prediction.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from dateutil.parser import parse\n",
    "import datetime\n",
    "import re\n",
    "\n",
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.model_selection import StratifiedShuffleSplit\n",
    "from sklearn.utils import check_array\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from scipy import sparse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "interviews = pd.read_csv(\"data/raw/Interview.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date of Interview</th>\n",
       "      <th>Client name</th>\n",
       "      <th>Industry</th>\n",
       "      <th>Location</th>\n",
       "      <th>Position to be closed</th>\n",
       "      <th>Nature of Skillset</th>\n",
       "      <th>Interview Type</th>\n",
       "      <th>Name(Cand ID)</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Candidate Current Location</th>\n",
       "      <th>...</th>\n",
       "      <th>Are you clear with the venue details and the landmark.</th>\n",
       "      <th>Has the call letter been shared</th>\n",
       "      <th>Expected Attendance</th>\n",
       "      <th>Observed Attendance</th>\n",
       "      <th>Marital Status</th>\n",
       "      <th>Unnamed: 23</th>\n",
       "      <th>Unnamed: 24</th>\n",
       "      <th>Unnamed: 25</th>\n",
       "      <th>Unnamed: 26</th>\n",
       "      <th>Unnamed: 27</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>13.02.2015</td>\n",
       "      <td>Hospira</td>\n",
       "      <td>Pharmaceuticals</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>Production- Sterile</td>\n",
       "      <td>Routine</td>\n",
       "      <td>Scheduled Walkin</td>\n",
       "      <td>Candidate 1</td>\n",
       "      <td>Male</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Single</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>1</th>\n",
       "      <td>13.02.2015</td>\n",
       "      <td>Hospira</td>\n",
       "      <td>Pharmaceuticals</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>Production- Sterile</td>\n",
       "      <td>Routine</td>\n",
       "      <td>Scheduled Walkin</td>\n",
       "      <td>Candidate 2</td>\n",
       "      <td>Male</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Single</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>2</th>\n",
       "      <td>13.02.2015</td>\n",
       "      <td>Hospira</td>\n",
       "      <td>Pharmaceuticals</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>Production- Sterile</td>\n",
       "      <td>Routine</td>\n",
       "      <td>Scheduled Walkin</td>\n",
       "      <td>Candidate 3</td>\n",
       "      <td>Male</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Uncertain</td>\n",
       "      <td>No</td>\n",
       "      <td>Single</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>13.02.2015</td>\n",
       "      <td>Hospira</td>\n",
       "      <td>Pharmaceuticals</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>Production- Sterile</td>\n",
       "      <td>Routine</td>\n",
       "      <td>Scheduled Walkin</td>\n",
       "      <td>Candidate 4</td>\n",
       "      <td>Male</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Uncertain</td>\n",
       "      <td>No</td>\n",
       "      <td>Single</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>4</th>\n",
       "      <td>13.02.2015</td>\n",
       "      <td>Hospira</td>\n",
       "      <td>Pharmaceuticals</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>Production- Sterile</td>\n",
       "      <td>Routine</td>\n",
       "      <td>Scheduled Walkin</td>\n",
       "      <td>Candidate 5</td>\n",
       "      <td>Male</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Uncertain</td>\n",
       "      <td>No</td>\n",
       "      <td>Married</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  Date of Interview Client name         Industry Location  \\\n",
       "0        13.02.2015     Hospira  Pharmaceuticals  Chennai   \n",
       "1        13.02.2015     Hospira  Pharmaceuticals  Chennai   \n",
       "2        13.02.2015     Hospira  Pharmaceuticals  Chennai   \n",
       "3        13.02.2015     Hospira  Pharmaceuticals  Chennai   \n",
       "4        13.02.2015     Hospira  Pharmaceuticals  Chennai   \n",
       "\n",
       "  Position to be closed Nature of Skillset    Interview Type Name(Cand ID)  \\\n",
       "0   Production- Sterile            Routine  Scheduled Walkin   Candidate 1   \n",
       "1   Production- Sterile            Routine  Scheduled Walkin   Candidate 2   \n",
       "2   Production- Sterile            Routine  Scheduled Walkin   Candidate 3   \n",
       "3   Production- Sterile            Routine  Scheduled Walkin   Candidate 4   \n",
       "4   Production- Sterile            Routine  Scheduled Walkin   Candidate 5   \n",
       "\n",
       "  Gender Candidate Current Location     ...      \\\n",
       "0   Male                    Chennai     ...       \n",
       "1   Male                    Chennai     ...       \n",
       "2   Male                    Chennai     ...       \n",
       "3   Male                    Chennai     ...       \n",
       "4   Male                    Chennai     ...       \n",
       "\n",
       "  Are you clear with the venue details and the landmark.  \\\n",
       "0                                                Yes       \n",
       "1                                                Yes       \n",
       "2                                                NaN       \n",
       "3                                                Yes       \n",
       "4                                                Yes       \n",
       "\n",
       "  Has the call letter been shared Expected Attendance Observed Attendance  \\\n",
       "0                             Yes                 Yes                  No   \n",
       "1                             Yes                 Yes                  No   \n",
       "2                             NaN           Uncertain                  No   \n",
       "3                             Yes           Uncertain                  No   \n",
       "4                             Yes           Uncertain                  No   \n",
       "\n",
       "  Marital Status Unnamed: 23 Unnamed: 24 Unnamed: 25 Unnamed: 26 Unnamed: 27  \n",
       "0         Single         NaN         NaN         NaN         NaN         NaN  \n",
       "1         Single         NaN         NaN         NaN         NaN         NaN  \n",
       "2         Single         NaN         NaN         NaN         NaN         NaN  \n",
       "3         Single         NaN         NaN         NaN         NaN         NaN  \n",
       "4        Married         NaN         NaN         NaN         NaN         NaN  \n",
       "\n",
       "[5 rows x 28 columns]"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "interviews.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date of Interview</th>\n",
       "      <th>Client name</th>\n",
       "      <th>Industry</th>\n",
       "      <th>Location</th>\n",
       "      <th>Position to be closed</th>\n",
       "      <th>Nature of Skillset</th>\n",
       "      <th>Interview Type</th>\n",
       "      <th>Name(Cand ID)</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Candidate Current Location</th>\n",
       "      <th>...</th>\n",
       "      <th>Are you clear with the venue details and the landmark.</th>\n",
       "      <th>Has the call letter been shared</th>\n",
       "      <th>Expected Attendance</th>\n",
       "      <th>Observed Attendance</th>\n",
       "      <th>Marital Status</th>\n",
       "      <th>Unnamed: 23</th>\n",
       "      <th>Unnamed: 24</th>\n",
       "      <th>Unnamed: 25</th>\n",
       "      <th>Unnamed: 26</th>\n",
       "      <th>Unnamed: 27</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1229</th>\n",
       "      <td>07.05.2016</td>\n",
       "      <td>Pfizer</td>\n",
       "      <td>Pharmaceuticals</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>Niche</td>\n",
       "      <td>Biosimiliars</td>\n",
       "      <td>Scheduled</td>\n",
       "      <td>Candidate 1230</td>\n",
       "      <td>Male</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Single</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>1230</th>\n",
       "      <td>06.05.2016</td>\n",
       "      <td>Pfizer</td>\n",
       "      <td>Pharmaceuticals</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>Niche</td>\n",
       "      <td>Biosimiliars</td>\n",
       "      <td>Scheduled</td>\n",
       "      <td>Candidate 1231</td>\n",
       "      <td>Male</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Married</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>1231</th>\n",
       "      <td>06.05.2016</td>\n",
       "      <td>Pfizer</td>\n",
       "      <td>Pharmaceuticals</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>Niche</td>\n",
       "      <td>generic drugs – RA</td>\n",
       "      <td>Scheduled</td>\n",
       "      <td>Candidate 1232</td>\n",
       "      <td>Male</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Single</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>1232</th>\n",
       "      <td>06.05.2016</td>\n",
       "      <td>Pfizer</td>\n",
       "      <td>Pharmaceuticals</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>Niche</td>\n",
       "      <td>generic drugs – RA</td>\n",
       "      <td>Scheduled</td>\n",
       "      <td>Candidate 1233</td>\n",
       "      <td>Female</td>\n",
       "      <td>Chennai</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Uncertain</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Single</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>1233</th>\n",
       "      <td>NaN</td>\n",
       "      <td>﻿﻿</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Date of Interview Client name         Industry Location  \\\n",
       "1229        07.05.2016      Pfizer  Pharmaceuticals  Chennai   \n",
       "1230        06.05.2016      Pfizer  Pharmaceuticals  Chennai   \n",
       "1231        06.05.2016      Pfizer  Pharmaceuticals  Chennai   \n",
       "1232        06.05.2016      Pfizer  Pharmaceuticals  Chennai   \n",
       "1233               NaN          ﻿﻿              NaN      NaN   \n",
       "\n",
       "     Position to be closed  Nature of Skillset Interview Type   Name(Cand ID)  \\\n",
       "1229                 Niche        Biosimiliars     Scheduled   Candidate 1230   \n",
       "1230                 Niche        Biosimiliars     Scheduled   Candidate 1231   \n",
       "1231                 Niche  generic drugs – RA     Scheduled   Candidate 1232   \n",
       "1232                 Niche  generic drugs – RA     Scheduled   Candidate 1233   \n",
       "1233                   NaN                 NaN            NaN             NaN   \n",
       "\n",
       "      Gender Candidate Current Location     ...      \\\n",
       "1229    Male                    Chennai     ...       \n",
       "1230    Male                    Chennai     ...       \n",
       "1231    Male                    Chennai     ...       \n",
       "1232  Female                    Chennai     ...       \n",
       "1233     NaN                        NaN     ...       \n",
       "\n",
       "     Are you clear with the venue details and the landmark.  \\\n",
       "1229                                                Yes       \n",
       "1230                                                Yes       \n",
       "1231                                                Yes       \n",
       "1232                                                NaN       \n",
       "1233                                                NaN       \n",
       "\n",
       "     Has the call letter been shared Expected Attendance Observed Attendance  \\\n",
       "1229                             Yes                 Yes                 Yes   \n",
       "1230                             Yes                 Yes                 Yes   \n",
       "1231                             Yes                 Yes                 Yes   \n",
       "1232                             NaN           Uncertain                 Yes   \n",
       "1233                             NaN                 NaN                 NaN   \n",
       "\n",
       "     Marital Status Unnamed: 23 Unnamed: 24 Unnamed: 25 Unnamed: 26  \\\n",
       "1229         Single         NaN         NaN         NaN         NaN   \n",
       "1230        Married         NaN         NaN         NaN         NaN   \n",
       "1231         Single         NaN         NaN         NaN         NaN   \n",
       "1232         Single         NaN         NaN         NaN         NaN   \n",
       "1233            NaN         NaN         NaN         NaN         NaN   \n",
       "\n",
       "     Unnamed: 27  \n",
       "1229         NaN  \n",
       "1230         NaN  \n",
       "1231         NaN  \n",
       "1232         NaN  \n",
       "1233         NaN  \n",
       "\n",
       "[5 rows x 28 columns]"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "interviews.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#the last row is redundant, drop it\n",
    "interviews = interviews.drop([1233])\n",
    "\n",
    "#last 5 columns are empty, drop them\n",
    "interviews = interviews.drop([\"Name(Cand ID)\", \"Unnamed: 23\", \"Unnamed: 24\", \"Unnamed: 25\", \"Unnamed: 26\", \"Unnamed: 27\"], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1233 entries, 0 to 1232\n",
      "Data columns (total 22 columns):\n",
      "Date of Interview                                                                                     1233 non-null object\n",
      "Client name                                                                                           1233 non-null object\n",
      "Industry                                                                                              1233 non-null object\n",
      "Location                                                                                              1233 non-null object\n",
      "Position to be closed                                                                                 1233 non-null object\n",
      "Nature of Skillset                                                                                    1233 non-null object\n",
      "Interview Type                                                                                        1233 non-null object\n",
      "Gender                                                                                                1233 non-null object\n",
      "Candidate Current Location                                                                            1233 non-null object\n",
      "Candidate Job Location                                                                                1233 non-null object\n",
      "Interview Venue                                                                                       1233 non-null object\n",
      "Candidate Native location                                                                             1233 non-null object\n",
      "Have you obtained the necessary permission to start at the required time                              1029 non-null object\n",
      "Hope there will be no unscheduled meetings                                                            986 non-null object\n",
      "Can I Call you three hours before the interview and follow up on your attendance for the interview    986 non-null object\n",
      "Can I have an alternative number/ desk number. I assure you that I will not trouble you too much      986 non-null object\n",
      "Have you taken a printout of your updated resume. Have you read the JD and understood the same        985 non-null object\n",
      "Are you clear with the venue details and the landmark.                                                985 non-null object\n",
      "Has the call letter been shared                                                                       988 non-null object\n",
      "Expected Attendance                                                                                   1228 non-null object\n",
      "Observed Attendance                                                                                   1233 non-null object\n",
      "Marital Status                                                                                        1233 non-null object\n",
      "dtypes: object(22)\n",
      "memory usage: 221.6+ KB\n"
     ]
    }
   ],
   "source": [
    "interviews.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "interviews = interviews.rename(index=str, columns={\n",
    "    \"Have you obtained the necessary permission to start at the required time\": \"question_permission\",\n",
