Starbucks_Capstone_notebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Starbucks Capstone Challenge\n",
    "\n",
    "### Introduction\n",
    "\n",
    "This data set contains simulated data that mimics customer behavior on the Starbucks rewards mobile app. Once every few days, Starbucks sends out an offer to users of the mobile app. An offer can be merely an advertisement for a drink or an actual offer such as a discount or BOGO (buy one get one free). Some users might not receive any offer during certain weeks. \n",
    "\n",
    "Not all users receive the same offer, and that is the challenge to solve with this data set.\n",
    "\n",
    "Every offer has a validity period before the offer expires. As an example, a BOGO offer might be valid for only 5 days. You'll see in the data set that informational offers have a validity period even though these ads are merely providing information about a product; for example, if an informational offer has 7 days of validity, you can assume the customer is feeling the influence of the offer for 7 days after receiving the advertisement."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Sets\n",
    "\n",
    "The data is contained in three files:\n",
    "\n",
    "* portfolio.json - containing offer ids and meta data about each offer (duration, type, etc.)\n",
    "* profile.json - demographic data for each customer\n",
    "* transcript.json - records for transactions, offers received, offers viewed, and offers completed\n",
    "\n",
    "Here is the schema and explanation of each variable in the files:\n",
    "\n",
    "**portfolio.json**\n",
    "* id (string) - offer id\n",
    "* offer_type (string) - type of offer ie BOGO, discount, informational\n",
    "* difficulty (int) - minimum required spend to complete an offer\n",
    "* reward (int) - reward given for completing an offer\n",
    "* duration (int) - time for offer to be open, in days\n",
    "* channels (list of strings)\n",
    "\n",
    "**profile.json**\n",
    "* age (int) - age of the customer \n",
    "* became_member_on (int) - date when customer created an app account\n",
    "* gender (str) - gender of the customer (note some entries contain 'O' for other rather than M or F)\n",
    "* id (str) - customer id\n",
    "* income (float) - customer's income\n",
    "\n",
    "**transcript.json**\n",
    "* event (str) - record description (ie transaction, offer received, offer viewed, etc.)\n",
    "* person (str) - customer id\n",
    "* time (int) - time in hours since start of test. The data begins at time t=0\n",
    "* value - (dict of strings) - either an offer id or transaction amount depending on the record"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ⬇ Install"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "!conda update pandas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 🌴 Environment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import math\n",
    "import json\n",
    "import os\n",
    "import sagemaker\n",
    "from sagemaker import image_uris, model_uris, script_uris\n",
    "from sagemaker import hyperparameters\n",
    "from sagemaker.estimator import Estimator\n",
    "from sagemaker.utils import name_from_base\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "PROCESSED_DATA_DIR = 'processed_data'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 💿 Data Overview"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read in the json files\n",
    "portfolio = pd.read_json('data/portfolio.json', orient='records', lines=True)\n",
    "profile = pd.read_json('data/profile.json', orient='records', lines=True)\n",
    "transcript = pd.read_json('data/transcript.json', orient='records', lines=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 📂 Portfolio"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>reward</th>\n",
       "      <th>channels</th>\n",
       "      <th>difficulty</th>\n",
       "      <th>duration</th>\n",
       "      <th>offer_type</th>\n",
       "      <th>id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>[email, mobile, social]</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>ae264e3637204a6fb9bb56bc8210ddfd</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10</td>\n",
       "      <td>[web, email, mobile, social]</td>\n",
       "      <td>10</td>\n",
       "      <td>5</td>\n",
       "      <td>bogo</td>\n",
       "      <td>4d5c57ea9a6940dd891ad53e9dbe8da0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>[web, email, mobile]</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>informational</td>\n",
       "      <td>3f207df678b143eea3cee63160fa8bed</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5</td>\n",
       "      <td>[web, email, mobile]</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>9b98b8c7a33c4b65b9aebfe6a799e6d9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>[web, email]</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3</td>\n",
       "      <td>[web, email, mobile, social]</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>discount</td>\n",
       "      <td>2298d6c36e964ae4a3e7e9706d1fb8c2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2</td>\n",
       "      <td>[web, email, mobile, social]</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>fafdcd668e3743c1bb461111dcafc2a4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0</td>\n",
       "      <td>[email, mobile, social]</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>informational</td>\n",
       "      <td>5a8bc65990b245e5a138643cd4eb9837</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>5</td>\n",
       "      <td>[web, email, mobile, social]</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>bogo</td>\n",
       "      <td>f19421c1d4aa40978ebb69ca19b0e20d</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2</td>\n",
       "      <td>[web, email, mobile]</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>discount</td>\n",
       "      <td>2906b810c7d4411798c6938adc9daaa5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   reward                      channels  difficulty  duration     offer_type  \\\n",
       "0      10       [email, mobile, social]          10         7           bogo   \n",
       "1      10  [web, email, mobile, social]          10         5           bogo   \n",
       "2       0          [web, email, mobile]           0         4  informational   \n",
       "3       5          [web, email, mobile]           5         7           bogo   \n",
       "4       5                  [web, email]          20        10       discount   \n",
       "5       3  [web, email, mobile, social]           7         7       discount   \n",
       "6       2  [web, email, mobile, social]          10        10       discount   \n",
       "7       0       [email, mobile, social]           0         3  informational   \n",
       "8       5  [web, email, mobile, social]           5         5           bogo   \n",
       "9       2          [web, email, mobile]          10         7       discount   \n",
       "\n",
       "                                 id  \n",
       "0  ae264e3637204a6fb9bb56bc8210ddfd  \n",
       "1  4d5c57ea9a6940dd891ad53e9dbe8da0  \n",
       "2  3f207df678b143eea3cee63160fa8bed  \n",
       "3  9b98b8c7a33c4b65b9aebfe6a799e6d9  \n",
       "4  0b1e1539f2cc45b7b9fa7c272da2e1d7  \n",
       "5  2298d6c36e964ae4a3e7e9706d1fb8c2  \n",
       "6  fafdcd668e3743c1bb461111dcafc2a4  \n",
       "7  5a8bc65990b245e5a138643cd4eb9837  \n",
       "8  f19421c1d4aa40978ebb69ca19b0e20d  \n",
       "9  2906b810c7d4411798c6938adc9daaa5  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "portfolio.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Convert the channels column to multiple boolean columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "portfolio_df = portfolio.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('O')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "portfolio_df['channels'].dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([\"['email', 'mobile', 'social']\",\n",
       "       \"['web', 'email', 'mobile', 'social']\",\n",
       "       \"['web', 'email', 'mobile']\", \"['web', 'email']\"], dtype=object)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "portfolio_df['channels'] = portfolio_df['channels'].astype(str)\n",
    "portfolio_df['channels'].unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the above analysis it appears that we have 4 unique channels which are web, email, mobile and social"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "vals_dict = {\n",
    "    'email': [],\n",
    "    'mobile': [],\n",
    "    'web': [],\n",
    "    'social': []\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(len(portfolio_df)):\n",
    "    channels_lst = portfolio_df.loc[i, 'channels']\n",
    "    for channel in vals_dict.keys():\n",
    "        if channel in channels_lst:\n",
    "            vals_dict[channel].append(True)\n",
    "        else:\n",
    "            vals_dict[channel].append(False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>email</th>\n",
       "      <th>mobile</th>\n",
       "      <th>web</th>\n",
       "      <th>social</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   email  mobile    web  social\n",
       "0   True    True  False    True\n",
       "1   True    True   True    True\n",
       "2   True    True   True   False\n",
       "3   True    True   True   False\n",
       "4   True   False   True   False"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_df = pd.DataFrame(vals_dict)\n",
    "temp_df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>reward</th>\n",
       "      <th>channels</th>\n",
       "      <th>difficulty</th>\n",
       "      <th>duration</th>\n",
       "      <th>offer_type</th>\n",
       "      <th>id</th>\n",
       "      <th>email</th>\n",
       "      <th>mobile</th>\n",
       "      <th>web</th>\n",
       "      <th>social</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>['email', 'mobile', 'social']</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>ae264e3637204a6fb9bb56bc8210ddfd</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10</td>\n",
       "      <td>['web', 'email', 'mobile', 'social']</td>\n",
       "      <td>10</td>\n",
       "      <td>5</td>\n",
       "      <td>bogo</td>\n",
       "      <td>4d5c57ea9a6940dd891ad53e9dbe8da0</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>['web', 'email', 'mobile']</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>informational</td>\n",
       "      <td>3f207df678b143eea3cee63160fa8bed</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5</td>\n",
       "      <td>['web', 'email', 'mobile']</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>9b98b8c7a33c4b65b9aebfe6a799e6d9</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>['web', 'email']</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   reward                              channels  difficulty  duration  \\\n",
       "0      10         ['email', 'mobile', 'social']          10         7   \n",
       "1      10  ['web', 'email', 'mobile', 'social']          10         5   \n",
       "2       0            ['web', 'email', 'mobile']           0         4   \n",
       "3       5            ['web', 'email', 'mobile']           5         7   \n",
       "4       5                      ['web', 'email']          20        10   \n",
       "\n",
       "      offer_type                                id  email  mobile    web  \\\n",
       "0           bogo  ae264e3637204a6fb9bb56bc8210ddfd   True    True  False   \n",
       "1           bogo  4d5c57ea9a6940dd891ad53e9dbe8da0   True    True   True   \n",
       "2  informational  3f207df678b143eea3cee63160fa8bed   True    True   True   \n",
       "3           bogo  9b98b8c7a33c4b65b9aebfe6a799e6d9   True    True   True   \n",
       "4       discount  0b1e1539f2cc45b7b9fa7c272da2e1d7   True   False   True   \n",
       "\n",
       "   social  \n",
       "0    True  \n",
       "1    True  \n",
       "2   False  \n",
       "3   False  \n",
       "4   False  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "portfolio_df = pd.concat([portfolio_df, temp_df], axis=1)\n",
    "portfolio_df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "portfolio_df = portfolio_df.drop('channels', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>reward</th>\n",
       "      <th>difficulty</th>\n",
       "      <th>duration</th>\n",
       "      <th>offer_type</th>\n",
       "      <th>email</th>\n",
       "      <th>mobile</th>\n",
       "      <th>web</th>\n",
       "      <th>social</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ae264e3637204a6fb9bb56bc8210ddfd</th>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4d5c57ea9a6940dd891ad53e9dbe8da0</th>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>5</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3f207df678b143eea3cee63160fa8bed</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>informational</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9b98b8c7a33c4b65b9aebfe6a799e6d9</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0b1e1539f2cc45b7b9fa7c272da2e1d7</th>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2298d6c36e964ae4a3e7e9706d1fb8c2</th>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fafdcd668e3743c1bb461111dcafc2a4</th>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5a8bc65990b245e5a138643cd4eb9837</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>informational</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>f19421c1d4aa40978ebb69ca19b0e20d</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2906b810c7d4411798c6938adc9daaa5</th>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  reward  difficulty  duration     offer_type  \\\n",
       "id                                                                              \n",
       "ae264e3637204a6fb9bb56bc8210ddfd      10          10         7           bogo   \n",
       "4d5c57ea9a6940dd891ad53e9dbe8da0      10          10         5           bogo   \n",
       "3f207df678b143eea3cee63160fa8bed       0           0         4  informational   \n",
       "9b98b8c7a33c4b65b9aebfe6a799e6d9       5           5         7           bogo   \n",
       "0b1e1539f2cc45b7b9fa7c272da2e1d7       5          20        10       discount   \n",
       "2298d6c36e964ae4a3e7e9706d1fb8c2       3           7         7       discount   \n",
       "fafdcd668e3743c1bb461111dcafc2a4       2          10        10       discount   \n",
       "5a8bc65990b245e5a138643cd4eb9837       0           0         3  informational   \n",
       "f19421c1d4aa40978ebb69ca19b0e20d       5           5         5           bogo   \n",
       "2906b810c7d4411798c6938adc9daaa5       2          10         7       discount   \n",
       "\n",
       "                                  email  mobile    web  social  \n",
       "id                                                              \n",
       "ae264e3637204a6fb9bb56bc8210ddfd   True    True  False    True  \n",
       "4d5c57ea9a6940dd891ad53e9dbe8da0   True    True   True    True  \n",
       "3f207df678b143eea3cee63160fa8bed   True    True   True   False  \n",
       "9b98b8c7a33c4b65b9aebfe6a799e6d9   True    True   True   False  \n",
       "0b1e1539f2cc45b7b9fa7c272da2e1d7   True   False   True   False  \n",
       "2298d6c36e964ae4a3e7e9706d1fb8c2   True    True   True    True  \n",
       "fafdcd668e3743c1bb461111dcafc2a4   True    True   True    True  \n",
       "5a8bc65990b245e5a138643cd4eb9837   True    True  False    True  \n",
       "f19421c1d4aa40978ebb69ca19b0e20d   True    True   True    True  \n",
       "2906b810c7d4411798c6938adc9daaa5   True    True   True   False  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "portfolio_df = portfolio_df.set_index('id')\n",
    "portfolio_df.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 📃Transcript"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(306534, 4)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>person</th>\n",
       "      <th>event</th>\n",
       "      <th>value</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>78afa995795e4d85b5d9ceeca43f5fef</td>\n",
       "      <td>offer received</td>\n",
       "      <td>{'offer id': '9b98b8c7a33c4b65b9aebfe6a799e6d9'}</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>a03223e636434f42ac4c3df47e8bac43</td>\n",
       "      <td>offer received</td>\n",
       "      <td>{'offer id': '0b1e1539f2cc45b7b9fa7c272da2e1d7'}</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>e2127556f4f64592b11af22de27a7932</td>\n",
       "      <td>offer received</td>\n",
       "      <td>{'offer id': '2906b810c7d4411798c6938adc9daaa5'}</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>8ec6ce2a7e7949b1bf142def7d0e0586</td>\n",
       "      <td>offer received</td>\n",
       "      <td>{'offer id': 'fafdcd668e3743c1bb461111dcafc2a4'}</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>68617ca6246f4fbc85e91a2a49552598</td>\n",
       "      <td>offer received</td>\n",
       "      <td>{'offer id': '4d5c57ea9a6940dd891ad53e9dbe8da0'}</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             person           event  \\\n",
       "0  78afa995795e4d85b5d9ceeca43f5fef  offer received   \n",
       "1  a03223e636434f42ac4c3df47e8bac43  offer received   \n",
       "2  e2127556f4f64592b11af22de27a7932  offer received   \n",
       "3  8ec6ce2a7e7949b1bf142def7d0e0586  offer received   \n",
       "4  68617ca6246f4fbc85e91a2a49552598  offer received   \n",
       "\n",
       "                                              value  time  \n",
       "0  {'offer id': '9b98b8c7a33c4b65b9aebfe6a799e6d9'}     0  \n",
       "1  {'offer id': '0b1e1539f2cc45b7b9fa7c272da2e1d7'}     0  \n",
       "2  {'offer id': '2906b810c7d4411798c6938adc9daaa5'}     0  \n",
       "3  {'offer id': 'fafdcd668e3743c1bb461111dcafc2a4'}     0  \n",
       "4  {'offer id': '4d5c57ea9a6940dd891ad53e9dbe8da0'}     0  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(transcript.shape)\n",
    "transcript.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "val_dict = {\n",
    "    'offer_id': [],\n",
    "    'amount' : []\n",
    "}\n",
    "for i in range(len(transcript)):\n",
    "    temp = dict(transcript.loc[i, 'value'])\n",
    "    key = list(temp.keys())[0]\n",
    "    val = temp[key]\n",
    "    if key == 'offer id' or key == 'offer_id':\n",
    "        val_dict['offer_id'].append(val)\n",
    "        val_dict['amount'].append(None)\n",
    "    else:\n",
    "        val_dict['offer_id'].append(None)\n",
    "        val_dict['amount'].append(val)\n",
    "\n",
    "temp_df = pd.DataFrame(val_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>person</th>\n",
       "      <th>event</th>\n",
       "      <th>time</th>\n",
       "      <th>offer_id</th>\n",
       "      <th>amount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>78afa995795e4d85b5d9ceeca43f5fef</td>\n",
       "      <td>offer received</td>\n",
       "      <td>0</td>\n",
       "      <td>9b98b8c7a33c4b65b9aebfe6a799e6d9</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>a03223e636434f42ac4c3df47e8bac43</td>\n",
       "      <td>offer received</td>\n",
       "      <td>0</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>e2127556f4f64592b11af22de27a7932</td>\n",
       "      <td>offer received</td>\n",
       "      <td>0</td>\n",
       "      <td>2906b810c7d4411798c6938adc9daaa5</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>8ec6ce2a7e7949b1bf142def7d0e0586</td>\n",
       "      <td>offer received</td>\n",
       "      <td>0</td>\n",
       "      <td>fafdcd668e3743c1bb461111dcafc2a4</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>68617ca6246f4fbc85e91a2a49552598</td>\n",
       "      <td>offer received</td>\n",
       "      <td>0</td>\n",
       "      <td>4d5c57ea9a6940dd891ad53e9dbe8da0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             person           event  time  \\\n",
       "0  78afa995795e4d85b5d9ceeca43f5fef  offer received     0   \n",
       "1  a03223e636434f42ac4c3df47e8bac43  offer received     0   \n",
       "2  e2127556f4f64592b11af22de27a7932  offer received     0   \n",
       "3  8ec6ce2a7e7949b1bf142def7d0e0586  offer received     0   \n",
       "4  68617ca6246f4fbc85e91a2a49552598  offer received     0   \n",
       "\n",
       "                           offer_id  amount  \n",
       "0  9b98b8c7a33c4b65b9aebfe6a799e6d9     NaN  \n",
       "1  0b1e1539f2cc45b7b9fa7c272da2e1d7     NaN  \n",
       "2  2906b810c7d4411798c6938adc9daaa5     NaN  \n",
       "3  fafdcd668e3743c1bb461111dcafc2a4     NaN  \n",
       "4  4d5c57ea9a6940dd891ad53e9dbe8da0     NaN  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transcript_df = pd.concat([transcript, temp_df] , axis=1)\n",
    "transcript_df = transcript_df.drop('value', axis=1)\n",
    "transcript_df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>person</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9942</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9465</th>\n",
       "      <td>8dbfa485249f409aa223a2130f40634a</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6268</th>\n",
       "      <td>5e60c6aa3b834e44b822ea43a3efea26</td>\n",
       "      <td>48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13901</th>\n",
       "      <td>d0a80415b84c4df4908b8403b19765e3</td>\n",
       "      <td>48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8077</th>\n",
       "      <td>79d9d4f86aca4bed9290350fb43817c2</td>\n",
       "      <td>48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2582</th>\n",
       "      <td>28681c16026943e68f26feaccab0907f</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12657</th>\n",
       "      <td>bd2cdd691aca4bb0a0e039979ee5de5c</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11013</th>\n",
       "      <td>a42ed50acc4d4b25bca647c9e0b916ad</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11938</th>\n",
       "      <td>b1f4ece7d49342628a9ed77aee2cde58</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11470</th>\n",
       "      <td>ab25fd6fbd5040f880751921e4029757</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 person  size\n",
       "9942   94de646f7b6041228ca7dec82adb97d2    51\n",
       "9465   8dbfa485249f409aa223a2130f40634a    49\n",
       "6268   5e60c6aa3b834e44b822ea43a3efea26    48\n",
       "13901  d0a80415b84c4df4908b8403b19765e3    48\n",
       "8077   79d9d4f86aca4bed9290350fb43817c2    48\n",
       "2582   28681c16026943e68f26feaccab0907f    46\n",
       "12657  bd2cdd691aca4bb0a0e039979ee5de5c    46\n",
       "11013  a42ed50acc4d4b25bca647c9e0b916ad    46\n",
       "11938  b1f4ece7d49342628a9ed77aee2cde58    46\n",
       "11470  ab25fd6fbd5040f880751921e4029757    44"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t_df = transcript_df.groupby('person', as_index=False)['time'].size().sort_values('size', ascending=False)\n",
    "t_df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Person id with the maximum number of transactions in the dataset : 94de646f7b6041228ca7dec82adb97d2\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>person</th>\n",
       "      <th>event</th>\n",
       "      <th>time</th>\n",
       "      <th>offer_id</th>\n",
       "      <th>amount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2276</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer received</td>\n",
       "      <td>0</td>\n",
       "      <td>f19421c1d4aa40978ebb69ca19b0e20d</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16010</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer viewed</td>\n",
       "      <td>6</td>\n",
       "      <td>f19421c1d4aa40978ebb69ca19b0e20d</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24531</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>30</td>\n",
       "      <td>None</td>\n",
       "      <td>7.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24532</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer completed</td>\n",
       "      <td>30</td>\n",
       "      <td>f19421c1d4aa40978ebb69ca19b0e20d</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42252</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>102</td>\n",
       "      <td>None</td>\n",
       "      <td>1.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55475</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer received</td>\n",
       "      <td>168</td>\n",
       "      <td>9b98b8c7a33c4b65b9aebfe6a799e6d9</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75256</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer viewed</td>\n",
       "      <td>186</td>\n",
       "      <td>9b98b8c7a33c4b65b9aebfe6a799e6d9</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>77624</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>192</td>\n",
       "      <td>None</td>\n",
       "      <td>2.62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81725</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>204</td>\n",
       "      <td>None</td>\n",
       "      <td>0.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93913</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>246</td>\n",
       "      <td>None</td>\n",
       "      <td>2.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96627</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>258</td>\n",
       "      <td>None</td>\n",
       "      <td>1.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97913</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>264</td>\n",
       "      <td>None</td>\n",
       "      <td>2.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99190</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>270</td>\n",
       "      <td>None</td>\n",
       "      <td>6.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99191</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer completed</td>\n",
       "      <td>270</td>\n",
       "      <td>9b98b8c7a33c4b65b9aebfe6a799e6d9</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100437</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>276</td>\n",
       "      <td>None</td>\n",
       "      <td>5.87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>101667</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>282</td>\n",
       "      <td>None</td>\n",
       "      <td>3.77</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102890</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>288</td>\n",
       "      <td>None</td>\n",
       "      <td>0.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>104013</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>294</td>\n",
       "      <td>None</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109101</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>324</td>\n",
       "      <td>None</td>\n",
       "      <td>2.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110041</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>330</td>\n",
       "      <td>None</td>\n",
       "      <td>0.52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>113102</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer received</td>\n",
       "      <td>336</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>152891</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer received</td>\n",
       "      <td>408</td>\n",
       "      <td>fafdcd668e3743c1bb461111dcafc2a4</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>164148</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer viewed</td>\n",
       "      <td>408</td>\n",
       "      <td>fafdcd668e3743c1bb461111dcafc2a4</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>168269</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>414</td>\n",
       "      <td>None</td>\n",
       "      <td>9.66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>171735</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>420</td>\n",
       "      <td>None</td>\n",
       "      <td>1.35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>171736</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer completed</td>\n",
       "      <td>420</td>\n",
       "      <td>fafdcd668e3743c1bb461111dcafc2a4</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>177767</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>432</td>\n",
       "      <td>None</td>\n",
       "      <td>1.84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>180401</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>438</td>\n",
       "      <td>None</td>\n",
       "      <td>0.93</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>185283</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>450</td>\n",
       "      <td>None</td>\n",
       "      <td>2.79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>187480</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>456</td>\n",
       "      <td>None</td>\n",
       "      <td>1.70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>193520</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>474</td>\n",
       "      <td>None</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>198779</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>492</td>\n",
       "      <td>None</td>\n",
       "      <td>1.40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>198780</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer completed</td>\n",
       "      <td>492</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>203846</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer received</td>\n",
       "      <td>504</td>\n",
       "      <td>9b98b8c7a33c4b65b9aebfe6a799e6d9</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>222483</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer viewed</td>\n",
       "      <td>516</td>\n",
       "      <td>9b98b8c7a33c4b65b9aebfe6a799e6d9</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>222484</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>516</td>\n",
       "      <td>None</td>\n",
       "      <td>0.27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>235588</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>546</td>\n",
       "      <td>None</td>\n",
       "      <td>1.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>237713</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>552</td>\n",
       "      <td>None</td>\n",
       "      <td>2.70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>247375</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer received</td>\n",
       "      <td>576</td>\n",
       "      <td>4d5c57ea9a6940dd891ad53e9dbe8da0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>258696</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>576</td>\n",
       "      <td>None</td>\n",
       "      <td>4.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>266174</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>588</td>\n",
       "      <td>None</td>\n",
       "      <td>0.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>275001</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>606</td>\n",
       "      <td>None</td>\n",
       "      <td>5.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>275002</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer completed</td>\n",
       "      <td>606</td>\n",
       "      <td>9b98b8c7a33c4b65b9aebfe6a799e6d9</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>282196</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>offer viewed</td>\n",
       "      <td>624</td>\n",
       "      <td>4d5c57ea9a6940dd891ad53e9dbe8da0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>282197</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>624</td>\n",
       "      <td>None</td>\n",
       "      <td>6.18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>293425</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>660</td>\n",
       "      <td>None</td>\n",
       "      <td>1.78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>295012</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>666</td>\n",
       "      <td>None</td>\n",
       "      <td>0.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>298164</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>678</td>\n",
       "      <td>None</td>\n",
       "      <td>0.42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>300877</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>690</td>\n",
       "      <td>None</td>\n",
       "      <td>0.73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>302162</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>696</td>\n",
       "      <td>None</td>\n",
       "      <td>4.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>303445</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>transaction</td>\n",
       "      <td>702</td>\n",
       "      <td>None</td>\n",
       "      <td>0.90</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  person            event  time  \\\n",
       "2276    94de646f7b6041228ca7dec82adb97d2   offer received     0   \n",
       "16010   94de646f7b6041228ca7dec82adb97d2     offer viewed     6   \n",
       "24531   94de646f7b6041228ca7dec82adb97d2      transaction    30   \n",
       "24532   94de646f7b6041228ca7dec82adb97d2  offer completed    30   \n",
       "42252   94de646f7b6041228ca7dec82adb97d2      transaction   102   \n",
       "55475   94de646f7b6041228ca7dec82adb97d2   offer received   168   \n",
       "75256   94de646f7b6041228ca7dec82adb97d2     offer viewed   186   \n",
       "77624   94de646f7b6041228ca7dec82adb97d2      transaction   192   \n",
       "81725   94de646f7b6041228ca7dec82adb97d2      transaction   204   \n",
       "93913   94de646f7b6041228ca7dec82adb97d2      transaction   246   \n",
       "96627   94de646f7b6041228ca7dec82adb97d2      transaction   258   \n",
       "97913   94de646f7b6041228ca7dec82adb97d2      transaction   264   \n",
       "99190   94de646f7b6041228ca7dec82adb97d2      transaction   270   \n",
       "99191   94de646f7b6041228ca7dec82adb97d2  offer completed   270   \n",
       "100437  94de646f7b6041228ca7dec82adb97d2      transaction   276   \n",
       "101667  94de646f7b6041228ca7dec82adb97d2      transaction   282   \n",
       "102890  94de646f7b6041228ca7dec82adb97d2      transaction   288   \n",
       "104013  94de646f7b6041228ca7dec82adb97d2      transaction   294   \n",
       "109101  94de646f7b6041228ca7dec82adb97d2      transaction   324   \n",
       "110041  94de646f7b6041228ca7dec82adb97d2      transaction   330   \n",
       "113102  94de646f7b6041228ca7dec82adb97d2   offer received   336   \n",
       "152891  94de646f7b6041228ca7dec82adb97d2   offer received   408   \n",
       "164148  94de646f7b6041228ca7dec82adb97d2     offer viewed   408   \n",
       "168269  94de646f7b6041228ca7dec82adb97d2      transaction   414   \n",
       "171735  94de646f7b6041228ca7dec82adb97d2      transaction   420   \n",
       "171736  94de646f7b6041228ca7dec82adb97d2  offer completed   420   \n",
       "177767  94de646f7b6041228ca7dec82adb97d2      transaction   432   \n",
       "180401  94de646f7b6041228ca7dec82adb97d2      transaction   438   \n",
       "185283  94de646f7b6041228ca7dec82adb97d2      transaction   450   \n",
       "187480  94de646f7b6041228ca7dec82adb97d2      transaction   456   \n",
       "193520  94de646f7b6041228ca7dec82adb97d2      transaction   474   \n",
       "198779  94de646f7b6041228ca7dec82adb97d2      transaction   492   \n",
       "198780  94de646f7b6041228ca7dec82adb97d2  offer completed   492   \n",
       "203846  94de646f7b6041228ca7dec82adb97d2   offer received   504   \n",
       "222483  94de646f7b6041228ca7dec82adb97d2     offer viewed   516   \n",
       "222484  94de646f7b6041228ca7dec82adb97d2      transaction   516   \n",
       "235588  94de646f7b6041228ca7dec82adb97d2      transaction   546   \n",
       "237713  94de646f7b6041228ca7dec82adb97d2      transaction   552   \n",
       "247375  94de646f7b6041228ca7dec82adb97d2   offer received   576   \n",
       "258696  94de646f7b6041228ca7dec82adb97d2      transaction   576   \n",
       "266174  94de646f7b6041228ca7dec82adb97d2      transaction   588   \n",
       "275001  94de646f7b6041228ca7dec82adb97d2      transaction   606   \n",
       "275002  94de646f7b6041228ca7dec82adb97d2  offer completed   606   \n",
       "282196  94de646f7b6041228ca7dec82adb97d2     offer viewed   624   \n",
       "282197  94de646f7b6041228ca7dec82adb97d2      transaction   624   \n",
       "293425  94de646f7b6041228ca7dec82adb97d2      transaction   660   \n",
       "295012  94de646f7b6041228ca7dec82adb97d2      transaction   666   \n",
       "298164  94de646f7b6041228ca7dec82adb97d2      transaction   678   \n",
       "300877  94de646f7b6041228ca7dec82adb97d2      transaction   690   \n",
       "302162  94de646f7b6041228ca7dec82adb97d2      transaction   696   \n",
       "303445  94de646f7b6041228ca7dec82adb97d2      transaction   702   \n",
       "\n",
       "                                offer_id  amount  \n",
       "2276    f19421c1d4aa40978ebb69ca19b0e20d     NaN  \n",
       "16010   f19421c1d4aa40978ebb69ca19b0e20d     NaN  \n",
       "24531                               None    7.41  \n",
       "24532   f19421c1d4aa40978ebb69ca19b0e20d     NaN  \n",
       "42252                               None    1.47  \n",
       "55475   9b98b8c7a33c4b65b9aebfe6a799e6d9     NaN  \n",
       "75256   9b98b8c7a33c4b65b9aebfe6a799e6d9     NaN  \n",
       "77624                               None    2.62  \n",
       "81725                               None    0.59  \n",
       "93913                               None    2.28  \n",
       "96627                               None    1.56  \n",
       "97913                               None    2.74  \n",
       "99190                               None    6.98  \n",
       "99191   9b98b8c7a33c4b65b9aebfe6a799e6d9     NaN  \n",
       "100437                              None    5.87  \n",
       "101667                              None    3.77  \n",
       "102890                              None    0.33  \n",
       "104013                              None    1.00  \n",
       "109101                              None    2.98  \n",
       "110041                              None    0.52  \n",
       "113102  0b1e1539f2cc45b7b9fa7c272da2e1d7     NaN  \n",
       "152891  fafdcd668e3743c1bb461111dcafc2a4     NaN  \n",
       "164148  fafdcd668e3743c1bb461111dcafc2a4     NaN  \n",
       "168269                              None    9.66  \n",
       "171735                              None    1.35  \n",
       "171736  fafdcd668e3743c1bb461111dcafc2a4     NaN  \n",
       "177767                              None    1.84  \n",
       "180401                              None    0.93  \n",
       "185283                              None    2.79  \n",
       "187480                              None    1.70  \n",
       "193520                              None    0.99  \n",
       "198779                              None    1.40  \n",
       "198780  0b1e1539f2cc45b7b9fa7c272da2e1d7     NaN  \n",
       "203846  9b98b8c7a33c4b65b9aebfe6a799e6d9     NaN  \n",
       "222483  9b98b8c7a33c4b65b9aebfe6a799e6d9     NaN  \n",
       "222484                              None    0.27  \n",
       "235588                              None    1.02  \n",
       "237713                              None    2.70  \n",
