2mf8Gzsh1+Mks+8bM/xDwnW739z0JEv3xoXlg+rfcsDcBNgOuA77WbEeXabsDb88NYFvKj+aIHNZdYJoL/AoYBgwB3p3p++XC3inLmArMadyxNGzI1cD0IjAp6/RFyqWalfJ2Uqcdc+W/N+v0DeD2yobe3fidzdNmwERKIBlK2XHNbKjXHbkM16MEweNy2Hb54/ko5Yx8BLBzDvs5cHZObyglGBzbbH2x4o95ldZbtawcfh/wT5SDln2yftvk8JnAI8CuOfxs4LROltcK02qx7KXAHpTL5ms1KfMG4Du5/Pei7OhXCkyV8r5e6T8CuLjSP45ykLZbLqspwJ2V7eEhym9ki1y3g4BbgH/J6b+RciDxnspO5hngA5n3JOCyZuuok+U1gRI4j8/y12mhjgdRzlrWAA6lbN+b5LBzKYFVWdaemb4pZaf3Mco29ynKzn1od9trkzo3LvNWyr6HEkDW7JiPhjJb3ZYnAlcBG+b87wBs2mzdd7HM51ACWFB+t+ok3yXApyv9pwDfy+6DKfuRN+Y8fRu4tJJ3EjA8h32NcqC5ZmWbeZ6yX1wj19ONwIE5fANgj27no7sMdfnQPDD9BXh/pX8i5VS240exsJsyjwFOb7bTaci3DfACsEGTYb8Gjq/0D6McSb6hcSOvbMjVwHRrZdhGWYdhjXk7qf8JrHjUOCh/NOO6G7+reWqSdxzwYMM8fLnS/yXgt9n9rx3LtKGMwTm96pHX+4Dbm60vutjpdbfeWDEwfYASPFQZfi5wTHbPBH5SGfYR4KZOptsYmFope2oXy/WNlKPstStp59DzwPRL8sCsknYfuSPIZfrxyrD3AHc15P9X4JTs/i7w+8qwXYFlrayjynr4K7nTaqWOTcr4E7BPdp9JOaPYvCHPZ4HLG9JuBA7ubnttMr3GZd5K2V/t5vfT6ra8H+VsdncaAkrjuu9mekOAv6PJWWwlz2RgbiX/Y+TZFXAp8IlK3jUpB9CbNSlHlIOXN1W2mYsa8lxLCWAbt1L/iKh344euSBJl539fJfk+YGQX42wv6QJJD0t6knLZYJMWJrcl8EhEPNVk2BbVOkTEMsoRVqf1aPBQpfuZ/F6/WcYWpv0yJVi3Mu1O50nSBpKmSbo/l9NFrLycGuvdUectKZckm9V1TWBBtiJbBvyWckTapV6st47p3h/5C0mN20ln89IXZS/qZvwlEfFcw/g9tTXw1Y7lm8t4RBf12RoY3ZD/S5TfVYeeLptXx4+IF1uto6TDsiVpx7DtWL6uvwisC9yYeToaBq3wO0h9uY67K7urdbyCbrblC4BTKZdzH5b0M0mruryJiBci4nfARySN7yTbb4C3SxpJCYiLI2J+Dtsa+HllHSyhnPmOynn4iqQ7JD1BuQWwNiv+HhuXx2TKFaU7JV0jaZ/u5mHABqbcGTxEWYgdtqLsmKEc2Tb6T8qlk20jYkPKJQa1MLlFwKadbCQPVOsgaSjlVPwvlKPFQZLWquR/A61rNg9dTXsQ5Qfzl07HWK6reTqGshG+PZfTeFpbTh3lbtsk/UHKxr1tRAzLz9CI2LiFMrtab60so60a0qrbSW+0UnZX9XsQ2ETS2g3j99Qi4JuV5TssItaNiHM6qc8iyhWGav4NIuLDLU6vu2XfLE+ndZT0Rsp9xynARlFaJC4k13VE/CUiPkO51Hc0ME3SVjT8DlJfruPuym5lOXTodFuO4ocR8TbKjnxn4B97MI0Og2n+WyQinqac3X+ccsn0V5XBi4BPNayjdSLiekkfAD5PuU84jHKV51lW3D+sUNeIuD0iDqIchJ4MnCNpSFcVH7CBKZ0OHCtpY0mbUk4X/yuHPczKO94NgCci4mlJO1BO07sVEX8GLgd+ImmopCGSOprnng58VuU/Q2tTmppeEhEPUTbqJcAnJA2S9A+0fibVMQ9/08XwM4APS3p3thg7hnJKPq+X87QB5ahymaRNgE6bqTYxA/igpA/nPI+QtFMeNU8DfiRpExVb5obenU7XW0Q8T7lR3Nly+gOwRv43ZXBObzzl5n1v9bbsOymXqr6Ry/99lMtfPTUV+Lyksbl815f0IUnrdpL/j/Dq/3bWznnYSdKuLU6vu+1zVeu4PuUy+BLKcj2CcsZE1vMgSVvkQWnH/7teAmYBb5N0QM7DJErwuHAV69ZMX5fd6bYsaVwul8GUg9oXKA2YoJtlnfuf8bkeh0j6DOVe1h+6qMsM4HDKNvfrSvrPga9LelOWPVzSRyv1f5GyjoZQAmv1wKpZ3SZJ2jiv6DxBCVxdtlwc6IHpm5QWaQuAmyitm/4th91M2ajuy1PSjSiXAg6X9DSlNd8ZqzCtQyiXou6inKl9DiAifk+5eT2LEojeQDkC6bi0djilRdCjlMtc16/CNE8CJkl6XNK/NQ7MU+/DKKf+Syg3zydG680zm84T8H3KqfljlJ3X+a1WOCLuptzr+yrlNH8e5SYuwBcoy2geZQO9kMqOpwvdrbdvAmflev5QQ32eAz4IHJDz80PgoKxnr/S27NzBHkS517YU+GeWH1j1pD5XUM4k/oOy476TckTc9Gg7Dxb2o7Tguo+yDZ1C65e5TgBOyOV+VG/rGBE3UHaK8yhnk9uw4kHWO4Drczs4C5gSEQ9ExMPAhygHpo9Rmk1/MC+r90obyu5qWx4GnEZZLvdQ1snJOWwq5dLbMkkzm5Q7iLI+llAa80wBPhoRt3ZRl0spjRP+GBEPdiRGxOmUe3nn5OXGmyj3UwF+RzmgvTvr+GhOsysfBO6Q9BRlX/mx7vZRWvHyuJmZvV5IuhL4WUT0+ICoHQb6GZOZmfWApD0pLUN/0991aVTrJz+YmVnfy8uB+wBHRg+eHNJuvpRnZma14kt5ZmZWK6+7S3mbbLJJjB49ur+rYWY2YFx//fWPRsSI1TW9111gGj16NPPmdfs3HzMzS5J680SSVeZLeWZmVisOTGZmVisOTGZmVisOTGZmVisOTGZmVisOTGZmVittDUySvihpgaRbJZ2ej2TfJl8WdZekMzreyyFprexfmMNHV8r5SqbfocpLpiRNyLSFko5p57yYmdnq0bbApPJmxKOBsRGxI+Wx7AdT3ld0UkSMobwW4bAc5TDg8YjYjvK6hxOznO1zvB0o7w35Wb7nZxDlsfH7AtsDh2ReMzMbwNp9KW8wsE6++GpdyjtW3g+cncOnA/tn98TsJ4fvJUmZPjMins+X2y0Eds/Pwoi4JyJeAGZmXjMzG8Da9uSHiPiLpO8D91NevXsR5SV5yyoviVrM8je6jiTfFR8RL6m8T37jTL+6UnR1nEUN6Xv0tt67/d8ZvS3iNeP6703q7yqY2etQOy/lDaecwWwDbAGsR7ns1qjj8ebqZNiqpjeryxRJ8yTNW7Kku5ctmplZf2rnpby9gT9HxJJ8hfM5lFc4D8tLewCjKK/ahnLGsyVADh9Ked30q+kN43SWvpKImBoRYyNi7IgRq+05hGZm1gPtDEz3A+MkrZv3ivYCbqO8Z/6AzDMZOC+7Z2U/OfySKC+LmgUcnK32tgHGANcC1wFjspXfEEoDiVltnB8zM1sN2nmP6RpJZwM3AC8BNwJTgf8BZkr6dqadmqOcCvxK0kLKmdLBWc4CSWdSgtpLlDcuvgwg6ShgNqXF37SIWNCu+TEzs9Wjra+9iIhjgWMbku+htKhrzPsccGAn5ZwAnNAk/Xzg/N7X1MzM6sJPfjAzs1pxYDIzs1pxYDIzs1pxYDIzs1pxYDIzs1pxYDIzs1pxYDIzs1pxYDIzs1pxYDIzs1pxYDIzs1pxYDIzs1pxYDIzs1pxYDIzs1pxYDIzs1pxYDIzs1pxYDIzs1pxYDIzs1ppW2CS9CZJN1U+T0r6gqSNJM2RdFd+D8/8knSypIWS5kvatVLW5Mx/l6TJlfTdJN2S45wsSe2aHzMzWz3aFpgi4o6I2CUidgF2A54BzgWOAeZGxBhgbvYD7AuMyc8U4BQASRtRXs++B+WV7Md2BLPMM6Uy3oR2zY+Zma0eq+tS3l7A3RFxHzARmJ7p04H9s3siMCOKq4FhkjYH9gHmRMTSiHgcmANMyGEbRsRVERHAjEpZZmY2QK2uwHQwcHp2bxYRDwLk96aZPhJYVBlncaZ1lb64SfpKJE2RNE/SvCVLlvRyVszMrJ3aHpgkDQE+BJzVXdYmadGD9JUTI6ZGxNiIGDtixIhuqmFmZv1pdZwx7QvcEBEPZ//DeRmO/H4k0xcDW1bGGwU80E36qCbpZmY2gK2OwHQIyy/jAcwCOlrWTQbOq6RPytZ544An8lLfbGC8pOHZ6GE8MDuHPSVpXLbGm1Qpy8zMBqjB7Sxc0rrAB4C/ryR/FzhT0mHA/cCBmX4+sB+wkNKC79MAEbFU0reA6zLf8RGxNLs/B5wGrANckB8zMxvA2hqYIuIZYOOGtMcorfQa8wZwZCflTAOmNUmfB+zYJ5U1M7Na8JMfzMysVhyYzMysVhyYzMysVhyYzMysVhyYzMysVhyYzMysVhyYzMysVhyYzMysVhyYzMysVhyYzMysVhyYzMysVhyYzMysVhyYzMysVhyYzMysVhyYzMysVhyYzMysVhyYzMysVtoamCQNk3S2pD9Jul3SOyRtJGmOpLvye3jmlaSTJS2UNF/SrpVyJmf+uyRNrqTvJumWHOdkSWrn/JiZWfu1+4zpR8CFEfFmYGfgduAYYG5EjAHmZj/AvsCY/EwBTgGQtBFwLLAHsDtwbEcwyzxTKuNNaPP8mJlZm7UtMEnaEHg3cCpARLwQEcuAicD0zDYd2D+7JwIzorgaGCZpc2AfYE5ELI2Ix4E5wIQctmFEXBURAcyolGVmZgNUO8+Y/gZYAvxS0o2SfiFpPWCziHgQIL83zfwjgUWV8RdnWlfpi5ukr0TSFEnzJM1bsmRJ7+fMzMzapp2BaTCwK3BKRLwN+CvLL9s10+z+UPQgfeXEiKkRMTYixo4YMaLrWpuZWb9qZ2BaDCyOiGuy/2xKoHo4L8OR349U8m9ZGX8U8EA36aOapJuZ2QDWtsAUEQ8BiyS9KZP2Am4DZgEdLesmA+dl9yxgUrbOGwc8kZf6ZgPjJQ3PRg/jgdk57ClJ47I13qRKWWZmNkANbnP5nwd+LWkIcA/waUowPFPSYcD9wIGZ93xgP2Ah8EzmJSKWSvoWcF3mOz4ilmb354DTgHWAC/JjZmYDWFsDU0TcBIxtMmivJnkDOLKTcqYB05qkzwN27GU1zcysRvzkBzMzqxUHJjMzqxUHJjMzqxUHJjMzqxUHJjMzqxUHJjMzqxUHJjMzqxUHJjMzqxUHJjMzqxUHJjMzqxUHJjMzqxUHJjMzqxUHJjMzqxUHJjMzqxUHJjMzqxUHJjMzqxUHJjMzq5W2BiZJ90q6RdJNkuZl2kaS5ki6K7+HZ7oknSxpoaT5knatlDM5898laXIlfbcsf2GOq3bOj5mZtd/qOGN6X0TsEhEdr1g/BpgbEWOAudkPsC8wJj9TgFOgBDLgWGAPYHfg2I5glnmmVMab0P7ZMTOzduqPS3kTgenZPR3Yv5I+I4qrgWGSNgf2AeZExNKIeByYA0zIYRtGxFUREcCMSllmZjZAtTswBXCRpOslTcm0zSLiQYD83jTTRwKLKuMuzrSu0hc3SV+JpCmS5kmat2TJkl7OkpmZtdPgNpe/Z0Q8IGlTYI6kP3WRt9n9oehB+sqJEVOBqQBjx45tmsfMzOqhrWdMEfFAfj8CnEu5R/RwXoYjvx/J7IuBLSujjwIe6CZ9VJN0MzMbwNoWmCStJ2mDjm5gPHArMAvoaFk3GTgvu2cBk7J13jjgibzUNxsYL2l4NnoYD8zOYU9JGpet8SZVyjIzswGqnZfyNgPOzRbcg4H/jogLJV0HnCnpMOB+4MDMfz6wH7AQeAb4NEBELJX0LeC6zHd8RCzN7s8BpwHrABfkx8zMBrC2BaaIuAfYuUn6Y8BeTdIDOLKTsqYB05qkzwN27HVlzcysNvzkBzMzqxUHJjMzqxUHJjMzq5WWApOkua2kmZmZ9VaXjR8krQ2sC2ySTbU7/tS6IbBFm+tmZmavQ921yvt74AuUIHQ9ywPTk8BP21gvMzN7neoyMEXEj4AfSfp8RPx4NdXJzMxex1r6H1NE/FjSO4HR1XEiYkab6mVmZq9TLQUmSb8CtgVuAl7O5I5XTZiZmfWZVp/8MBbYPp/OYGZm1jat/o/pVuAN7ayImZkZtH7GtAlwm6Rrgec7EiPiQ22plZmZvW61GpiOa2clzMzMOrTaKu9/210RMzMzaL1V3lMsf235EGBN4K8RsWG7KmZmZq9PrZ4xbVDtl7Q/5TXpZmZmfapHTxePiN8C728lr6RBkm6U9Pvs30bSNZLuknSGpCGZvlb2L8zhoytlfCXT75C0TyV9QqYtlHRMT+bFzMzqpdVLeR+p9K5B+V9Tq/9p+kfgdsqDXwFOBE6KiJmSfg4cBpyS349HxHaSDs58B0naHjgY2IHyzL6LJb0xy/op8AFgMXCdpFkRcVuL9TIzsxpq9Yzp7yqffYCngIndjSRpFPB/gF9kvyhnWmdnlunA/tk9MfvJ4Xtl/onAzIh4PiL+DCykXEbcHVgYEfdExAvAzFbqZGZm9dbqPaZP97D8fwf+Gei4R7UxsCwiXsr+xcDI7B4JLMrpvSTpicw/Eri6UmZ1nEUN6Xv0sJ5mZlYTrb4ocJSkcyU9IulhSb/Js6Guxvkg8EhEXF9NbpI1uhm2qunN6jJF0jxJ85YsWdJFrc3MrL+1einvl8Asyj2ekcDvMq0rewIfknQv5TLb+ylnUMMkdZypjQIeyO7FwJYAOXwosLSa3jBOZ+kriYipETE2IsaOGDGiu3k1M7N+1GpgGhERv4yIl/JzGtDlHj4ivhIRoyJiNKXxwiUR8QngUuCAzDYZOC+7Z2U/OfySfGjsLODgbLW3DTAGuBa4DhiTrfyG5DRmtTg/ZmZWU60GpkclfTKbfg+S9EngsR5O81+AL0laSLmHdGqmnwpsnOlfAo4BiIgFwJnAbcCFwJER8XLepzoKmE1p9Xdm5jUzswGs1WflfQb4CXAS5T7OlUDLDSIi4jLgsuy+hyZ/zo2I54ADOxn/BOCEJunnA+e3Wg8zM6u/VgPTt4DJEfE4gKSNgO9TApaZmVmfafVS3k4dQQkgIpYCb2tPlczM7PWs1cC0hqThHT15xtTq2ZaZmVnLWg0uPwCulHQ25R7Tx2hyz8fMzKy3Wn3ywwxJ8yj/RRLwET+TzszM2qHly3EZiByMzMysrXr02gszM7N2cWAyM7Naccs6a6v7j39rf1ehNrb65i39XQWzAcFnTGZmVisOTGZmVisOTGZmVisOTGZmVisOTGZmVisOTGZmVisOTGZmVisOTGZmVittC0yS1pZ0raSbJS2Q9K+Zvo2kayTdJekMSUMyfa3sX5jDR1fK+kqm3yFpn0r6hExbKOmYds2LmZmtPu08Y3oeeH9E7AzsAkyQNA44ETgpIsYAjwOHZf7DgMcjYjvKK9xPBJC0PXAwsAMwAfiZpEGSBgE/BfYFtgcOybxmZjaAtS0wRfF09q6Zn6C8OuPsTJ8O7J/dE7OfHL6XJGX6zIh4PiL+DCwEds/Pwoi4JyJeAGZmXjMzG8Da+qy8PKu5HtiOcnZzN7AsIl7KLIuBkdk9ElgEEBEvSXoC2DjTr64UWx1nUUP6Hp3UYwowBWCrrbbq3UyZ9ZM9f7xnf1ehNq74/BX9XQVro7Y2foiIlyNiF2AU5QznLc2y5bc6Gbaq6c3qMTUixkbE2BEjRnRfcTMz6zerpVVeRCwDLgPGAcMkdZypjQIeyO7FwJYAOXwosLSa3jBOZ+lmZjaAtbNV3ghJw7J7HWBv4HbgUuCAzDYZOC+7Z2U/OfySiIhMPzhb7W0DjAGuBa4DxmQrvyGUBhKz2jU/Zma2erTzHtPmwPS8z7QGcGZE/F7SbcBMSd8GbgROzfynAr+StJBypnQwQEQskHQm5bXuLwFHRsTLAJKOAmYDg4BpEbGgjfNjZmarQdsCU0TMB97WJP0eyv2mxvTngAM7KesE4IQm6ecD5/e6smZmVht+8oOZmdWKA5OZmdWKA5OZmdWKA5OZmdWKA5OZmdWKA5OZmdWKA5OZmdWKA5OZmdWKA5OZmdWKA5OZmdWKA5OZmdWKA5OZmdWKA5OZmdWKA