    "    \"Hope there will be no unscheduled meetings\": \"question_no_unscheduled_meetings\",\n",
    "    \"Can I Call you three hours before the interview and follow up on your attendance for the interview\": \"question_follow_up\",\n",
    "    \"Can I have an alternative number/ desk number. I assure you that I will not trouble you too much\": \"question_alternative_number\",\n",
    "    \"Have you taken a printout of your updated resume. Have you read the JD and understood the same\": \"question_resume_printout\",\n",
    "    \"Are you clear with the venue details and the landmark.\": \"question_venue\",\n",
    "    \"Has the call letter been shared\": \"question_call_letter\"\n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Date of Interview', 'Client name', 'Industry', 'Location',\n",
       "       'Position to be closed', 'Nature of Skillset', 'Interview Type',\n",
       "       'Gender', 'Candidate Current Location', 'Candidate Job Location',\n",
       "       'Interview Venue', 'Candidate Native location', 'question_permission',\n",
       "       'question_no_unscheduled_meetings', 'question_follow_up',\n",
       "       'question_alternative_number', 'question_resume_printout',\n",
       "       'question_venue', 'question_call_letter', 'Expected Attendance',\n",
       "       'Observed Attendance', 'Marital Status'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "interviews.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "interviews.update(interviews[[\n",
    "    \"question_permission\", \"question_no_unscheduled_meetings\", \"question_follow_up\",\n",
    "    \"question_alternative_number\", \"question_resume_printout\", \"question_venue\", \"question_call_letter\"\n",
    "]].fillna(\"Unknown\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-- Date of Interview\n",
      "06.02.2016               220\n",
      "04/12/16                  86\n",
      "13.04.2016                84\n",
      "25.06.2015                43\n",
      "02.09.2016                43\n",
      "21.4.16                   38\n",
      "22.01.2016                34\n",
      "07.05.2016                31\n",
      "03.04.2015                30\n",
      "14.01.2016                28\n",
      "26.02.2016                26\n",
      "27.02.2016                26\n",
      "05/11/2016                25\n",
      "20.08.2016                25\n",
      "25.5.2016                 24\n",
      "23.02.2016                22\n",
      "02.12.2015                19\n",
      "19.06.2015                18\n",
      "19.03.16                  17\n",
      "30.01.2016                16\n",
      "09-01-2016                15\n",
      "13.06.2016                15\n",
      "11.5.16                   14\n",
      "21.4.2016                 14\n",
      "30.1.16                   14\n",
      "25-5-2016                 14\n",
      "22.03.2016                13\n",
      "13/03/2015                13\n",
      "25.05.2016                12\n",
      "3.6.16                    12\n",
      "                        ... \n",
      "28.08.2016 & 9.30 AM       2\n",
      "03/06/16                   2\n",
      "28.08.2016 & 10.30 AM      2\n",
      "24.05.2016                 2\n",
      "18 Apr 16                  2\n",
      "28.08.2016 & 09.30 AM      2\n",
      "20 Apr 16                  2\n",
      "19 Apr 16                  2\n",
      "30.1.2016                  1\n",
      "12.04.2018                 1\n",
      "30-1-2016                  1\n",
      "12.04.2016                 1\n",
      "28.8.2016 & 03.00 PM       1\n",
      "4.5.16                     1\n",
      "28.08.2016 & 09.00 AM      1\n",
      "28.8.2016 & 12.00 PM       1\n",
      "28.08.2016 & 10.00 AM      1\n",
      "28.08.2016 & 11.00 AM      1\n",
      "28.8.2016 & 02.00 PM       1\n",
      "28.8.2016 & 09.30 AM       1\n",
      "12.04.2023                 1\n",
      "12.04.2022                 1\n",
      "28.8.2016 & 04.00 PM       1\n",
      "12.04.2021                 1\n",
      "12.04.2020                 1\n",
      "12.04.2017                 1\n",
      "12.04.2019                 1\n",
      "28.8.2016 & 11.00 AM       1\n",
      "22 -Apr -16                1\n",
      "30.1.16                    1\n",
      "Name: Date of Interview, Length: 96, dtype: int64\n",
      "-- Client name\n",
      "Standard Chartered Bank            904\n",
      "Pfizer                              75\n",
      "Hospira                             75\n",
      "Aon Hewitt                          28\n",
      "Flextronics                         23\n",
      "ANZ                                 22\n",
      "Hewitt                              20\n",
      "UST                                 18\n",
      "Prodapt                             17\n",
      "Standard Chartered Bank Chennai     17\n",
      "Astrazeneca                         15\n",
      "Williams Lea                        11\n",
      "Barclays                             5\n",
      "Aon hewitt Gurgaon                   2\n",
      "Woori Bank                           1\n",
      "Name: Client name, dtype: int64\n",
      "-- Industry\n",
      "BFSI                        949\n",
      "Pharmaceuticals             165\n",
      "IT Products and Services     45\n",
      "Electronics                  23\n",
      "IT Services                  23\n",
      "Telecom                      17\n",
      "IT                           11\n",
      "Name: Industry, dtype: int64\n",
      "-- Location\n",
      "Chennai       754\n",
      "Bangalore     292\n",
      "chennai        86\n",
      "Hyderabad      38\n",
      "Gurgaon        33\n",
      "Noida          15\n",
      "- Cochin-       9\n",
      "chennai         3\n",
      "CHENNAI         1\n",
      "Gurgaonr        1\n",
      "Delhi           1\n",
      "Name: Location, dtype: int64\n",
      "-- Position to be closed\n",
      "Routine                1023\n",
      "Niche                   163\n",
      "Dot Net                  18\n",
      "Trade Finance            11\n",
      "AML                       8\n",
      "Selenium testing          5\n",
      "Production- Sterile       5\n",
      "Name: Position to be closed, dtype: int64\n",
      "-- Nature of Skillset\n",
      "JAVA/J2EE/Struts/Hibernate                220\n",
      "Accounting Operations                      86\n",
      "Fresher                                    86\n",
      "AML/KYC/CDD                                84\n",
      "CDD KYC                                    52\n",
      "Routine                                    47\n",
      "Oracle                                     43\n",
      "JAVA/SPRING/HIBERNATE/JSF                  42\n",
      "Java J2EE                                  33\n",
      "SAS                                        27\n",
      "Oracle Plsql                               25\n",
      "Java Developer                             25\n",
      "Lending and Liabilities                    22\n",
      "Banking Operations                         22\n",
      "Java                                       21\n",
      "Core Java                                  17\n",
      "Java J2ee                                  16\n",
      "T-24 developer                             15\n",
      "ALS Testing                                15\n",
      "Senior software engineer-Mednet            15\n",
      "SCCM                                       14\n",
      "Analytical R & D                           13\n",
      "COTS Developer                             13\n",
      "Sr Automation Testing                      13\n",
      "Regulatory                                 12\n",
      "Hadoop                                     12\n",
      "testing                                    11\n",
      "Java                                       10\n",
      "Dot Net                                     9\n",
      "RA Publishing                               9\n",
      "                                         ... \n",
      "Java,SQL                                    3\n",
      "Biosimillar                                 3\n",
      "L & L                                       2\n",
      "11.30 AM                                    2\n",
      "Production Support - SCCM                   2\n",
      "LCM -Manager                                2\n",
      "Java, Spring, Hibernate                     2\n",
      "Lending&Liablities                          2\n",
      "Submission Management                       2\n",
      "RA Label                                    2\n",
      "Banking operations                          2\n",
      "Biosimilars                                 1\n",
      "Production                                  1\n",
      "sccm                                        1\n",
      "Technical Lead                              1\n",
      "Java, J2Ee                                  1\n",
      "9.00 Am                                     1\n",
      "Manager                                     1\n",
      "12.30 Pm                                    1\n",
      "9.30 AM                                     1\n",
      "SCCM-(Network, sharepoint,ms exchange)      1\n",
      "SCCM – Sharepoint                           1\n",
      "SCCM – SQL                                  1\n",
      "Tech Lead- Mednet                           1\n",
      "11.30 Am                                    1\n",
      "JAVA, J2ee                                  1\n",
      "10.00 AM                                    1\n",
      "Sccm- networking                            1\n",
      "JAVA,J2ee                                   1\n",
      "BaseSAS Program/ Reporting                  1\n",
      "Name: Nature of Skillset, Length: 92, dtype: int64\n",
      "-- Interview Type\n",
      "Scheduled Walk In    456\n",
      "Scheduled            371\n",
      "Walkin               189\n",
      "Scheduled Walkin     189\n",
      "Walkin                27\n",
      "Sceduled walkin        1\n",
      "Name: Interview Type, dtype: int64\n",
      "-- Gender\n",
      "Male      965\n",
      "Female    268\n",
      "Name: Gender, dtype: int64\n",
      "-- Candidate Current Location\n",
      "Chennai       754\n",
      "Bangalore     292\n",
      "chennai        86\n",
      "Hyderabad      38\n",
      "Gurgaon        34\n",
      "Noida          15\n",
      "- Cochin-       9\n",
      "chennai         3\n",
      "CHENNAI         1\n",
      "Delhi           1\n",
      "Name: Candidate Current Location, dtype: int64\n",
      "-- Candidate Job Location\n",
      "Chennai          893\n",
      "Bangalore        259\n",
      "Gurgaon           35\n",
      "Visakapatinam     21\n",
      "Noida             15\n",
      "- Cochin-          9\n",
      "Hosur              1\n",
      "Name: Candidate Job Location, dtype: int64\n",
      "-- Interview Venue\n",
      "Chennai       852\n",
      "Bangalore     277\n",
      "Hyderabad      40\n",
      "Gurgaon        35\n",
      "Noida          15\n",
      "- Cochin-       9\n",
      "Hosur           5\n",
      "Name: Interview Venue, dtype: int64\n",
      "-- Candidate Native location\n",
      "Chennai          595\n",
      "Hyderabad        172\n",
      "Bangalore        151\n",
      "Gurgaon           26\n",
      "Cuttack           25\n",
      "Cochin            24\n",
      "Pune              22\n",
      "Coimbatore        21\n",
      "Allahabad         20\n",
      "Noida             17\n",
      "Visakapatinam     16\n",
      "Nagercoil         16\n",
      "Trivandrum        14\n",
      "Kolkata           14\n",
      "Trichy            13\n",
      "Vellore           12\n",
      "Mumbai             7\n",
      "Chitoor            6\n",
      "- Cochin-          5\n",
      "Pondicherry        5\n",
      "Ahmedabad          5\n",
      "Chandigarh         5\n",
      "Delhi              4\n",
      "Vijayawada         4\n",
      "Tirupati           4\n",
      "Salem              3\n",
      "Hosur              3\n",
      "Warangal           3\n",
      "Ambur              2\n",
      "Patna              2\n",
      "Delhi /NCR         2\n",
      "Tanjore            1\n",
      "Tuticorin          1\n",
      "Baddi              1\n",
      "Panjim             1\n",
      "Bhubaneshwar       1\n",
      "Lucknow            1\n",
      "Ghaziabad          1\n",
      "Anantapur          1\n",
      "Belgaum            1\n",
      "Faizabad           1\n",
      "Mysore             1\n",
      "Hissar             1\n",
      "Agra               1\n",
      "Kurnool            1\n",
      "Kanpur             1\n",
      "Name: Candidate Native location, dtype: int64\n",
      "-- question_permission\n",
      "Yes               917\n",
      "Unknown           204\n",
      "No                 79\n",
      "Not yet            19\n",
      "Na                  5\n",
      "yes                 4\n",
      "Yet to confirm      4\n",
      "NO                  1\n",
      "Name: question_permission, dtype: int64\n",
      "-- question_no_unscheduled_meetings\n",
      "Yes         949\n",
      "Unknown     247\n",
      "Na           20\n",
      "No            6\n",
      "yes           5\n",
      "Not sure      4\n",
      "Not Sure      1\n",
      "cant Say      1\n",
      "Name: question_no_unscheduled_meetings, dtype: int64\n",
      "-- question_follow_up\n",
      "Yes        951\n",
      "Unknown    247\n",
      "Na          20\n",
      "No          10\n",
      "yes          4\n",
      "No Dont      1\n",
      "Name: question_follow_up, dtype: int64\n",
      "-- question_alternative_number\n",
      "Yes                          936\n",
      "Unknown                      247\n",
      "No                            27\n",
      "Na                            19\n",
      "No I have only thi number      2\n",
      "na                             1\n",
      "yes                            1\n",
      "Name: question_alternative_number, dtype: int64\n",
      "-- question_resume_printout\n",
      "Yes                      940\n",
      "Unknown                  248\n",
      "Na                        19\n",
      "No                        16\n",
      "Not Yet                    4\n",
      "Not yet                    2\n",
      "yes                        2\n",
      "No- will take it soon      1\n",
      "na                         1\n",
      "Name: question_resume_printout, dtype: int64\n",
      "-- question_venue\n",
      "Yes                    946\n",
      "Unknown                248\n",
      "Na                      19\n",
      "No                      14\n",
      "yes                      2\n",
      "No- I need to check      2\n",
      "no                       1\n",
      "na                       1\n",
      "Name: question_venue, dtype: int64\n",
      "-- question_call_letter\n",
      "Yes               932\n",
      "Unknown           245\n",
      "Na                 19\n",
      "No                 17\n",
      "Not Sure            8\n",
      "Need To Check       3\n",
      "Not yet             2\n",
      "yes                 2\n",
      "no                  1\n",
      "Havent Checked      1\n",
      "na                  1\n",
      "Not sure            1\n",
      "Yet to Check        1\n",
      "Name: question_call_letter, dtype: int64\n",
      "-- Expected Attendance\n",
      "Yes          882\n",
      "Uncertain    250\n",
      "No            59\n",
      "NO            34\n",
      "10.30 Am       1\n",
      "yes            1\n",
      "11:00 AM       1\n",
      "Name: Expected Attendance, dtype: int64\n",
      "-- Observed Attendance\n",
      "Yes     701\n",
      "No      401\n",
      "yes      81\n",
      "NO       35\n",
      "no        7\n",
      "No        6\n",
      "yes       1\n",
      "no        1\n",
      "Name: Observed Attendance, dtype: int64\n",
      "-- Marital Status\n",
      "Single     767\n",
      "Married    466\n",
      "Name: Marital Status, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "for column in interviews.columns:\n",
    "    print(\"-- \" + column)\n",
    "    print(interviews[column].value_counts())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def parse_date(date_string):\n",
    "    try:\n",
    "        preprocessed = re.sub(r\"[-–]\", ' ', date_string)\n",
    "        preprocessed = re.sub(r\"\\s{2,}\", ' ', preprocessed)\n",
    "        preprocessed = re.sub(r\"(&.*)\", ' ', preprocessed)\n",
    "        return parse(preprocessed, ignoretz = True, dayfirst = True)\n",
    "    except (ValueError, TypeError) as e:\n",
    "        raise RuntimeError('Cannot be parsed as a date: ' + date_string) from e\n",
    "        \n",
    "        \n",
    "assert parse_date(\"06.02.2016\") == datetime.datetime(2016, 2, 6, 0, 0)\n",
    "assert parse_date(\"04/12/16\") == datetime.datetime(2016, 12, 4, 0, 0)\n",