       "247375  4d5c57ea9a6940dd891ad53e9dbe8da0     NaN  \n",
       "258696                              None    4.13  \n",
       "266174                              None    0.31  \n",
       "275001                              None    5.64  \n",
       "275002  9b98b8c7a33c4b65b9aebfe6a799e6d9     NaN  \n",
       "282196  4d5c57ea9a6940dd891ad53e9dbe8da0     NaN  \n",
       "282197                              None    6.18  \n",
       "293425                              None    1.78  \n",
       "295012                              None    0.41  \n",
       "298164                              None    0.42  \n",
       "300877                              None    0.73  \n",
       "302162                              None    4.96  \n",
       "303445                              None    0.90  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# person with the most number of transactions\n",
    "person_with_max_transactions = t_df.loc[t_df['size'].idxmax(), 'person']\n",
    "print(f'Person id with the maximum number of transactions in the dataset : {person_with_max_transactions}')\n",
    "\n",
    "temp_df = transcript_df.loc[transcript_df['person'] == person_with_max_transactions]\n",
    "display(temp_df)\n",
    "\n",
    "\n",
    "sns.set_style('whitegrid')\n",
    "fig, ax = plt.subplots()\n",
    "sns.lineplot(\n",
    "    data=temp_df,\n",
    "    x='time',\n",
    "    y='event',\n",
    ")\n",
    "sns.scatterplot(\n",
    "    data=temp_df,\n",
    "    x='time',\n",
    "    y='event',\n",
    "    hue='event',\n",
    "    palette={\n",
    "        'offer received': 'red',\n",
    "        'offer viewed': 'orange',\n",
    "        'offer completed':'green',\n",
    "        'transaction': 'gray'\n",
    "        },\n",
    "    s=100\n",
    ")\n",
    "fig.set_figheight(10)\n",
    "fig.set_figwidth(20)\n",
    "ax.set_title(f'Transactions timeline for person with id {person_with_max_transactions}')\n",
    "ax.set_xticks(range(1,750,20))\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Person id with the maximum number of transactions in the dataset : da7a7c0dcfcb41a8acc7864a53cf60fb\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>person</th>\n",
       "      <th>event</th>\n",
       "      <th>time</th>\n",
       "      <th>offer_id</th>\n",
       "      <th>amount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>263150</th>\n",
       "      <td>da7a7c0dcfcb41a8acc7864a53cf60fb</td>\n",
       "      <td>transaction</td>\n",
       "      <td>582</td>\n",
       "      <td>None</td>\n",
       "      <td>0.35</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  person        event  time offer_id  amount\n",
       "263150  da7a7c0dcfcb41a8acc7864a53cf60fb  transaction   582     None    0.35"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# person with the minimum number of transactions\n",
    "person_with_min_transactions = t_df.loc[t_df['size'].idxmin(), 'person']\n",
    "print(f'Person id with the maximum number of transactions in the dataset : {person_with_min_transactions}')\n",
    "\n",
    "transcript_df.loc[transcript_df['person'] == person_with_min_transactions]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- I can remove all entries for the users who has only one entry in the transcript in the dataset, since they all haven't been influenced by an offer.\n",
    "- I can do this since I want only to include the offers that influenced a user. A user with only 1 entry in the transcript eith made only one transaction with a certain amount or recieved an offer without even viewing the offer.\n",
    "- I can also remove all the entries of a user who has not viewed any offer i.e. has no offer viewed entry. this data is not relevant for the usecase studied."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Remove the entries of users with no offers viewed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "t_df = transcript_df.groupby('person', as_index=False)['event'].apply(\n",
    "    lambda e: (e == 'offer viewed').sum()\n",
    ").rename(columns={'event': '#viewed_offers'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "persons_with_no_viewed_offers_lst = t_df.loc[t_df['#viewed_offers'] == 0, 'person'].to_list()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape before cleaning : (306534, 5)\n",
      "Shape after cleaning : (305264, 5)\n"
     ]
    }
   ],
   "source": [
    "transcript_df_viewed = transcript_df[~transcript_df['person'].isin(persons_with_no_viewed_offers_lst)]\n",
    "print(f'Shape before cleaning : {transcript_df.shape}')\n",
    "print(f'Shape after cleaning : {transcript_df_viewed.shape}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Converting event based data to tabular form"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Steps:\n",
    "- Group transcripts by person id and offer id.\n",
    "- For each group loop through the offers.\n",
    "- Append to 4 lists {person_id, offer_id, offer_viewed?, offer_competed?}\n",
    "- Whenever we find an event received append person_id, offer_id, offer_viewed=0, offer_completed=0.\n",
    "- Whenever we find offer_viewed make the last offer that wasn't viewed in the list = 1 (traverse the list from the end to the beginning)\n",
    "- Whenever we find offer_completed make the last offer that wasn't completed in the list = 1 (traverse the list from the end to the beginning)\n",
    "\n",
    "Assumption : Always associate the offer viewed and the offer completed events to the closest event that wasn't viewed or completed.\n",
    "Also the events are travesed one group at a time.\n",
    "\n",
    "I should end up with a dataframe that looks like this:\n",
    "\n",
    "+-----------+----------+--------------+-----------------+\n",
    "<br/>\n",
    "| Person Id | Offer Id | Offer Viewed | Offer Completed |\n",
    "<br/>\n",
    "+-----------+----------+--------------+-----------------+\n",
    "\n",
    "Finally remove all the offers that weren't viewed from the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>person</th>\n",
       "      <th>event</th>\n",
       "      <th>time</th>\n",
       "      <th>offer_id</th>\n",
       "      <th>amount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>78afa995795e4d85b5d9ceeca43f5fef</td>\n",
       "      <td>offer received</td>\n",
       "      <td>0</td>\n",
       "      <td>9b98b8c7a33c4b65b9aebfe6a799e6d9</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>a03223e636434f42ac4c3df47e8bac43</td>\n",
       "      <td>offer received</td>\n",
       "      <td>0</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>e2127556f4f64592b11af22de27a7932</td>\n",
       "      <td>offer received</td>\n",
       "      <td>0</td>\n",
       "      <td>2906b810c7d4411798c6938adc9daaa5</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>8ec6ce2a7e7949b1bf142def7d0e0586</td>\n",
       "      <td>offer received</td>\n",
       "      <td>0</td>\n",
       "      <td>fafdcd668e3743c1bb461111dcafc2a4</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>68617ca6246f4fbc85e91a2a49552598</td>\n",
       "      <td>offer received</td>\n",
       "      <td>0</td>\n",
       "      <td>4d5c57ea9a6940dd891ad53e9dbe8da0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             person           event  time  \\\n",
       "0  78afa995795e4d85b5d9ceeca43f5fef  offer received     0   \n",
       "1  a03223e636434f42ac4c3df47e8bac43  offer received     0   \n",
       "2  e2127556f4f64592b11af22de27a7932  offer received     0   \n",
       "3  8ec6ce2a7e7949b1bf142def7d0e0586  offer received     0   \n",
       "4  68617ca6246f4fbc85e91a2a49552598  offer received     0   \n",
       "\n",
       "                           offer_id  amount  \n",
       "0  9b98b8c7a33c4b65b9aebfe6a799e6d9     NaN  \n",
       "1  0b1e1539f2cc45b7b9fa7c272da2e1d7     NaN  \n",
       "2  2906b810c7d4411798c6938adc9daaa5     NaN  \n",
       "3  fafdcd668e3743c1bb461111dcafc2a4     NaN  \n",
       "4  4d5c57ea9a6940dd891ad53e9dbe8da0     NaN  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# remove all transactions\n",
    "transcript_df = transcript_df.loc[transcript_df['event'] != 'transaction']\n",
    "transcript_df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "trans_df = {\n",
    "    'person' : [],\n",
    "    'offer_id' : [],\n",
    "    'is_offer_viewed' : [],\n",
    "    'is_offer_completed' : []\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "trans_grps = transcript_df.groupby(['person', 'offer_id'], as_index=True)\n",
    "\n",
    "i = 0\n",
    "for (person, offer_id), group_df in trans_grps:\n",
    "    person_lst = []\n",
    "    offer_id_lst = []\n",
    "    offer_viewed_lst = []\n",
    "    offer_completed_lst = []\n",
    "\n",
    "    for _, row in group_df.iterrows():\n",
    "        \n",
    "        if row['event'] == 'offer received':\n",
    "            person_lst.append(person)\n",
    "            offer_id_lst.append(offer_id)\n",
    "            offer_viewed_lst.append(0)\n",
    "            offer_completed_lst.append(0)\n",
    "\n",
    "        elif row['event'] == 'offer viewed':\n",
    "            for i in range(len(offer_viewed_lst)-1, -1, -1):\n",
    "                if offer_viewed_lst[i] == 0:\n",
    "                    offer_viewed_lst[i] = 1\n",
    "                else:\n",
    "                    break\n",
    "        elif row['event'] == 'offer completed':\n",
    "            for i in range(len(offer_completed_lst)-1, -1, -1):\n",
    "                if offer_completed_lst[i] == 0:\n",
    "                    offer_completed_lst[i] = 1\n",
    "                else:\n",
    "                    break\n",
    "        else:\n",
    "            raise ValueError('Unknown value for event')\n",
    "    \n",
    "    trans_df['person'] = trans_df['person'] + person_lst\n",
    "    trans_df['offer_id'] = trans_df['offer_id'] + offer_id_lst\n",
    "    trans_df['is_offer_viewed'] = trans_df['is_offer_viewed'] + offer_viewed_lst\n",
    "    trans_df['is_offer_completed'] = trans_df['is_offer_completed'] + offer_completed_lst"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(76277, 4)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>person</th>\n",
       "      <th>offer_id</th>\n",
       "      <th>is_offer_viewed</th>\n",
       "      <th>is_offer_completed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0009655768c64bdeb2e877511632db8f</td>\n",
       "      <td>2906b810c7d4411798c6938adc9daaa5</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0009655768c64bdeb2e877511632db8f</td>\n",
       "      <td>3f207df678b143eea3cee63160fa8bed</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0009655768c64bdeb2e877511632db8f</td>\n",
       "      <td>5a8bc65990b245e5a138643cd4eb9837</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0009655768c64bdeb2e877511632db8f</td>\n",
       "      <td>f19421c1d4aa40978ebb69ca19b0e20d</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0009655768c64bdeb2e877511632db8f</td>\n",
       "      <td>fafdcd668e3743c1bb461111dcafc2a4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>00116118485d4dfda04fdbaba9a87b5c</td>\n",
       "      <td>f19421c1d4aa40978ebb69ca19b0e20d</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>00116118485d4dfda04fdbaba9a87b5c</td>\n",
       "      <td>f19421c1d4aa40978ebb69ca19b0e20d</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0011e0d4e6b944f998e987f904e8c1e5</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0011e0d4e6b944f998e987f904e8c1e5</td>\n",
       "      <td>2298d6c36e964ae4a3e7e9706d1fb8c2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0011e0d4e6b944f998e987f904e8c1e5</td>\n",
       "      <td>3f207df678b143eea3cee63160fa8bed</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             person                          offer_id  \\\n",
       "0  0009655768c64bdeb2e877511632db8f  2906b810c7d4411798c6938adc9daaa5   \n",
       "1  0009655768c64bdeb2e877511632db8f  3f207df678b143eea3cee63160fa8bed   \n",
       "2  0009655768c64bdeb2e877511632db8f  5a8bc65990b245e5a138643cd4eb9837   \n",
       "3  0009655768c64bdeb2e877511632db8f  f19421c1d4aa40978ebb69ca19b0e20d   \n",
       "4  0009655768c64bdeb2e877511632db8f  fafdcd668e3743c1bb461111dcafc2a4   \n",
       "5  00116118485d4dfda04fdbaba9a87b5c  f19421c1d4aa40978ebb69ca19b0e20d   \n",
       "6  00116118485d4dfda04fdbaba9a87b5c  f19421c1d4aa40978ebb69ca19b0e20d   \n",
       "7  0011e0d4e6b944f998e987f904e8c1e5  0b1e1539f2cc45b7b9fa7c272da2e1d7   \n",
       "8  0011e0d4e6b944f998e987f904e8c1e5  2298d6c36e964ae4a3e7e9706d1fb8c2   \n",
       "9  0011e0d4e6b944f998e987f904e8c1e5  3f207df678b143eea3cee63160fa8bed   \n",
       "\n",
       "   is_offer_viewed  is_offer_completed  \n",
       "0                0                   1  \n",
       "1                1                   0  \n",
       "2                1                   0  \n",
       "3                1                   1  \n",
       "4                1                   1  \n",
       "5                1                   0  \n",
       "6                1                   0  \n",
       "7                1                   1  \n",
       "8                1                   1  \n",
       "9                1                   0  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trans_df = pd.DataFrame(trans_df)\n",
    "print(trans_df.shape)\n",
    "trans_df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>person</th>\n",
       "      <th>offer_id</th>\n",
       "      <th>is_offer_viewed</th>\n",
       "      <th>is_offer_completed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>44584</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44585</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>4d5c57ea9a6940dd891ad53e9dbe8da0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44586</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>9b98b8c7a33c4b65b9aebfe6a799e6d9</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44587</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>9b98b8c7a33c4b65b9aebfe6a799e6d9</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44588</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>f19421c1d4aa40978ebb69ca19b0e20d</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44589</th>\n",
       "      <td>94de646f7b6041228ca7dec82adb97d2</td>\n",
       "      <td>fafdcd668e3743c1bb461111dcafc2a4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 person                          offer_id  \\\n",
       "44584  94de646f7b6041228ca7dec82adb97d2  0b1e1539f2cc45b7b9fa7c272da2e1d7   \n",
       "44585  94de646f7b6041228ca7dec82adb97d2  4d5c57ea9a6940dd891ad53e9dbe8da0   \n",
       "44586  94de646f7b6041228ca7dec82adb97d2  9b98b8c7a33c4b65b9aebfe6a799e6d9   \n",
       "44587  94de646f7b6041228ca7dec82adb97d2  9b98b8c7a33c4b65b9aebfe6a799e6d9   \n",
       "44588  94de646f7b6041228ca7dec82adb97d2  f19421c1d4aa40978ebb69ca19b0e20d   \n",
       "44589  94de646f7b6041228ca7dec82adb97d2  fafdcd668e3743c1bb461111dcafc2a4   \n",
       "\n",
       "       is_offer_viewed  is_offer_completed  \n",
       "44584                0                   1  \n",
       "44585                1                   0  \n",
       "44586                1                   1  \n",
       "44587                1                   1  \n",
       "44588                1                   1  \n",
       "44589                1                   1  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# view the offers for the person with the most offers\n",
    "trans_df.loc[trans_df['person'] == '94de646f7b6041228ca7dec82adb97d2']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we remove all offers that weren't viewed and didn't influence the users"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>person</th>\n",
       "      <th>offer_id</th>\n",
       "      <th>is_offer_viewed</th>\n",
       "      <th>is_offer_completed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0009655768c64bdeb2e877511632db8f</td>\n",
       "      <td>3f207df678b143eea3cee63160fa8bed</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0009655768c64bdeb2e877511632db8f</td>\n",
       "      <td>5a8bc65990b245e5a138643cd4eb9837</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0009655768c64bdeb2e877511632db8f</td>\n",
       "      <td>f19421c1d4aa40978ebb69ca19b0e20d</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0009655768c64bdeb2e877511632db8f</td>\n",
       "      <td>fafdcd668e3743c1bb461111dcafc2a4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>00116118485d4dfda04fdbaba9a87b5c</td>\n",
       "      <td>f19421c1d4aa40978ebb69ca19b0e20d</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             person                          offer_id  \\\n",
       "1  0009655768c64bdeb2e877511632db8f  3f207df678b143eea3cee63160fa8bed   \n",
       "2  0009655768c64bdeb2e877511632db8f  5a8bc65990b245e5a138643cd4eb9837   \n",
       "3  0009655768c64bdeb2e877511632db8f  f19421c1d4aa40978ebb69ca19b0e20d   \n",
       "4  0009655768c64bdeb2e877511632db8f  fafdcd668e3743c1bb461111dcafc2a4   \n",
       "5  00116118485d4dfda04fdbaba9a87b5c  f19421c1d4aa40978ebb69ca19b0e20d   \n",
       "\n",
       "   is_offer_viewed  is_offer_completed  \n",
       "1                1                   0  \n",
       "2                1                   0  \n",
       "3                1                   1  \n",
       "4                1                   1  \n",
       "5                1                   0  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trans_df = trans_df.loc[trans_df['is_offer_viewed'] == 1]\n",
    "trans_df.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 🙍🏻‍♂️Profile"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>id</th>\n",
       "      <th>became_member_on</th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>20170212</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>F</td>\n",
       "      <td>55</td>\n",
       "      <td>0610b486422d4921ae7d2bf64640c50b</td>\n",
       "      <td>20170715</td>\n",
       "      <td>112000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>38fe809add3b4fcf9315a9694bb96ff5</td>\n",
       "      <td>20180712</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>F</td>\n",
       "      <td>75</td>\n",
       "      <td>78afa995795e4d85b5d9ceeca43f5fef</td>\n",
       "      <td>20170509</td>\n",
       "      <td>100000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>a03223e636434f42ac4c3df47e8bac43</td>\n",
       "      <td>20170804</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age                                id  became_member_on    income\n",
       "0   None  118  68be06ca386d4c31939f3a4f0e3dd783          20170212       NaN\n",
       "1      F   55  0610b486422d4921ae7d2bf64640c50b          20170715  112000.0\n",
       "2   None  118  38fe809add3b4fcf9315a9694bb96ff5          20180712       NaN\n",
       "3      F   75  78afa995795e4d85b5d9ceeca43f5fef          20170509  100000.0\n",
       "4   None  118  a03223e636434f42ac4c3df47e8bac43          20170804       NaN"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "profile.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>id</th>\n",
       "      <th>became_member_on</th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>20170212</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>38fe809add3b4fcf9315a9694bb96ff5</td>\n",
       "      <td>20180712</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>a03223e636434f42ac4c3df47e8bac43</td>\n",
       "      <td>20170804</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>8ec6ce2a7e7949b1bf142def7d0e0586</td>\n",
       "      <td>20170925</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>68617ca6246f4fbc85e91a2a49552598</td>\n",
       "      <td>20171002</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age                                id  became_member_on  income\n",
       "0   None  118  68be06ca386d4c31939f3a4f0e3dd783          20170212     NaN\n",
       "2   None  118  38fe809add3b4fcf9315a9694bb96ff5          20180712     NaN\n",
       "4   None  118  a03223e636434f42ac4c3df47e8bac43          20170804     NaN\n",
       "6   None  118  8ec6ce2a7e7949b1bf142def7d0e0586          20170925     NaN\n",
       "7   None  118  68617ca6246f4fbc85e91a2a49552598          20171002     NaN"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_df = profile.loc[profile['gender'].isna()]\n",
    "gender_ids = temp_df.index\n",
    "temp_df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>id</th>\n",
       "      <th>became_member_on</th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>20170212</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>38fe809add3b4fcf9315a9694bb96ff5</td>\n",
       "      <td>20180712</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>a03223e636434f42ac4c3df47e8bac43</td>\n",
       "      <td>20170804</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>8ec6ce2a7e7949b1bf142def7d0e0586</td>\n",
       "      <td>20170925</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>68617ca6246f4fbc85e91a2a49552598</td>\n",
       "      <td>20171002</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age                                id  became_member_on  income\n",
       "0   None  118  68be06ca386d4c31939f3a4f0e3dd783          20170212     NaN\n",
       "2   None  118  38fe809add3b4fcf9315a9694bb96ff5          20180712     NaN\n",
       "4   None  118  a03223e636434f42ac4c3df47e8bac43          20170804     NaN\n",
       "6   None  118  8ec6ce2a7e7949b1bf142def7d0e0586          20170925     NaN\n",
       "7   None  118  68617ca6246f4fbc85e91a2a49552598          20171002     NaN"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_df = profile.loc[profile['income'].isna()]\n",
    "income_ids = temp_df.index\n",
    "temp_df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Is all ids with missing genders the same as the ones with missing incomes ?: True\n"
     ]
    }
   ],
   "source": [
    "print(f'Is all ids with missing genders the same as the ones with missing incomes ?: {np.all(gender_ids == income_ids)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[118]\n"
     ]
    }
   ],
   "source": [
    "temp_df = profile.loc[profile['income'].isna()]\n",
    "print(temp_df.age.unique())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "All ids with missing genders and income have the same age 118, which might be logical since old people might not fill their complete information during account setup on the rewards app"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>id</th>\n",
       "      <th>became_member_on</th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>14825</td>\n",
       "      <td>17000.000000</td>\n",
       "      <td>17000</td>\n",
       "      <td>1.700000e+04</td>\n",
       "      <td>14825.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>17000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>M</td>\n",
       "      <td>NaN</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>8484</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>NaN</td>\n",
       "      <td>62.531412</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.016703e+07</td>\n",
       "      <td>65404.991568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>NaN</td>\n",
       "      <td>26.738580</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.167750e+04</td>\n",
       "      <td>21598.299410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>NaN</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.013073e+07</td>\n",
       "      <td>30000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.016053e+07</td>\n",
       "      <td>49000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>58.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.017080e+07</td>\n",
       "      <td>64000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>73.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.017123e+07</td>\n",
       "      <td>80000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>NaN</td>\n",
       "      <td>118.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.018073e+07</td>\n",
       "      <td>120000.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       gender           age                                id  \\\n",
       "count   14825  17000.000000                             17000   \n",
       "unique      3           NaN                             17000   \n",
       "top         M           NaN  68be06ca386d4c31939f3a4f0e3dd783   \n",
       "freq     8484           NaN                                 1   \n",
       "mean      NaN     62.531412                               NaN   \n",
       "std       NaN     26.738580                               NaN   \n",
       "min       NaN     18.000000                               NaN   \n",
       "25%       NaN     45.000000                               NaN   \n",
       "50%       NaN     58.000000                               NaN   \n",
       "75%       NaN     73.000000                               NaN   \n",
       "max       NaN    118.000000                               NaN   \n",
       "\n",
       "        became_member_on         income  \n",
       "count       1.700000e+04   14825.000000  \n",
       "unique               NaN            NaN  \n",
       "top                  NaN            NaN  \n",
       "freq                 NaN            NaN  \n",
       "mean        2.016703e+07   65404.991568  \n",
       "std         1.167750e+04   21598.299410  \n",
       "min         2.013073e+07   30000.000000  \n",
       "25%         2.016053e+07   49000.000000  \n",
       "50%         2.017080e+07   64000.000000  \n",
       "75%         2.017123e+07   80000.000000  \n",
       "max         2.018073e+07  120000.000000  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "profile.describe(include='all')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Convert became_member_on to year, month, date"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "profile_df = profile.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "object\n"
     ]
    }
   ],
   "source": [
    "profile_df['became_member_on'] = profile_df['became_member_on'].astype(str)\n",
    "print(profile_df['became_member_on'].dtype)\n",
    "def change_date_format(date_str):\n",
    "    out_str = f\"{date_str[:4]}-{date_str[4:6]}-{date_str[6:]}\"\n",
    "    return out_str\n",
    "\n",
    "profile_df['became_member_on'] = profile_df['became_member_on'].apply(change_date_format)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "profile_df['became_member_on'] = pd.to_datetime(profile_df['became_member_on'], infer_datetime_format=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "profile_df['became_member_on_year'] = pd.DatetimeIndex(profile_df['became_member_on']).year\n",
    "profile_df['became_member_on_month'] = pd.DatetimeIndex(profile_df['became_member_on']).month\n",
    "profile_df['became_member_on_day'] = pd.DatetimeIndex(profile_df['became_member_on']).day\n",
    "profile_df = profile_df.drop('became_member_on', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>id</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>F</td>\n",
       "      <td>55</td>\n",
       "      <td>0610b486422d4921ae7d2bf64640c50b</td>\n",
       "      <td>112000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>38fe809add3b4fcf9315a9694bb96ff5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2018</td>\n",
       "      <td>7</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>F</td>\n",
       "      <td>75</td>\n",
       "      <td>78afa995795e4d85b5d9ceeca43f5fef</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>5</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>a03223e636434f42ac4c3df47e8bac43</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>M</td>\n",
       "      <td>68</td>\n",
       "      <td>e2127556f4f64592b11af22de27a7932</td>\n",
       "      <td>70000.0</td>\n",
       "      <td>2018</td>\n",
       "      <td>4</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>8ec6ce2a7e7949b1bf142def7d0e0586</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017</td>\n",
       "      <td>9</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>68617ca6246f4fbc85e91a2a49552598</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>M</td>\n",
       "      <td>65</td>\n",
       "      <td>389bc3fa690240e798340f5a15918d5c</td>\n",
       "      <td>53000.0</td>\n",
       "      <td>2018</td>\n",
       "      <td>2</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>8974fc5686fe429db53ddde067b88302</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016</td>\n",
       "      <td>11</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age                                id    income  \\\n",
       "0   None  118  68be06ca386d4c31939f3a4f0e3dd783       NaN   \n",
       "1      F   55  0610b486422d4921ae7d2bf64640c50b  112000.0   \n",
       "2   None  118  38fe809add3b4fcf9315a9694bb96ff5       NaN   \n",
       "3      F   75  78afa995795e4d85b5d9ceeca43f5fef  100000.0   \n",
       "4   None  118  a03223e636434f42ac4c3df47e8bac43       NaN   \n",
       "5      M   68  e2127556f4f64592b11af22de27a7932   70000.0   \n",
       "6   None  118  8ec6ce2a7e7949b1bf142def7d0e0586       NaN   \n",
       "7   None  118  68617ca6246f4fbc85e91a2a49552598       NaN   \n",
       "8      M   65  389bc3fa690240e798340f5a15918d5c   53000.0   \n",
       "9   None  118  8974fc5686fe429db53ddde067b88302       NaN   \n",
       "\n",
       "   became_member_on_year  became_member_on_month  became_member_on_day  \n",
       "0                   2017                       2                    12  \n",
       "1                   2017                       7                    15  \n",
       "2                   2018                       7                    12  \n",
       "3                   2017                       5                     9  \n",
       "4                   2017                       8                     4  \n",
       "5                   2018                       4                    26  \n",
       "6                   2017                       9                    25  \n",
       "7                   2017                      10                     2  \n",
       "8                   2018                       2                     9  \n",
       "9                   2016                      11                    22  "
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "profile_df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>id</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>14825</td>\n",
       "      <td>17000.000000</td>\n",
       "      <td>17000</td>\n",
       "      <td>14825.000000</td>\n",
       "      <td>17000.000000</td>\n",
       "      <td>17000.000000</td>\n",
       "      <td>17000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>17000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>M</td>\n",
       "      <td>NaN</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>8484</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>NaN</td>\n",
       "      <td>62.531412</td>\n",
       "      <td>NaN</td>\n",
       "      <td>65404.991568</td>\n",
       "      <td>2016.635000</td>\n",
       "      <td>6.683529</td>\n",
       "      <td>15.881176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>NaN</td>\n",
       "      <td>26.738580</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21598.299410</td>\n",
       "      <td>1.177624</td>\n",
       "      <td>3.487098</td>\n",
       "      <td>8.775390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>NaN</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>2013.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>49000.000000</td>\n",
       "      <td>2016.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>8.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>58.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>64000.000000</td>\n",
       "      <td>2017.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>16.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>73.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>80000.000000</td>\n",
       "      <td>2017.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>23.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>NaN</td>\n",
       "      <td>118.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>120000.000000</td>\n",
       "      <td>2018.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>31.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       gender           age                                id         income  \\\n",
       "count   14825  17000.000000                             17000   14825.000000   \n",
       "unique      3           NaN                             17000            NaN   \n",
       "top         M           NaN  68be06ca386d4c31939f3a4f0e3dd783            NaN   \n",
       "freq     8484           NaN                                 1            NaN   \n",
       "mean      NaN     62.531412                               NaN   65404.991568   \n",
       "std       NaN     26.738580                               NaN   21598.299410   \n",
       "min       NaN     18.000000                               NaN   30000.000000   \n",
       "25%       NaN     45.000000                               NaN   49000.000000   \n",
       "50%       NaN     58.000000                               NaN   64000.000000   \n",
       "75%       NaN     73.000000                               NaN   80000.000000   \n",
       "max       NaN    118.000000                               NaN  120000.000000   \n",
       "\n",
       "        became_member_on_year  became_member_on_month  became_member_on_day  \n",
       "count            17000.000000            17000.000000          17000.000000  \n",
       "unique                    NaN                     NaN                   NaN  \n",
       "top                       NaN                     NaN                   NaN  \n",
       "freq                      NaN                     NaN                   NaN  \n",
       "mean              2016.635000                6.683529             15.881176  \n",
       "std                  1.177624                3.487098              8.775390  \n",
       "min               2013.000000                1.000000              1.000000  \n",
       "25%               2016.000000                4.000000              8.000000  \n",
       "50%               2017.000000                7.000000             16.000000  \n",
       "75%               2017.000000               10.000000             23.000000  \n",
       "max               2018.000000               12.000000             31.000000  "
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "profile_df.describe(include='all')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is clear from the above description that:\n",
    "- 2175 users haven't specified their age.\n",
    "- 2175 users haven't specified their income (still don't know if they are the same as the ones who didn't specify their age)\n",
    "- We have users that became members in the region from 2013 to 2018, I can try to derive info such as if the user being older has an effect on whether or not he might accept the offer.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>14825.000000</td>\n",
       "      <td>14825.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>54.393524</td>\n",
       "      <td>65404.991568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>17.383705</td>\n",
       "      <td>21598.299410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>18.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>42.000000</td>\n",
       "      <td>49000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>55.000000</td>\n",
       "      <td>64000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>66.000000</td>\n",
       "      <td>80000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>101.000000</td>\n",
       "      <td>120000.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                age         income\n",
       "count  14825.000000   14825.000000\n",
       "mean      54.393524   65404.991568\n",
       "std       17.383705   21598.299410\n",
       "min       18.000000   30000.000000\n",
       "25%       42.000000   49000.000000\n",
       "50%       55.000000   64000.000000\n",
       "75%       66.000000   80000.000000\n",
       "max      101.000000  120000.000000"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_df = profile_df.loc[:, ['age', 'income']]\n",
    "temp_df = temp_df[temp_df['age'].notna()]\n",
    "temp_df = temp_df[temp_df['income'].notna()]\n",
    "temp_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>14825.0</td>\n",
       "      <td>14825.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>84.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>58.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>408.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>NaN</td>\n",
       "      <td>65404.991568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>NaN</td>\n",
       "      <td>21598.299410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>NaN</td>\n",
       "      <td>30000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>49000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>64000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>80000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>NaN</td>\n",
       "      <td>120000.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            age         income\n",
       "count   14825.0   14825.000000\n",
       "unique     84.0            NaN\n",
       "top        58.0            NaN\n",
       "freq      408.0            NaN\n",
       "mean        NaN   65404.991568\n",
       "std         NaN   21598.299410\n",
       "min         NaN   30000.000000\n",
       "25%         NaN   49000.000000\n",
       "50%         NaN   64000.000000\n",
       "75%         NaN   80000.000000\n",
       "max         NaN  120000.000000"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_df['age'] = temp_df['age'].astype('category')\n",