5OZmdWKA5OZmdWKA5OZmdWKA5OZmdVK2wKTpC0lXSrpdkkLJP1jpm8kaY6ku/J7eKZL0smSFkqaL2nXSlmTM/9dkiZX0neTdEuOc7IktWt+zMxs9WjnGdNLwD9FxFuAccCRkrYHjgHmRsQYYG72A+wLjMnPFOAUKIEMOBbYg/JK9mM7glnmmVIZb0Ib58fMzFaDtgWmiHgwIm7I7qeA24GRwERgemabDuyf3ROBGVFcDQyTtDmwDzAnIpZGxOPAHGBCDtswIq6KiABmVMoyM7MBarXcY5I0GngbcA2wWUQ8CCV4AZtmtpHAospoizOtq/TFTdKbTX+KpHmS5i1ZsqS3s2NmZm3U9sAkaX3gN8AXIuLJrrI2SYsepK+cGDE1IsZGxNgRI0Z0V2UzM+tHbQ1MktakBKVfR8Q5mfxwXoYjvx/J9MXAlpXRRwEPdJM+qkm6mZkNYO1slSfgVOD2iPhhZdAsoKNl3WTgvEr6pGydNw54Ii/1zQbGSxqejR7GA7Nz2FOSxuW0JlXKMjOzAWpwG8veEzgUuEXSTZn2VeC7wJmSDgPuBw7MYecD+wELgWeATwNExFJJ3wKuy3zHR8TS7P4ccBqwDnBBfszMbABrW2CKiD/S/D4QwF5N8gdwZCdlTQOmNUmfB+zYi2qamVnN+MkPZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWKw5MZmZWK20LTJKmSXpE0q2VtI0kzZF0V34Pz3RJOlnSQknzJe1aGWdy5r9L0uRK+m6SbslxTpbU2dtyzcxsAGnnGdNpwISGtGOAuRExBpib/QD7AmPyMwU4BUogA44F9gB2B47tCGaZZ0plvMZpmZnZANS2wBQRlwNLG5InAtOzezqwfyV9RhRXA8MkbQ7sA8yJiKUR8TgwB5iQwzaMiKsiIoAZlbLMzGwAW933mDaLiAcB8nvTTB8JLKrkW5xpXaUvbpLelKQpkuZJmrdkyZJez4SZmbVPXRo/NLs/FD1IbyoipkbE2IgYO2LEiB5W0czMVofVHZgezstw5Pcjmb4Y2LKSbxTwQDfpo5qkm5nZALe6A9MsoKNl3WTgvEr6pGydNw54Ii/1zQbGSxqejR7GA7Nz2FOSxmVrvEmVsszMbAAb3K6CJZ0OvBfYRNJiSuu67wJnSjoMuB84MLOfD+wHLASeAT4NEBFLJX0LuC7zHR8RHQ0qPkdp+bcOcEF+zMxsgGtbYIqIQzoZtFeTvAEc2Uk504BpTdLnATv2po5mZlY/dWn8YGZmBjgwmZlZzTgwmZlZrTgwmZlZrTgwmZlZrTgwmZlZrTgwmZlZrTgwmZlZrTgwmZlZrTgwmZlZrTgwmZlZrTgwmZlZrTgwmZlZrTgwmZlZrTgwmZleTx9MAAAMuUlEQVRZrTgwmZlZrTgwmZlZrQz4wCRpgqQ7JC2UdEx/18fMzHpnQAcmSYOAnwL7AtsDh0javn9rZWZmvTG4vyvQS7sDCyPiHgBJM4GJwG39Wiszq73/ffd7+rsKtfGey/+3v6uwAkVEf9ehxyQdAEyIiMOz/1Bgj4g4qiHfFGBK9r4JuGO1VnTVbQI82t+VeA3x8uxbXp59ayAsz60jYsTqmthAP2NSk7SVIm1ETAWmtr86fUPSvIgY29/1eK3w8uxbXp59y8tzZQP6HhOwGNiy0j8KeKCf6mJmZn1goAem64AxkraRNAQ4GJjVz3UyM7NeGNCX8iLiJUlHAbOBQcC0iFjQz9XqCwPmsuMA4eXZt7w8+5aXZ4MB3fjBzMxeewb6pTwzM3uNcWAyM7NacWDqJUknSfpCpX+2pF9U+n8g6Us9KPerle7Rkm7tfW0HJklvkDRT0t2SbpN0vqQ39qK8X/TXE0IkhaQfVPq/LOm4bsZ5r6R3VvqPkDSpj+vV5TYm6YuSnpM0tJI2VtLJLZS9hqSTJd0q6RZJ10napgd1/IKkdVd1vHaQ9DVJCyTNl3STpD3aOK3jJe2d3bVZBu3kwNR7VwLvhPIDpPxZbofK8HcCV/Sg3K92n6U1kgZsIxdJAs4FLouIbSNie8qy2ayVcXOdVNMGRcThEdHy00Hy0Vd95XngI5I2WYVx3ktuYwAR8fOImNGHdWrFIZRWsB+u1GNeRBzdmLHJ9nYQsAWwU0S8NctY1oM6fAHo952ypHcAHwR2jYidgL2BRW2a1qCI+GZEXJxJtVgG7ebA1HtXsHynsQNwK/CUpOGS1gLeAtwo6f/mkeJ8Sf/aMbKk30q6Po++pmTad4F18kjs15l1kKT/zHwXSVon824r6cIs4w+S3pzpp0n6oaRLgRNXy5Joj/cBL0bEzzsSIuImyjKdK+mGPAqfCK8e+d8u6WfADcCWkp7Oo85rgHdIukzS2Mw/XtJVWc5ZktbP9HslfVPSH4ED+3B+XqK0wvpi4wBJfyfpGkk3SrpY0maSRgNHAF/M7eFvJR0n6cs5zi6Srs7t6lxJwzP9MkknSrpW0p2S/rayfP6Q83tD9UysM5K2BdYHvk4JUB3p75X0++w+TtJUSRcBjUFzc+DBiHgFICIWR8Tjkg6TdFKlvM/mNruepP+RdHOeZR0k6WhKcLs0t+nu1t3/y2HzJO2qciXjbklHdL+KurU58GhEPJ/z82hEPCBpN0n/m7/F2ZI2z/psl+vz5qzrttVll3l+IulTlfq/uu3lb/mAxmXQ2fLrg/nrfxHhTy8/wL3AVsDfU3Yi3wL2A/YELgfGU3ZGohwM/B54d467UX6vQwlqG2f/05XyR1N2aLtk/5nAJ7N7LjAmu/cALsnu03I6g/p7+fRy2R4NnNQkfTCwYXZvAizM5TsaeAUYV8kbwMcq/ZcBY3O8y4H1Mv1fgG9W1uk/t2F+ngY2zPKHAl8Gjsthw1neUvZw4AfZfRzw5UoZr/YD84H3ZPfxwL9X5rFj/P2Ai7N7XWDt7B4DzKtsY7d2UuevA9/IbfdeYNNMfy/w+0qdrgfWaTL+qBzvJuAHwNsyfT3gbmDN7L8SeCvwUeA/K+MPrayTTSrrvKt197nsPimX0QbACOCRPliH6+e83An8DHgPsGbWf0TmOYjy9xWAa4APZ/fauQ5eXXaZ/hPgU822Pcpv+YAmy6Dp8uvv32xffAbsJZ6a6ThreifwQ2Bkdj9B2VjG5+fGzL8+ZadwOXC0pI7LI1tm+mNNpvHnKGcKUHYAo/MI8Z3AWdKrT2daqzLOWRHxcq/nrp4E/D9J76YEopEsv7x3X0RcXcn7MvCbJmWMozyV/opcfkOAqyrDz+jrSgNExJOSZlCC7rOVQaOAM/JIewjw567KUbnfMywiOp7AOR04q5LlnPy+nhJ4oOxAfyJpF8pyaeVe3cGUHesrks6hnEH+tEm+WRHxbGNiRCyW9Cbg/fmZK+nAiJgr6RLgg5Jup+xgb5H0PPB9SSdSdt5/aDKt7tZdxx/tbwHWj4inKFcynpM0LCJ6cimxY36elrQb8LeUM/ozgG8DOwJzsj6DgAclbQCMjIhzc9znACq/1850u+1FxF+bLb+ezVW9ODD1jY77TG+lnPUsAv4JeBKYRjk6+k5E/Ed1JEnvpVyffkdEPCPpMsoRVTPPV7pfppxhrQEsi4hdOhnnrz2Yl7pZABzQJP0TlCPg3SLiRUn3snzZNc73c50EaAFzIuKQJsOaldOX/p1yqfGXlbQfAz+MiFm5bRzXy2l0bDMvs/y3/kXgYWBnyvbzXFcFSNqJcrA0pxIA7qF5YOp0eUW57HUBcIGkh4H9KWf7v6DcM/wTuSwi4s7c8e8HfEfSRRFxfGPV6Hrddcz7K6z423mFPtjv5fZ0GXCZpFuAI4EFEfGOFSopbdhJES+x4q2Uxt99q9veSsvvtcD3mPrGFZSboUsj4uWIWAoMA95BOYqbDXymcg18pKRNKZdyHs+g9GbKUWCHFyWt2dVEI+JJ4M+SDsxyJWnnvp65fnYJsJakz3YkSHo7sDXlssyLkt6X/avqamBPSdtlueuqF639VkVuI2cCh1WShwJ/ye7JlfSnKJeiGst4Ani84/4RcCjQ3fsLhrL8fs+hlCP7rhxCudQ4Oj9bACMltby88x7PFtm9BrATcF/OwzWUKwUfB07PPFsAz0TEfwHfB3bNoqrLod/WnaQ3SRpTSdoFuB0YodIwAklrStohf6OLJe2f6WuptKq7D9g++4cCe7U4+RW2hWbL77XAgalv3EK55n11Q9oTUW6MXgT8N3BVHl2dTdm4LgQGS5pPuS9VHX8qMF/LGz905hPAYZJuppxdTOyLGaqLKBfPPwx8IG9eL6CcSZwPjJU0j7IM/tSDspcAnwJOz3VwNfDmPqp6K35A2W46HEe5LPsHVnwNwu+ADysbPzSUMRn4XtZ/F8p9pq78DJgs6WrKZbzujswPprSKrDo301u1KfA7lebo8ylnCz+pDD8TuCIiHs/+twLXSroJ+BrlMhmU38QFki7t53W3PjBd5a8L8ymXFL9JObM/MX+LN7G8UdShlEv28ylXV94QEYso8z0f+DXLL/N359VlUElrXH4Dnh9JZGb9KlunnRQRc/u7LgPRa3H5+YzJzPqFpGGS7gSefS3tVFeX1/Ly8xmTmZnVis+YzMysVhyYzMysVhyYzMysVvwHW3tdkrQx5Q+eAG+g/Al1SfbvHhEvrKZ6bE/5/8krlKcr3Ls6pmtWZ278YK97Kq+deDoivt8P0/465Xf4rW7yDXoNP17KbAW+lGdWIek7ko6s9J8o6R8k7Z1PdP5t/rHyp8pn9EjaV8ufcn2GpPWalLurypPD50v6jaShkj4EHAUcIenihvyDJS2T9G1J1wK7S3q7lj+9+gJJm2XeI1SeXH+zylO2O548f7DK07lv1vIncq8jabrKE9lvUHnWIJIOl3S2ylOx75L0nTYtYrNuOTCZregXlCcKdLyH6UCWP+plD8r7cN5KeZ3JxHy01DHAXhGxK+Wf/P/YpNz/Av4pyvt77gC+ERGzcnrfi4i9m4wzFLghInanPFfvR8BHI2K3LK/jLOusiHh7ROxMedr0pzL92KzXzix/j9LRwAtR3ot0KPArSUNy2M6UpxfsBHyy4zFCZqub7zGZVUTE3ZKekvRWyvP3ro3y7iCAqzvuAUmaCbwrR9seuFLLH3L6x2qZeT9r7YjoSJ8O/KqF6rzA8scBvYXyvq+Ltfzp1Ytz2E6Sjqc8n3EDyutOoDzDcYaks1j+pPF3Ad/LeV0g6QFguxx2cT6FG0l/orzK5YEW6mnWpxyYzFZ2KuWsYzRQfSJ84w3ZoDzl+sKIOLSL8rp9x0Enno3lN4EFzI+IxmflQXkx374Rcaukw1n+MODPUs7yPgjcrPKk8K7q0vgEe+8frF/4Up7Zyn4D/B3loajVez/jJG2Vl/g+RjkzuhJ4j6S/AVB5+2r1ydNExKPAs1r+tthWngLe6DbKU713z+kMkbRDDlsPeEjlafQfr4zzN/leqm8Aj1PeWXU55aG3SHoL5W2sC1exLmZt5SMiswYR8Zyky4GH8vUQHa6kPBF8B8q7eGZFREg6jPKCv457NV8F7moo9lDglGyYsBD49CrW6XlJBwAnq7x8bnDWZQHlydbXAvdT3gfW8W6fkyRtQzlLuijPqO4G/iOfcv8iMCkiXlD3L64zW23cXNysgco7g24C9o+IezJtb+CoiNi/Xytn9jrgS3lmFdno4W7KfaN7+rs+Zq9HPmMyM7Na8RmTmZnVigOTmZnVigOTmZnVigOTmZnVigOTmZnVyv8HApxUd+kYLg4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.barplot(x=all_cncl_reason, y=cncl_count)\n",
    "plt.xlabel('Type of reason')\n",
    "plt.ylabel('count')\n",
    "plt.title('Total count of cancellation for different reason for last 3 years')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['FL_DATE'] = pd.to_datetime(df['FL_DATE'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "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>FL_DATE</th>\n",
       "      <th>ORIGIN</th>\n",
       "      <th>ORIGIN_CITY_NAME</th>\n",
       "      <th>ORIGIN_STATE_NM</th>\n",
       "      <th>DEST</th>\n",
       "      <th>DEST_CITY_NAME</th>\n",
       "      <th>DEST_STATE_NM</th>\n",
       "      <th>CANCELLED</th>\n",
       "      <th>CANCELLATION_CODE</th>\n",
       "      <th>CRS_ELAPSED_TIME</th>\n",
       "      <th>ACTUAL_ELAPSED_TIME</th>\n",
       "      <th>AIR_TIME</th>\n",
       "      <th>DISTANCE</th>\n",
       "      <th>CARRIER_DELAY</th>\n",
       "      <th>WEATHER_DELAY</th>\n",
       "      <th>NAS_DELAY</th>\n",
       "      <th>SECURITY_DELAY</th>\n",
       "      <th>LATE_AIRCRAFT_DELAY</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-02-01</td>\n",
       "      <td>ORD</td>\n",
       "      <td>Chicago, IL</td>\n",
       "      <td>Illinois</td>\n",
       "      <td>SFO</td>\n",
       "      <td>San Francisco, CA</td>\n",
       "      <td>California</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>298.0</td>\n",
       "      <td>336.0</td>\n",
       "      <td>280.0</td>\n",
       "      <td>1846.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>31.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     FL_DATE ORIGIN ORIGIN_CITY_NAME ORIGIN_STATE_NM DEST     DEST_CITY_NAME  \\\n",
       "0 2017-02-01    ORD      Chicago, IL        Illinois  SFO  San Francisco, CA   \n",
       "\n",
       "  DEST_STATE_NM  CANCELLED CANCELLATION_CODE  CRS_ELAPSED_TIME  \\\n",
       "0    California        0.0               NaN             298.0   \n",
       "\n",
       "   ACTUAL_ELAPSED_TIME  AIR_TIME  DISTANCE  CARRIER_DELAY  WEATHER_DELAY  \\\n",
       "0                336.0     280.0    1846.0            0.0            0.0   \n",
       "\n",
       "   NAS_DELAY  SECURITY_DELAY  LATE_AIRCRAFT_DELAY  \n",
       "0       31.0             0.0                  0.0  "