    "assert parse_date(\"21.4.16\") == datetime.datetime(2016, 4, 21, 0, 0)\n",
    "assert parse_date(\"15 Apr 16\") == datetime.datetime(2016, 4, 15, 0, 0)\n",
    "assert parse_date(\"09-01-2016\") == datetime.datetime(2016, 1, 9, 0, 0)\n",
    "assert parse_date(\"09 01 2016\") == datetime.datetime(2016, 1, 9, 0, 0)\n",
    "assert parse_date(\"25 – Apr-16\") == datetime.datetime(2016, 4, 25, 0, 0)\n",
    "assert parse_date(\"21-Apr -16\") == datetime.datetime(2016, 4, 21, 0, 0)\n",
    "assert parse_date(\"28.8.2016 & 12.00 PM\") == datetime.datetime(2016, 8, 28, 0, 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class DateParser(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, attribute_name = \"Date of Interview\"):\n",
    "        self.attribute_name = attribute_name\n",
    "        self.__days_of_week = [\"Monday\", \"Tuesday\", \"Wednesday\", \"Thursday\", \"Friday\", \"Saturday\", \"Sunday\"]\n",
    "        self.__months = [\"January\", \"February\", \"March\", \"April\", \"May\", \"June\", \"July\", \"August\", \"September\", \"October\", \"November\", \"December\"]\n",
    "        \n",
    "        \n",
    "    def __create_date_columns(self, value):\n",
    "        date = parse_date(value)\n",
    "        return pd.Series({\n",
    "                \"interview_month\": self.__months[date.month - 1],\n",
    "                \"interview_day_of_week\": self.__days_of_week[date.weekday()]\n",
    "            })\n",
    "    \n",
    "    def fit(self, X, y = None):\n",
    "        return self\n",
    "    \n",
    "    def transform(self, X):\n",
    "        result = X[self.attribute_name].apply(\n",
    "            lambda date: self.__create_date_columns(date)\n",
    "        )\n",
    "        dataframe = pd.concat([result, X], axis=1)\n",
    "        return dataframe.drop([self.attribute_name], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def parse_city_name(city_name):\n",
    "        preprocessed = re.sub(r\"(\\W)\", '', city_name)\n",
    "        return preprocessed.title()\n",
    "\n",
    "class CityNameParser(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, attribute_name):\n",
    "        self.attribute_name = attribute_name\n",
    "    \n",
    "    def fit(self, X, y = None):\n",
    "        return self\n",
    "    \n",
    "    def transform(self, X):\n",
    "        result = X.copy()\n",
    "        result[self.attribute_name] = X[self.attribute_name].apply(\n",
    "            lambda name: parse_city_name(name)\n",
    "        )\n",
    "        return result\n",
    "    \n",
    "assert parse_city_name(\"Chennai\") == \"Chennai\"\n",
    "assert parse_city_name(\"- Cochin-\") == \"Cochin\"\n",
    "assert parse_city_name(\"CHENNAI\") == \"Chennai\"\n",
    "assert parse_city_name(\"chennai\") == \"Chennai\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class MergeITIndustries(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, attribute_name):\n",
    "        self.attribute_name = attribute_name\n",
    "    \n",
    "    def fit(self, X, y = None):\n",
    "        return self\n",
    "    \n",
    "    def transform(self, X):\n",
    "        result = X.copy()\n",
    "        result[self.attribute_name] = X[self.attribute_name].apply(\n",
    "            lambda name: \"IT\" if name.startswith(\"IT\") else name\n",
    "        )\n",
    "        return result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#sorry, just gut feeling, due to the lack of a domain expert\n",
    "class MergePositionType(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, attribute_name):\n",
    "        self.attribute_name = attribute_name\n",
    "        self.mapping = {\n",
    "            \"Routine\": \"Routine\",\n",
    "            \"Dot Net\": \"Routine\",\n",
    "            \"Selenium testing\": \"Routine\",\n",
    "            \"Niche\": \"Niche\",\n",
    "            \"Trade Finance\": \"Niche\",\n",
    "            \"AML\": \"Niche\",\n",
    "            \"Production- Sterile\": \"Niche\",\n",
    "        }\n",
    "    \n",
    "    def fit(self, X, y = None):\n",
    "        return self\n",
    "    \n",
    "    def transform(self, X):\n",
    "        result = X.copy()\n",
    "        result[self.attribute_name] = X[self.attribute_name].apply(\n",
    "            lambda name: self.mapping.get(name, \"Routine\")\n",
    "        )\n",
    "        return result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#sorry, just gut feeling, due to the lack of a domain expert\n",
    "class MergeSkillsetType(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, attribute_name):\n",
    "        self.attribute_name = attribute_name\n",
    "        \n",
    "    def __map_skill(self, skill):\n",
    "        name = skill.lower()\n",
    "        if \"java\" in name: #I checked the dataset, there were no JavaScript, so don't worry ;)\n",
    "            return \"Java Developer\"\n",
    "        elif \"testing\" in name:\n",
    "            return \"Tester\"\n",
    "        elif \"oracle\" in name:\n",
    "            return \"Oracle\"\n",
    "        elif \"aml\" in name or \"kyc\" in name or \"cdd\" in name or \"regulatory\" in name:\n",
    "            return \"Regulatory\"\n",
    "        elif \"sccm\" in name:\n",
    "            return \"IT Infrastructure\"\n",
    "        elif \"banking\" in name or \"lending\" in name or name == \"l & l\":\n",
    "            return \"Banking Operations\"\n",
    "        elif \"software engineer\" in name or \"developer\" in name:\n",
    "            return \"Other software engineer\"\n",
    "        elif \"lead\" in name or \"manager\" in name or \"management\" in name:\n",
    "            return \"Management\"\n",
    "        elif \"r & d\" in name or \"r&d\" in name:\n",
    "            return \"R&D\"\n",
    "        elif \"analyst\" in name:\n",
    "            return \"Analyst\"\n",
    "        elif \"publishing\" in name:\n",
    "            return \"Publishing\"\n",
    "        elif \"production\" in name:\n",
    "            return \"Production\"\n",
    "        elif \"fresher\" in name:\n",
    "            return \"Fresher\"\n",
    "        elif \"accounting operations\" in name:\n",
    "            return \"Accounting Operations\"\n",
    "        elif \"routine\" in name:\n",
    "            return \"Routine\"\n",
    "        elif \"sas\" in name:\n",
    "            return \"SAS\"\n",
    "        elif \"hadoop\" in name:\n",
    "            return \"Hadoop\"\n",
    "        elif \"dot net\" in name:\n",
    "            return \".NET\"\n",
    "        elif \"etl\" in name:\n",
    "            return \"ETL\"\n",
    "        elif \"emea\" in name:\n",
    "            return \"EMEA\"\n",
    "        elif \"global labelling\" in name:\n",
    "            return \"Global Labelling\"\n",
    "        elif \"biosimiliars\" in name:\n",
    "            return \"Biosimiliars\"\n",
    "        elif \"product control\" in name:\n",
    "            return \"Product Control\"\n",
    "        else:\n",
    "            return \"Other\"\n",
    "        \n",
    "    def fit(self, X, y = None):\n",
    "        return self\n",
    "    \n",
    "    def transform(self, X):\n",
    "        result = X.copy()\n",
    "        result[self.attribute_name] = X[self.attribute_name].apply(\n",
    "            lambda name: self.__map_skill(name)\n",
    "        )\n",
    "        return result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class ParseInterviewType(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, attribute_name):\n",
    "        self.attribute_name = attribute_name\n",
    "\n",
    "    def __convert_interview_type(self, name):\n",
    "        interview_type = name.lower()\n",
    "        if \"scheduled\" in interview_type or \"sceduled\" in interview_type:\n",
    "            return \"Scheduled\"\n",
    "        else:\n",
    "            return \"Walk-in\"\n",
    "        \n",
    "    def fit(self, X, y = None):\n",
    "        return self\n",
    "    \n",
    "    def transform(self, X):\n",
    "        result = X.copy()\n",
    "        result[self.attribute_name] = X[self.attribute_name].apply(\n",
    "            lambda name: self.__convert_interview_type(name)\n",
    "        )\n",
    "        return result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class ParseYesNoAnswer(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, attribute_name):\n",
    "        self.attribute_name = attribute_name\n",
    "\n",
    "    def __convert_answer(self, name):\n",
    "        name = str(name).lower()\n",
    "        if name == \"yes\":\n",
    "            return \"Yes\"\n",
    "        elif name == \"unknown\":\n",
    "            return \"Unknown\"\n",
    "        else:\n",
    "            return \"No\"\n",
    "            \n",
    "    def fit(self, X, y = None):\n",
    "        return self\n",
    "    \n",
    "    def transform(self, X):\n",
    "        result = X.copy()\n",
    "        result[self.attribute_name] = X[self.attribute_name].apply(\n",
    "            lambda name: self.__convert_answer(name)\n",
    "        )\n",
    "        return result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class ParseExpectedAttendanceAnswer(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, attribute_name):\n",
    "        self.attribute_name = attribute_name\n",
    "\n",
    "    def __convert_answer(self, name):\n",
    "        name = str(name).lower()\n",
    "        if name == \"yes\":\n",
    "            return \"Yes\"\n",
    "        elif name == \"uncertain\":\n",
    "            return \"Uncertain\"\n",
    "        elif name == \"no\":\n",
    "            return \"No\"\n",
    "        else:\n",
    "            return \"Yes\"\n",
    "            \n",
    "    def fit(self, X, y = None):\n",
    "        return self\n",
    "    \n",
    "    def transform(self, X):\n",
    "        result = X.copy()\n",
    "        result[self.attribute_name] = X[self.attribute_name].apply(\n",
    "            lambda name: self.__convert_answer(name)\n",
    "        )\n",
    "        return result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "class RelativeLocation(BaseEstimator, TransformerMixin):\n",
    "    def __create_columns(self, value):\n",
    "        return pd.Series({\n",
    "            \"current_same_as_location\": value[\"Candidate Current Location\"] == value[\"Location\"],\n",
    "            \"current_same_as_job_location\": value[\"Candidate Current Location\"] == value[\"Candidate Job Location\"],\n",
    "            \"current_same_as_venue_location\": value[\"Candidate Current Location\"] == value[\"Interview Venue\"],\n",
    "            \"native_same_as_location\": value[\"Candidate Native location\"] == value[\"Location\"],\n",
    "            \"native_same_as_job_location\": value[\"Candidate Native location\"] == value[\"Candidate Job Location\"],\n",
    "            \"native_same_as_venue_location\": value[\"Candidate Native location\"] == value[\"Interview Venue\"]\n",
    "        })\n",
    "    \n",
    "    def fit(self, X, y = None):\n",
    "        return self\n",
    "    \n",
    "    def transform(self, X):\n",
    "        result = X.apply(\n",
    "            lambda row: self.__create_columns(row), axis = 1\n",
    "        )\n",
    "        dataframe = pd.concat([result, X], axis=1)\n",
    "        return dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "preprocessing_pipeline = Pipeline([\n",
    "    (\"extract_interview_date\", DateParser()),\n",
    "    (\"normalize_location\", CityNameParser(\"Location\")),\n",
    "    (\"normalize_candidate_current_location\", CityNameParser(\"Candidate Current Location\")),\n",
    "    (\"normalize_candidate_job_location\", CityNameParser(\"Candidate Job Location\")),\n",
    "    (\"normalize_interview_venue\", CityNameParser(\"Interview Venue\")),\n",
    "    (\"normalize_candidate_native_location\", CityNameParser(\"Candidate Native location\")),\n",
    "    (\"merge_it_industries\", MergeITIndustries(\"Industry\")),\n",
    "    (\"merge_position_type\", MergePositionType(\"Position to be closed\")),\n",
    "    (\"merge_skillset_type\", MergeSkillsetType(\"Nature of Skillset\")),\n",
    "    (\"parse_interview_types\", ParseInterviewType(\"Interview Type\")),\n",
    "    (\"convert_permission_answer\", ParseYesNoAnswer(\"question_permission\")),\n",
    "    (\"convert_unscheduled_meetings_answer\", ParseYesNoAnswer(\"question_no_unscheduled_meetings\")),\n",
    "    (\"convert_premeeting_call_answer\", ParseYesNoAnswer(\"question_follow_up\")),\n",
    "    (\"convert_alternative_number_answer\", ParseYesNoAnswer(\"question_alternative_number\")),\n",
    "    (\"convert_printout_answer\", ParseYesNoAnswer(\"question_resume_printout\")),\n",
    "    (\"convert_venue_answer\", ParseYesNoAnswer(\"question_venue\")),\n",
    "    (\"convert_call_letter_answer\", ParseYesNoAnswer(\"question_call_letter\")),\n",
    "    (\"convert_expected_attendance_answer\", ParseExpectedAttendanceAnswer(\"Expected Attendance\")),\n",
    "    (\"convert_observed_attendance_answer\", ParseYesNoAnswer(\"Observed Attendance\")),\n",
    "    #(\"add_relations_between_locations\", RelativeLocation()) # results are getting worse ;)\n",
    "])\n",
    "\n",
    "preprocessed = preprocessing_pipeline.transform(interviews)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "split = StratifiedShuffleSplit(n_splits = 1, test_size = 0.2, random_state = 1234)\n",
    "for train_index, test_index in split.split(preprocessed, preprocessed[\"Observed Attendance\"]):\n",
    "    train_set = preprocessed.iloc[train_index]\n",
    "    test_set = preprocessed.iloc[test_index]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train_set_features = train_set.drop(\"Observed Attendance\", axis = 1)\n",
    "train_set_label = train_set[\"Observed Attendance\"].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Stolen from https://github.com/scikit-learn/scikit-learn/blob/19e2772/sklearn/preprocessing/data.py#L2871 ;)\n",
    "# it will be be available in the next version of scikitlearn\n",
    "class CategoricalEncoder(BaseEstimator, TransformerMixin):\n",
    "    \n",
    "    def __init__(self, encoding='onehot', categories='auto', dtype=np.float64,\n",
    "                 handle_unknown='error'):\n",
    "        self.encoding = encoding\n",
    "        self.categories = categories\n",
    "        self.dtype = dtype\n",
    "        self.handle_unknown = handle_unknown\n",
    "\n",
    "    def fit(self, X, y=None):\n",
    "        if self.encoding not in ['onehot', 'onehot-dense', 'ordinal']:\n",
    "            template = (\"encoding should be either 'onehot', 'onehot-dense' \"\n",
    "                        \"or 'ordinal', got %s\")\n",
    "            raise ValueError(template % self.handle_unknown)\n",
    "\n",
    "        if self.handle_unknown not in ['error', 'ignore']:\n",
    "            template = (\"handle_unknown should be either 'error' or \"\n",
    "                        \"'ignore', got %s\")\n",
    "            raise ValueError(template % self.handle_unknown)\n",
    "\n",
    "        if self.encoding == 'ordinal' and self.handle_unknown == 'ignore':\n",
    "            raise ValueError(\"handle_unknown='ignore' is not supported for\"\n",
    "                             \" encoding='ordinal'\")\n",
    "\n",
    "        if self.categories != 'auto':\n",
    "            for cats in self.categories:\n",
    "                if not np.all(np.sort(cats) == np.array(cats)):\n",
    "                    raise ValueError(\"Unsorted categories are not yet \"\n",
    "                                     \"supported\")\n",
    "\n",
    "        X_temp = check_array(X, dtype=None)\n",
    "        if not hasattr(X, 'dtype') and np.issubdtype(X_temp.dtype, np.str_):\n",
    "            X = check_array(X, dtype=np.object)\n",
    "        else:\n",
    "            X = X_temp\n",
    "\n",
    "        n_samples, n_features = X.shape\n",
    "\n",
    "        self._label_encoders_ = [LabelEncoder() for _ in range(n_features)]\n",
    "\n",
    "        for i in range(n_features):\n",
    "            le = self._label_encoders_[i]\n",
    "            Xi = X[:, i]\n",
    "            if self.categories == 'auto':\n",
    "                le.fit(Xi)\n",
    "            else:\n",
    "                if self.handle_unknown == 'error':\n",
    "                    valid_mask = np.in1d(Xi, self.categories[i])\n",
    "                    if not np.all(valid_mask):\n",
    "                        diff = np.unique(Xi[~valid_mask])\n",
    "                        msg = (\"Found unknown categories {0} in column {1}\"\n",
    "                               \" during fit\".format(diff, i))\n",
    "                        raise ValueError(msg)\n",