    "temp_df.describe(include='all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'age vs income')"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZgAAAEWCAYAAABbgYH9AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8/fFQqAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOxdeXgUVfY9ve/dSTpLh4Q0hCQkJCQsYRFZNCCChkUWtxEVUH4qCA4IuIKgo7IMKoIzIG6gowgoCCiiIAIDiEEIWyCEkJWks3d6339/VKpS1VUVAoLODH2+z086VfXerVfV/d5999xzBYFAIIAQQgghhBBCuM4Q/tkGhBBCCCGE8L+J0AQTQgghhBDCDUFoggkhhBBCCOGGIDTBhBBCCCGEcEMQmmBCCCGEEEK4IQhNMCGEEEIIIdwQhCaYEEL4L8Dly5fRs2dP+Hy+P9uUEEJoNwShPJgQQgghhBBuBEIeTAghhBBCCDcEoQkmhJsWa9euxbBhw9CzZ0/cdddd+OGHH6hjPp8Pb775Jvr164ecnBx8+umn6Nq1K7xeLwDAYrHghRdewMCBAzFo0CC89dZbnNtXJpMJmZmZaGpqov529uxZ9OvXDx6PB6WlpXjooYfQu3dv9OvXD8888wynrRUVFYz+J02ahLfffhv3338/evbsiSlTpqChoYE6Py8vD/fffz+ys7MxZMgQfPXVV5Td8+bNQ//+/XH77bfjvffeg9/vBwB89dVXuP/++/H6668jOzsbQ4cOxW+//YavvvoKQ4YMwS233IKvv/6a6sPtdmPJkiW47bbbMGDAACxYsABOp/PaHkYI/5MITTAh3LTo2LEjPvvsMxw7dgwzZszA3LlzUVNTAwD48ssvsX//fmzbtg1ff/01fvzxR8a18+fPh1gsxu7du7F161b8+9//xqZNm1h9xMTEoEePHti9ezf1t+3bt+POO++ERCLBO++8g1tvvRW//vor9u/fj4ceeqjd9u/YsQNvvPEGDh8+DI/Hgw8//BAAEa95/PHH8dBDD+Hw4cPYunUr0tLSAACvvvoqLBYLfvzxR2zYsAHbtm3Dli1bqDZPnjyJrl274pdffkFubi5mz56NU6dO4YcffsCyZcuwePFi2Gw2AMCyZctw6dIlbN26Fbt370ZNTQ1Wr17dbvtD+N9HaIIJ4abFyJEjERMTA6FQiLvuugtGoxEnT54EAHz33Xd4+OGHYTAYoNPpMG3aNOq6uro67N+/Hy+88AKUSiX0ej0effRR7Ny5k7OfUaNGYceOHQCAQCCAb7/9FqNGjQIAiMViXL58GTU1NZDJZMjOzm63/ePGjUPnzp0hl8sxYsQIFBQUACAmsAEDBiA3NxcSiQTh4eFIS0uDz+fDt99+izlz5kCtViM+Ph6TJ0/GN998Q7UZHx+P8ePHQyQS4a677kJVVRWmT58OqVSKgQMHQiqVoqysDIFAAJs2bcILL7yAsLAwqNVq/N///R/vGIRwc0L8ZxsQQgh/FrZu3YqPPvoIlZWVAAC73Y7GxkYAQE1NDWJjY6lzDQYD9e/Lly/D6/Vi4MCB1N/8fj/jfDruvPNOvPrqqzCZTCgtLYVAIKAmkrlz5+Kdd97BhAkToNPpMHnyZEyYMKFd9kdFRVH/VigUsNvtAICqqiokJCSwzm9sbITH40GHDh2ov3Xo0AEmk4n6rNfrqX/L5XIAQGRkJPU3mUwGm82GhoYGOBwOjBs3jjoWCASo7bYQQgBCE0wINykqKyvx0ksv4eOPP0bPnj0hEokwZswY6nhUVBSqq6upz/R/GwwGSKVSHDlyBGLxlb9CWq0Wt956K7777jsUFxfj7rvvhkAgoPp57bXXABBxk8mTJ6NPnz4wGo3XfG+xsbGUJ0ZHeHg4JBIJLl++jKSkJADEZBQTE3PVfYSHh0Mul2Pnzp3XdH0INwdCW2Qh3JRwOBwQCASIiIgAAGzZsgUXLlygjo8cORLr16+HyWRCc3Mz3n//fepYdHQ0br31Vrz55puwWq3w+/0oKyvD0aNHefsbNWoUtm3bhu+//57aHgOIrThy8tLpdBAIBBAKf9/XctSoUTh06BC+/fZbeL1eNDY2oqCgACKRCCNGjMBbb70Fq9WKyspKfPTRRxg9evRV9yEUCjFx4kS8/vrrqK+vB0AQGg4cOPC7bA/hfwuhCSaEmxJJSUmYMmUK7r//fgwYMACFhYXo1asXdfzee+/FrbfeitGjR2Ps2LEYMmQIxGIxRCIRAGDp0qXweDy466670KdPH8ycORO1tbW8/eXk5KCkpASRkZFITU2l/n7q1ClMnDgRPXv2xJNPPokXX3wRHTt2/F331qFDB7z//vv46KOP0LdvX4wdOxbnzp0DALz88stQKBQYNmwYHnzwQeTm5mL8+PHX1M/cuXNhNBpx7733olevXnj00Udx6dKl32V7CP9bCCVahhBCO/Dzzz/jlVdewU8//fRnmxJCCP81CHkwIYTAAafTiZ9//hlerxcmkwmrV6/GsGHD/myzQgjhvwohDyaEEDjgcDjw0EMPobi4GHK5HLfddhtefPFFqNXqP9u0EEL4r0FoggkhhBBCCOGGILRFFkIIIYQQwg1BKA+mBSdOnIBMJrshbbtcrhvW9v8KQmPUPoTG6coIjdGVcT3HyOVyoUePHpzHQhNMC2QyGaXXdL1RUFBww9r+X0FojNqH0DhdGaExujKu5xiREkVcCG2RhRBCCCGEcEMQmmBCCCGEEEK4IQhNMCGEEEIIIdwQhCaYEEIIIYQQbghCE0wIIYQQQgg3BDdsgnn++edxyy23IDc3l/rbkiVLMGLECIwaNQrTp09Hc3MzdWzNmjW44447cOeddzIUWU+fPo1Ro0bhjjvuwGuvvQYyL9TtduOZZ57BHXfcgYkTJ6KiooK65uuvv8bw4cMxfPhwRonXEEK40fB6/cgvb8Su01XIL2+C18tfH8XvD6C41orDF+tQXGuF339tOc/BfbrdPt52g/v0ev3tPvda7fuzcDX2X+u9Xum6//Yx/L24YTTlcePG4aGHHsL8+fOpv916662YM2cOxGIxli1bhjVr1mDu3LkoKirCzp07sXPnTphMJkyePBnff/89RCIRXnnlFSxevBg9evTA448/jv3792PIkCHYtGkTtFotfvjhB+zcuRPLly/H22+/jaamJqxatQpbtmyBQCDAuHHjkJOTA51Od6NuNYQQABA/9FvzK/HS1tNwevyQS4R4bWwGxmbFQSxmruX8/gB2nanG7C9PUOeuuLcHRqQbIBQKrrlPo16B6bcnY8G206x2ATD6NOoVeDonmWEv37nXat+fhasZ32t9Fle67no94/9m3LAJpk+fPgyvAgCjAmCPHj2wa9cuAMCePXtw9913QyqVomPHjlTp2ri4OFitVvTs2RMAMHbsWOzZswdDhgzB3r17MWPGDABExcDFixcjEAjg4MGDuPXWWxEWFgaAmNQOHDjA8KRCCMHr9eNMlRlVZididQqkx2qpSSD4WEqkCmdMFlQ3O9EpQgmvP4DqZuJYWowGFWYHTM1OSEVC7D5zGUsnZMHh8iJKI4VSKsZ3Z6ph0MqQbtDgQp0NVWYnItUy6ocHAMKVUpyrboZQAFa7MVo5OulV1I+S2+3DyctmVDc7Ea2R4d29F6h2cjPjqMkFAJweP2Z/eQLqSb2hkIjw4cGLmDowEQIB0DVGg7mb81nnps4cBADYerwMayb1RqPNgwiVBDvyKxAXJofd7YNBK4fF6cFlM3scYrRyJIQrUdZo57T/ap5LW+PQFkrqbYzxpd9bYpT6ms8FiImlpN6Gknobzlc3I1wpRZXZCafHjyW7CqgxUkpFV9Xu/yL+tETLLVu2YOTIkQCIQkVZWVnUsZiYGJhMJojFYkapWoPBQJV3NZlMVIlasVgMjUaDxsZGmEwmxjVkW1eCy+VqM2Ho98DpdN6wtv9X8EeOkVgqxYlGCV7edoZaWb46Jh09wj0AwDq2eHQ6Vu8rgk4uwQP9jFi0nX2stN4BuUSIRaPT8ffd56jPs4YmY/3hUjTa3Vg8Oh0/FlRh99k6PDs8hfrhidXJMam/EStbJgqjXoHptyVhwTet/bw5thvSNC4IRCJsza9kHJuZk4wNR0pRZXZCJhZS7VJj6/HjRLkZ7+0rwsLcdPxzP2Ev3Qb6uZeqG6FVSDA0LRb/t+EY415nfnEcbm8AD99ixDt7LjCOfZlXhrxSM4x6BWbcnsQYQ9J+v893Vc8leHzb047T6cTl6kbee3PVlTP+ftmj4jm3gXWuUCRCgUWG57aeZY0/ANyXnYD71h6B0+Nvc3yD2/2j8Ud93/6UCeYf//gHRCIRVUmPS29TIBDw/v1ar2kLoUz+Pxd/5Bjllzfi5W1HGCvLl7edwcZp/QGAdWzBN2cwdWAiuho0mBe04iePrf6pCE6PHwuDPr+z5wL1ecE3Z7BmUm/sPlsHo14FuYSYDMb1iqcmF6DFC2mZQMh+ntt6Ft/OHIQ6ixMLvvmVcWzl3tY+Oke2tktCLhGiUyTxI7poR6t9dBvo50aGqeH2+rHwm+Oc9wqAmlzox5ZOyEJe6XHkZsZRk0Sw/W2t3LmeS/D4tqedgoICRKo1vPeWFlTQTVZr5Ty3syECiVEJjHOLa614bsMBzvEHwHiObY1vsA1/NP6oTP4/fIL5+uuvsW/fPnz88cfUD7/BYGDUPDeZTIiOjmb9vbq6GtHR0dQ1VVVVMBgM8Hq9sFgsCAsLg8FgYJSuNZlM6Nu37x90dyH8N6DK7ES4UopxveJBrj22HKtAtdmJAMC56kw1aBAIBDiP0dcv9M+ZcVo8NrgL/P4A3n2gJ97ffxE2lxczcpIgEQnwzr2ZUMgkaLJ78NGj2Sgy2WCyutA1RoOUaDUGpUQz7CuqscLl9XHaIGoJ8VQ12TEzJ5n6oSNX2Jeb7ACIrbhUg4ayYfGobqgwO+EPACIBEKGUwmRxwecLcI4R+W8uGxxuLwBAIOA+XmNxtrnlVN7o4LwuIVyBGTlJlA1XaueyRwWN14fnR6TijV3nGOPg8bFJF530Kqx6sCdOVpipceger0MnvYp1rqnZybCRfMYIBBClkeFsZRN+KqwDAFQGPQujXoH5I9JganaiuNba7u2+6wn6GMn+ABv+0Alm//79eP/99/Hpp59CoVBQf8/JycGcOXMwefJkmEwmlJSUIDMzEyKRCCqVCidOnEBWVha2bt2KSZMmUdd8/fXX6NmzJ77//nv0798fAoEAAwcOxIoVK2A2mwEABw8exOzZs//I2wzhPxwJEQrWFs+socmIj1DA7wfnqrPW4kSD3cN5jO40k58z47R4oK+R8njkEiEWjkpHg9WFVXuLkG3U4d7sBMz68iTClVKWPQtHpeOfPxcxttpK6qxodvk4bRiYFIn0WC3CVVLM3ZxPxVkCAWBjXhlyM+MQq5Pj4VuMVNzFqFfgiSFJWLu/mOp39h0puFhjhUgAzjECAJub2waNjPg5CVdIuFfuKra4Ij0Q/tigRM7rKs0OrNpbRNlg0MrbbId+L7OGJsPm9lHjMCLDwLoWANzeAGMcVtzbg/O8GK2cspHrGS8anQ4A+KmwDlaXD9vzKzF1YCI0chE0csmfGvD/M0gHN4ymPHv2bNx///24dOkSBg8ejE2bNuHVV1+FzWbD5MmTMWbMGCxYsAAAkJycjJEjR+Kuu+7CY489hgULFlC1z1955RW89NJLuOOOO5CQkIDBgwcDACZMmICmpibccccd+Oijj/Dss88CAMLCwvDUU09hwoQJmDBhAqZPn04F/EMIAQAcbj9ri+edPRfgdPshADBraDLkEuKrQf6oxYUrseVYBWbmMI8tHJWOHScrWZ8fG9wFi3Ywt4kWbT+DiJYf2YcHJFLbYON6xbPsWbT9DHIz4xj2uX0BThteG5sBmViIAACZWIjptyXhg4PFWLW3CB8cLMb9fRLw1W8VmJgdjy9+LcPUgYmYkZOEZ4en4p8/FzH6XfFDIXz+ANy+AOcYAYQXETxGi0anQy4RYkZOErp10GD2HSmsMay3u1jU6FOVZizZVQCnx895b7OGJmNTXgXDhnqrm0UD5wrUf/ZLKTqGKwEAYiHwcm43yiuh04fpNgCthIt9hTUsanEnvQor7u0BuUTI+YwXfnMGkwd2BgBsz6/E0znJ+OBgMSxOH17dcZYV8C+pt3G+ozeC3sxHZuCz4XrghnkwK1asYP1t4sSJvOc/+eSTePLJJ1l/7969O3bs2MH6u0wmw8qVKznbIieXEELggsnC3OYAiC+byeKEUCDA+sOlDA9g/eFSPDMsGVVmJzYcIY6lGjSIUEmw5LtzyM2Mo879/JdSPDs8FT4/93aaxUVsIzlcXuo435ZS8Nab0+tn2JAQoUCDzQ2xUEAFluUSId4Y151auaukInSP0+LFu9OgkAghF4tY22ckQYDej1TETRboGKHE7DtSYHV68Na9PdDs9EArl6DO6sJT/yJiNusOCPH8iFTMuD0JTq+fGkOb24d1B4rx2tgMvLv3AuWd0W2g35taJsZrOwso20gbfr5Qi5V7ihg08OCtq1idHPdlJ+BZmndBeiVcK3l6oJ5OuAhe5QuFAoxINyB15iCcudzMOUbNTi++mNYP0RqCTdcrIRyFJku7tw1vlKcRPEZt2XC9EJLrD+GmQ5RaxrkVE6WWQSoWodHuxuqfiqhjRr0CBp2cigNsz6/EoORMiAQCFNZYcbKymdHOuWoLUg3cQeaOEUosGdcdCXoljHoFSusd1LH2bL2REAkBBID4cCWLavz8V6eowDh57dIJWRCLhIwgdDBBgN5P50gVjHoFNXmS9x2plqHW4oJGLkZNsxM1Vje6xoix7UQFRc9WysT45FAx+iVGYfVPRYjVyTExOx7J0Rosm5CFTw4VIzczjgrcb8wrw/N3paHQZKH6yc2MQ6pBg0a7m/Hs5BIhyDBKuFKKsgY7fiqsQbRGjmyjDv0SoygK9vLd51heiVwihF4lZa3k+QL1Tg+TekxSpROj1Gi0uTmfm0ErR3anCOpv5I8317nRGvZ239XSptsL+vbelWy4XghNMCHcdAgAnIFwCID0WC1eG5vBTFy8LZlF1/3+1GWkx4VhYW46tU0ilwgpGvCBQgkWjkpnUZqf/+oktXInKbjkllMw7Xf1vtYfffIzGUchz505NKld3s/lJgd8PlmbBAE6rXrOHUl46rYkLPyGaf+HB4uw+2wdtR24/3wNDBopxvdKYMabctPh8/tY9pLHPH6CZkx6GsGxqs9/KYVOLuJ8ThuOlPJQu5Ox4JvTrHMBplfCN2YiIeAPML1J0j66h0h6E+kGLRaPzmD0uXh0BjIMWtY7R26tBXsl7SESkPb9Xk/jamy4XhAEuHi9NyFuJE02RFO+Mv7IMSqutWLyx0cZW1s7TlbitbHdoZVLGImV0RoZHv7wKGvV98Ej2RALhZi3JZ9qp0+ncLy7p5BaRXfUydAhXIUmhwcxWhnmbs6nPBaynfWT+8JkcSFSLaFYZIEAcMHUhCkDk1BjcSJKLcPRizXolxQDt9eHaS2THQDMyEnCugPFDPuMegXmDE9leASv39MdEpGQ817WT+mLWosLQoEAtRbCKxnQRY/nvzrJ8mDmDE/FzM+PU9cunZAFnUJMTcD0dj+Z3BcNNjfqrU5EqGSwOFu9m9nDU3HoYj0r2ZO8dk1LYui6A0VIjgljeCWl9Q5Mv52IM5HXBX8m2yG9EvoxrjGTS4TYOK0/nF4/5m1ufaZkn/RxOHKxFi/npqPe5kZVkwOdo1SotbgRppTgsyOX8NzI9DZZbjUWJ6I17KRRepJpQVUzvsyroLYH5RLhFenZ7QFpw6XqBnQ2RFwXFllb392QBxPCTYdOehXmj0hjrORmDU3GvM0niYTIMRlY/RMRI3j7vh6cq8mKRgcqmxworXdQ20skq4jh0bSsxifd0okxuZDtlNTbMG/LKWrFveUYEdCe1N+IRz86ymAnzd+Sj/G9OjLsIQPj9JX8U7clsbyJyiYHbE4vp0dwptKMRTsKGDakxKhxX3YC69xAwM+w3+H2wunhpk6X1tvw6ZFSPNDPiNmbmPZYnB6s2lvE6038WtKIdQeKsXh0Br7MK6USOEkvJThuxRfH4vJKgseMjOWkxxJyUnT5nBdGdmWMg1GvwBODkxgeTXAca8pAbk9DKBQgMUrNeYxLZoiepHu9PA3SBlddOSvH50YgNMGEcNNBKBRgWNdofDq1H6rNTmgVYnx08BL1A7H6pwtYPCYD1U1OxOrknLGIKLUMSqmYEUc5WdkM6W9l+PDRPqi3uiETC1HWYMPgrtGI0XLHfeRS4isYHAfYmFdGEQ0A4L19RcjNjOOMjew9V401k3rj15JGpMSwk0EX7TiDDx/tg0MX63HkYi0RK3F7oZQS3sTDAxKp+NLGvDKM6xUPnUKCeXtPsuIUayb1BgAqrqKUiqFTiDnvLUYnxzPDUvDUv35j2fPx5L6YkZOE5GjuWFUgQCZansaGKX3hCwSgkIiw+qdCLJ2QBbVUxHld8OekaA3UUlGbYxYIAF8cLUVytBr1NjfKG+yU/EtsGDPGlZsZx2KO0eNY1xrTOFNlpiYXst139lzAB49kI1an+FNyZq4HQhNMCDcdvF4/vjl1mbFaXNiy5VFrdeO+7ARqy+fe3rEs2ZbFo9OxI78S354xMaRX5BIh7s1OwIrd5zCpfycU1dgZK99Fo9MZMY2FuelYt/8iZRcZO9HKRJzeg1AINNpdeGJwEivuU15va9MjqLe6sP98DWfexvpDxcgrNTP6OV/FzXo6X21hxVWMegUrFrFodDre3VOI/l2iONs5dLEOq/YWwahXsOJY9NgJEXtw4e7MDjh8sQ67zxL/ZcZpGddtz6/kiIcQsj1TB3bmHLPSljEj4zlcXsmlOlu7PCWZWPi7Yhqklllwu1anF4lJ/726ZaEJJoSbDlyrxUU7CKmT89UWBovo7qx4RnyBWFUTki9fnaiivIPyBjsUUjF25Jdj/og0+APA/K9OUdeV1jvw3r4irJ/SF402NyJUUizZVcBioHWN0SBaI8MjHx1l9Lly7wUsnZAFAcBijS3acQaftHgEt3aJpBIG6e1qFRIMSonmzNuYOjAReaVmhpfS7PByegRdDRq8Ma47nvj0GOPeVu+7QI2DUa/C8u8LkFdqRv8uUZztkEyw0noH/rm/CB8+2gfNDg/OVjUztpvkEiGiNETuEJ0FNSglGv/cX8Sgk3+ZV4r1U/qipkUYc1NeKeYMT4VeJcHUT/I4nzcAllQP3Stx+/zt8pT6d45Abuaga/Y0YnUKbkaa7sYxvP4IhAqOhXDTIXi1GKuTt2xNBdA7IQzhSil1rNHm4VxZNtkJYcxwpRRenx8ysQhhCjGyE8IQCAA1FhfrutJ6B6xOL7QKCYQCAcb1isfMoUmYkZOEF0Z2xbv398SlOhscHh9SotWYfjtxbEZOEu5Kj0GUWooAAnhsUCJiaT88To8fjXY3OoYp4IcfSydkUsmKRr0C7/2lF4pMVqTFaljthiul1LYROQ6mZhcqm+x4dnhXRtLjX4el4Lktp2DiWG2X1jtQb3WjvNGBy00O5JUSShpcyZMzc5Lx1W8VjGtPV5rRaHMhVqugqMkU48znw+GLdQgEgDWTemHm0CQkhCtwb3ZHqGVEQrZAAFQ2udBoc6OLtBm9jRG4LTUW8zbn41hpE+czLGtJMBQJ+fOQtudX4rWxGZT92/MrsTA3nXE/L+d2g9XtwdWCnkwZphQz+gmOC/23IuTBhHDTgb5aDKa70oOrVWYn9Gop58oyQimltoqmBVGY527Jx6isOM7rzpssWLLrfIvicDLW7i+mpGKe/uI4gxxASsWQsjKTP/6VM7AslwhxtqqZSj58dUwGZg9Lhh+ARi7BU5/9xtkuea/EmLDHgZ6wKRQAipYfP52SWwqmotGOVXuLMGtoEnWcTJ6cNjgRydEa6BRivLztNCN5kvQArG4vVuw+T8WIFFIx1h8qxoCkKKzcU0TVr6FLuswaSpASGu1uQnEhTAFYmhgJkbVWF6e9CXpVm3GgQUmRGNczjkqWrLE4oZCI8GqLYKhMLERStBpLdhVQ49nehEiuZMo1k3ph4+P9Ud3shEEnR3qsjlVH6L8NIZpyC0I05T8Xf+QY0Rk7UwcmctJbpw1OxMo9RVg6PgO+gICRz7JwVDqEAT8qm13YdqKSRQDIzYzDV79V4InBiai3uykBRb1KCpvbhyW7zjNotW1RbFf/VISVD/RkBO7pxz84WMyYEOnHALTZLvl51tBkWF2+K55r1Cswd3gq3D4/FFIR3vyugDVRLdl1njP35eXcbrA4iVW+UCDAih8KqWOz70hB304RaLC70GT34oWvT1HHXhiZimanF06vn0FTptu4bEIWzpss2J5fibfu7YFGsxWdY1spuFw/5q+OyUCjzYVmlw9amQgRahkj9+nVMd0hEQlYNWjobU0dmIgdJ9nP/6NH+16RThxMlb+aa68HrreacoimHEIILRCLhRibFYfkaDUqeBR80ztosWR8d6hkYqz5+SJjVb1u/0U8PKATMuN1nNIrwpZFp9PrZ6y2/zoshfoxoQeL+Wq4kOfSZWXox1MNaiyfkIW/fcuWUxEIwCv3EpyEqVdJEaFsW66GS3qFnDQsTh/WHy7F+N7xAIgtyPWHS/HJ5L6ot7kACBir/OdHpOLte3vgdBVRYE0mEuJ4WSMW7SigtvR8vgC0SjFOlJmxqiXjn4sS7PT4cd5kwboDxZiZk4y8kga8/t15ljdBejNEXpEcxXVWvLytdUKZM7wrpg1OhFJKiFJO25BH9cknFVNSb4VKyn7+DTbXFSeJepuLk8jRnmv/mxCaYEK4KSEWC5HVMRwur5+ThhyukMLh9iNKI4PZ6cH5agt13Oz0IFYnh0TILb2ydEIWxvWKx1s/FjKOvfVjIRVYBoBsow4PD0iETi7mtIHcW1DKuGnAURoZai1ulpyKUa9ASoyGl8obLEFT2uCAQIA2beAKhL+64yyDnktvt9HuRpRGBpFQgIc++IVx3Ru7zmHphCys2tvqRZH0Z51cArvLB4EAEELAGkM+aRv62JPnzv7yBOKm9WdJvBTXWvH05621bnIz4zBv80nKm3z7R+Z98rVjcXqwcu9xln1kXSGgNbExuCKnlEe2h37t9QKXDX8UQhNMCDc1bG4vJ4W10NSMl7adRbZRx6IpL8xNx8VaKzQyCeeqv9rs4KWzVpuJ7Z0asw33ZhOU4ZRoNZ4YksTYhls0Oh2b8soAAJ8cKsbi0ekMGxaNTseafRdRXG9j0J/JapjzNucjXCm9ogQN6RGkxqg5pGEy8GPBZQBE/KUtei6ZEEm2S9J1q3notyV1NsbnBqubSlQlvaS2JF2C7edrd8+5Gio2RXoiwVIs9GfF99y42rG5uBNM7W5CBqct0Uq7u+1rrxf4bOjSQo640QhNMCHclCBXdSqpGIt2sBMByZVwv8QoVnXJRTvO4L0He0HB4yH0iA+Dy+vHOo5jGXE6LBnfHUa9isrUH5QSTU0uZB8LvyGSEU1mJ6K1hFQMkcDpQqRahk2/llGFrd7bV4SlE7JQaLIwEi3JrappgxMRp1OgrNGBL/PKsGx8Fi7V29AxQokVu89hXK949E4IYyVELvjmND6Z3BdD02Jh1HOLX2Z11BEexb4L+PvEHqizuhDbEqAWCgUw6LgFFl1eprRNjE6Oxwd3QVGNhUpy9Ae4KcG9jeF4+74eEIsEWPvzRUbsid5utlGH3sZwLBnXHUqZGB8evAi9SkolyNLlYOj9tEWrJj2a1JmDoOLwLI16BSQiIQ5frINSKmaUAaBfyyc8qZSKcPhiHWJ1cvj8QI2F8DoSwpUoa7SzPKErgU8481+PZF7x2uuB0AQTwk0H+qruhZGpba6w+Va0v5U3UZTV4CTBE+WN8INbUPN0RRNe/+483rm/VYKGr49DF+uoVXOwUObMnGTE6hpRZXaitN6BQpMFq/YWYUYOc9VfZXZi5R7i7+S2ksnixPwtp/DamG4Y3ysBi3acgWBQIqcNNS3nrnwgkzdZkWx3P01Gn1ypd9DIOZMw39vXShx4YnASpnAw5LgkXRaPTseCbacZ5IJaK7FNSJYBAIjJZSItYZa097ypGd+cqMT025OxgBaDWTwmAwu2neYUHqV7SfRxEQkFLJmeJwYnUVuCfDGjGosTfTvpWcKTr43NwMwvjsPtDbASWenyNVfDVuMTzmx0XF9PiQ+hCSaEmw70VV2NlVtynb4S5otjkEmCpPdArx4pEACF1U1YM6k3Gm0eRKgk+PTIJSTHhAEAojWyq1o1r95XxBCwJCVduOIfbcVd5BIhYrRyfDGtH8RCISM+wrUaj9HIsWRcdxi0Sjy8uTX5M1wpRVWzA8nRRPnl7fmVDHu3Hi9DpFqKOqsbDTYXo7Lkprwy6l64pG3ocZYT5fX48NE+qG12waCTY9n3BRSLzOkh5FTefzgbWrkYXaM0SI5Wo7LBjkitnCHsSdqbatBi5tAUfHrkEkOKJ+9SLTZMIYRHO+jk+PbpgaixuihaMr109Pb8SkoOZu+5aqpMgSFMzkrK5YoZRWvkLOKBQiLCzC+OU0Ke9GJvuZlxrMTg9sr383lKcpnsDynbHJpgQrjpQF/V7T9fw/JC6HGK7fmVLIkX+oq2tN6BohoLVdKXHtMY0T2WJfN/5CKxrVXeYKf63X++hiXtvzA3HZ8fJfrgkrSfmZNMxT9IuRrS3uD7IWnMpA06hQhdDWE4VFRHjQOXaOb025LwcMs2Hj0ewpUzQ7f39pRIDE2LpX7gSXu3HGtVBx6QZGlT2kYgILyQYWkdWN5NZZOL4RFcbnLg8W/OMAqZLRnfvU17SS+qtN5BCYROotlLFjIDgPv7Ghnew2tjM5DQUinzfpr0zrPDUzjvhcwfosemAKb45eGLddTE2V4hz/bI93NJ9M8amoxnt5ymRDRvZMnk0AQTwk2HK0mO/FhQhTfGZaK8wQ6DTo7PWla7CREKVDY5WFImSS2reDJPo8rsxPN3pbFW5nSJGUWL0OTSCVnQysVY+M1phg3/3E+IW56sbOaVMvl4cl+smdQb7+4pZJQe2NIiuFlrcSFaK4MARGEyvVqKrb+VoatBAwCMGAI9IZKo1imjYkQAGPEQLnsWtSQfnqxsxqMDO7e5kqePGV+SY9cYDfp0Cm+XR6CQiuH0+Km8ptU/FUEpFbdpLymRs/onYpwXBsXZXtp6GsnRamjkEpb38NLW0+iVEA4AjGNGvYrzXvon6qkKl3weQ7CnwdVO8Of2iGrSPaXSehuOlzcxcqauRyGzthCaYEK46UBf1QkEYEjuA4TsPgDIxCIIBQKM752AWV+cQLhSism3dmJImfx1WAreaMlDoa+U+XJXmh1ezMhJgkwixB3dCCmTRaPSWTYAoNhSfFImlY0OrNxbiCeHJOEVemGzMRlUvCbbqMNf7+gKIAABAK8PFFPJ7fPh+RGpjGTQBL0SZQ12eLwBhCul1NaQWibCX4el4K0fC9uUxgf45XUEAu4x44pVvf5tAZ4ZmtxmP6TnRAqGkn0AwPv7L1LMuyuVpA4+HquTY1yveJQ3OhCl9lOkA/q1NRYnRY8mUdlk57wXp8cLkZCdkU+nD8fq5NQ7GRwHIuVqgmMwV0s3dnn98Ael1bfXE7pWhCaYEG460Fd11c0OrDvQujrMjNPigX5GxrbM4tEZeOmuVFjdPujVUkwbnAh/AAz5FIAIqG/MK8OKiVkIV3FLzAgAajvthZFE3fr4cG6hw76dIjAjJwl9jBGcx2O0MtzfJwFev59hk04hxv19EqBXSwEIKKFHcossLpxY+UYqpXD5mMmgs+9IwSeHSvHmuAxWNv4LI1Ox8r6ecPEIQKa0eCWxOu7SBNnGcAgFiZxjRgp5XqixUB4iX/7PwKRIdI3RIBAA1u6/SAmGBseiJGJCkaFHfFi78oGcnitLB5Hnkt4DvV2ry4ft+ZUMT3RjXhmc3jgW+QEAiz686sGe2Pn0INRanTBo5RjezYBaK1GcjC5X05YnFAwumnKwzNCNLJkckoppQUgq5s/FnzVGpfVWHLrYQMU/Vj3Qk8rDICGXCCml5bZkZYgf8G6I1SlhdnigkonxyvZWxtPCUenweH2osRIe0JGLtXh0APFjJBQKWNIrcrEQi3YUYPHobpBJRIw40GtjM6BVSOD2+vHNiXLkZnWEw0VUjMwvq0NOtw5osBH9rP2Z+SO8fkpf9O2sx7HSBvxl3S+s+5k6MBG3JEbgsfV5rGNv39sDi3acxeRbOzHkXp4bkYo1+4tRZXbi9XvSYXZ4GZMT/X7YY5aB1fsuQCeXYPLAzrhUZ4M/AESrpVBIxVTGPemdjc3sALFYiL3nTThZYaa8r24dtCiutcHm9qFvp3C83MI2y4zT4okhXeAPADaXFyq5GGIh8MZ35xgxGHLb7ErPmG+S4GJ7cU1On0zuC7lEhJlf/MaSvbkeVSvpKK614q6VBzif8QcHi69LDCYkFRNCCDy43OTE57+UUlIwfNspbq8PGpmY81icToEZOUmIVkvhCwjwOE1m5LWx3aFViKGRiVFeb8Oi785RP0ZPDE5ql/SK3eODQCDAtMGJEAuFSI5W480W6RWuCpaLR6djxe5zVI2XhbnpwNFSnKxshtPjh6nZBYC/BolAANRZ3ZzHvIEAXr47Db4AGF5TuEqC1Bg1qsxO1Fk9+PxoGWMlv/5wKR7om0C1Ex9GjJlQAHSKkuPvE3ug0e6GzeVjeFTvPtATnz1GFIaL0cqR2UEHqVQEr5dQtKafu3hMBr74tYxB5SbJGM1OL3MbcXQGHh/YGdUWN+GJigV4f1I26m3c992zYxhnHIXOBDNo5SiqtVLjkhqj4ZTxOVBUR0nbcFGYr+cEw0dTTo9V49uZ115eoL3475bqDCGE3wmlVAypuOULFgCiWujDdMglQnQIUyCro451zKhXQCEj1mnGSBUrYfKlrYRwY53VjdU/X8TUgUT1yGeHp7Jqs7y64ywsTh9W/1SERrub2sKJC1Ni3cFi+PxA50gV3txVgNzMOKqd4AD1gm/O4OEBidTnRTvO4LHBXah7idHKcPhiHWJ47jUQACJUEs5jGrkYKrkYy3efo2jJ/gCw4odCPDqwMwDCxrgwGboaNOgYpkCqQYO4MBk6RaqoMYvRydExTIHkaA2WfnceESopukSp8dxXzCqar+08C7FQgAiVFBEqKaUuzFXTZ8G208jNjKM+r9x7AeN6xeOxwV2oyaV1jE5Dq5Rh1d4irNxThB2nLkMiFiJcyX3fRr0K/RMjkRilZvwgk0yw/omR8AeAGf86jpV7irBqbxHOmSwsGR9yfJ0ePzbmleH5u9Ko0glGveK6b1eR5IFgGwxKsO7lRiDkwYRwU0MgCGAijQLMV2GxuNYKrULCovLSvQc6NZaE00PUrfd4/QxxQz5KKxkIp1Ol6yxO6tpnhiUz2uGj+TrcXtZnsl1Tsx1Pf36SsxIluaVzwdTEkqdZPDodnx25hHt6JnAKNTpayAMuj4cxpiQt2OnxUPTn4ATIZocbDo+fFWy/LzuBUWmS3NKpaWbX23F6COka+meREHB7uWVZ3F7CXpJW/ehHRznlddobUA/2FrgSRUmvhYt6Tqc/Xy9w0ZRX3NsDKr/luvbDh9AEE8JNCZLB43T7GR5Aab0DW34rw5pJvVFtdlJ1658emgK7ywe1TIS37u2BZocHhjA5FmxrpRfH8siiJEQoIRIKMHdzPnVuZryO89w+ncKREpOFg4XVWDwmA9VNTiToldTKPi6oRjyfnIpCKmZ8jg9XYOmELKxvuZcZOUkAgB8LLmP9lL4wNbvQQSeD2eHFM0OTERsmx/GSOkqeRq+WoaSmCVMHJkEA4K9B8iMkbXpGThI6hKsokgRAJDlWNjnQLVaLpeOzWNU6F+04gw1T+kIoEDDuhYteTNJq+WrSZHXUUfe2Pb8SvY3hVL5Q8LlkpUw6rZour5PRQYekaHWb20h0JhidGg20Ehg2TuuPBpsbx8ubqC2x6bcnse6NpD9f7RYZn6AmAFZCJ7nFd/5801X1ca0ITTAh3HSgM2sWj8lgrZpzUg2MFTZZX76yyYWHbzHi9ZYf1xdGdmWs5LONOlZS5qLR6Vj780WM6x3POJfPUyqoasYnh0oxqb+RYQO58q0MKi/AJ6ey/lAxgFYq75LvzlGB/mOljVi1t4iKAz384VGq6FmwMCad7nxvdgIe+egor7zOBVMzVu0tQmKkijoezMri87gOFNVhUx6TnstHzzY1O2Fzezm9g/NVzRRLb/GYDLy7pxAjMzpwnltcYwXAplWT8jqrH+zZ5o99MEPLqFew6MTzR6She1wYAMDh8VNbZnzioabmq4vBtCWoSZ9kyITOPxqhCSaEmw50qRiFRHTFVTPJLgLAkPCIDfImiDLBZVRteo1MjAabC+N6d0S0Roa5m1vjC3wyMwtz0/Hm+O68CYZxQZRmcpX8yeS+KKm3QSEV42RZHebdmQaTxQWdQoyPDl6iJhejXsFKDHV6iGRE+r2RsRzSvj6dwintrnAVNw25S7QGS8Z3R1xYq43B48nncfn8xL18d6oKyydkQSgEotQybDtRyWJaSURCyMUibMwrY1GC6TGYBdtOY+O0/nB6/ZhH8x7Jc/8+sQf++VAvRPBQyvVqaZtyKsFCkqX1Dry79wI+ndqPEv1MiVTjt7JGVLfkuuyaNQjVzU5IhNxelVJ6dSrHfGKWNzJ58moQmmBCuOlA3yv/raSOEWtoqz47+W8Sl+psrHPzSs3IK2nE50fLrrhyD5aZmTU0GS9uPY2J2fGcNoiE3Ml892UnoLZFlDJY4oVs95zJCqlYwGKckZ4RH0OurN7GksHhSygkhTxfG5NOHQ9m5XF5XItGp2PXqSrE6uQYkRHLYNaRMjgU1Ts3HXM2nWAJQgazxkj7HR4fesWHM8QtSRbZ2v0XsPtsHYZ3i2TFohaOSsfrOwtQWGPlpfJyMbRK6x2U6Cfh9RkZ7ZI06/zLTZxj6PEx27sS+FhiNzJ58moQmmBCuOlAl+W4Lc2Ad/cUUjTlhAglRX0lQTJ/yAA8eczNk3Do8rZ/5X5rl0h0i9XibFUzlS/Bd27/RD28Pj9e3naatRp/Y1wmlozvjo4RSkb8w+khBCGXTchCAOAVlkzrwC3ZktAS3Kafy5dQSHoPOqUU7x8sxtSBiSwpfHpiJem5vdci5Hm+2sIpQUMvRfB3WslkeikCpUxMyfTQ7Y/WyFFgasbqny4w7F297wLmDE/F7rN12H2W0IfbMKUvaiwuVgInn0fAJyRJzhEPD0hkywVtO43ESBX0KhmnBzYiw9DmuxsMPhtuZPLk1SA0wYRw04HOrGm0eQivo/Q4AOC5kV0pSRR6LGL1viK4vQGWhAcp806eu2xCJioaHYhQSvHMsGR4fQE4vYTg4fMjUvHGrnPUuX8dlgKrywOPz49NeRUMWZZnh3fF8t3naXv5qbA4vahosLOS+RaOSkeRyYIaqxsysYhzRWt3e4EAt3cmEgJNNjfnivpyk5117qa8Cjw5JBF1NkJiRiwEpg/pghqrGzNykuD3tzLmuFhZ92UnUFIxJAQAEsIVnPY5WxhxTreXsV1GL0WwPb+S5aWsuLcHhAKgvNHBKcUT8AcoQsCWYxXwBQKIUEkx+8t8jOsVj8Fdo6ljXB4BF0Prr8NS4A8EWsYhwBtn6ZUQjvkj0lixk/bKv5CB/XqbC0vGZ2L+lpPX1M6NRmiCCeGmA51ZU2d1sVaAipbMbUoORirE0vFZqLW6EKaUUNLzKqkIOrmIOlcrE1EeAxk0p9eTn31HClbe3xMnK82UzExlowNx4QpW/Y9nhqUwbIhQSfD+/iLklZph1Cuw6sFecHt9iFLLcKnOSiVwrp+SzbmiJRI4vZzH+naKgFAowIofC3m9EvLcbgYtHhlghEQsojw9o16BJ4YkUfc6a2gStp6oZMjhzxqajC6RaoSpJJi7OZ/laVyosfB6blaXF6v2FmHW0CTO411jNBD2iEOMVoqdTw9CiakBnWIicKneihHvHMDTOdzXycRCxvZkjEYOgQCsrbdZQ5Nh0LI9gmCGllIiwonyJrze8ixWPdiTs98YLVuu//fIvxj1CqydlA2JSHBVxcj+CISkYloQkor5c/FnjZHb7cNPF2rg8QZgc3nRIUyBl7adYgWWZw1NhtXlY8iITL89ifdz8DGynRm3J2H57kIARMB96fgsNNjdOFfVjC/zKigKK9e1SydkYebnx6nPn0zuC7FIgDU/X6CkYpKj1filpIH1A5kYpcYr35zh/PEEgNRYDarNLmbJgFHp+OfPrfGP2XekwKhXwe31M8gNwfbG6uQMD0ckAJKiNYjRyqBViHCyvJmRX0Pm3gBg6YCRff7fhmOI1clZE/H8EWkoqrHC7SM8yo8e7QtXXTlkkR0piZT5I7oCAK90DfksVkzsAV8gwKgjQ471zqcHoUt02zGNizVW3P1uqywLqWu3KEiINFYnhVYuQ3qslkocvRrwyb9cjczM9fy+haRiQgihDZgdXsbWCpeER5coNVxeZiJgcABbJm5drfJJzji9rT/C97XQfoP7bUuuhv75QFEdjlysZSQ1rnygB9YfLmXJtDwzLJnK8Vg2IQvnW+IfpIRLvdXNkMyJUsvw9o+tZQDkYiFkIiFmfXEcjwVVv+Sy1+Fhi2i+/m1BSw2SLCKXyOlBtEaGF74+TY31hiOljNIIMpEQTg+xRUa3P4AALE4vS8Sx3uaCGszgt83tw5ZjFbzSNeSz+MsHv7DujRzrWqvzihNMjYUZcD9Z2Qz8UoqPJ/dBvdWNMKUEb/1wnpLwIWvOXO0k858e2KcjNMGEcFPj5GUzNbkA3DVHso06hCklqLe6MWtoEuVpAGDUdu/RkZk8ybU9Qu4X8NV4IenQ9HYBImmQTAyktxUcSI7RyNFod1O2x+rkmJgdD6VUTMUqzlVbGPeW3SkcDTY3nro9CVVNDtTbPVBKxTDopOhq0FDVGsmk0o5hCtY4BFO9yRgWQCRaWl1EAmd5kwMfHizG7OGpcHn9UErFDDmVKrMTH7QQBMjaMWsm9aaOS8UCBACIhULYnB7MuD2JmrQ35pWhTyeiTktw0iN9TEh7yaRMOl2b77m1J2jOFXAvrLFCIREhTCmhVK3J503WnMnqGM5qq63kyf/0wD4doQkmhJsawatOgFmFkEwwpFdnJLd0jlysxfTbkqjtHnryJFdtd3rlybbo0Ecu1lLqvvRrqxuJgDvd2wmumWKy2CnaNV/y5Jd5ZdS9TcxOYJQmmDWUqDzZaHdj8eh0irUVnFRKH4ft+ZUMmi/93viqX5IJnHwJpyTd2Onx43w1IWtCULCTMW9zPmeMa2ZOMiwuD9QAXF4fRVrgqhhKJs+S3gTZJxeNur1B84RwJSvRcvGYDMz84jhGZcVxPu9qsxNZHZntXCl5kk/+5T8lsE/HDYvBPP/889i3bx/0ej127NgBAGhqasJf//pXVFZWIi4uDm+//TZ0Oh0AYM2aNdi8eTOEQiFeeuklDBo0CABw+vRpPP/883A6nRgyZAhefPFFCAQCuN1uzJs3D2fOnEFYWBjeeustxMcT6rNff/01/vGPfwAAnnzySdxzzz1XtDcUg7l+CF59JYQrUdZo51yNkSDHyOv140yVuaUYlQJpMRpUmB0wNRNqtRanB5c5jgW363b7cPKymUhw08rRvUWFlwR53OPzYzKN1gsQP+AfT+6L0jobOkWqMG9LPsubeHZ4Kov2CxA/gmS9eVVLf0a9CpEqKZZ+X4B+iVEQCICuMRpGHIPsd+2k3pCJRQw5FbpNNRYnwhQSfHTwEn4qrMO7D/RktDMjJwkS+NA/KQb1NjfOVzezPK43xmWipplIBOTqh/QejHoFS66GjE2RnlGqQQu1TITjJXW4JTkGNRZXS1JpPlVfniueRPcQyTFzuL0w291wtzDvyLFePiELDTY3IlRSqqwxX7vrp/SFxWKDVqPCs5uJ55ZqIDwUetXPX4pr8fCARBSaLFQ/uZlxWP1TEXVvPTuGwahXtev9BYjYyLzNJ/DwgEQ43F4YdHJqEpuRk4R1B9j2bpzWn+XBtCfGQn7HrpYgQOK/PgYzbtw4PPTQQ5g/fz71t7Vr1+KWW27BtGnTsHbtWqxduxZz585FUVERdu7ciZ07d8JkMmHy5Mn4/vvvIRKJ8Morr2Dx4sXo0aMHHn/8cezfvx9DhgzBpk2boNVq8cMPP2Dnzp1Yvnw53n77bTQ1NWHVqlXYsmULBAIBxo0bh5ycHGoiC+HGgovhEkyr5Utc83r92JpfyVoBrv7pAmdiHXmMDEKT7Xq9fmw9eZmZWNeS4CaViuB2+6jjL9yVylpBLxqdjvlb8nlX7kRdeDvcXjYNlSt58tUdZyEVC3i9HbpHMH/LKfylXwLnavdkRRNe//YcFYCvt7nxyaFihjxNYXUThqXFcsZ2AOA+msfCJ9tC6qrdl53AKVcDMIPxXF4eWfOeL55EF6Ukx+xUeRNGdI9lSe3IJULcmRGLXaerrhjjqrW4MP1fpygZnEU7zuCxQYkMmjLpVQUnnJImNdrdSIhQYlBSFIRCwRWlWEjU21zISTWw2q1scnF6Rq+NzUB6LPt3qT0xlj9T/uVqcMMmmD59+qCiooLxtz179mDDhg0AgLFjx2LSpEmYO3cu9uzZg7vvvhtSqRQdO3aE0WjEyZMnERcXB6vVip49e1LX7NmzB0OGDMHevXsxY8YMAMCdd96JxYsXIxAI4ODBg7j11lsRFhYGALj11ltx4MAB5Obm3qhbDYGGknobluwqoAKqwSt1p8ePrcfLEKWWwmRxoWOYAn4EUOHWwlLeiHeD4hLkvj8AtpTJttNUEh4AbD1ehki1FPVWNyob7VSpW6fHjy9/LUUnvZJaYX/5aymcHj/ClTJ8cqiYCG63xBo+O3KJWu2mxmqpgldkvyv3XsAHj2SzxBmBloqLXSKRHE1I1Hv9QHw4IcN+qrwO6yf3RbXFCYNGjt1nKrFmUm80O4g6NGt/vkjY62UncBr1CnQ1EBIvBo0MnfQKPD64C2K0MpjtdnzS4t3EaOSYuyWfQRHemFeGcb0I7749yZ+BAH+MiKw8uXz3OaoPuowMee7Cb85gzaTe8PkCnPcSLEo5sEsk+hgjqFo69HY+ndoPABCrY8rkcNmuV8tac1t+IxI61dIrywGt3HsBayb1xoycJAQCwLt7L6BXAuFZ0N9nAPjw4EXEhclhd/sYHo1UJGQkTwLA3nPVeP6uNCKpFAHMHpaMBL0KsTo50mN1nAH+PzLG0las53rgD43B1NfXIzqaSF6Kjo5GQ0MDAMBkMiErK4s6LyYmBiaTCWKxGAZDa2arwWCAyWSiromNjQUAiMViaDQaNDY2wmQyMa4h27oSXC4XCgoKfv9NcsDpdN6wtv/TUBvQtClLT0qjT+IRWORicHHJtJCfSW+BjCfQV9FkW1FqKcb3Yh5bmJsOt7cUNc1OxqqTvvJty6ZjpU2QiwWcyYknyhtx+GI9azW+eHQ61h0swu6zddSqP9hDqLW6caaiieGVcEnck/EPMlay6RjR7nsP9uT0uIRCwO5mytZzrarJdvnkaopqLFBKRZx9BI9RvdWNxqAETnJ86feyeHQ6ln5fgMEpMTxeiQMFBQXQqlRUfIkrxjUzJxknyxsZ0jbVTXbsOFnFiMHo5NyyOEUmK1btbSUCXKpugEQkZImUPjE4iVFC4M2x3ZCmccHsU7HGZWFuaxyL/KyTBiC1VuPChWrO75BQJMKbY7vhua1nGX14Gi+joM7Hec3Vwul04nxhIQosMlY/aRoX/L7r089/RJCfKwwkEAh4/36t17QFmUwWisFcB7jLG7Hy81PUF9ioVzFWY3RpdC6BxWAGF7miDpZpIY+1JctBttXVoGEdIyVIBADDw8rNjGMVAuOyyeX1w+0Dr2QK/T7Jdha0rOp3n627or1/p3kIKTFs+9/Z02oTvV29WobZm9jtfvBINhpsHsYYkrItH0/uizqrC2KBAFVmB8b3jke2MZx3vINLBvCNUXyYArE6OZ776mSb97KgRUy0c6SKs89ItRxpneOQX96I1fuOsxI4I5RSlDU6WGKXK/dewIYpfWHUq1FrcWDFxCxYXF504uknKUbN+NzZEAGL08N4n7nej+e2nsW3MwdB5/Rg5d6TrPeMHBfy88Zp/ZHWsQPPN4hAsj+AzAT9NcdYroSCggJIwjvguQ0HOO/larbe2lo8/6ETjF6vR01NDaKjo1FTU4OIiAgAhGdSXd06m5tMJkRHR7P+Xl1dTXlABoMBVVVVMBgM8Hq9sFgsCAsLg8FgwNGjRxlt9e3b9w+6wxDsbh9SotV4bHAXOFxehCklWDyqGxZsJ1ZJdGl0vn10YwRRTlckALpEq/H33edZMi3BMQGHy8sbT3C4uY8FAgG4fW3ntpDnilp2Mox6BV7OTUeT3Q21TIwYjQx/+7aAZVNytJqznWaHFzNyiJoqbdlLjxnMyOGPlbS268GMnCTUWrkLcTXY3KhqcuDVMRmMGvcP9jXC4vTA5w/gb98VUHI1xbVWLJ2QiaIaK5UsGaGU4p/7i/FAX3aMKFwpRVrLFp5IQMioWN0eWF0+TBnQGfV2N6VKkBKtxqCUaMr+LccqkNFBC6EA+Ns93fHi16co+0jplcMX6+D1B+D2BliSLzNyiID//BGp6GpQIzm6ByJUEvxUUI1AAPD4/dApZfjnz4S+2MoHenB6ng63j7K/e7wOnfQq/HKpvl3vR6HJgnCllNqW5XpO5Ge7+8rewY2KsZBbYpc9KqgtLk57r2c+zR86weTk5GDr1q2YNm0atm7diqFDh1J/nzNnDiZPngyTyYSSkhJkZmZCJBJBpVLhxIkTyMrKwtatWzFp0iTqmq+//ho9e/bE999/j/79+0MgEGDgwIFYsWIFzGYzAODgwYOYPXv2H3mbNzU6hMnxQD9mAHXhqHS8MDIVNVY3YrRMqXeulWRFk4Pa5nhtbAam3toZMokILo+PStDrEKbAS1tPUV8OpUzM2VafTuGQiril0UUCAS7U2zmPBX/ubQzHukd6o8nuwYx//Ubd26LR6XjvwZ74rdyM1BgNlrUILpIlh4PbEQBtyp706RTOKiGgkHDbL2/Zv5dLhIhQSbFqbxHWT+nDHVeBACt+vIClEzLx4sg0mKwuyMVCKCRCPP35cbxzf0/GduXwbpG4o1sHVrIkwBb5JDPs6SrIc+/sis6RKoQpAqg2O1myMnTK8KyhyShvsOH1787DqFfgvQd74VKdDQ6PDwqJEI98xKRRk6Kg5L2lx2rx3l96weL0YMrHeaytN5KKvDA3HThaCgEEnEKTc4enUu/dint7AOCPhwR/PlXZjA8OFnPaR99UuVGxlPaAi/7MZe/1tO+G0ZRnz56No0ePorGxEXq9Hk8//TSGDRuGZ555BlVVVYiNjcU777xDBeP/8Y9/YMuWLRCJRHjhhRcwZMgQAMCpU6comvLgwYPx8ssvQyAQwOVyYe7cuSgoKIBOp8Nbb72Fjh0JQvnmzZuxZs0aAMATTzyB8ePHX9HeEE35+iC/rBH3vX+E9QVcPiELMz4/juHdIjEsLZY3TyP4hSdptRWNDsSFyXGxxooaqxuZcTpcrLVS15IxmODckc+PEh5OsGTHK6PSEaOTodbiZrSrlYkQoZYxmGyvjsnAxl9L8cywrnhx6ykWZXn+nal46l/HGVImt3SOwF2ZHRjewqLR6YjWylBjdiFMJUWD1YmqZhflIXSJVsPm9CAlRoPCGhtl7wsju8IXYEudAMTfZg1NRi9jGA5cqEdChByBgIAxDq+NzYDZ7kGjw4MjF2sxZ3gq/n2xHqkGDZZ9T8QHXro7FRuOlFL3NqCLnqHKTD7HaYMTsSmvgvHcZg5N4lSg/nRqP3j9ATxKo0K3R0JHLiGSK4+VNnK2O21wIlbuKWKMQ0YHHYMgQG/n15JGqGUiBAJAp0gVdAox6qxuzNt8kvE+uL0+1Fjd1HNdOj4LMVoZimqtOFlhpvTmgt8Pevwp2L5guZ1ZQ5MxMsOATpFqFiX/WqVj2gs++jPdXj6GXFv4U2jKK1as4Pz7J598wvn3J598Ek8++STr7927d6fyaOiQyWRYuXIlZ1sTJkzAhAkTrsLaEK4XqngolkIh8MW0fsTq8WhpC3vKA71aSq2ou8Zo8DpNZZekynIlAgoFhKovfRW661QV1kwivAyBQIB1LZLrsTo5RAhg+YQs2NxehCmkCMCPJz/9jdVuo92NpeO7U4KWQgGRPT46Kw42l5czuO3xt8Y01h8uxdpJveHw+ODzBYJEM0V45ZszVILhjNuTGR7C3Du7Yt2BS5iYHY9952oo2RalVIzXvy3glH+ZOjAR6w+XQt/iwcglQiwdl4GPHu2DBpsbMokIr+44wwgyk/ks9B9HAQSMe+ukV3I+x26xWrxwVyoqGh1UFn1CBPe5pfU2VDYx34crSeiQny0OLzrpVZznxukUFNtr/eFSvHR3GswOD+e5JrMTW45VsBYyr43NwOoHe6LO6oZWLoFAEMAzG5nEjuPlRF2fp3OYz2nZhEwi9qOSoqzBwSI39OwYRr3rf9t5lpF/s/5wKXomhCE+TMmi5F+rdEx7wUd/Ju29EbGe/4ggfwj/O9CrpJwyJ5FqGRpsbkSpZShrtOPopUakGjTUlgZA7KXTZUO46KRkbRMAiAtrlU4RCIDiehuOXmqEQABGUtu4XvFUDAgAVj7Qs82g+frDJXh6aAqqm5xQysT45FAxHh6QCK1CwktvJREXJoNcTCRYTvviGOMLLZe0JhjmZsax6M/Lvj+PqQOJCcns9OB8tYWienNJnSilYggERJ8xOjlFz/37j0StE51CzCIa0OurAASN9u/3ZqHZ4UFJnY2aWOlVKel9himl8Pv9jOqc7z7ArRocG6ZAlIa7+iVrC4+1jSSDL8CmOMslQpQ1Ohhkgki1jJMEYtQrEB+hxN/GZuCplm1Nchxe2noaayb1RnmjA11jiGdMTuCkt5MQrmTErMhr524mSAs2t4PTG4tQSWF3+6CQimB2eliyPXa3D/kVTSxKflvSMdcDMVo553fTqFfdsHya0AQTwnWFL+BjyZwsGp0Om8uDJz79rWVvPAOr913AgUIJgz5KSI5cubrkeZMF6w4UU8l85GqclEFxewOMdoPbaYsQkBmnxfheTPrwwtx0yETABZOF8zqytjuVcPjRUV7RRPKL3VYC4o9nqxlU6WyjjjEupMf1+rcFVAIni+7c7MTlRu56JCS1m6TccnmIWpmIMxBeVGPBl7+WM8a3iqfC5fmqZohEAkZC6fb8SgYFmxxfUkKH/NzocMPt9bPaDT531tBkPLPxBOLCZIwxIqndUz7+lfdZHCttpTST7bZVKZPupcjEQnx+tIwzeXLmF8epd/K1sRl4dy93ojBXu1zSMdcLCeFKVtLza2MzkBCuvDEdIjTBhHCdIRaKqR8PgJl0R34mqwk6XF402t3UqjkQAH4sILa5qs1OGPUq3uqSTo+fqoRIrsZX7yvC4jEZ8PkCFD1UIACSo5lVFfkIAYEA8NjgLgyKMAD8c38Rlo3PglbBXbs9rYMOb9/XAwadHMu/L+AVhDTqFUiJ0VACi0a9glUWoFOkCoNSoqkkQYfLi9gwOVbuIWq1JIQrUGl2UHGq6be3qgOQ47ty7wV8+GgfON0+XqoxwE25JT25ZpcPRy7WMrbpPjlUjKeHpuAv/YyIUEnx3oM9YXH6EKaStFYFdXkpr69fYhQl00Lf3tsUVNFyy29lhCSN2QmFVIz1LR5jlFqG5bvPM67d8lsZlo7PQo3FhQs1FmociDEuw4eP9kGtxQWDlimD09Y4kJ4dX0JvMAWbTBR1++IhFBJU6c6RKkSqZXjzu7PUM3V6/Hh37wX8fWIPODw+PL4+r8125RIhDDoiwH4jEiDLGu3U5ELa8NLW0+iVEB7yYEL474A7SNIeIF7kgipiEiDjKsFyGluOtf4Qp8aGYeWeIkZt9+BVH18756ua0ezysaRB6O0Ey6vQyQXP8UjDnChvhB/gtKfgshmLdhRwJmmS7ZKeBt1eslImudpdNDodTo8HnfUKxIex723DkVKM7x3PSAbk84SOFNdj3zm2yCOd2s13rUAA7D9fgwf6Gln20gUiF+amw+P3YcORS4ySAXSChUws5KwmOSDJwvAeFmw7zRiHr4+V474+HVnjuTA3HU12Fy7V2bByD7PNvFIzjhTXY+WeIiybkEndG1dCKX0cyPuWiYUs2jp5jE5TD04UXTgqHe/9VITCGislDUME7ol39KEPfsFTt3FTzcl25RJCzigtRntFsctrxZ8h8x+aYEK4rjDouGmdFidR06MtmfrVPxVheLdIDEwipFZitIRo4tSBiUg1aHChxkJtKUy/PYk3HlLd5GR4D1VmJ/aeq8Ynk/uiyuxEhzA5dp8mZFrMDg/ClVJcbrBhfO94RGvkmL/lFKtdcsXNlVi5eEwG5ZXQ9/IB4Itfy7BkfHeIRUIGK8vpIRIMyXblYiHMdjcS9CpEqKS8iYzkeF4ppuHzA4NSovHPn4sYsj1cdevJa8kYQZyOkKGhS9iT9k4dmIi8UjO16v/w0T6cSaOkRxBMaSb77G0Mx5Lx3aGUinGgsJoS1SS9n0cGJEIpE2PdwWLKiyK9m6eHpiCjg7ZNr0ROo3pXmZ3YcKQU0wYnolusFuFKKeZtYVfV7BSpIp4FR7tJ0YTnyZUoumg7ca8nK5sZ7zL9XedLIu2fqKcqfX525BKyjUT8hZxcyD5mf3kCqTwJkO31dv4Mmf/QBBPCdQWXlDgpSgmAV6ZDIxe3UJg7UJIu2UYdJRD5zLBkyMUiigTAF585VWHG8t2FLO9hfC9mca/FozMYq2bSi1JLue0rqbNh//ka1op60ejW1TefzMzZqmaWTAvZbqHJgi3HKjCpvxErfmx7f14kJJhz9ITT4LhVsLfDJfJI/uhtz6+k4iNXK9tDfq5qcqK0wca7OufyouiekFGvwFMcMaTKJjs+OVSNe4M8I3K8r5R4+9WxcoaX2mh3Iy5MgY1HyzA0LZrTM2q2u3jLLJByL8HSR+S9kuKddK+E/o5WcsSpFo1uLVtA9mN2uOH0cHtRXJ7G1Xg7f4bMf2iCCeG6gqvWeLxOga4xalSbnQhXcccxenYMQ++EMEqOHQD6JUZh9T5iBR4XpmTs5ffmkTJJaPmykPGE9x7sBavbxyFPcpoh4UGuPK1u7rr1Lq+f8Aj2t3oEKTEa6ocH4JeZobwUnrjPlbw68txbu0QiTqeA2eGmKMKBAPAlLaZBCk9yFQIjpWHWT+6LqpbyB8taYkapBrYoKZ9sD93+uHAFOoTJOWNlvY3h6BEfxoiHBQJErCw3Mw55pWbkZsaxYnbkmOWVmkHGVcob7OgYoaR+kAGC8jttcCLidAooZWKGd9YtLgzv7Sti9PteS786pYwVF/rnfiJ+98ywZJjtzPHd8lsZFevrHq/jfI6dIlXUv4emRmNAFz0UEjE1LlaXj+H9DuiiZ9wL6fVtmNL3qjyNknpbu70d+nfzUnUDOhsirjstORihCSaEGwbyx0gsFiKrYziyOraoy45IpaRDSAkSi9MNZ5D8vUAAagUevPp+YWRXzn31y0126nqnx4/Tl5vh9XOvCINX4wIBOD0CkvU2KiuOJeFCD9LzxTRK6mzYcqwCfx2WQlV6lEuI5L5//FzEW4yKvj//yqh0lDfaUN6ickBHrE4OMnOiuNaK2XekYP6WU5yr8Rm3J2PzsVJ8eawK2UYdHr6lM4pqrbDzMOvoNgQzuGbmJKPO4kKDzcUqtPXqmAwU11gRrpJhVFYcI8ZGjtWVxgwg4ip5JY1Y8UMhZt+RArc3gOm3J0EgAMIVEsRHKOFw+yARCbBwVDecKCcSIlMNGnz1G1tWRiAALtXZWHGhWJ0cfn8AAoEAaR10eOfH1tLGSydkorKl2JvD7cPzI1Lxxq5zrPeO9Ah0Cgnsbh90CjFWPdgTJyvMEAuFeG5kGt78roDwdiOUjHeHvO8aiwu9jREsT2PVgz0RCACHL9YxtsGuNq5CStC46sqRGJXAOn69EZpgQriuuJLLHq2Ro6jGxpIg0SvFCAj4JVLo++jGCCU0cjH+FpR8SBc7BIgfwYwOWkjF3FIrwavxfp3CkdlBBz/8jATJKK0Er9/THf4AsK4d8Q8u74e8H3q7YUoxXhmVDquLm+2V0rLvLxQAXr8fiVFqGLQKhg1cMi3PjUjFR49mo7LJiWiNjJE0KpcIcVdmHHp3ikRilBIldQ6s3V+MxwYlthl7CF7Jk+M9sXdH7DtvwsO3dGbcm1QswM5Tl1kVI8ls9+CxD+6XXFTTx08rF/EqP0jFAjx1WxLjveKTbREJ2bGnh28x4snPfmNMkM8MkyJcKcM5k5XR5+w7UhjMx415BLNt59ODcKneihHvENnyZC0kuk1LxmciLkyOAE+phBitnLULYNDKcbbKgrvfPcD6Tv2nl0++YVIx/20IScX8PpCBxpJ6G/LLm1j03JX394Td7YNSIsLMjcdZyV7vjE+DVq3G92dN1JeZTyIlMUoNAcCQipFLhHhjXHdUm52wuX0QCQCDVoaVey9i8gAjq53gc6M0MnQ1aGAyu7CMVv2QtG/xmAwIAZy+3Ey1Q8YP6LL6s+9IwcVaG+Wd6VVS/OPnYkzMjmdtIxn1Ciwdn4VqswNuX4AhK8P147hhSl9cqLHA6wugzuamVuqk3AsJwlNJgsPj50wE/PDRPiivt1NVKnMz46CRi6CRS/Dqjlbp9kWj02G2u9Hs8lECkB5vAM0OD1RyMZweLzRyYsuTntBJ9kPKtJBjmJsZhw8OFjPYc3zaZADxvN65vweUEjHqbW7E6uSc1UXJRUXwvRr1CswdnopzJgtEAiBBr0StxYXESDXj3eGTutk4rT8AUNL89GN0eZUl4zPRIUwOqUiEV3ecvmLV0m9nDkK8ToFdBdUMMdEu0WpkxulQZWYG69uqcNlJr7omxtl/fUXLEG4ecHktwVUUyS8pX4XIZpcPjoAL6w+XUl6JTiHF2y15GXSpjedGplL/Jo/plRJ4vH7GJPLqmAz87Z502Fx+htTKwC56lDXYGecSnoQHYpGA0z6n2we3z8/ok5SnWTupN6rMTiREKHCu2spYsS4clY5HBhgRrpQyfiBICitJPDDqFVj9ICHYKBIKGZI5QEsSXrMTDrcfXn+A0QdXMN7p9bdJYV65pwjv3J/FqnWy4t4esLu9qLW4oJCKsPCb1mNxYQrG1uGi0ekwO1xodgg5+zGZnYzaLIlRKrxzX09IxcCYHnHEBBmjwdr9F1nPeN6dXfHmPekw272Y9c2JNt8doRCwunyc40v37GbfkYJPDhG1bugyQ3E6Bbf9zS5e1etOehVm35GC5Gg13txVwAjU/3M/MXnyVQytsTjRSa+CSCBkPMe/3dMdj3x0lFWh9UrbYMExzxsdV7ka3DhltRBuGpTU2/DhwYtYOiELS8Z1x7IJWdh7rhrjesWzAtixYUrOgLZYKodIKIBULMDqn4qwam8RypsckIpbvyiClq0XgUAAkVBAyaes2lsEvUbOSjh8edtpyCXEGop+rkAgYJ37yvYzEAkImY8T5fWYOjARM3KS8NigRGzMK0OYUooYrZxlj8XlQZPDi/JGwgtZsotJ7V20/QwsTh/KG4kfDRLB41Ja78D0f/2GaK0c500WhmQO0LJ9opGjWwctFcehjx9ZsZI8t0+ncPTqGMbokzxGUnnDlFKWDbO/PAGpWASry0eJQQIEgSF4zBZ+cwbx4SrEaGVE5vztSZiRQ/xn1Csgl4oZNmoVEsilQjz52XGs3EM8i3MmCwprrNSzWf1TERrtbsSGKRCtU2LBN6ev+O6kRGuoxFW+8XV6/FjxQyHG9YqHP8B8H8qbHJzjFKGSQqeUcB7Tq6Vwef3465cnWIF60qMiK4aSiNXJMXNoEuxuH05VmrH0+wKGfS9+fYpRz2b2lydQUm+jtsGCbQjeBiP3ovz+AIprrTh8sQ7FtVb4/X/eJlVoggnhd8PscGN8L4JOOv+rU5i7OR/jeyUgRitl0Ykv1XFTWvNKGvHwh0cx/bYk6ofiyMVaTL+NUN9dtbcI6w4U46nbkvD1sXJ8cqgYi0dnUF88vpovtRYXtueXY3FLbXcAqGxycJ57tKQRczfnY2haLKPP+7IT0Oz0wO3z4qkgeyZmJ+Dvu89h1d4iHCtt5CUTkMl+pA18NGvC3kqqFj0Aitq77mARfi3h7oORsNdC5X1x62nMGprMaGdhbjoOFNYAAAouc0vflNTZWN4PnzdUa3GhutnBGhfyOdHPNZmdqDYzV+PB40J6JWaHG3UWV7vend/KmzB3cz7j3eEbX65nQdK1g20IIACzw8NpX5Pd3WaiavC9kXGetfuLMeXjPNy39jDuy05ArE7OeS35mfR2Vtzbg2EDSS8mdw/uWnkAD7z/C+5aeQBb8ysx+eOj1OddZ6r/tEkmtEUWwu+Gzw8GPTdcKUVVswPdYrXopFcxJFH4Eu9cLQoA9ORDrqS2hd+cwXsP9oLN7UN+WR02TOkLk8WFaB5RRYNWjtysjgz5l1ieZNBAgFip0+mtABHEXTY+C75AgFMGh6TytlXjnqQIk1ImsTpuam+URobczDjsKajC+sl9YbI4EaORY/OxUiTHhPEm7JGJix0jlFhBo06TVN7kaA3OVVvw0/kqzLmzK6qbnIgN4x4HV8v2WnsIDOFKKSQiGSYHJZGS4/JTYR11boyW8FKHd4vE7rPE38lxIeM1ZND8jXGZ0MqZfbp9fk6xxq4xGjw2KJGSCvq1pJGXxk4+i73nqrF+Sl+Yml2I0cqwOa+UJYvTPV4HnULCWTtmzvBUAOC0h/QkyHvbOK0/nB4/Q7qG9L6CaeCklBDZVrSGCPoP6xqNT6f2Q3WzE7FaObp30FHxmWCa8ktbmRT8tpI0bzRCE0wIvxsNttaVZjCdmE7zLa13tJkYCLQmH67aW8RbyfG38iasO1CMhbnpeHZzPkrrHfj0sWwOenE63D4vSoJoqSkxSl4bHhlg5NznP3PZDCVPEmbwipWrxr1cQtR2n/MlkUE+rkcsp72bfy3Dt2dMmJmTjMIaC17aegbv3J+FTpFarNxL1JkJlrlZPDodnx25hN1n6yh7SbmSKrMTK/cUYcn47jhQSMi/kAF5LhHNhaPS8fkvpTA7PYx+iOeWQW1Zkfc2f8tJTMyOv6JXNTMnGScrm/D2jxeweHQ6AFD2Tr8tiSFBQ45DhErM6LOwugnTb0tm2LBwVDre33+RkmkprrFi1d4iTho7+SyMegXG90qgEnoJYcxkVpG8ZbvOtUj8MPtcPDoDO/LLUW12swgKi8dk4MtfS6n7nj8iDd3jwliVMbnGaPGYDCqvSi5pFaL0ev345tRlTml/vvgMlycUmmBC+K+EXtXqPXDtfS/45jRD3JAUtGx2eBnyL0D7BCHJFSslktjkhFQkRmG1CR8+2gd1Vhei1DJ8/VsZ7s6KR+dIFZHzMSARDpcXMRol5m3hlqDhqze/dEIWBGhbap5csZL3GqOWISlGhbiwVMRo5fjgYFFrrZtwFTUOTXYPIlRSVDbYMCw9FmN7xWNHfgXG9zZiybjuiNUpsOIHIu5yT2+mN0YmLs4ZnordZ+tYK2NS/kUpFWPOnV3xbotoJl06hpFw+HMRnh2einPVFuwpqMInk/uipN4GhVSMS6YmrJ/SF/VWNwqqmymWG5/nRqc4kxRyp8ePL/PKMO/ONIzKikeMVo6K+mbMG5FGeRM2lxsdIlRIidFgR345NUYxWjmjeBkZ46LLtKyf0hfvPtADMdpWOaAmmwcRKglMzQ48MzQZCpmY4RkT8aXTrHbfe7AXxCIBJTRKjtGPBZfx2KAk1FhcAIgS0Ccrm4l3fdtpbJzWHw6Pj0oyPlXZBPB4hIOTo1qSMkWUCjNpAylEaXF6WCKV7+69AGOEEl4/d0mD/5QqmqEJJoTfDT8C1GqxrZrlq/a2Jkz+34ZjlDwJGdAmJdbbEoSky9QHiw5ySW80WJ3w+QMMMcaZQ5MojyYzTosH+rbaUFrPvc9fUmdDSa2F0+tYva91i+OJwUlYt/8iAOCBvkZMDirh22DzIK/UDJGQWMHvPlvH4/WlY96WfFZhMIfLyyke6XB7GfaKhGBU2KRW5kEsJ662zpss+OAg4SE2OdyYv+UUZePDHx7Fm+MyGUKTXJ4bXV6Fbj9ZDuFhhmxPOjbnEcmf5LX7zxNxInKMAODvE7PaXK07PX4cvdSA5bsLKa8k+P3YdboK3TuGtSu+RHrKdJYe+b6Q3g9pL46WUpOM3e3DLV0i4fX6qaJic4dzJwYHEED/xEgcvljHk3jpRJOdWUyNZMhN+vAowpVSVjItWSKAfCdvtBxMWwhNMCH8buhVMmqfumuMps0VFd3DIStAThuciDSDFjqlhFcQ0uHysmTqg2VZ6PEQp4dg9Kyf3BcCARgSNPQV98nKZuBoKZZPyAIE4I3luLx+3JZmwPpDreKLZLyDXrHwn/sJOZKuBnb8aME3Z/Dx5L4orbchIUJJxWC4vT7mvZBeCV+pAYVUzPhMyrQEF9oiZVvaihn16hiGpROyKGHJ4OcWoZIwrqPHlw5drKdEP3Mz45DRQYuC6mbqB/r5u9I4x+XDR/vgy2NVlI1c8jrx4dxF0Mh3Sy4RwtGGV7LwmzP44JFsCAUCzvgXV7vBHuFjg7twCnsunZCFmZ8fh1wihKRl3+tMlZnyPKK1cpa3uDGvDH07RwBoW4hSIWE+c77vUPc4HbpEqZEQrkSvhPD/CNpyiEUWwu9GJ70KL+d2g0hIiPotHMVk5LwyKh07TlYCINg94UopRWkd35vISbC5vahsdDCOzchJQrhSikKTBeVNDqzc07rFxLfqTAhXMK41WVyoCWIjBbOICmussLt9EAsAm9uLV4LsXzSasN/h8qKyyYXz1RaUNzqI3Bhv616EQAC4vQGIhPxFzcob7Ji/5RROVTTh+RGpmDk0CQnh3HkY9H30WzpHYGCSHv5AAB8+mo1so46yb/HoDKw/VEx9nj8iFWcvN6M26L6D2w0eBzIXp8HuhlAAPNzfCJfbjxk5SUiL1eCWzhFY+UBPODw+LB6TwRijp25LQnkdUXitrMGGe3t3hEgI2N0+pBm0FL2bj+1XbyW2m2J1cqKqpz8AtVSEdY/0wvopffD2fT0gFoLFrps1NBlf/VZB3bdaJsKMHP4xbbC5YXN78ddhKUwW2Sg2i+yr3yoAEO9rmoHY7lNLRQhXSlntOlo07F68Kw0+fwC7TlfB7vHhls4RmH57ElxeH16+uxt0cqLaqVgIPDYwERYn4TlzMcWWjM9Evc2FMKUYr41tHe9ghhwZZ/P7A0iMUkMsFiIxSo3+iZFIjFL/qTkxIQ8mhOsCt5dI/gtXSvHkkESGbIjX78cLd3XDqUoz0jvooLhFxMrOr7W4EKEUc0qB6BRiXDa72sVqqjQ7qOS+WUOTUdFIiCRyrbg/mdwXdVYXwpQSvPXDeUrd9/mRqVg+IQs2txcqqRj+gA8Te3eEMVLJ3nIalY5//szcwuvRUQePj3tvPEYrw4ycJCTFqCnJHD6ZFnmLQu/tKZEY0T2WYmqRAeHHB4uBgAASMTAgKQr9u0RBKAAiVBKs+OESRmXFtStm9NGjfXD2cjN0Sgmm/6tVLmXR6HSsP1KEvFIzso06xjZjtlGHDx7JRqPdg2i1DGWNNrz0TQEVNH9iCFO25eXcbrA4PbwMPr1axtoqzDbqEKGSMQLsS8Z3x8r7esLj90MlE6Owuhnje8dT973hcCXySs2YNTSJ+94hwIx/HccLI1MpQUu5WAiRIEAIZ4YpUNnkoDwuLikeLpUFg06O+Xd2hVQsZKh2B1ddJSuGNtrdWDQ6nSowRpeHMTU74fEF8PK2U9R1ayb1wsbH+6O62YkIlZTTAzPo/jPkYegIScW0ICQVc+2gS1lMvz2JU57krXt74ExVMzJitXiGRqskj8+4PQlZHcMYVf8AYlW9bHwWShrsiNXJKRFCLlmWKI0MFqcXNrePqqseqZah2eFBhFqK5786RX3xXxmVjjSDBpfNTpyrbqakbbjsN+oVeHVMBgBgGockSjDV9NnhKfjmxGU80NfIkIR/ZVQ6YnQy1JhdSNArqQmDKwYza2gyRALg9e/OY/2UPpz9EvVsvIwtG4Ao3fz00BQ02dwsCRquIme7TlXhnt4dWe3IJUJq62flAz05j3/wSDYA4IWWJEE6eSCYnDF1YCJ+Ka7Fw7d0QhHtuXWJViNCKUGdlVBSWPvzRSJoz9Pn8glZ0CkkeHxDHq+9XPGnV0al41+/ELESuaRV7mXm0CRsO0FIzsjEQkZ2Pp+MDF0qhmTeTRvchZqI2no/yM9yiRCfTu0Hj89/RWkYutySQSvH6ctmXKDJzCRFqzEyPRZicfs2pUJSMSH814BOlZSJuWVDLtYS9FE++Qyn1496q5szmMkICI/JwGODRBAKhGh2ehir5FdGpeOLX8s4658/O7wrPpnSByV1duiVUngDftz3/hFGwHXDkVLW1htpw7QNx3hruwdTQpscXiq2s3RCFgKBAORiEQLw48lPCQ+BPg6kkOfUgYlIiFCgrIGINb1wVxpm5CShyeHh7LfJ7oEnqAIjGUQng9vktpfL64MAAgiFAbw6JgONdg/ClRI02Vw4Z7LCw1OJ1NMiNMm35ddg80AuEXJSu4PlazrpldDJDfD6wXhur43NwItbTzPIGThaytsnBECTw815jCQ7kLGJ5ROycM5kgVAAeHx+1FrdNHtUmJGThIw4HeRiEcP+l3O7Qa+SwuMLcPaTHK3BkvHdoZKKIRELMHVQZ+AKiZfBn50eP4prrZi35RQVjOeShiHfQVJuiUtE8+8Te/zHyMPQEZpgQvjdoAcoO0eqOJPPyDotfIHlrjEaRKqZtWI4g9/bTmPtpN6QiITYfKyUVQeelNoIrqu+fPd5rJnUG/O3nMInU/pgxobjjOP0ipH39o7F2F4JqLO4EKOVY0tLP2qpiNP2YEoo+flkZTPmtVTkDA76B49DldmJDw4WY9mELEIfK0wGrYL4esZouLeVojQyeIOSD/t0CmcIT5ISNB9P7otaiwvLd3F7Fh14ki5jw4htFz5yQaRaCpFQ0K56NjKJCGanDyt+ZFYMpScGkkm6jw/uAp1CzElTj1BJgQD3FiSd7NBod6Og2sLpPRj1Chh0chTX2aCTS1j2v7rjLNZM6o3TJY2c/ZwLanfZhCzoeWodke8DvWLojJwkbM+vRIxOTiVWLtlVgFSDBjFaOYNWHxsmx8o9rfJAuZlxLNrynE0nkBb75yRTtoVQkD+E3w16gNLqdOOJwUzZkCcGJ6HJTgRx95+vYclykMlyH/37IkPShU/u42hJIyZ//CuGpsXi77vPseRp+AgAxTVEELrRxu0RCASA0+VCdqdITPn4V8z84gQe+egosjtFYv2hYk7plcWjWwkMXJ/JYHHwapxLIoWkYK87UIx7sxOwYNtprNpbhHlb8hnjQvbTZHOiqtHOGO+aZu7AfkWjHXM353PKk4iEQIPNzSuJAgCfHCpmBdgXjkrHG98W4OilBs4+gxMtLzfZ25RXIbcK1+4vxtOfH8e0DccY8i/kfT//1Um8+d05VmB+8eh0BtmBHqin90NWHp3y8a9YtbcI+eVNPB6im1dGJrjdS3U2nKpoYo3hwlzifaBLxcz/6hTWHSjG9NuS8NmRSwxJogabC/E6Be7tY6Skl6ZtOIbxvRKQGacFwE9wqbE48Z+GkAcTwu+GUCjA8LQYXkkMksYJgFUVkkzuy82Mw+qfihCmkLRKpGi55VRI+igXLXnNpN6osbg5V5LdOujwzn09EBvGXCFqFWLUW13Qq+XQq6VUjgNpP0mVnvn5cQat+vTlZvxYUIVl47MoezfnlSI3Mw4JEcxgsVImZnl2e89V4+PJfdFgc+FctYUROKbTlEvrHfiypRIl2c/hCyb06RINlZzw6sjxjA9XcHqQURoZr2dByKqIsPR7Jo1277lqzLszDW/f1wMxWhnOVNTjw0f7oN7qQqRahk2/luFkZTMGd41m9XnkYi36J+qRGKlGhEqCT49cQnJMGK8EDV9lT5LaTdabn9+SGwQ4gF+Y9PKi6kbMHp6Keivhec7dnM9QpJZLCAp2sARR11huar1Bq8Czw1OxPb+cSp4UC4WYs+kEq91OkSoUmiw4crGWITlzoLAayydkwen1M+KL5L1NHZjISJDdOK0/CkzNWP1Tq4o4QNDf5wxPxczPj1N9BtsrABG/CaYlk6U0yDH8I3NiQhNMCL8bfn8AuwtMmP3lCSwZ351zdVXfsrpSSIScyX0yMSEI2ClSS8Vcso06lixKsKwMSxKjmbuu+sycZNRanJi18QReH9OVVet94ah0/GNfER4bzB1noe/tr9xThOUTM/HVbxVE8iEtRkTalxqjxojusVQC5/b8cky/LYmVpPnhwSKkGMJYVSrp9xarkyMn1cDoZ9HodLi8XlQ3ORnxD0JOJYklq29qbs0QD65SuWDbabx0dzdGnXoy6fXhj44yPk+hMdkW5qajqNaGyw02Rp9c55JkguJ6GytZlWRa8VX2rGy0Y86mk3h2eApju+xkZTNmfH4cz41MxbLvz2H6bclUn/f2juUc7x35lYjXqxj9NLV4b8ExJJPFiXmbT2LxmAykxWghlYpw9FI9Z7ypttmJ/ecJKR5monAGnt2cz3tvwe+v3e2Dw+Pl7AMBf8u7VMmqIDpraDKe2XgCjXY3ox4MXwHALjIR/giEJpgQfjfodcGjeeIF6R10eLvFe+DaV+8UqcK4XvEMYcGuMRpszy+nZFqCZWW44h9x4Qo8MywZ0VoZ3rmvJ5weH4RCAT45VIzZw1MxIycJiTHhWP59ARW/idJK4fMB0wZ3QbSWO9EyOJExSiPjXHGTHkJXg4Yh6cIl3EmuYMk26X0GS+bQk/QAor78a2O7o3OUBlu+L2CMWbDUzcJvCM+O3Pcf2CUSXSLViNXJse4g4T36/AE4vV68c39PNNnd6BihZCS9csn1k56pTiFmxH34pP0/fLQPxEIBNueVMuw9UFhNFXTjqhgaF67EknHdkaBXcnpnHcKIv63e17rq79MpHO/uKaSesVYpQW2zE8MzYhGulGLbiUrqHRQKhZyClgShJBGrf7qAxEgVPD4/xEIBr/jloJRoVvLvgm9O8z5jrvc3RitHg83N6AMg+lg+IQtfTOuHaI2cSqYsrbfheHkTw/uli1vSv5ukTbO/PIF/PZKJPwKhCSaE3w0666XW6uKUDZlLkz2hy5XQ9+d1chGrANYTgwnpmGBZGfrKl/y8eHQ6Vuw+h8omF4tFtmh0OmrNdqzaW4TuLUwrervkuVwCkMF7+6RHwBcjIhMt6Z7a7DtSeFewwVIrwZI5fIW2CqstWH+khOF58LH08svNWHeAqCa59PsCiupNv5ZcCa8/XIpnhiUz2uHb9/f5/DAHsdz4zi1vsGPhN2cYDDMyx+f/NhzDM8OSOT2JgstmLNpRgGyjjtMrAXzQBL07Kx/ogZxUA8ObmDU0GYu2n0Wj3c0QYP3kUDGm356MBduYIpp///48JaJ5ucmOmV/ks4q00d/f9pQIoF/HJ+lSa3Fy9hEIELIyJBKj1DA1OxmyPWSfpLglnxhmo8OHPwKhCSaE3w06i0wA5govJUZDaVIBzJUvvbb7otEZUEhEjPhNbmYc9eMXLCtz+nIz9hRUYcn4LJTW2yjZlrxSM6bfnsRikZGreAAIV0gx6wtiVTeuVzzj3LxSMwAiCZOU2vjgYBH6JRKJjIEA4T28fk93xGi5pUv6J+pZwph8MvukfPyGI8S9dYvVIkwpZYg6xvIIcK6Z1JsxRgA/S49eDmHqwETklZpZ1zo9REXQZS1eCVc7wZ87hCl4j3F5gcFxIHr+jdXlw/b8Sl5p/H6JUSzPiJSZEUDEGKMotYxRMI28N7LfBd+cxqdT+8Hr91OilF1j1KhodCAQANbuJ3JxAFBjDQBCAbe38/o93SEVCXljhvRnfCVJF6GAm5W3YUrfNr979D5JcUu+4+GKP2aLLMQiC+F3g84ie3//RUbxqaIaC9zeAEv+paiGEL/84GAx/tLPiOe/OoVfS5hspOCVMBn/sLm9+OBgMUZ2j4MAAcjEInh9flQ2uTivA1rjMwBgokmocJ2bV2pGSb0NT39+Ar9casDus3VU9cOvfqtAbmYcKpucOF1pxot3pbGYTOermnH2cjM+fjQb//hLLywZ1x3hSgnr3DfHdUf3DlosGdcdL96dhm6xGiAQYEm88BXaqrW4WPbzFfAiWU9OT6ucDp+cynmTBSv3FGLp+O6YOTSJkkhZMr47o92Xc7uhyeGGze1hSa8EM84WjU5Hk91FPX8yDuRweyl5IKlIiOdGpGHHyUrq3bgvOwGXm+xtPtdaiwul9TakRKux8oGeWDKuOzw+P6ekCz0HpcbipLaohEIBNHIJREIBZnx+nJpcyHMLqiwAgPf3X8STQ1rf7w8OFuPpnGT066SHxcUuTkayyACCNi0XixCulLQp6VJr5WYC1rbI6dDRVjEyruNGvQJrJ2Wj0eH/Q6pdhjyYEH436DIXZocb5Y0OSiomI04HhYQtDdMlSo0ZOUkQCgC9WopHBhhh1HOv8rlWzase6IkGmxuPfPQro931h0t5r5NJiFVblIaZr8B1rkZGfDXoBdK4Mu4XjU7H7GHJaHb5IGxhSK3eVwKdXAKlzMioFbJodDrm39kV9XYP9EoJXF4/ngtSF/jXL6W4PTWaYRNfkTapWMSyn5R/eX9SNi6bHQwmG3kuKaezcFQaZ7u3JEZgSEokzldbGMl8z41IxayhyfD4AkiKVmMJrRY9XXolEAD2FFTh/UnZaLC7EaGU4p0956l6L7OGJiOrow5xYd2RGKVibWe+nNsNaqkIhTU2Sua/rfchWiNDuEoMvVrG2hILlnShC2P6/cAD7/9CJS6+tPU0r2yPxUmQPGqtbnj9foYUUphSAqFQAL1Kjo15ZxnxJTI3i+7tjMgwtPl94st74pLcp3/3uMQt6ccbbC5UNjkxbUOrwjedEHAjEJKKaUFIKub6IK+kAXM2naC+VN1iNZj9JVs+Y8btSVi+u5D6PHVgInRyEXyB1iRJUtMq+Ed616kqjO/dkSXLYdQrMHd4KorrbEiJUeON71p/AF8dk4GNv5Yir9SMt+/tDocXWLT9DCsGQ+6/J0UrUVxjR7haimaHh0oG5JLB4ZICCU6sJI8tn5CFGZ8fx6oHenLKiiydkIU3vi3A5Fs7UTVgyJgMX2GwZ4al4FJdq/SKXiVFhzAFzlZZkN5Bg0KTFTa3jzr2j5+LUWV2Ys2k3iiutbIm/8QoNS6YLFjVQv8Otj8AMLajyGN0+ZTFo7shVqdEg90NhVSEsnobmhzEj/SRi7WYOrALzlQ1IzNOh5lfHOd8P1b9VESoDthcaHb5oJWJYAhToIgmkZIYpYLL40OsTskpHUO3iZxwiBhMOmwuL+rtHoa0DdciYtmETFQ0OmBz+5Bq0GDZ9+xk1W9nDoJICPxW1kQ9C7VUhNgwOeZvOXVVP+hutw/bTl5mSPy8OiYDYzI7QColFhVc1OMrTRJXkqBpbzvBCEnFhPCHocHuZgQo25KGoX9OiVFDKhJi8Y6zjP3tz38pxfsPZ6PZQRScitZKYdQrUccjK0MXJSQFFu1uH2QSIR4ZYET/LnZ4AwJ8/ksJIePiD6DB5sJb9/ZAs9MDhVSMdfsvYni6Act3FxI/+uO749Op/VBldvAGcYM/tyVzsmR899bPQcdJOrRMJGSskuVSIeURRKqluFRrxeCu0dDKRPB4/QxP46/DUlBca8XnR8tYk+dfh6VQ/TXZ3Fh/uJQx3usPl2L2HSnwB7jtK6yx8B4jM9Sj1VL4AgI8viGPewLPTWdofXG11TlShY8e7YNLdVas+LF1wTHjdqZEyqtjMuD0+nnldOLDCJtUUhHSO2jxyIBOsLu9kIiF+HhfCateDV22p3ucFqkGDc5ctrAo78EyODUWJ2RiIetZvDomA58/1g8un7/d0vlisRBhSgnLSyJ1xviox1eauK4kQXMjPJrQBBPCdUW4gim7wRd0Dq49LhQIEKaUoNHuZuTIGPUKyMRCiIQCSERCdNAqkRChxrGSBgZllYue+2rLZEVKgywZn4XzJjuiNTKYnR7M/Pw4ZuQkYd0Btlfy2GAV1c76wyWYd2catHJJu+4lEOCXVolSy2BzehGtlXFSbhVSMcb1iscbu85xejdCgYAhLMlFAHjrR4KeG0xgII+tmdQb1U1OJOiVrPEmtiDliOGhaydFa+BwefG3senoHKVCTbMbESoJduRXwNAiK5OgV+K5r4gAO5cN7alJE6GSQigQYA0taJ6bGUet6sm2Xt52Gmsm9YZCwi3jI2+hl9vcPry87TQWj8nAryWNePvHQiq5N5h4ABAqEoEAUGd1Y84mJs2XK1lVAAG8vgCLhPDyNoJMQGd/XQlE/I/p1Rn1CoQrpVS11g8PXqSOhyulOFdNCHh20qt4J7HggD8XzZ5Ocb4eaNcEc+nSJbzyyiuor6/Hjh07cO7cOezduxdPPfXUdTEihP8deLxMcUAueubi0cza44tGp+OrY+W4LzuOQRGmkv1o1QPJWuTdYjSMWul8K2FSguS+7ASKmUWuov+5v4g3KZMMLNMrMHJVD1w8Op11L5vyynCgsAYLR6Uz67WPTseyForw8G6RnAmRXx8rZ1VcJO+lqMYCnVzSLg/xcpODNyh+rLQRK/cUcVKyF41Ox7r9xSiut2FhbjqDwhxcpTJ4y+mzI5ew+2wdY5XfljQM3/sxMycZ+RVNePvHCwxvgU9I1Wz3wCPzt6uqZpPdTVWplNGUh50e7iqgfONLT1YlkxwnD+jEeW6jzc36nrQFPk/joQ9+Ydyb21uKWqubtaXH54