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['Delay'] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[(df['CARRIER_DELAY'] > 0) | (df['WEATHER_DELAY'] > 0) | (df['WEATHER_DELAY'] > 0) |\n",
    "       (df['SECURITY_DELAY'] > 0) | (df['LATE_AIRCRAFT_DELAY'] > 0),'Delay'] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    8799320\n",
       "1    1564553\n",
       "Name: Delay, dtype: int64"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Delay'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['status'] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[(df['CANCELLATION_CODE'] == 'B') ,'status'] = -1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 1    10283561\n",
       "-1       80312\n",
       "Name: status, dtype: int64"
      ]
     },
     "execution_count": 107,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['status'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10283561, 20)"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sch_df = df[df['status']==1]\n",
    "#sch_df = sch_df.sample(frac=1, random_state=1).head(1000000)\n",
    "sch_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "#delay_df = new_df[new_df['status']==1]\n",
    "cancel_df = df[df['status']==-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.concat([sch_df,cancel_df])\n",
    "df = df.sample(frac=1, random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(15,4));\n",
    "\n",
    "ax = fig.add_axes([0,0,1,1]);\n",
    "\n",
    "df[df['status'] ==1].groupby('FL_DATE')['ORIGIN'].count().plot.line(c='b', label=\"scheduled\");\n",
    "df[df['status'] ==-1].groupby('FL_DATE').count()['ORIGIN'].plot.line(c='r', label=\"cancelled\");\n",
    "#new_df[new_df['status'] ==2].groupby('FL_DATE').count()['ORIGIN'].plot.line(c='g', label=\"cancelled\");\n",
    "plt.xlabel('Date')\n",
    "plt.title('Trend over last 3 years')\n",
    "\n",
    "ax.legend();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 1    10283561\n",
       "-1       80312\n",
       "Name: status, dtype: int64"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['status'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [],
   "source": [
    "city_weather_list = os.listdir(weather_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'weather_data/'"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "weather_path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.FL_DATE = pd.to_datetime(df.FL_DATE)\n",
    "df.set_index('FL_DATE', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['Month'] = df.index.month\n",
    "df['Date'] = df.index.date"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_col_to_str(df):\n",
    "    for col in df.columns:\n",
    "        df[col] = df[col].apply(str)\n",
    "        \n",
    "    return df\n",
    "\n",
    "\n",
    "def remove_letters(df):\n",
    "    for col in df.columns:\n",
    "            #df = df[df[col] != 'T']\n",
    "            #df[col] = df[col].str.sub(\"\\D\", \"\", \"aas30dsa20\")\n",
    "            #df[col].replace(regex=True,inplace=True,to_replace=r'\\D',value=r'')\n",
    "        df[col] = df[col].str.replace(r'[A-Za-z]','')\n",
    "    return df\n",
    "\n",
    "\n",
    "def to_float(x):\n",
    "    try:\n",
    "        return float(x)\n",
    "    except:\n",
    "        return float(''.join(list(x)[:-1]))\n",
    "def make_col_numeric(df):\n",
    "    for col in df.columns:\n",
    "            print(col)\n",
    "            df[col] = df[col].apply(to_float)\n",
    "        \n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20/20 [02:14<00:00,  6.32s/it]\n"
     ]
    }
   ],
   "source": [
    "all_feature_df_list = []\n",
    "reqd_cols = ['DATE','HourlyDewPointTemperature', 'HourlyDryBulbTemperature', 'HourlyPrecipitation', 'HourlyRelativeHumidity', 'HourlyStationPressure', 'HourlyVisibility', 'HourlyWetBulbTemperature']\n",
    "\n",
    "for index,i in enumerate(tqdm(city_weather_list)):\n",
    "    weather_df = pd.read_csv(weather_path+i,low_memory=False)\n",
    "    weather_df = weather_df[reqd_cols]\n",