    "                le.classes_ = np.array(self.categories[i])\n",
    "\n",
    "        self.categories_ = [le.classes_ for le in self._label_encoders_]\n",
    "\n",
    "        return self\n",
    "\n",
    "    def transform(self, X):\n",
    "        X_temp = check_array(X, dtype=None)\n",
    "        if not hasattr(X, 'dtype') and np.issubdtype(X_temp.dtype, np.str_):\n",
    "            X = check_array(X, dtype=np.object)\n",
    "        else:\n",
    "            X = X_temp\n",
    "\n",
    "        n_samples, n_features = X.shape\n",
    "        X_int = np.zeros_like(X, dtype=np.int)\n",
    "        X_mask = np.ones_like(X, dtype=np.bool)\n",
    "\n",
    "        for i in range(n_features):\n",
    "            Xi = X[:, i]\n",
    "            valid_mask = np.in1d(Xi, self.categories_[i])\n",
    "\n",
    "            if not np.all(valid_mask):\n",
    "                if self.handle_unknown == 'error':\n",
    "                    diff = np.unique(X[~valid_mask, i])\n",
    "                    msg = (\"Found unknown categories {0} in column {1}\"\n",
    "                           \" during transform\".format(diff, i))\n",
    "                    raise ValueError(msg)\n",
    "                else:\n",
    "                    # Set the problematic rows to an acceptable value and\n",
    "                    # continue `The rows are marked `X_mask` and will be\n",
    "                    # removed later.\n",
    "                    X_mask[:, i] = valid_mask\n",
    "                    Xi = Xi.copy()\n",
    "                    Xi[~valid_mask] = self.categories_[i][0]\n",
    "            X_int[:, i] = self._label_encoders_[i].transform(Xi)\n",
    "\n",
    "        if self.encoding == 'ordinal':\n",
    "            return X_int.astype(self.dtype, copy=False)\n",
    "\n",
    "        mask = X_mask.ravel()\n",
    "        n_values = [cats.shape[0] for cats in self.categories_]\n",
    "        n_values = np.array([0] + n_values)\n",
    "        feature_indices = np.cumsum(n_values)\n",
    "\n",
    "        indices = (X_int + feature_indices[:-1]).ravel()[mask]\n",
    "        indptr = X_mask.sum(axis=1).cumsum()\n",
    "        indptr = np.insert(indptr, 0, 0)\n",
    "        data = np.ones(n_samples * n_features)[mask]\n",
    "\n",
    "        out = sparse.csr_matrix((data, indices, indptr),\n",
    "                                shape=(n_samples, feature_indices[-1]),\n",
    "                                dtype=self.dtype)\n",
    "        if self.encoding == 'onehot-dense':\n",
    "            return out.toarray()\n",
    "        else:\n",
    "            return out\n",
    "\n",
    "    def inverse_transform(self, X):\n",
    "        check_is_fitted(self, 'categories_')\n",
    "        X = check_array(X, accept_sparse='csr')\n",
    "\n",
    "        n_samples, _ = X.shape\n",
    "        n_features = len(self.categories_)\n",
    "        n_transformed_features = sum([len(cats) for cats in self.categories_])\n",
    "\n",
    "        # validate shape of passed X\n",
    "        msg = (\"Shape of the passed X data is not correct. Expected {0} \"\n",
    "               \"columns, got {1}.\")\n",
    "        if self.encoding == 'ordinal' and X.shape[1] != n_features:\n",
    "            raise ValueError(msg.format(n_features, X.shape[1]))\n",
    "        elif (self.encoding.startswith('onehot')\n",
    "                and X.shape[1] != n_transformed_features):\n",
    "            raise ValueError(msg.format(n_transformed_features, X.shape[1]))\n",
    "\n",
    "        # create resulting array of appropriate dtype\n",
    "        dt = np.find_common_type([cat.dtype for cat in self.categories_], [])\n",
    "        X_tr = np.empty((n_samples, n_features), dtype=dt)\n",
    "\n",
    "        if self.encoding == 'ordinal':\n",
    "            for i in range(n_features):\n",
    "                labels = X[:, i].astype('int64')\n",
    "                X_tr[:, i] = self.categories_[i][labels]\n",
    "\n",
    "        else:  # encoding == 'onehot' / 'onehot-dense'\n",
    "            j = 0\n",
    "            found_unknown = {}\n",
    "\n",
    "            for i in range(n_features):\n",
    "                n_categories = len(self.categories_[i])\n",
    "                sub = X[:, j:j + n_categories]\n",
    "\n",
    "                # for sparse X argmax returns 2D matrix, ensure 1D array\n",
    "                labels = np.asarray(_argmax(sub, axis=1)).flatten()\n",
    "                X_tr[:, i] = self.categories_[i][labels]\n",
    "\n",
    "                if self.handle_unknown == 'ignore':\n",
    "                    # ignored unknown categories: we have a row of all zero's\n",
    "                    unknown = np.asarray(sub.sum(axis=1) == 0).flatten()\n",
    "                    if unknown.any():\n",
    "                        found_unknown[i] = unknown\n",
    "\n",
    "                j += n_categories\n",
    "\n",
    "            # if ignored are found: potentially need to upcast result to\n",
    "            # insert None values\n",
    "            if found_unknown:\n",
    "                if X_tr.dtype != object:\n",
    "                    X_tr = X_tr.astype(object)\n",
    "\n",
    "                for idx, mask in found_unknown.items():\n",
    "                    X_tr[mask, idx] = None\n",
    "\n",
    "        return X_tr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "encoding = Pipeline([\n",
    "    (\"one_hot_encoder\", CategoricalEncoder())\n",
    "])\n",
    "\n",
    "encoded_train_set = encoding.fit_transform(train_set_features)\n",
    "\n",
    "label_encoder = LabelEncoder()\n",
    "encoded_train_set_labels = label_encoder.fit_transform(train_set_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_predict\n",
    "from sklearn.metrics import roc_auc_score, roc_curve\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "\n",
    "def score_classifiers(classifiers, input_data = encoded_train_set.toarray(), labels = encoded_train_set_labels):\n",
    "    roc_auc_results = []\n",
    "    for (classifier, name) in classifiers:\n",
    "        predictions = cross_val_predict(\n",
    "            classifier,\n",
    "            input_data, labels,\n",
    "            cv = 10\n",
    "        )\n",
    "        roc_auc = roc_auc_score(predictions, labels)\n",
    "        fpr, tpr, _ = roc_curve(predictions, labels)\n",
    "        roc_auc_results.append((roc_auc, name))\n",
    "        plt.plot(fpr, tpr, linewidth = 2, label = name)\n",
    "\n",
    "    plt.plot([0, 1], [0, 1], 'k--')\n",
    "    plt.axis([0, 1, 0, 1])\n",
    "    plt.title(\"ROC\")\n",
    "    plt.xlabel(\"False positive rate\")\n",
    "    plt.ylabel(\"True positive rate\")\n",
    "    plt.legend(bbox_to_anchor=(1.04,1), loc=\"upper left\")\n",
    "    plt.show()\n",
    "\n",
    "    for (roc_auc, name) in sorted(roc_auc_results, key=lambda x: x[0]):\n",
    "        print(\"ROC AUC (\" + name + \"): \" + str(roc_auc))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkYAAAEWCAYAAACDu1gGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a15a71eb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC (Naive Bayes): 0.575431034483\n",
      "ROC AUC (Multi-layer Perceptron): 0.617277344078\n",
      "ROC AUC (k-nearest Neighbors): 0.631152761854\n",
      "ROC AUC (Random Forest): 0.632190000252\n",
      "ROC AUC (Stochastic Gradient Descent): 0.647268700787\n",
      "ROC AUC (Support Vector Machine): 0.682321466412\n",
      "ROC AUC (Logistic Regression): 0.689316978795\n"
     ]
    }
   ],
   "source": [
    "score_classifiers([\n",
    "    (GaussianNB(), \"Naive Bayes\"),\n",
    "    (RandomForestClassifier(), \"Random Forest\"),\n",
    "    (LogisticRegression(), \"Logistic Regression\"),\n",
    "    (SVC(), \"Support Vector Machine\"),\n",
    "    (SGDClassifier(max_iter = 1000), \"Stochastic Gradient Descent\"),\n",
    "    (KNeighborsClassifier(), \"k-nearest Neighbors\"),\n",
    "    (MLPClassifier(), \"Multi-layer Perceptron\")\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[False False False False False False False False False False False False\n",
      " False False False False False False False False False False False False\n",
      " False False False  True False False False False False False False False\n",
      " False False False False False False False False False False False False\n",
      " False False False False False False False False False False False False\n",
      " False False False False False False False False False False False False\n",
      " False False False False False False False False False False False False\n",
      " False False False False False False False False False False False False\n",
      " False False False False False False False False False False False False\n",
      " False False False False False False False False False False False False\n",
      " False False False False False False False False False False False False\n",
      " False False False False False False  True False False False False False\n",
      " False False False False False False False False False False False  True\n",
      " False False False  True False  True  True False]\n",
      "[  8  57   3 110  23  26  15  14  58  97   5  20  39  31  28  61  60 124\n",
      "  83 148 107 157  81 131  41  65 103   1  92 122 114 159  24  99  71  46\n",
      "  67  51  35 143 145 105 152 111 134  40   7 138  33  69  68 121 142 144\n",
      " 123  72 130  80   2 113  64  29  48 155 104  87 108  21  56 106  45  38\n",
      "  10   4  13  59  12 140 135 117 100 150  54  50 129  96 146 126  95  37\n",
      "  53 156 118  88  74 112 158  94  32  73 153  16 149 115  77   6  66  27\n",
      "  47  25 128  78 101 102 109 136  98   9  75  49 141 127  36  34  79  84\n",
      "  89  19 137 125  86  42  76 132  62  93  30 154   1  22  44  85 119 151\n",
      "  43 116  63  91 133  52  55 120  82  70 147   1  90 139  17   1  18   1\n",
      "   1  11]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.feature_selection import RFECV\n",
    "\n",
    "feature_selection = RFECV(RandomForestClassifier(), step=1, cv=10)\n",
    "feature_selection = feature_selection.fit(encoded_train_set.toarray(), encoded_train_set_labels)\n",
    "\n",
    "print(feature_selection.support_)\n",
    "print(feature_selection.ranking_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "only_important_features = encoded_train_set.toarray()[:, feature_selection.support_]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkYAAAEWCAYAAACDu1gGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10abd8588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC (k-nearest Neighbors): 0.619961846004\n",
      "ROC AUC (Random Forest): 0.688537653528\n",
      "ROC AUC (Logistic Regression): 0.689904975134\n",
      "ROC AUC (Support Vector Machine): 0.702453796204\n",
      "ROC AUC (Stochastic Gradient Descent): 0.708684016242\n",
      "ROC AUC (Multi-layer Perceptron): 0.719141747692\n",
      "ROC AUC (Naive Bayes): 0.786661941048\n"
     ]
    }
   ],
   "source": [
    "score_classifiers([\n",
    "    (GaussianNB(), \"Naive Bayes\"),\n",
    "    (RandomForestClassifier(), \"Random Forest\"),\n",
    "    (LogisticRegression(), \"Logistic Regression\"),\n",
    "    (SVC(), \"Support Vector Machine\"),\n",
    "    (SGDClassifier(max_iter = 1000), \"Stochastic Gradient Descent\"),\n",
    "    (KNeighborsClassifier(), \"k-nearest Neighbors\"),\n",
    "    (MLPClassifier(), \"Multi-layer Perceptron\")\n",
    "], only_important_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'degree': 3, 'kernel': 'poly', 'shrinking': True}\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeUAAAEWCAYAAABYNo/VAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nwT/48AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10ab56eb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC (Best SVC): 0.736343119885\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from scipy.stats import randint\n",
    "\n",
    "param_grid = {\n",
    "       'kernel': [\"linear\", \"poly\", \"rbf\", \"sigmoid\"],\n",
    "       'degree': randint(1, only_important_features.shape[1]),\n",
    "       'shrinking': [True, False]\n",
    "  }\n",
    "\n",
    "random_search = RandomizedSearchCV(estimator=SVC(), param_distributions=param_grid, n_iter=1000)\n",
    "random_search.fit(only_important_features, encoded_train_set_labels)\n",
    "best_classifier = random_search.best_estimator_\n",
    "\n",
    "print(random_search.best_params_)\n",
    "score_classifiers([\n",
    "    (best_classifier, \"Best SVC\")\n",
    "], only_important_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'bootstrap': True, 'criterion': 'entropy', 'max_features': 5, 'min_samples_split': 7, 'n_estimators': 2}\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAh0AAAEWCAYAAADCTyW5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxE5GRAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10abfbc18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC (Best Random Forest): 0.689178334097\n"
     ]
    }
   ],
   "source": [
    "# limited feature set\n",
    "classifier = RandomForestClassifier()\n",
    "param_grid = {\n",
    "        'n_estimators': randint(1, 20),\n",
    "        'criterion': [\"gini\", \"entropy\"],\n",
    "        'max_features': randint(1, only_important_features.shape[1]),\n",
    "        'min_samples_split': randint(2, 8),\n",
    "        'bootstrap': [True, False]\n",
    "  }\n",
    "\n",
    "random_search = RandomizedSearchCV(estimator=classifier, param_distributions=param_grid, n_iter=1000)\n",
    "random_search.fit(only_important_features, encoded_train_set_labels)\n",
    "best_classifier = random_search.best_estimator_\n",
    "\n",
    "print(random_search.best_params_)\n",
    "score_classifiers([\n",
    "    (best_classifier, \"Best Random Forest\")\n",
    "], only_important_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'bootstrap': False, 'criterion': 'entropy', 'max_features': 1, 'min_samples_split': 6, 'n_estimators': 11}\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAh0AAAEWCAYAAADCTyW5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10ae69208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC (Best Random Forest): 0.6216875412\n"
     ]
    }
   ],
   "source": [
    "#full feature set\n",
    "classifier = RandomForestClassifier()\n",
    "param_grid = {\n",
    "        'n_estimators': randint(1, 20),\n",
    "        'criterion': [\"gini\", \"entropy\"],\n",
    "        'max_features': randint(1, encoded_train_set.toarray().shape[1]),\n",
    "        'min_samples_split': randint(2, 8),\n",
    "        'bootstrap': [True, False]\n",
    "  }\n",
    "\n",
    "random_search = RandomizedSearchCV(estimator=classifier, param_distributions=param_grid, n_iter=1000)\n",
    "random_search.fit(encoded_train_set.toarray(), encoded_train_set_labels)\n",
    "best_classifier = random_search.best_estimator_\n",
    "\n",
    "print(random_search.best_params_)\n",
    "score_classifiers([\n",
    "    (best_classifier, \"Best Random Forest\")\n",
    "], encoded_train_set.toarray())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjYAAAEWCAYAAAB18t2eAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nVs3+gKzogbwAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10aba9d30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC (AdaBoost Random Forest): 0.598898530175\n",
      "ROC AUC (AdaBoost Decision Tree): 0.620360644258\n",
      "ROC AUC (Voting): 0.659336956522\n"
     ]
    }
   ],
   "source": [
    "#full feature set\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "from sklearn.ensemble import VotingClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "voting_all = VotingClassifier(estimators=[\n",
    "    (\"Naive Bayes\", GaussianNB()),\n",
    "    (\"Random Forest\", RandomForestClassifier()),\n",
    "    (\"Logistic Regression\", LogisticRegression()),\n",
    "    (\"Support Vector Machine\", SVC()),\n",
    "    (\"Stochastic Gradient Descent\", SGDClassifier(max_iter = 1000)),\n",
    "    (\"k-nearest Neighbors\", KNeighborsClassifier()),\n",
    "    (\"Multi-layer Perceptron\", MLPClassifier())\n",
    "], voting='hard')\n",
    "\n",
    "score_classifiers([\n",
    "    (AdaBoostClassifier(base_estimator=RandomForestClassifier()), \"AdaBoost Random Forest\"),\n",
    "    (AdaBoostClassifier(base_estimator=DecisionTreeClassifier()), \"AdaBoost Decision Tree\"),\n",