WQAX/S8+GjVZMU5+uBdk0wL7/8MubNm4cFCxYAAFJTU/Hss89e8wTz8ccfY9OmTRAIBEhJScEbb7wBh8OBv/71r6isrERcXBzefvtt6HREUaU1a9Zg8+bNEAqFeOmllzBo0CAAwOnTp/H888/D6XRiyJAhePHFFyEQCOB2uzFv3jycOXMGYWFheOuttxAfH39NtobQPpB7wiKRgDPoHCy1Qe5hhyulqGxyYGKfBESopVj37wJGshw9gc/pIYQRE8KVEAoFjIqFbYlovnBXGmuV/8/9RVg8JgPNDi8kIgEVfA8EgBPl9Xj01i5EYqhOTlGng6nSepUUS78vYHgh7+0j2rU4vIjRSrB+SqvMyea80hYaNJAcE8aZjPjPh3pTZIHgFWxStAZ6lQRTP8m74n13CFOwKkKSx+gSLsdL6vDho31Q2+xCuEqCmmYHJmZ3hFouxo78CsLbMTvRMUJJlaMm7WXSfgkaOF32pD0JhuT7EUxbf3Z4Kh4blIiNeWUY1yseq38q4qV6G7RyQAAGfbhPp3As2HaaYS9JN6b/m94OVxVQvvEdmhqNnh3DUNloR4RKhmeGJqMTh31GvQI6pQSHL9YhWiOHSEjcc1vxjvZ4GiTV/3y1pd1eCD3gf6m6AZFhWk5aNReZ4FrRLpqyw+FAZmYm428i0bXxqE0mE9avX48tW7Zgx44d8Pl82LlzJ9auXYtbbrkFu3fvxi233IK1a9cCAIqKirBz507s3LkT69atw6JFi+DzEUlCr7zyChYvXozdu3ejpKQE+/fvBwBs2rQJWq0WP/zwAx599FEsX778mmwNoX0g94TvWnkAf9t5liVC+HROMtJjdeifGIkGm4v60tPFDZ/67Dc8/OFR5KQa8NVvFVi1twjVTdxJYqUNdlystXJ6SvR+SfHI0nomzZdcEf7fhmN4+vPjeGbjCTi9fmw5VoFfimsxLC0WUz7+Fc9sPIF/X6xjXEtSpc9UNeNEeSNyUg0MYc/7shNwvqoZf/uW8FQe/vAonv78BB7+8Ch6JkRSgoUamZjz3s5VW4itHBrN16hX4KmWxMtjpU1XvO9Fo9Ph9ngIYVEOQcjnvzqJZzaewLQNxxAbrsaK3ecwq+VzrdWDV3cWYNqGY+jdKRK1zQ7M33IKpyvMnDXj6bTfJruHcUwkBOUhMm3IYAiCPnVbEv6++xxF+31iSBLe33+RGk+yCmRVk52Tgt3ocMPUTCQnks/iFI+9JN3Y6fGjzuJitHO+il0FlGt8XxubgfRYHcKUEoiEIszeRIhS5pc3siqEPjGY8MAfeP8X3P3uAXx3uhqzv8zHXSsPYNeZak6acEK4ss0KlqT9Trf3qoUvhUIBEqPU6CC2oXucrk2K8/VAuzyY8PBwlJWVQdDyNu3atQtRUVHX3KnP54PT6YRYLIbT6UR0dDTWrFmDDRs2AADGjh2LSZMmYe7cudizZw/uvvtuSKVSdOzYEUajESdPnkRcXBysVit69uxJXbNnzx4MGTIEe/fuxYwZMwAAd955JxYvXoxAIEDZH8L1RUm9DUt2EZ5HQrgCLq+PCporpYSMfq+EcCRGqRGubKUItyW1svqnIt44RrRWxkpkpFeprLW6gJZkuSqzk0V/5ur3nT0XsOah3lBIRVi6q4CxgqULY5JlAQIB/gRIvvgHvXZ7WgdugcW0WA1GZcXhx4IqfPhoH9Q0uxCjlVExjWDKMpcH8N6+Iiwbn4U5d3ZlCGFyJb2ubvG4fi1pBAB88Wurx7Apr4wSuwyuRMklPR+mlDDuZUCXSCRHqxGtkWH1gz1RZ3XDoJNjZ34F5gxPhcPtRUKEEh/9+yKDykuKn56sbMbGvDK8MS4TS8Z1p2Trg5Mcl4zPQq3FxfBg+CjvJN1YLhEitoUAQAp7zh6eimaHB7OGJlEF6MjxXTOpN34taUQgAHxxtBTJ0WrYXD5GompzUKIoOd50yRf6+AZ7Gm63Dycvm1Hd7IRaJsYLI1NRY3UjOZr7XUmMUsMYqeKMIbbHC7kSxfl6oF0TzMKFC/Hyyy+juLgYgwYNQnx8PJYtW3ZNHcbExGDKlCm4/fbbIZPJcOutt2LgwIGor69HdHQ0ACA6OhoNDQ0ACI8nKyuLcb3JZIJYLIbB0MonNxgMMJlM1DWxsbHEDYrF0Gg0aGxsREREBK9dLpcLBQUF13RPV4LT6bxhbf8noDagYUlb0PfnZ+Yko6KuCa66cjQHwqi9cr7VF7m/vT2/nBUjICU8zE4P69j025IwjyZJMzMnGbVWNypbVr6kfQoJ917+r6WNWHegmNrfPlnZjMsNNpYw5uLR6ThysY43JlBSZ+O9N2cLS6zJzhZYnDU0Gc9tOQWpWEDJydPjEhuOlHLGLO7LTsAb3xYwFH7LG+2otbgZcjUzcpIYq3q6J0fvRyYWMmJPTo8fw7tFUuPNxQxbPDodO/OJZE7y8/LvC1iyPXT5n+B7o9tPj53RxyG4Iufi0elweTw4UFjNEtwMfj/Id5KMYRS3JPtyiXPS39+nbkvCxwcv4afCOsrrvm/tEVYBui3HKhjxED6JH6Gw9X24VN0AV105pAoFjtUKme96bjr2n6/hfOavjkmHztcIv8+HN8d2w3Nbz1LH3hzbDZ7Gyyio45eDCf5N0gFwuYDzdbyXXBPaNcF07NgRH3/8Mex2O/x+P9Tqaw8Amc1m7NmzB3v27IFGo8GsWbOwbds23vO50nQEAgHv39u6pi3IZLJQHsw1wl3eiJWfM4tI0ffnSRnytI7xcJc3YfFOwtvp1TGMk0mVFE2wssh9dD5pf3KVX9vsQpRWhhW7z1FtkSWTnxmaDKVMTIlmOlxeJOiVvHGBYE9jTK94VhyIjDf8ylOMyuX188rSx0cosWR8dxi0Cmw4XMIQ8iSTAqffnsQoaQCAEYugy8pEKKWYu4UtSy8Ri3i9pLY8OTI28djgLow8ntysjtRqPNXAZuyRki23JscgSiPDx/++yCnbwyVPszGvDM/flYZCk4V6B0j5/r3nqhlF5bbnl+ONcZmoaXYhWiOFVCxEldmFCdlGrDvALAPxY0EVFUOKVEuhlokRH66EViHGZ0cuYfrtKfhiWj8oJCKKqkt/f5dNyMK5agve20e8cz8V1rHGLNi7PVBYzWDpTaYJhtI9XKAlPqNRo9QqQURAitX7mGNKfw835pXh06n9UG91waCTIy1GiwqzA3XNTvQwyrHz6UGotbbfC7neJZP50K4Jprm5GVu3bkVlZSUV/wCAl1566aqNOXToEOLj4ylvYvjw4Th+/Dj0ej1qamoQHR2Nmpoa6rjBYEB1dTV1vclkQnR0NOvv1dXVlAdkMBhQVVUFg8EAr9cLi8WCsLCwq7Y1hPbB7vZxrtbp+/M2F/HepMVoKGHBs50j8NRtSQw5/kUtwpJ5pWZKnj1Y2l8gINhdt3eNpVadL4zsipxUA1buvcC7wia3h14Y2ZWTuUQvA0B6GnzFyZrsHqoYFZ1pRQpwur0BFuNsUYsYZ16pGff2jqU8o8cGJTLk+oOFG0n7SNITKTlSUm/D3oIqXjba2J5xjPsMtpfPk2uyEzlG9GN08c4ZOdzMqiPF9VRS48ycZJyqtLI8Ob6S1MGlEz7/pRT3ZsdToqRcYxjsaSzMTcfnRwnPk/Q0/m/DMZ73IQNdozSQy8U4HBRnI+/nUp2NQVsm/0+ee6aiCRNZ3m0G3t1TSL2/fB4u4TUlM5hhXPk1DrcXcokQ80ekoVdCeJsS/DcyI/9a0a4JZtq0acjKykJKSgqEwnbxAnjRoUMH5Ofnw+FwQC6X4/Dhw8jIyIBCocDWrVsxbdo0bN26FUOHDgUA5OTkYM6cOZg8eTJMJhNKSkqQmZkJkUgElUqFEydOICsrC1u3bsWkSZOoa77++mv07NkT33//Pfr37x+Kv9xA8Anq0WU5pC2/jhVmB778lShBHK6UsIowLWwp7tW/i4V3Hz0QAGuFTY+HcMU/6AW86HvlqQYNZxkAo16Fd+5nxx7I4zFaGXIz47DlNyLuU1JvQye9Ch+0yN8LBIBcLMTb9/aANxCARk6smknRzGD5fnofcW2IW5IxA1JCZWCKAesPFVNFrhRSMdYfKka/xChOGfotv5VRBbz47s2gVcAfVJI4OB7GdZ2v5SNpb1sssra8qEXbzxAliNVSRskA+vuRV3qc5SnTV/z0drnfh9NIjFIhu1MElFLuWF+nSBX1b9KrpsdD6EKd9HYp+7zcEj9ZHXXI6qhjecZc+TUJEUp8O3MQwyvhk+C/nvkr1wvtmmBcLheef/7569JhVlYW7rzzTtxzzz0Qi8VIS0vDfffdB5vNhmeeeQabN29GbGws3nnnHQBAcnIyRo4cibvuugsikQgLFiygGGyvvPIKRVMePHgwBg8eDACYMGEC5s6dizvuuAM6nQ5vvfXWdbE9BG4E8+uD97tn5iSj2enGrtNViFBJoZFJcL7aglSDhtvzAdAxTAGZRIjXxmTgpW2nWe0Gy8nTBSHbk3tB7pWnRKvxQD9mGYCFuYQw44UaKww6KRaPzqAo0QSLqDsu1VohFgL39UmA2+dHeaMDEUophqUaUGF2wh8AmhweCAUCeP0BLPu+BE8OScIr29my+sF77MGsN9L+UxVmrNpLeAikSoEAhDhnXulxxvkDkqLw/v6LnFI7y1tycVY+0IPTk2u0u1HZ6MAb93TH818TW5+fHCqmYhpbjlVg9h0plJSNXEIoBHx8qITqP1wpRZpBg+I6G9Y+3As2px/NDg90Sgn+8ZeeOF5uhj+ANgU3jR4l5zFS7SD4udI9T7qnIRAQ9ozrFc94B0zNxILC7fOxyiXMH5EGu9uLWUOT0DlSRXm/Rr0CKyZmwR8A/IFAm/bx5fg02d1wuv2c15KeklGvwMJcQs4mGHwS/Nczf+V6oV0TzJgxY/Dll1/itttug1TaqlB6rdtOM2fOxMyZMxl/k0ql+OSTTzjPf/LJJ/Hkk0+y/t69e3fs2LGD9XeZTIaVK1dek20hXBukYgGmDU5E1xgNyhqIJMXxveOp1fb9fRKwZNd56kduU14ZUg3cMYILNRZqq+XNcd3x7PAU2N1+pBk0uFRnxfje8egQxpTKD2ZXteVRkcyg5ROyYHN74fH6sGJiFiyuVg9gQFIUVu0twpCUftAqRC1lb30IU0rx6o4zjCDzS1tPUT8+TwxJYkiFzL4jBZnxOiyfSOQskPIfmXE6yka6lHucToFItZTT/r6dI7B8YiZUUjEltTJraBLnudnGcPgDBGNz3cPZMDs8iFLLsI5WeiBMIcHf886zWFm5mXH44GAx5t7ZFRum9sXlJicMWjm2HCM8z0DAD4FAwJAyUUhadzbIZMVnN+cjJVqNKI0sKOE0A4eKapFXam7Tfr7s/ODib3RPuXOkCjNykpAS0/puqWUi1hbZrKHJiGspNRChlFGenkYugkYuYSyWFo1Ox9RbO6PG6ka0WopmpxevtHhZbdnHl+MzZ3gqtAoh57W3dolEt1gtbG4flY8TvAV2JYn+/yS0S+zys88+w1tvvQWtVtt6oUCAPXv23FDj/kiExC6vHcW1Vkz++ChyM+MwKFmP42VNrC8zACzZdR5Aq+zJuv0X8UBfI2fBK/p21fIJWSiotmDHyUpq+6m3MQzVZhf1w0XmiizkYTktHJWOf/7cyj6afUcKfP4ABAIgSiNDab2dElDsFquGQipBg82DWJ2cYqZNvz2pTbFLvuPvP5wNmViIzXmluDsrHo02DzqGK3CsrJHhBSwclQ6zw41exnAcK2lk2D/7jhT0TghHldmJikY7AiDkT9QyEYQCAaudz39pjUVMzI4nCAEqKVxeL34taYI/ABg0UkglYkahLXocQC4RYv3kvrhUZ4NeLYXd48O8zSd5RT9JYcmZQ1sn2ZUP9OQU/SS3smJ1coawp1wixN/u6Q6NXAyFRICqJherwNiXeWWcMZhZQ5OR3kGDhz/MY7RLCmcG27Dx8f6we3zQyMU4dLEeK34o5L23tZN642hJI/p2CsfL204jNzMOGpkYyTFqLKYtOF4dkwGH24saqxsiAdAlWo2/7z5PHX9tbAbMdg86RylRZ3GzYmcerw9VzW5OG75t2QK7HjGY6x3k/11ilx999BF2797dJs03hJsX9TYXFZROCFdwCiiO792qpOD0+OH2+oiaG0dLW5hUAegUEszfcorBiHJ6/IAAyIzXQSUVMSRS9p2rYcQevj5Wjvce7AW72we5lPjxsbkJGX0RAnhhZDecumxGeqwWNc1OvPHDOYQrpZh8ayfqB9GoVyAuLAmzNp5k/ejy0ZJJyRG+rblfLjWgsLoJw9JiqX33hblpLEFLj9eHTw6VwqBljqFcLIRMJMRfaAHhWUOTseVYBRrtbjw/IpViS8Xq5Fj+/XlGoDtYpocsF2zUK/DCyDQsn5AFnz+AwhorK8h88GId5U0uGp2Oz6b2RXUzd70SUpMtObp165NP9JO+zRU8DjanB8u/P49Guxtv3ZuFjyf3ganZhUi1FFvyyvHwgEQ8MiCA0nqmp7z+cCmeHd6V1W6sjnsb7peSBrz+7TnIJUI8P4IoN5Cg596Wq7W4sGpvEd59oAeLgPHqmAzoFBLEaGUob3QwVJD/dk93vPdgL1Q2OSEVC7Hwm9PU2D8zLIVx324vIfIZrZVz2mBqdlJ1Y250/sr1QrsmmKSkJCgUihtty/8cvF4/zlSZUeHWwl3ehLQYDSrMDk6J7WD57YRwJcoa7VQw1ucHaizEsXidAgWmZlSZnYjVKZAeq6WUVsk+uY5x9dOWDfR+4sOJwO/lJna7UpGQopNKhALcmx1PJaoBwPBukRjQRY+kKEKe5NMjlxClkQEg5OPDFGKYHV7IJSIqFkKCcP1laLC54fL6EK6UoqolxlFYY8XMz1tjD0a9Ala3DzEaGeZtyae8HX8AWP3zRbwxjlCjUMpE+PDQJSrIT693zyWaSVJL1VIRJ606q6MOM3KS0DWGueVHeg/J0RoMSoqktpgcLi86Raqw7PsC9EuMomz84N+XMK5XPCJUEkjFAtp9qfDJoWJWwt4Ld6XhvMmCDw9dwuIxGZi/5RTefaAnzE4Ppt+exLo3AFi97wJF887NjMOsjcQqeEYO2/vKNuoYsjKfHrmEzvokRGm4hTqj1DJ4owKI1siQbdShX2IUYsO4t3PIbSQ+Yc9lE7Jw3mTBBweLMe/ONAgAiIVCHCtvwlcnqrB+Sh+WV2LUKxDXUkyNrMXSLzEKHXhsyIjTUSKlHx66hDnDUxHFsz1JJmVGa+SYG1Qp8+Vtp7G+pdrks5uY786LX5/Cp1P7QaeQsKq1Pv/VKVY/G6f1h0bObYNS2qqeQmbktyfmwvWd/6PQrglGJBJh7Nix6NevHyMGcy005ZsFXq8fW/Mr8dLW09TKOLjuN+nWAmC5vGS9brc3wEpUC26HFIAEwOiTfkwsFrbpWgfbkG3U4d4+RizYdppzy4ners8fYNFJya2L7nFqSnqFvs3h83qpeuzTWlb1XOKLi0enY+5mZvIkV8IhvY79sgmZnDTfBiuxCqUnwD0zLLld4pFFNRbsO1fDGTQnaalGvQKLRqfzbtPRqdJkwiF9e5BMwtv6WwW7/kuLOCd9HCqb7JRwY7ODkGlptLuoducO5070U8uJrz3d4zpT0UTZTj3/oCTHRaPT4fZ74feDZTtdyJNev+ZAoZpT9HNHfjkAfhmU8yYLtudXElIrNJHSxaMzsHrfBVwwWVnP/4nBSZyCpl0Nak4hTFJbjRyX6iY7nG4PZ2D+fFUzVu0tQiceD6fO6oJQIOA8VlxrRVkjs9QDnzdsd/so+4KFRv3XULqL7zvfRfbHlExuVwzm66+/5vz7Pffcc90N+rNwPfYkHQ4PTlU3w9Qi77F0VwElbjj99iRcMDXhL/07o9HmoVaEUwcSQU56ohfQurcPgBF7CF5hk+eun9IX9VY3Sutt1P48QBR3em5kN3h8fiilYqz+6TxyszoyZE+WTugBAJi3+QSVNEZKrudmxrES6wBitbh8QhbqbW5EqKRYd6CI1e5TtyVDJRMzVm50e50eH1buKWQlqt2dFY9qsxMGnRzv7imkVvkAqFwOm9uHYlMTbkmKQXXLyoyMlXwypQ+DAkr2uWZSbzzy4a9494GeWN6SlNk7gSluyCffTz6LKxUcI8sCeLx+zgJYV4rXkGKXL9OEGrn6occxso06PHtnGkrrbNRza2scPpncFyV1NsSGyal+Vj7QE3+nJaoO6KJnUISp5za5LyAApXDdnnvLjNNi2uAuRMxLKMD6Q8WYPTwV5Q12GPUqalJo73hzvevBtG/6GEWppfjo3xep716MVoaP/n0Ru8/WMc5dM6k3Uba55b0nt3h/Ka7F00NT8GtLDGYax5h+OrUfJCIh7lt7mNOGwpYJk2yXjOUEP+NPp/aD0+PDi1tPMWzYcbISf5/YA9mdri5MQY+P0r9Db93TFb2S4tq+uJ343TGYe+65B263GyUlJQCAzp07QyKRtH3RTQaHw4Ptp6sZdFa65IhRL0dCRCxDlmPR6HSUNlhR0chNO9TIxVBKRO1aYR8sqmNU7ttyrIKSHCGTueiiiXQbmx1uBACGF7JsQgbVb7AkBpkcN4kmo78oSLZ+YW46Kprs0MqlvPvJjXY3y/NZmJuO0nobFmw7i8Wj06nkSfpq8lKdDb9cIkQp6atb0rspuGzh7JMUOrQ6W4uiCTjkPviSMMf3jm8zBgMApfUO5JU08O77k19yvnjNyQoz3ttXxJl0F1zYrKTORkm6PMoxDvVBRdnI6+qsLsz/6hTlTazeV8RIpASApCg157XVFifEQu6VOl8s6mRlM2a01N4hE0qLTBYs+OYsXr47jXe8H+ibwNnP0UsNVDE48l3nk8qvNjsgFvgZ8S/yPas2u4k4YMu5x0oboZQyv2+kZ0ReS/dS6V5VhoEgQAVT2hePTse+gmo02j2spGIur/S30gY4vQHOBOOrlf0HmPFR+vhaPX9MIeN2TTC//PILnnvuOcTFxSEQCKCqqgpLlixBnz59brR9//EgBeo8Pj9DPt7pYUo9xIerOJPGPny0D8oanJx72j07hsEfCDDkJvjkw8lEMIDYn//b2AyIRYIWL4k/vrBoxxmsndQbcokIW36jiwWqseon7kRAruS4hbRERrLdNZN6s0QpybaiNXJEqKR44etTjBjBP/cX4Y1xmcQ+ukHD8H6cntaEw6QYNWei2ppJvdHs8OKFkV3h9gWoypnb8yspoUNjpBqLaGMaPPZ7z1Xj48l9UVpvg0Enx4JtpxmstuB7IRPyyM8JehXvXj59v4CvLfJegj2j4MJmnSNVmHNnV96EPT66c7hKSp27el8RlozPQpPdjedGdKU8Ar6CYwaNHL5AgPN9DR4HvnuXS4RIMWiJmIZWTsWXEiIUqGxyUBMrlzy/Ua9AL2M4lozrDq1CjHqrCwtyuyGC5167GjSQiISYvelX3u8mea7PD1g5BCvpnlFpvQPv7SvC+sl9YbI4EamWoc7qwKGSemjlEqzed4FBcFm9rwhzhqcCACspk7SBpDDvPVeNhwckQgAwBDeDn1sw2oqrSkVCTjmgDZN7c7Z1vdGuCWbJkiX44IMPkJhIuK2XLl3CnDlz8NVXX91Q4/7T4Xb7sPXkZSxoqZTHtYJyeYitqmAZcPJ4rcWFyw021p774tHp2HKsFFkdI664wqbLq5ArlMIaCz4/WsZYNfF5P0dLGqn9bnLfl9zb33CklJB9p+0J8+2bB6+wa5pdOFxUyxlX+eFMJbrHR3Curuoszjb3u2stLoiF3HvYx0obsSmvAg/fYqSCwHR6KwA02FpX9vvP17DiCYtGp8PUZMP8LacY9Ge+4mTVLcXJyM+Xm+wQIMD5nN7bR0wYXDIz9EJn5DgD4PY8R6Xj/f0X8Zf+Rs5xEAnBilOQfZTWWQFwF2IjPYLt+ZWcz23dwSIMTI5mjdnC3HQ0213UvbUlNLlodDrm00RJybgKGW8kiR5VQSKl5DjQ40ILR6XjH/uK8NTtXTjjFk6PD012brkfMimT7jUBYDDvuOReSusdqGp2YtYXJ2j3VohnhiVzeh5Otxfgic8U0QQ3g0VAgwVjuWIwV6Is88k4NdtcrLZuBNo1wXg8HmpyAYgtMo/H08YVNwdOXjZTwXYDz4ovIULZwj7hPh6lkbUpqCgTM5PN+OTZczPjkFdqplYoSydkITczjppcAH7vJxDgFiGky33QxRf55MODV+cyiQix4SpKdLDJ7kGYUoLPjlxCblZH6BRizP/qFKtPshBU8L2T7UrFIoQpuOU9fH7wSsWQEjRR6tZnMSglmnXfC1vGHmhdsX74aB+Y7R6UN9gw4/YkOL1+KnGO9LjoiYqAgCHTkhKjwfpDxYxYGp0ZxryW6ZUGr6KdHkJOZerARF6Zk97GcDQ7vAwWGT3RD+AvW0B6Tqv3FVH0Z7oETbhSxukJr5/cF+/c3wNhSglOlNQRpRMsLoQpJShrsGF873jOkgELviFYWDUWF6QiIZ4dnoImhxexYUqG/VwF6BZtJ56rQirGG98x75X0hhUSbsmWTnoVlozvjk56FebRBEPJpNf0WC3CVNyeEcmCpI8Z37NIjFKzpHfIY4OSCIalWCiktrLp7ZKCmxvzytC3Mzv+ciXZGL6kzHDF75P8ai/aNcFkZGTghRdewJgxYwAA27dvR0ZGxg017L8B1TTJhhKeErNOrxcdwxTwBfxYOr47imptVEJflygVnB4vrE4fp5RFs8MLqdjHWoVyybOT18Xq5BjXKx5Ojw+pBg1F6yXbvJr4QrhSit4JYWh2evHs8FS8v/8ilV/x12EpeOvH1uS4V8dkYNVPFwC0ylw0OTzonRCGs5VNMDu8cLp9aBYIoJZKoFdJUGcl8hxqLS7UWN2UjQ3/z957x1dR5f//r9v7vemFhCSkhwRCCQFZRSkiuhQXEMsuCOrycVcWbIhiQdCPCmJDdAVREXWtsLiiIivgAtKkhRYIIRCSkJ7c3svvj7lnMuVMQMHd/X5+nMeDx4ObO3fmzJkzM+/3eb/ez7fTj1kjctHp9uH5iX1wpq1rzHKTTFArZNCqGSUbVzF3oViJJ4rdaHP6LlgyoMXeZeH5gxEEQmHIZUCcUcM75sLxxawlrJQD828qQiQMBMJhzB1diKoWB5vn0GD1sdYtLUdFaOWTB7HQ8yTXOCNWx+TzcJAuWpUciyaUoK7dBac/hP8ZloNFG47zxoh4XBfyRGvbPfjpbCcPxDkkJ5GH5eH+rsPth9cfgkMuQ69EEzpcfkQQwR/e2ctuJywZQH7b7vJDBiBGr4JSbkCrww+NiilWR8Z78aQ+9P4iAo8/JBm3sHv8WDShhKe8XDKpL3OeUeTLQ6PyUN3GJNsa1QrkJhvh9ofQYvdS7xlvIIhXb+2HOIMK7+04A5kMeHvbadEz4NmbS6BTydFo94rm6/O/68O8DMNAKBzh3avk3IhU+7ExhfAFQ9h1uo2XxlDVTI83EmwMDeP08pR+MIQd+He0i3rBLFy4EB999BE++OADRCIRDBo0CHfcccev3bf/+saFBbY4/Nh9ulUEHRyam4hlm6sxuncCRhf34GFEnr25BG0OH3rG6akoC41SDp2KDywsSGZyG4R4dq1S3u1Di1s8aeXUgXD5Qqhssosgj2SyEtwHF1exYGwxsJepA65T8ZPjgqEQ5t/UG/5gCP5QhPe7hQJ57n3X5bJlf8kyx7aTLahqcWLOyDz4AkFenQ4heuW9H88iLUaDaVf1YvtQnGrGCxsru42VNNg8WL6lGnNG5mL9oQa2ZABtW41KwRsHbpB3+R0DYHX7YVQr4QuF8Oe/HeQtI70RpSlzEzi7uxYtDh+STBp4A2HcPyoPBrUSWrUctw3KYBVSpI+0azz3hgK8N70M5zo8MGtVkMsimPNllzz+lSn9cKrFCV8wzHows0bkoijVfMFYidArjUTEWB7y3Zk2Fw8HtGb3aQzJSaRuK/xc2WjniVTW7KqFWinDQ9cXsNdYCs4ZgQxn21z066iU47G/H0NZpgXv3DkIHS4/eli0qGlziUQqBF/D7cOUsnR2rnC9wAn90ngJqDJEcLjBDvUBPlb/XIcbk97aJZo7Jq0K9Z1unmFAo1gUJJuYeyIU5t0zJI1hXGmaZIwTkC4qdvKktdtn2+VqFyVTdrvd0Gg0LGQyFArB7/f/n0q+/CUy5bNtTuyq6cDCr+h5DwRH4vKHJGWf79xZBo8/hFkfH+R9V5ZpwZxRBeh0+ZFgVOPlf55kcwyEuRgPXp+PrHgDOl1+2Dx+UXCbezPMvaEAPWP18AWZ2NCpFifCEcCsUSDWoGGzkLm4D25/l0wuRXWLg/rd0smlMOmUVGksWdKTklmT+ipfVTRg8aS+2FHdzr5MhcHksX3TUJDCXzaaN4bJ4H5tMx3Xz715yUvjtc1dsEshK8sfDKLJ4Udhigmrf6wRSaUfji4zCfu3+3QrG6ilnSfBqWhVcsy/sRC5yUZYXQGcbHbwgrqZ8To8NLoQVc0O6FRyxOlVWLihUhJlMmt4LpZuqsJ9w3Oxp6ZVVIFzcHYi3tlRgyWT+0Itl8PmCSAjXodDdTbeOBGIpscfQlaCAa9+X8XjrpGXp3B8595QgPRYPY402NgxWjShBJWNdpi0KjwT9aK4OTJCo4H7YKVJ9GN1KsQZ1WyCYma8Dk+NLcapZid6JRpQ2+7iIWfmjMxDXpIR96zZz9tvUYoJD3eDr+Fuu+5AfbdziWz73vRBUcm+EskmPVocXqgU/GUvsu1bfxgIjVLOE+/Q5gc5zi1l6dT7jfTv3mHZaHf7WS+/T7oFIwqSqZn9bPJ3hxvpcQZRIvYvaZcsU54+fTree+89GAyMSsTr9eLuu+/GJ598ckkd+3+9Ndq8+HgPk50dCoWRbNHg1Sn9YPMG0MOiQ02rE8//k8lSzoyjB6zrOz1osPKTsIj0lKDsybLHn4eroJTLsfS7LkAhwYjM/uQg+2DlBrdnj8hDzzgmC9mgViBOr2a3FVrYL07uy+JVuLgPbn9liKAwmf4dWQ6ifUeCmVJCgyablxUWVLc4sXxLNZ6dUCyZKChEkLj8IazdX89TpM0azmSdGzVKPPt1l3fTaPNiza5avHxLKYLhCKxuPwu7jNGpAEQw59MuD4CWENlgdUMOiPq3YCzjrY0rTaOeZ3o0I1wuY5aDHlt3RJREiuh+uQHfZyaU4MFRebDo6LJvZVQRkGxWU6XfyRY1BmT0R7vTj0f+cZg1Ij7fx4yZRilHXpKRhWhqVXI8PLqA9aIiEWD36Ta8MLEv2l1+mLRKdq6QeTXnk4O8MXJ4Anjh25MMgXhKP4TCYajkckQQ4Xm/GoVcdD7khSIc34dHF+Dd6WVweIOwe4K4j+MpE9wLiY+t2VWLR28s4O23MMUouSxKozSTufLSLQwPb0DPGDy+/qhoKetchxtv/lCN+4bn4a7V+9nxpR1nX20ntBKJlv17xuCTmYORYNDAGwwhO8EAxQVk4d5gmHcfvzylH2hNmPxNPCGSMP1rtIvG9ZOXCwAYDAZ4PJ5ufvH/j5Zs1rK4kmW398dj67qSozLi9PjxdAtWTB2ITldAElehUytFwXdhrRNvIIynvjzKBp6rWpysfv++4blYvpV5idGC2yRovnwLkwD3arSmeUasDudtHnbd1xsIY+mmk1g0oQRNVq+kxNWsU0GtkLMoEK5VHwpDspIjqRUiJTRINGpwzzXZbNAcALKTjCJp97Itp/Du9EFod/pF++l0+3lr8MTKK0wxiRA0nW4/ghEGdPnI2i70hxDOKCV+IJUoacHu7uqg1HZ4eMmSXGk3V1QhDL4/+eVR9pi0/fZNZ7AnuUkm0Zgt3HAMq2eUwxcM4Y0f+GINMmb3Dc/FA4Jg8dJNJ0VS6euKUmAReKn3Dc9ll3rIbz/ddw6LJ5WysurFGyvx0OhCHDlvF3lgmfE6vDCpD5qsXtbjikRAlaIv3XQSSyaXItmkwXPfVPIMind3nmExOGRcUsw6nrQbkCHBSBfcSFGaO91+xBrU6Bmjg05Nxxnp1EqM7ZvGxnkA8Vwn6KA0i04SoWPSqph4n1yG3qkWlKTFoKLOSt02K8GAiQPS2VgoGSOp2jDHGm3sy4Vs+8T6o8hLMqK0Zyx+jXZRry2dTodjx46xn48ePQqt9r8PDf3vbiSAplXJIUMEt5Zl4J0dNVi+pRo2lxcjo8ld9396CHO/qMCi8cXQqpghJ5blqm2n2eA7+c7jpwMCW+w+LNtcxduPsO4F7Xcd0QB6nF7F9nHeuiNYsa0GU4dkItWi5dVnn7fuCB5dd1jUX1IzfsbqnzClLAMbDjdg+ZZqrNpeg3uH5WJ7VYvoXIg1u+4AU6+d9v2ckXl4fP1RrNpeg1vLMuDwMgrFNglpd7vTh68q6nj9I9JY4X7XHaiHLSrzFPbJ5fWjw8X/zh/kyzqlxvRsm0sy2C2T0c+TOw7cbYWfEwwqSS/wuW8qMWdknmgePbbuMJZvqUZdh5v622a7F3ZPgDdHl246gQVjmTGTOk+uVJrIaCvqbN2OEVf+TObHrWUZaLV7Jbcl827uFxW4pSwDxxusktL+c+0u1LY7eedCjmHRKnjj3en2d83Ra3Px9rbT6HR6RHOFi6/hzh3u+ErdF6SCqPDcuHOALMuu3Mbce09+eZQde25/D9R24Pa39+CmZdux8VgTwuEIW69GuG2T1S0tUnF4IWzEkBRuy/XGLne7KA9m/vz5mDNnDluSuLW19UoRL/ADaG1OHw+CF2PU4f7Pu5IEa9s9eOOHaraaoEYhR6PNg2EFzJgeqmtn68tLWTdJZk0UPXMO704fhLoON1IEgU/a75ItjDEglZxIs5pJf9fMKEeby4cTTQ7euvNTgsTKt7ZVszGDCCJ4cFQeUmP0AMATJZDg9uoZ5eh0+VHZZOftl3goALOERLMAlXI57rkmF5uONojqtRORRTAURq9EA+ZZCtEjRou5X1TwArVbTjTh4RuK0OnyQy4Du9zTM1bfbVCa9CHZpBFVeQS6EiInDUyHXM7st1eCARadCo+uOywSZxArmey3MMUs6T2Gwl1LNjOHZSMvyYQIwJP9SkllY/SM58mdo7XtHry1jZmTrQ4f9Xe/yUlAOMJIhJ/68ijuvjobySYNNRGQ/JYmf1625RSLwSnLtLAIJeG2sXo1Gqwe3DIoA/ESEuE+6RYo5TI89Y99omMIq34SCTkAvPUvRs5vNmix6scTPEHOhoo6/H5IL4wsTIVJp0SH04f7R+ZBp1HyxpfcF8I0ARKTy4zXsdtyUwpidSoeOoiMvbBWDJGpcz2ReINGVJn0033nsOy2/vAGwzwEDcAYW7TaMIkSnluCUSPa9nK1i3rB9O3bF99++y3OnDmDSCSC7OzsK6iYaCNU01MtfLlgq0NsLdS2MyTld7bX4PbBmVj83Ul4A12BT7K0UZZpEeEoCEIFYCoYHqjtxNJNVbxtaYmAC8cXwxcIRvtEtwhJLIfW30a7B15/GMs2V1N/B0jUVR9bjE63D1/sqxfFMO67Lhfz1lZg0oCe1P2ebmEklB/uPsMm7AkD91y4Jdnv7BF5aLV7seNUK8b0ScVdq5kb+rXbSqn4D1qCYV6Cjic1/aqiQRJgmRmv4yUUSqF4Ol0+PP9tJTWZ9o0fqnmiA28gjPk3FkjKyQHmwbVsczUevD4fwXCYJ/ulSWUXji/G6h1nMJYSF6pt9+C81YMXvzspmjuLxhfjx1PNWL71DJ78bZEoHkKC0MLESin5877aTqzazlTGBJiaLtxthQo5OpalGE99eRSTBvSkHoNb9XPR+GK8vOkELwFZo5SjzeHjVQElx+WiZBjwaAgNNo9IVl3b7kFVs4Mn3z7Z7IieG5M0SuJYfxmRh5c2ncDdv+lFHXsSmxReY3I+LQ4vyrPiMW9MkUhq3CctBuFwhCflJnGVjFg9hE0uBzWV4lcKvwC4SBUZABw4cAANDQ0IhULs326++eZfq1//9vZLYZesKqPTgwiAlf9ickU+nTmECnkk8QPu2v19w3N5ahkAqGpiwJjcJLc7h2azlFkSK8lKMLBo+oxYHQKhEOIMGjh8QejVjFV/19W5qG1zISvBQO3Tm3cMgE6t4EH+AEYR9fANRWh1+FDd4hBZrBcCNy6ZzFRxZFVN/iCLXqlt9+D12/tTlVbvTh+EunY3UmK0OHS2DYNykhAKR1jRAxkz2jHJ+DZa3ayarjwrFh9Gkzs9viBSYrr6wP3tiqkDoVMpsGp7NfKSY1hr8VSzFdN/kwMvp8ogaVy1V3dKwV01HazCjFise2pa8cgNRfAEQzzPctaIXJ5VStR0Uv0VgifLMi145IYiNDt8sOiUeG/HGWytapMc7xVTB6LJ6oVMFkGqRY8Olx8xBhW+rqhnwKNWLzLi9VTV04qpA7H3TCda7S5MHJjJlpfoDobJvcYZ8Xq8tOkEBmcnsnFB7jwjIM8WhxfJpi6gqRSUlKwQpJi1LN1Z2F+1gq/g6m4uBUNhEdySe82NGgUiESBOr0ad1YPdp1sxb0wR2l1+pFp0bImONqePOiYf3j0YwXAYOpUCsz85KLrG788oR6JJw+a9COu/1LQ6cdOy7aL9fkOJwdS0OruAtpxUiiWT+11SqeVLVpHNnTsXdXV1KCwsZKXKMpns/9QL5pc0mipjwbhiYE8tmqwuqvW1+sfTyE+J4U0Ik1ZBVUudbLTjuW9Pstbic99UotPtxyIOIn72yFw2wezRGwsQDgPPfsO3oAmWQ2hxc+MfD12fy8PKEA9BaOUTdAUB9QHSCXtn21ywaBUYUZjCWvWkvwDQIECBcNe7udDMx9YdxgOjCi4qNnKgthMf7TnHU9NJeTtCmORPZxkLe8HYYny8t5ZXtOuu1T+JoJ8AY4WebXNh+ZZqZEc5YsI+HWmwswq581Y3z/I9VG+FQoARWbu/XmTJ03A2q3ecgcMXEF3TKWUZWPJdJTJi9RiSk4BdZzoAgI1bCb2o1zdXYXy/NKjkCnYZhxyDvIglyxg0O7HuANNfMldG904QHYdrnXsDYeyuaWfluMSTo6npRhSmsPvlzh2ppOHDdZ147tuTWHpLX97LhRy3w+mHPxC8KPRRu9OH7VXNPIAlM5fy8MgXFVSP9t5hubz8GoJtaXf5qP0FIhiSnYBwOCLyUuaMzMP9nx5Cp9vP7kf4Imi20+MqJNGS27LiDbjr6hyRJ/Rr1oe5qBfM0aNH8c0330DGjUheaVRVxsKvjuGdO8sQDEfw0e4zLCIl0aTB+gPnkJccIypMlRajpxa5WjOjHG/9YQCON9qp8Y99tTaeUoW2H6JqemNrNW/9WCWX8eIfiWYdz4oe2zeNfUCQfb22+RTevGMADBolXvyukrWwpdAxvRIMMAsUR9z+0sCCQowI6X+cQUU9hvBzRrxBpKajnYsQJkniIeSYJCdCGCOQUvMAkIyd9IjRscdcMrmU912KRQeLAHtDwwGtPSDGAz00ujCat8GPyRHjY3B2Ik81RnA1wtjDtKEMGJMGYyVjJKn+M2tFYzS2tCezLEQBWJLfEVWhNxAWxfOk4oLcPjTavCzSJc2iw7lODy+JNC1GJ3ktfEGmTAQZB2Eck2ybZNJiyqAsvPDtcd62ZD7T5plQcXihOMqYEqYWEzeeW9vuwsE6K++el1KGSaFgaDEY7jHONHWgV0rcr14J86JeMHl5eWhtbWWD/Fca04SqjL5pZtwzLAcdLj+SzVrcc002dlR3IBwBfIEQtld3YuIAAxqsbjw5tjebfFbbTlcjnbd5IJfL8Pm+eh5GZtvJFhSmMJwqnYrR/z+/8US3qibSyPpxVryeF/9oFpTBlfIQWp0+tDo8uKUsg+ftCC3W//1dHyjlMqgUMuQnGVlZNdfyXLufn8TGtVBJI7gahzeIZ2/ugyfWM3LYryoa8MyEEl552gdG5TOgSUHfL6SQolnYBP/CtW5pVvO8MYVINKqweGIfqBUyPDy6AEs3neRZqVyA5bloHI30V6OUw+0L4bExhbxkuV4JBhbpopQDUwb2ZH8rkzH4GpK30WD1wesPoa7TA4NGiYxYPf4yMh82dwCDJ5SwS2SzRuTyYg+kMR41PU8rI1oh0qhRiPBAC8cXIxIOoyiVjyTiov+JB0ikvbRYQ6xezc5nMs4yGZMjI/TsuOPf6fZDp1Lg1c2n0On24+HRBSxCKcmkpsYbrB4/3thSjTsGZ7IebWa8ToRxee53feD0BaBWyOEPRnCyycGiZUifLnaedRdHoXkPvmAYYUHgIlavRqvDJyImS6FgpLwSEjP2tdUhOzGDus3lbBf1guns7MRvf/tb9O3blxfcf+utt361jv2/0HpYuqykvmlm3F6eyVuGWTS+hMVPzL+xQOROvzKlH1z+oCTUsabNBb1aIcokXjCumF2T16qYTOhlt/WHL0hHeNBwHwY1/5g6Fb0Pws+tDh+Wb63GszcX473pg9Di8CFWr4LHH8DMYdmwaFXolWDAwg3Hupa5osuGhxvsHBUZEysxcxL28iklh7m4msx4Hf76h4Fw+YJINmtQ0+LkJezpVHK20NrFnEtukgnL7+iPqmaHyMImpXe5fRJazS5/EHEGFWas7lpWen5iHxbWSANYFkaR+9ooBmjOJwfx6BimtAA3WW7h+GJ8tPdcV/XLa3N5CbRzRuYh3qAWjdHo3gmi2icLxxd3Ow6RiPT1J3gdrYohD7w6pR+ONtohlzFG06vfMw93bnY7V13HHbPCFBMMGiWv/AHp/1wBRTg3yQiNUk717FZMHYiKOhuKUkw40+ZkFHsyIN6oRqKRedE5vEFsPdEgQjfdfXUOWp1+BMNhdu6kmNRQK2S8uSSTAa99fwo2b4BHzpgzMpfXp4uZZwkGjSSyhVuuXEhF5r6Ip12VycZPuUtvF9rvf7pdVJB/79691L+Xl5df9g79p9ovCfJX1HVi5+l2vLaZWfqQqqg3++ODeHh0vqiGuFYlx0u3lKKh083DtHDjHb8fnIHP99dRcSnc5Z1Zw3Px0Z5zuPvqXjwLmiDiycN+8aQ+OG9lrKAEoxrHztvh8ocQb1Ah3qDGI2u7MBz3XZfHK55E0DeLNzLZ2c9O6IPzVg9SY7T4cPcZ5CXHYGhOHMtMIi0zXodnJpRg79lOKGRAvEGNoh5mtDn8qOtwscF4nUqORKMa9VYvwhGgMMWEF78TB7ff/P0AqBRyXtCffPf21DI0RgPNr33fhdcRFnsi4zL9qix4g2ERCiQ/2Yi7398vUnhxb/zHbiqiXvOlk0sx6+ODorjVg9fno0+aGbe/vZeH4vnr7wfwkhzJfi4koiAvWy7K5jc58WxFS+6279xZhiP1NvSI0bGJpWQ+aBQKBMJh+IIR1kPk9p374iVIGm7SYJ3Vw0MSlWVaMPWqLJzmgF2zEw3QKOQ41+FGSowO1VFEkdQ1fmVKP7yz4zSmXdUL1a3OLkBskhFmrRJWdxBKhYwV1QjHfv6NBdCqFGhzdXmF8QY1QuEIOj0B7Kxu5VVvpQkY3rxjAFz+EG/ZmTsfqDEYAcZpzsg8DOkVh2aHD6kWnSSaRSpYf/fV2VDIQUXF0AL5F9suRwXfi9nXRXkw/5deJJezNVg9WLOrFndfnY1wOEJ1j8kyhjcYpn5f2cTIG5dM6sNa8oXJJvxv1NXXqegCAG4VRW+AQYUkGtWIN6hZS8ygZrD2E/qlsbwxEkuh8brm31iI9+8ahCabDylmDRqsbhHSwxMJsbJkEhDmPsAzYvmVHMm2MzkW9QOj8nGwthPv7awVBePvvZYPt6QF4w+cs6JnLL1i5E+1HV3B4wklmDlMxfC8lDIsnVwKlz8Ig1qJcDiEWwb2RKxBg8Ub+Zj3Nbtqcf+oPAB8rIzLH8LZdjfbHyGuhvQhFI7gwevzRRa2RiGHOxDCrBG5yIrvEgTYPfR6JReqfrm/tkuUQCojcuv4cMesvtODVzefwqIJxXjjjv5oc/rRM06Hug4P5nGMir/+fiDc/iBMWiXmrT0iQqJ4g3TgJhdJ1CfNgvNWD+86zr2hAKu2n4FaKcOs4XkXvMbBcBije6fAHwqLALEL/nGMJwJBVJBBxmzWiFwkW7Rosvl4v31gVD7ykw0oUiuQatbxhCe08T1QZxUt0wnRMUAXkkivVuLtbadFc8kTCLFzUgrNIhWs75tmhloCK9NsFwfy/9taty+Y22+/HR9//DH69+/PC/BHIhHIZDIcOHDgV+/gf3NLNGpY1Mbrt/cXuceZ8TqkWLRMdcZkEy8JC+AHlh9Ze4SVi5p1SqiVzHjnJhvZBDduxcUHo8FMoKv6ZUacHqFwCFfnJrBrtbtONSM3OQYeX5Ctwe4N0LEyz317AiumDsSpFidi9Sq8trla7D3cMQDzbyrC0k0n2N9y687oNUpetUMiseUe55XvmQDrxAHp+OQnMSmau61UMF4qoZAXPP7yKFbPKIcnEMYj646KtmWsQxkVMcPFhhCsjFal4HkStERLrUoOmVwGXzCMv3xyUPTdmhmMscYlA0vt50LVL7miBFqQnDtmOrUyOiaMgGHe2iP42x8H402OCAAAnv6KKZ6nUsipSJRIhEmO5AasAWYpcNGEEizfwtwLq3ee4X2/ZtdZzL+JsXJpghZhfy06NQAZrx4MALy+5RTrvZNzJ6sEWpUc8UYNAAdi9Gq8/M8q9rdGjQLhSATeQAQGjRxvbetaTZBCuuQlmWDRKUX3bafbj1AErBqQzCWZjI9xAphnALfa7OtbTrFoFm4lSr2af98AzH2dl2xCp0uMRdKq5FAp6Aks3VW4/He3bl8wH3/8MQDg4MGD3W32/9sWAdig49vbTmPBuGLWPSbSWOFa+JsSckxikQqlmwTvIZSscqv6LRpfjHd2VONIgzO6nn2E1weaPLc7q1iY+MXzHuqsIiuZu6+tlU2ihELafs5bPSy6hpZQyN2WFoxPNIqDuLREtXanT1KGqpBDdN3IdVqzs4Z3TBKo5waZ399ZQ5X97qhqQpKFLlnecboNy7dU8xI41+2vE0naF44vxufRCpzcbWnn6g2ERR4td8wIkoh8R6qsCucW2W9lox3fH2+intsbP1Tj9vIM6u8cHoL48VK/b7C6JSsscvs7Z2Qe5q09jBlDM6n7kcv5v/X4g+x3LdFqqFxQKs1b584zrniAti3tvuVWMeWWFxBKmoVVKmePyEOHyyeKuZRlWnDf8DxezRqSMHmug16Z1OkTF328UIXLf3e76ETL/+vtl6xJchOXwuEI4g1KqJVKNthGS2okFQIbrB58LpG4CDCWz5JJjKSVu58LJTVyv5PallhbtEQ1oSVJPgvX3LlJg9xEQCEsUmq/SyaXiqCJUtu+O30QrG4/TjQ5eGNWlmnBX0bmoylav33uFxUij2vNjHJ4giG8vrmKh7DfXtWEiQMz0WTzItWigUouR4PNi2SzRpRoyU2CJedNEja/rqjH1fkpPNnvXVfnotPlx4kmu2RyKgBMGZiKiQMz4QuE8PqWKhFin4zv0Jx4rP7xNMaW9oQMzJKZMPZAyh0AzAtpyaRStDh8kAFYuY2/7ZoZ5Wiye5EikRBJPILRvRNwz9W5aI4mOe6qbka/rARRoiJ3vpJyCFJz3+4JUpM9CfKnwdZ1X0glhgqx+u/cWYYOV4Ads9kfH8T7dw1i51Z390xVM7PMtft0Kx4cXShK6OXOwZ2n29n5QBIte1g0yIgzoNMdQLJZg8Pn2pAUY4LHH0RWPD2x+e2pZdCq5HiYM19pydZfVTTgvekMUunBzw+x35E+vHRLP5Rl8atcXmzi5X9VDOZKo7eMWD1uEyjHiGV051VZVEutstGB93ee7Va6SeIWd763F3NHX1yCoccf/Fny3M/30evLCz0AhRzUQDfB0hOrjlh5Qlgkdz/cc213eOEP0qGO3G0XRnEfDVYfr167ViXHpAEZbCLgqjsHUq1duy+ADpcPt3BQNrQk0gXjivHxnlrYvAGRB0bOzaJV4fbB4utNpLEkNiGVnMod3+H5CRiYlYDp7+3F/aPyeMmoXCt5+ZZqFCQb0a9nvKhee6vTzya9knEg16bT7cNz31Ri6pBMVLU4eeM5N5p4S6s37w0wCbKpFi369YzHNM65zB7BwE7vuZqPPeHO1+5iGpWNDny8t1bkGREPa1hBEi8JVUp2z5V6LxhXjMXfnkBVi5OFTgJ8LJLUfcDFtCwaX4KXN53AlEGZ1G0bOj08z/75byqRaFTj9vJM3LOGn5z69/112FrVhkfHFFL3tedsh2gVQCrZusPlQzgiLgsxe0QeIhD7BlKxnP9UvObKC+YS2rlOtyjRctmWUyx6hbZu2iNGh0abFxuPNmLp5FJEAJh1Sry+uYq1dLmJa0lmrSimQdsviRnQvhN+zk1iYIwAA2NMj9Ujwaimwhhzk0yYf1NRtwmc3kAYb/5QjRcnlSIkUXucrEMT6e6iCSWS6Hnutm/+UM2uuRPIY+9UM+IManyxrxYPjS6ExxeEWauiJrK9OKkUJo2KZ0nSEi9JbffZHx/EZ/vO4f0Z5Wi0edEjRotV25k+FKaYuo0fSNW4f/OOAVAoZHh9cxWbz0TUXt4APUGW4O5fvbUfks1avL/zjGi/L0bnz5qdNTzLl9Si5wI3s+INiDOoeQmymfEG6vj7gkyMbsuJJh5M9P2dNZg4IB1JguQ+4XlLJWX2iNFRYZFvbavmSbnJ76QqZw7IjMVrt/aDQi7jeWckMXRrVZtI+n/BeN0/mDinTqWgxkJ6JRjwyczBLNKl0eYVqQi9ASY5dfWMcowpSRVBa7mrAKQ0xWM3FaGq2YGiVDMVRPvpzCEAZNS5/ZvceAibVGxSr2YILCzaym+Gv856WQqOddeuvGAuoUlZCwfqrEiRSPSyu31ItWhxY59Utqoe+a7VyazzcmMGnW4fDxMiBQBcs7MGDVYfzysRQgiFFjdps0bkotFKt6Ce/6ZSssa9EHY5LWqN0/AvXG+HYHBONNovqn8kvsCFPK49UMeLLy0YK4Yxzh6RhxPNDmgEKpzuvMBUixYjClN4OQfE0pQaB+JxScV6DtRZUdVkxZSyDPZcuRas0FLn4u65Y+gP1rIPU2+Aqdcer1dhRGGK6LxrWp2sxb1wfDE+++kcrsqJ521LhapGPbkpZenUwmVyhNHp9nWLWpHCuJA4Fg0WSfOqd59upc71fx49D4NOw/u98Fpsr+qKIdEgsDRvXSYD1u2vE0naF00oQd8eFqjVCh7SJRiiK0MP11vx3DcnognITExGKg5EvFQpb9LtD0GpkFHndiDE3x4Ai/anbftfW3DsSqM3KUxDJAJY9BqeAoZYaosmlEh6BO/PKMfZdhd6JRhYeWWsXsPbtrbdgzc52P8YvQrbTzbhkTFFaLH70CNGgzV3Md8lmrQ4WtfGAwA+sraCioy3C7AtxMLursY9id5xLVgh/qUg2YT3ORY2F6Nekh6Dlzd1YdN7xunxGMWLIigW8tkXDIu8kHZ3gHdcbpJjYQrd66N5gVKo+e6KiBGPSwqZMzgrDtfkJvC8KO5+hZY6rQ9ctRTZb0GyCfEUxAvB1pPPC/5xjFogbXB2Ik9FFokwSPtnb+4DjULOLo8J52g4EsGj6w7zCpdx+09D3QgTToUKuSHZ8chLMkGjlLOJqgQdxO3fGz8w99BPZzu7vRbc8gLEECKlE8w6FXWeRSJA77QY9uVCzvupL4+iLDMW2YlGUYkOWh8KUkwACNr/FBXXT64TQQd5JZKkk80M8qU7zAy3dYek+U8UHLvygrmEJsQ0ZMbrMG9MEapbnJBHcR5c+SsAHKm3SebEtDl9qOv0ICuhC71CW4eubffA6vYjEgFkAEJhIBRmVmT9IeDdHdXYdLwNWpUcq+4cwMjKAUAGPD2+N1zeMFy+IAxaJYAI6js9CIUjuGtoLzy/kZEJp5jUrOdEswAJlhwQ41QuBGpcMLYYL3xTiT9clYUGq4/FcFh0Kl4ftComdyHBqMKrt/ZDnEEFq8uH5zdWibwJ4XHJb1fvPItEoxpPjyvG0191Ifi5ah9iua/adhrXFSYhVq/moXnW7q9nLewFY4vQaPexyXvJZg0vBiO0Hh8eXQCXLwirh54nRbPcpTwhgq/Rqpik1+e+qcRDEpav0xvE4ol9oNcwuRkef5CKAyJIF4DxnCYOSEen2w+lRJneVocP4UhE9Dvhed87LJfntZJYlFbFlH9evvUUey4kzuYPRnDvdTlIMetg0gQRo1PColWJ7iGrO0Cdk0sm94VaLkPPGB0QASzaLuoIKal9/6g8LPzqOKYPzeJhbxZNKMEbW09Jlrqu5ch+e5i06HD50ekOUL0FK0feTby1WB093ni2zcXOMeG+uMiXi8XMZMUbJLfddLyJ2ocmmxelPUW7uiztygvmEptaySAm+qZb0Gz3iWiowkzogmSTpLVyMrpssGq7HCumDsB70wchILEOfbzRjmWbq9mA9QyebLkEN5akQqmQ4XynX0SCFT5YvznciKoWJx68Ph+v3dYfRxpsyEk24aXvuryLGL0aj9xQAJNWhbpONyw6BV66pR+aosor4nHRIIRrD5zDe9MH4VyHGxlxDJ79cIMdFp1KtGzw4PX5bMIpwanctZofRC1MNrLjILSal04uxYlmB+TRGjcARGgQuQyw6BR4dUo/2LwBGNRKQBbB7FF5iNOroVMpRJn9eUlG9M+woM0R4CXvPTOhBIXJRjTavNQ+xBvVaHF4ESdR7GlIdjxyEo1INKnx5G+LEGtQI86goWZuk5wquQzoEaPD0+OLEaunQ0AjEWDeuiPsNU4ya1DX6eFtq1N1oViEyZNCJArZr1mnxF6B9yA878JkE1ZzvFatUg6FDLh/ZB4abB4EQyG8fEs/VLc6ecBNl9cPpy/Ek+pyMUOkD6lmLR69sRBJJjVW/GEAfMEI0mK0qGxy4JH1h7sMG0pWfYxezZ47mQ9xehXMWgUm9EuT9EIP1llZSsGUskw89Y+juOeabKrX/FC0+Bj3WniDIcl4F3cMP7irHKFIRIR8uVgcTHfomFQLHQCaYvn1qhNfkSlH2y+VKc9YvRdj+6bh2vwEquRz5rBsNrdlzsg8lGXF4nynB1q1AoFghPUk3L4AXvm+WiRpzU/UIRCR8dZNuS8uKQnme9MHQSGT8ZY5upNrzuagTRZvPIkXfleMTk+Qh8F4dEwRXP4gzra78VVFA5ZOLmXqXpi1OFhnxQscz4Ornnrjjv5QKxVod/qRYFRDLo+gw8mc958+FMuUV04diEarF6kxWiyLkoG536+YOhBPfXmU+gAh48INqBo0Sjbexd2PUA79ypR+MOuUItSNVsXU5fAFQ6LaIKS/0979iX1BvvfjWTTavOibZsbMYTmIgHnIhiJhHG908PApL/+ziuWNPTqmCKdanIjTK6FWKUUP2o8FD9qVUwfCrFXhRJOD9c6ktn17WhmONdiQbNHisXVMnhQXp5Jm0cHm8bPYHqNGAblMhpf/yYdbbjzSiBPNTpEHMP/GQti9QcnfchVTWpUc795ZhjtW7WXjAK9vOYVFE0qosnUueueZCSXIiNfhbKsbCSY1jBolWhx+JJs1vPtPaq6/fnt/+AJhvLjpBPsCHJQVi4+i9YIi4TBkchmLsiGImb/+q4aJAd7eHy9Ff2vSKmDSqlhoLenfpz/VskXOiGd026AMAOAZLs9MKEGnywe7L8RicMYUpUAdDchf7vZrxWCuyJR/pdbu8rHBt8IUI9X9TLPoWEUUwUb8cKIFtw/O5D0cF4wrZkF95LcyGWDzhWDWqlhri4uRAaQD1g1WjwgxIbVtINi1JhunV2PWiFzIZHIWg0NupAcEIL6TzQ48sf4Y3p9RhlgD00elXI6SHmYEwxHcPyoP6bFaNHT68NQ/DnI8rGJ8X9koqotD+rCPk3BKC27bPQEsmlCCl747yYv1PCeQCxNrXEo2y6VMewNhHGu0SyJofjzdhnCEPn42T4D1LDTRKDOBnz7Mg58W48tDDeyy0bwxhbhtUAZkMiBGp2bHNzNeh8duLOShbdy+AFqdft5x957txO7Trbjnmmx22/RYHRZ/e4KXTe4NMImtr24+hYdHF7AeYrJZiyY7g1MhQWguUPOxMYVYNa0M7S4/Ek0avP2v09ha1cZQCJRdHkDfNAvOtLlEv112a3/YvAEeXof0p8PdNWZqpRx/vLoXbG46MgcyYPGkPojTq+EJhnDnu13eOpfZdzFzvaLeBrNGLGhZOJ4RoviDEcz4TZYIMcPuFxFR4vPLU/ohFA7jVIsTgVAIc8cUwuoKIMWiRVGyGWWZsahqdmDhV8fZ+apVyhEIhfDy9119+N/f9fl10flKOW4uTUNekhENnW6kxepRnGr5v6cis9vteOKJJ1BVVQWZTIbnnnsOvXr1wgMPPICGhgakpaXh1VdfhcViAQCsWLECX3zxBeRyOZ544glcc801AJg6NY899hi8Xi+uvfZaPP7445DJZPD7/XjkkUdw7NgxxMTE4JVXXkF6evplPw+1Qs4G1BJM9CUQnYYZYpmMWU7LTTKhX3oMrzKiN8DIZLnJcrtPtyI/2YR4g4pnUc8akQu1Uob7hud2K1tONGmglaA007YlVRRJf/WaLlxNd/VqXr+9H3RqJZ6K8qHuG87EW4h1mBGnxxs/8APWT/2j61xp/eHKR7nBbeIRQMbQfwtTDChIMcHjC8LCwetcjGw2M16H/GQTD7dTkGwS1Wbh9kkmo4+fWacSYUMKUkwiCauw7snijSdYGfB5m4dF3o/tm4b7P72wxxWJANOGZvO2ff125qFO5gc5N4KKWbrpJBt8jzGoMS/qzdDQQc9vPIHVM8pxqsUJuUwGlRJYdnt/hMOMFD0ciaDdHYBGJWe9GfLbd3eewfMT+8IXCsOkYYjghHT9VUUDEo0auGIY+fNXFXX4/ZBeUCvk1PE1aZWY9beD0STew7zjvLaZEUpISa5pS4eplPlMJM4AWM+LfPfK91Xs2CcYNZjL6UNtuwcPfnYI704fhL98fAhalRyfzhyCG0oS2OOS/BMukui+4blYtKGSd5zH/34EmXF6BELMkmUoDKaS5wVwLz8HDaNUylHaMxZqZxOKeqZRt7mc7Vesxizd/vd//xfXXHMNNm7ciC+//BI5OTlYuXIlrrrqKmzatAlXXXUVVq5cCQCorq7G119/ja+//hqrVq3CwoUL2bLNTz/9NBYtWoRNmzbh7Nmz2LZtGwDg888/h9lsxj//+U9Mnz4dS5cu/VXOwxcM49ayDLyzowYrfziNheOLoVUxQ0os1jU7a6JxlRr8+bpcrNlZgwN1Vqp1RTAtq7bXYEpZBl7adAL1HR7etttOtuDeYYz7v3xLNZZuOoFFguMuGFeMl747iePnbZg9Io/97quKBiwYy9+WYDlWba/BfdflYkNFHZZvqcb7O5nP7+yokUx423G6DX/5+BCmvbsXt5ZlINWiZRPGSP9mrP6J/Y77WxKo5faPeEbrDtTztg0EQ4xHMJjxCGb97SDmflGBsqwEPPJFBeatO4KZH+zHfdflIjNeJ7Jet51s4Z03F6FDxvu+63KxtbIJH+4+IxpP0idaf0llSW5/FXJIJpwKvabqFgfmrTuCFdtqMHVIJlIt2ouuX7PuQL0IuElktmT8ybwjCYjeAJOouHxLNU42OtjfSh1z39kOdp6NKkrFI19U4P5PD+H+Tw/BGwxj7f567K/t5P2WyKzvWv0T/vLxQby6mQnor91fj1Xba3DvsFycarZj3rojmPtFBUYWpeL1zVVY+S/xPcQdXym4qEzWVR21u7lOxqy7uklS46BTdUnlad83Wru8M7c/BGEjgiDSHykhx7ZTrXjwswp8e7QJv319O25/ew9uWrYdG481ISwsEoMuNMxNyy687X+i/ds9GKfTiZ9++gkvvPACAECtVkOtVmPz5s344IMPAAA333wzpk6dirlz52Lz5s347W9/C7VajZ49eyIzMxOHDx9GWloanE4n+vfvz/5m8+bNuPbaa7FlyxbMmjULAHDDDTdg0aJFLKDzcja5TMZayr3TYkSSzzeiSYL7am0iK+lClvtn0WREty+ER8cUIAJGCSMEQlIT1/7FHDfBpOVJpYlk+O6rs5ERq0NHVO1Cki7fiFZJ3HS8DYOzE1kZcK8EunXI7S+R3Ep5O0LrO0avEgkCMuL1omTPzHgdesbq8efhuahstPOsfGGyJPGM7J6gKDmVnDdZcxdKWN/4oRrP/a4POlwB3pgJ5dqkvyTZc9epZvxuYE+MKUllkxGHZMdDLgM1YU8oz6WNodT8GJqTgLwkExqtbvhDEUwamI7UGHEi7vukaiUnQXJwdiK2VrUhM16HPukWzBqRi8JUE/tbKU84I97Awlq51zVWr4YnEML8m4pg0SlRlmlhY2VCD5Jse//IPNRZPXhrWzUbCCf3Beu9R8J4e2oZrB6mCuy6/bXYWtUGQBouGolAJI8nVUCJ9J9bVdMfCkteGykvtbxXHJbf0R9JJg31typll/GiUsix8WgjD88vl8swuigZn84cgkabF3EGNbtcKpwPNG9SqqLl2XYXKyy60La09muDMf/tL5i6ujrExcXhsccew4kTJ1BcXIzHH38c7e3tbMXMpKQkdHR0AACam5tRWtpVZjY5ORnNzc1QKpVISenSgaekpKC5uZn9TWpqKgBAqVTCZDKhs7MTcXF8bg+3+Xw+VFZW/qxzaQ2YeRYgV7pJmtBilcmkE9FI4lffNDMmDcjggTLnjMzD2v31POQIeejREteIVUcDZS7ccAz3j8pDJAJRHxDFT3CTE1vs3m77S85Np5KjoZPvcZHvuNb3ovHF+LqC8VI63X5olUxVwnuu7kXtrxBX0h2ss8nmxUe7a0XSaO6YvXZrKTVxrc3pw+lWF+86Hqu38hLvOt1+pFp0eHNrNYuVeUQQZ1n942kML0gW9YELsJQaQ5mMDrdcOL4Ybn8AClkYMXo1+3Ity7Tw0DbPTuhNTZAEusoqkHlVlmlhzy1WrxbJfrkJstw4Fg3XzxQ1O4d9tTaedS6F9m+188sANNu8PBSLXM7k7CwYW4w9Z62obffgq4o6UWIuFzQpRPwsGl+MjUcasPF4Kw/NtPt0K3Xbz/adg0mjoo79uXYnHv7iKEb3TqCihP6+v471jP/wzh72u2cmFKNfbADhUAiVDg0eXX+cN76kzAJ3Pkgl9J5p6oCvrY739/MBOlSVti1vG68XJ6uqRH164ebeKDL5EA6JvbBf0v7tL5hgMIjjx4/jySefRGlpKZ599ll2OYzWaCI3mUwm+ffuftNd02g0P1tF5q+zXtACpFWT5FruRSlm6DUKXpW/e4bliNbvyVrzG1urJRH2wuM4fSFUNVmxYupAdLoCiDOosKGiHmtmMFLIGZQEvXenDwIAnteSZBZ7Qks3nRAlqpVlxSEgIcccmBmLxZP6sJUF595QhOGFqahssrMP/k4PP1kyP5kfx4jVq+ENMpawXsNYzaTQFtAVa7gmP0lUG507ZtzYA/f7FVMHIhxx8fq/taoNKiWwekY52pw+INIFj7xveC4+iyYUEm/hq4o6TP9NDqwuP042O1iPi1jqSyaXYkiOg/U0uPk2X1U0YEDPGAzKisX2k0286/bh7jP4/ZBeiDPq8ODnXWoprqcJAD3jDSLkyMINDL7kodGFvPEcnJ3IPkhJnROm8qQZCQY1yywD+HEsWiLogn8cw5oZ5WiMWsIkSC6VuEoSDAHG6k+OSrCBLqmvNxDG1pONeHFSKQPcNDPYHiKdN2mV6HD6cP/IPJzr9KCx04n3Z5Sz8tzd1c3QazWYNDAdEUTw4Kg8WHRq6DRKanxszV3lCIYjeIyTREpwRc9P7Ev15Mi5E2yPcL9PfnkMn84cApNWhVe+3MvzmoknxwXGdpfY3CslTlTmWNPqvOhtua2yshKq2B549IPtvP4+uv74zy5k1p1h/m9/waSkpCAlJYX1SsaMGYOVK1ciPj4eLS0tSEpKQktLC+ttpKSkoKmpif19c3MzkpKSRH9vampiPaCUlBQ0NjYiJSUFwWAQDocDMTExl/1cipJNLGKbZgESvDnQtZ5MLFhiuZ+3erBm91melRySQFCQiSn0CIglRD4Tq65PmlFUPnfR+GKs2lGNkrQ46jFaHcwEb4yuaS/bcqpbT4hrjc1bWwF/MELFcjwqKF513urBku9O8gCWX1U08PZ7IauZjC+xABeNL8aGijpJdRoZM27sgft9ZaOD6l0OL0jFvLUVGFeaxvMSk81qnrdAPARuKQWu5+QNdAEWmXyKDD70cVwxXv2+CjZvgOdpkPG1eXyQoXtloDCZkntdq5qdvO+ECB0uiudEOMxbvuGOi3TteR/mfHII828sYLcVHoNs22Rj9i30qsiYWd1+DM9PwMiiVJ4HS8MOyeUhvLODKZ3w4neV2Fdro4/v2GJ8tKcWwwqSqH0KBMNweIPUlYhGKwO7nH8jHWDZYmcMJCmv2h8KU71mgorh3lO0JNLuEiu5yd7dbStsUqirFsflA2P+218wiYmJSElJQU1NDbKzs7Fr1y7k5OQgJycH69evx8yZM7F+/XqMHDkSADBixAg89NBDmDFjBpqbm3H27Fn07dsXCoUCBoMBhw4dQmlpKdavX4+pU6eyv/n73/+O/v3747vvvsOQIUMue/wFAOptHjZXgWsBpll00GmUPAghsYQeGl2IITld+IxFE0owtm8aPuegNdJi6QlRxEvRqvhAyLUHmP002bwwa1VodXgx/6YiKkbkqagVXdXsoB4jxczIqlNj9KzXksb5P3d9m+yHID3IA4k7DrS4ilYlR1xUkk22LUoxI96oxpKNlexxyjJju7Wahaoscm5GCdAoGbOCVLq36fAGRZiTq3MTMI9jyXN/lynwFriF10gfuZ4Ttw9C78wbYJSEJAYj3M/CDYyHAIkYAfmsU9HPPdGkxclmvrUrFVvLSjCI5gd3XKTGN8GkAcDHDvXraaFuW5oeg8WT+iAz3sAy17hjtmLqQEy/uhfVGxMWV1sxdSDzN879RR3fDd3HQNXKqPqym3OTmjtFPczYebqdfk9FBS40T+7DuwezEM1nov3jom0KU0zIjDf8osTKCzUp1FWS6fIlXv5HVGRPPvkkHn74YYwbNw6VlZW49957MXPmTPz4448YPXo0fvzxR8ycORMAkJeXhxtvvBE33XQT7rnnHjz11FNQKJhEpKeffhpPPPEErr/+emRkZGDYsGEAgMmTJ8NqteL666/He++9h4cffvhXOQ+aBUCCtjIwqBjSZFF0DLFg99S04pkJfWB1B1CYYoJJo2IvRqfLL1IrzRnJKGC0KiY5ihQ8UsqZoKA3EAYigFbN8IvCkQjanX66JwRGYv3KlH7IjNexx5g9Ig9Wj59VDc0bU4iiFBN8wRCeGluMDYcbsHxLNd7ZUYNpV2WhoZPpg9cf5Fm7ZBw0KgUCoTDuGtpLpOapbnbytvWHGAtyXN80VgG1bHMVqyqSspqFMa6qZgde/b5KpARbMrkv219fMIyHRxfwvn9ybG+YtArMGZmLuaML2fFtc/rgD0Zw3/Bc0Zi1UNArUn3Uqpi8F9KH7hRRUvvpcPthdQfw0i2lmD0yF7NG5MKsUWDJ5L7sZ5vbx1NiZcbr8MYdA9Dm9MGoVuD53/Vhv2u0uvHYmEL2t3NG5uKxMYU4b3VTFXO3lmXg+W8q8er3VSKF1tPjiuEPBjFrRC5DXxjaC+/sqMHxRjtVKWjzBoAI0OGiz9FWhw+dru5LSZPPFXU2vLG1mheLrGqme6kZcTqYNQo8PY7f/wdG5SMUicDhC4iUbAvGFqM2WiLASrk3Z4/Igz26vPvMhBLed89P7INQOIK6TkaGft9wZqxnjchFrF4Na/SaAjI8ODqf9bK9gRDyko24Nj+J9SZqWp3YdboNNa1OnkpMLpchO9GIIdkJLC/tYppQ2fZzvJ+Lbf+RPJiioiKsW7dO9Pf333+fuv2f/vQn/OlPfxL9vU+fPtiwYYPo7xqNBsuWLbv0jl6gcS0A2hLOgnHFeOtf1TweU3aiEYsm9IZGqcBMQU17kpS3/I7+PGAdQW3Mv6kI9Z1uqBQyXoLWMxNK8Nw3x9l6JSSBc/kd4jLOWpUcp1ocbCLjk2N7w6hWoKrFxeDtJ5di1ohclGfFoMnm4y0xPHtzH8TolQiHAaevK8ufixWhjQMX/8IFH17MtpsrG7FqWhlCYXoZAFrsCQBUnERAs0YBXyDMW3J4fmIfzBmZB5lMhvwkIxZu6Krz/syEEmw63oR9tTasnDZQhLMhYybE1pM+CD8XJJswZ2Qe4gwqzFt7QjRmwv5LKZmON9rx+b56XiJgZrxOVOP+2ZuL8fItpQiEw8zLMZpzpVXJMfeGAqyePgi1HW70StDjUJ2N99sHr8+HRiFnPZaVUwdi79lOlGfF4slonLDR5gX21rLlJpRyGSKIYOYHB3j7WXZbf+jUCjyx/ogIp/Lkb3tj3rojeHRMAfVc02N1UMnpeTHCa06ep1qVHDE65pHGxeBwt211+LB8azWe+G0RZg3PZbBNUSTR9PcYb78s04KVU8vQbPdCo1JgTVSFBwByuZwKk3xhYl/8vjwDWlXXvEs1a6CQyXDHqj24f1SeaB7NGZmH81YPnvrHcWaJd0IJLxH35Sn9APx6VSovxfu52HYFFRNtvwQVEw5H8O3RJjz0+SHcfXW2ZPVIbjB+zsg8pMXqWdwECZpzA4d908z4/ZBMnpLl2ZtLYHMH2IC7UN5IS+4juSNSOBXy25duYZaD4g1qGDQKPPT5EXxwVzmP/kq2fW/6ILQ5/bz+cguSSY0DF5lDXrxj+6ZhT00rr5LjVxV1+MOQXmi0emHQKhEOh5Bs1uFogx0WvYqHuXh6XDH+ynmBL72lL5JMWrQ7/ahtd/Gk3bTKiETSrFLI0Gj1oN3NlKD9qqKBJfb26WHBnE8Pin/7h4EwaBSoaXOzy6RERUQrFz1xQDpvXKhF3MYVw+bxQw7Aolfzrj+5bhMHpPMqH3KriQrHOxQG9Vq8PbUMe8524OrcePahyv3+r78fgEN1VmQlGPDejjM43GAXzaWyTAseuL4AjTYvkkwaPBkteEZaZrwOz0wogVopx8FzVirb7Z41+zFvTAEUcj5W5sHr85GTaIBSLsPJZqcIV8MtXzxnZB4ABsHy4PX5yIo3YOYH+3kYHC7yxeUPYfHGk8iM1+GFiX3RbPch2awRVegk99Q7O2pYOGdtuwdlmRbcUpbBXpvMeB2eGlsMfzAEg0aJZZurWOEJkbiT4m6EdMA9xitT+uFPHx2QfF58M5tJKr+YKpU/p5Hn3eWQKV9BxfxKLRyOIBxhIIoZcfqLcufTY/VQK/n1HWgoEw3HApfLALcviFU7zlBlyuQ4wmWXVqcfCkRYjEiiUYPH1x/lxUO8gTAqmxxYtb0GD4zKR2qMDosn9YHdS1+eaHf6caqFv/xAYilLJzMFx2i/y4/GHvr2sMDq9mFCvzT0TjEhPUbHk9QuHF/MPqyIZ3dn9CGYGa/DK1P64VSLE0aNAjF6JSb0S2O9FI8/zPKoyMNn7f56EUaE9Gk/B0kjlIGfbLRj+ZZqSczMT7WdWLW9Bq/d2g+v3dofVo8fPSw6uH1+ZhzCEVS1ONnrJFz24o4ZAWMGgiG8v5Pht70Q9bBc/hAPDyRV+VA4HzLi9JDL6ETkn2qZ5Em5xFLc/nNWrNrOBM1pc0mvkaPNEWAJE8I+kETLmR/sxz3XZGPt/nqexb9mVy0eGs3gV2RRvA53rmsUcgRDEYQi4u/8wTCevbkEe850svu6f1QeZg7LhkbBLKW+eUc/aNUKnGp2iZAvMllXIih5qUhd44w4HR65oQBmXdc8K0w2Ye3+OiyZXIpIROwhcqXH3Jo/UgR1uzfA+yx8XjTbu66p8LeXGoz/tTwjbrvygrmEdqzRhq8q6vH7Ib0kqzMK3fkTTQ4My0vAsi1dyVHhCD8pb1BWbLe16rmV8ICuJDFhItrEAel441+n2f3q1UpWscXdb0GyCfdck42/7a3F8xP7oqLehmF5CdTzSTRpGHij4LtOtx+VTQ4M7hVL/V2cUQ00AyadEt9XnsfNAzIQCkVw/2d0ZMcbW6tFQfPadg8e+IzxFrMSDDxEyn3Dc/Hy9/xaFxfCiHCTHIUycCKjFV4b7nh7A2HM+fQQlkwuxby1R6I15BmMyKwRudhT04rHbiqCxxdEaoyWl4wIMOgg0qNwBHjnxzOYf1MRTjY78Mr3VWwlz1kjctnrdrGJrGqlAolGdbfnLVV5kpwbEU3M/vggby6V94plidzcPqyYOhBNUUjph7vPMF51somHSOHOI9r5kO/fmz4ICrkMsz8We4+rZ5QD6MIvnevwsCKKD+4qh93hh1GjFeFrXvm+Cksml14QJURAqYgwAEoCwvT4gjDrlHD4Apj98UERUNMboIsQuhMWkEq03LHnflbIZNBICDd0KgV2nW77xZ7HpSZpXkz7jwT5/680TyCAkVEZ8Ly1RzBnJD/4t2g8ExgnnxeOL8bxBis6XPzgsBD/IkRvAF3WDbG+hJiTPTWteHvbaSzgBC8tAmzL0k0nqKiY576pxKrtNbi1LIOthNjm8FGDme1OH97fWUMNhG6vasHZdpfoGKTex/It1WhzelCWlYC7Vv+EPWc7uvX6pILdGqVc5K11F2AXYkTIuQiRNFwZ+Hmrh3ptCOpke1ULu60nWqeFiyA5Vm/FLdHrRFA2U8oyUJbJ8PWILJV7HW8ty0CD1c3+36RlxCzcgLsU5kSIkTlvdcMarVcidd4XQvV4A101aLhzqaLORu3DkXobe64ji1Kxp6YVz31TKbovZo/IQ00LI/Kobaefz09nO9HioEuuD9dbJa/FiSY77v7bERw6Z6X+9rzVI1mBU6uSs0uXK7fVYN66I3jqy6MYGUXkkHO7JXodL1Z4opCDlR7znw8lWLOzhv28UPC8mD0iDwfqOvHMhqNYJBAPLJpQgmc2HLskPEx3MuXL1a54MJfQ5DIFL1Ht2yONbOAz3qjG6h9P82TKn+87h4dvKEKny485I3Px2b56NNq8osTA7ixLKbnu6hnlaLJ50SNWg3fuLEOHKyBCmNe2M5gOEns41eLgxWO4lRCNWiU1mLloQgm1EuLaA+dwz7AceHxBKOQRrIwmCSaaNVj942nWck806TFvbQXuvjobPWN0vHHgnif3vIXjkJVgoHqMUp4GrcomLVGUHDczXoeiVBObWMdF83Ct1MMNdmTG69g6Lb0SDMiMZ+rO/25gT3oy34xyZhkpRsu7NsTaXTK5lPd/oEsivGZGuWRdEa5snYgoDBoF7xoKz5vs9/0Z5ehw+XlJr2S/6XF6LJ7UBz3j9KzkXUrenBFVHxFPlHg/5L4AALNOiQ0V9bh5QAYWT+zDom6EMaS+6Rbo1QrqNU2JEsqNGgWsHj/+OCwHAPD+zhq2D6kxdKl/nzSzSDDCju9d5fAFw/jjmq7YI016TpJKPQH6tRDO39/kJCAvyYgeMTqsuWsQWux+pFi0KE42oSDFiCabF/FGDVZuO8WeqyGK7E+L0eMvI/N5aKNIBHhj6ykehuqXeB7/DpnylRfMJbQ2Z5eFlWrRYkxJKg/PLlyXnjokk1dnnQRuL6amOdn2ljI6RmLn6TZ8vq+eFzimJYXVtntwsomJoSzbLK5p3urwAQDOtrmpa/217W5qJcSpQzLxyBcVLPb9iS+P8353pMGJRpsXNo9ftF9u7RhucipTebJYlCxndfuw41QL7zsa/oPURbF5A7wETlqiKOkD+U6Y+CeMcWiUcjaoz92WBKGlZMiNdi/mfHoIS2/pS/2eVDj0BhgoJcDc9Pddl4sl31WiweqjJvRykw/JGDZ2unhjQhMh3FqWgWa7B38/0IAxfVLZpTiyny9+Ood1hxrx5G+L2P42cJJwuWN0PirtJv33+IPU+2LR+GI8Fg1+SwXuH1t3BGkxGiqWZc3OGjRYfSKRxMLxxQiFmT5K4Y1ONDqwZvdZEQ7mvuvyUJJiQsV5+0V5xmfbXfhwdy0WjCvmiWhoydWPRPOouHOyqsXJxjtKe8oQDkdwc/8MPPjZIfYeIuc2e2TuRWGofm5M5lKSNC+2XXnBXEJL4FQpnDggHVtONIkggxMHpOONrdVUz+O1zafw5h0DoFDIqBbV6hnlqG13scmTC8b1RoxeTa12mJtkwv0j83jYdykrrl96jKQlTND9WQl6PCmwmgjCo8XBr0U+cUA6aykXpph4SZlAVMY5qQ+arF7E6tW8+BMZh7enliGMCA6dbcPzE/ui1e5DaowWX+yr5deM31aNFyb2RWGKGau2M96Y1R1AslkrSthb+NUxvHnHALj8IXxVUcfDiLz4XaVIBn7/yDwqRoQW4+jXMwblveIwby1/2zd/qMaLk0rR4fZTvbMYPVPGVyNRSqFXgoEtnVCeFYfXb+/HIlKIF8h6BDIgyaTBqu3VPE+5qsmKFyeVos3lR6fLh3enl6HF7kesQYWvK+p5YNRDde3ol5GDiQN7otHqZqW7BNa6aEIJesQZ0LuHme0vDSxJvCbuuejUyotKkCWwy05XAEqFDCv/dRqNNi8mDkgXAU25wFghEHJB1JOfNSIXeclGyflb2+7BmwJA7Gf7apGVoIfNE8D8GwvYwmsFySaqh6WNIonW7j/Homv0aiUqzrVh8aRSZjUhhrlu5LdkThLPbvHGSiSZNGhz+pBq0WFUQRK+mX0Nzls9uIfjRXW3osH9/HM9j3+HTPnKC+YSmtvfVZNbiA0h1nYozEDjpPDcHW4/Opw+kbVFLMt5a4+w279+ez/UtjpE29LwGR/srpW0NDvcfiz86jjVEn6Ko+CiASNdXj+LQiffJRhUrFdy/6g8qudT0+LEwg2VkqiNqmbGsryvG9QKaR0uP1QKOfr1jGe9BymP4EAdo4haMLYYZ1sdmPf3Y5IezKubT0l6iNwYBylxIFT0kfjYNIqXSryzHyoZvNHWyiaxdzauGG9vO42qFicWji/Gkij2hOsFAujWUyaFzqT6wMWTEJyKcLzX7u96KZISEmaNgp1La/fXU72HNwWW+5qdNRiSk0gdT6H1faTehqWbqtg+tDr93cY41Ao6goaUF5h/YwF1HhIvi5uUSTzwae/upRZeWzS+BG/8cKrLCxlbjFXbTuOmvqkYUZjC3vNEwjxdgLZpsvl5BfOIZ3drWQYPjEmqSwqTRGkrGqQKKBnvX+p5kCTNyxXUF7YrL5hLaHq1irXcc5NMIizLwg3H8O70QYxFJVHrOz1Wj/TYrrhEd7W9440aJBg1eJMD4yPoea6VRCxuKUtz0YQSHqYlL8kEs07J2w+J13CtvE/3ncOSSaV4YWJfvMSFX6aa2XOXUjmR2I7dF6R7TmYtFcFP8x7ijRrIwUdvJEkgPogiauGGY2wfuOfm8QfR4eoqW8DF03D3k5tkwuKJfdBg84jiVqR/3XmpTn8Ia3bWYNrQbLy3qw7XFaXwxpBbZuFwg5211MkaO1eNREOOkD5cDCh1zYxyvHZbPyagTYkDcdE2RHFmF8wlgEGZxOnVbP7SQ6ML4fEHkWLR4qPdZzA4OxEDesZclPXNjd9cSHkViUhjbsh+hP0VelncPnCvGw2V/9Q/jmLNXeVszsySjZU43GDHH4fl8Ob6tKHZVDwN8VjIcaU8uyfWH0VekhGpFv7KA1nR+PDuwQiGw0gyaZERq8eAjNhfzfO4XO3KC+YSms0TYK2kzHh6HkxDp4e1kmjeBCM1jlDR7lyFyZNje8PhCcCiU+KO8kxWgvnw6HwWZUJu/LX761nsu7Bu/TMTSngFskJhJsdGJgMsWhXuG57G2w9B25D+Wj1+yGUyzPhNFkxaNTpcAdg9AcTq1Zg4IB2eQIjNfeDGLOzRssJGjQIPjMrn1XN/ZkIJVHIZClNM3XoPmfE6LBhXjA6XDyatil0KBIBTzc5uSwp4AwzIkBB7t51sgRwMHLJ3qgk1rYx3cLbNhWcmlODJL7sSOkmMY1xpGtbur+cRkNfur2f7J+WlOrxByOUyZMTqEW9QYfHEPlDJZd2uq3sDYRSmMC81vUaJt7edhkIOatnm/CQjBveKRUYss19yLYTzgezX6glALmNQN1LbalVM3sjqnWfZvwupC8TrUytleGBUPqpbHAhHAJvbj7Gl6ahuceJMmwsvTu6LU5wa91kJBrz6fRU7t7meRd80M+4ZlsNWzXz5llI8+HmXt/b0uGLYPX74giG8MqUfXthYyXoWD4zKh0YpZ+cZ9z7henLCWIlCDnYcMiRKZne6/BhX2gPhcAR3XZ2Do+cPiRR9UnE3osQjXuqqbadxXaEYuBmrV8PuDcLuCeDd6YPw8qYTrAc7/6YiKOUytDkD0KmYxzbxPITJkhmxepzrdP9qNV5+TrvygrmEFqdXsx5MipluQWtUjBpECFHkWlQyGVOjgljUOrUSGyrqMG1oNoblu5CbZMRizo00/8ZCdq18YGYMFHKZKFM6O9EYjS1EeIlqwVAIJ5qdPwttw1UnvTCxL/5+4BzKshLx0OfM8tSi8b1FSybcZRutSg4ZZOyLav6NhXhlSj/UdriRl2TEoiimRQqfclV2PHKTjAiEIvjzRwdEDwwh6j8jTscrMEX2Q5ZEyLm+yFlW5CZaLpnUB8tu7Y9Ojx894/RYf4DxJhONKhjUCtFYl/a0IBzJRVGqmdr/qhYnNhxuwH3X5bLJid2hYsj/uUifBeOK0TvFBKsnwPsdybCfGV0qnH9jARVJwj1Gq8OLx9cfw6czB1O3LelhhlyWzVZx5M7ft6eW4bzNw8YF7x+ZB18whGA4QsXXxOrVohr3i8YXY/pVWWh3B9h5dcvAnuzyHneZ+elxxVjxhwH4qdaK4lQz2l0+HibpybG94fAGEAxFoFPJMeeTg+x3ZJ4da7QjRqdERpwB828qQgTAjqomPBP15HMTDdCpmOt6zzXZ1OtCgJXcuEWb08vbVqogWs84PV67rR/iDWqcbXNiWEGSyFMm8miiYNOqGCnyA9drEGdQ42iDA3M+2S1aTpPLZaJkSbJ8xkXOXM7kyZ/TrqBiou2XoGJOtziw90w7Gu0+9E4xweYNigoVfb7vHGuF0B7gpFjSg9fn43Sri7XyeqeaUdPmQpxejfM2Dy9YXJZpweyR+Wi0epEWq8Pj64+IgpCzhufCHwrjy0MNPJnn7tOteGh0IXzBMO79UDqZE2A8hkfHFOFYox0KGZN0ZtEqoVMp2TV+APjr7wfgAU7CFndf7+yowdPjivG3PbXsOjQ597RYPW9JgYZPmTMyD1qlHPEmLW9b0r9nJpSg0eqFRa+CLxjCvLVHRCoc4cuIdq7cz1qVnGVwxRtUiDdq8MgXh/Hw6Hws3VQlOs9HbijAog2VmH9jASDjY0+IFyBEvOhUcpg0Sjz37QlqrITW3zUzynGu0w2jRonnv2UMjuW392fjMQBESBKSNNgzVo/6TjdSzBpkJhhwx9t78fHMwZhBQcW8dmt/HDlvw+7TrZg+NBsnmh2M5xGvQ6JJh4ZODwxaJYAI6js9KEo18xKDuQmIwmREcoylk0sx6+OD7HkrZEBKjF50jbUqBm0z9d29mDemgLd8xb1uCjmo4pc5I/NYNMxTY4txqtkJTyCItBgtWhx+eINhDMmOwz3RFz/N8CIPc6WSnzZ4sLYDp1oc0KtVcPmCiDeq4Q6E8Eg02Zbc86kWDVy+EB76XBrjNHtkLrX/q2cMglIuZ2M13O9InRkaRoaGnOHGWX7J806qXUHF/ErN6QtCqVCwE4MLyUsya7B6B7MOPSQnEZEI8PGeWjw8uhBuf5CVA08amA6DWoEOV4BnAabF6KkeQaKRERPMvICMNjPeAINGDq1SIRIEPLruMMaVplHdeU30JiJByAc4ltHcGwrw/DcncP+oPN5vHRJLA/nJRiydXAqh4eQNhNEzTo9wmI+VIXGhN+8YgAN1VhYFcnt5Bgxafo17Lo6E+0JfPYNhpVl0Sha10jvFhGe+rhQhcoSBZnLu3kAYh+ttrLfzzIQSPDgqDzF6NfU8zToVZo3IRc84PVrsPp7HqI3uk4Z4eWBUPl67rT+ONNjY7e4flQeDWsmiYbjH2XG6jQcpdXgDCArGkIskoT0sn5lQwkp5WyVqxxw5b2OFES/y5M8lrDFDHp7fHG5EglHD2w83OC8VqA+FI6xnvGZXLR6/qQhh0DFDVo8/qmykV28sSjFBoaBjcbLiDXjttlIEhEiXccX4fH8dats9kMu6UDGkGCBRRGbE6VCcahG9XADA6g0gFJHxBBcLxxfjjTv6o83ph06tZJfDhEubhxvswJ5arJlRjk63n7r06Q2EUdvuhkJOP7dmuw9uf4j63aVKmC9Xu5LJfwnNFwyza/UA0GD1YV9tB8LhCOyeIBy+Ls4QwVpEACACBEJhrNlVi+VbquH0hbB4Y1cyHxPsFqM4Jg5Ixz3DcqjVGicOSAfAPFRmj8yFUi6DUcOIELjbLtxwjBfo5DatikliBOgJnS9+dxITB6RDr1byfptiZuqUc1HkmfE6VDU7Mevjg1j83Qk8dEMB+93o3gmIM6ihlMswZ2QuUi1d8spOtx8KhYwds7QYDcqyYpEaXYIkTaqqYjgC6FUKPLH+KJw+RsFn1NIROcJAc2lPC9t3btD5yS+PIiVGzwoJhPtRKxVYvqUacQYNnt94Ass2V0fLDVTj+Y0nMHFAOtJi9KL+vvJ9FXzBcJSyUIXXNp/CuQ4PTjQ7qP3lom2e2XAcDm8ICrmM2iepMXryy6NQKxi7MlHifLjCCDJXvAEm2M39vPCrY7hnWI7kuND+Tz7L5MyS6Rtbq6FWyhBnVCPRSN9PYlR+G29QS3yvkfwuzqCGRafGE1/y76e3/sXUjiECHFKCAWBeMu/sqEGCUQ2m8Aa9GdRK1gMh+13wj2Nw+cOYt/YIZn98EFUtTva6CftX1eKERiXHDSWpkmOoUyslv4szqNlkSeF3lyphvlztygvmEhq3eiCxFgliYu4XFZhSlsHWUPmqogFTOOiQFdtqMO2qTKRatKLgcHfyTI+f7i0o5OBhLv700QFMe3cvbi3L4D3AyX6kMCGkDopUwFomA4Ok4eBgPIEg7ruOjlMhnsb/fLCf/W5UUSoeW3cYf/7bQd44kPX5p748yo7ZLVEZ7SMCFI9ORZepNlo9qGpx8BA5Jyg1SRaMFWM5KhvtbN+tbh9vv9UtDjTZvCIMzoKxTB0UAGh30T0CJjGVjkThJlMSTMu2ky2i40ihbdbtr+Nhe0hyqlYlXUOHABRbnZ5uj0OzhIWffYEQzrbxUTxExq5VySUQKV0CFm4V0Oe/qeShjsi2bq8Py7cwiaO08a9pdWDuFxWiGkCLokmO+wToJSFuae4XFbh3WC6vNtKCccV4/ptK3LpyF9ZXNCAY5I8jALRKoGx8gZBoPKXut0D07dOnh0WEgyFyaCJgEf42ggi1psuzN5fw5vblTp78Oe3KEtkltGSzBmWZFkwbmg2jWoE/R11wgJloJFGNAACf/FIMY3x7WhkiEXqtE+HnssxYNoNciM8YmBmLgZmxogqANJlvJNK1FEBkyieaHPh03zm29vjQnATqmnAkwrj36gMMuqTZ4UWsXoM5n0hXD7xQot1rmxkZs0ohx8sc9DwX0yGsGJoRr6eOUYJJA7VSwUv2TI3R4/2dNTxkyvuCaqNEcEH6zq0ZT7yH9Di9qF77W9uqsXgSU0OHm3jL/e2grFioFfT6JOVZcXjjjv6Qy2RodXgxaWA6278LoW0Kkk3sd9w+fbbvHCPJlkjQSzZrsXhSHySb9HhkU0W3wgipejvkc5JZA7sniHd+rOH1gVvxVKuU49Up/RAIR9gA+4OjC9Hq8CHRpGEl7mTZiOCWzDoltp9sQlZxGl69tR9SLFq8s6NaNP4kefKNH5jE2yabF5nxBslKpDTPjqQUtNp9UMhlWLntNBszJPLh0p6x4Dap6peZ8Xp8MnMwlHI5Hvr8EDueG48yCbLBcAQKuQzv76zBmJIUAIBarcDNfXsgO8GAZruXnQ/DCpLYEh1CyfWYkhRqsuR/k4T5ygvmElphigFTyhjVyz3XZFOtpAshRzpcfrz8z5O8xMWvKhpEKIsFY4vx8qaTyE00iPAZi8YX46PdZ3BTnwsnCRKcOMAsR2mVCjRY3XhnB7PmTmTJBclGSVyNViXHiMIUNDm8mP3xIbx+e39Jy10KCS+0hH8628nizRusPknE/bLNDF04HAnzxoycW7PdC4cnKIp3cDHq3dVtIf0RehYf7K5FQbKRKi1udXixfEs1eqeYqGP26FoGeyJMrFw0vhhzORgRYckAIWZIuN/nvqnELWXp1D79dLaT9YSE8vfaDifmrT2KZbf1Y39LjiFExXCxJ8LPpNKj1x+iJjU+L4gjLb2lLx7+/DAAICXGgGWbxeUQDjfYMevjg5g1IhfbTraIkkYXjC3Gx3u7BCMAWLxObbuHLcHwWvTcAHGiopR3fqC2E97okqXwuyabF6U9eX+GJxCkzkFPIIhr85MRDIbxlxF5eGL9UcTq1bixDz9B9tmbS5ARq2f3p1YrUJYVh+2nmlHf4WMFIMTL4z4PuF4JLVny10ye/DntygvmEtqJJhcvVnIhK4nrTRB1j0ohx8OjC7G9qokFIaZYtPjnsQasmDoQNk8AcXo1znW4MKwgSYTyJx7B+zPKEZbwhAZmxrLgRprlvmBsMZZMLsWGijqMLe2JWSNyYdGpsL2tme1DolENuUyGVIsOcQYVNlTUozjNjFkjcpEsIdEelBUrGhfyWWgJk5IBW040saUIpDAdkQhg1KpFVt1b2xiPUacKipI939rGIFyaHUxuwK5TzVgyuRThcISHJyHHyIg38MZs4oB0JEjg7+ONDF6HCwjNiNXxkjInDkjHGxxAaH6yiaUvkD4KEyLJ/7kSd48vyNtvdxiRww12tvIkZECcQY0tx8+jb0YCZo3I5VngXI+2d6oZOrUCX1fUY9GEElhdAcQYVNh+sonxyG1exOhU0KjkcHpDSEzW8DwLKY+LIPq58aTu+k9LGqUlLvqCYdF+k7o5t0CIfp8M7hUHbyCEVZTvUjjLzMFgGMcabdCplFh74BQvvWDNzhrMG8MoqpRKOW4uTUNekhF2b5AH0fQGuhIr3f4Qks1apFt0qGy2w+ULo8nehXwiaJsP7ipHKBL5j3slP6ddecFcQuPGYIRWklSMgBsrEaI2CBqEBlEk1q2UR1BRb8Wu6jaR50O8m03H26iW+5yReXh8/VGolTL8+bpcHvZiStQDy08yiipjcrEyUlDKl747iVanX2TVCy1hYo13uv0i7A3ZVijtnjE0i2q5W91+NFr5a+NChAuxCLn5FnNGMngSBulSwvaBjMNT/ziG7VUtIrjhwvHFePtfp7G1qg3D8xNYS/Oea7J5lrAQEDprRC7vxUmuIzchkut53nddLg6ebYNBp+Htl4YR4WJbqlqccHiDCIRDePjzCh4Ys6rJyrtunW4/0mJ0+HTvOVyVE4+sBLPIAz/ZaMeGw43U+fBZVJIvhRmqbnaKvEWal0WAkBlx9KRHbuIi16vm7vdsm5O33063H6lmHTpdPqw9UC+6TxaOL8bijXSY6LM3l6A4lSmzEAyGsb6iAU+sP4r5NxbxUDGkD1xdgFIpR2nPWOw63UY9l80nWrBsc3VUJJPHVkcVrnjUtnsQDEdwVU4C/l9qV14wl9C4lrswpiFl1f8mJ4EaKyF48321B6nIFGLdSll8BSkm2DxBEUb/jR+q8fzEvshPicHQnHi8HLX6M+J0PERKfjK/3DIXe0GzJLlxFLL+vWZGOVqcPiAC3hr2p9GYQJPNC51aiR1VjCXs9oVQ2WTn5XsICzY9FQUhNtm8vOS+TAlUSLJZx6qRpLxJGoKdxMPCkQi2n2zC8xP7oq7DjV4JBhanwo0RQAYkGTVYtaMaW6vaAAC902LY8S9IpqOBLvSZm2jJRfCT67jzdDvvd8IE3kFZsXh9cxXPSyVxCuF123Sc6ffqGeVodXiRYNQACGNSWU8kmTRUjMyKqQNRIMh74V6nn852imJIxFN+eHQhXpxcyvNurslPwlvbBHGVf1Xj2Ql9oFHR41ZZCQYsv6M/YvUqBENh3D8qD1nxBjyytoLdb4PNJ0peXrOzBncOzca0odkiTM+bP1SzBd5IrK841YIUi4YnUz7WaGPLdrc4fVQcDYmr8J8VdDQ+8bjG9k1jXy7c8ebmZiWb/zNKsEtpV14wl9CCoZDYSrLo8Pa207j32hyRZfnYmEIEQgw2hRqXAGPZ9oyhW24ZcTp0uv0i1MoDo/IhB5ARq2NxJtzlCbsngNxEI2QATBqG5tts96G4hwlVzQwiRS5jkCPX5CdBJgOMagWu6hWH3w3sCUiUQebGUfzBCJz+IBQyQKtWwBYtBatVyfGXqOVr94WgkAFlmfE4Um9DklkLYY0kWnymyepFndWDJFMICUYN/KEInL6AaHxnj8iD3RNAKBzBk2N745kNx6lr7lLqKqvbj5PNTuw9a8M1BSlARKwUIjGC12/vj0AojEkDM7DtVAd7HG5Mg9s/oZdH8/q41vhjYwrZOJBMBmRH19sz4vR4Z3oZTjc70eL0RxMgDaxXQtBBBSkmluq9vUrFxinyk4woj2Jl9BoltlY2wR8Mwe0LoV3mh8cfRFaCAVaXeI7G6tWQgUEkvTi5FG9zjAhvIAy7h/EsznW4MKFfOv73m0r23B68Ph/PfVMJtVKGeWMKEQ4Drmilz+1VLbzSDxMHpKPT7UesQYklk/qiurULM5OTZES/tBjsr+vE899U4p5hOazs/75rc1AfXTY0qhX4/eAs1jAiScOnWpzITjTAolXx5oM/GGHnXWGyEQMzY2H3BOEPhhEMhtkXTKPNy2JljBoFZg7LYecZiY3IZRBVmqSh8bkeV3fK0cupBCNYmfMBAzStzl99qe3KC+YSmlGrxtoDVayVZNIo0eHy4feDMxFvVGPxd11WkkGtgC8Uxj1r9kniKE5FA+yPjimgfn+uw4N3dtTwEBhapRw6lRx/lBATaFVyHG+0swl6hC5Aq6fBJQ0QMiwRMHRnbZMlPy7GhSQCuv0hqBQyfLT3XBel+dpcHq1W2F9hfKbB5mECvoMzWS7VnJG5WH9IbD1O6JfGLjm8PKUfqluc6NfTclHeg16tYPltxEL/8O5y6rYnmhx4Z0cNFo0vwad/HIJznW4km7Ws8o7r0aZZdNEHfRNbXiBGz6Dz355WhvNWD8+bHJoTh6MNdnaMuFJe4ZJpp9uPuTcU4LZBGXD5QyhOM0fJAxW86xoIhtjscXJuZMmUm6y6aHwxkoxqUTkBco252y4YWwxEA+5alRz1nW42OXXh+GI8OCoPdl8IchlTgO/OoZkwqhVweoMiijT21KLV6eeJGTLjdfjLiDweZualW/pBqZSjR4wWtw/O5L1A7r02l7ft3BsK8MkfhyAYDqPB6mWThsm23CU+gtMZnp+AMX1SeUuDi8aXYHyfVGi1SqTH6nj3TWa8Dq9M6QeVQoZeCUacaXdizGvbeS8cgmnhqr10KgVmf3JQFKcSzrOrsuMwsX/aZXkRhMMREVbm18bIXEHFRNsvQSdUnOtERb0VbS4/a2HF6dV4a1sN/jIiB54Ak5zoDfBREDRV0IPX5+O9H8+i0ebFvDEFUMjpyBHyECYIDCnEBMG0CJEjmfE6VjotRNCQ372xtRrLbu/P3rypFi2mD83ieU3c2Eh3fSDuPfm/FDqE9FcKp/PYTUUXxMrMv7EQdm8Q3mjQ96sKBs2yp6YVU8oyWUGG1AOmX08L2pwBfBUVO3h8QWQl6HH0vJ29jrQX4poZ5Thv8yI9VovaDg8e//sR0XVVK2XUgmgpFg3mrzvKG9+nxhZhSfR4ALodMzK+M4dlY9nmaklsz8u3lCIcAQ8rI7Xf96YPgkmjxPbqNl7RK9o1XjK5FI98UcE+oBdvPCnq3/D8BMy4uhdsniCSTRoeZohsu3RyKSqbHDycTkGyiSfWIPP3pVv6odXpQ22bCxEALn+IFRYIBSEbZl0NuVzGw6lInfeckXno3YO//Ee+W3NXOULhCPQqBWZ/elCUJvDOtEGQy2WYsXqv6Lv3ppeLFF3Chz0tBiOFqPmlrabVScXKCDEyP7ddQcX8Sq3V5YMnEOZZTQ+MygcABMMRxOnVLDakMLmLFMzHURhxqsUJjaJrEslkgEYhpyJHgC7UyuJJfaBXKyWX05ZOLuUhRy4kneYuTwnJsFolvz8mrRLP/a4PdtV0ID+JTkHmBqzJ/6WWArLi9bj76mx8vKeWVSqlWrSYt/YIGm1eUX+4WJlWpw89LFocb3SIPKO0WCaRNcWixttTy9Dh9iNGr8JL353keT9rdtXC5U/Hqu01WCioEPn87/rg4dH5SDRqUdXiFEnNz7a78MjaI8wDd1JfFlEjlwGpFh3m31QIi06F9QfqeTGBVdtO4/dDMkXja9apLmpJjzu+vVPNeO22fgiF6cuZoUgEAB85olHShSgNVg/Odbjx+b56dozSLPRl20gkgruvzsaaXbWYNDCd951GKWc9gpkXqNsTQQT5SQae3FkoYabVUJGSdpP9trl8LJngQuNZmGKCmRNIfAAAS7FJREFU3UNPZD7bxlxjqToznR7GA6V91+HyiR7gtPyVdIsOBclMCeUUi1YSUfNLW7PdSz23XxMjc+UFcwnNrFXhb3u7Ki4CwN/21uJ/by6BQaPEnRxLbc1dg0TB2Xd21GDF1IHskg6R5/ZONeNJTm0WImm+f2Qe6qwe7D7digSjBsFQRFI6e67DA5kMPOTIhaTT3OUpvUbJJnRmxOpw3ubB5xxvJzNehxcnlaJnjE4y4YwbsOb6ybRE0XiDGrUdbti8AWiUimiSoILtP41U2+n2Y/85K97YWo3Xb+/PegDcc3vzjgEIR4Blm09h2tBsnGpxoiDZBJs3wFOgkT56A11VE4nQ4NXNVVg0oQR2TxDC+12rYqobkmM+svawKLF1yeRS7D3biW+ONWPdoUbeOKRYtLh/ZB5sHj9TRCsYRpKpK4HXE41TXGh84/RqNEUZeLRtLToVtCr+spdUTZUeMTpolMzYc1VvtG1j9SoM7hULn9+HQVmxbHmB93fWoLSnBaU9LTyPQKpuT7xRg3annzc/hYIWqXo7NGk32W+SSQuHl16DSPjZqFFKEpHJNU6VqHf04d2DoVLIqPfXh3cPFsVkAHH+SjgcgUmrgtsfgkmruuzLVlJCg18TI3PlBXMJzeELUC2WqhaHCAD43o4zVGnk6h1neOgKoWcBQLSctmh8MZulnBmvEwWLub+9mAQzEkjkyoe3VzV1m4wolP3SqjN+vKeWPU8im22xuaiJohsqGvDNsWYsHF+MpVG5dlmmhR0zWgXIReOL8X0l88CWQrEcqLMycZVhuaJqoyTxUnhuXO+A5vVxqccE5yEcT+5njz8okhNzpejCKoqjeyewWCGp8eUKAhaM7UrYLMu0UK/F0u9OwuYN8L5rlKh4Wnnejm+PnufNVylRwry1R9jlv/8RxHK+rqhHkoUPqJSq22N1+UX1VS42QVJK2k0SGQ/Wd4oEF0Jp9OwReTh4rhMnGu2i81w4vhh/318HAGjo9FD70Onyw6ilrybsrmlnq3VKxTz+HfERmtDg18bIXInBRNsvicHsPdPOk3ICzMR+844B0KkVmMGpcAkAUwamYuLATFYS+vlP57DuUGO3a+wAul1/BxhLePGkUtS2u9BgFXsaiyeVotXhRZJZizsl+rv/nBWnmq34/ZBeaLJ60TNez6vQyd1WqZBh2eYqDM5O5JUBmDY0m0WDqBUypMcZYNAwCXuTB2aiyeFFiklLXYNfMXUg7nz3J9biJ4mW7++sweDsRJT3Yip3cuW3Gw43YNGEElYaK1yvv9AYrplRjnaXH5VNdt6YXUzM6K0/DIRepWBzl6SuDTmf2R8fRFmmBY/cUISm6JIIMRKEx+DGv0gj17HZ5kWcUY1Gqxs6tUpUifS+4bm8OAYZJyLBzYzXsRJsglOhbQtAtJ89Na14+IYidArGTGqMiGx51fau72aNyGVjY9z9PnJDEdpdfpxosvPigpnxOjw0upCdD1LXmHjg3HpLe2pa8eTYYrS7/KhudsAfisAbDKMg2dQVZ/MHoVczHte0odmQAVi66YSof2Rul2fF8lYXSB8+uKsc8UbNL0LnA9LxkfdnlCPRpLlsRcSIiuxMUwd6pcRdFvHAlRjMr9Q63f5urWaulZQZr8PArAReve45I/OQaunsNilTCuPNtZJr2z1otnvx6venRLiPP1+Xyz7I5t9YQLUez7S5sO4AU7GQWKHC9W/uuRGUu9ADOG91s5UPuUWhFowtRlWzHU98eRzLbutH3a/VHWD/L6yi+cHuWmTE6qiJlU02b7cVQz/YzcQGaMfce7YDH+05h2lXSSNSpK7NiSYH9Go5Jg3IwNHzfK+Kh1OJVjDMjNdh0oAMHvaE9E8YE6BVRqxt9+AcJ9azcHwx1u2vQ5+eMbyHnTChk/t3sp/6DjfmrT2CKQNTJb3U28szqPs51mBDvFGDZZu7/i4Vy7F7AiIvhHiT3Pviz9fl8saF6yHed10uGw8b3TsBi8aXsGINoafMxdMQIc2tK3fz9rt2fz36pBkxqqgHz6NlvOg65KfE8M6b7OdC3q/LH8TA+LhupchkXGgxD6n4yPbqNqzaXnPZioiRZTlfWx2yEzN+1m9/SbvygrmElmLm184mahmrO4BnJpRgQ0U9mySYYtHy1qJj9Wp4AiE8ObY34g1qSYBlSAJrIZTyJpo0mDQwHRFE8OCoPPSMMyDeoGaXTgBxnXKtUg6lHEiN0eGFSX14/esO4eENhEUJkZ/uO4dFE0ow/6YiHsIF6EK4zBqRixQLfR04Rq9i/8/F0h+qa8dLU0rR7vRjzV2D8N6OM2xio1bF4MyBroTDFVMHsgmcQnCj8JieAB+iWZhiQijMLA8unlSKZruXpTwLf1uUasJPZztFyXyHz7Xh+Yl90e70I0anQqPNjWEFSaJEVrI+Tzws7vWXirlwYz0kMbeq2XFR8QVuvCYz3oDFE/sgI16P9348jSWTSyEDcKrFwY6ZVHwmN5nJp7qYWE6iSYtJA9MRq1di1bQydLgYgcW2k014d/og1HW40TOO7yl7A0xcZWUUfvr3aDVRjz+IzHgD3t3BAC0Jvuaj3WfwyJhCmLRKPLPhGFv+WejtkPtt/k1FiDeqRcd8ijOe3Gsh3A+Z+8LKtMtu60+VIr+xtQqP3VQEjy8Ii16FFrsXbn8INa1OnldCSmBI3W9PrD/Ku98e/OwQCi9R/QVAVG75cufFXHnBXEIrSDSyFtVVveJ4ahliZbbaPZi39ijPIxDKlMsyLSKYHQEJfrGvXrReTIMQPrbuMGvdLBxfjM9+OocxfVJ51i23rjpZ9yeehtBjoSFILiZOcf+oPGpcqqbFieVbqkV4Em4sSniM4fkJGFmUysv/WDi+GACw60wHFnKw71qVHLeWZeBko50JZiq7BAJSa+7kOFyIJgFuHq7rxHPfnsSbd/SnjoMtWqKZi5whcRVavsrvB2dQLVSdSo79Zzt4158a06LEejz+IM53uH5WAufC8cV4Mbqsx7X6E41q3F7e5clJxWdsngD+vp+PWpHa9nBdJ9bur8e0qzKxkFsKeHwJO18XT+pDHZdGmxcL/nGMhw56586BGF7Az1FZMLYYgXAYxakW3FaeyWbZd3e/SXnn/mAI20628DwsqW25XvaiCSUoSjYD4Afu/f4QRvXuIUISPf53Bs1EQJjkmj97cwn7ubv7jXy+VPXXvyPuc+UFcwntZKsDb/zAWKHX5MVj+nt8q2jBPxgEOMD3CIRqmMHZiSJ0CbGSuMBCmQxINGqwq7o5aiX7kMhZy+ce9+6rs0VWERcrIgN4lpnQY7kYj4BYxdzzSZNQ2ayYOhAAWDzJu9MHodXhQ6xehRa7B1MGZeBP1+ViLgf3Mf3qXlSkzuoZgzD96l74aDe/Yuin+xhrN0anxMvfV4nw8VyUCQ3GSKzFZVtOsbj+eJOGlzBLjrNkUikeGl2IDRV17H5pINLXNp/Ci5NLYdbRLdQ+6Rb0z4jB3e93gRAJeodr5b+86QSPIEy8twkD0vHUl0dFuP41M8rRaPfCoFYAYCplpli0eH1zFRszEnpRXGRLYVTJKDzv5yf2RXF6DGSI4O2pZbB5AkgyazD3iwrRtmP7pmHigHRemWPGW+iyxqUsd11Ufs/1lI0aFe7bcFB0n6y5iyklTR7OF7rfpLzzRJMG1+Qn8Qr6SW3Lw/hsPYWyzFjRw76y2S7CvxDVGwBef2vbPXh9yyl8OnMIOlx+HKyzXrB0wqWqv862u9iXC+nf5fKMSLvygrmERiB0b2ytRn6ykWrpkNLIa/fX47ExhWh3+5Fm0eGea7JZpIuULt8fDGHWiFwoZEAEEbj9QbSc90Kr6aqhMXtkLvzBCO4bnstaOGv317OFwZ7/XR+caXexiaClPS3wBSLwBcN4ZUopWh0+tDj9MGoUIgTNfdfl4mSjHVq1AmkxOsk4BVfdI1QCCccBYF4yhakxWLa5GmWZFjxwfQE6XAFEAMwfU4BjTQwaRJiTQDAirQ4flHIZxvbpgeo2N8IRQCkH7rs2B+faXbDqlFSP8KXvTuJwg11k0WbG6zBvTBGqW5yYNSIXx+qtiDeo8Oqt/aCQgRenKMu04Lnf9UGrwwe5DLi9PBM+fxg9Y3QIhyMsAZd77m5/EJFImGrlewNh+ILisrfcWAmhKnBjPSS2c9fV2fAHu548MhlTWbXJ7sUL357An67NZhOBDRolda6QWB839tA3zSwCVi4Y11XOoW+aGfcMy4EMDGJl6pAsLN0kTkaVin+R45PidcLjEG+Nu229hIKr1eGDTYBfEt5vF/LOHxiVj5oWp+helPLkhaUIaN4EyS2TOnfaNfcEQrg2PwmeQJh3v5EYDPn8S9VfXFSM0eGjztfLmRdz5QVzCS3V0hWDSTDQ81Hi9Gr2sy/ET8rkusC0357rcPPqwmfF6+HxM3W6SVJeaU8LdCoFL6OdZFW3Ov3ocPvZY5ZlWtAjRsd/8I4rxraTLahqcWL+jYVYNa0Mu890MMmdKjk+2nsO/mAED4zKxdLJpXD5gzColYjRK/DomCLYvQFenMIfClPPRa1U8D7/JicexT3MsHmCrPVOXlxfHmpAbbsHy+/oz+6LRj9YyNmWnAups16WacHbU8vQ6fYj2azFi99Vsh4A15OLRMJw+kK8ZYKF44sx/+9dteefn9gHr9/WH6FIGHZPSNDfEny2r5ZdchKSE7QqOftCJCh/rpV/26AM5CQaqWOWatGxVvLmykY2nsdFEqXFiokGc0bmISVGizuHZkKlVLDXvzuCgT/Ij/WRonIrpw5Eo80Lg1oJtz8Aqy+KnCnnB76fn8gko1o9QZGH2F1MiHocXwCtTr9oW72aDr+UATjd4hR9R+43IeqIXP+lk0txotkBuYwRcyRbdEgwhanbfnj3YLQ7fYg3anhFxEgfEgyabp8PwnMnqQHC75JM2l+tiBhtSYw2Xy9nXswVmXK0/RKZcm27E4fqrDjd6sJvcuJQ0+bh3bwLxhUjLUaDvWetKEw28TAdQJeEccPhBuqNL7zwb08rg1Ihw3s7TrMok4x4vUgOrVXJMWt4LvyhML481CUJFS7hkG3fvGMADtRZsft0K/54TQ6OnGcexER67PEFu8XKcLEtJLbDfeA9M6EEHn+QhTOmmDXISjAgEAKe/PIIT9zAlTsnGFTQqZXskt+F5Nq0z2/9YSD0GjmabD5Ut3RBE3v3MCMSBjQqOf700YEL7nfl1IGIANTxe3f6INS1u9kEw6G5iSz7jbzsSaMZAoWpJjTZfKK5k2rRYF+tlYE8Jhrw0j+r2NwnAm4sTjVh0dfHRbLZv/5+AJy+EG+5UkpOPGt4LjYda8LskXlQK+XocAUQZ1DB6vbh+W+reJLhR8cUIRiOoLrFIZITPzOhBHvPdsKsUcCiV2PBP45JzodOl4+Fn8Yb1Pjrv2pEc+udHTVYMqkPGqxeuPwhlGXGoJEyTr3idQiGgfNWN+qtXpac8TAHdUQr2Ca8v8jfhP3lxiWONlix/VSb6Dpek5eAkrQY3rXmov2Fx1UrZXh4dAFOceZkn3QLchONaLSJA+7dBeMvNlAvJYUmmKFfGoO5IlP+lVqHi5Epr9xWg3CEqW0hRIFcV5iE5VvElfuALqTL2L5p+HhPLV6cXIqTUb3/cwIX3BsIY8+ZDuw+3cpCKMkSGW2/PeP0UMpl3aI3yLbcZMQ5n3ZZN8LaLFJYGaLEenFyKdz+INy+IF6Z0g92bwA9YnSo63Cx1fnIA+bRdUcwbUgWtfIk95iP31SE96YPQqtTut4997NGgNQ5XG/D2gN1mDWcD00kS3zjStMuar/7ajslcSn1HW7MW3eE7X+sXsl6HlyEytr99TwPhnzX6vDj4z21knOHLJEsmdwXzXYfAqEIyxsTXhfSp2Pn7fAGw7z+Si3FeoNhxBvUaHf5RYnAhclGVvZ7a1kG9bgAk3jLFbg8flMRZg3PhTcYhlYpxytT+iEYDiPZrEWz3cuTsRO8ErdPhSlGvHFHf7Q7/byH+aJxvfHyLaVw+LrGaVhBEgspJdeYO9e5aKaseD0MagWe+Vp8fxWmmPDKraVIMWsxuncKWp1ib6G+kyn2JryO2QkG0QuGW3CMoI9MWhX6Z8QgxazFsfMO3px89uYS3PneXpEUGYBkML6774QvCSkpdP+eMfhk5uBfpZDZf+wFEwqFMGnSJCQnJ2PFihWwWq144IEH0NDQgLS0NLz66quwWJgiPytWrMAXX3wBuVyOJ554Atdccw0A4OjRo3jsscfg9Xpx7bXX4vHHH4dMJoPf78cjjzyCY8eOISYmBq+88grS09O7684vasFQhFf/46uKBpxscrAPJ5s3cMHKfec6PGwCnFmnRM8YHSw6JQ/xQraNRPh1Wrrbb4xeBZVCjoc+v/C2kQhTj4Ib5AWArScbWTAmsc4nDkgXYWUAJifoRPTcuYl0qRYtVnAgid5AGE9+eRRLJpci0ajmeV/egFj+/L/fVEYloeIlEFrgs7SnhakuqVEgEgHSY/V4eHShSHjwfWUjKydefkd/rPxXF3qett9QmI6r0arkSLZoMWtELgBGTPDg6EIAzJLOjKGZsOjU0GuUSIvpWkaRyYBrcmNxdW4CWh0+PHxDAU+CnRmvYwPJAPD6llNYPKkUFp1KJCSgIVIy4g0XLWHOTzahd6oZJ5vs7Jq8N8AIKlZMHYitVW3dYoYAiL5btaOGN3fe2XEa04Zm419VbTwvyhsI45WoIENY6VWpkPOqgALAG/86zSaNkv5fk5+EsX3TWM+GNtcJmumdO8sgl8mo91ecQY3SnrFsxUqrOwCdSolgMIx6G5NrlmjU8BA65LfcipfcRgqOlfYEb78quRxLvqvkjcPrW06xSaUAsHhjJQpTTAAgGYwn23HHiPxOGEeRQsVkxhv+77HI1qxZg5ycHDidTD2SlStX4qqrrsLMmTOxcuVKrFy5EnPnzkV1dTW+/vprfP3112hubsaMGTPw3XffQaFQ4Omnn8aiRYvQr18//PGPf8S2bdtw7bXX4vPPP4fZbMY///lPfP3111i6dCleffXVy34ONm8XKiY/ySha5iL1wwF6sJC4y8IKlt3hX+4fmcebIFIVDZvtHmhVygtuS/Z759BMkTexkFO1kpxPIBxiPYDP9p0DAN5nfzAiGgeahV3d4kAwxMeIkO+E3sPZNhe17yThjfSBVFwk0lguCZh7HJr8mSDjq1qcPLQNt/+JRjW1xj1RZpFKjkKZ8qubT0UrZXYRqEf3TsCoolRe4i2RYNe0u0Rom9kj8uD2B9Bqpyf3chEpJOlVWC2SJmHmVu+keakkAbY7TIuwyioNr7NofDF+qGxCjzj6NadVepUCSyoF50oTE0jN9dOtLvzjUD1VKi+X0Ze1Fk0owWc/MXE2WrInV6Ys1Wj75Y61FC6qw8V4rLQxa3F4oRCsUnB/J3xp/P8GFdPU1IR58+bh3nvvxerVq7FixQrccMMN+OCDD5CUlISWlhZMnToV3333HVasWAEA+J//+R8AwN13341Zs2YhLS0N06ZNw8aNGwEAGzZswN69e7Fo0SJ2m/79+yMYDOI3v/kNdu/eDZlM2vW7VFTMhZDqADC6dwLuujoXrQ4f4g0qKBQyNFp9SLHwpcYAWKxIi8MHyMBa2K/f3l+EyyjLtOAvI/PRZPXCpFOiw+lDvFGLeKMaSzbyUSYEFdLu9ONUi4PFfdD2S4tFrJ5RDn8whA0V9fhtaTproX5VUYfpv8mBPxjGvR+K4xRCtM1DowuhkstEaJDu4irEui1KMbPnRnA1QswJ91qQ/Bby+f27BknGUn6sbsepZivuuToXzQ5mTXvV9mo25pVk0cAfiCAQDiPJqOGhYi4Gq0/+L9WHt6eVQaOUY+4XFaK4ypoZ5Wh2+KjXiZv4dyHcC4lx5Seb2JcLd19kqfarigYsmVSKJhsD0eR6m+RaFCabEGtQ49FoXkt347Bi6kCoFXK8tOkEDzP0VUUDFk8qhS8Y4o2J1Jxcc1c5WqLCiUarB+3ugCTan5C5CQ7mLyPzsfdMJxWn89ItTF0Xkv0v7DtX4i78LUHyS8VDKuo6qfu9EJLonTvLYNSoMPuTA6Lr9M3sa+DwBqj7/XTmEJT2jIWw/f8CFfPcc89h7ty5cLlc7N/a29uRlJQEAEhKSkJHRwcAoLm5GaWlpex2ycnJaG5uhlKpREpKV2nSlJQUNDc3s79JTU0FACiVSphMJnR2diIuLk6yTz6fD5WVlT/rPFoDFvbCSq1v61SMuVWWaWEtVlrgU2jNjChMEeEzWp1+fFVRx7O+CIJECGN844fjrNUMnGNVTvcOy8UL31SyxZ3IUoEULFLoTdR1uPHkl0clLfd7rsnu1sImaBBh8hkXHsn1Srhy6E63HzqVAos2HEdajAZTyjJEyYgf763FsIKkbq3ZTkq1Rm+Akbu+s4NJtDwUTbRcMbUfRhal8vq7cHwxNh5pRJ+eMbyX98Vg9cn/pfqw50wHm+wp9PpanT6s218ngqYuEpQXIEXlhuQkUnEvQ3IYqfGsEbm8hxY5zslmB1Ztr8Gi8cX457EGrPqxjudV0+Yv1zuTwusQLhl3W/LbdftrUZ6dyPudlOR975kOFh5JElnVSpnIK/nzdbkiD9zjD0ridFrsHhHanxxzf20nG0ul/fZMUwcCnedR6dDg0fVdFS5fuLk3ikw+nPdKe26AtIe4q4aZD89MKMbyrV1j9sLNvRHoPI82Hz0u2GZ1otLZBKkWF/TC1+bCyTbJTS5L+7e/YLZu3Yq4uDiUlJRgz549F9ye5mDJZDLJv3f3m+6aRqP52R6M4wy/PjptfbNPOhMT4Cq4aMln3HV0KSw5SdjjJtZJJfeRfT31j2NRy9eLZJMWu6qbcc+wHHh8QcQa1HjrD4zaSAqjLkruMmtwzzXZePOHaknQJG0/JJ5AQ6aQczvR5MDaA+fY/UYiQFWTFUsmMfk62ujL+v6ReUiN0eLD3Wf4tdy3VbOWu3D9/VBdO1bPKEeL3SuJq0k2a3H31dk4VNeO3w/phcUT+yBWp8WcT/i16Rf84xjevGMAXP4QMuN1IstSagy5/48zqCS3Fc4H8l2iUYPfDUhHklmDN+/oj1anHykWLT7afYZnUZNrY/fQMfXCayrVB4JPwY91bPInLUmXbLt6Rjlq213IjKejY7j75cbZyG+FCBopybtHYq6/8UM1Vs8oR3PUSJsrSEBeuOEYm/Ar3G9mvA4JJi06XH7MGZkr8qovFEvtlcIYr49+sJ03Lo+uP44P7x6MOCUd+TQ0JwFpMTpJJBEZsye/PIZPZw6BJxDiBeNVrfTYZK+UWGQn9oRU+yUrNt3tS6pdvmo2F9kOHDiALVu2YMSIEXjwwQexe/duPPzww4iPj0dLSwsAoKWlhfU2UlJS0NTU9SZubm5GUlKS6O9NTU2sB5SSkoLGRgbjHgwG4XA4EBMTc9nPpSjFgEXjS6BVybF2fz3mjMxjH4LEK6lstGP5lmo0WbsUHFKW7oWsmTNtLlSed7AW1PIt1aios13Qam60e/GXjw9hyXeVSI014pEvKjBv3ZFoyVYfVv7rNFZtr8ai8cW8/i8aX4wNhxvYzwvGMcmKq7bXsFiW5VuqcarFwfaBeAvCcXj+m0oGFdPsoPb3ZDNTgnhEYQqqmxmszPEGK/JSYnDne3sx6+ODmPXxQZxqceLVzacw84P9GF6Qiu1VLVi+pRpvRK07hZxR8y0Y13UuXO9x9ieHsGRjJfVcd55qxp6aVowqYnAk89YdwZluygDM/aIC9w7LRWa8DgAjsV4o2O+CccXYXtXCWtBkPD/cfUbUh9kj8rDuQL1oPpD+zV1bgdmfHML0935Cs8OPj3bX4n8+2I9+PeOx7kA9bxyO1NtYb4d7jIWca0oQOt31wR8Msedd2+5BVbMDJyWu4c7TbZi39ggqz9uoc4C7X6Fn3Gz3orbdxesPiRlJ7Ue4r9p2D3aebsOcTw/hUJ2V6p1VNjpE+yUx0Gnv7sW9Hx7Aim01mHZVJvvQXzCWuYYA2LgWt0+kLECTRGLlgXOdONpAH5PK8zbMW3sEyzZXiecO57jeQBieQAhDshOQnWhkl7VIXIX7u187rvJz2r/dg3nooYfw0EMPAQD27NmDd999F0uXLsXixYuxfv16zJw5E+vXr8fIkSMBACNGjMBDDz2EGTNmoLm5GWfPnkXfvn2hUChgMBhw6NAhlJaWYv369Zg6dSr7m7///e/o378/vvvuOwwZMuSCHswvaTVtHny2j5GXahQy1La7WGkmF5cBiBVINKtjSHY80mJ0khZ2VoLhoq08rtVMQJJCBZo3EMbCr46xOHmgC+GSaNJgfRQ0KJcBiAArt3Uprbg4Fa5VR6tFT8OyCPtbkGzC3Vdns7gXgI6K4VqsBKdD+q5VyfGbnASUZcbyMCdCL49Z1jrHenaJJi2+P9aA0owEDOyVwNv2QhDChRuO4f0Z5Tjb7kKveAPmruUjU976VzUeHl1I9c72nW3D6hnl6HD5cKLJIUKDDMyMxeJJfVisPs8a51w3KRVZn3SLCCPzeXR8Sbxm7QEm4VQll1FxQIkmjei8pZIEiZXf7g6wUNWMOB0arJ4LYk8STUzBMS4olWBvuOUbpBA/wj7YfXTvrbiHGS/d0g+f/XSWlYVLATeJV00848MNdlyTn8RTW0YijMJvQEYsNEp6ImhBCgNG5YJmhc+HwdmJrCJV6JEfbrAzqweUBEhaUubllhpfSvuvyYOZOXMm7r//fnzxxRdITU3Fa6+9BgDIy8vDjTfeiJtuugkKhQJPPfUUFAomK/zpp59mZcrDhg3DsGHDAACTJ0/G3Llzcf3118NiseCVV175VfrcaPNiX60N+2oPYmK/VAzJSeCtAT//uz5osnsxa0Qu2pxevHprKY43OqCUy/HKrf3wwreVvPXhF76pZFEmQgXMk2N7IxIOAzJ+ETEayPHJsb3h8AYwZ2QuchKNaOp0Y9aIXIQlyul6/EEAQJPNj0AoDK8/BLsngCE5CahucSAr3oAXvzuJiQPSMayA8RLX7q+HLxBiJcFczEyn2w+tUoFXNzNoi8fG5CPGoEGHK4BksxpZ8XqemuaZCSUIh8MoSjUhMzYHiIQxa0QubG5+nILgSRCJsGTlSCTC4nRykox48btKTL2qF2+dfPFEMVBxX60NzQ4f3L4QWuHD3rM2aDUa9Izhr2nTUCZCCOF5q4exQm/vR12fJ94ZN26lVcnx2JhCdLr8ONfhRqpZy0ODPDOhhI0fvHprPyrihVy3WL0aAzNisHhiHxi0SihkwAsbT+DP19LjBUNzOaDG8cVosrqRGqPjAULJeTo8ga7xjSZ7+oMRPHh9Pl7+ZxdW6IFR+QhHr4VOJcddQ3vh+Y0n2HgNHzNUgjd+OMX5XIxgKARPICTq7/D8BCSZNJAhgmSzBo+NKcDxpq7kRJKkScbT4QuyfXhsTCGe33iCd18EQiH0sOhw84B0dDiDQASieUaQRG4fE6/xByO81QUumkevluO2QRmoanZArZTRi6m5/TzQLG3MesbQy1GQl3l3XgmtMmZNq/NXIyT/nHYlkz/afsmaZEWdFbeu3AVvIIy+aWbMHpUHtUKOTncAaTFaVNTb8OJ3XXymp8cV46//6grUPXtzH5h1SsRoVaLCVURpdbbNhdwkIxZvZF5Gc0bmYv0hsTJo+tBsNNu9yE404OmvjnGOUYJP9jISy7f+MAD3f3pIZGG9MqUf/vpDNW4fnMmTF5MHwe3lGYhExFnoChnw3LcnoVXJMf/GQiSatLB7A7yiZ6QmOzcovWRSXzi8ATh8IeQlGbFoQ1d/F4wrRigUwhNfHsff/liOu1bvY8f39vJMkUQ4PVaL36/6iX0op8VoIJfLeeBRKTXS0smlmPXxQfa4gWAIiSYNHvhMrNIjBbFoxcneubMMu2o6cE1uPO7kHJd8/+YdA+D2hxCOhHCuwwtvMAyDWgGFXMY+pLnZ+cFwGHlJRrZscV6SEXvOdIjGPy/JiCe/PEYNuAMRpFh0bPKj8LwJIqVXggGvfl+FcaVpokJgGw43YEK/NDbL+9mbS5CdYECz3YtgmMH7hyNASaoZLQ4vL5n2wevzkRVvwIkmB0rTzfD4w7B5AzColTBo5fD5I+xnuRxYvPEExpWm8dR+tLkjRPMsntQXaTFaqORy7DnbwXvpPXh9PjLjDTjZ5ODdQ1oVIy1+Y+sp9p4iuVpSWf/X5CXA6QvCqFFSM/nX7KrFlLJ00b254XADHhpdiNkfH2SVd3lJJqjkMt6YcfvAvVak4NjFviQulpB8uWMwUvu68oKJtl+Kitl5uoNdrlgTrYrn8QWRlWDAnYLKjdzqfCQRMCvegESTGmfbXTjX4WEts4IUE2rb3YjTq3mYFmGuAHPT9YZRo4JMLsOJRrHslyylLBhbBG8wLLo58pONUCnkeJJTMRLoSpjUqeRsOV/StCo5Xr6lFMebmKSw3adbcffVOTjV4kRespH1ztbcNYj6kFsxdSBsniBe4kg+yTHn3VCIP//tIN6bPhAnmpx4bTOzHCes8qhVMbLeqe/sZT+/P6MckEXQZPPidCsD+Yw3qBBnUGPe2iOcm64UCpkMVneATSKdPTIfGqUcZ9vdIl5bikUDjVKOijobdfzufn8/XvhdMey+EO8ht2BcMWweP5y+EHafZgyBE80OFKaY8OJ3YokwV9K8YupANFm9yEky4A/viKuAvn5bfxw5b2OzwbnfvT+jHFa3Hw5fCE+s7zrv+TcWwu4NwhsMs9ftLyPzYXX5IZfLeOgSIcIlM16HFyeXosPlRyVnnj08Op86P2YNz4UnEOZJgsk1FibwDs5OZIvekYe71NwRLot2J9d9e2oZAqEw/vw3aRxQqkWLGb/Jwsv/rJJEEn36xyFwB0LQq5V4Y+tJVrYu7L/43uS/EJ8eV4y/7anF6OIU3pjR7utfgm2RwsEIq2j+u14w/zVLZP8vthaHDwpEsHRyKQxaprqhFMKFm0hFk3lywY2Z8Tr86dpcSRnzml21LPiwV6Iep1vcePDzCuq23CWwdndAEldi1iioCVtyOeD0iWm/3kAYJ5q7lloWjC3GCxwL8amxvWH3BmDziKszegPRBL5IhHrMYNTmaXUGWCwHIvTlvQ6Xn/e5vtMNlz8IDQfySLyb5bf3R7vLj16JetS0uPG0ICn2bLsLz31zAo/fVMSibrjYljSLjooJefy3RVg8qQ8SjGr4OjwsiFQuAwLBEN7fybCn7h2Wy/KxhNeJ9J8rzthf24llm6vx4uS+1HN3B0IoTDZRv/vxdBuWbWYIEcvvGACr248UswbHzjvYBxs576e+PAqLVoXfD8nkjRkX4ULmL8n74vbfLFGL3qxVQqeiX+OaFicWbqhk+wCERUgXqblD5jP53Gz3imjK5LufajsuqiqsRiFn44a0bTefbGHHUyizJwnI5N58844BaHX6YNaqoFXLMDSXKSkhlwGBUBitTr8I40N++/6MckQQ+cWxFCkczOUkJP+c9m9Xkf1fajLI8NRXxzHr44PQq1TUOhKkTRyQztJ0mWz8EGKjpGVvgJG+koDf2L5p7MOPfL9syylMHMDgbjrdfqgUciACICzjITKE22bGM6KBWSNyUZBsYjEXRHHU6fajINmEwlQzPt13jrefT/edQ1GqGQXJJt65APyAqjcQxlvbqvHQ6ELMGpGL+0cxhanSYvRINmtYlRX3t4kmDZLMWip+JMnE9LdnrI7tb4xeRe0DETCQz6kxDGJFWF/nyS+Pwh0IY97aI/D4w6LxXbjhGDLjmVyF//2mEkfP2zFv7RHM/vggbN4AcpNM0GuUmDE0k81tksmAtBgNEo0aIAKYtCq8u/MMlm1mxveHEy2IM2gwb0whFk0o4c0P4XUi/acFrJPNGslzN2qV3V6b2nYPZv3tAM60uRGKAH/byzzAZ43IxT3XZLOB5HuG5YjG7JXvq9j+SaFi5t9UhIJUM7UP+Slm5CYbqb/LTjKyn9ceOIecRBNevbUfXpzcFz1jNNCqFJLnTaqYks8qhZwlF0uNA+07bj2jd3ee4eGApPYztm8atXZTZjQ+0un2Y/85K+atPYKj5+3404cH2fnw+b56tDp9uH9kHgqSTSjLtOC+4bmYNYL5lxajEQkgfm4jOBhh/y8nIfnntCsezCW0DlcXgLHRxq9VIUzus2jpHgLX0zBpmcvRXcIeWWMnFQGlatzLZKAiaIRJenNG5uG5byrR6fZT0RX/88F+xOrVmDMyj+pRARf2zmiJdat/PI3BvRKpfa88b8PyLdVIMXWhWd7bcYaaYPjR7jMAuoLSNa1OqJUK6n69Ucu3xU4HZ7bYfez/uYmhBNtCzo14AGR873xPbNWTCpEPXwBKypUicytPcsfX6vaLzn3heGbpbUdVC7X6JUEUceeD0xukzkGNUg4/pSaNN9CVKCwlnT/Z7MDSTSeoQgibxw+Hl+6FkPpAfdPMmDQgQ5RU/NqW00iL0VCRLtwqprNH5MHu9WNolhjhwh1DYfCdxGAA8O7NC811qXuzxe7rdltabIfEOLlxodmfHBTBLn+OF/OfwMF01668YC6hJZt1bP3u1Bh+7QeS3McmOZq17PICIE6m06qYOu+kcfdFPg/KisWw3HKs2lHNrt0Ty160bWYshubE8+SXte0evPlDNdbMKEe7m1lH5yLLP913Do/dVMSThHoD4rr1CQYGkUKrf06vYMhIeVuikuB1+2uRlxwjWcs9O8mIWSNyYdFr8FVFHSsnTYvV4u1pZbC6A0g2afBFdD/5KTG8ipZCKTfZb0a8Aa/e2g89JGreJ5k1mDUiF19VNLCJoYOyYlmpb2GKiSej5Y6R8JoWpDAYFrJtXpKJV+sdYGIRRIqcbNZApZBj/k1FSDZr8MW+WnZ8k806PMKRPxvUCnS4/OilM+B3AzJ4xxHKW8m5RSJAnEHNg5+S/q6YOlByzMqz4vDqrf2QbNZgZ3WrCPESiRYqe2tbtahO/YuTSpllIsp+E00aUQIy6RNXis6VkyebtfhiXy2mDc3GLQO74h990wtwrMmOfWdbsWLqQFhdTJXNlziSZiKdL0m1INmiQX6CEdkJBjTZvUgyadh7kzvXS1ItiDOqqfVfhOeTGafHpzOHRHPLurYl17wwxURJTj3KTzj9kv/5l1SX/G+TLV95wVxCK0o24b7heXjqy6MYmGHhWZkkuY9YZlIWLPFKZo/Ig62bGvJzRubh0bVHWJxKk82Pww12rD8gBvfNGZmHR9cdwS1l4mqCte0e7DjdhnAEWL6lSxIpBdsjHg23bn2KyYdJAzKodculLDwSE5h/YwGyEsxYtuUU1lGsRRIT4Mq3uRgUbj0NYdVKAnl0e/0UqGMxlkaVegywUGwZr95xBrvOdGBR1EreV2vDgrFFrHV7/6i8br1Qcq4KORAJh3nblmVaRP1dFJUt+4MRioKPgYfuq7VBq5SzElaaJSzsAwCqZ5SfTMeVtDl8cPlCVJgoyYYn3pqUp1Tb7uHXqR9fjMN1bTBoNaK5vIgDUV08SSwhJ/cFwMjJD9Vb8dw3J1CWacEUwRxdMK4YSzaegM0bwJ853jq5pq1OxhjrdPuRatFh+dZTqGpxshUia9s9eHh0Pq8PZK4/eH0+1n5bh7+MyGNl9UySphh22S89BkqlHPPGFLHew+7TrezzQQqhJEw4FX7+JbEToWz5P9muvGAuodXbPGzNbYVCzkuUEnoPUoiJAT1j2ATDRRNK2DoiW0824t3pg9Du9EMWhV2SBwg3wTA11sAmo8nASEeJVyJ1zIGZseh0BXiYeqk1dmKVAl0Wq0Wv4VljwuPQjknWsFNj9DwvIByJ4MFRebDo1NBplDz4ojfAxHaWTCpFi8MHi06J93acYcfhzSgapLbdBb1aie1VTfhtaTpkkOHJL4+IEvamDc3GkBzGO9te1cRauzEGFT7afQa902KwtaoNb0RRK01WL3olGth68/lJJp4aSeiFAozFOjAzFh5/CL5giMXfD85OFK3dE2RKQYoYn/PUP46xhcxSOR5Xd9h8bqIlQZAkmTUwapRIj9WzMQ3htekZq0coEsGj6w6zY5afbMKanTWsxyXEAXkDXaUVSCIg8cZ0aiU2VNTh7qtz0eLwIUanxPszytEcxfSs2t7lgXeXyEr+X5JmYWOIoj58dQwvRhN+hd8t+MexLjCmTIZGmwfDCpIwrCAJr285xWL/vUF6srIvGEZtuwevbznFYlp0KgU+2n2GmTvuAOINakQiYXx/shlpMXqMKkjCN1HvQadS8JRtNA9WmHBKPhNJs9sfQk2r878qefLntCsvmEtoXMVGfYeblyiVm2jkTVgpfHhVC5OER6xZKez77BEM7JLEa0g8QaeSs8mes0bk8rwS2jG51iOxALGnVnKNnWuVEovVpOGrhrhYeILMkVrDbrF7qV5Ap8ePBpuHJ9slXhU3xrFgbDHaXYz3RtAgRN1D4k1/vo6fYEisfvIQJ+PLtXaJYk6Imud6MLKLAHned514vx/sru02rubx0eMUpJAZN3Z2IcwQGSOb24eNRxoxpk8q/vQP5qX4zp0DRd7EgrHF6HT74PTx4zPzbyzAiMKUCxar43rgj649wsbvpg7JxJ0SYFeuB05LZBXGotoc3m6L9p2M5vTQvqvrcKPV6UM4DNGck8ul7xPunK1t98DtD+GqnAT8dLYdWQlmNjYpPLdnby7BzaVpyE40YtfpNrZP20624N5huXxPLloGgFw34lVdLsnyf0O78oK5hMYt4JMtqKsuhBly68Bz16kXjC3Gksml2FBRx1rYNCCkMF6THqfHrBG56JNukfQehMck8QQhcuS96YMgk8mo+RRcpRixWNNi+fEmIT4DAOaMzENWvAF6jQJPfXmU9TqSzFqRpblsyym8c2cZgmE+EJBmrXO9N6G6hyzfCGM7wv2M7ZtGVXQtmVwq2jbBpL0gyHNoTgLCEXR73aR+G4lIFzLTRtVS3NiZNxiibjskOx7hSFcM5vmJffHHa3N4XrRKIWfRMKRyJsnd6hGjx7ItXdewINXMi41IecMFySa8P6Mcj6ytYK8xdwxpMTnuNTzcYIf6wDmsnlGONmeXpzFpYDovrtZdHyIRICwxviqlAmkxeuqcI6gjcp98OnMIWhw+HGmwidA2ejVDD5FB1u25PbH+KPKSjCjtGcvzzq7JTxLNuae+PMoDWGbE6jEgIxatDh8vh84b+GXxmP+GdkWmfAmNC5qzugN4/KYizB7JSA6bbB4WhAkwk/S+63Lx0qYTWL6lGu/sqMHMYTmoanGgqtmBJpufvRheP92i5arI6tqc7N8fHl0ArYoO3Ly1LIMFTTZZvVQAYJvTB5sn0C2gkNuHRqubt61C3oU/X76lGmt21cLpC8HhC+Jogw13De3FbitVFqDF7oMtqmTj7pe2rccfFPWPa9k3UPrH3Y+UFxCJRDAwIwb5SUZWPuoNdKmraCDPReOL8dne2m5BnqTK59PjxDDDDYcbGCte8N3zv+uDFruHlbD6gxG0On3wBEJ47nd9RPshniaBXTbbvGhz+BCrV7Pn0mT3YOpVWahucaCu04PTLQ786bpcxOlVaHf58VS0P8u3VONIPR+iSjv3B0bl47lvKlHT6sCTY3tjzV2D8Oqt/VDSw8z+VrJMc9QD16rkmNAvHcGopaBWybF611n2HvnTtbmwunw8JBFtjtL698yEEuhVcoQjEbw4uRR908y8PviDDOpozshcLJpQDKNGBYWcMRyFyJxAtH9c5ahMBt74zhqRi/wkI+zeIHadboMvGMJjYwqhVcklx4ELsFQq5UxNGYmcr9p2F3adbkNNqxPh8P8b+fFXPJhLaFzFhsMbgN0b4CWqvXZrKVbPGIRmuw+xehXksjCemVACbyCMUDjCS0xcMK4YL0bjD3NG5krGa16Z0g8RRPDgZ13LPfePymeT+xQy4PXb+qOyyYHsBANe5ChppCxlhUyG4+dtWH+oC8ZHvB2hemZAzxgYtUo2LkEUUmS/tCD0g9fn44O7y/GvqjaUZcVS++APhXC+3cvrw1U58VSvqkcMk1EuBdF0+kI8sCC3f8JtuZ+rmp14Z0cNFowrxltRpA/3WnBBnkUpZsQZ1Fi7vxbrDjV2u9+CZBPk/dJg1ikxZ2QeXH4mQXI1J8YRCIbwypR+8IfC6BGjxeF6G5Zu6iICzBmZB5VCjpkf7EdmvA5v3jEADl8AMpkca6KZ5NxjNtm96J8Rw1tqyYzXYdbwPHZMyzItSIv5/9o79+imqnyPf5OmeTVt0leaQmmhLYXa8lCKA3MryquUAbFLuKKLi1zQES+DMNQHUq66RMGrCxnldi4LBhkERy+LgduOgEyl1QGHkYe8oRTTWl5tU9qmafN+3j/SczgnOSkwNAZnfp//kuzs8zu/vZO999m//f0peBv3b5fkw2R1QRuQXpeZ5f/miZE439wFsQiQ96SWjJFHocPsZOsJ7L+C+z49K/BEZTSkEjGe3XacteGtx/IhlYgQJRL3yMjcDHEvmzoU62ffD6fXi0sGvkDojuNXWBmcBGU0pBIRFn92kvdoDkcvs3tGXTYXymv8j1cHJCixcPsJ1k/rnhgJfasZDrf/PFhxvj/3VErczdW7ShYV5N/nH87GLzn3Ujo5B0snDkZavFLQD0LnU0KlNj55tZOV7fmpPDIjqZge/l7pBKfTgzNNJrSY7LzMk4C/U2yaOwpHG40A/LlN/m3MIIhFIvYHxcCVkVFEixErk/C0nZZNysHWw414/IE0nvSGUCY/ebRfBiNQuoKJwuFFVz2Wj8SYaDjdXrg8HuhvWOH1AVqVFPExUlY6JE4WhRxdHNrMDqSq5bhutOL1P13gDXI/tFkwWCtsz/YFD+Iv37dhQk4yjjQG62qNGhiPpZ+dwrJJ2VBKo2FxuDEwWYkTlzt5ZV+bfh9S4mSQSaLg9Xnx3eVOeH2AShqFFLUcK3afRbxSiv94OBNtFif72aAkJc43dweVDXzm3myys/5jora4dTESKh6vD/3jlei0Olh741VSmKwuvFZ5LqjdhOr91fgstHY7WJtydCp0WFzQxckF+8fbJcPQZLRBrYxGa5cdSbFy1LV0ITM5Buu+vMT7E06OlcPn86HO0M1KuvxqfDab0dLmcCM9UcnLUsm01XuzRuD9qosomzoULg9gcbgRI5fA6nCh3eLCu/vreP2sMDsxKNsl0+8E92AezcN9ulg89j+HUf7U/exZocD++9uv9LxsrUzm1LljMnHD7IBG4T/czNz3f07LRWKMDF02F9ITlbwsm9x7e+WPp1E2dShydLG43OYPolhffSkoBPuloqHQt3ZjWJoaE4akQCwW8VIfLx6fzZN7CZWV8rlxmdh5/BorR3OrfRUhPTFmX4r72I6RfwmVRbM3SCrmJ4DT6UHFmSY2kixwlmR3eXHmmv/QYEGGGv9akI7nejahuZ1QKER42aQcrH/yfpy5buLNFmMDDmwGhlgCnEdZPTH9G+Y8gNZuB/qpFbA6nFg7awTsbg9S1XL85ss6VieJK1dTkKHG7NF+6RDmT4JJhcz8SZRNHYpWsxMx0ii0m53YdLAhZDjmkR86UF6jx4B4YbmVtHgFklVSeHwi3uHEncf90jYyiZgnWMhsqHNXjC8VDcHv5o6C1eVBh+XmatJfdnBQ2aUTByMhRoorHbagUGOZ5ObTY5vLGySholPLECUGz155tBjvzhyGV6YMQYfVxWs3brsAQLJKColYzKuX+ROZ87N0wf7BnRkvnTgYG/b4s5a+OSMPz/zLILYtbC4vlnH+nJjBMyVO2qucEWsjfPhg9khcMpiDNNk0Cgmv7KAkJdrMTl49XDkjs90NlSwKHzwxkhW3tDpcaOn2q4x7Ifw4SCYRs4cwuQeFFz2SjV9uv+kHRjk8R6tCa7dT8L657erz+bB04mDYXF5WEJUJ+gg8KHq904qNBxuw7omRrG1isQgaZTSeG5eJZJXsth69pmkUmPfzDCSqpDwZIakk9CAglYjYskNTYrF6Xy1vtW53+UOYBybG3Ja4ZaSgPZi74EyTiR1cgJsbmM+OywLgV+F9ICMe7z4+DC8WDcXOHikWZhOaQWgz+zcHLsHu9qK8Ro/11Xq8s/8iyn6Ri/tS43DqajvemzUC7z4+DKMy4nl1AfxwR6PVL2J41WiDDz6s2V+HxZ+dhFQShWc+Ps4qONtdfLmap3+eyc7EBTdqPz8PTYwM5TV6mB0erK2q4z0SCbSHyeyZqpYLytUkqmR4ccoQnuyN1we2rMPt3+hkZqTcTX3GprVVdTDa3IiTR/Pq8Zc9F1TW7PDgqtGGj75pCHrUNjAphm0bJg0Bt21EIr9kSaBMz/JdZ5EUK+e1GyO3wm2XZ8dlBcnVfFjtl45hMkL21j+Ysky7ZfTYmxavDLKXkaTJSIzpVc6IsTFRJYPT7QuWRPn8PDKSVLyyqWq5oKSL0eqEQhKFhBgpFn5yAs//4QSW7zqLxZ+dxOufX0C8QoryGj00CmEJoIFJMXh2XBbPXiGZlrf2XEB/jRIKqQQb/qIXvG9uvRplNMwOD89HoYI++mmUsLv8/a6x3Z/evbHdgsWf+uVfrnbaBG0PfH25w4Zuuwcr/+8cKxuzvlqPt/ZcwNnrnUH7KtxrlNfocdHQze4JcevVxsrR2G5hBxfGdq69kYYGmLugJYSwnM3pZlcsC7Yew/LdZzF/6zHMfCAdw/vH3XITmqmnsc3Ce11n6MZrlefYHPHLd5/Fq7vPBGXCWzrRv/HJrKherzyH8ho9ntv+HWYXpCNVLQ8ZGsvMsLmf32qjlmt/qIyWTPbL1yrPCWZ9/K99tfjuspF3HW5dMgn/mXQom660W9hQ7luVjRIjZCbSlk4rAITML3+jy46mTlvI9g/0KbNCZLJJhpJmEYmCgxRC2cCssuwuLwwme1B20cB6AyVyhNrqzRl5+N1f6tEYIhijrUfihfHT2esmOD1uwcyZTq8b34cIfGiz+Ov53mAW7C9NnVbYAoJdAvsAU1edoRvztx5j+zb3s8Dw7cvtltsO+mB+f8xqAeAfTQj03+enr+Ptkvyge9l94lrQNZhV6exN3+Kp3x3BL9Yfwv7zLfB6fUGClULtxMi/9CZueS9Aj8jugtQQm3Fp8Qq8MDEnSAKDCc+8ZOi+rU1ohzs4HDNwFseEsDLqyoy68MxRachJiQ06uMiEzYba8Gdm2LeTgTOlR0STa39gRssrRhsvcx835LbV7OBlyhw3RCu4sbxx7ihIQ2QLDHxdMDABPp9w/vPA19naWL8wp9UZlImUEe4MDANnvpvXXx3yOoFijA8OTMC47CS89+dadu+sf4C0ENf/gUEKoWxgVlny6JshzaFCeR8cGA9ZdFRI/7aY7GzY8s8yk0MegEzqkXjhZmT0ekWorm1mDx9qlP6Dq3PGDEJ8jPDhzhiZ316jzSWY6fGN6XmQRIl43w0lLcRkFxU6cMpkiWXubdEjgzFYGysYPBLq98fdjOf6hdvXh/VXIytZxYYaMwctudIx3GsIrUqZUOTATX5uGDUT0szss4QKCIiUuGUgtIK5C4b1U2PVY/wZy5sz8rD5YEPI/Nx2pxu7vruGJ0en46NvGlBeo8faqouCOeKZ2W5vMyEAbFjqBwf84n0fVn+P8ho99K3dgmHJUWIIhsZyZ9gfH25gbRKa5TMHQxn7uXnKjVYnFNH+jJYffdOA2QXpvHDny+02HG3swMWWbizuOQsBCM/Unh+Xjff/XIdOizNotiiUT/7UVSPe2Vd7y9zuq2b4JWg+OPA9PD6gvOeRHWMvE9pd29wlOMPutDrxvkC7CYkxXmg24WhjB45fNrGPBi80BdfLrDw/P30dz4/LZvtHXQgbmjqt7Mx888H6kD5cMmEwGtutuNxuDlppzC5Ix6u7zrLK0ccvmyAS3czkGeizrX+t5/lp94lr+P03P2BibioWbv8OS//3FBZu/w6TclPxh29/CFqNMfY0tJrZtvnV+MHsvX70TQOefzgbHxy4hPXVl3i/r8DweO7vgtu3b9qbj3VVF7F811m88sfTmJirw8qKc/5249TL5KcJ9RSAKxbp9Hh4NjDZW+OV0bxQ4zGZSRjWX4PlxbmCv6FQTy2YfRXm+ANjz/LiXAzrr2FDmpn9FaGykRS3DISiyHroiygyjTIaSpkYVzt6EoNtCU4StW3BgzB0OZCsksLqdOOG2QmdWo5DdS14aIiOnQEeqmtB8bA0tJsdON98M4vi4gnZvKx/3Hpbux2QRolxpcOCTps7ZIQZY0OqWgaH24s2sxPJsTJs/Ws9Bqdo2JnkjS4LSh5Ix41uB1LiZLA43LC5vNDGynDmShu0mljYnG6kJyjxftVFNgpHLhFDGiVCqkYJAFhbFZxYa9uCB2F3eYISSmUkKvDuzBG40SOMyeiH/fdT92MtJzkZk8nzhYk5bI57Jrrut1/pMbx/HJ4blwWRCEiOlaHq3HXWv8mxMlScuILkuBhWPBIAEpRSKGUSnr2M+GVglsJ3Z46Aw+3BqcY2/HxwClq7HUiOlWHzIT3Ph3vO+AfCqCgRb0UbWC/jM7VCiitGG4403MDLU3LR0uXvSy//8XSQDW8/NgyyaDHe2y+cDZU50MuNOrzU0ok5Ywah0+qPVnulR2uM2zbcCK6XpuSy0UkHzl/H8PQkVpKIm9mz6L4kLCjMhqHLzvbfKfn94XB7sbLibJDt65+8n52Np6kVqDV0ocVkhy5ODk1MNJpN9qDPklQylO48henD+yM9QcHLnMrYvn0BI0mjQL4uDk3ddr/IqkqOKDF6xC0DrqmWIzclDtdMNsGy3KishhtmzN96NOh+fv/vDwoegmQivFq7/ffm8QI3zHYooiVsNlyu7wMjw25HsPJOyjJQRssfmbt1ODd80e7y4olRqSgYmMwXxZuRj0fzdZDJJLzIj4xEBU9Qj5mFFOfp4HR68KezzWw9TEQUt15GngIAz4beyorFoiAbGGE+rr1cOfFQ3/WHP2fwrvPGo3n47MhlmOyukPfm9fp49sqj+alsuX7J0aoEUzpzswWGstfr9fGi/YT8whXR5PpBqOyqGXn4tr4N+84beBE7gX3gTutlyhqtTqyakY8DtU2outCGovuSMCm3X5DAYsnwfpBIxEFRRFwhx0AbuG1RkKHGE6MzeG0eaAPjX24bC6dk4LdFqL7Cbf87jXLihu8K2cDUW1d3sc/+PHuz4W7up6/q+XuhAeZHpi8c7nZ7cb7ZxM6KspOUON9iZmeAw3RxUCj8CbICZx3p8UpcMVoFZyF2uxtnm00wdPlXEnm6WHzfZmGvk5eqhqRnwzfQhiHJsai70S1YNtCGwFndnXy3X6wc51q6/LPHODkSVVJ2BtjbvTH2Xjda0T9eyZtJBn43VS1Hp8WFli5h+3qzl1lpMm3Bnd0mxchgd3tw3WgLms1qY+VIVcl67s3vf51GhutG4dki1/+pajli5dEwhPBvTpIK5w3dMHTZ0U8th0gkwvVOW5B9XP+2mGzQqRUY3k8Nac/Kq7e+FDgbD2yLQJskYhGu9fihN/+mxSvg9vr+rr5yN8KNoVYE3Hr78s/zVjbczf30pV/uFBpgfmTC2SnD3eH/ESAf3R7kp1tDPro1P9YAQ5v8BEEQRFigAYYgCIIICzTAEARBEGGBBhiCIAgiLNAAQxAEQYQFiiLr4dSpU5DJZJE2gyAI4ieFw+HAyJEjBT+jAYYgCIIIC/SIjCAIgggLNMAQBEEQYYEGGIIgCCIs0ABDEARBhAUaYAiCIIiwQAMMQRAEERZogOlDmpubMXfuXEydOhXTpk3Dxx9/DADo7OzE/PnzUVRUhPnz58NkMt2ipn98PB4PSkpKsHDhQgDkIyG6urqwZMkSFBcXY+rUqTh58iT5KYCtW7di2rRpmD59OkpLS+FwOMhHAFasWIGxY8di+vTp7Hu9+WXjxo2YPHkypkyZgkOHDvWZHTTA9CFRUVF49dVX8cUXX2DHjh349NNPodfrsWnTJowdOxZVVVUYO3YsNm3aFGlTI862bduQlZXFviYfBbN69Wo89NBD2L9/PyorK5GVlUV+4mAwGLBt2zbs2rULe/bsgcfjwd69e8lHAB5//HFs3ryZ914ov+j1euzduxd79+7F5s2b8eabb8Lj8fSJHTTA9CFarRZ5eXkAAJVKhczMTBgMBlRXV6OkpAQAUFJSggMHDkTQysjT0tKCr7/+GrNmzWLfIx/xMZvNOHbsGOsjqVSKuLg48lMAHo8HdrsdbrcbdrsdWq2WfARg9OjRUKvVvPdC+aW6uhrTpk2DVCrFgAEDkJGRgTNnzvSJHTTAhIlr166htrYWI0aMQHt7O7RaLQD/INTR0RFh6yLLmjVr8PLLL0Msvtn9yEd8rl69ioSEBKxYsQIlJSVYuXIlrFYr+YlDSkoKFixYgPHjx6OwsBAqlQqFhYXkoxCE8ovBYIBOp2PLpaSkwGAw9Mk1aYAJAxaLBUuWLEFZWRlUKlWkzbmn+Oqrr5CQkID8/PxIm3JP43a7ceHCBTz11FOoqKiAQqH4p3zU0xsmkwnV1dWorq7GoUOHYLPZUFlZGWmzfnIIqYWJRH2TupkGmD7G5XJhyZIlePTRR1FUVAQASExMRGtrKwCgtbUVCQkJkTQxopw4cQI1NTWYMGECSktL8e233+Kll14iHwWg0+mg0+kwYsQIAEBxcTEuXLhAfuJw+PBhpKWlISEhAdHR0SgqKsLJkyfJRyEI5RedToeWlha2nMFgYFc6dwsNMH2Iz+fDypUrkZmZifnz57PvT5gwARUVFQCAiooKTJw4MUIWRp4XX3wRBw8eRE1NDdatW4cxY8Zg7dq15KMAkpOTodPp0NDQAAD429/+hqysLPITh379+uH06dOw2Wzw+Xzko1sQyi8TJkzA3r174XQ6cfXqVTQ2NmL48OF9ck1SU+5Djh8/jjlz5iAnJ4fdXygtLcXw4cPx61//Gs3NzUhNTcWHH34IjUYTWWPvAY4cOYItW7Zg48aNMBqN5KMAamtrsXLlSrhcLgwYMADvvPMOvF4v+YnD+vXrsW/fPkgkEuTm5mL16tWwWCz/9D4qLS3F0aNHYTQakZiYiBdeeAGTJk0K6ZcNGzZg165diIqKQllZGR5++OE+sYMGGIIgCCIs0CMygiAIIizQAEMQBEGEBRpgCIIgiLBAAwxBEAQRFmiAIQiCIMICDTAEQRBEWKABhiAIgggLkkgbQBAEsGjRIrS0tMDhcODpp5/G7NmzsXPnTmzevBlarRYZGRmQSqV4/fXX0dHRgTfeeANNTU0AgLKyMowaNSrCd0AQwdAAQxD3AGvWrIFGo4HdbsesWbPwyCOPYMOGDdi9ezdiYmIwb948DB06FIA/T8y8efNQUFCApqYmPPPMM/jiiy8ifAcEEQwNMARxD7B9+3Z8+eWXAPyZUSsrKzF69GhWyqO4uBiNjY0A/CKPer2e/a7ZbIbZbCblbuKegwYYgogwR44cweHDh7Fjxw4oFArMnTsXgwYNQn19vWB5r9eLHTt2QC6X/8iWEsSdQZv8BBFhuru7oVaroVAoUF9fj1OnTsFms+HYsWMwmUxwu92oqqpiyxcWFuKTTz5hX9fW1kbCbIK4JSR2SRARxul0YtGiRTAYDBg0aBCMRiMWL16MxsZGbNmyBVqtFpmZmdBoNFi2bBk6OjqwatUq1NfXw+PxoKCgAKtWrYr0bRBEEDTAEMQ9isViQUxMDNxuNxYvXoyZM2di8uTJkTaLIG4b2oMhiHuU8vJyHD58GA6HA4WFhZg0aVKkTSKIO4JWMARBEERYoE1+giAIIizQAEMQBEGEBRpgCIIgiLBAAwxBEAQRFmiAIQiCIMLC/wNaKzxCYJWXFQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "ax = sns.scatterplot(x=temp_df.age, y=temp_df.income)\n",
    "ax.set_title('age vs income')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "From the above plot it is clear that the average income increases as the age of the user increasers and thus we can divide the people into three categories based on the age and income of different users\n",
    "- below 40\n",
    "- between 40 and 50\n",
    "- above 50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "def categorize_users(age):\n",
    "    if age <= 40:\n",
    "        return '<=40'\n",
    "    elif age > 40 and age <= 50:\n",
    "        return '40-50'\n",
    "    else: \n",
    "        return '>50'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>age_category</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>14825.0</td>\n",
       "      <td>14825.000000</td>\n",
       "      <td>14825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>84.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>58.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>&gt;50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>408.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9132</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>NaN</td>\n",
       "      <td>65404.991568</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>NaN</td>\n",
       "      <td>21598.299410</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>NaN</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>49000.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>64000.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>80000.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>NaN</td>\n",
       "      <td>120000.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            age         income age_category\n",
       "count   14825.0   14825.000000        14825\n",
       "unique     84.0            NaN            3\n",
       "top        58.0            NaN          >50\n",
       "freq      408.0            NaN         9132\n",
       "mean        NaN   65404.991568          NaN\n",
       "std         NaN   21598.299410          NaN\n",
       "min         NaN   30000.000000          NaN\n",
       "25%         NaN   49000.000000          NaN\n",
       "50%         NaN   64000.000000          NaN\n",
       "75%         NaN   80000.000000          NaN\n",
       "max         NaN  120000.000000          NaN"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_df['age_category'] = temp_df['age'].apply(categorize_users)\n",
    "temp_df.describe(include='all')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "get the average income for each group and set it to the clients with None groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "avg_incomes = pd.Series(temp_df.groupby('age_category').income.mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "def impute_income(row):\n",
    "    age = row['age']\n",
    "    income = row['income']\n",
    "    if pd.isna(income):\n",
    "        if age <= 40:\n",
    "            return avg_incomes['<=40']\n",
    "        elif age > 40 and age <= 50:\n",
    "            return avg_incomes['40-50']\n",
    "        else:\n",
    "            return avg_incomes['>50']\n",
    "    else:\n",
    "        return income"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "profile_df['income'] = profile_df.apply(impute_income, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>id</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>14825</td>\n",
       "      <td>17000.000000</td>\n",
       "      <td>17000</td>\n",
       "      <td>17000.000000</td>\n",
       "      <td>17000.000000</td>\n",
       "      <td>17000.000000</td>\n",
       "      <td>17000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>17000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>M</td>\n",
       "      <td>NaN</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>8484</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>NaN</td>\n",
       "      <td>62.531412</td>\n",
       "      <td>NaN</td>\n",
       "      <td>66069.420094</td>\n",
       "      <td>2016.635000</td>\n",
       "      <td>6.683529</td>\n",
       "      <td>15.881176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>NaN</td>\n",
       "      <td>26.738580</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20243.750515</td>\n",
       "      <td>1.177624</td>\n",
       "      <td>3.487098</td>\n",
       "      <td>8.775390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>NaN</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>2013.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>51000.000000</td>\n",
       "      <td>2016.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>8.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>58.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>68000.000000</td>\n",
       "      <td>2017.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>16.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>73.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>76000.000000</td>\n",
       "      <td>2017.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>23.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>NaN</td>\n",
       "      <td>118.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>120000.000000</td>\n",
       "      <td>2018.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>31.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       gender           age                                id         income  \\\n",
       "count   14825  17000.000000                             17000   17000.000000   \n",
       "unique      3           NaN                             17000            NaN   \n",
       "top         M           NaN  68be06ca386d4c31939f3a4f0e3dd783            NaN   \n",
       "freq     8484           NaN                                 1            NaN   \n",
       "mean      NaN     62.531412                               NaN   66069.420094   \n",
       "std       NaN     26.738580                               NaN   20243.750515   \n",
       "min       NaN     18.000000                               NaN   30000.000000   \n",
       "25%       NaN     45.000000                               NaN   51000.000000   \n",
       "50%       NaN     58.000000                               NaN   68000.000000   \n",
       "75%       NaN     73.000000                               NaN   76000.000000   \n",
       "max       NaN    118.000000                               NaN  120000.000000   \n",
       "\n",
       "        became_member_on_year  became_member_on_month  became_member_on_day  \n",
       "count            17000.000000            17000.000000          17000.000000  \n",
       "unique                    NaN                     NaN                   NaN  \n",
       "top                       NaN                     NaN                   NaN  \n",
       "freq                      NaN                     NaN                   NaN  \n",
       "mean              2016.635000                6.683529             15.881176  \n",
       "std                  1.177624                3.487098              8.775390  \n",
       "min               2013.000000                1.000000              1.000000  \n",