    "    weather_df.DATE = pd.to_datetime(weather_df.DATE)\n",
    "    weather_df.set_index('DATE', inplace=True)\n",
    "    weather_df = make_col_to_str(weather_df)\n",
    "    weather_df = remove_letters(weather_df)\n",
    "    #weather_df = make_col_numeric(weather_df)\n",
    "    weather_df = weather_df.apply(pd.to_numeric, errors='coerce')  \n",
    "    weather_df.fillna(weather_df.mean(), inplace=True)\n",
    "    weather_df = weather_df.groupby(weather_df.index.date).mean()\n",
    "    city_code = i.split('_')[1][:-4]\n",
    "    temp_city_w = pd.merge(df[df.ORIGIN==city_code], weather_df, how='left', left_index=True, right_index=True)\n",
    "    all_feature_df_list.append(temp_city_w)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_df = pd.concat(all_feature_df_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_df = final_df.sample(frac=1, random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ORIGIN                              0\n",
       "ORIGIN_CITY_NAME                    0\n",
       "ORIGIN_STATE_NM                     0\n",
       "DEST                                0\n",
       "DEST_CITY_NAME                      0\n",
       "DEST_STATE_NM                       0\n",
       "CANCELLED                           0\n",
       "CANCELLATION_CODE            10218590\n",
       "CRS_ELAPSED_TIME                   12\n",
       "ACTUAL_ELAPSED_TIME            170199\n",
       "AIR_TIME                       170199\n",
       "DISTANCE                            0\n",
       "CARRIER_DELAY                 8380934\n",
       "WEATHER_DELAY                 8380934\n",
       "NAS_DELAY                     8380934\n",
       "SECURITY_DELAY                8380934\n",
       "LATE_AIRCRAFT_DELAY           8380934\n",
       "Delay                               0\n",
       "status                              0\n",
       "Month                               0\n",
       "Date                                0\n",
       "HourlyDewPointTemperature           0\n",
       "HourlyDryBulbTemperature            0\n",
       "HourlyPrecipitation                 0\n",
       "HourlyRelativeHumidity              0\n",
       "HourlyStationPressure               0\n",
       "HourlyVisibility                    0\n",
       "HourlyWetBulbTemperature            0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_df.isnull().sum(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_df.dropna(axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = final_df.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "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>ORIGIN</th>\n",
       "      <th>ORIGIN_CITY_NAME</th>\n",
       "      <th>ORIGIN_STATE_NM</th>\n",
       "      <th>DEST</th>\n",
       "      <th>DEST_CITY_NAME</th>\n",
       "      <th>DEST_STATE_NM</th>\n",
       "      <th>CANCELLED</th>\n",
       "      <th>DISTANCE</th>\n",
       "      <th>Delay</th>\n",
       "      <th>status</th>\n",
       "      <th>Month</th>\n",
       "      <th>Date</th>\n",
       "      <th>HourlyDewPointTemperature</th>\n",
       "      <th>HourlyDryBulbTemperature</th>\n",
       "      <th>HourlyPrecipitation</th>\n",
       "      <th>HourlyRelativeHumidity</th>\n",
       "      <th>HourlyStationPressure</th>\n",
       "      <th>HourlyVisibility</th>\n",
       "      <th>HourlyWetBulbTemperature</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-16</th>\n",
       "      <td>DFW</td>\n",
       "      <td>Dallas/Fort Worth, TX</td>\n",
       "      <td>Texas</td>\n",
       "      <td>LGA</td>\n",
       "      <td>New York, NY</td>\n",
       "      <td>New York</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1389.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>2016-11-16</td>\n",
       "      <td>48.008287</td>\n",
       "      <td>69.664524</td>\n",
       "      <td>0.002465</td>\n",
       "      <td>51.039878</td>\n",
       "      <td>29.284557</td>\n",
       "      <td>9.962225</td>\n",
       "      <td>57.533681</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-19</th>\n",