    "    (voting_all, \"Voting\")\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjYAAAEWCAYAAAB18t2eAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzs3XdYlGf2N/DvPTMwhV4FhqqIigIW\nxACDZe0lmuhaorH3xJQ11p8mm7jGXbPGZLNvIklMk7gxUZMYE1vUqAEEA0pREQUUBOll6MOU+/1j\nAMHQVAYUz+e6vGSeuZ9nzoM4c7jbYZxzEEIIIYR0BYLODoAQQgghpL1QYkMIIYSQLoMSG0IIIYR0\nGZTYEEIIIaTLoMSGEEIIIV0GJTaEEEII6TIosSGEEEJIl0GJDSEtYIzdYoxVMcbKGWM5jLEvGWOm\nDZ4PYoydZoyVMcaUjLHDjDHve65hzhh7nzGWUXudlNrHth1/R4QQ0rVRYkNI657mnJsC6A9gAICN\nAMAYCwRwAsAhAE4APADEA4hgjHWvbWMM4BSAvgDGATAHEASgEEBAx94GIYR0fYx2HiakeYyxWwCW\ncM5P1j5+B0BfzvlExtjvABI55y/cc85RAPmc83mMsSUA3gbQg3Ne3sHhE0LIE4d6bAhpI8aYM4Dx\nAFIYYzLoe172N9H0OwCja78eBeAYJTWEENIxKLEhpHU/MsbKANwGkAfg7wCsof//k91E+2wAdfNn\nbJppQwghxAAosSGkdc9wzs0ADAfQG/qkpRiADoBjE+0dARTUfl3YTBtCCCEGQIkNIW3EOT8L4EsA\nOzjnFQDOA5jeRNMZ0E8YBoCTAMYyxkw6JEhCCHnCUWJDyP15H8Boxlh/ABsAzGeMvcwYM2OMWTHG\ntgIIBPBWbfsw6IewDjLGejPGBIwxG8bY/zHGJnTOLRBCSNdFiQ0h94Fzng9gD4DXOefhAMYCmAr9\nPJp06JeDKzjnN2rbq6CfQHwNwK8ASgFcgH44K7rDb4AQQro4Wu5NCCGEkC6DemwIIYQQ0mUYLLFh\njH3OGMtjjF1u5nnGGPugdnv5BMbYQEPFQgghhJAngyF7bL6Efgv55owH0LP2zzIAuwwYCyGEEEKe\nAAZLbDjn5wAUtdBkCoA9XC8KgCVjjPb7IIQQQsgDE3Xia8uhXwZbJ7P22J92aWWMLYO+VwcmJiaD\nevfu3SEBEkLI40THOUrLi1GpKkE1r4aKcagK1dBWaQEtCjjndp0dIyGG1pmJDWviWJNLtDjnnwD4\nBAD8/f15TEyMIeMihJDHglqrQ0xSPCLi9iCl7AJSjEqgNjKBhMsA6LvkBb+WwKzCFOGHktM7N1pC\nOkZnJjaZAFwaPHYGcKeTYiGEkEeeTseRlFmAyJhvcD3nODIEGbgu4dAwBsgAQABpQQ3ywgow8i9B\n+MeG9+C0wBMAwFhTv0sS0vV0ZmLzE4BVjLF9AIYAUHLOqVggIYTU4pwjvaAC0fHhuJr6He6oE5Ei\nrUCBSAjUFukQcKC32gi+st6ovmGN/7czDGq1GgErJ8HJ3rNzb4CQTmCwxIYx9g30RQNtGWOZ0FdE\nNgIAznkogCMAJgBIAVAJYKGhYiGEkMdFXmk1opJu4urlfcgu/x3Z4hwkSwTQSRggAQAhbLRAf6Ej\nQtxHYpT/AhTklGPp0qX47bdvMGLECHz66afo0aNHZ98KIZ3CYIkN5/y5Vp7nAF401OsTQsjjQFml\nRlRqPpISTiAr72eUiFKQLKtBsUgIWAKAEELO4aOVIdB2IMb0nw0vF0WjoaXfEs8jNjYWn3zyCZYs\nWWLQYafY2Fh7kUi0G0A/0CavpOPpAFzWaDRLBg0alNdUg84ciiKEkCdOtVqLP24V4dLVJGTf3I8K\nHotsWTGSxUJwq7qERAh7rQABMneM8pqIp/rOgonYvNF1Ll++jIsXL2LevHl45plnkJaWBhsbG4PH\nLxKJdjs4OPSxs7MrFggEVJOHdCidTsfy8/O9c3JydgOY3FQbSmwIIcSANFod4jOViL6ehazkI6iq\n/A0Vstu4LNOh1EpY20oEI87hyywxTB6IYf1mw8Per8mel5qaGmzbtg3btm1Dt27dMGPGDEgkkg5J\namr1o6SGdBaBQMDt7OyUOTk5/ZprQ4kNIYS0I52OIzm3DBE38pFyLRaqwp8BaRIyZGW4bmYEmAH6\n3S6EkOtEUFj1xlCvKfD3fBoyY5MWrx0dHY3FixfjypUreP755/Hee+9BIpF0xG01JKCkhnSm2p+/\nZodBKbEhhJCHlFFYiYjUAlxMvoWqjKOQGUVDaZqNeKkAFU51779GEHNgsHE3hLiGQNF3Dlyt2r5q\nKSsrCyEhIejWrRt+/vlnTJw40TA3Q8hjjhIbQgi5T3ll1TifWojIG7koTDkPS91vgGkKbsiqkeZs\nVNtK//bqwSQIsfVFsNdUDPIYBbFQfF+vdf36dXh5eUEul+Pbb7/FyJEjYW5u3vqJXdyePXss58+f\n3+PixYtXBgwYUN1Um2nTprlPmjRJuXDhwuLmrjNt2jT3qKgoMzMzM61KpWJTp04tevfdd9tt65Gw\nsDBLb2/v6kGDBv0pxtWrVzt9/fXXttbW1hq1Ws3WrVuXvXz58pZKEbVZW+79QQiFwkE9e/asqnt8\n6NChlF69etW052vU2bJli/3f/va3AjMzM939nEeJDSGEtKK0Wo2o1EJEphbi+o1rsCo7DUvTeJTI\n8pHgIEKV4G6vjIwzPGXijGC3kVD0ngknc+cHes2SkhKsW7cOu3fvxpkzZzB06FA8++yz7XdTj7l9\n+/ZZDxw4sDwsLMx6wIABD7W569atWzMXLlxYXFlZyby8vPotXbq0sHfv3u3yYf3jjz9aajQaZVOJ\nDQCsWLEid8uWLbmJiYniwMBA7wULFhSLxeJHdqhPLBbrrl27dvV+z1Or1TAyMmq9YQMff/xxt6VL\nlxZRYkMIIQ+pWq1FbHoxIlIK8EfKHUiyw+FmEg21yS2UWmiRaFv3Bm0MAPASmkJhPwiK3n9Ff3kw\njIT39wZ+r59++gkrV65ETk4O1q5di8GDBz/kHXUtSqVSEBMTY3ry5MnkKVOmeO7cufMOAOh0OixY\nsMA1IiLCzMXFRaXfVURvzZo1jseOHbNUqVQCf3//8r1796YLBI2naVRWVgoAoO6D9NChQ2YbNmxw\n0Wq18PPzq9yzZ0+6VCrlzR1/4YUX5MePH7cUCoV8+PDhpdOnTy8+efKkZVRUlNn27dsdDx48mNq3\nb19VU/fk4+OjkkgkuoKCAqFcLte8++67tl988YWdWq1m7u7uqgMHDtw0MzPTTZs2zd3MzEwbHx9v\nkp+fb/SPf/wjc+HChcUt3Xtz8crlcp9nn322KDw83Eyj0bDQ0ND0DRs2yNPT08UvvfRS7rp16/Lb\n8u9RWVnJ5s2b55aQkCATCoV45513bj/99NNlH3zwgc3Ro0ctVCqVoLKyUhAVFXX99ddf7/bDDz9Y\n19TUsIkTJ5a89957d0pLSwWTJ0/unp2dbazT6di6devu5ObmGuXl5RkNGzbMy8rKShMdHX29rT8f\nlNgQQp54Gq0OCVlKRKYUIDKlAMUZCegnOg8T0yuwNCnBJU9jxAkEqK2+BDMIEGjWHQqPcQju9Szs\nZfbtFsuSJUvw2WefwcfHB4cOHYK/v3+7Xbu9uW/4ZZAhrnvrXxNjW3p+7969lsOHD1f6+vqqLC0t\nteHh4TKFQlEZFhZmmZKSIk5OTr6SmZlp5OPj03fBggWFALB27dq8HTt2ZAPAM88847Fv3z6L2bNn\nKwFg8+bNztu3b3fMyMgQL1q0KE8ul2sqKyvZ8uXLPU6cOJHs6+urevbZZ93//e9/261Zsya/qePL\nly8vPHLkiFVaWtplgUCAgoICoa2trXbUqFElbRkSCg8Pl7m5uVXL5XINAMyZM6f4tddeKwCAl19+\n2emDDz6w3bRpUx4A5ObmGsXExFyLi4uTPPvss54LFy4sbu7em7uPN954Iw8AXFxcauLi4q4tXrzY\nZdGiRe7R0dHXqqqqBP369evbVGKjUqkEvXv39q49V/Xrr7+mbt++3R4Arl+/fvXSpUuSCRMm9ExN\nTb0MABcvXjRNSEi40q1bN+33339vnpKSIklISEjinGPUqFGeR48eNc3NzRU5ODioz5w5kwIAhYWF\nQhsbG+2uXbu6nT179rqjo6Om7T89lNgQQp5AnHNczy1HREoBIlMLkJSWAR9NDOSmf8DNJBNF7gzH\njOreHvWrjvoaWSHY8Skoev0VPg4DIRK039tn3W/XjDH4+/vDzc0N69evh7Gxcbu9Rlfy3XffWb/y\nyit5ADBt2rSisLAwa4VCUXn27FmzGTNmFIlEIri7u6sDAwPL6s45evSo2c6dOx2qq6sFJSUlIm9v\n7yoASuDuUJRSqRSEhIR4/frrryampqY6Z2dnla+vrwoAFixYUPjhhx/ajx49uqyp4xs3bswTi8W6\nWbNmuU2cOFE5c+ZMZVvuJTQ0tNuePXvsMjMzjQ8ePHij7nhsbKz0jTfekJeVlQkrKiqEw4YNq7/e\n5MmTS4RCIQYNGlRdWFhoBADN3Xt8fLykqXgB5AHAjBkzSgDAx8ensqKiQmBlZaWzsrLSicViXV1y\n1jDepoaiIiMjTV966aU8ABgwYEC1k5NTTWJiogQAQkJCSrt166YFgGPHjpmfO3fO3Nvb2xvQ95Bd\nu3ZNMnLkyLJNmza5rFy5Uj5lyhTluHHjytvyvWsOJTaEkCfC7aJKRKQUICK1EBdSciGvvAIf6QW4\nmyaDO1YgWiLWF5PUV36BJRMhyLI3FJ5PI8hjLGykhtkn5vbt21ixYgVmzZqFuXPnYsWKFQZ5HUNo\nrWfFEHJycoRRUVHm169fl65atQparZYxxviuXbsygaaLfVZWVrLXXnvNLTo6+qqnp6d69erVTtXV\n1X9aLmxhYaELDg4uO3v2rOmECRNKm3r9hkM8DRkZGSEuLi7pp59+Mt+3b5/Vrl277KOiolodPqmb\nY/PVV19ZLl261GP06NGJMpmML1u2zOPAgQMpgYGBVR988IHN2bNnzerOkUgk9UE0jKepe28u3nuv\nJRAIYGxsXN9YIBBArVa3aQvrll5DJpPpGrZ79dVXs9euXVtwb7uLFy9ePXjwoMWmTZvkJ0+eLK3r\nXXsQtB02IaRLyi9T4af4O9hwMAEh75zGrHe+Q8zhf8Ml/SX8xfIlKD1343u3q/jWRosLUgl0YPAT\n2+OFHtPwzfi9OPN8DLZP/gZPe882SFKj0+mwa9cu9O3bF2fOnIFK1eTUC3KPsLAwq6lTpxbeuXMn\nMSsrKzEnJyfB2dm55sSJE6bDhg0r279/v7VGo0F6erpRVFSUGXB37oyDg4NGqVQKDh8+bNXUtdVq\nNWJjY009PT1V/fv3r87KyjK+fPmyGAD27NljExISUtbccaVSKSgqKhLOnDlTGRoaejspKUkGAKam\nptrS0tJWP2vnz59f4uPjU/Hhhx/a1MXs6uqqVqlUbN++fdatnd/cvTcXb1u+122lUCjKv/76a2sA\nSEhIEGdnZxv7+vr+abL0+PHjS8PCwmyVSqUAAG7evGmUlZUlunXrlpGZmZnuhRdeKHr11Vdz4+Li\nZABgYmKirWt7P6jHhhDSJZRVqxGdVoSI1AJEphQiIzcfAYKr6CO9gIkmqUjyUOOkWAwtY6gbXrJh\nxgi28UGI1zN4ymU4LCWWHRLrjRs3sGTJEpw7dw6jRo3CJ598Ag8Pjw557cfd/v37bdatW9fot/kp\nU6YUh4WFWYeFhWWcOnXKvFevXn09PDyqAwICygDA1tZWO2fOnHxvb+++zs7ONX5+fhUNz6+bY6NW\nq5lCoSidN29eiUAgQGho6K3p06f3qJt0u2bNmnypVMqbOp6XlyeaNGmSp0qlYgCwdevW2wAwZ86c\nopUrV7qHhoZ2O3DgQLOThwHgzTffzJ47d2731atXF2zYsOFOQEBAH7lcXtOnT5/K8vJyYXPnAcDc\nuXNLmrp3mUzWZLwP9t1v2rp16/Lmzp3r5uXl5S0UCvHxxx/fkkqlf+rGmTp1aumVK1ckgwcP7l0b\nm27v3r03r127Jt64caOzQCCASCTiH330UToAzJ8/v2D8+PE97e3t1fczeZi11k31qPH39+cxMTGd\nHQYhpJNVq7W4mF6MyNRCRKQWICGzBD15BoaIYmBpmoBckyJEScUoFN39PBAC6C+TQ+E2EgrPp+Fl\n5QUB6/iO60OHDmHBggXYuXMnFixYYNCilXUYY7Gc84eeiRwfH3/Lz8/vT0MJhHSk+Ph4Wz8/P/em\nnqMeG0LIY0Gr40jMUtZP+I25VQyZpgQKQSKGyWIQZHcTcTKGH8TG4IwBkAEAugml+qXYPadgiDwY\nZsZmLb+QgcTHxyMuLg7z58/HlClTkJaWBiurJkdECCEPgRIbQsgjiXOOG3l1K5cKEZVWiKrqagxk\nN+BvHItA86vIMCnDeakEZ4RC1A0vGYFhoJk7QjzGIdh9DHpY9uiQHpHmqFQqbN26Ff/617/g6OiI\nmTNnQiKRUFJDiIFQYkMIeWRkFlciMkU/tBSZWoj8MhVcWC5CBHF4UXYJ5VZZuCA1Qpi4bhm0vmik\nXGQGhVMgFD0mIcBxCGRGss67iQbOnz+PxYsXIykpCfPmzcPOnTs7o2glIU8USmwIIZ2msFyFyNRC\nRKYWICKlEBlFlZChGoGCK5hvFAux7XUkyWpwRiJBmVCAukRGDAH8Lb0Q0n0Cgl1HwM3crVN7ZZqS\nlZWFYcOGwcHBAUeOHMH48eM7OyRCngiU2BBCOky5SoPoNH3NpYiUAlzLKQODDt4sA+MEcXC2jEee\nNB+RMmN8bGwM/VuU/m3K3dgKCuehUHQfj0HdBkEiejR7PpKSktCnTx/I5XJ89913GDlyJMzMOmde\nDyFPIkpsCCEGo9JocTG9pLZHpgDxmUpodRy2UEIhSMQs2UUw01TESjgOSSWoFAgAmAIApEyEITb9\noOg+HsHOQ+Fs9mDFJDtKcXExXnvtNXzxxRc4d+4cQkJC8Mwzz3R2WIQ8cSixIYS0G62O43KWEhGp\nBTifWog/bhWhWq2DETTwFyTjb6J4OJhfRqpxCSKkUpwyNgIgrT/fU2oPhctfoHAfhQH2A2AsfDxK\nCvzwww944YUXkJ+fj40bN1LRyg6wZ88ey/nz5/e4ePHilQEDBjRZOXvatGnurdVpmjZtmntUVJSZ\nmZmZVqVSsalTpxa9++67D7zr7b3CwsIsvb29q5uq7r169Wqnr7/+2tba2lpTVVUl6NWrV9U///nP\nrOYqgbfk1q1bRitWrHA5duxYWnNtBgwY0PvSpUvX7vfaDc2dO9f1jz/+MFWr1SwrK0vs7u5eDQDr\n16/Pbq0eVkehxIYQ8sA450jNL0dEin5oKSqtEKXVGgAc7iwHMwQJGGSZAKX4Fs5LRPhKIka1QADA\nHABgKjBGoN0ABHcfh2C5Ag4mDp15Ow9k0aJF+OKLL9C/f3/88ssvGDhwYGeH9ETYt2+f9cCBA8vD\nwsKsBwwYcOdhrlVXK6qyspJ5eXn1W7p0aWHv3r1r2iPOH3/80VKj0SibS1bqSioAwKeffmo1duzY\nXgkJCVecnJzuq/Cju7u7uqWkBgAeNqkBgLCwsAwASE5ONp40aVLPe+tG1VGr1TAyergq9w+KEhtC\nyH3JKqnSL8GuXYadV6bfSNUUlQgSXMEosyuQSS8j3qgKEVIJfjAyAnB3jkkfE2cEu42CwnU4fO18\nYSTonDe/h9GwaOVTTz2Fnj17Ys2aNZ32Rv6kUSqVgpiYGNOTJ08mT5kyxXPnzp13AH2ZigULFrhG\nRESYubi4qBpuQLtmzRrHY8eOWapUKoG/v3/53r170wWCxpsz1pVeMDMz0wHAoUOHzDZs2OBSt2Pv\nnj170qVSKW/u+AsvvCA/fvy4pVAo5MOHDy+dPn168cmTJy2joqLMtm/f7njw4MEWdx5eunRp8ZEj\nRyw+++wz69dffz3v999/l61evdqlsrJSYGVlpdm7d+8tNzc39eXLl8XLli1zKywsFAmFQr5///40\nkUjEJ02a1PPGjRtXYmJiJAsXLvRQq9VMp9Ph4MGDqT4+PiqZTDagsrLykk6nw8qVK51Pnz5twRjj\na9euzV66dGnxzz//bLZlyxYna2trdXJystTHx6fyxx9/vHnv96k5gwYN6hUUFFR+/vx508mTJxcv\nXLiwaPHixW537twxZozh/fffzxg5cmSFUqkULFq0yPX69etSjUbDXn/99ay6SuvtgRIbQkiLiipq\ncL52d9/IlALcKqwEADDo0I/dwnzZZXjLEpEuzESE1Bj/kkigZiLUJTMWQgmCHIYg2H00guXBsJXa\nduLdPLz09HQsX74cs2fPxrx587Bs2bLODqnzvGkxyDDXVbZYXHPv3r2Ww4cPV/r6+qosLS214eHh\nMoVCURkWFmaZkpIiTk5OvpKZmWnk4+PTd8GCBYUAsHbt2ry6worPPPOMx759+yzqPkzrSipkZGSI\nFy1alCeXyzWVlZVs+fLlHidOnEj29fVVPfvss+7//ve/7dasWZPf1PHly5cXHjlyxCotLe2yQCBA\nXWXsUaNGlbQ2HNbQgAEDKq9duyZRqVTs5Zdfdv3ll19SnJycNJ9++qnVmjVr5Pv37781e/ZsjzVr\n1uTMmzevpLKykmm1Wnbnzp36z/P//ve/di+88ELuypUri6qrq5lG07jzZ8+ePZaJiYnSpKSkK9nZ\n2aKAgIA+Y8aMKQeApKQkaVxcXJq7u7t60KBBvX/99VfTsWPHtrnadmlpqSAmJiYZACZOnNh9/fr1\nOSNHjqyo6+G5cePGlfXr1zuNHTtWefDgwVv5+fnCwYMH93nmmWdKZTJZu5RCoMSGENJIhUqDCzeL\n6ithJ2XfLXJsh2LMEV/BaLOrqGJXcMFYg0MyKUJFIgAWAAAGwMe8OxTuoxHsHIJ+Nv0gFLRY5uax\nUFe0csOGDeCcY/r06Z0d0hPru+++s37llVfyAGDatGlFYWFh1gqFovLs2bNmM2bMKBKJRHB3d1cH\nBgbWF3s8evSo2c6dOx2qq6sFJSUlIm9v7yoASuDuUJRSqRSEhIR4/frrryampqY6Z2dnla+vrwoA\nFixYUPjhhx/ajx49uqyp4xs3bswTi8W6WbNmuU2cOFE5c+bMB+qBqOtlSkhIEN+4cUP6l7/8xQvQ\n//zZ2dmpi4uLBbm5ucbz5s0rAfS1oAA0SggCAwMrduzY4ZiZmWk8a9asYh8fn0a9RL///nv998nF\nxUUzZMiQ8vDwcJmFhYXOx8enokePHmoA6Nu3b2Vqaup9TXSbM2dOUd3XERER5qmpqfXLF5VKpbC8\nvJydOXPG/PTp0+Y7d+50BACVSsVSUlKM676nD4sSG0KecCqNFpcySvT7yaQUIO52CTQ6/fukGDUY\nbnQD0yyuwU54CZdRgAipBH+TiKFhdyf9WotMECxXQOH6FwQ6BcJK0rV21U1OTsaSJUsQHh6OMWPG\n4OOPP4a7u3tnh9X5WulZMYScnBxhVFSU+fXr16WrVq2CVqtljDG+a9euTABN7mdUWVnJXnvtNbfo\n6Oirnp6e6tWrVztVV1f/aXzFwsJCFxwcXHb27FnTCRMmlP7pQribeNzLyMgIcXFxST/99JP5vn37\nrHbt2mUfFRXV5sKNdeLi4mSDBg2q5JwzT0/Pqri4uEbzYoqKilodF1qxYkVRSEhIxQ8//GAxfvx4\nr48++ujW5MmT65O8lmpEisXi+ieFQiE0Gs19bRBlamqqa/g6cXFxSRKJpNELcs7xww8/tDgs9zA6\nvvobIaRTaXUciZlKhJ5NxdzPouH31gnM+iQKH5y6gZj0IrgjE5ttzuAnx//gPbtX4NxtF96zOo+l\n1hr8x9oSMVIJOBNgoFUfvNR/FfZN2offZkdi2/AdmNB9QpdLagB9YnPlyhV8+eWXOHbsGCU1nSgs\nLMxq6tSphXfu3EnMyspKzMnJSXB2dq45ceKE6bBhw8r2799vrdFokJ6ebhQVFWUG3J074+DgoFEq\nlYLDhw83+UOqVqsRGxtr6unpqerfv391VlaW8eXLl8UAsGfPHpuQkJCy5o4rlUpBUVGRcObMmcrQ\n0NDbSUlJMgAwNTXVlpaWtumz9ssvv7T8/fffLRYtWlTk6+tbXVRUJDp58qQJoO/ViImJkVhbW+sc\nHBxqwsLCLAGgqqqKlZWVNbr+1atXjfv06aPavHlz3pgxY0ri4uKkDZ8fNmxY2YEDB6w1Gg3u3Lkj\nunDhgmlISEijiuftITg4uHT79u12dY8jIyOlADBixIjSHTt22Ncdj4iIkDZ1/oOiHhtCujj9yqWK\n+r1kotKKoKxS1z9vjgrMsU7FONkVCGpi8YegHGelUvxXbAwds6hvZ29sAYXLcAQ7D8UQxyGwEFs0\n9XJdxqVLlxAXF4eFCxdi8uTJSEtLg6WlZWeH9cTbv3+/zbp16xotx54yZUpxWFiYdVhYWMapU6fM\ne/Xq1dfDw6M6ICCgDABsbW21c+bMyff29u7r7Oxc4+fn1+hDvG6OjVqtZgqFonTevHklAoEAoaGh\nt6ZPn96jbpLwmjVr8qVSKW/qeF5enmjSpEmeKpWKAcDWrVtvA/qhmZUrV7qHhoZ2O3DgwJ96KUJD\nQ7t99913NlVVVQIvL6+q48ePJ9etiNq3b1/qyy+/7FpWVibUarVs5cqVuf7+/tVff/31zaVLl7r9\n4x//cDIyMuL79+9PFQrvDveGhYVZ79+/30YkEnE7Ozv1P//5z0arxubOnVsSGRlp2qdPn76MMf7W\nW29lurq6ahISEtrxXwrYvXt3xqJFi1y9vLxstVotCwoKKgsKCsp455137ixbtszFy8vLW6fTMTc3\nt+pTp06lttfrspa6pB5F/v7+PCYmprPDIOSRlq2sQkSKfmgpIrUAuaV330sF0GGURRb+apGMHtpY\nJKlSES4VI1IqQXGDN0cRE2CgrS+CXUdAIVegp2XPR65sgSFUV1djy5YteOeddyCXy5GcnNwl6jsx\nxmI55/4Pe534+Phbfn5+Be0REyEPKj4+3tbPz8+9qeeox4aQLqC4ogbn0/R7yZxPLURaQeNeZW+T\nMsyxS0UQi0NReSzOCzX4QiR1uYeeAAAgAElEQVTBFZkxuLl1fTsniS0UriMQLFdgiOMQmBiZdPSt\ndKqIiAgsXrwYycnJWLhwId59990ukdQQ8iShxIaQx1CFSoMLt4r0y7BTCnA1uxQNO19txDo873Ab\no8VXYFUejQs1dxChleAjqQSltneTFWMmgn+3gQh2HgaFswIe5h5PRK9MU7KysjBixAjI5XIcP34c\nY8aM6eyQCCEPgBIbQh4DNRod4m6X6DfGS9WvXFJr72YyxiKGKY6lmGKWBG9VDFJL4hFRI8CbQimS\nLYwB2NS3dZU51vbKBGOww2BIRe06b++xc/XqVXh7e0Mul+PgwYMYMWIETE1NOzssQsgDosSGkEeQ\nTsdxNbu0NpHR11yqrNHWPy9gQLBciBnWKXhKFwdtQQQiqpU4wKSIkkpQYX930YdUYIzBDgFQuAxF\nsFMwXM1dO+OWHjlFRUVYvXo1vvrqK5w9exZDhw7F008/3dlhEUIeEiU2hDwCOOe4WVCBiNq9ZM6n\nFaKkUt2oTS87KWY45mG4KAGOxecRX5KECC7GF1IJUq0b98r0MHNDsMswKOQKDOw2EGKhuIPv6NF2\n8OBBvPjiiygsLMSmTZsQEBDQ2SERQtoJJTaEdJIcZXXt7r76Cb/ZysY18uSWUkx0VWOC7Cp6V8Qg\nPysC4dlqvCeT4oJEjCqH+u0hIBNK8JRTIBTOIQh2CoaTqVNH385jY8GCBfjqq68wcOBAHDt2DP37\n9+/skAgh7YgSG0I6SEllDaLSCvWVsFMLkJbfeOWSjYkxhnnIMNniJgZqLkKUeRaxtzNxVCrFZpkE\n6d3MGrXvZdkTwc4hUMgV6G/XH0ZCKsDYnIZFK4OCgtCnTx+89tprEInoLfBxExAQ0Gv9+vXZ06ZN\nq98ZeMuWLfbXr1+XfP311xn3tk9OTjb+7bffTFesWFEEAOfOnZN9/vnnNl9++eXtjoybdByD/q9m\njI0D8B8AQgC7Oef/uud5VwBfAbCsbbOBc37EkDER0lEqazT441Zx/V4yV+40XrlkYizEEA9rTOpW\nhGCWALu833Hr9h+IyBPhG6kEMaYS1JjXb84JMyMTBMkVCHYKRrA8GPYy+yZeldzr5s2bWLZsGZ5/\n/nnMnz//yS5a2QVMnz698JtvvrFumNgcPHjQevv27ZlNtb9x44b422+/ta5LbIYOHVo5dOjQyo6K\nl3Q8gyU2jDEhgA8BjAaQCeAPxthPnPOrDZptBvAd53wXY8wbwBEA7oaKiRBDUmv1K5cia3tkLmUU\nN165JBRgoJslRrqKMFJ8Be7FUai6+Rsu5CvxiVSCCKkUWU52ja7Z19obitpemX62/SASUA9DW2m1\nWnz44YfYuHEjBAIB5syZ09khkXYwd+7c4m3btsmrqqqYVCrlycnJxnl5eUZjxowpX758ufPp06ct\nGGN87dq12UuXLi3etGmTPC0tTdK7d2/v5557rmDQoEFV7777brfffvstZfXq1U63b982Tk9PF9+5\nc8d4xYoVuZs3b84DgLVr1zoeOHDA2tHRscbGxkYzYMCAyi1btuR29v2T1hnyXTIAQArnPA0AGGP7\nAEwB0DCx4QDMa7+2ANBo22dCHmU6HUdSTml9InPhZuOVS4wBvs4WUHhYYqxlBrwr/oDo1inciL2K\nszIx3pZKcdFGDA27m8xYGVsgyFnfKxPkFAQbqU1TL01akZSUhMWLF+P8+fMYP348QkND4epKq8Ha\nm89XPoMMcd3E+YnNFtd0cHDQ+vn5VRw8eNDi+eefL/nqq6+sJ0+eXLxnzx7LxMREaVJS0pXs7GxR\nQEBAnzFjxpS//fbbWXWJDAD8/PPPjcZ0U1JSJJGRkcklJSXCPn369Fu7dm1+dHS09PDhw1aJiYlX\n1Wo169+/v/eAAQOol+cxYcjERg6g4RhmJoAh97R5E8AJxthLAEwAjGrqQoyxZQCWAaA3J9JpOOe4\nVVhZv7vv+bRCFFXUNGrjaW+K4B42GOlQhcGai5Bm7EVpwu+IEqrxvUyKcKkEec4O9e0FYPCz84VC\nroBCroC3jTcEjGrTPqyUlBQkJycjLCwMc+bMeWI3HeyqZsyYUfTtt99aPf/88yXff/+99e7du299\n8cUXNjNmzCgSiURwcXHRDBkypDw8PFxmYWGha+laY8aMKZFKpVwqlWqsra3VmZmZojNnzpiOHz++\nxNTUlAPgo0ePLumgWyPtwJCJTVPvJPcWpnoOwJec83cZY4EAwhhj/TjnjX4QOeefAPgE0NeKMki0\nhDQht7S6tnikfhn2nXtWLjlZSBDkaYuh7hKEiK7B6s5h6FJP4tqVTIRJpYiQSRDvaAFtgw9WW4kN\ngmsTmUCnwC5fTLKjxMbGIj4+HosWLcLTTz+NmzdvwtzcvPUTyQNrqWfFkObMmVOyefNml/DwcFl1\ndbVAoVBUfv755w/UvSkWi+s/U4RCITQazWNXQ5E0ZsjEJhOAS4PHzvjzUNNiAOMAgHN+njEmAWAL\nIM+AcRHSLGWlGufTCusrYafes3LJSmaEoB62COphhREWuXAsiARLPY3i49E4LxYhQipBhIkUhRaO\n9eeImBD+9gMQLA+GQq5AL6te1IPQjqqqqvDWW29hx44dcHFxwezZsyGRSCip6cIsLCx0Tz31VNmS\nJUvcp06dWgQAw4YNK/v000/tVq1aVZiXlye6cOGC6QcffHA7PT3duLy8XNjaNRsaPnx4+cqVK90q\nKyuz1Wo1O3nypOW8efPyDXM3pL0ZMrH5A0BPxpgHgCwAswDMvqdNBoCRAL5kjPUBIAFAPzykw1TV\naPHHrSJEpuqTmctZSuga/LImq125FNTDFiFOOniVx0CQdgDa30/jikaJH6VSREglSHTuBt4gWXGQ\nOSBYHowQeQgCHANgZmzWxKuTh3Xu3DksWbIEN27cwOLFi7Fjxw4qWvmEmDVrVtH8+fN7fPPNN2kA\nMHfu3JLIyEjTPn369GWM8bfeeivT1dVV061bN61IJOK9evXynj17dsGgQYOqWrv2sGHDKseNG6f0\n9vbuK5fLVb6+vhUWFhba1s4jjwaDdrkxxiYAeB/6pdyfc87fZoxtARDDOf+pdiXUpwBMoR+mWsc5\nP9HSNf39/XlMTIzBYiZdm1qrQ0JmiX4vmZQCXMooQY327sinkZBhgKsVgnvYQuFhBj8kQ5R2Gkg9\nhYK8K4isnSdzXipBifDuL4FGAiMM6jaofq5Md4vu1CtjYFlZWXB3d4eLiws+/fRTjBw5srNDeqQx\nxmI55/4Pe534+Phbfn5+Be0R06NMqVQKLCwsdGVlZYLAwMBeoaGh6QqFgiYQPyLi4+Nt/fz83Jt6\nzqBrR2v3pDlyz7E3Gnx9FUCwIWMgTzadjuNaThkiU/U1l6LTClFxz8olH7kFgjxtENzdBoMtiiFN\nPwuknoIm+nfECzSIkEkQLpUiyc250bWdTZ3rE5nBDoMhM5J19O09kRITE+Hj4wO5XI4ffvgBI0aM\ngImJSesnEnIfnn/+ebcbN25IVSoVmzVrViElNY8P2hSDdCmcc2QUVdbv7huVWojCe1Yu9bAzQVAP\nWwR72uApuTEsc84DqQeBo6eQU5aJo1IJwmVSRDlaoUx4d4WSWCjGYIfB9cmMq5kr9cp0oIKCAvzt\nb3/D119/XV+0ctKkSZ0dFumiDh8+fLOzYyAPhhIb8tjLK61GZGphfSXsrJLGQ+iOFpL6RCaouzUc\nKq4BqUeBC6ehzryAaGMhIqRS/G4iQYqVvNG57ubu9YnMoG6DIBHR/I2OxjnH/v37sWrVKhQXF+Pv\nf/87hgy5d+cI0oF0Op2OCQQCWjpEOoVOp2MAml3GT4kNeewoq9SITiusT2Zu5JU3et5SZoTA7jYI\n8rRFcA8beIjLwNJ+A1JOASd/Q1ZNCcKl+rky0S4OqBLc7ZWRiqQY4jgEIfIQBDkFwdnM+d6XJx1s\n/vz5CAsLg7+/P06dOgUfH5/ODulJdzk/P9/bzs5OSckN6Wg6nY7l5+dbALjcXBtKbMgjr1qtRcyt\nYkSkFiAypQCJ96xckhoJEeBhre+R6WELbzsxBLfPA6k/APtPQ5V/BTESCcKlEoRbSXHLuHGvTE+r\nnlA4KRAsD8ZA+4FUTPIR0LBo5bBhw+Dr64tXX32VilY+AjQazZKcnJzdOTk5/QDQbpKko+kAXNZo\nNEuaa/DYbUREq6K6Po1Wh/hMJc7XbowXm1GMGs09K5dcrBBUm8j0d7aAcUkqkHoKSDkFfiscGVAj\nvHbSb4xEgmrB3bkwpkamCHQKhEKuQJBTEBxMHJoKg3SStLQ0LF26FM8//zwWLlzY2eF0Ge21KoqQ\nRx39+kM6Heccybll9bv7Rt8sQrlKU/88Y0BfJ3MEe9oiqIcNAjysIdOWAzfPAomngB9Po7I0E39I\nxPhdJkWEgyUyjRr3uvSx7gOFXN8r42vnCyMB9co8arRaLf773/9i06ZNEAqFmDdvXmeHRAh5DFFi\nQzpFRmElImp39z3fxMql7rYm+iXYPWzxVHcbWEmFwJ1LQMpPQPgp8MwYpIoEiJBKEC6TINbKGeoG\nK5QsxBYIcgqq75Wxldp29C2S+3D16lUsWrQI0dHRmDhxIkJDQ+HsTPObCCH3jxIb0iHyy1T6vWRq\nl2FnFjdeudTNXIzgHrb6Cb+eNnC0kALKLCD1JHDkFJB2BuXVJYiWSvC7VIoI527IaTDfgoHB19an\nvmxBX5u+EAruaxd10olu3ryJ1NRU/O9//8OsWbNoGT0h5IFRYkMMorRajei0ovoemeTcskbPW0j1\nK5eCPfWrl7rbmoBpqoH0COC8fqdfnn8NycZGCJdKEWEhQVw3F2gafN5ZS6z1w0tOwQhyCoKlxLKD\n75I8jD/++ANxcXFYunQpJk6ciLS0NJiZUekJQsjDocSGtItqtRax6cX1lbATMkv+tHJpsIc1gnro\nh5e8ncwhZADyrwE3vgSOnQLSI6HU1eC8VL+CKcLVGQUNNsgTMiEG2vnVz5Xpbd0bAkaLMh43lZWV\neOONN/Dee+/Bzc0Nc+fOhUQioaSGENIuKLEhD0Sj1SExS1m/l0xMeuOVSyIBw0BXy/q9ZPq7WkIs\nEgKVRUDaKSD2FJD6G3SlWbhqbKxfwWRngUSxGLoGvTL2Mvv6XpmnnJ6CuTFVbH6cnTlzBkuWLEFq\naiqWL1+O7du3U9FKQki7osSGtAnnHNdzy+t3941OK0RZg5VLAODtaF4/tDTY3RqmYhGg1QBZscDv\n+qXYuHMRRQyIkEoQIZMi0s0FxQ2WYosEIvjbD6zvlelp2ZPmW3QRmZmZGD16NNzc3HD69GmMGDGi\ns0MihHRBlNiQZt0uqqwfWopMLURBuarR8+42stoeGVsE9rCBtYmx/omSDCAxDEg9DaSdhUalxGWx\nsX7Sr6M9rhobgzfIVZxMnOrLFgQ4BsDEiAoadiXx8fHw8/ODs7MzDh06hOHDh0Mmo4KhhBDDoMSG\n1CurVuO35HxE1vbKZBQ1LmZrbyau30smyNMWckup/omaSuDWmfoN8lB4A/lCgX7Sr7kE52WuKG0w\nFcZYYIzBDoMRLA9GsDwYHuYe1CvTBeXn5+OVV17BN998gzNnzmDYsGGYMGFCZ4dFCOniKLEhAPST\nf6d8GIG0/Ir6Y+YSEQJ72NQXkOxhZ/rnBKQ0G9g9EijNwiWxMc7IpIiQOyHZuPGPlpu5G4Kd9Eux\n/R38IRVJO+K2SCfgnGPfvn14+eWXoVQq8dZbbyEwMLCzwyKEPCEosSEAgE/OpSEtvwLOVlLMGeKG\nYE8b9HWygFDQQk8K58DhV4DSLHzl4I4d0ruTh6UiKQIcAvT7yjgp4GLu0gF3QR4Fc+fOxd69