       "25%               2016.000000                4.000000              8.000000  \n",
       "50%               2017.000000                7.000000             16.000000  \n",
       "75%               2017.000000               10.000000             23.000000  \n",
       "max               2018.000000               12.000000             31.000000  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "profile_df.describe(include='all')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Imputing values for gender\n",
    "\n",
    "One idea to impute missing values in gender is to use predictive modelling. A Decision tree model can be used to predict the gender of the user given its age and income.\n",
    "I decided to impute missing data and remove it since we have only 17000 rows of data so losing ~2000 rows of the dataset is regarded as a great loss of information."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>id</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>F</td>\n",
       "      <td>55</td>\n",
       "      <td>0610b486422d4921ae7d2bf64640c50b</td>\n",
       "      <td>112000.000000</td>\n",
       "      <td>2017</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>38fe809add3b4fcf9315a9694bb96ff5</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2018</td>\n",
       "      <td>7</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>F</td>\n",
       "      <td>75</td>\n",
       "      <td>78afa995795e4d85b5d9ceeca43f5fef</td>\n",
       "      <td>100000.000000</td>\n",
       "      <td>2017</td>\n",
       "      <td>5</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>a03223e636434f42ac4c3df47e8bac43</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>M</td>\n",
       "      <td>68</td>\n",
       "      <td>e2127556f4f64592b11af22de27a7932</td>\n",
       "      <td>70000.000000</td>\n",
       "      <td>2018</td>\n",
       "      <td>4</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>8ec6ce2a7e7949b1bf142def7d0e0586</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>9</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>68617ca6246f4fbc85e91a2a49552598</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>M</td>\n",
       "      <td>65</td>\n",
       "      <td>389bc3fa690240e798340f5a15918d5c</td>\n",
       "      <td>53000.000000</td>\n",
       "      <td>2018</td>\n",
       "      <td>2</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>None</td>\n",
       "      <td>118</td>\n",
       "      <td>8974fc5686fe429db53ddde067b88302</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2016</td>\n",
       "      <td>11</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age                                id         income  \\\n",
       "0   None  118  68be06ca386d4c31939f3a4f0e3dd783   70598.226018   \n",
       "1      F   55  0610b486422d4921ae7d2bf64640c50b  112000.000000   \n",
       "2   None  118  38fe809add3b4fcf9315a9694bb96ff5   70598.226018   \n",
       "3      F   75  78afa995795e4d85b5d9ceeca43f5fef  100000.000000   \n",
       "4   None  118  a03223e636434f42ac4c3df47e8bac43   70598.226018   \n",
       "5      M   68  e2127556f4f64592b11af22de27a7932   70000.000000   \n",
       "6   None  118  8ec6ce2a7e7949b1bf142def7d0e0586   70598.226018   \n",
       "7   None  118  68617ca6246f4fbc85e91a2a49552598   70598.226018   \n",
       "8      M   65  389bc3fa690240e798340f5a15918d5c   53000.000000   \n",
       "9   None  118  8974fc5686fe429db53ddde067b88302   70598.226018   \n",
       "\n",
       "   became_member_on_year  became_member_on_month  became_member_on_day  \n",
       "0                   2017                       2                    12  \n",
       "1                   2017                       7                    15  \n",
       "2                   2018                       7                    12  \n",
       "3                   2017                       5                     9  \n",
       "4                   2017                       8                     4  \n",
       "5                   2018                       4                    26  \n",
       "6                   2017                       9                    25  \n",
       "7                   2017                      10                     2  \n",
       "8                   2018                       2                     9  \n",
       "9                   2016                      11                    22  "
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "profile_df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    age        income gender\n",
       "0   118  70598.226018   None\n",
       "2   118  70598.226018   None\n",
       "4   118  70598.226018   None\n",
       "6   118  70598.226018   None\n",
       "7   118  70598.226018   None\n",
       "9   118  70598.226018   None\n",
       "10  118  70598.226018   None\n",
       "11  118  70598.226018   None\n",
       "17  118  70598.226018   None\n",
       "23  118  70598.226018   None"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>55</td>\n",
       "      <td>112000.0</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>75</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>68</td>\n",
       "      <td>70000.0</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>65</td>\n",
       "      <td>53000.0</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>58</td>\n",
       "      <td>51000.0</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>61</td>\n",
       "      <td>57000.0</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>26</td>\n",
       "      <td>46000.0</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>62</td>\n",
       "      <td>71000.0</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>49</td>\n",
       "      <td>52000.0</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>57</td>\n",
       "      <td>42000.0</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    age    income gender\n",
       "1    55  112000.0      F\n",
       "3    75  100000.0      F\n",
       "5    68   70000.0      M\n",
       "8    65   53000.0      M\n",
       "12   58   51000.0      M\n",
       "13   61   57000.0      F\n",
       "14   26   46000.0      M\n",
       "15   62   71000.0      F\n",
       "16   49   52000.0      M\n",
       "18   57   42000.0      M"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "temp_profile_df_notna = profile_df.loc[profile_df['gender'].notna(), ['age', 'income', 'gender']]\n",
    "temp_profile_df_na = profile_df.loc[profile_df['gender'].isna(), ['age', 'income', 'gender']]\n",
    "display(temp_profile_df_na.head(10))\n",
    "display(temp_profile_df_notna.head(10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import tree\n",
    "y = temp_profile_df_notna.gender\n",
    "X = temp_profile_df_notna.drop('gender', axis=1)\n",
    "\n",
    "clf = tree.DecisionTreeClassifier()\n",
    "clf = clf.fit(X,y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age        income gender\n",
       "0  118  70598.226018      F\n",
       "2  118  70598.226018      F\n",
       "4  118  70598.226018      F\n",
       "6  118  70598.226018      F\n",
       "7  118  70598.226018      F"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_profile_df_na['gender'] = clf.predict(temp_profile_df_na.drop('gender', axis=1))\n",
    "temp_profile_df_na.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "profile_df.loc[profile_df['gender'].isna(), ['gender']] = temp_profile_df_na['gender']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>id</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>F</td>\n",
       "      <td>55</td>\n",
       "      <td>0610b486422d4921ae7d2bf64640c50b</td>\n",
       "      <td>112000.000000</td>\n",
       "      <td>2017</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>38fe809add3b4fcf9315a9694bb96ff5</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2018</td>\n",
       "      <td>7</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>F</td>\n",
       "      <td>75</td>\n",
       "      <td>78afa995795e4d85b5d9ceeca43f5fef</td>\n",
       "      <td>100000.000000</td>\n",
       "      <td>2017</td>\n",
       "      <td>5</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>a03223e636434f42ac4c3df47e8bac43</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age                                id         income  \\\n",
       "0      F  118  68be06ca386d4c31939f3a4f0e3dd783   70598.226018   \n",
       "1      F   55  0610b486422d4921ae7d2bf64640c50b  112000.000000   \n",
       "2      F  118  38fe809add3b4fcf9315a9694bb96ff5   70598.226018   \n",
       "3      F   75  78afa995795e4d85b5d9ceeca43f5fef  100000.000000   \n",
       "4      F  118  a03223e636434f42ac4c3df47e8bac43   70598.226018   \n",
       "\n",
       "   became_member_on_year  became_member_on_month  became_member_on_day  \n",
       "0                   2017                       2                    12  \n",
       "1                   2017                       7                    15  \n",
       "2                   2018                       7                    12  \n",
       "3                   2017                       5                     9  \n",
       "4                   2017                       8                     4  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>id</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>16995</th>\n",
       "      <td>F</td>\n",
       "      <td>45</td>\n",
       "      <td>6d5f3a774f3d4714ab0c092238f3a1d7</td>\n",
       "      <td>54000.0</td>\n",
       "      <td>2018</td>\n",
       "      <td>6</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16996</th>\n",
       "      <td>M</td>\n",
       "      <td>61</td>\n",
       "      <td>2cb4f97358b841b9a9773a7aa05a9d77</td>\n",
       "      <td>72000.0</td>\n",
       "      <td>2018</td>\n",
       "      <td>7</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16997</th>\n",
       "      <td>M</td>\n",
       "      <td>49</td>\n",
       "      <td>01d26f638c274aa0b965d24cefe3183f</td>\n",
       "      <td>73000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>1</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16998</th>\n",
       "      <td>F</td>\n",
       "      <td>83</td>\n",
       "      <td>9dc1421481194dcd9400aec7c9ae6366</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>2016</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16999</th>\n",
       "      <td>F</td>\n",
       "      <td>62</td>\n",
       "      <td>e4052622e5ba45a8b96b59aba68cf068</td>\n",
       "      <td>82000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>7</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      gender  age                                id   income  \\\n",
       "16995      F   45  6d5f3a774f3d4714ab0c092238f3a1d7  54000.0   \n",
       "16996      M   61  2cb4f97358b841b9a9773a7aa05a9d77  72000.0   \n",
       "16997      M   49  01d26f638c274aa0b965d24cefe3183f  73000.0   \n",
       "16998      F   83  9dc1421481194dcd9400aec7c9ae6366  50000.0   \n",
       "16999      F   62  e4052622e5ba45a8b96b59aba68cf068  82000.0   \n",
       "\n",
       "       became_member_on_year  became_member_on_month  became_member_on_day  \n",
       "16995                   2018                       6                     4  \n",
       "16996                   2018                       7                    13  \n",
       "16997                   2017                       1                    26  \n",
       "16998                   2016                       3                     7  \n",
       "16999                   2017                       7                    22  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>id</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>17000</td>\n",
       "      <td>17000.000000</td>\n",
       "      <td>17000</td>\n",
       "      <td>17000.000000</td>\n",
       "      <td>17000.000000</td>\n",
       "      <td>17000.000000</td>\n",
       "      <td>17000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>17000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>M</td>\n",
       "      <td>NaN</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>8484</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>NaN</td>\n",
       "      <td>62.531412</td>\n",
       "      <td>NaN</td>\n",
       "      <td>66069.420094</td>\n",
       "      <td>2016.635000</td>\n",
       "      <td>6.683529</td>\n",
       "      <td>15.881176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>NaN</td>\n",
       "      <td>26.738580</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20243.750515</td>\n",
       "      <td>1.177624</td>\n",
       "      <td>3.487098</td>\n",
       "      <td>8.775390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>NaN</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>2013.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>51000.000000</td>\n",
       "      <td>2016.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>8.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>58.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>68000.000000</td>\n",
       "      <td>2017.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>16.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>73.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>76000.000000</td>\n",
       "      <td>2017.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>23.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>NaN</td>\n",
       "      <td>118.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>120000.000000</td>\n",
       "      <td>2018.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>31.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       gender           age                                id         income  \\\n",
       "count   17000  17000.000000                             17000   17000.000000   \n",
       "unique      3           NaN                             17000            NaN   \n",
       "top         M           NaN  68be06ca386d4c31939f3a4f0e3dd783            NaN   \n",
       "freq     8484           NaN                                 1            NaN   \n",
       "mean      NaN     62.531412                               NaN   66069.420094   \n",
       "std       NaN     26.738580                               NaN   20243.750515   \n",
       "min       NaN     18.000000                               NaN   30000.000000   \n",
       "25%       NaN     45.000000                               NaN   51000.000000   \n",
       "50%       NaN     58.000000                               NaN   68000.000000   \n",
       "75%       NaN     73.000000                               NaN   76000.000000   \n",
       "max       NaN    118.000000                               NaN  120000.000000   \n",
       "\n",
       "        became_member_on_year  became_member_on_month  became_member_on_day  \n",
       "count            17000.000000            17000.000000          17000.000000  \n",
       "unique                    NaN                     NaN                   NaN  \n",
       "top                       NaN                     NaN                   NaN  \n",
       "freq                      NaN                     NaN                   NaN  \n",
       "mean              2016.635000                6.683529             15.881176  \n",
       "std                  1.177624                3.487098              8.775390  \n",
       "min               2013.000000                1.000000              1.000000  \n",
       "25%               2016.000000                4.000000              8.000000  \n",
       "50%               2017.000000                7.000000             16.000000  \n",
       "75%               2017.000000               10.000000             23.000000  \n",
       "max               2018.000000               12.000000             31.000000  "
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display(profile_df.head(5))\n",
    "display(profile_df.tail(5))\n",
    "profile_df.describe(include='all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "ax = sns.scatterplot(\n",
    "    x=profile_df.age, \n",
    "    y=profile_df.income, \n",
    "    hue=profile_df.gender, \n",
    "    palette=dict(F='#F9429E', M='#03E5CE', O='#BDBBFF')\n",
    ")\n",
    "ax.set_title('age vs income')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "it appears from the plot that there is no clear relation in the dataset between the gender and the income of the reward application users"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Todo : make a stacked barplot that maps age to gender to see which use the application more for different age categories\n",
    "temp_df = profile_df.loc[:, ['age', 'gender']]\n",
    "sns.set(style='white')\n",
    "temp_df['age_category'] = temp_df.age.apply(categorize_users)\n",
    "temp_df.drop('age', axis=1, inplace=True)\n",
    "sns.countplot(\n",
    "    data = temp_df,\n",
    "    y='age_category',\n",
    "    hue='gender',\n",
    "    palette=dict(F='#F9429E', M='#03E5CE', O='#BDBBFF'),\n",
    "    order=['<=40', '40-50', '>50']\n",
    ")\n",
    "\n",
    "plt.ylabel('Age Category')\n",
    "plt.title('Genders through different age categories')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Probably imputing the values of missing genders with all Females since they were all the same with age 118 caused the number of females to be more than the number of males for the age category >50"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 🖇Dataframe Joining"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "trans_df shape : (58941, 4)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>person</th>\n",
       "      <th>offer_id</th>\n",
       "      <th>is_offer_viewed</th>\n",
       "      <th>is_offer_completed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0009655768c64bdeb2e877511632db8f</td>\n",
       "      <td>3f207df678b143eea3cee63160fa8bed</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0009655768c64bdeb2e877511632db8f</td>\n",
       "      <td>5a8bc65990b245e5a138643cd4eb9837</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0009655768c64bdeb2e877511632db8f</td>\n",
       "      <td>f19421c1d4aa40978ebb69ca19b0e20d</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0009655768c64bdeb2e877511632db8f</td>\n",
       "      <td>fafdcd668e3743c1bb461111dcafc2a4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>00116118485d4dfda04fdbaba9a87b5c</td>\n",
       "      <td>f19421c1d4aa40978ebb69ca19b0e20d</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             person                          offer_id  \\\n",
       "1  0009655768c64bdeb2e877511632db8f  3f207df678b143eea3cee63160fa8bed   \n",
       "2  0009655768c64bdeb2e877511632db8f  5a8bc65990b245e5a138643cd4eb9837   \n",
       "3  0009655768c64bdeb2e877511632db8f  f19421c1d4aa40978ebb69ca19b0e20d   \n",
       "4  0009655768c64bdeb2e877511632db8f  fafdcd668e3743c1bb461111dcafc2a4   \n",
       "5  00116118485d4dfda04fdbaba9a87b5c  f19421c1d4aa40978ebb69ca19b0e20d   \n",
       "\n",
       "   is_offer_viewed  is_offer_completed  \n",
       "1                1                   0  \n",
       "2                1                   0  \n",
       "3                1                   1  \n",
       "4                1                   1  \n",
       "5                1                   0  "
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(f\"trans_df shape : {trans_df.shape}\")\n",
    "trans_df.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Add person information and offer information to each row in the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>person</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>F</td>\n",
       "      <td>55</td>\n",
       "      <td>0610b486422d4921ae7d2bf64640c50b</td>\n",
       "      <td>112000.000000</td>\n",
       "      <td>2017</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>38fe809add3b4fcf9315a9694bb96ff5</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2018</td>\n",
       "      <td>7</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age                            person         income  \\\n",
       "0      F  118  68be06ca386d4c31939f3a4f0e3dd783   70598.226018   \n",
       "1      F   55  0610b486422d4921ae7d2bf64640c50b  112000.000000   \n",
       "2      F  118  38fe809add3b4fcf9315a9694bb96ff5   70598.226018   \n",
       "\n",
       "   became_member_on_year  became_member_on_month  became_member_on_day  \n",
       "0                   2017                       2                    12  \n",
       "1                   2017                       7                    15  \n",
       "2                   2018                       7                    12  "
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "profile_df = profile_df.rename({\"id\": \"person\"}, axis=1)\n",
    "profile_df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset shape : (58941, 10)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>person</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "      <th>offer_id</th>\n",
       "      <th>is_offer_viewed</th>\n",
       "      <th>is_offer_completed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>2298d6c36e964ae4a3e7e9706d1fb8c2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>2906b810c7d4411798c6938adc9daaa5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>fafdcd668e3743c1bb461111dcafc2a4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>fafdcd668e3743c1bb461111dcafc2a4</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age                            person        income  \\\n",
       "0      F  118  68be06ca386d4c31939f3a4f0e3dd783  70598.226018   \n",
       "1      F  118  68be06ca386d4c31939f3a4f0e3dd783  70598.226018   \n",
       "2      F  118  68be06ca386d4c31939f3a4f0e3dd783  70598.226018   \n",
       "3      F  118  68be06ca386d4c31939f3a4f0e3dd783  70598.226018   \n",
       "4      F  118  68be06ca386d4c31939f3a4f0e3dd783  70598.226018   \n",
       "\n",
       "   became_member_on_year  became_member_on_month  became_member_on_day  \\\n",
       "0                   2017                       2                    12   \n",
       "1                   2017                       2                    12   \n",
       "2                   2017                       2                    12   \n",
       "3                   2017                       2                    12   \n",
       "4                   2017                       2                    12   \n",
       "\n",
       "                           offer_id  is_offer_viewed  is_offer_completed  \n",
       "0  0b1e1539f2cc45b7b9fa7c272da2e1d7                1                   0  \n",
       "1  2298d6c36e964ae4a3e7e9706d1fb8c2                1                   1  \n",
       "2  2906b810c7d4411798c6938adc9daaa5                1                   0  \n",
       "3  fafdcd668e3743c1bb461111dcafc2a4                1                   1  \n",
       "4  fafdcd668e3743c1bb461111dcafc2a4                1                   0  "
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# join person information\n",
    "dataset = pd.merge(profile_df, trans_df, on='person')\n",
    "print(f\"Dataset shape : {dataset.shape}\")\n",
    "dataset.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Portofolio df shape : (10, 9)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>offer_id</th>\n",
       "      <th>reward</th>\n",
       "      <th>difficulty</th>\n",
       "      <th>duration</th>\n",
       "      <th>offer_type</th>\n",
       "      <th>email</th>\n",
       "      <th>mobile</th>\n",
       "      <th>web</th>\n",
       "      <th>social</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ae264e3637204a6fb9bb56bc8210ddfd</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4d5c57ea9a6940dd891ad53e9dbe8da0</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>5</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3f207df678b143eea3cee63160fa8bed</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>informational</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9b98b8c7a33c4b65b9aebfe6a799e6d9</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           offer_id  reward  difficulty  duration  \\\n",
       "0  ae264e3637204a6fb9bb56bc8210ddfd      10          10         7   \n",
       "1  4d5c57ea9a6940dd891ad53e9dbe8da0      10          10         5   \n",
       "2  3f207df678b143eea3cee63160fa8bed       0           0         4   \n",
       "3  9b98b8c7a33c4b65b9aebfe6a799e6d9       5           5         7   \n",
       "4  0b1e1539f2cc45b7b9fa7c272da2e1d7       5          20        10   \n",
       "\n",
       "      offer_type  email  mobile    web  social  \n",
       "0           bogo   True    True  False    True  \n",
       "1           bogo   True    True   True    True  \n",
       "2  informational   True    True   True   False  \n",
       "3           bogo   True    True   True   False  \n",
       "4       discount   True   False   True   False  "
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "portfolio_df = portfolio_df.reset_index()\n",
    "portfolio_df = portfolio_df.rename(dict(id='offer_id'), axis=1)\n",
    "print(f\"Portofolio df shape : {portfolio_df.shape}\")\n",
    "portfolio_df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "# join offer information\n",
    "dataset = pd.merge(dataset, portfolio_df, on='offer_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset shape : (58941, 18)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>person</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "      <th>offer_id</th>\n",
       "      <th>is_offer_completed</th>\n",
       "      <th>reward</th>\n",
       "      <th>difficulty</th>\n",
       "      <th>duration</th>\n",
       "      <th>offer_type</th>\n",
       "      <th>email</th>\n",
       "      <th>mobile</th>\n",
       "      <th>web</th>\n",
       "      <th>social</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>68be06ca386d4c31939f3a4f0e3dd783</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>a03223e636434f42ac4c3df47e8bac43</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>a03223e636434f42ac4c3df47e8bac43</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>a03223e636434f42ac4c3df47e8bac43</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>F</td>\n",
       "      <td>61</td>\n",
       "      <td>aa4862eba776480b8bb9c68455b8c2e1</td>\n",
       "      <td>57000.000000</td>\n",
       "      <td>2017</td>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>F</td>\n",
       "      <td>62</td>\n",
       "      <td>31dda685af34476cad5bc968bdb01c53</td>\n",
       "      <td>71000.000000</td>\n",
       "      <td>2016</td>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>744d603ef08c4f33af5a61c8c7628d1c</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>F</td>\n",
       "      <td>40</td>\n",
       "      <td>440cf1fd7580490c971d8c651ed962af</td>\n",
       "      <td>71000.000000</td>\n",
       "      <td>2016</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>M</td>\n",
       "      <td>64</td>\n",
       "      <td>4b0da7e80e5945209a1fdddfe813dbe0</td>\n",
       "      <td>100000.000000</td>\n",
       "      <td>2017</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>2b826eba31074a059d63b0ae8f50b7d5</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>9</td>\n",
       "      <td>7</td>\n",
       "      <td>0b1e1539f2cc45b7b9fa7c272da2e1d7</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age                            person         income  \\\n",
       "0      F  118  68be06ca386d4c31939f3a4f0e3dd783   70598.226018   \n",
       "1      F  118  a03223e636434f42ac4c3df47e8bac43   70598.226018   \n",
       "2      F  118  a03223e636434f42ac4c3df47e8bac43   70598.226018   \n",
       "3      F  118  a03223e636434f42ac4c3df47e8bac43   70598.226018   \n",
       "4      F   61  aa4862eba776480b8bb9c68455b8c2e1   57000.000000   \n",
       "5      F   62  31dda685af34476cad5bc968bdb01c53   71000.000000   \n",
       "6      F  118  744d603ef08c4f33af5a61c8c7628d1c   70598.226018   \n",
       "7      F   40  440cf1fd7580490c971d8c651ed962af   71000.000000   \n",
       "8      M   64  4b0da7e80e5945209a1fdddfe813dbe0  100000.000000   \n",
       "9      F  118  2b826eba31074a059d63b0ae8f50b7d5   70598.226018   \n",
       "\n",
       "   became_member_on_year  became_member_on_month  became_member_on_day  \\\n",
       "0                   2017                       2                    12   \n",
       "1                   2017                       8                     4   \n",
       "2                   2017                       8                     4   \n",
       "3                   2017                       8                     4   \n",
       "4                   2017                       9                    11   \n",
       "5                   2016                       2                    11   \n",
       "6                   2017                       8                     1   \n",
       "7                   2016                       5                     4   \n",
       "8                   2017                       9                     9   \n",
       "9                   2017                       9                     7   \n",
       "\n",
       "                           offer_id  is_offer_completed  reward  difficulty  \\\n",
       "0  0b1e1539f2cc45b7b9fa7c272da2e1d7                   0       5          20   \n",
       "1  0b1e1539f2cc45b7b9fa7c272da2e1d7                   0       5          20   \n",
       "2  0b1e1539f2cc45b7b9fa7c272da2e1d7                   0       5          20   \n",
       "3  0b1e1539f2cc45b7b9fa7c272da2e1d7                   0       5          20   \n",
       "4  0b1e1539f2cc45b7b9fa7c272da2e1d7                   0       5          20   \n",
       "5  0b1e1539f2cc45b7b9fa7c272da2e1d7                   0       5          20   \n",
       "6  0b1e1539f2cc45b7b9fa7c272da2e1d7                   0       5          20   \n",
       "7  0b1e1539f2cc45b7b9fa7c272da2e1d7                   1       5          20   \n",
       "8  0b1e1539f2cc45b7b9fa7c272da2e1d7                   1       5          20   \n",
       "9  0b1e1539f2cc45b7b9fa7c272da2e1d7                   0       5          20   \n",
       "\n",
       "   duration offer_type  email  mobile   web  social  \n",
       "0        10   discount   True   False  True   False  \n",
       "1        10   discount   True   False  True   False  \n",
       "2        10   discount   True   False  True   False  \n",
       "3        10   discount   True   False  True   False  \n",
       "4        10   discount   True   False  True   False  \n",
       "5        10   discount   True   False  True   False  \n",
       "6        10   discount   True   False  True   False  \n",
       "7        10   discount   True   False  True   False  \n",
       "8        10   discount   True   False  True   False  \n",
       "9        10   discount   True   False  True   False  "
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(f\"Dataset shape : {dataset.shape}\")\n",
    "dataset = dataset.drop('is_offer_viewed', axis=1)\n",
    "dataset.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Checking if there are duplicate rows with the same features but different or same class labels.\n",
    "- if they have the same class label we can leave only 1 of the rows.\n",
    "- if they have  different class labels then the given features aren't enough for the target value, for simplicity, I decide to remove these columns if they are not too much."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "      <th>is_offer_completed</th>\n",
       "      <th>reward</th>\n",
       "      <th>difficulty</th>\n",
       "      <th>duration</th>\n",
       "      <th>offer_type</th>\n",
       "      <th>email</th>\n",
       "      <th>mobile</th>\n",
       "      <th>web</th>\n",
       "      <th>social</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age        income  became_member_on_year  became_member_on_month  \\\n",
       "0      F  118  70598.226018                   2017                       2   \n",
       "1      F  118  70598.226018                   2017                       8   \n",
       "\n",
       "   became_member_on_day  is_offer_completed  reward  difficulty  duration  \\\n",
       "0                    12                   0       5          20        10   \n",
       "1                     4                   0       5          20        10   \n",
       "\n",
       "  offer_type  email  mobile   web  social  \n",
       "0   discount   True   False  True   False  \n",
       "1   discount   True   False  True   False  "
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset = dataset.drop(['person', 'offer_id'], axis=1)\n",
    "dataset.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset Shape before dropping duplicates : (58941, 15)\n",
      "Dataset Shape after dropping duplicates : (48270, 15)\n"
     ]
    }
   ],
   "source": [
    "print(f'Dataset Shape before dropping duplicates : {dataset.shape}')\n",
    "dataset = dataset.drop_duplicates(keep='first')\n",
    "print(f'Dataset Shape after dropping duplicates : {dataset.shape}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Make sure that no two rows with the same features have the different labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "features = dataset.drop(['is_offer_completed'], axis=1).columns.to_list()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "      <th>reward</th>\n",
       "      <th>difficulty</th>\n",
       "      <th>duration</th>\n",
       "      <th>offer_type</th>\n",
       "      <th>email</th>\n",
       "      <th>mobile</th>\n",
       "      <th>web</th>\n",
       "      <th>social</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>131</th>\n",
       "      <td>F</td>\n",
       "      <td>20</td>\n",
       "      <td>35000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>145</th>\n",
       "      <td>F</td>\n",
       "      <td>20</td>\n",
       "      <td>46000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>10</td>\n",
       "      <td>11</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>388</th>\n",
       "      <td>F</td>\n",
       "      <td>22</td>\n",
       "      <td>45000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>11</td>\n",
       "      <td>15</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>457</th>\n",
       "      <td>F</td>\n",
       "      <td>22</td>\n",
       "      <td>73000.0</td>\n",
       "      <td>2014</td>\n",
       "      <td>6</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>550</th>\n",
       "      <td>F</td>\n",
       "      <td>24</td>\n",
       "      <td>37000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>5</td>\n",
       "      <td>31</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46982</th>\n",
       "      <td>O</td>\n",
       "      <td>55</td>\n",
       "      <td>84000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>12</td>\n",
       "      <td>29</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46988</th>\n",
       "      <td>O</td>\n",
       "      <td>56</td>\n",
       "      <td>33000.0</td>\n",
       "      <td>2018</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47018</th>\n",
       "      <td>O</td>\n",
       "      <td>57</td>\n",
       "      <td>91000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>23</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>5</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47026</th>\n",
       "      <td>O</td>\n",
       "      <td>58</td>\n",
       "      <td>73000.0</td>\n",
       "      <td>2015</td>\n",
       "      <td>8</td>\n",
       "      <td>31</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47120</th>\n",
       "      <td>O</td>\n",
       "      <td>64</td>\n",
       "      <td>76000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>11</td>\n",
       "      <td>13</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>956 rows × 15 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      gender  age   income  became_member_on_year  became_member_on_month  \\\n",
       "131        F   20  35000.0                   2017                       2   \n",
       "145        F   20  46000.0                   2017                      10   \n",
       "388        F   22  45000.0                   2017                      11   \n",
       "457        F   22  73000.0                   2014                       6   \n",
       "550        F   24  37000.0                   2017                       5   \n",
       "...      ...  ...      ...                    ...                     ...   \n",
       "46982      O   55  84000.0                   2017                      12   \n",
       "46988      O   56  33000.0                   2018                       5   \n",
       "47018      O   57  91000.0                   2017                       8   \n",
       "47026      O   58  73000.0                   2015                       8   \n",
       "47120      O   64  76000.0                   2017                      11   \n",
       "\n",
       "       became_member_on_day  reward  difficulty  duration offer_type  email  \\\n",
       "131                       6      10          10         7       bogo   True   \n",
       "145                      11       5           5         5       bogo   True   \n",
       "388                      15       2          10        10   discount   True   \n",
       "457                      12      10          10         7       bogo   True   \n",
       "550                      31      10          10         7       bogo   True   \n",
       "...                     ...     ...         ...       ...        ...    ...   \n",
       "46982                    29       5           5         7       bogo   True   \n",
       "46988                     1      10          10         7       bogo   True   \n",
       "47018                    23      10          10         5       bogo   True   \n",
       "47026                    31       5          20        10   discount   True   \n",
       "47120                    13       2          10         7   discount   True   \n",
       "\n",
       "       mobile    web  social  size  \n",
       "131      True  False    True     2  \n",
       "145      True   True    True     2  \n",
       "388      True   True    True     2  \n",
       "457      True  False    True     2  \n",
       "550      True  False    True     2  \n",
       "...       ...    ...     ...   ...  \n",
       "46982    True   True   False     2  \n",
       "46988    True  False    True     2  \n",
       "47018    True   True    True     2  \n",
       "47026   False   True   False     2  \n",
       "47120    True   True   False     2  \n",
       "\n",
       "[956 rows x 15 columns]"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_df = dataset.groupby(features, as_index=False)['is_offer_completed'].size()\n",
    "temp_df.loc[temp_df['size'] == 2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are 956 rows in which the input feature aren't enough to give information about whether the offer would be completed or not, for simplicity I decide to remove these rows."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset Shape before removing rows with same features but different labels : (48270, 15)\n",