       "      <td>IAH</td>\n",
       "      <td>Houston, TX</td>\n",
       "      <td>Texas</td>\n",
       "      <td>ORD</td>\n",
       "      <td>Chicago, IL</td>\n",
       "      <td>Illinois</td>\n",
       "      <td>0.0</td>\n",
       "      <td>925.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2017-03-19</td>\n",
       "      <td>63.632470</td>\n",
       "      <td>72.527653</td>\n",
       "      <td>0.002963</td>\n",
       "      <td>75.860489</td>\n",
       "      <td>30.109542</td>\n",
       "      <td>9.583841</td>\n",
       "      <td>66.879557</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           ORIGIN       ORIGIN_CITY_NAME ORIGIN_STATE_NM DEST DEST_CITY_NAME  \\\n",
       "2016-11-16    DFW  Dallas/Fort Worth, TX           Texas  LGA   New York, NY   \n",
       "2017-03-19    IAH            Houston, TX           Texas  ORD    Chicago, IL   \n",
       "\n",
       "           DEST_STATE_NM  CANCELLED  DISTANCE  Delay  status  Month  \\\n",
       "2016-11-16      New York        0.0    1389.0      0       1     11   \n",
       "2017-03-19      Illinois        0.0     925.0      0       1      3   \n",
       "\n",
       "                  Date  HourlyDewPointTemperature  HourlyDryBulbTemperature  \\\n",
       "2016-11-16  2016-11-16                  48.008287                 69.664524   \n",
       "2017-03-19  2017-03-19                  63.632470                 72.527653   \n",
       "\n",
       "            HourlyPrecipitation  HourlyRelativeHumidity  \\\n",
       "2016-11-16             0.002465               51.039878   \n",
       "2017-03-19             0.002963               75.860489   \n",
       "\n",
       "            HourlyStationPressure  HourlyVisibility  HourlyWetBulbTemperature  \n",
       "2016-11-16              29.284557          9.962225                 57.533681  \n",
       "2017-03-19              30.109542          9.583841                 66.879557  "
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 1    10283561\n",
       "-1       80312\n",
       "Name: status, dtype: int64"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['status'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10363873, 19)"
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.sample(frac=1, random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 10363873 entries, 2017-05-03 to 2016-10-11\n",
      "Data columns (total 19 columns):\n",
      "ORIGIN                       object\n",
      "ORIGIN_CITY_NAME             object\n",
      "ORIGIN_STATE_NM              object\n",
      "DEST                         object\n",
      "DEST_CITY_NAME               object\n",
      "DEST_STATE_NM                object\n",
      "CANCELLED                    float64\n",
      "DISTANCE                     float64\n",
      "Delay                        int64\n",
      "status                       int64\n",
      "Month                        int64\n",
      "Date                         object\n",
      "HourlyDewPointTemperature    float64\n",
      "HourlyDryBulbTemperature     float64\n",
      "HourlyPrecipitation          float64\n",
      "HourlyRelativeHumidity       float64\n",
      "HourlyStationPressure        float64\n",
      "HourlyVisibility             float64\n",
      "HourlyWetBulbTemperature     float64\n",
      "dtypes: float64(9), int64(3), object(7)\n",
      "memory usage: 1.5+ GB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [],
   "source": [
    "categorical_features = list(df.select_dtypes(include=['O']).columns)\n",
    "for column in categorical_features:\n",
    "    df[column] = df[column].astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [],
   "source": [
    "category_columns = df.select_dtypes(['category']).columns\n",
    "df[category_columns] = df[category_columns].apply(lambda x: x.cat.codes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 1    10283561\n",
       "-1       80312\n",