ezFk\nyBB89tln6Nu3b2eHRAh5glBiQ5CtrMKuM6kAgH//1Q+BPWzadmL8PuDGcVw3scJ/TISATofZvWdj\nuMtwDOw2EGKh2IBRk0eJTqcDYwyMMYwYMQKDBg3Cyy+/DKGQNkkkhHSsVhMbxpgUwKsA3DjnKxhj\nngB6cs6PGjw60iG2H72GKrUW4/s5tD2pKc0Gjq2HGsAmV0+oq3Mx3Ws6Ng7ZaNBYyaMnJSUFS5cu\nxdy5c7Fo0SIsXry4s0MihDzB2rK72ecAGABF7eM7ALYZLCLSoS5mFOPHuDswFgnwfxP6tO0kzoGf\nXwWqlfi4+0Bcq86F3FSO1/xfM2yw5JGi0WiwY8cO+Pj44NKlSzA2Nu7skAghpE1DUT05588xxqYD\nAOe8ktESli5Bp+N46/BVAMAShQdcrNu4BDfhW+D6MVw2tcZuFIGBYWvwVlqm/QS5fPkyFi5ciJiY\nGEyZMgUfffQRnJycOjssQghpU2JTwxiTAOAAwBjzAFDT8inkcfBjXBbib5fA3kyMF0Z4tu2k0mzg\n6DpUM4b/c3aDVlWIed7z4O/gb9hgySMlIyMD6enp2LdvH2bMmEHL9Qkhj4y2JDb/AHAMgDNj7CsA\nwwAsMWhUxOAqVBr86+g1AMC6cb31uwS3psEQ1H+7++GmqhAeFh54acBLBo6WPAqio6MRHx+PZcuW\nYcKECUhLS4OpqWlnh0UIIY20OsemdpLwdABLAfwAIIBzftLQgRHD2nUmFXllKvg5W2DqAHnbTqod\ngooxs0YYL4GQCbFNsQ0SEdX66coqKiqwevVqBAYG4p133oFKpd+BmpIaQsijqNXEhjF2gnOezzk/\nxDn/kXOexxg70RHBEcO4XVSJT35PAwC88bQ3BC3tVVOnLAc4ug6VjGGzkxwcHEt8lqCfbT8DR0s6\n0+nTp+Hr64v33nsPK1aswMWLFyEW0zJ+Qsijq9nxB8aYMQAJgG6MMTPoV0YBgDkA1w6IjRjIv45e\nQ41Ghyn9nTDIzbr1EzgHDuuHoHZ090VWTQl6W/fGct/lhg+WdJrMzEyMHTsWHh4eOHv2LIYOHdrZ\nIRFCSKtamljxIoDVAOwBXMHdxKYUQKiB4yIGEpVWiF8SsyExEmD9uN5tOynhO+D6UUSY22A/L4FI\nIMLbirdhJDQybLCkU1y6dAkDBgyAs7MzDh8+jGHDhkEqpRIYhJDHQ7NDUZzz9zjnLgDWc85dOecu\ntX/6cs7f78AYSTvR6ji21C7vXjGsB5ws2/BhVTsEpRQwvOHQDQDwYv8X4WXlZchQSSfIzc3FzJkz\nMXDgQJw9exYAMG7cOEpqCCGPlVaXwnDO32eM9QbgDf3QVN3x/xkyMNL+9sfcxtXsUjhZSLB8aI/W\nT6gfgirB9u79kKcuha+dLxb2XWj4YEmH4Zxj7969eOWVV1BeXo6tW7ciKCios8MihJAH0paSCpsB\njAHQG8BxAGMBhAOgxOYxUlqtxo4TyQCA9eN7Q2rchho+tUNQpyxscJiXQiKU4O3gtyEUUP2frmT2\n7NnYt28fAgMD8dlnn6FPnzbuQE0IIY+gtuxjMxNAfwAXOedzGWOOAD42bFikvf2/0ykoKK+Bv5sV\nJvu1YYfY2iGoQoEAW+zsAG0lXh30Ktwt3A0eKzG8hkUrx4wZg8DAQLz44otUtJIQ8thrS62oKs65\nFoCmdnVUDoDuhg2LtKebBRX4IuImAP3y7lZ3ieUc+Plv4NUl2OrWG0XaSgxxGILnej/XAdESQ7t+\n/TpGjBiBzz//HACwcOFCqsRNCOky2pLYXGKMWUJfDDMGwAUAFw0aFWlXb/+SBLWW46+DnOHrbNn6\nCYn7geQj+MXSBidRDhMjE2wJ3gIBa8uPC3lUaTQavPPOO/Dz80NCQgJNCiaEdEktDkXVFrt8k3Ne\nAuBDxthxAOacc0psHhPhNwpwMikXJsZCrBvbq/UTynKAI2uRKxRim401oFNh/eD1cDKlAoePs4SE\nBCxatAixsbF49tln8eGHH8LR0bGzwyKEkHbXYmLDOeeMsZ8BDKp9nNIhUZF2odHqsOXnKwCAF//i\nCXvzVkofNBiC+rtHb5TpKjHUeSie8XymA6IlhpSZmYnbt29j//79mDZtGhWtJIR0WW0ZW7jAGBv4\nIBdnjI1jjCUzxlIYYxuaaTODMXaVMXaFMUYrrdrR/y5k4HpuOVyspVgU7NH6CbVDUAesbBGBSliI\nLfBm4Jv0IfiYioyMRGiofi/NuqKVf/3rX+nfkxDSpbUlsVFAn9wkM8YuMsYuMcZaHYpijAkBfAhg\nPPR74DzHGPO+p01PABsBBHPO+wJ49b7vgDSppLIGO3+9DgDYNKEPJEatTAwtywWOrMVtkRD/trYA\nAGweshl2MjtDh0raWXl5OV555RUoFAq8++679UUrTUxMOjkyQggxvLYs937QcYgAACmc8zQAYIzt\nAzAFwNUGbZYC+JBzXgwAnPO8B3wtco/3T95ASaUaT3W3xti+Di03rh2C0lWXYLNHL1TpqjDWfSzG\neYzrmGBJuzlx4gSWLVuGjIwMvPjii9i2bRsVrSSEPFHasvNw6gNeWw7gdoPHmQCG3NPGCwAYYxEA\nhNBPVD5274UYY8sALAMAV1eqv9maG7llCItKh4ABb0zq2/rQQ+IBIPkXfG1th4uogo3EBpuGbOqY\nYEm7uX37NiZOnIgePXrg3LlzUCgUnR0SIYR0OEOu323q05Tf81gEoCeA4QCeA7C7dml545M4/4Rz\n7s8597ezo6GRlnDO8Y9fkqDVccwKcIW3k3nLJ5TlAkfXIs1IhP9YmgIA3gx6E1YSqw6IlrSH2NhY\nAICLiwuOHDmCuLg4SmoIIU8sQyY2mQBcGjx2BnCniTaHOOdqzvlNAMnQJzrkAf2WnIdz1/NhJhHh\ntdGtFKqsHYLSVBVjk7MHargWz3g+g+EuwzskVvJwcnJyMH36dPj7+9cXrRw9ejQkklZWvxFCSBfW\npsSGMebMGBtR+7WYMdaWWYh/AOjJGPNgjBkDmAXgp3va/Aig7rq20A9NpbU1eNJYjUaHrT8nAQBe\nGdkTNqatzK24fBBI/gWf2djhMlRwMHHAusHrOiBS8jA45/jqq6/g7e2Nw4cPY9u2bVS0khBCarWa\n2DDGFkGfkOyuPeQG4FBr53HONQBWQV84MwnAd5zzK4yxLYyxybXNjgMoZIxdBfAbgLWc88L7vw0C\nAHvO30JaQQW625pgXqB7y43LcoEja3DN2Aih5vo89R/B/4CZsZnB4yQPZ9asWViwYAG8vb0RFxeH\njRs3wsjIqLPDIoSQR0JbVkW9DP0Kp2gA4JxfZ4zZt+XinPMjAI7cc+yNBl9zAKtr/5CHUFiuwn9O\n3QAAbJ7UB8aiFnJWzoFfVqOmqhj/5+EJDWrwXO/n8JTjUx0ULblfDYtWTpgwASEhIXjhhRcgEFCZ\nC0IIaagt74rVnPOauge1+9PQDl+PmHd/vY6yag2GetlhRK9W8s7LB4FrP+MjWzvcQA1czVzx6kDa\nQuhRde3aNQwdOhSfffYZAGD+/PlYtWoVJTWEENKEtrwzRjDG1gGQ1M6z+RbAz4YNi9yPpOxS7LuQ\nAaGA4fWJfVpe3l2eBxxZgzixMb4wk0HABHhb8TZkRrKOC5i0iVqtxrZt2+Dn54erV6/C1NS0s0Mi\nhJBHXlsSm3UAygBcA/AKgFMAaJOTRwTnHFsOX4WOA3OfckPPbi3MkaldBVVVXYLNTi7QgWN+3/no\nb9+/4wImbRIXF4eAgABs2rQJkydPxtWrVzFr1qzODosQQh55bZljMwHAbs75LkMHQ+7f8Ss5OJ9W\nCEuZEV4d1cpK+dohqP/Y2SMdanhaemJV/1UdEyi5Lzk5OcjJycHBgwcxderUzg6HEEIeG23psZkB\nIIUx9gVjbGztHBvyCKhWa/H2Ef3y7tWjvWApM26+cXkecGQtoiVi7DWVQMREeFvxNoyFLZxDOlR4\neDg++ugjAMC4ceOQmppKSQ0hhNynVhMbzvlc6PeXOQxgEYA0xliooQMjrfs84iZuF1XBq5spZge0\nUGqidgiqvLoYrzvKAQDL/JbB28a7+XNIhykrK8OqVasQEhKC999/v75opUxG854IIeR+tWlZBedc\nBf3eNV9Cv/HeDAPGRNogr7QaH55OAQC8PskbImEL/5S1Q1D/trNHNjTwtvHGEp8lHRQpacnx48fR\nr18/fPTRR3jllVdw8eJFKlpJCCEPodU5NoyxUdDvGjwKQASAPQBmGzgu0op/H09GRY0Wo/p0Q0jP\nFupn1Q5BnZNK8L2JGMYCY2xTbIORgDZ062y3b9/GpEmT4OnpifDwcNo9mBBC2kFbJg+vALAPwEuc\n8yoDx0PaICGzBPtjM2EkZNg0sU/zDWs34itRleDvbm4AtHh54MvoYdmjw2IljXHO8ccffyAgIAAu\nLi44evQoFAoF1XcihJB20pY5Nn/lnB+gpObRULe8GwAWBnvAw7aFsl1XvgeSDmObnT0KoMVA+4F4\nvs/zHRQpuVd2djamTZuGIUOG1BetHDVqFCU1hBDSjppNbBhjZ2v/LmaMFTX4U8wYK+q4EElDhxOy\nEZNeDBsTY6z6i2fzDcvzgF/W4JiJDEdlxpCKpNgavBVCAS1q62icc3zxxRfw9vbG0aNHsX37dgQH\nB3d2WIQQ0iW1NBQ1ovZv244IhLSuqkaLf9Uu7147thfMJc3Mk6kdgiqoKcHbrq4AdFjjvwYu5i4d\nFyypN2PGDBw4cAAhISHYvXs3vLy8OjskQgjpsprtseGc62q//Ixzrm34B8BnHRMeaejjc6m4o6yG\nt6M5pvu3kKRc+R486TDesrNHCXQIdAzEdK/pHRcogVarhU6n/y/09NNP46OPPsKZM2coqSGEEANr\ny3Jv34YPajfoG2yYcEhz7pRUIfRsKgDg7097Qyhoph5UeT7wyxocMjXBGakxzIzMsCV4S8v1o0i7\nSkpKQkhISH3Rynnz5mHlypVUtJIQQjpAS3Ns1jPGigH4NpxfAyAfwJEOi5AAALYfu4ZqtQ4TfBww\npLtN041qh6Cya5TYbqcfQdw4ZCMcTBw6MNInl1qtxtatW9G/f38kJyfDwsKis0MihJAnTktzbN4B\n8C6AfwLYUHewdiiKdHiceBQAACAASURBVKDY9CIcirsDY5EAG8e3sLz7yg/QJf2E1x0dUQ6Ov7j8\nBZO6T+q4QJ9gly5dwoIFC5CQkICZM2figw8+gL29fWeHRQghT5yWEhtPzvkNxlgYgL51B+uGNDjn\nCQaOjQDQ6Tjeql3evSykO1ysm9lmvzwfOLIG35qZIlpiBCuxFV4PfJ2GoDpIbm4uCgoK8OOPP2LK\nlCmdHQ4hhDyxWkpsNgBYDODDJp7jAIYaJCLSyPeXspCQqYS9mRgrh7ewsd6R15Beo8R7LnIAHK8H\nvg5bKS1oM6Rz584hMTERL774IsaNG4eUlBRIpdLODosQQp5ozSY2nPPFtX+HdFw4pKEKlQbvHLsG\nAFg/rjdMxM38c13+Htqrh7DZyRFV4JjYfSJGu43uwEifLKWlpdiwYQN27doFLy8vLFmyBGKxmJIa\nQgh5BPz/9u47Oqpqe+D4d8+kEQiBEJpUpSd0Qm9BijRBAUEUBBuKz6dSFUFEQJ6iok+li0+eP1F8\nWAiQAFKkhyIWSEJVCIhIbwmknt8fM0KAJAzIzGQm+7PWLO7MPXPv5iwgm1PuvuE2DRHpISJB9uOX\nRORLEanj/NDUtO/3cex8CnXKFeH+emWyb2SfgpobHMRP/r6UKFCCUY1GuTbQfCQ6Oprw8HBmzpzJ\n0KFDtWilUkrlMY7sPx1njDkvIs2Ae4H5wEznhqUOnUpm9rrfANv2bktO27ujh7E37RwfhhQFYFyz\ncQT7624cZzh06BDdu3cnODiYjRs38s4771CwYC4lLZRSSrmcI4nNX7ugugLTjDFfAfpfVCebFJ1A\nanom99W9g/rli2bfKO4b0uIXMrpEcdKAnlV60rKszhzeTsYYYmNjAShXrhzLly9n+/btNG7c2M2R\nKaWUyo4jic0fIjIVeBCIFhE/B7+nblHsryeJ2XmUAr5WXuxUPftGF47DkmHMLhJMgp8PZQqVYUTD\nEa4N1MsdOXKE++67j6ZNm14uWtmmTRv8/PzcHJlSSqmcOJKg9AbWAJ2NMaex1Y56KfevqFuVkWV7\n9+DISpQOzmFBavRw4tLPM6uobdppQvMJFPTVaZHbwRjDRx99RFhYGMuXL+ftt9/WopVKKeUhctvu\nDYAx5oKIxAORIhIJrDPGxDg9snzqy22HSPjjHGWKFGBQq7uybxT3DSkJ3/JymTJkAP1q9KNhKa1y\ncbv06tWLr7/+mtatW/PRRx9RuXIuVdSVUkrlKY7sinoW+BIob399KSLPODuw/OjcpTTeXrYbgJc6\nVSfA13p9o6QTsGQYHxYpwq++VioWrsjz9Z93caTeJ2vRyvvuu48ZM2awatUqTWqUUsrDODIVNQho\nZIx52RjzMtAYeNq5YeVPH6zcy8mkVBpWLErX2qWzb7RkGNszLjC3SGEsYmFSi0kE+AS4NlAvs3Pn\nTpo3b365aGX//v156qmntGilUkp5IEf+5RYgLcv7NPtn6jb67UQSn2w8gAiM7RqefSmEuG9ITljI\n6BKhGODxmo9Tq3gtl8fqLVJTU3nttdeoX78++/fvp2jRHHafKaWU8hg3XGMDfArEishX2BKa+4C5\nTo0qH3p9STxpGYbeEWWpVTab59AknYAlw5kSUoTDPlaqFq3K4DqDXR+ol/jhhx8YOHAgO3fu5KGH\nHuK9996jePHi7g5LKaXU3+TI4uHJIrIa+OsBKU8bY7Y6N6z8Ze2e46xIOEYhfx+G31Mt+0bRw9lo\nLjC/cAl8LD5MajEJX6uvawP1IidPnuTMmTMsWrSIrl21ArpSSnkLR0ZsAFLsr0z7r+o2Sc/IZMJi\n2/buf7SpTImgbNbLxH3LuYRvGVvmDlu7uv+gWkgOCZDK0erVq9mxYwfPPfccHTp0YO/evQQE6Pok\npZTyJo7sihoNfA6UBsoC80REixHdJp9tTmTvsQuUDwnksRYVr29g3wX1ZkhR/vSxUju0NgPDB7o6\nTI929uxZnnrqKe6++26mT59OSootN9ekRimlvI8ji4f7AQ2NMWOMMaOBRsAjzg0rfziTnMq7K/YA\nMLpLDfx9stneHT2clSQRFVSIAGsAE1tMxMfi6ECbWrRoEWFhYXz00UcMHz6cH374QYtWKqWUF3Pk\nJ+TBa9r5AL86J5z85b0VezmTnEazSsXoEFby+gZx33IqYSHjy9qmoF5o8AJ3Bt/p4ig916FDh+jZ\nsyfVq1fn22+/pWFDfYihUkp5O0cSm2QgTkSWAQboAKwXkSkAxpihTozPa+358zyfxh7EIjD23rDr\nt3cnncAsGcbE0BBOWS00LNWQvtX7uidYD2KMYdOmTTRr1uxy0cpmzZppfSellMonHJmKWgKMAzYB\nscB4YBUQZ3/lSEQ6ishuEdknIjnWlxKRXiJiRCTC4cg9mDGGCYvjycg09G1UnuqlCl/fKHoE0ZLM\ndwUDCfQJZELzCVhEHxiXm8OHD9OtWzeaN29+uWhlZGSkJjVKKZWPOLLde86tXFhErMBUoD1wGNgq\nIlHGmPhr2gUBzwGbb+U+nmjVrmOs23uCwgE+DG1f9foG8Qs5tmshr5e1PX14ZMORlClUxsVReo7M\nzExmz57NiBEjSE9PZ8qUKbRo0cLdYSmllHIDZw4BNAL2GWN+NcakAl8A3bNpNwGYDFxyYix5Rmp6\nJhOXJADwfLuqFCt0zULWpJOYxUN5NTSE8xYLLcu0pEeVHm6I1HP07NmTp59+moYNG7Jz506GDBmC\n1ZrNQmyllFJez5mJTRngUJb3h+2fXSYi9YByxpjFuV1IRAaJyDYR2Xb8+PHbH6kLzd14gN9OJHFX\n8YI80rTC9Q2ih/O15SLrAwtQ2K8w45qNy768Qj6Xnp5+uWhlz549mT17NitWrOCuu3KoiK6UUipf\ncDixEZGb3SOb3U9jk+V6FuBdYNiNLmSMmWWMiTDGRHjyY+9PXEjh/ZV7AXilSxi+1mu6P34hh3cv\nZHKxEABGNx5NicASrg4zz/vll19o2rQps2fPBqBfv3488cQTmgAqpZRy6AF9jURkB7DX/r6OiHzg\nwLUPA+WyvC8LHMnyPgioCXwvIgeAJkCUNy8gfmf5Hs6npBNZrThtql+TsCSdJHPJMF4JLUayRWhf\noT2d7uzknkDzqJSUFF599VUaNGjAwYMHtbaTUkqp6zgyYvM+0BU4CWCM+Rlo48D3tgJVROROEfED\nHgSi/jppjDlrjAk1xlQ0xlTEtuOqmzFm203+HjxC3JGzfLE1ER+LMKZL2PUNYkYwz3qJbQUCCAkI\n4ZUmr+gIRBZbt26lfv36jB8/nr59+5KQkECPHrr2SCml1NUceY6NxRhz8Jofshk3+pIxJl1EngWW\nAVbgY2NMnIiMB7YZY6Jyv4L3MMYwflE8xkD/ZhWoXKLQ1Q3io/h190LeK2PbBTWu6TiKBhR1Q6R5\n1+nTp7lw4QLR0dF06qQjWUoppbLnSGJzSEQaAca+hfufwB5HLm6MiQair/lsbA5tIx25pidauvMo\nm387RdFAX15oe8327qSTpC8ZypjQYqSI0K1SN9qUd2RAzPutWrWKHTt28Pzzz9OhQwf27Nmj5RCU\nUkrlypGpqMHAUKA88Ce2tTCDnRmUN7mUlsHr0bbt3UPbVyU40PfqBjEj+I9PCjsC/CkZWJIXG73o\nhijzljNnzvDkk0/Stm1bZs6ceblopSY1SimlbuSGiY0x5pgx5kH7ephQ+/EJVwTnDeas/43Dpy9S\nrWQQfRuVv/pkfBS790QxrWgwABOaT6CwXzZPIc5HFi5cSFhYGB9//DEjR47UopVKKaVuyg2nokRk\nNlm2af/FGDPIKRF5kT/PXWLq6n2ArR6UT9bt3UknSV0ylJdDi5EuQp9qfWh6R1M3RZo3JCYm8sAD\nD1CjRg2ioqKIiPDaDXJKKaWcxJE1NiuyHAcA93P1g/dUDiYv3U1yagbtw0rSvHLo1SdjRjLDL5U9\n/sGUCyrH0Ab5s5aoMYb169fTsmVLypcvz4oVK2jSpInWd1JKKXVLHJmKmp/lNRfoAWSzX1ll9fOh\nM3y1/TC+VmF05xpXn0xYxC97o5gTXBhBmNh8IoG+ge4J1I0SExPp0qULrVq1uly0slWrVprUKKWU\numW3UlLhTiCbWgDqL8YYxi+21fp8rMWdVAwteOVk8ikuLh7C6NBiZIowMHwg9UvWd1Ok7pGZmcm0\nadMIDw9n7dq1vP/++1q0Uiml1G3hyBqb01xZY2MBTgEvOTMoTxf18xF+OHia0EL+PNum8tUno0fw\nvn8aB/wKUCn4Lv5R7x/uCdKNevTowcKFC2nfvj2zZs2iYsWK7g5JKaWUl8g1sRHbU/nqAL/bP8o0\nxly3kFhdkZyazhsxuwAYcU9VggKybO9OWMSWfYv4v9IlsYqF11tOwt+aP3b8pKenY7FYsFgs9OnT\nh+7duzNw4EB9urJSSqnbKtepKHsS840xJsP+0qTmBmau+ZU/zl4i/I7C9GqQpVRW8imSlgzlldBi\nAAyq/RThxcLdFKVr/fzzzzRu3JhZs2YB0LdvXx599FFNapRSSt12jqyx2SIi+WsRyC06cuYiM9fu\nB+DVe8OxWrL84I4ZyVv+6Rzx9aFGSA2erP2km6J0nUuXLjFmzBgiIiI4fPgwpUqVcndISimlvFyO\nU1Ei4mOMSQdaAE+KyH4gCRBsgzma7FzjjZhdXErLpEvt0jS6M+TKiYTFrN23iK9KlcDX4sOkFpPw\ntfjmfCEvsGXLFgYMGMCuXbsYMGAAU6ZMISQk5MZfVEoppf6G3NbYbAHqA/e5KBaPtu3AKaJ+PoK/\nj4VRnapfOZF8irNLhjAu1PZD/Z/1nqNy0co5XMV7nDt3josXL7J06VLuueced4ejlFIqn8gtsREA\nY8x+F8XisTIzDa8tsm3vHtTqLsoWzfJMmpgXmVQgg+M+AdQrXpdHwh5xU5TOt3z5cuLi4hgyZAjt\n2rVj9+7dWg5BKaWUS+WW2BQXkRwfh2uMmeKEeDzSV9sPs+P3s5QqHMDgyEpXTuxawvL9i4guWZwC\nVn8mtngdq8XqvkCd5PTp0wwdOpRPPvmE8PBwnnnmGfz9/TWpUUop5XK5LR62AoWAoBxeCriQks7k\nZbsBeLFTNQL97Lli8ilOLH6BCfYpqKERwylfuHxOl/FYX3/9NWFhYXz66aeMGjWKbdu2aUKjlFLK\nbXIbsfnDGDPeZZF4qKmr93H8fAp1yxWhe50ylz830SN5LTCTM1YrTUo1oXe13m6M0jkSExN58MEH\nqVmzJtHR0dSrV8/dISmllMrnchux0YeM3EDiyWTmrPsNgFfvDcPy1/buXUtYdCCa7wsGUsgnkAkt\nJmCRW6lekfcYYy7XdSpfvjyrVq1i8+bNmtQopZTKE3L7advWZVF4qEnRCaRmZNKjXhnqlS9q+zD5\nFEejh/BGiO39S41fplRB73h+y8GDB+nUqRORkZGXk5sWLVrg6+vdW9eVUkp5jhwTG2PMKVcG4mk2\n7T/J0rijFPC1MrLjle3dJuZFxgZmct5qIbJsJN0qdXNjlLdHZmYmH374IeHh4axfv54PPviAli1b\nujsspZRS6jo3LIKprpeRaXhtURwAz0RWolRwgO3Ermi+PBDNptAQivgG8WqzV72ibMB9993HokWL\nuOeee5g5cyYVKmhxd6WUUnmTJja34Iutiew6ep4yRQrwZKu7bB8mn+JQ9Au8E1IEgFeajSO0QKgb\no/x70tLSsFqtWCwW+vbtS69evejfv79XJGpKKaW8l3esaHWhsxfTeGf5HgBGda5OgK/tuTQZMS8y\nuqDhosVCp4qd6FCxgzvD/Fu2b99Oo0aNmDFjBmArWvnII49oUqOUUirP08TmJn2wci+nklJpVDGE\nLrVK2z7cFc2nB2P4MSCAUP+ijG4y2r1B3qKLFy8yatQoGjVqxNGjRylXrtyNv6SUUkrlIToVdRP2\nH7/AJxsPIAJj7w2zjWAkn2JfzBDet09BvdZiIsH+wW6O9ObFxsYyYMAA9uzZw2OPPcbbb79N0aJF\n3R2WUkopdVM0sbkJry9JID3T0CeiHDXL2JKXNPsUVJoIPSv3oFXZVm6O8tYkJSWRlpbGd999R7t2\n7dwdjlJKKXVLNLFx0Jo9x1m16xiF/H0Yfk8124e7Y/jo0FLiixbhjgLFGd5whHuDvElLly4lLi6O\nYcOG0bZtW3bt2oWfn5+7w1JKKaVuma6xcUBaRiYTFtuqd//z7soUD/KH5FPExbzArCK2kZsJLd+g\nkF8hd4bpsJMnTzJgwAA6derE3LlzSU1NBdCkRimllMfTxMYBn8UeZN+xC1QoFsjA5hUBSIl5kTEF\nIV2Eh6v3pVHpRu4N0gHGGBYsWEBYWBjz5s1jzJgxbN26VRMapZRSXkOnom7gdFIq767YC8DozjXw\n97HC7himHl7OviKFqVjwDp5vMMTNUTomMTGRhx56iNq1a7N8+XLq1Knj7pCUUkqp20pHbG7g3RV7\nOHsxjeaVi9E+rCRcPM2PMUP4JDgIC8LE1pMp4FPA3WHmyBjDqlWrAKhQoQLff/89sbGxmtQopZTy\nSprY5GLPn+f5bHMiFoGxXcMREZJjRjK6EBgRHqv5KHWK590E4bfffqNDhw60bdv2ctHKZs2a4eOj\nA3VKKaW8kyY2OTDGMGFxPBmZhocbV6BaqSDYHcO7v3/HIV9fqgZVZHDdf7g7zGxlZGTw73//m5o1\na7J582amT5+uRSuVUkrlC/pf9xysTDjGur0nKBzgw5D2VeHiaTYtHcIXRYLwwcKkyLfxs+bNRbfd\nu3dnyZIldO7cmRkzZugThJVSSuUbmthkIyU9g4lLbNu7X2hXlZCCfpz/+hnGFrLVShpcdzDVQqq5\nM8TrZC1a2b9/f/r27ctDDz2k9Z2UUkrlK06dihKRjiKyW0T2ichL2ZwfKiLxIvKLiKwUkQrOjMdR\nczce4MDJZCoVL0j/phVg91LePLKSoz4+1CpShcdqPeHuEK+ybds2IiIimD59OgB9+vTh4Ycf1qRG\nKaVUvuO0xEZErMBUoBMQBvQVkbBrmv0IRBhjagMLgMnOisdRJy6k8MHKfQC80jUM39SzrF72AguD\nCuEvViZGvo2PJW8MdF28eJEXX3yRxo0bc/z4cSpUyBN5oVJKKeU2zhyxaQTsM8b8aoxJBb4Aumdt\nYIxZbYxJtr+NBco6MR6HvLN8N+dT0mlTrTiR1UpwOnoErxW0ddPzDYZwV/Bdbo7QZtOmTdSpU4fJ\nkyfz2GOPER8fT9euXd0dllJKKeVWzhx6KAMcyvL+MNA4l/aPAzHZnRCRQcAggPLly9+u+K6z8/ez\nfLH1ED4WYUzXMMzupUw8uoqThQoSERLOw2H9nXbvm3Xx4kUyMzNZsWIFbdu2dXc4SimlVJ7gzMQm\nuwUeJtuGIv2ACKB1dueNMbOAWQARERHZXuPvMsYwfnE8xsCA5hWpVCiNmC+GsLxwQQLFlwmRb2MR\n9+6Oj46OJi4ujhEjRnD33XeTkJCAr6+vW2NSSiml8hJn/qQ+DGTdZ1wWOHJtIxFpB4wGuhljUpwY\nT65idh5ly2+nCCnox3Ntq3A8ejgTC9pysxGNX6JskPtmyU6cOEG/fv3o0qULn3322eWilZrUKKWU\nUldzZmKzFagiIneKiB/wIBCVtYGI1ANmYktqjjkxllxdSstgUnQCAEPbV6Vw4krGHVvDOauV5sXr\n0bPqA26JyxjDF198QY0aNfjyyy959dVX2bJlixatVEoppXLgtKkoY0y6iDwLLAOswMfGmDgRGQ9s\nM8ZEAW8BhYD/2bcmJxpjujkrppx8tO5XDp++SPVSQTxYM4hvPhnC2qACBFn8eK31W27bNp2YmMiA\nAQOoU6cOc+bMoVatWm6JQymllPIUTt23bIyJBqKv+WxsluN2zry/I/48d4lp3+8HYGzXMP5cPpw3\n7bugXm76KiULlnRpPMYYVq5cSbt27ahQoQJr1qyhYcOGWK1Wl8ahlFJKeaJ8XyvqzaW7SE7N4J7w\nkjTJ2MrYY+tItlhoX6oJXSrd69JY9u/fT9u2bWnfvv3lopVNmjTRpEYppZRyUL5ObH46dIavt/+O\nn9XCmLZl+HzFMLYUCCDEGsCY1m+6bAoqIyODKVOmUKtWLX744QdmzpypRSuVUkqpW5A3HqHrBsYY\nXlsUB8BjLe4kfdMo3g20JTJjW0wiJCDEZbHce++9xMTE0LVrV6ZPn07Zsm5/TqFSSinlkfLtiM3C\nn47wY+IZQgv5849y+xhzYiMpFgv3lmlN24rtnX7/1NRUMjMzARg4cCDz5s0jKipKkxqllFLqb8iX\niU1yajpvxOwCYHTb0sxfO5JfAvwpYS3IS60mOf3+W7ZsoUGDBkybNg2A3r1707dvXy1aqZRSSv1N\n+TKxmbHmV46eu0StMsGEHZnM1EBbN0yIfIvCfoWddt/k5GSGDRtG06ZNOX36NJUqVXLavZRSSqn8\nKN8lNr+fucjMNbbt3W/U/Z3RJ2NJF6F3+Q40K+u8Bbvr16+nVq1aTJkyhSeffJK4uDg6derktPsp\npZRS+VG+Wzz8r+gEUtIzeSA8iFU7R7I70I+yPkEMazHBqfdNS0vDarWyevVqIiMjnXovpZRSKr/K\nV4nN1gOnWPzLH/j7WHjAdwaDCghiYOLd/ybQN/C232/RokUkJCQwcuRI2rRpQ3x8PD4++arLlVJK\nKZfKN1NRmZmG8YviARhf6xATz28nQ4RH7upGg9INb+u9jh8/zkMPPUS3bt34/PPPLxet1KRGKaWU\ncq58k9gs2H6YHb+fpXJQBgdOv8dvfr7c5VuEfzYfe+MvO8gYw7x586hRowYLFixg/PjxbN68WYtW\nKqWUUi6SL4YQzl9KY/LS3QC8UPq/vGyxYDUwqf1U/K3+t+0+iYmJPProo9SrV485c+YQHh5+266t\nlFJKqRvLFyM2U1fv58SFFB4pGceHmXEYEZ6s8gDhxWv/7WtnZmaybNkyACpUqMC6devYsGGDJjVK\nKaWUG3h9YnPwZBIfr/+NIJKRgE/43deHGn7FGNRk1N++9t69e7n77rvp2LEja9euBaBRo0ZatFIp\npZRyE69PbCZFJ5CakckLZefydaAVXwOvd5iJr9X3lq+Znp7OW2+9Re3atfnpp5+YM2eOFq1USiml\n8gCvXmOzcd8JlsX9yd0B25hXYD/gw7PV+1GlWLW/dd2uXbuybNkyunfvzrRp07jjjjtuT8BKKaWU\n+lu8dsQmPSOT8YvjCSKZkJLzOebjQx3/EgxoNPyWrpeSknK5aOUTTzzB/Pnz+eabbzSpUUoppfIQ\nr01svth6iF1Hz/NQ6GyWBVoJMPD6PbOwWm5+/UtsbCz169dn6tSpAPTq1YvevXtr0UqllFIqj/HK\nxObsxTSmfLeHZr6biQk5DMCQ8MepUPTmik4mJSUxZMgQmjVrxvnz56lSpYozwlVKKaXUbeKVic37\nK/eSlnSa4Dv+x2mrlcYFSvNgxHM3dY1169ZRq1Yt3nvvPQYPHszOnTvp2LGjkyJWSiml1O3gdYuH\n9x27wNyNB+hTdCZRgT4UNMKEjnOwyM3lcOnp6fj6+rJmzRpatWrlpGiVUkopdTt5XWLz+pJ4Inw2\nsrr4UcDCi7UHU7pwOYe+++2335KQkMCoUaNo06YNcXFxWt9JKaWU8iBeNRX1/e5jbNt9gIDSX3He\naqF1YDnuq/f0Db/3559/0rt3b+6//34WLFigRSuVUkopD+U1iU1aRiYTFsfTMWQ6PwT6EGyEcZ0+\nznXnkjGGTz/9lLCwMBYuXMjrr79ObGysFq1USimlPJTXDEl8uukgxc59x5qKxwALY+q9QGihUrl+\nJzExkSeeeIKIiAjmzJlD9erVXROsUkoppZzCK0ZsTiWlMnvFdqylv+GixULHgnfSsc5j2bbNzMwk\nJiYGsBWt3LBhA2vXrtWkRimllPICXpHYvPvdHhoX+oCdBayEGmF054+zbbdnzx4iIyPp3Lkza9as\nASAiIkKLViqllFJewuMTm11Hz7H7p/9jXbETAIyLeJEigaFXtUlPT+fNN9+kdu3a7Nixg//85z+6\nhVsppZTyQh69xsYYw+SoWNJLR5FqsXJ/UBVa13z4unZdunRh+fLl9OjRg6lTp1KqVO5rb5RSSinl\nmTx6xOa7+D8JuDCRvQFWSmVaGNnpyhTUpUuXyMjIAGDQoEEsWLCAr776SpMapZRSyot5bGKTkp7B\n/6I/YFPIKQAmNhlDoQJFANiwYQN169a9XLSyZ8+e9OzZ022xKqWUUso1PDaxmfv9jxwt8i3pIvQN\nqk7jGg9w4cIFnnvuOVq2bMmlS5eoUaOGu8NUSimllAt55Bqb4+dT+CFuKAeKWCiXYWFIl49Zs2YN\nAwYMIDExkWeffZZJkyZRqFAhd4eqlFJKKRfyyMRm+leTiQ0+hcXApOavUcA/CIDAwEDWrVtH8+bN\n3RyhUkoppdzB46aiklNS2JD+BZki1NwZTPRX8QC0bt2aHTt2aFKjlFJK5WNOTWxEpKOI7BaRfSLy\nUjbn/UVkvv38ZhGpeKNrnjh3gINJGZx4P5F572zkm2++uVy0Uh+0p5RSSuVvTktsRMQKTAU6AWFA\nXxEJu6bZ48BpY0xl4F3gzRtd90xyCvte3supnRf517/+xcaNG7VopVJKKaUA566xaQTsM8b8CiAi\nXwDdgfgsbboD4+zHC4APRUSMMSani6adSKN81WIsj9pAtWrVnBO5UkoppTySMxObMsChLO8PA41z\namOMSReRs0Ax4ETWRiIyCBhkf5uSuOfkTi1aCUAo1/RVPqZ9cYX2xRXaF1fo/wRVvuDMxEay+eza\nkRhH2mCMmQXMAhCRbcaYiL8fnufTvrhC++IK7YsrtC+uEJFt7o5BKVdw5uLhw0C5LO/LAkdyaiMi\nPkAwcMqJMSmllFLKizkzsdkKVBGRO0XED3gQiLqmTRQwwH7cC1iV2/oapZRSSqncOG0qyr5m5llg\nGWAFPjbGxInI9DWMCgAAB35JREFUeGCbMSYKmAN8KiL7sI3UPOjApWc5K2YPpH1xhfbFFdoXV2hf\nXKF9ofIF0QESpZRSSnkLj3vysFJKKaVUTjSxUUoppZTXyLOJjTPKMXgqB/piqIjEi8gvIrJSRCq4\nI05XuFFfZGnXS0SMiHjtVl9H+kJEetv/bMSJyDxXx+gqDvwdKS8iq0XkR/vfk87uiNPZRORjETkm\nIjtzOC8i8r69n34RkfqujlEppzPG5LkXtsXG+4G7AD/gZyDsmjbPADPsxw8C890dtxv7og0QaD8e\nnJ/7wt4uCFgLxAIR7o7bjX8uqgA/AkXt70u4O2439sUsYLD9OAw44O64ndQXrYD6wM4czncGYrA9\nQ6wJsNndMetLX7f7lVdHbC6XYzDGpAJ/lWPIqjsw1368AGgrItk98M/T3bAvjDGrjTHJ9rex2J4Z\n5I0c+XMBMAGYDFxyZXAu5khfPAlMNcacBjDGHHNxjK7iSF8YoLD9OJjrn6nlFYwxa8n9WWDdgf8a\nm1igiIiUdk10SrlGXk1ssivHUCanNsaYdOCvcgzexpG+yOpxbP8j80Y37AsRqQeUM8YsdmVgbuDI\nn4uqQFUR2SAisSLS0WXRuZYjfTEO6Ccih4Fo4J+uCS3Pudl/T5TyOM4sqfB33LZyDF7A4d+niPQD\nIoDWTo3IfXLtCxGxYKsSP9BVAbmRI38ufLBNR0ViG8VbJyI1jTFnnBybqznSF32BT4wx74hIU2zP\nz6ppjMl0fnh5Sn75d1PlY3l1xEbLMVzhSF8gIu2A0UA3Y0yKi2JztRv1RRBQE/heRA5gW0MQ5aUL\niB39O7LQGJNmjPkN2I0t0fE2jvTF48CXAMaYTUAAtgKZ+Y1D/54o5cnyamKj5RiuuGFf2KdfZmJL\narx1HQXcoC+MMWeNMaHGmIrGmIrY1ht1M8Z4Y/E/R/6OfIttYTkiEoptaupXl0bpGo70RSLQFkBE\namBLbI67NMq8IQp4xL47qglw1hjzh7uDUup2ypNTUcZ55Rg8joN98RZQCPifff10ojGmm9uCdhIH\n+yJfcLAvlgEdRCQeyABGGGNOui9q53CwL4YBs0VkCLapl4He+B8hEfkc29RjqH090auAL4AxZga2\n9UWdgX1AMvCoeyJVynm0pIJSSimlvEZenYpSSimllLppmtgopZRSymtoYqOUUkopr6GJjVJKKaW8\nhiY2SimllPIamtioPEtEMkTkpyyvirm0rZhTRWNXE5EIEXnffhwpIs2ynHtaRB5xYSx1vbWStVJK\nZSdPPsdGKbuLxpi67g7iZtkfCPjXQwEjgQvARvu5Gbf7fiLiY6+Xlp262MpsRN/u+yqlVF6kIzbK\no9hHZtaJyHb7q1k2bcJFZIt9lOcXEali/7xfls9niog1m+8eEJE37e22iEhl++cVRGSl/XorRaS8\n/fMHRGSniPwsImvtn0WKyGL7CNPTwBD7PVuKyDgRGS4iNURkyzW/r1/sxw1EZI2I/CAiy7Krviwi\nn4jIFBFZDbwpIo1EZKOI/Gj/tZr9KbzjgT72+/cRkYIi8rGIbLW3za46ulJKeSxNbFReViDLNNQ3\n9s+OAe2NMfWBPsD72XzvaeDf9tGeCOCw/TH6fYDm9s8zgIdzuO85Y0wj4EPgPftnHwL/NcbUBj7L\nct+xwD3GmDrAVU97NsYcAGYA7xpj6hpj1mU5lwD4ichd9o/6AF+KiC/wAdDLGNMA+Bh4PYc4qwLt\njDHDgF1AK2NMPXtMk4wxqfbj+fb7z8dWT2yVMaYhtnILb4lIwRyur5RSHkenolRelt1UlC/woYj8\nlZxUzeZ7m4DRIlIW+NoYs1dE2gINgK32shMFsCVJ2fk8y6/v2o+bAj3sx58Ck+3HG4BPRORL4Oub\n+c1hK8rYG3gDW2LTB6iGrZDnd/Y4rUBOtXz+Z4zJsB8HA3Pto1MG+2P0s9EB6CYiw+3vA4DyQMJN\nxq6UUnmSJjbK0wwB/gTqYBtxvHRtA2PMPBHZDHQBlonIE4AAc40xoxy4h8nh+Lo2xpinRaSx/V4/\n2RMuR83HVt/ra9ulzF4RqQXEGWOaOvD9pCzHE4DVxpj77VNg3+fwHQF6GmN230ScSinlMXQqSnma\nYOAPY0wm0B/biMZV7NM7vxpj3sdWzbg2sBLoJSIl7G1CRKRCDvfok+XXTfbjjVwptPowsN5+nUrG\nmM3GmLHACaDcNdc6DwRldxNjzH5so06vYEtyAHYDxUWkqf36viISnkOcWQUDv9uPB+Zy/2XAP8U+\nHCS2yvBKKeU1NLFRnmYaMEBEYrFNQyVl06YPsFNEfgKqY1sbEw+MAZbbF+l+B1y3KNfO3z7i8zy2\nESKA54BH7d/tbz8HtjUqO+xbzdcCP19zrUXA/X8tHs7mXvOBftimpbCvi+mFbUHwz8BPwHULpLMx\nGfiXiGzg6mRvNRD21+JhbCM7vsAv9pgnOHBtpZTyGFrdW6ksROQAEGGMOeHuWJRSSt08HbFRSiml\nlNfQERullFJKeQ0dsVFKKaWU19DERimllFJeQxMbpZRSSnkNTWyUUkop5TU0sVFKKaWU1/h/5W7I\nF1yhEVcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10ae37828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC (Voting): 0.695028236076\n",