      "Dataset Shape after removing rows with same features but different labels : (46358, 15)\n"
     ]
    }
   ],
   "source": [
    "print(f'Dataset Shape before removing rows with same features but different labels : {dataset.shape}')\n",
    "dataset = dataset.drop_duplicates(subset=features, keep=False)\n",
    "print(f'Dataset Shape after removing rows with same features but different labels : {dataset.shape}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Save dataset before processing in a csv file to avoid having to process the data every time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset Shape : (46358, 15)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "      <th>is_offer_completed</th>\n",
       "      <th>reward</th>\n",
       "      <th>difficulty</th>\n",
       "      <th>duration</th>\n",
       "      <th>offer_type</th>\n",
       "      <th>email</th>\n",
       "      <th>mobile</th>\n",
       "      <th>web</th>\n",
       "      <th>social</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>F</td>\n",
       "      <td>61</td>\n",
       "      <td>57000.000000</td>\n",
       "      <td>2017</td>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>F</td>\n",
       "      <td>62</td>\n",
       "      <td>71000.000000</td>\n",
       "      <td>2016</td>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>F</td>\n",
       "      <td>40</td>\n",
       "      <td>71000.000000</td>\n",
       "      <td>2016</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>M</td>\n",
       "      <td>64</td>\n",
       "      <td>100000.000000</td>\n",
       "      <td>2017</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>9</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>F</td>\n",
       "      <td>56</td>\n",
       "      <td>88000.000000</td>\n",
       "      <td>2018</td>\n",
       "      <td>4</td>\n",
       "      <td>28</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>3</td>\n",
       "      <td>30</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age         income  became_member_on_year  became_member_on_month  \\\n",
       "0      F  118   70598.226018                   2017                       2   \n",
       "1      F  118   70598.226018                   2017                       8   \n",
       "2      F   61   57000.000000                   2017                       9   \n",
       "3      F   62   71000.000000                   2016                       2   \n",
       "4      F  118   70598.226018                   2017                       8   \n",
       "5      F   40   71000.000000                   2016                       5   \n",
       "6      M   64  100000.000000                   2017                       9   \n",
       "7      F  118   70598.226018                   2017                       9   \n",
       "8      F   56   88000.000000                   2018                       4   \n",
       "9      F  118   70598.226018                   2017                       3   \n",
       "\n",
       "   became_member_on_day  is_offer_completed  reward  difficulty  duration  \\\n",
       "0                    12                   0       5          20        10   \n",
       "1                     4                   0       5          20        10   \n",
       "2                    11                   0       5          20        10   \n",
       "3                    11                   0       5          20        10   \n",
       "4                     1                   0       5          20        10   \n",
       "5                     4                   1       5          20        10   \n",
       "6                     9                   1       5          20        10   \n",
       "7                     7                   0       5          20        10   \n",
       "8                    28                   1       5          20        10   \n",
       "9                    30                   0       5          20        10   \n",
       "\n",
       "  offer_type  email  mobile   web  social  \n",
       "0   discount   True   False  True   False  \n",
       "1   discount   True   False  True   False  \n",
       "2   discount   True   False  True   False  \n",
       "3   discount   True   False  True   False  \n",
       "4   discount   True   False  True   False  \n",
       "5   discount   True   False  True   False  \n",
       "6   discount   True   False  True   False  \n",
       "7   discount   True   False  True   False  \n",
       "8   discount   True   False  True   False  \n",
       "9   discount   True   False  True   False  "
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset = dataset.reset_index(drop=True)\n",
    "print(f'Dataset Shape : {dataset.shape}')\n",
    "dataset.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "filename = 'dataset.csv'\n",
    "dataset.to_csv(filename)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Load dataset from csv file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "      <th>is_offer_completed</th>\n",
       "      <th>reward</th>\n",
       "      <th>difficulty</th>\n",
       "      <th>duration</th>\n",
       "      <th>offer_type</th>\n",
       "      <th>email</th>\n",
       "      <th>mobile</th>\n",
       "      <th>web</th>\n",
       "      <th>social</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>F</td>\n",
       "      <td>61</td>\n",
       "      <td>57000.000000</td>\n",
       "      <td>2017</td>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>F</td>\n",
       "      <td>62</td>\n",
       "      <td>71000.000000</td>\n",
       "      <td>2016</td>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>70598.226018</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age        income  became_member_on_year  became_member_on_month  \\\n",
       "0      F  118  70598.226018                   2017                       2   \n",
       "1      F  118  70598.226018                   2017                       8   \n",
       "2      F   61  57000.000000                   2017                       9   \n",
       "3      F   62  71000.000000                   2016                       2   \n",
       "4      F  118  70598.226018                   2017                       8   \n",
       "\n",
       "   became_member_on_day  is_offer_completed  reward  difficulty  duration  \\\n",
       "0                    12                   0       5          20        10   \n",
       "1                     4                   0       5          20        10   \n",
       "2                    11                   0       5          20        10   \n",
       "3                    11                   0       5          20        10   \n",
       "4                     1                   0       5          20        10   \n",
       "\n",
       "  offer_type  email  mobile   web  social  \n",
       "0   discount   True   False  True   False  \n",
       "1   discount   True   False  True   False  \n",
       "2   discount   True   False  True   False  \n",
       "3   discount   True   False  True   False  \n",
       "4   discount   True   False  True   False  "
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset = pd.read_csv('dataset.csv', index_col=0)\n",
    "dataset.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "check for dataset imbalance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_style('whitegrid')\n",
    "dataset['is_offer_completed'].value_counts().plot(kind='bar')\n",
    "plt.title('Is offer completed')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the above plot it is apparent that the dataset is balanced before splitting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 46358 entries, 0 to 46357\n",
      "Data columns (total 15 columns):\n",
      " #   Column                  Non-Null Count  Dtype  \n",
      "---  ------                  --------------  -----  \n",
      " 0   gender                  46358 non-null  object \n",
      " 1   age                     46358 non-null  int64  \n",
      " 2   income                  46358 non-null  float64\n",
      " 3   became_member_on_year   46358 non-null  int64  \n",
      " 4   became_member_on_month  46358 non-null  int64  \n",
      " 5   became_member_on_day    46358 non-null  int64  \n",
      " 6   is_offer_completed      46358 non-null  int64  \n",
      " 7   reward                  46358 non-null  int64  \n",
      " 8   difficulty              46358 non-null  int64  \n",
      " 9   duration                46358 non-null  int64  \n",
      " 10  offer_type              46358 non-null  object \n",
      " 11  email                   46358 non-null  bool   \n",
      " 12  mobile                  46358 non-null  bool   \n",
      " 13  web                     46358 non-null  bool   \n",
      " 14  social                  46358 non-null  bool   \n",
      "dtypes: bool(4), float64(1), int64(8), object(2)\n",
      "memory usage: 4.4+ MB\n"
     ]
    }
   ],
   "source": [
    "dataset.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Change datatypes for efficiency"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "dtypes_dict = {\n",
    "    'gender' : 'category',\n",
    "    'age' : 'int8',\n",
    "    'income' : 'float32',\n",
    "    'became_member_on_year': 'int16',\n",
    "    'became_member_on_month' : 'int8',\n",
    "    'became_member_on_day' : 'int8',\n",
    "    'is_offer_completed' : 'int8',\n",
    "    'reward' : 'int16',\n",
    "    'difficulty' : 'int16',\n",
    "    'duration' : 'int16',\n",
    "    'offer_type' : 'category',\n",
    "    'email' : 'bool',\n",
    "    'mobile' : 'bool',\n",
    "    'web' : 'bool',\n",
    "    'social' : 'bool'\n",
    "\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 46358 entries, 0 to 46357\n",
      "Data columns (total 15 columns):\n",
      " #   Column                  Non-Null Count  Dtype   \n",
      "---  ------                  --------------  -----   \n",
      " 0   gender                  46358 non-null  category\n",
      " 1   age                     46358 non-null  int8    \n",
      " 2   income                  46358 non-null  float32 \n",
      " 3   became_member_on_year   46358 non-null  int16   \n",
      " 4   became_member_on_month  46358 non-null  int8    \n",
      " 5   became_member_on_day    46358 non-null  int8    \n",
      " 6   is_offer_completed      46358 non-null  int8    \n",
      " 7   reward                  46358 non-null  int16   \n",
      " 8   difficulty              46358 non-null  int16   \n",
      " 9   duration                46358 non-null  int16   \n",
      " 10  offer_type              46358 non-null  category\n",
      " 11  email                   46358 non-null  bool    \n",
      " 12  mobile                  46358 non-null  bool    \n",
      " 13  web                     46358 non-null  bool    \n",
      " 14  social                  46358 non-null  bool    \n",
      "dtypes: bool(4), category(2), float32(1), int16(4), int8(4)\n",
      "memory usage: 1.3 MB\n"
     ]
    }
   ],
   "source": [
    "dataset = dataset.astype(dtypes_dict)\n",
    "dataset.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "      <th>is_offer_completed</th>\n",
       "      <th>reward</th>\n",
       "      <th>difficulty</th>\n",
       "      <th>duration</th>\n",
       "      <th>offer_type</th>\n",
       "      <th>email</th>\n",
       "      <th>mobile</th>\n",
       "      <th>web</th>\n",
       "      <th>social</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>70598.226562</td>\n",
       "      <td>2017</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>70598.226562</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age        income  became_member_on_year  became_member_on_month  \\\n",
       "0      F  118  70598.226562                   2017                       2   \n",
       "1      F  118  70598.226562                   2017                       8   \n",
       "\n",
       "   became_member_on_day  is_offer_completed  reward  difficulty  duration  \\\n",
       "0                    12                   0       5          20        10   \n",
       "1                     4                   0       5          20        10   \n",
       "\n",
       "  offer_type  email  mobile   web  social  \n",
       "0   discount   True   False  True   False  \n",
       "1   discount   True   False  True   False  "
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 💪 Training"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dataset Splitting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = dataset['is_offer_completed']\n",
    "X = dataset.drop(['is_offer_completed'], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since the dataset is not too large we do a 90% train, 10% test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, stratify=y, test_size=0.1, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "      <th>reward</th>\n",
       "      <th>difficulty</th>\n",
       "      <th>duration</th>\n",
       "      <th>offer_type</th>\n",
       "      <th>email</th>\n",
       "      <th>mobile</th>\n",
       "      <th>web</th>\n",
       "      <th>social</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>M</td>\n",
       "      <td>59</td>\n",
       "      <td>74000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>M</td>\n",
       "      <td>55</td>\n",
       "      <td>42000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>26</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>F</td>\n",
       "      <td>77</td>\n",
       "      <td>92000.0</td>\n",
       "      <td>2016</td>\n",
       "      <td>3</td>\n",
       "      <td>22</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age   income  became_member_on_year  became_member_on_month  \\\n",
       "0      M   59  74000.0                   2017                       4   \n",
       "1      M   55  42000.0                   2017                       8   \n",
       "2      F   77  92000.0                   2016                       3   \n",
       "\n",
       "   became_member_on_day  reward  difficulty  duration offer_type  email  \\\n",
       "0                     3       5          20        10   discount   True   \n",
       "1                    26       2          10        10   discount   True   \n",
       "2                    22      10          10         7       bogo   True   \n",
       "\n",
       "   mobile    web  social  \n",
       "0   False   True   False  \n",
       "1    True   True    True  \n",
       "2    True  False    True  "
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train = X_train.reset_index(level=0, drop=True)\n",
    "X_train.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAEICAYAAACqMQjAAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8/fFQqAAAACXBIWXMAAAsTAAALEwEAmpwYAAAklUlEQVR4nO3df1RUdf4/8CeMjoVF41DgVVw5UtiUx8RBqZT4OGikDlJrG5xR++mWHqOza5TTL4Yl6TRl4tLxR7blbrvYDzVlGTHsZGWbpS2e2tjRSENNGGaSH4WSgHfe3z/8eleEyyAMDODz8Zdz3/fH6868vc+57zuXGySEECAiImpHcKALICKivoshQUREqhgSRESkiiFBRESqGBJERKSKIUFERKoYEn2AyWTCnj172m3797//jeTk5F6uqG8RQuCpp57CpEmTcPfddwe6nHbt3bsXt912W69s6/jx4xg7dizOnDlz0cuePn0aixYtgtFoxGOPPQYAyMvLQ3x8PKZMmeLvUts4dOgQfvvb3/p1nVlZWVi9erVf13mxDh48iPT09IDW0FMGBboA6lhcXBxKSkp8zvfqq6/i6NGjWLFiRS9U1btKS0vx+eef49NPP0VISEigy+m2QH5WH3zwAU6cOIG9e/di0KBBcLlc2LBhAz7++GOEhYX1+Pb//Oc/46GHHlJem0wmLF++HLfeemuX15mTk+OP0i7K2LFjsXPnTowePRoAcP311+PKK6/Erl27YDKZer2ensQzCfKpK99Y/amyshIjR44cEAERaFVVVYiKisKgQWe/H1ZWVkKn03UpIIQQ8Hq9nZ7f4/Fg7969mD59eqeXCXTfuxgpKSl49913A12G3zEk+ogDBw4gJSUFRqMRf/jDH9DU1ASg7TDG+vXrkZCQgNjYWCQnJ+OLL77A7t278dprr2HHjh2IjY3FnDlzAAButxuLFi3C5MmTMWPGDLz33nvKek6fPo1ly5Zh0qRJmDlzJl5//fVW2zGZTFi/fj1SUlIwYcIEnDlzBuvXr8f06dMRGxuLWbNm4cMPP1Tmf//995Geno4XXngBcXFxSEpKwv79+/H+++8jMTERt9xyC7Zu3aq6/2q1btq0Cc8++yy+/vprxMbGIj8/v93l33vvPcycOVOp7b///S8A4PDhw1iwYAHi4uIwe/ZsfPTRR8oyVqsV2dnZWLhwIWJjY5Geno6ffvoJubm5mDRpEu644w44nc5W78lrr72GWbNmYdKkSXjqqaeUz6m9/cnIyMDNN98Mk8mEt956CwBUP6uGhgY8/fTTmDp1KhISEpCXlwdZlgEAsizDbrcjPj4eSUlJ+PTTT1Xfx472OT8/H2vWrFG2/c477+DBBx+Ex+NBbGwsrFYrAODrr79Geno64uLiMGfOHOzdu1dZ94IFC5CXl4f09HTcdNNN+PHHH/H+++8jKSkJsbGxMJlM+Oc//9luXXv27MENN9yAIUOGAACeeOIJVFVVYdGiRYiNjcXrr7+uDKVt2rQJ//d//4f77rsPAPDYY49hypQpMBqNmDdvHr7//vtWn2NeXh6A//1/efPNN3HLLbdg6tSp2LJli+p71VHtmzdvxsyZMzFp0iQ89NBDqKysBADMmzcPAJCamorY2FgUFxcDAOLj4/HFF1+gubm5w8+n3xEUcNOmTRNz584V1dXVoq6uTtxxxx1i48aNQgghvvzyS5GQkCCEEOLw4cPitttuE9XV1UIIIX788Udx9OhRIYQQ+fn54vHHH2+13nnz5gmbzSZOnz4tnE6niI+PF3v27BFCCPHyyy+LefPmifr6euFyuYTZbFa2c66mOXPmiKqqKvHrr78KIYQoLi4W1dXVQpZlsX37dnHTTTcJt9sthBBiy5YtwmAwiM2bN4szZ86IlStXisTERJGdnS2amprEZ599JiZMmCBOnjzZ7nvQUa1btmwR6enpqu9fcXGxmDp1qvjmm2+E1+sVR44cEcePHxfNzc1i+vTpYu3ataKpqUns2bNHTJgwQRw+fFgIIcSyZcvE5MmTxbfffitOnz4tFixYIKZNmya2bt2q7MP8+fNbvSezZ88WVVVVoq6uTqSlpYmVK1e2+ZxkWRZ33XWXePXVV0VTU5M4duyYMJlMYvfu3aqf1eLFi8Vzzz0nTp06JU6cOCHmzp0r3n77bSGEEBs3bhTJycnKdufPny9iYmJES0tLm/fC1z5fuO3z6xZCiOrqajF58mTxySefCFmWxb/+9S8xefJkUVNTI4QQYv78+SIxMVGUl5eLlpYW8csvv4jY2Fhl/W63W5SXl7f7Ob344osiOzu71bRp06aJzz//XHn9448/ipiYGPHEE0+IU6dOKX1v06ZNoqGhQTQ1NYnly5eLOXPmKMssW7as1edgMBjEqlWrRHNzs/jkk0/E+PHjRX19fZt6Tp06pVr7hx9+KKZPny4OHTokWlpaxOrVq0VaWpqybExMjDhy5EibdcbGxooDBw60u//9Fc8k+ogFCxYgIiICOp0O06ZNw4EDB9rMo9Fo0NzcjMOHD6OlpQWRkZH4zW9+0+76XC4XSktLkZmZiSFDhsBgMOB3v/sdCgsLAQA7duzAI488gquuugrDhw/Hvffe225NkiThsssuAwDMnDkTERERCA4OxqxZszB69Gj85z//UeaPjIzE3LlzodFoMGvWLLhcLixZsgRarRZTp06FVqvFsWPHLrpWXzZv3oyFCxdi/PjxCAoKwujRozFy5Eh88803aGxsxMMPPwytVotbbrkF06ZNw/bt25VlZ8yYgXHjxmHIkCGYMWMGhgwZgjvvvFPZhws/h3nz5kGSJOh0OixevLjVus759ttvUVtbi0cffRRarRajRo3CPffco3zjvNCJEyewe/duPP300wgJCUFYWBjuv/9+Zd07duzAfffdp2z3kUceUX0vOrPPHSksLMRtt92GxMREBAcHY8qUKRg3blyrs5e77roL1113HQYNGgSNRoPg4GB8//33OH36NMLDw3Hddde1u+6GhgYMHTq0U3VkZGQgJCRE6Xt33303rrjiCmi1WmRkZODgwYNoaGhod9lBgwZhyZIlGDx4MBITExESEoKKiop251Wr/Z133sHDDz+M6OhoDBo0CIsWLcKBAweUswk1Q4cOVa2rv+KF6z7immuuUf59+eWXw+PxtJln9OjRePrpp/Hqq6/i0KFDmDp1KqxWKyIiItrM6/F4cNVVV+GKK65Qpo0YMQJlZWVKuyRJStvw4cPbrOP8dgDYtm0bNmzYoPxHaWxsRF1dndJ+/rj2uf/cV199tTJtyJAhOHXq1EXX6ovL5Wo3LD0eD4YPH47g4P99FxoxYgTcbrdqzefXe9lll6GxsbHVOs9/T0aMGNHu51RZWQmPx4O4uDhlmizLrV6fr6qqCmfOnMHUqVOVaV6vV9nWhZ/ViBEj2l1PZ/e5I1VVVfjggw/w8ccfK9POnDmD+Ph45fX5tYSEhCAvLw9vvvkmnnnmGUycOBHLli1DdHR0m3WHhoa2+/m35/z+KMsy8vLy8MEHH6C2tlbZt7q6Olx55ZVtltXpdMo1F+Ds/6cLP0dftVdVVeGFF16A3W5X5hdCwO12Y+TIkap1nzp1qt2a+jOGRD+TkpKClJQUnDx5EllZWVixYgVefvllBAUFtZovPDwcP//8M06ePKkcfF0ulxIo11xzDaqrq3HttdcCAKqrq9ts6/x1VlZW4tlnn8Vf//pXxMbGQqPRIDU11S/75KtWXyRJavcMJTw8HNXV1fB6vcqBxeVyISoqqsu1ulwu5d9VVVUIDw9vt57IyEjs3Lmz3XVc+FkNHz4cWq0WX375ZauD2znXXHNNq+2e/+8LdXefJUlCamoqli9frjrPhfUnJCQgISEBp0+fxqpVq/Dcc89h48aNbZYbO3Ystm3b1qk6zt9GUVERPvroI2zYsAGRkZFoaGjApEmTIPzwB6zVapckCYsWLVKuGXWG2+1GS0sLxowZ0+26+hION/UjP/zwg3JhTKvVYsiQIdBoNADOfiOurKxUfm0iSRJiY2OxcuVKNDU14eDBg9i8eTNSUlIAnB06eu211/Dzzz/D7XbjH//4R4fb/vXXXxEUFAS9Xg8A2LJlS6uLh93hq1Zf7r77brz55psoKyuDEAJHjx5FZWUlxo8fj8svvxx/+ctf0NLSgr1792LXrl2YNWtWl2vduHEjqqurUV9fr1zEvtD48eNxxRVXYP369Th9+jRkWUZ5ebkyNHfhZxUeHo4pU6bgxRdfxMmTJ+H1enHs2DHs27cPwNnP6u9//zuqq6vx888/Y/369ar1dXef58yZg48//hifffYZZFlGU1MT9u7d2+6XCODsUNlHH32ExsZGaLVahISEKH3yQlOmTIHT6Wx1sf/qq6/Gjz/+2GFNp06dglarxbBhw/Drr79i5cqVndoXXzqqPT09HevXr1f6eENDA3bs2NFh3fv27cPNN98MrVbrl/r6CoZEP9Lc3IxXXnkF8fHxmDp1Kmpra/HHP/4RAHDHHXcAOPsLi7vuugsAsHLlSlRWViIhIQGPPvooMjIylBumlixZguHDhyMpKQn3338/kpOTO+zc1157LR588EGkp6fj1ltvRXl5OSZOnOi3feuoVl9mzpyJRYsW4fHHH8fEiROxZMkS/Pzzz9BqtVi7di12796Nm2++GX/605/w0ksvtTsU0llmsxkPPvggpk+fjlGjRmHx4sVt5tFoNFi7di0OHjyIpKQk3HzzzXj22Wdx8uRJAO1/Vi+99BJaWlqUX0499thj+OmnnwAA99xzD6ZOnYrU1FTcdddduP3221Xr6+4+S5KENWvW4LXXXsMtt9yCxMREvPHGG6o/dfV6vdiwYQMSEhIwefJkfPXVV7DZbO3Oe/XVVyM+Pr7VL8wefvhhrF27FnFxcXjjjTfaXe7OO+/EiBEjkJCQgNmzZ2PChAmd2hdfOqp9xowZWLhwIZYuXYqJEyfCbDZj9+7dyrKPPvoorFYr4uLilGtNRUVFA/KGuiDhj3M26vc2btyI4uJin2cUlzJ/3Ph1qTt06BCWLVuGzZs3txm26s++++47ZGVl8T4JGjg8Hg9KS0vh9Xrxww8/YMOGDRd1kxNRV1x77bXYsmXLgAoI4Oz1loEYEAAvXF+yWlpaYLPZcPz4cVx55ZWYPXs2LBZLoMsioj6Gw01ERKSKw01ERKRqQA03eb1enDp1CoMHDx5wY55ERD1FCIGWlhYMHTq01Y2YwAALiVOnTqG8vDzQZRAR9UsxMTFt7hgfUCExePBgAGd3dKDd0BIIZWVlGDduXKDLIGoX+6f/NDc3o7y8XDmGnm9AhcS5IaZzdyNT9/F9pL6M/dO/2hum54VrIiJSxZAgIiJVDAkiIlLFkCAiIlUMCSIiUsWQICIiVQwJIiJSxZAIkOYWOdAl+GQ0GgNdQqf0h/eSqL8aUDfT9SfawRqkPF4Y6DIGhKJX/POsbSJqi2cSRESkiiFBRESqGBJE1Ep/ucbDa2a9g9ckiKgVXi/zr/5+zYxnEkREpMpnSNTV1eH3v/89kpOTkZKSgkcffRS1tbUAgIqKCqSlpSE5ORlpaWk4cuSIslxPtBERUe/yGRJBQUFYuHAhSkpKUFRUhFGjRmHFihUAAJvNBovFgpKSElgsFmRlZSnL9UQbERH1Lp8hodPpEB8fr7yeMGECqqqqUFNTA6fTCbPZDAAwm81wOp2ora3tkTYiIup9F3Xh2uv14u2334bJZILL5UJERAQ0Gg0AQKPRIDw8HC6XC0IIv7fp9Xp/7jcREXXCRYXE888/j5CQEMyfPx9Op7Onauq2srKyQJfgU3/5+V5/UVpaGugSBgz2Tf/rz/2z0yFht9tx9OhRrFu3DsHBwZAkCW63G7IsQ6PRQJZleDweSJIEIYTf2y7GuHHj+OzbSwwPbNSX9fX+2dTUpPrlulM/gc3Ly0NZWRlWr14NrVYLAAgLC4PBYIDD4QAAOBwOGAwG6PX6HmkjIqLeFySEEB3N8P3338NsNiMqKgqXXXYZACAyMhKrV6/G4cOHYbVa8csvvyA0NBR2ux1jxowBgB5p8+VcGvaXMwnesOQf/f1mpb6IfdN/+kP/7OjY6XO46brrrsN3333Xblt0dDQ2bdrUa21ERNS7eMc1ERGpYkgQEZEqhgQREaliSBARkSqGBBERqWJIEBGRKoYEERGpYkgQEZEqhgQREaliSBARkSqGBBERqWJIEBGRKoYEERGpYkgQEZEqhgQREaliSBARkSqfDx2y2+0oKSlBZWUlioqKEBMTg+PHj2PJkiXKPA0NDTh58iT27dsHADCZTNBqtcoTjjIzM5GQkAAAqKiogNVqRX19PXQ6Hex2O6Kiony2ERFR7/MZEklJSbj33nsxb948ZVpkZCQKC//3eMPc3FzIstxqufz8fMTExLRZn81mg8ViQWpqKgoLC5GVlYW33nrLZxsREfU+n8NNcXFxkCRJtb25uRlFRUWYO3euz43V1NTA6XTCbDYDAMxmM5xOJ2praztsIyKiwPB5JuHLrl27EBERgRtvvLHV9MzMTAghYDQasXTpUoSGhsLlciEiIgIajQYAoNFoEB4eDpfLBSGEapter7+omsrKyrq7Wz3OaDQGuoQBpbS0NNAlDBjsm/7Xn/tnt0Niy5Ytbc4iCgoKIEkSmpubkZubi5ycHKxYsaK7m+q0cePGKddD6NLAAxv1ZX29fzY1Nal+ue7Wr5vcbje++uorpKSktJp+bnhKq9XCYrFg//79ynS3261cv5BlGR6PB5IkddhGRESB0a2Q2Lp1KxITEzFs2DBlWmNjIxoaGgAAQggUFxfDYDAAAMLCwmAwGOBwOAAADocDBoMBer2+wzYiIgoMn8NNy5cvx86dO3HixAk88MAD0Ol02L59O4CzIfHMM8+0mr+mpgYZGRmQZRlerxfR0dGw2WxKe3Z2NqxWK9asWYPQ0FDY7fZOtRERUe8LEkKIQBfhL+fG1frLNYmUxwt9z0Q+Fb2SGugSBhz2Tf/pD/2zo2Mn77gmIiJVDAkiIlLFkCAiIlUMCSIiUsWQICIiVQwJIiJSxZAgIiJVDAkiIlLFkCAiIlUMCSIiUsWQICIiVQwJIiJSxZAgIiJVDAkiIlLFkCAiIlU+Q8Jut8NkMmHs2LEoLy9XpptMJtxxxx1ITU1FamoqPvvsM6WtoqICaWlpSE5ORlpaGo4cOdLtNiIi6n0+QyIpKQkFBQUYOXJkm7b8/HwUFhaisLAQCQkJynSbzQaLxYKSkhJYLBZkZWV1u42IiHqfz5CIi4uDJEmdXmFNTQ2cTifMZjMAwGw2w+l0ora2tsttREQUGD6fcd2RzMxMCCFgNBqxdOlShIaGwuVyISIiAhqNBgCg0WgQHh4Ol8sFIUSX2vR6fTd3k4iIuqLLIVFQUABJktDc3Izc3Fzk5ORgxYoV/qyty8rKygJdgk9GozHQJQwopaWlgS5hwGDf9L/+3D+7HBLnhqC0Wi0sFgsWL16sTHe73ZBlGRqNBrIsw+PxQJIkCCG61Hax2nuYNw1sPLBRX9bX+2dTU5Pql+su/QS2sbERDQ0NAAAhBIqLi2EwGAAAYWFhMBgMcDgcAACHwwGDwQC9Xt/lNiIiCgyfZxLLly/Hzp07ceLECTzwwAPQ6XRYt24dMjIyIMsyvF4voqOjYbPZlGWys7NhtVqxZs0ahIaGwm63d7uNiIh6X5AQQgS6CH85d8rUX4abUh4vDHQJA0LRK6mBLmHAYd/0n/7QPzs6dvKOayIiUsWQICIiVQwJIiJSxZAgIiJVDAkiIlLFkCAiIlUMCSIiUsWQICIiVQwJIiJSxZAgIiJVDAkiIlLFkCAiIlUMCSIiUsWQICIiVQwJIiJSxZAgIiJVPp9MZ7fbUVJSgsrKShQVFSEmJgZ1dXV48skncezYMWi1WowePRo5OTnKo0ZNJhO0Wq3y8IrMzEwkJCQAACoqKmC1WlFfXw+dTge73Y6oqCifbURE1Pt8nkkkJSWhoKAAI0eOVKYFBQVh4cKFKCkpQVFREUaNGoUVK1a0Wi4/Px+FhYUoLCxUAgIAbDYbLBYLSkpKYLFYkJWV1ak2IiLqfT5DIi4uDpIktZqm0+kQHx+vvJ4wYQKqqqp8bqympgZOpxNmsxkAYDab4XQ6UVtb22EbEREFhs/hJl+8Xi/efvttmEymVtMzMzMhhIDRaMTSpUsRGhoKl8uFiIgIaDQaAIBGo0F4eDhcLheEEKpt54axOqusrKy7u9XjjEZjoEsYUEpLSwNdwoDBvul//bl/djsknn/+eYSEhGD+/PnKtIKCAkiShObmZuTm5iInJ6fNcFRPau9h3jSw8cBGfVlf759NTU2qX6679esmu92Oo0ePYtWqVQgO/t+qzg1PabVaWCwW7N+/X5nudrshyzIAQJZleDweSJLUYRsREQVGl0MiLy8PZWVlWL16NbRarTK9sbERDQ0NAAAhBIqLi2EwGAAAYWFhMBgMcDgcAACHwwGDwQC9Xt9hGxERBYbP4ably5dj586dOHHiBB544AHodDqsWrUK69atQ1RUFNLT0wEAkZGRWL16NWpqapCRkQFZluH1ehEdHQ2bzaasLzs7G1arFWvWrEFoaCjsdnun2oiIqPcFCSFEoIvwl3Pjav3lmkTK44WBLmFAKHolNdAlDDjsm/7TH/pnR8dO3nFNRESqGBJERKSKIUFERKoYEkREpIohQUREqhgSRESkiiFBRESqGBJERKSKIUFERKoYEkREpIohQUREqhgSRESkiiFBRESqGBJERKSKIUFERKp8hoTdbofJZMLYsWNRXl6uTK+oqEBaWhqSk5ORlpaGI0eO9GgbERH1Pp8hkZSUhIKCAowcObLVdJvNBovFgpKSElgsFmRlZfVoGxER9T6fIREXFwdJklpNq6mpgdPphNlsBgCYzWY4nU7U1tb2SBsREQWGz2dct8flciEiIgIajQYAoNFoEB4eDpfLBSGE39v0er0/9pWIiC5Sl0KirysrKwt0CT4ZjcZAlzCglJaWBrqEAYN90//6c//sUkhIkgS32w1ZlqHRaCDLMjweDyRJghDC720Xq72HedPAxgMb9WV9vX82NTWpfrnu0k9gw8LCYDAY4HA4AAAOhwMGgwF6vb5H2oiIKDCChBCioxmWL1+OnTt34sSJExg2bBh0Oh22b9+Ow4cPw2q14pdffkFoaCjsdjvGjBkDAD3S1hnn0rC/nEmkPF4Y6BIGhKJXUgNdwoDDvuk//aF/dnTs9BkS/QlD4tLUH/4T9jfsm/7TH/pnR8dO3nFNRESqGBJERKSKIUFERKoYEkREpIohQUREqhgSRESkiiFBRESqGBJERKSKIUFERKoYEkREpIohQUREqhgSRESkiiFBRESqGBJERKSKIUFERKoYEkREpKpLz7g+5/jx41iyZInyuqGhASdPnsS+fftgMpmg1WqVB1hkZmYiISEBAFBRUQGr1Yr6+nrodDrY7XZERUX5bCMiot7VrZCIjIxEYeH/nmCVm5sLWZaV1/n5+YiJiWmznM1mg8ViQWpqKgoLC5GVlYW33nrLZxsREfUuvw03NTc3o6ioCHPnzu1wvpqaGjidTpjNZgCA2WyG0+lEbW1th21ERNT7unUmcb5du3YhIiICN954ozItMzMTQggYjUYsXboUoaGhcLlciIiIgEajAQBoNBqEh4fD5XJBCKHaptfr/VUqERF1kt9CYsuWLa3OIgoKCiBJEpqbm5Gbm4ucnBysWLHCX5vrUFlZWa9spzuMRmOgSxhQSktLA13CgMG+6X/9uX/6JSTcbje++uorvPTSS8o0SZIAAFqtFhaLBYsXL1amu91uyLIMjUYDWZbh8XggSRKEEKptF2PcuHHKBXO6NPDARn1ZX++fTU1Nql+u/XJNYuvWrUhMTMSwYcMAAI2NjWhoaAAACCFQXFwMg8EAAAgLC4PBYIDD4QAAOBwOGAwG6PX6DtuIiKj3+eVMYuvWrXjmmWeU1zU1NcjIyIAsy/B6vYiOjobNZlPas7OzYbVasWbNGoSGhsJut3eqjYiIepdfQqKkpKTV61GjRmHbtm2q80dHR2PTpk0X3UZERL2Ld1wTEZEqhgQREaliSBARkSqGBBERqWJIEBGRKoYEERGpYkgQEZEqhgQREaliSBARkSqGBBERqWJIEBGRKoYEERGpYkgQEZEqhgQREaliSBARkSqGBBERqer2Q4dMJhO0Wq3yTOnMzEwkJCSgoqICVqsV9fX10Ol0sNvtiIqKAoAutxERUe/yy5lEfn4+CgsLUVhYiISEBACAzWaDxWJBSUkJLBYLsrKylPm72kZERL2rR4abampq4HQ6YTabAQBmsxlOpxO1tbVdbiMiot7nl2dcZ2ZmQggBo9GIpUuXwuVyISIiAhqNBgCg0WgQHh4Ol8sFIUSX2vR6fafrKSsr88du9Sij0RjoEgaU0tLSQJcwYLBv+l9/7p/dDomCggJIkoTm5mbk5uYiJycH999/vx9K67px48Yp10jo0sADG/Vlfb1/NjU1qX657vZwkyRJAACtVguLxYL9+/dDkiS43W7IsgwAkGUZHo8HkiR1uY2IiHpft0KisbERDQ0NAAAhBIqLi2EwGBAWFgaDwQCHwwEAcDgcMBgM0Ov1XW4jIqLe163hppqaGmRkZECWZXi9XkRHR8NmswEAsrOzYbVasWbNGoSGhsJutyvLdbWNiIh6V7dCYtSoUdi2bVu7bdHR0di0aZNf24iIqHfxjmsiIlLFkCAiIlUMCSIiUsWQICIiVQwJIiJSxZAgIiJVDAkiIlLFkCAiIlUMCSIiUsWQICIiVQwJIiJSxZAgIiJVDAkiIlLFkCAiIlUMCSIiUtWt50nU1dXhySefxLFjx6DVajF69Gjk5ORAr9fDZDJBq9Uqz5rOzMxEQkICAKCiogJWqxX19fXQ6XSw2+2Iiory2UZERL2rW2cSQUFBWLhwIUpKSlBUVIRRo0ZhxYoVSnt+fj4KCwtRWFioBAQA2Gw2WCwWlJSUwGKxICsrq1NtRETUu7oVEjqdDvHx8crrCRMmoKqqqsNlampq4HQ6YTabAQBmsxlOpxO1tbUdthERUe/r1nDT+bxeL95++22YTCZlWmZmJoQQMBqNWLp0KUJDQ+FyuRAREQGNRgMA0Gg0CA8Ph8vlghBCtU2v1/urVCIi6iS/hcTzzz+PkJAQzJ8/HwBQUFAASZLQ3NyM3Nxc5OTktBqK6kllZWW9sp3uMBqNgS5hQCktLQ10CQMG+6b/9ef+6ZeQsNvtOHr0KNatW4fg4LMjWJIkAQC0Wi0sFgsWL16sTHe73ZBlGRqNBrIsw+PxQJIkCCFU2y7GuHHjlAvmdGnggY36sr7eP5uamlS/XHf7J7B5eXkoKyvD6tWrodVqAQCNjY1oaGgAAAghUFxcDIPBAAAICwuDwWCAw+EAADgcDhgMBuj1+g7biIio93XrTOL777/HunXrEBUVhfT0dABAZGQkrFYrMjIyIMsyvF4voqOjYbPZlOWys7NhtVqxZs0ahIaGwm63d6qNiIh6V7dC4rrrrsN3333Xbtu2bdtUl4uOjsamTZsuuo2IiHoX77gmIiJVDAkiIlLFkCAiIlUMCSIiUsWQICIiVQwJIiJSxZAgIiJVDAkiIlLFkCAiIlUMCSIiUsWQICIiVQwJIiJSxZAgIiJVDAkiIlLFkCAiIlUMCSIiUtUnQ6KiogJpaWlITk5GWloajhw5EuiSiIguSX0yJGw2GywWC0pKSmCxWJCVlRXokoiILkndenxpT6ipqYHT6cSGDRsAAGazGc8//zxqa2uh1+s7XFYIAQBobm7u8Tr9QTdUE+gSBoSmpqZAlzDgsG/6T3/on+eOmeeOoefrcyHhcrkQEREBjeZsJ9VoNAgPD4fL5fIZEi0tLQCA8vLyHq/TH/6QKgW6hAGhrKws0CUMOOyb/tOf+mdLSwsuu+yyVtP6XEh0x9ChQxETE4PBgwcjKCgo0OUQEfULQgi0tLRg6NChbdr6XEhIkgS32w1ZlqHRaCDLMjweDyTJ9zeb4OBgXHnllb1QJRHRwHLhGcQ5fe7CdVhYGAwGAxwOBwDA4XDAYDD4HGoiIiL/CxLtXakIsMOHD8NqteKXX35BaGgo7HY7xowZE+iyiIguOX0yJIiIqG/oc8NNRETUdzAkiIhIFUOCiIhUMSSIiEgVQ4KIiFT1uZvpiIjU1NXVobq6GgAwfPhwDBs2LMAVDXwMCSLq844dO4bnnnsOTqcT4eHhAACPx4MbbrgBf/rTnxAVFRXYAgcw3idBnZKSkoKioqJAl0GXqPT0dFgsFpjNZgQHnx0l93q9KCoqwsaNG/Huu+8GuMKBi2cSpDh06JBqW11dXS9WQtRafX095syZ02pacHAwUlNTsXbt2gBVdWlgSJDCbDZj5MiR7f5N+fr6+t4viOj/0+l0cDgcmD17tvIXnoUQKCoqQmhoaICrG9g43ESKpKQkbNy4EREREW3aEhMT8emnnwagKiLgyJEjsNlsOHDggNI/3W43rr/+emRnZ/Nvu/UgnkmQ4vbbb0dlZWW7ITFjxowAVER0VlRUFP72t7+htrYWLpcLwNnHCvCvQ/c8nkkQEZEq3kxHRESqGBJERKSKIUFERKoYEkREpIohQUREqv4fC6YKvPiwiMMAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_train = y_train.reset_index(level=0, drop=True)\n",
    "y_train.value_counts().plot(kind='bar')\n",
    "plt.title('histogram of completed offers (train set)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Save the train dataset to a csv file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.exists(PROCESSED_DATA_DIR):\n",
    "    os.mkdir(PROCESSED_DATA_DIR)\n",
    "if not os.path.exists(os.path.join(PROCESSED_DATA_DIR, 'train')):\n",
    "    os.mkdir(os.path.join(PROCESSED_DATA_DIR, 'train'))\n",
    "pd.concat([y_train, X_train], axis=1).to_csv(os.path.join(PROCESSED_DATA_DIR, 'train', 'data.csv'), index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>income</th>\n",
       "      <th>became_member_on_year</th>\n",
       "      <th>became_member_on_month</th>\n",
       "      <th>became_member_on_day</th>\n",
       "      <th>reward</th>\n",
       "      <th>difficulty</th>\n",
       "      <th>duration</th>\n",
       "      <th>offer_type</th>\n",
       "      <th>email</th>\n",
       "      <th>mobile</th>\n",
       "      <th>web</th>\n",
       "      <th>social</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>M</td>\n",
       "      <td>59</td>\n",
       "      <td>59000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>12</td>\n",
       "      <td>31</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>M</td>\n",
       "      <td>36</td>\n",
       "      <td>46000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>7</td>\n",
       "      <td>30</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>M</td>\n",
       "      <td>36</td>\n",
       "      <td>73000.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>12</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  gender  age   income  became_member_on_year  became_member_on_month  \\\n",
       "0      M   59  59000.0                   2017                      12   \n",
       "1      M   36  46000.0                   2017                       7   \n",
       "2      M   36  73000.0                   2017                      12   \n",
       "\n",
       "   became_member_on_day  reward  difficulty  duration offer_type  email  \\\n",
       "0                    31       5           5         7       bogo   True   \n",
       "1                    30       5           5         5       bogo   True   \n",
       "2                    12       5           5         7       bogo   True   \n",
       "\n",
       "   mobile   web  social  \n",
       "0    True  True   False  \n",
       "1    True  True    True  \n",
       "2    True  True   False  "
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test = X_test.reset_index(level=0, drop=True)\n",
    "X_test.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_test = y_test.reset_index(level=0, drop=True)\n",
    "y_test.value_counts().plot(kind='bar')\n",
    "plt.title('histogram of completed offers (test set)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Save test set to a csv file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.exists(PROCESSED_DATA_DIR):\n",
    "    os.mkdir(PROCESSED_DATA_DIR)\n",
    "if not os.path.exists(os.path.join(PROCESSED_DATA_DIR, 'test')):\n",
    "    os.mkdir(os.path.join(PROCESSED_DATA_DIR, 'test'))\n",
    "pd.concat([y_test, X_test], axis=1).to_csv(os.path.join(PROCESSED_DATA_DIR, 'test', 'data.csv'), index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Uploading Data to S3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Upload metadata\n",
    "sagemaker_session = sagemaker.Session()\n",
    "BUCKET = sagemaker_session.default_bucket()\n",
    "PREFIX = 'Project/starbucks-dataset'\n",
    "# ---\n",
    "DATASET_LOCAL_DIR = PROCESSED_DATA_DIR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset directory path on S3 : s3://sagemaker-us-east-1-543661723968/Project/starbucks-dataset\n"
     ]
    }
   ],
   "source": [
    "dataset_dir_path = sagemaker_session.upload_data(\n",
    "    path=DATASET_LOCAL_DIR,\n",
    "    bucket=BUCKET,\n",
    "    key_prefix=PREFIX\n",
    ")\n",
    "\n",
    "print(f\"Dataset directory path on S3 : {dataset_dir_path}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Benchmarking and Training Model using Autogluon AutoML container from sagemaker"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Some of the following code is adapted from the sagemakers examples from aws sagemaker examples public repo\n",
    "https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/autogluon_tabular/Amazon_Tabular_Classification_AutoGluon.ipynb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture --no-stderr\n",
    "!pip install sagemaker ipywidgets --upgrade --quiet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sagemaker, boto3, json\n",
    "from sagemaker import get_execution_role\n",
    "\n",
    "aws_role = get_execution_role()\n",
    "aws_region = boto3.Session().region_name\n",
    "sess = sagemaker.Session()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_model_id, train_model_version, train_scope = (\n",
    "    \"autogluon-classification-ensemble\",\n",
    "    \"*\",\n",
    "    \"training\",\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "training_instance_type = \"ml.m5.large\"\n",
    "\n",
    "# Retrieve the docker image\n",
    "train_image_uri = image_uris.retrieve(\n",
    "    region=None,\n",
    "    framework=None,\n",
    "    model_id=train_model_id,\n",
    "    model_version=train_model_version,\n",
    "    image_scope=train_scope,\n",
    "    instance_type=training_instance_type,\n",
    ")\n",
    "# Retrieve the training script\n",
    "train_source_uri = script_uris.retrieve(\n",
    "    model_id=train_model_id, model_version=train_model_version, script_scope=train_scope\n",
    ")\n",
    "# Retrieve the pre-trained model tarball to further fine-tune. In tabular case, however, the pre-trained model tarball is dummy and fine-tune means training from scratch.\n",
    "train_model_uri = model_uris.retrieve(\n",
    "    model_id=train_model_id, model_version=train_model_version, model_scope=train_scope\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "s3://sagemaker-us-east-1-543661723968/Project/starbucks-dataset/\n"
     ]
    }
   ],
   "source": [
    "# Sample training data is available in this bucket\n",
    "training_data_bucket = sess.default_bucket()\n",
    "training_data_prefix = \"Project/starbucks-dataset/\"\n",
    "\n",
    "training_dataset_s3_path = f\"s3://{training_data_bucket}/{training_data_prefix}\"\n",
    "print(training_dataset_s3_path)\n",
    "\n",
    "output_bucket = sess.default_bucket()\n",
    "output_prefix = \"Project/starbucks-tabular-binary\"\n",
    "\n",
    "s3_output_location = f\"s3://{output_bucket}/{output_prefix}/output\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'eval_metric': 'accuracy', 'presets': 'best_quality', 'auto_stack': 'True', 'num_bag_folds': '0', 'num_bag_sets': '1', 'num_stack_levels': '0', 'refit_full': 'False', 'set_best_to_refit_full': 'False', 'save_space': 'True', 'verbosity': '2'}\n"
     ]
    }
   ],
   "source": [
    "# Retrieve the default hyper-parameters for fine-tuning the model\n",
    "from sagemaker import hyperparameters\n",
    "hyperparameters = hyperparameters.retrieve_default(\n",
    "    model_id=train_model_id, model_version=train_model_version\n",
    ")\n",
    "\n",
    "# [Optional] Override default hyperparameters with custom values\n",
    "hyperparameters[\"auto_stack\"] = \"True\"\n",
    "hyperparameters['eval_metric'] = 'accuracy'\n",
    "hyperparameters['presets'] = 'best_quality'\n",
    "hyperparameters['save_space'] = \"True\"\n",
    "hyperparameters['verbosity'] = 2\n",
    "print(hyperparameters)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2022-12-06 20:45:49 Starting - Starting the training job...\n",
      "2022-12-06 20:46:15 Starting - Preparing the instances for trainingProfilerReport-1670359549: InProgress\n",
      "......\n",
      "2022-12-06 20:47:15 Downloading - Downloading input data......\n",
      "2022-12-06 20:48:15 Training - Downloading the training image.....\u001b[34m2022-12-06 20:48:56,797 sagemaker-training-toolkit INFO     Imported framework sagemaker_mxnet_container.training\u001b[0m\n",
      "\u001b[34m2022-12-06 20:48:56,800 sagemaker-training-toolkit INFO     No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[34m2022-12-06 20:48:56,811 sagemaker_mxnet_container.training INFO     MXNet training environment: {'SM_HOSTS': '[\"algo-1\"]', 'SM_NETWORK_INTERFACE_NAME': 'eth0', 'SM_HPS': '{\"auto_stack\":\"True\",\"eval_metric\":\"accuracy\",\"num_bag_folds\":\"0\",\"num_bag_sets\":\"1\",\"num_stack_levels\":\"0\",\"presets\":\"best_quality\",\"refit_full\":\"False\",\"save_space\":\"True\",\"set_best_to_refit_full\":\"False\",\"verbosity\":\"2\"}', 'SM_USER_ENTRY_POINT': 'transfer_learning.py', 'SM_FRAMEWORK_PARAMS': '{}', 'SM_RESOURCE_CONFIG': '{\"current_group_name\":\"homogeneousCluster\",\"current_host\":\"algo-1\",\"current_instance_type\":\"ml.m5.large\",\"hosts\":[\"algo-1\"],\"instance_groups\":[{\"hosts\":[\"algo-1\"],\"instance_group_name\":\"homogeneousCluster\",\"instance_type\":\"ml.m5.large\"}],\"network_interface_name\":\"eth0\"}', 'SM_INPUT_DATA_CONFIG': '{\"model\":{\"ContentType\":\"application/x-sagemaker-model\",\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"},\"training\":{\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"}}', 'SM_OUTPUT_DATA_DIR': '/opt/ml/output/data', 'SM_CHANNELS': '[\"model\",\"training\"]', 'SM_CURRENT_HOST': 'algo-1', 'SM_MODULE_NAME': 'transfer_learning', 'SM_LOG_LEVEL': '20', 'SM_FRAMEWORK_MODULE': 'sagemaker_mxnet_container.training:main', 'SM_INPUT_DIR': '/opt/ml/input', 'SM_INPUT_CONFIG_DIR': '/opt/ml/input/config', 'SM_OUTPUT_DIR': '/opt/ml/output', 'SM_NUM_CPUS': '2', 'SM_NUM_GPUS': '0', 'SM_MODEL_DIR': '/opt/ml/model', 'SM_MODULE_DIR': 's3://jumpstart-cache-prod-us-east-1/source-directory-tarballs/autogluon/transfer_learning/classification/v1.0.2/sourcedir.tar.gz', 'SM_TRAINING_ENV': '{\"additional_framework_parameters\":{},\"channel_input_dirs\":{\"model\":\"/opt/ml/input/data/model\",\"training\":\"/opt/ml/input/data/training\"},\"current_host\":\"algo-1\",\"framework_module\":\"sagemaker_mxnet_container.training:main\",\"hosts\":[\"algo-1\"],\"hyperparameters\":{\"auto_stack\":\"True\",\"eval_metric\":\"accuracy\",\"num_bag_folds\":\"0\",\"num_bag_sets\":\"1\",\"num_stack_levels\":\"0\",\"presets\":\"best_quality\",\"refit_full\":\"False\",\"save_space\":\"True\",\"set_best_to_refit_full\":\"False\",\"verbosity\":\"2\"},\"input_config_dir\":\"/opt/ml/input/config\",\"input_data_config\":{\"model\":{\"ContentType\":\"application/x-sagemaker-model\",\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"},\"training\":{\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"}},\"input_dir\":\"/opt/ml/input\",\"is_master\":true,\"is_modelparallel_enabled\":null,\"job_name\":\"autogluon-classification-ensemble-train-2022-12-06-20-45-49-215\",\"log_level\":20,\"master_hostname\":\"algo-1\",\"model_dir\":\"/opt/ml/model\",\"module_dir\":\"s3://jumpstart-cache-prod-us-east-1/source-directory-tarballs/autogluon/transfer_learning/classification/v1.0.2/sourcedir.tar.gz\",\"module_name\":\"transfer_learning\",\"network_interface_name\":\"eth0\",\"num_cpus\":2,\"num_gpus\":0,\"output_data_dir\":\"/opt/ml/output/data\",\"output_dir\":\"/opt/ml/output\",\"output_intermediate_dir\":\"/opt/ml/output/intermediate\",\"resource_config\":{\"current_group_name\":\"homogeneousCluster\",\"current_host\":\"algo-1\",\"current_instance_type\":\"ml.m5.large\",\"hosts\":[\"algo-1\"],\"instance_groups\":[{\"hosts\":[\"algo-1\"],\"instance_group_name\":\"homogeneousCluster\",\"instance_type\":\"ml.m5.large\"}],\"network_interface_name\":\"eth0\"},\"user_entry_point\":\"transfer_learning.py\"}', 'SM_USER_ARGS': '[\"--auto_stack\",\"True\",\"--eval_metric\",\"accuracy\",\"--num_bag_folds\",\"0\",\"--num_bag_sets\",\"1\",\"--num_stack_levels\",\"0\",\"--presets\",\"best_quality\",\"--refit_full\",\"False\",\"--save_space\",\"True\",\"--set_best_to_refit_full\",\"False\",\"--verbosity\",\"2\"]', 'SM_OUTPUT_INTERMEDIATE_DIR': '/opt/ml/output/intermediate', 'SM_CHANNEL_MODEL': '/opt/ml/input/data/model', 'SM_CHANNEL_TRAINING': '/opt/ml/input/data/training', 'SM_HP_AUTO_STACK': 'True', 'SM_HP_EVAL_METRIC': 'accuracy', 'SM_HP_NUM_BAG_FOLDS': '0', 'SM_HP_NUM_BAG_SETS': '1', 'SM_HP_NUM_STACK_LEVELS': '0', 'SM_HP_PRESETS': 'best_quality', 'SM_HP_REFIT_FULL': 'False', 'SM_HP_SAVE_SPACE': 'True', 'SM_HP_SET_BEST_TO_REFIT_FULL': 'False', 'SM_HP_VERBOSITY': '2'}\u001b[0m\n",