       "Name: status, dtype: int64"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['status'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['ORIGIN', 'ORIGIN_CITY_NAME', 'ORIGIN_STATE_NM', 'DEST',\n",
       "       'DEST_CITY_NAME', 'DEST_STATE_NM', 'CANCELLED', 'DISTANCE', 'Delay',\n",
       "       'status', 'Month', 'Date', 'HourlyDewPointTemperature',\n",
       "       'HourlyDryBulbTemperature', 'HourlyPrecipitation',\n",
       "       'HourlyRelativeHumidity', 'HourlyStationPressure', 'HourlyVisibility',\n",
       "       'HourlyWetBulbTemperature'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df.drop(['status','CANCELLED','Delay'],axis=1)\n",
    "y = df['status']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import train_test_split, KFold, cross_val_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/bat/anaconda3/lib/python3.7/site-packages/sklearn/preprocessing/data.py:625: DataConversionWarning: Data with input dtype int8, int16, int64, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.partial_fit(X, y)\n",
      "/home/bat/anaconda3/lib/python3.7/site-packages/sklearn/base.py:462: DataConversionWarning: Data with input dtype int8, int16, int64, float64 were all converted to float64 by StandardScaler.\n",
      "  return self.fit(X, **fit_params).transform(X)\n"
     ]
    }
   ],
   "source": [
    "X = StandardScaler().fit_transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report, confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import itertools\n",
    "#plt.grid(b=None)\n",
    "\n",
    "\n",
    "def plot_confusion_matrix(cm,\n",
    "                          target_names,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=None,\n",
    "                          normalize=True):\n",
    "    \n",
    "    \n",
    "    \n",
    "    accuracy = np.trace(cm) / float(np.sum(cm))\n",
    "    misclass = 1 - accuracy\n",
    "\n",
    "    if cmap is None:\n",
    "        cmap = plt.get_cmap('Blues')\n",
    "\n",
    "    plt.figure(figsize=(8, 6))\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "\n",
    "    if target_names is not None:\n",
    "        tick_marks = np.arange(len(target_names))\n",
    "        plt.xticks(tick_marks, target_names, rotation=45)\n",
    "        plt.yticks(tick_marks, target_names)\n",
    "\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "\n",
    "\n",
    "    thresh = cm.max() / 1.5 if normalize else cm.max() / 2\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        if normalize:\n",
    "            plt.text(j, i, \"{:0.4f}\".format(cm[i, j]),\n",
    "                     horizontalalignment=\"center\",\n",
    "                     color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "        else:\n",
    "            plt.text(j, i, \"{:,}\".format(cm[i, j]),\n",
    "                     horizontalalignment=\"center\",\n",
    "                     color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label\\naccuracy={:0.4f}; misclass={:0.4f}'.format(accuracy, misclass))\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import IsolationForest\n",
    "rng = np.random.RandomState(42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = IsolationForest(behaviour='new', max_samples=100,\n",
    "                      random_state=rng, contamination='auto')\n",
    "\n",
    "clf.fit(X_train)\n",
    "y_pred = clf.predict(X_test)\n",
    "#y_pred_test = clf.predict(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7667133459112134"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_confusion_matrix(cm           = confusion_matrix(y_test, y_pred), \n",
    "                      normalize    = False,\n",
    "                      target_names = ['cancelled','scheduled'],\n",
    "                      title        = \"Confusion Matrix\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.3900591661486751"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "16943 /(16943 + 26494)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5446721822641666"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "13101/ (13101 + 10952)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}