      "ROC AUC (AdaBoost Decision Tree): 0.699486958706\n",
      "ROC AUC (AdaBoost Random Forest): 0.720396972505\n"
     ]
    }
   ],
   "source": [
    "#limited feature set\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "from sklearn.ensemble import VotingClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "voting_all = VotingClassifier(estimators=[\n",
    "    (\"Naive Bayes\", GaussianNB()),\n",
    "    (\"Random Forest\", RandomForestClassifier()),\n",
    "    (\"Logistic Regression\", LogisticRegression()),\n",
    "    (\"Support Vector Machine\", SVC()),\n",
    "    (\"Stochastic Gradient Descent\", SGDClassifier(max_iter = 1000)),\n",
    "    (\"k-nearest Neighbors\", KNeighborsClassifier()),\n",
    "    (\"Multi-layer Perceptron\", MLPClassifier())\n",
    "], voting='hard')\n",
    "\n",
    "score_classifiers([\n",
    "    (AdaBoostClassifier(base_estimator=RandomForestClassifier()), \"AdaBoost Random Forest\"),\n",
    "    (AdaBoostClassifier(base_estimator=DecisionTreeClassifier()), \"AdaBoost Decision Tree\"),\n",
    "    (voting_all, \"Voting\")\n",
    "], only_important_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdgAAAEWCAYAAADFO4ZdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzs3XlclOX+//HXBQi44i4KuOSSoqIs\ngppr7qlpYm6puOBSeepk2fKt3+nU6XhOnrZTx31PyyWx1NK0cssSFBAXVFwwYVAUFRFkHeb6/TFI\nZi6jMjMsn+fjwUPmnnvu+XA/YN7e133fn0tprRFCCCFE0XKwdwFCCCFEaSQBK4QQQliBBKwQQghh\nBRKwQgghhBVIwAohhBBWIAErhBBCWIEErBBCCGEFErBC3IVS6jelVJZSKkMplayUWqaUqnTT8x2V\nUtuVUulKqTSl1CallPct26iilPpEKZVQsJ1TBY9r2v4nEkLYigSsEPc2UGtdCWgL+AJvACilOgDb\ngA1APaARcBD4RSn1SME6zsBPQEugL1AF6AhcBgJt+2MIIWxJSScnIe5MKfUbEKq1/rHg8Sygpda6\nv1LqZ+Cw1vq5W16zBUjRWo9VSoUC/wQaa60zbFy+EMKO5AhWCAsppTyBfsAppVQFzEeiX91m1bVA\nr4LvewLfS7gKUfZIwApxb98opdKBROAi8DZQHfPfz/nbrH8euHF+tcYd1hFClHISsELc22CtdWWg\nG9Acc3imAiag7m3WrwtcKvj+8h3WEUKUchKwQlhIa70LWAZ8oLW+DuwFnr7NqsMwX9gE8CPQRylV\n0SZFCiGKDQlYIe7PJ0AvpVRb4HUgRCn1glKqslKqmlLqPaAD8E7B+iswDy2HKaWaK6UclFI1lFL/\np5R6wj4/ghDCFiRghbgPWusU4HPg/2mt9wB9gCGYz7OexXwbTyet9cmC9XMwX+h0HPgBuAbswzzM\nHGHzH0AIYTNym44QQghhBXIEK4QQQliB1QJWKbVEKXVRKXXkDs8rpdSnBW3jDiml/KxVixBCCGFr\n1jyCXYa5Ndyd9AOaFnxNBuZasRYhhBDCpqwWsFrr3cCVu6wyCPhcm4UDVZVScr+gEEKIUsHJju/t\ngfn2hRsMBcv+1PVGKTUZ81EuFStW9G/evLlNChRCiJLImK9JzcwlNTOX65eTMeVcB1P+Ja11LXvX\nVpbYM2DVbZbd9pJmrfUCYAFAQECAjoyMtGZdQghR4mTn5bPt6AXCogzsPnERNFRXCrfYrTSunM+u\nVXPO2rvGssaeAWsAvG567Amcs1MtQghR4mitiU5IZV1UEt8eOkd6thFj+iVSf5hDUI8BvD5tEt0e\n7YezkwNq1Rx7l1vm2DNgNwLTlFKrgSAgTWstTdGFEOIekq5msT7KwPoDSZy5dB0wh23VhN2c3DQP\nTEaGt51A75budq60bLNawCqlVmFujl5TKWXAPANJOQCt9TxgM/AEcArIBMZbqxYhhCjpMnONbDmc\nTFi0gb3xl7nRI6h2ZRe6uOezZ+lMIn7ZTffu3Vm4cCGNGze2b8HCegGrtR55j+c18Ly13l8IIUo6\nk0kTceYK66IMbDlynszcfACcnRzo09KdYD8POjWpybebNrLwcAwLFiwgNDQUpW53iUvRiIqKqu3k\n5LQIaIU0KzIBR4xGY6i/v//FW5+05xCxEEKI2/jt0nXWRxsIi04i6WpW4XL/BtUI9vOkv09dEk/H\nER3xPU6PjmXw4MHEx8dTo0YNq9fm5OS0yN3dvUWtWrVSHRwcynSvXZPJpFJSUryTk5MXAU/e+rwE\nrBBCFAPXsvP47tB5wqIMRJ5NLVzuUbU8Q/w8GOLnSaOaFcnNzWXmzH8yc+ZM6tSpw7Bhw3B1dbVJ\nuBZoJeFq5uDgoGvVqpWWnJzc6nbPS8AKIYSd5Js0e05dIizKwNbYZHKMJgAqODvSt5U7Q/08af9I\nDRwczEO+ERERTJw4kdjYWEaPHs3HH3+Mq6urrct2kHD9XcG+uO1QuQSsEELY2MkL6ayLNvDNgSQu\nXMspXN7hkRoE+3vSr5U7FV3++PGclJRE586dqVOnDt9++y39+/e3ddniPknACiGEDaRez2XToXOE\nRRk4aEgrXN6gRgWC/Tx5ytcDr+oV/vS6EydO0KxZMzw8PFizZg09evSgSpUqtiy9WAkMDHz0tdde\nOx8cHHztxrJ333239okTJ1xXrlyZcOv6cXFxzjt27Kg0derUKwC7d++usGTJkhrLli1LvHXdoiYB\nK4QQVpKXb2JnXAphUQZ+On6BvHzzyGplFycGtKlLsJ8n/g2q3faq36tXr/Lqq6+yaNEidu7cSZcu\nXXjqqads/SMUO08//fTlVatWVb85YMPCwqq///77htutf/LkSZc1a9ZUvxGwXbp0yezSpUumLWqV\ngBVCiCIWey6NsKgkNsQkcfl6LgAOCro0q8VQf096e9fBtZzjHV+/ceNGnn32WZKTk5kxYwbt2rWz\nVenF3pgxY1JnzpzpkZWVpcqXL6/j4uKcL168WK53794ZU6ZM8dy+fbubUkrPmDHj/KRJk1LffPNN\nj/j4eNfmzZt7jxw58pK/v3/Whx9+WGfHjh2npk+fXi8xMdH57NmzLufOnXOeOnXqhbfeeusiwIwZ\nM+quW7euet26dXNr1Khh9PX1zXz33Xcv3E+tErBCCFEEUtJz2BCTxLooA8eT0wuXN61diWB/8xBw\nnSr3viApNDSUxYsX07p1azZs2EBAQIA1y34oDV//zt8a2/3t3/2j7vScu7t7fps2ba6HhYW5jR49\n+ury5curP/nkk6mff/551cOHD5c/duxY7Pnz550CAwNb9O7dO+Of//xn0o1ABfj2228r37y9U6dO\nuf76669xV69edWzRokWrGTNmpERERJTftGlTtcOHDx/Ny8tTbdu29fb19b3vo14JWCGEeEA5xnx+\nOnaRsCgDO0+kkG8yDwFXrVCOJ9vUY6i/J6093O7Z+EEXtGVSShEQEECDBg147bXXcHZ2tvrPUBIN\nGzbsypo1a6qNHj366vr166svWrTot6VLl9YYNmzYFScnJ7y8vIxBQUEZe/bsqeDm5ma627Z69+59\ntXz58rp8+fLG6tWr5xkMBqedO3dW6tev39VKlSppQPfq1evqg9QpASuEEPdBa01M4lXCog1sOnie\ntKw8AJwcFD1b1GGovwfdm9fGxenOQ8A3S0xMZOrUqYwYMYIxY8YwdepUa5ZfpO52pGlNzzzzzNW3\n3nrLa8+ePRWys7MdOnXqlLlkyZIHuhHYxcWl8JYjR0dHjEajuvEfnodV1ttcCSGERc6nZTFn5yl6\nfrSLp+b8ysrwBNKy8vCuW4X/N8Cb8P/rwaKQAPq2qmtRuJpMJubOnUvLli3ZuXMnOTk593yNMHNz\nczO1b98+PTQ0tOGQIUOuAHTt2jV93bp11Y1GI+fOnXPat29fpc6dO193c3PLz8jIsOx/OwW6deuW\nsXXrVrfMzEyVlpbm8OOPP1Z9kDrlCFYIIe4gKzefrbHmBvt7Tl0qbLBfs5ILg9vWI9jfkxZ17/+W\nmZMnTxIaGsru3bvp2bMnCxYsoFGjRkVcfek2YsSIKyEhIY1XrVoVDzBmzJirv/76a6UWLVq0VErp\nd955x1C/fn1jnTp18p2cnPSjjz7qPWrUqEv+/v5Z99p2165dM/v27Zvm7e3d0sPDI8fHx+e6m5tb\n/v3WWGSHwrYiE64LIaxJa82+M1cIizaw+XAyGTlGAJwdHejlXYdgfw+6NK2Fk+ODDwBu2LCBcePG\n8dFHHzFu3DirNue/QSkVpbV+6CumDh48+FubNm0uFUVNxVlaWpqDm5ubKT093aFDhw6Pzps372yn\nTp1ue6HTwYMHa7Zp06bhrcvlCFYIIYDEK5mERRtYH51EwpXfP0fbelUl2N+TgT51qVrhwS86Onjw\nIDExMYSEhDBo0CDi4+OpVq1aUZQurGD06NENTp48WT4nJ0eNGDHi8p3C9W4kYIUQZVZGjpHNh86z\nLtrAvjNXCpe7V3EtbLDfpHalh3qPnJwc3nvvPf79739Tt25dhg8fjqurq4RrMbdp06YzD7sNCVgh\nRJmSb9LsPX2ZsGgD3x9JJivPfGrNtZwDfVu6E+zvScfGNXF0ePhh27179zJx4kSOHTvG2LFj+eij\nj+zRnL+omUwmk5KG/2Ymk0lhnhf2TyRghRBlwumUDMKiDHx9IInzadmFywMbVmeovyf9WrtT2bVc\nkb1fUlISXbt2xd3dnc2bN9OvX78i27adHUlJSfGuVatWWlkP2YL5YN2AI7d7XgJWCFFqpWXmmRvs\nRxs4kPB7rwCv6uUZ4utJsJ8n9Wv8ucH+wzh27BgtWrTAw8ODtWvX0qNHDypXrnzvF5YQRqMxNDk5\neVHBHKhl/VZPE3DEaDSG3u5JuYpYCFGqGPNN7D6ZQlhUEj8cu0BuwRyrFZ0d6e9jbrDfrmH1wjlW\ni0pqaiovv/wyS5cuZffu3XTu3LlIt/+wiuoqYmE5OYIVQpQKx5OvERZl4JuYc6Skm5s2KAWdmtQk\n2N+DPi3dqeBsnY+8r7/+mueee46UlBTeeOMNac4vAAlYIUQJdjkjhw0x5iHg2HOFs5fxSM2KhQ32\n61Utb9UaJkyYwNKlS2nbti3fffcdfn5+Vn0/UXJIwAohSpRco4ntxy8SFm1gx/GLGAsa7FdxdWJg\nG3N3JV+vqlZt3nBzc/727dvTtGlTXnnlFcqVK7qLpETJJwErhCj2tNYcSbrGuqhENh48R2qmucG+\no4Pi8ea1CfbzpEeL2nedY7WonD17lilTpjBq1CjGjh3L5MmTrf6eomSSgBVCFFsXr2Xz9YEkwqIN\nnLiQUbi8uXtlgv08GeRbj9qVbXNf6Y3m/K+//jpaa55++mmbvK8ouSRghRDFSnZePtuOXiAsysDP\nJ1MoGAGmekVnBrWtR7CfJy3rVbFJ/94b4uLiCA0NZc+ePfTu3Zv58+fTsGFDm72/KJkkYIUQdqe1\nJjohlXVRSXx76Bzp2eYG++UcFb0KhoC7PVobZyf73HYZFxdHbGwsy5YtY+zYsTYNd1FyScAKIezG\nkJrJ19FJrD+QxJlL1wuX+3i6EeznycA29ahe8cEb7D+MAwcOEBMTw/jx43nyySeJj4+natUHmhZU\nlFESsEIIm7qeY+T7I8msizKwN/5y4fLalV14yteDYH9PmtWxX+ej7Oxs3n33XWbNmoWHhwcjR47E\n1dVVwlXcNwlYIYTVmUya8DOXCYtKYsuR82TmmhvsOzs50KelO8F+HnRqUvOh5lgtCr/88gsTJ04k\nLi6O8ePH8+GHH5aG5vzCTiRghRBW89ul64VzrCZdzSpc7t+gGsF+nvT3qYtb+eJx72hSUhLdu3fH\nw8ODrVu30rt3b3uXJEo4CVghRJG6lp3Hd4fOExZlIPJsauFyj6rlC+dYbVSzoh0r/KOjR4/i7e2N\nh4cHYWFhdO/enUqVHm4OWCFAAlYIUQTyTZqfT6YQFp3Etthkcgoa7FdwdqRvK3eG+nvSvlGNIm+w\n/zCuXLnC9OnTWb58Obt27aJLly4MHDjQ3mWJUkQCVgjxwE5eSGddtIFvDiRx4VpO4fIOj9Qg2N+T\nfq3cqehS/D5mwsLCeP7557l8+TJvvvkmgYGB9i5JlELF7zdfCFGspV7PZeNBc4P9Q4a0wuUNalQg\n2M/cYN+retHOsVqUxo0bx/Lly/Hz8+P777+nbdu29i5JlFISsEKIe8rLN7EzLoWwKAM/Hb9AXr65\nvVJlFycGtDHPserfoFqxbcBwc3P+jh070qJFC15++WWcnOQjUFiPVX+7lFJ9gf8CjsAirfW/b3m+\nPrAcqFqwzuta683WrEkIYbnYc2msizKwMeYcl6/nAuCgoEuzWgz196S3dx2bNNh/GGfOnGHy5MmM\nHj2akJAQac4vbMZqAauUcgRmA70AA7BfKbVRa330ptXeAtZqrecqpbyBzUBDa9UkhLi3lPQcNsQk\nsS7KwPHk9MLlTWtXKpxjtU6V4n9vaH5+PrNnz+aNN97AwcGBZ555xt4liTLGmkewgcAprXU8gFJq\nNTAIuDlgNVCl4Hs34JwV6xFC3EF2Xj4/HTPPsbrrRAr5BR32q1Yox5Nt6jHU35PWHm7Fdgj4VseO\nHWPixIns3buXfv36MW/ePOrXr2/vskQZY82A9QASb3psAIJuWefvwDal1F+AikDP221IKTUZmAzI\nH4kQRURrTUziVcKiDWw6eJ60LPMcq04Oip4t6jDU34PuzWvj4lS8h4Bv59SpU8TFxbFixQqeeeaZ\nEvMfA1G6WDNgb/cbrW95PBJYprX+UCnVAVihlGqltTb94UVaLwAWAAQEBNy6DSHEfTiflsX6aPMc\nq/EpvzfY965bhWB/Twa1rUfNSi52rPDBREVFcfDgQSZMmMDAgQM5c+YMVapUufcLhbASawasAfC6\n6bEnfx4Cngj0BdBa71VKuQI1gYtWrEuIMicrN5+tscmERRvYc+oSBRfVUrOSC4Pb1iPY35MWdUtm\nGGVlZfHOO+/wwQcf4OXlxahRo3B1dZVwFXZnzYDdDzRVSjUCkoARwKhb1kkAegDLlFItAFcgxYo1\nCVFmaK3Zd+YKYdEGNh9OJiPHPMeqs6MDvbzrEOzvQZemtezeYP9h7N69m9DQUE6ePMnEiRP54IMP\npDm/KDasFrBaa6NSahqwFfMtOEu01rFKqXeBSK31RuBlYKFS6iXMw8fj9I0b1oQQDyTxSiZh0QbC\nog0kXvm9wX5br6oE+3sy0KcuVSvYZ47VopSUlESPHj3w8vLixx9/pEePHvYuSYg/UCUtzwICAnRk\nZKS9yxCiWMnIMbL50HnWRRvYd+ZK4XL3Kq6FDfab1C4dDewPHz5M69atAfj222/p3r07FSsWn8kD\niiulVJTWOsDedZQl0sZEiBIq36TZe/oyYdEGthw5T3ae+dpA13IO9G3pTrC/Jx0b18SxGDXYfxiX\nLl3ipZdeYuXKlYXN+QcMGGDvsoS4IwlYIUqY0ykZhEUZ+PpAEufTsguXBzaszlB/T/q1dqeya/GY\nY7UoaK356quvmDZtGqmpqbz99tsEBd16x58QxY8ErBAlQFpmHhsPnSMsykBM4tXC5V7VyzPE15Ng\nP0/q1yi+DfYfRkhICCtWrCAgIICffvqpcHhYiOJOAlaIYsqYb2L3yRTCopL44egFcvPNQ8AVnR3p\n72NusN+uYfViNcdqUbm5OX/Xrl3x8fHhr3/9qzTnFyWK/LYKUcwcO3+NsCgD38Sc41KGeY5VpaBT\nk5oE+3vQp6U7FZxL759ufHw8kyZNYvTo0YwfP56JEyfauyQhHkjp/SsVogS5nJHDhhjzHKux564V\nLn+kZsXCBvv1qpa3Y4XWl5+fz2effcabb76Jo6MjY8eOtXdJQjwUCVgh7CTXaGL78Qusi0piZ9xF\njAUN9qu4OjGwjbm7kq9X1TLRR/fo0aNMmDCBiIgI+vfvz7x58/D09LR3WUI8FAlYIWxIa83hpDTC\nogxsPHiO1Exzg31HB8XjzWsT7OdJjxa1i/0cq0XtzJkznD59mi+//JIRI0aUif9UiNJPAlYIG7hw\nLZuvDyQRFmXg5MWMwuXN3SsT7OfJIN961K5ctlr87d+/n5iYGCZNmkT//v2Jj4+ncuXK9i5LiCIj\nASuElWTn5bPt6AXCogz8fDKFghFgqld0ZlDbegT7edKyXpUyd7SWmZnJ3/72Nz7++GMaNGjAmDFj\ncHV1lXAVpY4ErBBFSGtN1NlUwqINfHvoPOnZ5gb75RwVvQqGgLs9Whtnp5LbYP9h7Ny5k9DQUE6f\nPs2UKVN4//33pTm/KLUkYIUoAobUTL6OTmL9gSTOXPp9jlUfTzeC/TwZ2KYe1SuW/Ab7D8NgMNCr\nVy8aNGjA9u3b6d69u71LEsKqJGCFeEDXc4xsOZJMWJSBvfGXC5fXruzCU74eBPt70qyODHsePHiQ\nNm3a4OnpyYYNG+jWrRsVKpTOrlNC3EwCVoj7YDJpws9cJiwqiS1HzpOZmw+As5MDfVq6E+znQacm\nNUv0HKtFJSUlhRdffJFVq1axc+dOunbtyhNPPGHvsoSwGQlYISzw26XrhEUbWB+dRNLV3+dY9W9Q\njWA/T/r71MWtfOlpsP8wtNasXr2aF154gbS0NN555x06dOhg77KEsDkJWCHuIjktm5fWxPxhCNij\navnCOVYb1ZR5SG81ZswYvvjiC4KCgli8eDEtW7a0d0lC2IUErBB3YEjNZNTCCBKuZFLB2ZG+rdwZ\n6u9J+0Y1SmWD/YdhMplQSqGUonv37vj7+/PCCy/g6Fi2GmYIcbN7BqxSqjzwV6CB1nqqUqoJ0FRr\nvcXq1QlhJ79dus6oheGcS8vGx9ONZeMDy/xVwHdy6tQpJk2axJgxY5gwYYI05xeigCVXYiwBFNCp\n4PE5YKbVKhLCzk5eSGfY/L2cS8vGv0E1VoYGSbjehtFo5IMPPqB169YcOHAAZ2fZR0LczJIh4qZa\n65FKqacBtNaZqqy1nhFlRuy5NMYs3seV67l0bFyDhWMDqOgiZ1JudeTIEcaPH09kZCSDBg1izpw5\n1KtXz95lCVGsWPLJkauUcgU0gFKqEZBr1aqEsIOYxKuMXRzBtWwj3R6txbzR/mWu6b6lEhISOHv2\nLKtXr2bYsGFlrt2jEJawJGD/AXwPeCqllgNdgVCrViWEje07c4UJy/aTkWOkT8s6fDrSFxcnCdeb\nRUREcPDgQSZPnswTTzxBfHw8lSpVsndZQhRb9zwHW3Ax09PAJOBrIFBr/aO1CxPCVvacvMTYJRFk\n5Bh5sk09/jfKT8L1JtevX2f69Ol06NCBWbNmkZOTAyDhKsQ93DNglVLbtNYpWusNWutvtNYXlVLb\nbFGcENb207ELTFi+n+w8E8MCPPl4eFvKSRemQtu3b8fHx4ePP/6YqVOnEh0djYuLi73LEqJEuOMQ\nsVLKGXAF6iilKmO+khigClDfBrUJYVWbD5/nhVUHMJo0Yzs04O8DW8r9rTcxGAz06dOHRo0asWvX\nLrp06WLvkoQoUe52DvZ5YDpQG4jl94C9Bsyzcl1CWNXXBwy8vPYgJg1TujzC6/2ay4U6BQ4cOICv\nry+enp5s2rSJrl27Ur58eXuXJUSJc8exMK31x1prL+A1rXV9rbVXwVdLrfUnNqxRiCK1al8C0wvC\n9cUeTSVcC1y4cIHhw4fj5+fHrl27AOjbt6+EqxAP6J5XEWutP1FKNQe8MQ8Z31j+pTULE8Ialv5y\nhnc2HQXgtb7NebZbYztXZH9aa7744gtefPFFMjIyeO+99+jYsaO9yxKixLOkVeJbQG+gObAV6APs\nASRgRYkyZ+cpZn0fB8DfB3oz7rFGdq6oeBg1ahSrV6+mQ4cOLF68mBYtWti7JCFKBUvugx0OtAWi\ntdZjlFJ1gfnWLUuIoqO15uMfTvDp9lMoBf96qjUjAsv2dXo3N+fv3bs3HTp04Pnnn5fm/EIUIUvu\nR8jSWucDxoKriZOBR6xblhBFQ2vNv7Yc59Ptp3B0UHw8rG2ZD9cTJ07QvXt3lixZAsD48eNl5hsh\nrMCSgD2glKqKuel/JLAPiLZqVUIUAZNJ87cNsSzYHY+Tg+J/I30Z7Oth77Lsxmg0MmvWLNq0acOh\nQ4fk4iUhrOyuQ8QFTf3/rrW+CsxWSm0FqmitJWBFsZZv0rwedoivogw4Ozkwb7QfjzevY++y7ObQ\noUNMmDCBqKgonnrqKWbPnk3dunXtXZYQpdpdA1ZrrZVS3wL+BY9P2aQqIR5CXr6J6WsPsungOcqX\nc2RRSACPNalp77LsymAwkJiYyFdffUVwcLDcliSEDVgyRLxPKeX3IBtXSvVVSsUppU4ppV6/wzrD\nlFJHlVKxSim5Mlk8lBxjPs9/Ec2mg+eo5OLE8gmBZTZcf/31V+bNM/eEudGcf+jQoRKuQtiIJQHb\nCXPIximlopVSB5RS9xwiVko5ArOBfpjvoR2plPK+ZZ2mwBvAY1rrlsBf7/snEKJAdl4+kz+PYtvR\nC1RxdWJlaBCBjarbuyyby8jI4MUXX6RTp058+OGHhc35K1asaOfKhChbLLlNZ/ADbjsQOKW1jgdQ\nSq0GBgFHb1pnEjBba50KoLW++IDvJcq46zlGQpdHsjf+MtUrOrNyYhDe9arYuyyb27ZtG5MnTyYh\nIYHnn3+emTNnSnN+IezEkk5Opx9w2x5A4k2PDUDQLes0A1BK/QI4Yr6g6vtbN6SUmgxMBqhfv2zf\nYiH+7Fp2HuOX7ifqbCq1K7vwRWgQTetUtndZNpeYmEj//v1p3Lgxu3fvplOnTvYuSYgyzZrzct3u\nRI++5bET0BToBowEFhXcEvTHF2m9QGsdoLUOqFWrVpEXKkqu1Ou5PLMwgqizqXhULc/aKR3KXLhG\nRUUB4OXlxebNm4mJiZFwFaIYsGbAGgCvmx57Audus84GrXWe1voMEIc5cIW4p5T0HEYuDOdwUhoN\nalRgzZT2NKxZds4zJicn8/TTTxMQEFDYnL9Xr164urre45VCCFuwKGCVUp5Kqe4F37sopSz5FNsP\nNFVKNSqYW3YEsPGWdb4Bbmy3JuYh43hLixdlV3JaNsMX7OV4cjqNa1Vk7ZQOeFarYO+ybEJrzfLl\ny/H29mbTpk3MnDlTmvMLUQxZ0ux/AjANcAMaAw2AOUDPu71Oa21USk3DPEGAI7BEax2rlHoXiNRa\nbyx4rrdS6iiQD8zQWl9+mB9IlH6JVzJ5ZlEECVcyae5emZWhQdSsVHYu5BkxYgRr167lscceY9Gi\nRTRv3tzeJQkhbkNpfetp0VtWUCoG8xXBEVpr34Jlh7TWPjao708CAgJ0ZGSkPd5aFANnLl1n1MJw\nzqdl08bTjeUTAqlawdneZVndzc35ly9fTnp6Os899xwODtY8yyNKE6VUlNY6wN51lCWW/HVma61z\nbzwouL9V7lQXNnfiQjrD5u/lfFo27RpWY2VoUJkI1+PHj9OlSxcWL14MQEhICNOmTZNwFaKYs+Qv\n9Bel1KuAa8F52DXAt9YtS4g/OpKUxvD5e0lJz6Fj4xosnxBIZddy9i7LqvLy8pg5cyZt2rTh6NGj\nVKpUyd4lCSHugyWNJl7FfA8rDocNAAAgAElEQVTqceBFzOdNZT5YYTPRCamELNlHeraR7o/WYu5o\nf1zLle6p1WJiYhg/fjwxMTEMHTqUzz77DHd3d3uXJYS4D5YE7BPAIq31XGsXI8StIuIvM2HZfq7n\n5tOvlTv/HeGLs1PpHxpNTk4mOTmZsLAwhgwZYu9yhBAPwJJPqmHAKaXUUqVUn4JzsEJY3e4TKYQs\n3cf13HwGta3HZyNLd7ju2bOHOXPmANC3b19Onz4t4SpECXbPTyut9RjM96duAiYA8UqpedYuTJRt\nPx69QOjySLLzTAwP8OKjYW1xciyd4Zqens60adPo3Lkzn3zySWFz/goVysZ9vUKUVhZ9Ymmtc4AN\nwDLMDSSGWbEmUcZ9d+g8U1dGkZtvYlzHhvxrSGscHUrnhetbt26lVatWzJkzhxdffJHo6Ghpzi9E\nKWFJo4memLsw9QR+AT4HRlm5LlFGrY828MpXBzFpmNL1EV7v27zUzl+amJjIgAEDaNKkCXv27JFu\nTEKUMpZc5DQVWA38RWudZeV6RBn2ZUQCb35zGK3hrz2b8mKPpqUuXLXW7N+/n8DAQLy8vNiyZQud\nOnWS/sFClEKWnIMdqrVeJ+EqrGnxnjP839fmcH2jX3P+2rNZqQvX8+fPExwcTFBQUGFz/p49e0q4\nClFK3fEIVim1S2vdVSmVyh+nmVOA1lpXt3p1okyYveMU/9kaB8A7T7YkpGND+xZUxLTWLFu2jOnT\np5Odnc3777/PY489Zu+yhBBWdrch4u4F/9a0RSGi7NFa89EPJ/hs+ymUgn8Pac3wdvXtXVaRGzZs\nGOvWraNz584sWrSIZs2a2bskIYQN3HGIWGttKvh2sdY6/+YvYLFtyhOlldaaf353jM+2n8LRQfHJ\n8LalKlzz8/Mxmcx/QgMHDmTOnDns3LlTwlWIMsSS23T+MGtOQaOJdtYpR5QFJpPm/204wqI9Zyjn\nqJg9ypdBbT3sXVaROXbsGJ07dy5szj927FieffZZac4vRBlzx794pdRrBedffZRSVwq+UoEUYLPN\nKhSlSr5J82rYIVaGJ+Ds5MCCMQH0bVXX3mUViby8PN577z3atm1LXFwcbm5u9i5JCGFHdzsHOwv4\nEPgX8PqNhQVDxELct7x8Ey+tieHbQ+cpX86RRSEBPNakdJziP3DgAOPGjePQoUMMHz6cTz/9lNq1\na9u7LCGEHd0tYJtorU8qpVYALW8svHHrhNb6kJVrE6VIjjGfaV8e4IejF6jk4sTS8e1o17D0XIh+\n4cIFLl26xDfffMOgQYPsXY4Qohi4W8C+DkwEZt/mOQ10sUpFotTJys1nysoodp9Iwa18OT6fEEgb\nr6r2Luuh7d69m8OHD/P888/Tt29fTp06Rfny5e1dlhCimLhjwGqtJxb829l25YjSJiPHSOjy/YTH\nX6FGRWdWTAzCu14Ve5f1UK5du8brr7/O3LlzadasGaGhobi4uEi4CiH+4J6XNSqlhiilKhd8/7pS\naq1Sqo31SxMlXVpWHmMXRxAef4U6VVxYM6V9iQ/XzZs307JlS+bPn8/06dOlOb8Q4o4suW/g71rr\ndKVUR2AgsAaYb92yREmXej2XZxaFE51wFY+q5Vk7pQNNale2d1kPJTExkUGDBuHm5savv/7Khx9+\nSMWKFe1dlhCimLIkYG9cNTwAmKO1DgPkv+ziji6mZzNiQThHkq7RoEYF1k7tQIMaJTOItNaEh4cD\n4OXlxbZt24iOjiYoKMjOlQkhijtLAva8Umo25inrNiulnC18nSiDzqdlMWJ+OHEX0mlSuxJrp3TA\no2rJPDd57tw5Bg8eTIcOHQqb83fv3h1nZ2c7VyaEKAksCcphwC7gCa11KubexK/f/SWiLEq8ksmw\n+XuJv3SdFnWrsGZye+pUKXkzxWitWbRoEd7e3mzbto0PPvhAmvMLIe7bPeeD1VpnKKWOAt2UUt2A\nn7XWW6xemShR4lMyeGZRBOfTsmnjVZXPxwfiVqGcvct6IEOHDmX9+vV07dqVRYsW0aRJE3uXJIQo\ngSy5ingasBaoX/C1Vin1nLULEyVHXHI6w+aHcz4tm8CG1Vk5seSF683N+QcPHsy8efPYvn27hKsQ\n4oEprfXdV1DqENBRa51R8LgS8KvW2ueuL7SSgIAAHRkZaY+3FrdxJCmNMYsjSM3Mo1OTmiwY608F\n53sOjBQrR44cITQ0lIkTJzJp0iR7lyOEVSilorTWAfauoyyx5BysAvJuepxXsEyUcVFnUxm5MJzU\nzDweb16bRSEBJSpcc3Nzeeedd/Dz8+P06dNUq1bN3iUJIUoRSz4NVwDhSqkwzME6GFhu1apEsbf3\n9GUmLt9PZm4+/Vq5898Rvjg7lZyLy6Oiohg3bhxHjhxh1KhRfPLJJ9SqVcveZQkhShFLLnKapZTa\nAdxomThVa73fumWJ4mzXiRQmfx5JjtHEU74e/GeoD06OJSdcAS5fvszVq1fZtGkTAwYMsHc5QohS\nyNLxvJyCL1PBv6KM+uHoBZ7/IprcfBMjA7345+DWODiUjDMGO3bs4PDhw7zwwgv07t2bkydP4upa\n8m4jEkKUDJZcRfwmsAqoC3gCXyql3rB2YaL42XTwHM+ujCI338S4jg2Z+VTJCNe0tDSmTJnC448/\nzty5c8nJMf8fUcJVCGFNlozrjQbaaa3f0lq/CQQCY61blihu1kUZeHH1AYwmzbPdGvP2QO/CuYGL\ns02bNuHt7c2iRYt45ZVXiIqKkub8QgibsGSI+Owt6zkB8dYpRxRHK8PP8tY3RwCY3qsZf3m8SYkI\n18TERIKDg2nevDnffPMN7dq1s3dJQogyxJKAzQRilVJbMU+03hvYo5T6CEBrPd2K9Qk7W/RzPO99\ndwyA/3uiOZO7NLZzRXentWbv3r107NixsDl/x44dpX+wEMLmLBki/g74O7AXCAfeBbYDsQVfd6SU\n6quUilNKnVJK3bF/sVJqqFJKK6XkJuhi5H/bTxaG6z8GtSz24WowGHjyySd57LHHCpvzd+vWTcJV\nCGEXltyms/hBNqyUcgRmA70AA7BfKbVRa330lvUqAy8AEQ/yPqLoaa35YFscs3ecRil4P9iHYQFe\n9i7rjkwmEwsXLmTGjBkYjUY++ugjOnXqZO+yhBBlnDVvXgwETmmt47XWucBqYNBt1vsHMAvItmIt\nwkJaa/7x7TFm7ziNo4Pik+Fti3W4AgQHBzN16lTatWvHkSNHeOmll3B0dLR3WUKIMs6aAesBJN70\n2FCwrJBSyhfw0lp/e7cNKaUmK6UilVKRKSkpRV+pAMBk0rz5zRGW/HKGco6K2aP8GNTW494vtAOj\n0VjYnD84OJiFCxfy448/8sgjj9i5MiGEMLM4YJVS93tvw+0uMy2cWUAp5QB8DLx8rw1prRdorQO0\n1gHSzs46jPkmXll3kC8jEnBxcmDB2AD6tnK3d1m3dejQITp06MDChQsBGD16NKGhoSXiymYhRNlh\nSaOJQKXUYeBkweM2SqnPLNi2Abh5bNETOHfT48pAK2CnUuo3oD2wUS50sr28fBMvrolhfXQSFZwd\nWTquHd0frW3vsv4kJyeHt99+G39/f86ePSu9g4UQxZolt+l8CgwAvgHQWh9USnW34HX7gaZKqUZA\nEjACGHXjSa11GlDzxmOl1E7gFa21zEVnQ9l5+Uz7Mpofj12ksosTS8e3I6BhdXuX9Sf79+9n3Lhx\nHD16lDFjxvDxxx9To0YNe5clhBB3ZEnAOmitz94y/JZ/rxdprY0Fk7VvBRyBJVrrWKXUu0Ck1nrj\nA1UsikxWbj6TV0Ty88lLVK1Qjs8nBOLjWdXeZd1WamoqGRkZbN68mX79+tm7HCGEuCdLAjZRKRUI\n6IJbb/4CnLBk41rrzcDmW5b97Q7rdrNkm6JoZOQYmbhsPxFnrlCzkjMrJgbRom4Ve5f1B9u3b+fw\n4cO8+OKL9O7dmxMnTkibQyFEiWHJRU7PAtOB+sAFzOdKn7VmUcK60rLyGLM4gogzV6hTxYXVkzsU\nq3C9evUqkyZNokePHsyfP7+wOb+EqxCiJLlnwGqtL2qtR2itaxZ8jdBaX7JFcaLoXbmey6iF4RxI\nuIpH1fKsndKBJrUr2busQhs2bMDb25slS5bw6quvSnN+IUSJdc8hYqXUQm66veYGrfVkq1QkrOZi\nejajF0Vw4kIGDWtU4MtJ7alXtby9yyqUkJDA008/TYsWLdi4cSMBAXJBuRCi5LLkHOyPN33vCjzF\nHxtIiBLg3NUsnlkUwZlL12lauxJfhAZRu4r950PVWrNnzx46d+5M/fr1+fHHH2nfvr30DxZClHiW\nDBGvuelrOTAE8LZ+aaKoJFzOZNj8vZy5dB3vulVYPbl9sQjXhIQE+vfvT5cuXQqb83fp0kXCVQhR\nKjxIq8RGQIOiLkRYx+mUDIbN34shNYu2XlVZNak9NSrZ95ymyWRizpw5tGzZkt27d/Ppp59Kc34h\nRKljyTnYVH4/B+sAXAHuOPWcKD6OJ19j9KIILmXkEtioOkvGtaOSiyVnBaxryJAhbNiwgV69erFg\nwQIaNmxo75KEEKLI3fXTVpm7S7TB3IkJwKS1/tMFT6L4OWxIY8ySCK5m5tG5aU0WjAmgvLP9Zpgx\nGo04ODjg4ODA8OHDGTRoEOPGjZP+wUKIUuuuQ8QFYfq11jq/4EvCtQSIOnuFUQvDuZqZR88WtVk4\n1r7hevDgQYKCgliwYAEAI0eOZPz48RKuQohSzZJzsPuUUn5Wr0QUib2nLzNm8T7Sc4z0b12XOc/4\n41rOPuGanZ3NW2+9RUBAAAaDAXf34jk7jxBCWMMdh4iVUk5aayPQCZiklDoNXMc8DZ3WWkvoFjM7\n4y4yZUUUOUYTQ3w9mDXUBydHa075e2f79u0jJCSE48ePExISwkcffUT16sVvEgEhhLCWu52D3Qf4\nAYNtVIt4CFtjk5n2ZTR5+ZqRgfX55+BWODjYbwj22rVrZGVl8f3339OnTx+71SGEEPZyt4BVAFrr\n0zaqRTygTQfP8dc1MeSbNOMfa8jfBnjb5fzmtm3biI2N5aWXXqJnz57ExcVJm0MhRJl1t4CtpZSa\nfqcntdYfWaEecZ++ikzktbBDmDQ8160xM/o8avNwTU1NZfr06SxbtoyWLVvy3HPP4eLiIuEqhCjT\n7naCzhGoBFS+w5ewsxV7f2PGOnO4vtK7Ga/2bW7zcF2/fj3e3t6sWLGCN954g8jISAlWIYTg7kew\n57XW79qsEnFfFv0cz3vfHQPgrf4tCO38iM1rSEhIYMSIEbRq1YrNmzfj6+tr8xqEEKK4utsRrNyk\nWAxprfnsp5OF4fqPwa1sGq5a68K+wfXr12f79u1ERERIuAohxC3uFrA9bFaFsIjWmv9sjePDH07g\noOA/Q30Y0952baHPnj1Lv3796NatW2HIdurUiXLlytmsBiGEKCnuGLBa6yu2LETcndaad789ypyd\np3F0UHwywpenA7xs8t4mk4n//e9/tGzZkj179vDZZ5/RuXNnm7y3EEKUVPbv/C7uyWTSvPnNEVbt\nS8DZ0YH/jfKld0vbdUUaPHgwmzZtok+fPsyfP58GDWQyJSGEuBcJ2GLOmG/i1XWHWH8gCRcnB+aP\n8afbo7Wt/r55eXk4Ojri4ODAyJEjGTp0KGPGjJH+wUIIYSH79NETFsk1mnhh9QHWH0iigrMjy8YH\n2iRco6OjCQwMZN68eYC5Of/YsWMlXIUQ4j5IwBZT2Xn5PPdFFJsPJ1PZxYkVE4Po0LiGVd8zKyuL\nN954g8DAQJKTk/Hyss05XiGEKI1kiLgYysrNZ/KKSH4+eYmqFcqxYkIQrT3drPqe4eHhhISEcOLE\nCSZMmMAHH3xAtWrVrPqeQghRmknAFjMZOUYmLN3Pvt+uULOSCytDA2nuXsXq73v9+nXy8vL44Ycf\n6Nmzp9XfTwghSjsJ2GIkLTOPkKX7iEm8insVV76YFETjWpWs9n7ff/89sbGxvPzyy/To0YPjx4/j\n7OxstfcTQoiyRM7BFhOXM3IYuTCcmMSreFYrz9opHawWrpcvXyYkJIR+/fqxfPlycnNzASRchRCi\nCEnAFgMXr2UzYkE4R89fo1HNiqyd0oH6NSoU+ftorVm3bh3e3t58+eWXvPXWW+zfv1+CVQghrECG\niO0s6WoWzywM57fLmTSrU4mVoUHUruxqlfdKSEhg1KhR+Pj4sG3bNtq0aWOV9xFCCCFHsHZ19vJ1\nhs3by2+XM2lZrwqrJ3co8nDVWrN9+3YAGjRowM6dOwkPD5dwFUIIK5OAtZNTFzMYNn8vSVez8K1f\nlS8ntad6xaIdqj1z5gy9e/emR48ehc35O3bsiJOTDFwIIYS1ScDawbHz1xg+fy8XruUQ1Kg6KyYG\n4Va+6Gakyc/P57///S+tWrUiIiKCuXPnSnN+IYSwMTmUsbFDhquMXbKPq5l5dG5akwVjAijv7Fik\n7zFo0CC+++47nnjiCebNmycdmYQQwg4kYG0o8rcrjF+6n/QcIz1b1GH2M764OBVNuN7cnH/MmDGM\nHDmSUaNGSf9gIYSwE6sOESul+iql4pRSp5RSr9/m+elKqaNKqUNKqZ+UUqV2HrRfT11izOJ9pOcY\n6e9Tl7mj/YosXCMjIwkICGDu3LkADB8+nGeeeUbCVQgh7MhqAauUcgRmA/0Ab2CkUsr7ltUOAAFa\nax9gHTDLWvXY0464i4xftp+svHyC/Tz5dIQv5RwfftdnZWXx2muvERQUREpKiszTKoQQxYg1j2AD\ngVNa63itdS6wGhh08wpa6x1a68yCh+GApxXrsYvvjyQz+fNIcowmngmqz3+G+uDo8PBHlnv37qVN\nmzbMmjWLCRMmcPToUQYMGFAEFQshhCgK1jwH6wEk3vTYAATdZf2JwJbbPaGUmgxMBqhfv35R1Wd1\nG2KSmL72IPkmzYTHGvH/BrQosmHbrKwsTCYTP/74Iz169CiSbQohhCg61gzY2yWJvu2KSo0GAoCu\nt3tea70AWAAQEBBw220UN2v3J/La+kNoDdO6N+Hl3s0eOlw3b95MbGwsM2bM4PHHH+fYsWOUK1d0\nt/cIIYQoOtYcIjYAN98f4gmcu3UlpVRP4E3gSa11jhXrsZnP9/7Gq2HmcH2ldzNe6fPoQ4XrpUuX\nGD16NP379+eLL74obM4v4SqEEMWXNQN2P9BUKdVIKeUMjAA23ryCUsoXmI85XC9asRabWbD7NH/b\nEAvAW/1bMO3xpg+8La01q1evpkWLFqxdu5a3336bffv2SXN+IYQoAaw2RKy1NiqlpgFbAUdgidY6\nVin1LhCptd4I/AeoBHxVcISXoLV+0lo1WZPWmk9/OsXHP54A4L3BrRjd/uGu6k1ISCAkJIQ2bdqw\nePFiWrduXRSlCiGEsAGldYk4pVkoICBAR0ZG2ruMP9Ba8/73cczbdRoHBbOGtmGo/4NdEK215qef\nfqJnz54AhIeH065dOxwdi7bbkxCibFFKRWmtA+xdR1kivYgfksmkeWfTUebtOo2Tg+LTkb4PHK6n\nT5+mR48e9OrVq7A5f/v27SVchRCiBJKAfQj5Js3/fX2YZb/+hrOjA3NH+zPAp979byc/n48++ojW\nrVsTFRXF/PnzpTm/EEKUcNKL+AEZ803MWHeIrw8k4VrOgQVjAujSrNYDbWvgwIFs2bKFAQMGMHfu\nXDw9S12/DSGEKHMkYB9ArtHEi6sPsOVIMhWcHVkyrh3tH6lxf9vIzcXJyQkHBwfGjRvHmDFjGDFi\nhPQPFkKIUkKGiO9Tdl4+U1dGseVIMpVdnVgZGnTf4bpv3z78/f2ZM2cOAMOGDWPkyJESrkIIUYpI\nwN6HzFwjocsj2X78ItUqlGPVpPb41a9m+eszM3n55Zfp0KEDqampNG7c2IrVCiGEsCcZIrZQenYe\nE5btZ/9vqdSs5MIXoUE86l7Z4tfv2bOHkJAQ4uPjmTJlCu+//z5ubm5WrFgIIYQ9ScBa4GpmLiFL\n9nHQkEZdN1e+CA3ikVqV7msbNyZE37FjB926dbNOoUIIIYoNCdh7uJyRw+jF+zh2/hpe1cvzZWh7\nvKpXsOi1mzZt4tixY7z66qt0796do0eP4uQku1wIIcoCOQd7FxeuZTN8QTjHzl/jkZoVWTulg0Xh\nmpKSwqhRo3jyySdZtWpVYXN+CVchhCg7JGDvwJCaybD5ezl1MYNH61RmzZQO1HUrf9fXaK358ssv\nadGiBevWrePdd98lIiJCmvMLIUQZJIdUt3H28nVGLYwg6WoWrTyqsGJCENUq3jskExISGD9+PL6+\nvixevJiWLVvaoFohhBDFkRzB3uLUxXSenreXpKtZ+NWvyheh7e8ariaTia1btwLQoEEDfv75Z375\n5RcJVyGEKOMkYG9y9Nw1hs8P52J6Du0fqc6KiUG4lb/zpOYnT57k8ccfp2/fvuzevRuAwMBAac4v\nhBBCAvaGg4lXGbkwnMvXc+nSrBZLxwVS0eX2I+hGo5H//Oc/+Pj4EBMTw+LFi6U5vxBCiD+Qc7DA\n/t+uMH7pfjJyjPTyrsP/Rvni4nTno9ABAwawdetWBg0axJw5c6hX7/5n0BFCCFG6lfmA/eXUJUKX\nR5KVl88An7p8PLwt5Rz/fGCfk5NDuXLlcHBwIDQ0lAkTJvD0009L/2AhhBC3VaaHiHccv8j4ZfvJ\nystnqL8n/x3he9twDQ8Px8/Pj9mzZwMwdOhQhg0bJuEqhBDijspswH5/5DyTV0SSazQxun19ZgX7\n4Ojwx8C8fv06L730Eh07diQ9PZ2mTZvaqVohhBAlTZkcIt4Qk8T0tQfJN2lCOzXizf4t/nQ0+vPP\nPxMSEsKZM2d47rnn+Ne//kWVKlXsVLEQQoiSpswF7Jr9Cby+/jBaw18eb8L0Xs1uO9RrNBopV64c\nu3btokuXLnaoVAghRElWpgJ2+a+/8fbGWABm9HmU57s3+cPz33zzDceOHeONN96ge/fuxMbGSv9g\nIYQQD6TMnIOdt+t0Ybj+bYD3H8L1woULDBs2jKeeeop169ZJc34hhBAPrdQHrNaaj384wb+3HEcp\nmPlUayZ0alT43IoVK/D29mbDhg3885//JDw8XJrzCyGEeGil+hBNa82/vz/O/F3xOCj44Ok2DPHz\nLHw+ISGB0NBQAgICWLx4Mc2bN7djtUIIIUqTUnsEazJp/r4xlvm74nFyUHw20o8hfp6YTCa2bNkC\nmJvz//LLL+zevVvCVQghRJEqlQGbb9K8sf4wy/eexdnRgXmj/envU5cTJ07QrVs3nnjiCXbt2gVA\nQECANOcXQghR5EpdwBrzTUxfG8OayERcyzmweFwA3ZrV4P3338fHx4fDhw+zdOlSufVGCCGEVZWq\nc7C5RhMvrDrA97HJVHR2ZMm4dgQ9UoM+ffqwbds2hgwZwuzZs3F3d7d3qUIIIUq5UhOw2Xn5PLsy\nih1xKVRxdWLBKB8CGlQFYPLkyUyePJng4GA7VymEEKKsKBVDxJm5RiYs28+OuBSqVSjHjLYw7snu\nhc35g4ODJVyFEELYVIkP2PTsPMYu3sevpy9T3dlE64T1hAzpR3Z2Ni1atLB3eUIIIcqoEj1EfDUz\nl7FL9nHIkEbFy3Fc3PpfVhoSmTZtGjNnzqRSpUr2LlEIIUQZVWID9lJGDqMXRXA8OR2v6uV54bGW\nvP1LRdb8/DOPPfaYvcsTQghRxpXIgL1wLZtRC8M5vOcHKuVc4KtV/8PdzZUhTxyWe1qFEEIUC1Y9\nB6uU6quUilNKnVJKvX6b512UUmsKno9QSjW81zZz800M/uA7whe8Sco3M3FLjqF6efOPIeEqhBCi\nuLBawCqlHIHZQD/AGxiplPK+ZbWJQKrWugnwMfD+vbZ74uw59s0KITs+kr+98x77IvZKc34hhBDF\njjWHiAOBU1rreACl1GpgEHD0pnUGAX8v+H4d8D+llNJa6zttNCf1IjUat2br+i/w92llncqFEEKI\nh2TNgPUAEm96bACC7rSO1tqolEoDagCXbl5JKTUZmFzwMOfy6UNHAtq0tkrRJUxNbtlXZZjsi9/J\nvvid7IvfPWrvAsoaawasus2yW49MLVkHrfUCYAGAUipSax3w8OWVfLIvfif74neyL34n++J3SqlI\ne9dQ1ljzIicD4HXTY0/g3J3WUUo5AW7AFSvWJIQQQtiENQN2P9BUKdVIKeUMjAA23rLORiCk4Puh\nwPa7nX8VQgghSgqrDREXnFOdBmwFHIElWutYpdS7QKTWeiOwGFihlDqF+ch1hAWbXmCtmksg2Re/\nk33xO9kXv5N98TvZFzam5IBRCCGEKHolvtm/EEIIURxJwAohhBBWUGwD1hptFksqC/bFdKXUUaXU\nIaXUT0qpBvao0xbutS9uWm+oUkorpUrtLRqW7Aul1LCC341YpdSXtq7RViz4G6mvlNqhlDpQ8Hfy\nhD3qtDal1BKl1EWl1JE7PK+UUp8W7KdDSik/W9dYpmiti90X5ouiTgOPAM7AQcD7lnWeA+YVfD8C\nWGPvuu24L7oDFQq+f7Ys74uC9SoDu4FwIMDeddvx96IpcACoVvC4tr3rtuO+WAA8W/C9N/Cbveu2\n0r7oAvgBR+7w/BPAFsw9CNoDEfauuTR/Fdcj2MI2i1rrXOBGm8WbDQKWF3y/DuihlLpd44qS7p77\nQmu9Q2udWfAwHPM9x6WRJb8XAP8AZgHZtizOxizZF5OA2VrrVACt9UUb12grluwLDVQp+N6NP9+T\nXyporXdz914Cg4DPtVk4UFUpVdc21ZU9xTVgb9dm0eNO62itjcCNNouljSX74mYTMf8PtTS6575Q\nSvkCXlrrb21ZmB1Y8nvRDGimlPpFKRWulOprs+psy5J98XdgtFLKAGwG/mKb0oqd+/08EQ+huM4H\nW2RtFksBi39OpdRoIADoatWK7Oeu+0Ip5YB5VqZxtirIjiz5vXDCPEzcDfOoxs9KqVZa66tWrs3W\nLNkXI4FlWusPlVIdMA71prcAAAVgSURBVN9/30prbbJ+ecVKWfncLBaK6xGstFn8nSX7AqVUT+BN\n4EmtdY6NarO1e+2LykAr4P+3d3chVlVhGMf/DzalGAhhXYUOfWgm6ZRmaRSCZVAgWNZc+JFSF3NT\nIdpFVBZJHyZkmYQShBIlKlhZBCalaY6pFKOZJpaJFFLZhZUFxfB2sdbgGeccPWOzZ86Mzw8Os2ad\nvfdaZ8PMO2vtNevdIukI6RnThj660Knan5H3I+LfiPgBOEgKuH1NNffiQWAtQETsAPqTEgGcb6r6\nfWJdo1YDrLdZPOWs9yJPi64gBde++pwNznIvIuJERAyOiPqIqCc9j54SEX1xk/NqfkbeIy2AQ9Jg\n0pTx4W7tZfeo5l4cBSYBSBpBCrC/dmsva8MGYFZeTXwzcCIijvV0p/qqmpwijuK2Wex1qrwXi4GL\ngXV5ndfRiJjSY50uSJX34rxQ5b3YCEyWtB9oBR6LiN96rtfFqPJezAPekDSXNCU6uy/+QS5pNemR\nwOD8vPlpoA4gIpaTnj/fBXwH/AXM6Zmenh+8VaKZmVkBanWK2MzMrFdzgDUzMyuAA6yZmVkBHGDN\nzMwK4ABrZmZWAAdYq1mSWiW1lLzqz3BsfaUMIt1N0lhJS3N5oqQJJe81SZrVjX1p6KuZY8xqXU3+\nH6xZ9ndENPR0Jzorb2zRtrnFROBPoDm/t7yr25N0Qd6Pu5wG0vaZH3V1u2Z2Zh7BWq+SR6rbJH2V\nXxPKHDNS0q486t0r6epcP6OkfoWkfmXOPSJpUT5ul6Srcv1QpVy7bTl3h+T6+yTtk7RH0tZcN1HS\nh3nE3QTMzW3eKukZSfMljZC067TPtTeXx0j6TNKXkjaWy3YiaaWklyVtBhZJGiepWSnfabOk4XlX\no2eBxtx+o6SBSjlDd+djy2UjMrMu4ABrtWxAyfTwu7nuF+COiLgBaASWljmvCXg1j37HAj/m7fEa\ngVtyfSswvUK7v0fEOGAZ8EquW0ZK8zUKeLuk3QXAnRExGmi3e1ZEHAGWA0sioiEitpW8dwC4UNIV\nuaoRWCupDngNmBYRY4A3gecq9HMYcHtEzAO+BW6LiOtzn57PqdsWkPIDN0TEGtJ+1Z9GxI2kbRQX\nSxpY4fpm9j94ithqWbkp4jpgmaS2IDmszHk7gCckXQ6sj4hDkiYBY4DdeTvJAaRgXc7qkq9Lcnk8\ncE8uv0XKNwuwHVgpaS2wvjMfjrT5/P3Ai6QA2wgMJyUs2JT72Q+otFfsuohozeVBwKo8Wg/y9nhl\nTAamSJqfv+8PDAEOdLLvZnYWDrDW28wFfgZGk2ZgOiRVj4h3JO0E7gY2SnqIlKZrVUQ8XkUbUaHc\n4ZiIaJJ0U26rJQf+aq0h7R+9Pl0qDkm6DvgmIsZXcf7JkvJCYHNETM1T01sqnCPg3og42Il+mtk5\n8BSx9TaDgGM5j+dM0givnTztejgilpKyh4wCPgGmSbosH3OJpKEV2mgs+bojl5s5lVBiOvB5vs6V\nEbEzIhYAx2mfCgzgD1IavQ4i4nvSKPwpUrCFlFLuUqWcpUiqkzSyQj9LDQJ+yuXZZ2h/I/Cw8vBY\nKROTmRXAAdZ6m9eBByR9QZoePlnmmEZgn6QW4BrSs9P9wJPAx3kx0Sagw+Kh7KI8An6UNGIGeASY\nk8+dmd+D9Azz6/wvQluBPadd6wNgatsipzJtrQFmcCpX6T+k9IuLJO0BWoAOC7nKeAl4QdJ22v/R\nsRm4tm2RE2mkWwfszX1eWMW1zewcOJuOWQmlRO1jI+J4T/fFzHo3j2DNzMwK4BGsmZlZATyCNTMz\nK4ADrJmZWQEcYM3MzArgAGtmZlYAB1gzM7MC/AcGqv1qOs/rQgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10aba1208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC (Voting): 0.613084302919\n"
     ]
    }
   ],
   "source": [
    "#full feature set\n",
    "voting_best = VotingClassifier(estimators=[\n",
    "    (\"AdaBoost Random Forest\", AdaBoostClassifier(base_estimator=RandomForestClassifier())),\n",
    "    (\"AdaBoost Decision Tree\", AdaBoostClassifier(base_estimator=DecisionTreeClassifier())),\n",
    "    (\"Random Forest (after random search)\", RandomForestClassifier(bootstrap = True, criterion = \"entropy\", max_features = 5, min_samples_split = 7, n_estimators = 2)),\n",
    "    (\"Support Vector Machine (after random search)\", SVC(degree = 3, kernel = \"poly\", shrinking = True)),\n",
    "    (\"Naive Bayes\", GaussianNB()),\n",
    "], voting='hard')\n",
    "\n",
    "score_classifiers([\n",
    "    (voting_best, \"Voting\")\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdgAAAEWCAYAAADFO4ZdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10af5d4a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC (Voting): 0.720569515053\n"
     ]
    }
   ],
   "source": [
    "#limited feature set\n",
    "voting_best = VotingClassifier(estimators=[\n",
    "    (\"AdaBoost Random Forest\", AdaBoostClassifier(base_estimator=RandomForestClassifier())),\n",
    "    (\"AdaBoost Decision Tree\", AdaBoostClassifier(base_estimator=DecisionTreeClassifier())),\n",
    "    (\"Random Forest (after random search)\", RandomForestClassifier(bootstrap = True, criterion = \"entropy\", max_features = 5, min_samples_split = 7, n_estimators = 2)),\n",
    "    (\"Support Vector Machine (after random search)\", SVC(degree = 3, kernel = \"poly\", shrinking = True)),\n",
    "    (\"Naive Bayes\", GaussianNB()),\n",
    "], voting='hard')\n",
    "\n",
    "score_classifiers([\n",
    "    (voting_best, \"Voting\")\n",
    "], only_important_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'weights': array([ 4, 16, 29,  9, 15])}\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiEAAAEWCAYAAACwgEcPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10a8ca128>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC (Best Voting classifier): 0.624028142199\n"
     ]
    }
   ],
   "source": [
    "classifier = VotingClassifier(estimators=[\n",
    "    (\"AdaBoost Random Forest\", AdaBoostClassifier(base_estimator=RandomForestClassifier())),\n",
    "    (\"AdaBoost Decision Tree\", AdaBoostClassifier(base_estimator=DecisionTreeClassifier())),\n",
    "    (\"Random Forest (after random search)\", RandomForestClassifier(bootstrap = True, criterion = \"entropy\", max_features = 5, min_samples_split = 7, n_estimators = 2)),\n",
    "    (\"Support Vector Machine (after random search)\", SVC(degree = 3, kernel = \"poly\", shrinking = True)),\n",
    "    (\"Naive Bayes\", GaussianNB()),\n",
    "], voting='hard')\n",
    "param_grid = {\n",
    "        'weights': np.random.randint(1, 40, size=(1000, 5))\n",
    "}\n",
    "\n",
    "random_search = RandomizedSearchCV(estimator=classifier, param_distributions=param_grid, n_iter=1000)\n",
    "random_search.fit(encoded_train_set.toarray(), encoded_train_set_labels)\n",
    "best_classifier = random_search.best_estimator_\n",
    "\n",
    "print(random_search.best_params_)\n",
    "score_classifiers([\n",
    "    (best_classifier, \"Best Voting classifier\")\n",
    "], encoded_train_set.toarray())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'weights': array([30, 21, 24,  9, 22])}\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiEAAAEWCAYAAACwgEcPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10aeeb7b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC (Best Voting classifier): 0.700582974549\n"
     ]
    }
   ],
   "source": [
    "classifier = VotingClassifier(estimators=[\n",
    "    (\"AdaBoost Random Forest\", AdaBoostClassifier(base_estimator=RandomForestClassifier())),\n",
    "    (\"AdaBoost Decision Tree\", AdaBoostClassifier(base_estimator=DecisionTreeClassifier())),\n",
    "    (\"Random Forest (after random search)\", RandomForestClassifier(bootstrap = True, criterion = \"entropy\", max_features = 5, min_samples_split = 7, n_estimators = 2)),\n",
    "    (\"Support Vector Machine (after random search)\", SVC(degree = 3, kernel = \"poly\", shrinking = True)),\n",
    "    (\"Naive Bayes\", GaussianNB()),\n",
    "], voting='hard')\n",
    "param_grid = {\n",
    "        'weights': np.random.randint(1, 40, size=(500, 5))\n",
    "}\n",
    "\n",
    "random_search = RandomizedSearchCV(estimator=classifier, param_distributions=param_grid, n_iter=500)\n",
    "random_search.fit(only_important_features, encoded_train_set_labels)\n",
    "best_classifier = random_search.best_estimator_\n",
    "\n",
    "print(random_search.best_params_)\n",
    "score_classifiers([\n",
    "    (best_classifier, \"Best Voting classifier\")\n",
    "], only_important_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [],
   "source": [
    "# mistake during encoding... removing 3 entries with unknown locations\n",
    "test_set = test_set[~test_set[\"Candidate Native location\"].isin(['Baddi','Mysore','Panjim'])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfYAAAEWCAYAAACUr7U+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10afef898>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC (Naive Bayes): 0.79375\n"
     ]
    }
   ],
   "source": [
    "# The final scoring using the test set\n",
    "test_set_features = test_set.drop(\"Observed Attendance\", axis = 1)\n",
    "test_set_label = test_set[\"Observed Attendance\"].copy()\n",
    "\n",
    "encoded_test_set = encoding.transform(test_set_features)\n",
    "encoded_test_set_labels = label_encoder.transform(test_set_label)\n",
    "\n",
    "selected_test_set = encoded_test_set.toarray()[:, feature_selection.support_]\n",
    "\n",
    "score_classifiers([\n",
    "    (GaussianNB(), \"Naive Bayes\") #and all of that just to use a Naive Bayes classifier with default parameters\n",
    "], selected_test_set, encoded_test_set_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}