      "\u001b[34m2022-12-06 20:48:57,246 sagemaker-training-toolkit INFO     Installing dependencies from requirements.txt:\u001b[0m\n",
      "\u001b[34m/usr/bin/python3 -m pip install -r requirements.txt\u001b[0m\n",
      "\u001b[34mProcessing ./lib/sagemaker_jumpstart_script_utilities/sagemaker_jumpstart_script_utilities-1.0.1-py2.py3-none-any.whl\u001b[0m\n",
      "\u001b[34mInstalling collected packages: sagemaker-jumpstart-script-utilities\u001b[0m\n",
      "\u001b[34mSuccessfully installed sagemaker-jumpstart-script-utilities-1.0.1\u001b[0m\n",
      "\u001b[34mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\n",
      "\u001b[34m--- Logging error ---\u001b[0m\n",
      "\u001b[34mTraceback (most recent call last):\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_internal/utils/logging.py\", line 177, in emit\n",
      "    self.console.print(renderable, overflow=\"ignore\", crop=False, style=style)\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_vendor/rich/console.py\", line 1673, in print\n",
      "    extend(render(renderable, render_options))\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_vendor/rich/console.py\", line 1305, in render\n",
      "    for render_output in iter_render:\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_internal/utils/logging.py\", line 134, in __rich_console__\n",
      "    for line in lines:\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_vendor/rich/segment.py\", line 249, in split_lines\n",
      "    for segment in segments:\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_vendor/rich/console.py\", line 1283, in render\n",
      "    renderable = rich_cast(renderable)\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_vendor/rich/protocol.py\", line 36, in rich_cast\n",
      "    renderable = cast_method()\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_internal/self_outdated_check.py\", line 130, in __rich__\n",
      "    pip_cmd = get_best_invocation_for_this_pip()\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_internal/utils/entrypoints.py\", line 58, in get_best_invocation_for_this_pip\n",
      "    if found_executable and os.path.samefile(\n",
      "  File \"/usr/lib/python3.8/genericpath.py\", line 101, in samefile\n",
      "    s2 = os.stat(f2)\u001b[0m\n",
      "\u001b[34mFileNotFoundError: [Errno 2] No such file or directory: '/usr/bin/pip3.8'\u001b[0m\n",
      "\u001b[34mCall stack:\n",
      "  File \"/usr/lib/python3.8/runpy.py\", line 194, in _run_module_as_main\n",
      "    return _run_code(code, main_globals, None,\n",
      "  File \"/usr/lib/python3.8/runpy.py\", line 87, in _run_code\n",
      "    exec(code, run_globals)\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/__main__.py\", line 31, in <module>\n",
      "    sys.exit(_main())\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_internal/cli/main.py\", line 70, in main\n",
      "    return command.main(cmd_args)\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_internal/cli/base_command.py\", line 101, in main\n",
      "    return self._main(args)\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_internal/cli/base_command.py\", line 223, in _main\n",
      "    self.handle_pip_version_check(options)\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_internal/cli/req_command.py\", line 190, in handle_pip_version_check\n",
      "    pip_self_version_check(session, options)\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_internal/self_outdated_check.py\", line 236, in pip_self_version_check\n",
      "    logger.warning(\"[present-rich] %s\", upgrade_prompt)\n",
      "  File \"/usr/lib/python3.8/logging/__init__.py\", line 1458, in warning\n",
      "    self._log(WARNING, msg, args, **kwargs)\n",
      "  File \"/usr/lib/python3.8/logging/__init__.py\", line 1589, in _log\n",
      "    self.handle(record)\n",
      "  File \"/usr/lib/python3.8/logging/__init__.py\", line 1599, in handle\n",
      "    self.callHandlers(record)\n",
      "  File \"/usr/lib/python3.8/logging/__init__.py\", line 1661, in callHandlers\n",
      "    hdlr.handle(record)\n",
      "  File \"/usr/lib/python3.8/logging/__init__.py\", line 954, in handle\n",
      "    self.emit(record)\n",
      "  File \"/usr/local/lib/python3.8/dist-packages/pip/_internal/utils/logging.py\", line 179, in emit\n",
      "    self.handleError(record)\u001b[0m\n",
      "\u001b[34mMessage: '[present-rich] %s'\u001b[0m\n",
      "\u001b[34mArguments: (UpgradePrompt(old='22.2.2', new='22.3.1'),)\u001b[0m\n",
      "\u001b[34m2022-12-06 20:48:59,799 sagemaker-training-toolkit INFO     No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[34m2022-12-06 20:48:59,813 sagemaker-training-toolkit INFO     No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[34m2022-12-06 20:48:59,826 sagemaker-training-toolkit INFO     No GPUs detected (normal if no gpus installed)\u001b[0m\n",
      "\u001b[34m2022-12-06 20:48:59,834 sagemaker-training-toolkit INFO     Invoking user script\u001b[0m\n",
      "\u001b[34mTraining Env:\u001b[0m\n",
      "\u001b[34m{\n",
      "    \"additional_framework_parameters\": {},\n",
      "    \"channel_input_dirs\": {\n",
      "        \"model\": \"/opt/ml/input/data/model\",\n",
      "        \"training\": \"/opt/ml/input/data/training\"\n",
      "    },\n",
      "    \"current_host\": \"algo-1\",\n",
      "    \"framework_module\": \"sagemaker_mxnet_container.training:main\",\n",
      "    \"hosts\": [\n",
      "        \"algo-1\"\n",
      "    ],\n",
      "    \"hyperparameters\": {\n",
      "        \"auto_stack\": \"True\",\n",
      "        \"eval_metric\": \"accuracy\",\n",
      "        \"num_bag_folds\": \"0\",\n",
      "        \"num_bag_sets\": \"1\",\n",
      "        \"num_stack_levels\": \"0\",\n",
      "        \"presets\": \"best_quality\",\n",
      "        \"refit_full\": \"False\",\n",
      "        \"save_space\": \"True\",\n",
      "        \"set_best_to_refit_full\": \"False\",\n",
      "        \"verbosity\": \"2\"\n",
      "    },\n",
      "    \"input_config_dir\": \"/opt/ml/input/config\",\n",
      "    \"input_data_config\": {\n",
      "        \"model\": {\n",
      "            \"ContentType\": \"application/x-sagemaker-model\",\n",
      "            \"TrainingInputMode\": \"File\",\n",
      "            \"S3DistributionType\": \"FullyReplicated\",\n",
      "            \"RecordWrapperType\": \"None\"\n",
      "        },\n",
      "        \"training\": {\n",
      "            \"TrainingInputMode\": \"File\",\n",
      "            \"S3DistributionType\": \"FullyReplicated\",\n",
      "            \"RecordWrapperType\": \"None\"\n",
      "        }\n",
      "    },\n",
      "    \"input_dir\": \"/opt/ml/input\",\n",
      "    \"is_master\": true,\n",
      "    \"is_modelparallel_enabled\": null,\n",
      "    \"job_name\": \"autogluon-classification-ensemble-train-2022-12-06-20-45-49-215\",\n",
      "    \"log_level\": 20,\n",
      "    \"master_hostname\": \"algo-1\",\n",
      "    \"model_dir\": \"/opt/ml/model\",\n",
      "    \"module_dir\": \"s3://jumpstart-cache-prod-us-east-1/source-directory-tarballs/autogluon/transfer_learning/classification/v1.0.2/sourcedir.tar.gz\",\n",
      "    \"module_name\": \"transfer_learning\",\n",
      "    \"network_interface_name\": \"eth0\",\n",
      "    \"num_cpus\": 2,\n",
      "    \"num_gpus\": 0,\n",
      "    \"output_data_dir\": \"/opt/ml/output/data\",\n",
      "    \"output_dir\": \"/opt/ml/output\",\n",
      "    \"output_intermediate_dir\": \"/opt/ml/output/intermediate\",\n",
      "    \"resource_config\": {\n",
      "        \"current_host\": \"algo-1\",\n",
      "        \"current_instance_type\": \"ml.m5.large\",\n",
      "        \"current_group_name\": \"homogeneousCluster\",\n",
      "        \"hosts\": [\n",
      "            \"algo-1\"\n",
      "        ],\n",
      "        \"instance_groups\": [\n",
      "            {\n",
      "                \"instance_group_name\": \"homogeneousCluster\",\n",
      "                \"instance_type\": \"ml.m5.large\",\n",
      "                \"hosts\": [\n",
      "                    \"algo-1\"\n",
      "                ]\n",
      "            }\n",
      "        ],\n",
      "        \"network_interface_name\": \"eth0\"\n",
      "    },\n",
      "    \"user_entry_point\": \"transfer_learning.py\"\u001b[0m\n",
      "\u001b[34m}\u001b[0m\n",
      "\u001b[34mEnvironment variables:\u001b[0m\n",
      "\u001b[34mSM_HOSTS=[\"algo-1\"]\u001b[0m\n",
      "\u001b[34mSM_NETWORK_INTERFACE_NAME=eth0\u001b[0m\n",
      "\u001b[34mSM_HPS={\"auto_stack\":\"True\",\"eval_metric\":\"accuracy\",\"num_bag_folds\":\"0\",\"num_bag_sets\":\"1\",\"num_stack_levels\":\"0\",\"presets\":\"best_quality\",\"refit_full\":\"False\",\"save_space\":\"True\",\"set_best_to_refit_full\":\"False\",\"verbosity\":\"2\"}\u001b[0m\n",
      "\u001b[34mSM_USER_ENTRY_POINT=transfer_learning.py\u001b[0m\n",
      "\u001b[34mSM_FRAMEWORK_PARAMS={}\u001b[0m\n",
      "\u001b[34mSM_RESOURCE_CONFIG={\"current_group_name\":\"homogeneousCluster\",\"current_host\":\"algo-1\",\"current_instance_type\":\"ml.m5.large\",\"hosts\":[\"algo-1\"],\"instance_groups\":[{\"hosts\":[\"algo-1\"],\"instance_group_name\":\"homogeneousCluster\",\"instance_type\":\"ml.m5.large\"}],\"network_interface_name\":\"eth0\"}\u001b[0m\n",
      "\u001b[34mSM_INPUT_DATA_CONFIG={\"model\":{\"ContentType\":\"application/x-sagemaker-model\",\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"},\"training\":{\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"}}\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_DATA_DIR=/opt/ml/output/data\u001b[0m\n",
      "\u001b[34mSM_CHANNELS=[\"model\",\"training\"]\u001b[0m\n",
      "\u001b[34mSM_CURRENT_HOST=algo-1\u001b[0m\n",
      "\u001b[34mSM_MODULE_NAME=transfer_learning\u001b[0m\n",
      "\u001b[34mSM_LOG_LEVEL=20\u001b[0m\n",
      "\u001b[34mSM_FRAMEWORK_MODULE=sagemaker_mxnet_container.training:main\u001b[0m\n",
      "\u001b[34mSM_INPUT_DIR=/opt/ml/input\u001b[0m\n",
      "\u001b[34mSM_INPUT_CONFIG_DIR=/opt/ml/input/config\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_DIR=/opt/ml/output\u001b[0m\n",
      "\u001b[34mSM_NUM_CPUS=2\u001b[0m\n",
      "\u001b[34mSM_NUM_GPUS=0\u001b[0m\n",
      "\u001b[34mSM_MODEL_DIR=/opt/ml/model\u001b[0m\n",
      "\u001b[34mSM_MODULE_DIR=s3://jumpstart-cache-prod-us-east-1/source-directory-tarballs/autogluon/transfer_learning/classification/v1.0.2/sourcedir.tar.gz\u001b[0m\n",
      "\u001b[34mSM_TRAINING_ENV={\"additional_framework_parameters\":{},\"channel_input_dirs\":{\"model\":\"/opt/ml/input/data/model\",\"training\":\"/opt/ml/input/data/training\"},\"current_host\":\"algo-1\",\"framework_module\":\"sagemaker_mxnet_container.training:main\",\"hosts\":[\"algo-1\"],\"hyperparameters\":{\"auto_stack\":\"True\",\"eval_metric\":\"accuracy\",\"num_bag_folds\":\"0\",\"num_bag_sets\":\"1\",\"num_stack_levels\":\"0\",\"presets\":\"best_quality\",\"refit_full\":\"False\",\"save_space\":\"True\",\"set_best_to_refit_full\":\"False\",\"verbosity\":\"2\"},\"input_config_dir\":\"/opt/ml/input/config\",\"input_data_config\":{\"model\":{\"ContentType\":\"application/x-sagemaker-model\",\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"},\"training\":{\"RecordWrapperType\":\"None\",\"S3DistributionType\":\"FullyReplicated\",\"TrainingInputMode\":\"File\"}},\"input_dir\":\"/opt/ml/input\",\"is_master\":true,\"is_modelparallel_enabled\":null,\"job_name\":\"autogluon-classification-ensemble-train-2022-12-06-20-45-49-215\",\"log_level\":20,\"master_hostname\":\"algo-1\",\"model_dir\":\"/opt/ml/model\",\"module_dir\":\"s3://jumpstart-cache-prod-us-east-1/source-directory-tarballs/autogluon/transfer_learning/classification/v1.0.2/sourcedir.tar.gz\",\"module_name\":\"transfer_learning\",\"network_interface_name\":\"eth0\",\"num_cpus\":2,\"num_gpus\":0,\"output_data_dir\":\"/opt/ml/output/data\",\"output_dir\":\"/opt/ml/output\",\"output_intermediate_dir\":\"/opt/ml/output/intermediate\",\"resource_config\":{\"current_group_name\":\"homogeneousCluster\",\"current_host\":\"algo-1\",\"current_instance_type\":\"ml.m5.large\",\"hosts\":[\"algo-1\"],\"instance_groups\":[{\"hosts\":[\"algo-1\"],\"instance_group_name\":\"homogeneousCluster\",\"instance_type\":\"ml.m5.large\"}],\"network_interface_name\":\"eth0\"},\"user_entry_point\":\"transfer_learning.py\"}\u001b[0m\n",
      "\u001b[34mSM_USER_ARGS=[\"--auto_stack\",\"True\",\"--eval_metric\",\"accuracy\",\"--num_bag_folds\",\"0\",\"--num_bag_sets\",\"1\",\"--num_stack_levels\",\"0\",\"--presets\",\"best_quality\",\"--refit_full\",\"False\",\"--save_space\",\"True\",\"--set_best_to_refit_full\",\"False\",\"--verbosity\",\"2\"]\u001b[0m\n",
      "\u001b[34mSM_OUTPUT_INTERMEDIATE_DIR=/opt/ml/output/intermediate\u001b[0m\n",
      "\u001b[34mSM_CHANNEL_MODEL=/opt/ml/input/data/model\u001b[0m\n",
      "\u001b[34mSM_CHANNEL_TRAINING=/opt/ml/input/data/training\u001b[0m\n",
      "\u001b[34mSM_HP_AUTO_STACK=True\u001b[0m\n",
      "\u001b[34mSM_HP_EVAL_METRIC=accuracy\u001b[0m\n",
      "\u001b[34mSM_HP_NUM_BAG_FOLDS=0\u001b[0m\n",
      "\u001b[34mSM_HP_NUM_BAG_SETS=1\u001b[0m\n",
      "\u001b[34mSM_HP_NUM_STACK_LEVELS=0\u001b[0m\n",
      "\u001b[34mSM_HP_PRESETS=best_quality\u001b[0m\n",
      "\u001b[34mSM_HP_REFIT_FULL=False\u001b[0m\n",
      "\u001b[34mSM_HP_SAVE_SPACE=True\u001b[0m\n",
      "\u001b[34mSM_HP_SET_BEST_TO_REFIT_FULL=False\u001b[0m\n",
      "\u001b[34mSM_HP_VERBOSITY=2\u001b[0m\n",
      "\u001b[34mPYTHONPATH=/opt/ml/code:/usr/local/bin:/usr/lib/python38.zip:/usr/lib/python3.8:/usr/lib/python3.8/lib-dynload:/usr/local/lib/python3.8/dist-packages:/usr/lib/python3/dist-packages\u001b[0m\n",
      "\u001b[34mInvoking script with the following command:\u001b[0m\n",
      "\u001b[34m/usr/bin/python3 transfer_learning.py --auto_stack True --eval_metric accuracy --num_bag_folds 0 --num_bag_sets 1 --num_stack_levels 0 --presets best_quality --refit_full False --save_space True --set_best_to_refit_full False --verbosity 2\u001b[0m\n",
      "\u001b[34mValidation data is not found. A certain fraction of training data is randomly selected as validation data. The fraction value is selected based on the number of rows in the training data. For details, see the official documentation: https://auto.gluon.ai/stable/api/autogluon.predictor.html#autogluon.tabular.TabularPredictor.fit\u001b[0m\n",
      "\u001b[34mWarning: path already exists! This predictor may overwrite an existing predictor! path=\"/opt/ml/model\"\u001b[0m\n",
      "\u001b[34mPresets specified: ['best_quality']\u001b[0m\n",
      "\u001b[34mBeginning AutoGluon training ...\u001b[0m\n",
      "\u001b[34mAutoGluon will save models to \"/opt/ml/model/\"\u001b[0m\n",
      "\u001b[34mAutoGluon Version:  0.4.3\u001b[0m\n",
      "\u001b[34mPython Version:     3.8.10\u001b[0m\n",
      "\u001b[34mOperating System:   Linux\u001b[0m\n",
      "\u001b[34mTrain Data Rows:    41723\u001b[0m\n",
      "\u001b[34mTrain Data Columns: 14\u001b[0m\n",
      "\u001b[34mLabel Column: target\u001b[0m\n",
      "\u001b[34mPreprocessing data ...\u001b[0m\n",
      "\u001b[34mWarning: Some classes in the training set have fewer than 10 examples. AutoGluon will only keep 2 out of 3 classes for training and will not try to predict the rare classes. To keep more classes, increase the number of datapoints from these rare classes in the training data or reduce label_count_threshold.\u001b[0m\n",
      "\u001b[34mFraction of data from classes with at least 10 examples that will be kept for training models: 0.9999760324041895\u001b[0m\n",
      "\u001b[34mSelected class <--> label mapping:  class 1 = 1, class 0 = 0\u001b[0m\n",
      "\u001b[34mTrain Data Class Count: 2\u001b[0m\n",
      "\u001b[34mUsing Feature Generators to preprocess the data ...\u001b[0m\n",
      "\u001b[34mFitting AutoMLPipelineFeatureGenerator...\u001b[0m\n",
      "\u001b[34m#011Available Memory:                    7238.09 MB\u001b[0m\n",
      "\u001b[34m#011Train Data (Original)  Memory Usage: 35.17 MB (0.5% of available memory)\u001b[0m\n",
      "\u001b[34m#011Inferring data type of each feature based on column values. Set feature_metadata_in to manually specify special dtypes of the features.\u001b[0m\n",
      "\u001b[34m#011Stage 1 Generators:\u001b[0m\n",
      "\u001b[34m#011#011Fitting AsTypeFeatureGenerator...\u001b[0m\n",
      "\u001b[34m#011#011#011Note: Converting 3 features to boolean dtype as they only contain 2 unique values.\u001b[0m\n",
      "\u001b[34m#011Stage 2 Generators:\u001b[0m\n",
      "\u001b[34m#011#011Fitting FillNaFeatureGenerator...\u001b[0m\n",
      "\u001b[34m#011Stage 3 Generators:\u001b[0m\n",
      "\u001b[34m#011#011Fitting IdentityFeatureGenerator...\u001b[0m\n",
      "\u001b[34m#011#011Fitting CategoryFeatureGenerator...\u001b[0m\n",
      "\u001b[34m#011#011#011Fitting CategoryMemoryMinimizeFeatureGenerator...\u001b[0m\n",
      "\u001b[34m#011Stage 4 Generators:\u001b[0m\n",
      "\u001b[34m#011#011Fitting DropUniqueFeatureGenerator...\u001b[0m\n",
      "\u001b[34m#011Useless Original Features (Count: 1): ['feature_10']\u001b[0m\n",
      "\u001b[34m#011#011These features carry no predictive signal and should be manually investigated.\u001b[0m\n",
      "\u001b[34m#011#011This is typically a feature which has the same value for all rows.\u001b[0m\n",
      "\u001b[34m#011#011These features do not need to be present at inference time.\u001b[0m\n",
      "\u001b[34m#011Types of features in original data (raw dtype, special dtypes):\u001b[0m\n",
      "\u001b[34m#011#011('object', []) : 13 | ['feature_0', 'feature_1', 'feature_2', 'feature_3', 'feature_4', ...]\u001b[0m\n",
      "\u001b[34m#011Types of features in processed data (raw dtype, special dtypes):\u001b[0m\n",
      "\u001b[34m#011#011('category', [])  : 10 | ['feature_0', 'feature_1', 'feature_2', 'feature_3', 'feature_4', ...]\u001b[0m\n",
      "\u001b[34m#011#011('int', ['bool']) :  3 | ['feature_11', 'feature_12', 'feature_13']\u001b[0m\n",
      "\u001b[34m#0110.4s = Fit runtime\u001b[0m\n",
      "\u001b[34m#01113 features in original data used to generate 13 features in processed data.\u001b[0m\n",
      "\u001b[34m#011Train Data (Processed) Memory Usage: 0.55 MB (0.0% of available memory)\u001b[0m\n",
      "\u001b[34mData preprocessing and feature engineering runtime = 0.44s ...\u001b[0m\n",
      "\u001b[34mAutoGluon will gauge predictive performance using evaluation metric: 'accuracy'\u001b[0m\n",
      "\u001b[34m#011To change this, specify the eval_metric parameter of Predictor()\u001b[0m\n",
      "\u001b[34mAutomatically generating train/validation split with holdout_frac=0.05991898952616063, Train Rows: 39222, Val Rows: 2500\u001b[0m\n",
      "\u001b[34mFitting 13 L1 models ...\u001b[0m\n",
      "\u001b[34mFitting model: KNeighborsUnif ...\u001b[0m\n",
      "\u001b[34m#011No valid features to train KNeighborsUnif... Skipping this model.\u001b[0m\n",
      "\u001b[34mFitting model: KNeighborsDist ...\u001b[0m\n",
      "\u001b[34m#011No valid features to train KNeighborsDist... Skipping this model.\u001b[0m\n",
      "\u001b[34mFitting model: LightGBMXT ...\u001b[0m\n",
      "\n",
      "2022-12-06 20:49:15 Training - Training image download completed. Training in progress.\u001b[34m#0110.8148#011 = Validation score   (accuracy)\u001b[0m\n",
      "\u001b[34m#0114.55s#011 = Training   runtime\u001b[0m\n",
      "\u001b[34m#0110.07s#011 = Validation runtime\u001b[0m\n",
      "\u001b[34mFitting model: LightGBM ...\u001b[0m\n",
      "\u001b[34m#0110.812#011 = Validation score   (accuracy)\u001b[0m\n",
      "\u001b[34m#0111.32s#011 = Training   runtime\u001b[0m\n",
      "\u001b[34m#0110.03s#011 = Validation runtime\u001b[0m\n",
      "\u001b[34mFitting model: RandomForestGini ...\u001b[0m\n",
      "\u001b[34m#0110.7828#011 = Validation score   (accuracy)\u001b[0m\n",
      "\u001b[34m#0117.38s#011 = Training   runtime\u001b[0m\n",
      "\u001b[34m#0110.21s#011 = Validation runtime\u001b[0m\n",
      "\u001b[34mFitting model: RandomForestEntr ...\u001b[0m\n",
      "\u001b[34m#0110.784#011 = Validation score   (accuracy)\u001b[0m\n",
      "\u001b[34m#0117.95s#011 = Training   runtime\u001b[0m\n",
      "\u001b[34m#0110.21s#011 = Validation runtime\u001b[0m\n",
      "\u001b[34mFitting model: CatBoost ...\u001b[0m\n",
      "\u001b[34m#0110.8204#011 = Validation score   (accuracy)\u001b[0m\n",
      "\u001b[34m#01113.75s#011 = Training   runtime\u001b[0m\n",
      "\u001b[34m#0110.02s#011 = Validation runtime\u001b[0m\n",
      "\u001b[34mFitting model: ExtraTreesGini ...\u001b[0m\n",
      "\u001b[34m#0110.7888#011 = Validation score   (accuracy)\u001b[0m\n",
      "\u001b[34m#0115.23s#011 = Training   runtime\u001b[0m\n",
      "\u001b[34m#0110.24s#011 = Validation runtime\u001b[0m\n",
      "\u001b[34mFitting model: ExtraTreesEntr ...\u001b[0m\n",
      "\u001b[34m#0110.7872#011 = Validation score   (accuracy)\u001b[0m\n",
      "\u001b[34m#0115.45s#011 = Training   runtime\u001b[0m\n",
      "\u001b[34m#0110.26s#011 = Validation runtime\u001b[0m\n",
      "\u001b[34mFitting model: NeuralNetFastAI ...\u001b[0m\n",
      "\u001b[34m#0110.7972#011 = Validation score   (accuracy)\u001b[0m\n",
      "\u001b[34m#01139.2s#011 = Training   runtime\u001b[0m\n",
      "\u001b[34m#0110.05s#011 = Validation runtime\u001b[0m\n",
      "\u001b[34mFitting model: XGBoost ...\u001b[0m\n",
      "\u001b[34m#0110.8008#011 = Validation score   (accuracy)\u001b[0m\n",
      "\u001b[34m#0113.21s#011 = Training   runtime\u001b[0m\n",
      "\u001b[34m#0110.03s#011 = Validation runtime\u001b[0m\n",
      "\u001b[34mFitting model: NeuralNetTorch ...\u001b[0m\n",
      "\u001b[34m#0110.81#011 = Validation score   (accuracy)\u001b[0m\n",
      "\u001b[34m#011106.13s#011 = Training   runtime\u001b[0m\n",
      "\u001b[34m#0110.05s#011 = Validation runtime\u001b[0m\n",
      "\u001b[34mFitting model: LightGBMLarge ...\u001b[0m\n",
      "\u001b[34m#0110.8132#011 = Validation score   (accuracy)\u001b[0m\n",
      "\u001b[34m#0112.26s#011 = Training   runtime\u001b[0m\n",
      "\u001b[34m#0110.04s#011 = Validation runtime\u001b[0m\n",
      "\u001b[34mFitting model: WeightedEnsemble_L2 ...\u001b[0m\n",
      "\u001b[34m#0110.8272#011 = Validation score   (accuracy)\u001b[0m\n",
      "\u001b[34m#0111.09s#011 = Training   runtime\u001b[0m\n",
      "\u001b[34m#0110.0s#011 = Validation runtime\u001b[0m\n",
      "\u001b[34mAutoGluon training complete, total runtime = 203.89s ... Best model: \"WeightedEnsemble_L2\"\u001b[0m\n",
      "\u001b[34mTabularPredictor saved. To load, use: predictor = TabularPredictor.load(\"/opt/ml/model/\")\u001b[0m\n",
      "\u001b[34m2022-12-06 20:52:38,201 sagemaker-training-toolkit INFO     Reporting training SUCCESS\u001b[0m\n",
      "\n",
      "2022-12-06 20:52:56 Uploading - Uploading generated training model\n",
      "2022-12-06 20:54:57 Completed - Training job completed\n",
      "ProfilerReport-1670359549: NoIssuesFound\n",
      "Training seconds: 458\n",
      "Billable seconds: 458\n"
     ]
    }
   ],
   "source": [
    "training_job_name = name_from_base(f\"{train_model_id}-training\")\n",
    "\n",
    "# Create SageMaker Estimator instance\n",
    "tabular_estimator = Estimator(\n",
    "    role=aws_role,\n",
    "    image_uri=train_image_uri,\n",
    "    source_dir=train_source_uri,\n",
    "    model_uri=train_model_uri,\n",
    "    entry_point=\"transfer_learning.py\",\n",
    "    instance_count=1,\n",
    "    instance_type=training_instance_type,\n",
    "    max_run=40*60,\n",
    "    hyperparameters=hyperparameters,\n",
    "    output_path=s3_output_location,\n",
    ")\n",
    "\n",
    "\n",
    "# Launch a SageMaker Training job by passing s3 path of the training data\n",
    "tabular_estimator.fit({\"training\": training_dataset_s3_path}, logs=True, job_name=training_job_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Benchmark table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Model</th>\n",
       "      <th>Training runtime</th>\n",
       "      <th>Validation runtime</th>\n",
       "      <th>Validation accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Weighted_Ensemble_L2</td>\n",
       "      <td>1.09</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.8272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CatBoost</td>\n",
       "      <td>13.75</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.8204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LightGBMXT</td>\n",
       "      <td>4.55</td>\n",
       "      <td>0.07</td>\n",
       "      <td>0.8148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>LightGBMLarge</td>\n",
       "      <td>2.26</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.8132</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LightGBM</td>\n",
       "      <td>1.32</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.8120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>NeuralNetTorch</td>\n",
       "      <td>106.13</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.8100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>XGBoost</td>\n",
       "      <td>3.21</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.8008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>NeuralNetFastAI</td>\n",
       "      <td>39.20</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.7972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>ExtraTreesGini</td>\n",
       "      <td>5.23</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.7888</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>ExtraTreesEntr</td>\n",
       "      <td>5.45</td>\n",
       "      <td>0.26</td>\n",
       "      <td>0.7872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>RandomForestEntr</td>\n",
       "      <td>7.95</td>\n",
       "      <td>0.21</td>\n",
       "      <td>0.7840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>RandomForestGini</td>\n",
       "      <td>7.38</td>\n",
       "      <td>0.21</td>\n",
       "      <td>0.7828</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   Model  Training runtime  Validation runtime  \\\n",
       "0   Weighted_Ensemble_L2              1.09                0.00   \n",
       "1               CatBoost             13.75                0.02   \n",
       "2             LightGBMXT              4.55                0.07   \n",
       "3          LightGBMLarge              2.26                0.04   \n",
       "4               LightGBM              1.32                0.03   \n",
       "5         NeuralNetTorch            106.13                0.05   \n",
       "6                XGBoost              3.21                0.03   \n",
       "7        NeuralNetFastAI             39.20                0.05   \n",
       "8         ExtraTreesGini              5.23                0.24   \n",
       "9         ExtraTreesEntr              5.45                0.26   \n",
       "10      RandomForestEntr              7.95                0.21   \n",
       "11      RandomForestGini              7.38                0.21   \n",
       "\n",
       "    Validation accuracy  \n",
       "0                0.8272  \n",
       "1                0.8204  \n",
       "2                0.8148  \n",
       "3                0.8132  \n",
       "4                0.8120  \n",
       "5                0.8100  \n",
       "6                0.8008  \n",
       "7                0.7972  \n",
       "8                0.7888  \n",
       "9                0.7872  \n",
       "10               0.7840  \n",
       "11               0.7828  "
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The following table data is manually copied from the training job logs\n",
    "import pandas as pd\n",
    "\n",
    "models_metadata = {\n",
    "    'Model' : [\n",
    "        'LightGBMXT',\n",
    "        'LightGBM',\n",
    "        'RandomForestGini',\n",
    "        'RandomForestEntr',\n",
    "        'CatBoost',\n",
    "        'ExtraTreesGini',\n",
    "        'ExtraTreesEntr',\n",
    "        'NeuralNetFastAI',\n",
    "        'XGBoost',\n",
    "        'NeuralNetTorch',\n",
    "        'LightGBMLarge',\n",
    "        'Weighted_Ensemble_L2'\n",
    "    ],\n",
    "    'Training runtime' : [\n",
    "        4.55,\n",
    "        1.32,\n",
    "        7.38,\n",
    "        7.95,\n",
    "        13.75,\n",
    "        5.23,\n",
    "        5.45,\n",
    "        39.2,\n",
    "        3.21,\n",
    "        106.13,\n",
    "        2.26,\n",
    "        1.09\n",
    "    ],\n",
    "    'Validation runtime' : [\n",
    "        0.07,\n",
    "        0.03,\n",
    "        0.21,\n",
    "        0.21,\n",
    "        0.02,\n",
    "        0.24,\n",
    "        0.26,\n",
    "        0.05,\n",
    "        0.03,\n",
    "        0.05,\n",
    "        0.04,\n",
    "        0.0\n",
    "    ],\n",
    "    'Validation accuracy' : [\n",
    "        0.8148,\n",
    "        0.812,\n",
    "        0.7828,\n",
    "        0.784,\n",
    "        0.8204,\n",
    "        0.7888,\n",
    "        0.7872,\n",
    "        0.7972,\n",
    "        0.8008,\n",
    "        0.81,\n",
    "        0.8132,\n",
    "        0.8272\n",
    "    ]\n",
    "}\n",
    "\n",
    "df = pd.DataFrame(models_metadata).sort_values(by='Validation accuracy', ascending=False)\n",
    "df = df.reset_index(drop=True)\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 🧪 Evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Loading the model artifacts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "2022-12-06 20:54:57 Starting - Preparing the instances for training\n",
      "2022-12-06 20:54:57 Downloading - Downloading input data\n",
      "2022-12-06 20:54:57 Training - Training image download completed. Training in progress.\n",
      "2022-12-06 20:54:57 Uploading - Uploading generated training model\n",
      "2022-12-06 20:54:57 Completed - Training job completed\n"
     ]
    }
   ],
   "source": [
    "training_job_name = 'autogluon-classification-ensemble-train-2022-12-06-20-45-49-215'\n",
    "tabular_estimator = Estimator.attach(training_job_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Deploying an endpoint for the best model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "inference_instance_type = \"ml.m5.large\"\n",
    "\n",
    "# Retrieve the inference docker container uri\n",
    "deploy_image_uri = image_uris.retrieve(\n",
    "    region=None,\n",
    "    framework=None,\n",
    "    image_scope=\"inference\",\n",
    "    model_id=train_model_id,\n",
    "    model_version=train_model_version,\n",
    "    instance_type=inference_instance_type,\n",
    ")\n",
    "# Retrieve the inference script uri\n",
    "deploy_source_uri = script_uris.retrieve(\n",
    "    model_id=train_model_id, model_version=train_model_version, script_scope=\"inference\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-------!"
     ]
    }
   ],
   "source": [
    "endpoint_name = name_from_base(f\"{train_model_id}-endpoint\")\n",
    "\n",
    "# Use the estimator from the previous step to deploy to a SageMaker endpoint\n",
    "predictor = tabular_estimator.deploy(\n",
    "    initial_instance_count=1,\n",
    "    instance_type=inference_instance_type,\n",
    "    entry_point=\"inference.py\",\n",
    "    image_uri=deploy_image_uri,\n",
    "    source_dir=deploy_source_uri,\n",
    "    endpoint_name=endpoint_name,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Quering the endpoint using the test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "bucket = sess.default_bucket()\n",
    "test_data_prefix = \"Project/starbucks-dataset/test\"\n",
    "test_data_file_name = \"data.csv\"\n",
    "\n",
    "boto3.client(\"s3\").download_file(\n",
    "    bucket, f\"{test_data_prefix}/{test_data_file_name}\", test_data_file_name\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mThe test dataset contains 4636 examples and 15 columns.\u001b[0m\n",
      "\n",
      "\u001b[1mThe first 5 observations of the data: \u001b[0m \n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Target</th>\n",
       "      <th>Feature_1</th>\n",
       "      <th>Feature_2</th>\n",
       "      <th>Feature_3</th>\n",
       "      <th>Feature_4</th>\n",
       "      <th>Feature_5</th>\n",
       "      <th>Feature_6</th>\n",
       "      <th>Feature_7</th>\n",
       "      <th>Feature_8</th>\n",
       "      <th>Feature_9</th>\n",
       "      <th>Feature_10</th>\n",
       "      <th>Feature_11</th>\n",
       "      <th>Feature_12</th>\n",
       "      <th>Feature_13</th>\n",
       "      <th>Feature_14</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>M</td>\n",
       "      <td>59</td>\n",
       "      <td>59000.00</td>\n",
       "      <td>2017</td>\n",
       "      <td>12</td>\n",
       "      <td>31</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>M</td>\n",
       "      <td>36</td>\n",
       "      <td>46000.00</td>\n",
       "      <td>2017</td>\n",
       "      <td>7</td>\n",
       "      <td>30</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>M</td>\n",
       "      <td>36</td>\n",
       "      <td>73000.00</td>\n",
       "      <td>2017</td>\n",
       "      <td>12</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>M</td>\n",
       "      <td>72</td>\n",
       "      <td>39000.00</td>\n",
       "      <td>2017</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>M</td>\n",
       "      <td>53</td>\n",
       "      <td>73000.00</td>\n",
       "      <td>2018</td>\n",
       "      <td>5</td>\n",
       "      <td>15</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1</td>\n",
       "      <td>M</td>\n",
       "      <td>90</td>\n",
       "      <td>108000.00</td>\n",
       "      <td>2018</td>\n",
       "      <td>6</td>\n",
       "      <td>22</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>54</td>\n",
       "      <td>37000.00</td>\n",
       "      <td>2014</td>\n",
       "      <td>6</td>\n",
       "      <td>26</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>5</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>118</td>\n",
       "      <td>70598.23</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>28</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1</td>\n",
       "      <td>M</td>\n",
       "      <td>48</td>\n",
       "      <td>70000.00</td>\n",
       "      <td>2017</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>discount</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>49</td>\n",
       "      <td>50000.00</td>\n",
       "      <td>2017</td>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>7</td>\n",
       "      <td>bogo</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Target Feature_1  Feature_2  Feature_3  Feature_4  Feature_5  Feature_6  \\\n",
       "0       1         M         59   59000.00       2017         12         31   \n",
       "1       1         M         36   46000.00       2017          7         30   \n",
       "2       1         M         36   73000.00       2017         12         12   \n",
       "3       1         M         72   39000.00       2017          9          9   \n",
       "4       1         M         53   73000.00       2018          5         15   \n",
       "5       1         M         90  108000.00       2018          6         22   \n",
       "6       0         M         54   37000.00       2014          6         26   \n",
       "7       0         F        118   70598.23       2017          8         28   \n",
       "8       1         M         48   70000.00       2017         12          5   \n",
       "9       0         M         49   50000.00       2017          8          6   \n",
       "\n",
       "   Feature_7  Feature_8  Feature_9 Feature_10  Feature_11  Feature_12  \\\n",
       "0          5          5          7       bogo        True        True   \n",
       "1          5          5          5       bogo        True        True   \n",
       "2          5          5          7       bogo        True        True   \n",
       "3         10         10          7       bogo        True        True   \n",
       "4          2         10          7   discount        True        True   \n",
       "5         10         10          7       bogo        True        True   \n",
       "6         10         10          5       bogo        True        True   \n",
       "7          5          5          5       bogo        True        True   \n",
       "8          2         10         10   discount        True        True   \n",
       "9         10         10          7       bogo        True        True   \n",
       "\n",
       "   Feature_13  Feature_14  \n",
       "0        True       False  \n",
       "1        True        True  \n",
       "2        True       False  \n",
       "3       False        True  \n",
       "4        True       False  \n",
       "5       False        True  \n",
       "6        True        True  \n",
       "7        True        True  \n",
       "8        True        True  \n",
       "9       False        True  "
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "newline, bold, unbold = \"\\n\", \"\\033[1m\", \"\\033[0m\"\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# read the data\n",
    "test_data = pd.read_csv(test_data_file_name, header=0)\n",
    "test_data.columns = [\"Target\"] + [f\"Feature_{i}\" for i in range(1, test_data.shape[1])]\n",
    "\n",
    "num_examples, num_columns = test_data.shape\n",
    "print(\n",
    "    f\"{bold}The test dataset contains {num_examples} examples and {num_columns} columns.{unbold}\\n\"\n",
    ")\n",
    "\n",
    "# prepare the ground truth target and predicting features to send into the endpoint.\n",
    "ground_truth_label, features = test_data.iloc[:, :1], test_data.iloc[:, 1:]\n",
    "\n",
    "print(f\"{bold}The first 5 observations of the data: {unbold} \\n\")\n",
    "test_data.head(10)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "content_type = \"text/csv\"\n",
    "\n",
    "\n",
    "def query_endpoint(encoded_tabular_data):\n",
    "    # endpoint_name = endpoint_name\n",
    "    client = boto3.client(\"runtime.sagemaker\")\n",
    "    response = client.invoke_endpoint(\n",
    "        EndpointName=endpoint_name, ContentType=content_type, Body=encoded_tabular_data\n",
    "    )\n",
    "    return response\n",
    "\n",
    "\n",
    "def parse_response(query_response):\n",
    "    model_predictions = json.loads(query_response[\"Body\"].read())\n",
    "    predicted_probabilities = model_predictions[\"probabilities\"]\n",
    "    return np.array(predicted_probabilities)\n",
    "\n",
    "\n",
    "# split the test data into smaller size of batches to query the endpoint due to the large size of test data.\n",
    "batch_size = 1500\n",
    "predict_prob = []\n",
    "for i in np.arange(0, num_examples, step=batch_size):\n",
    "    query_response_batch = query_endpoint(\n",
    "        features.iloc[i : (i + batch_size), :].to_csv(header=False, index=False).encode(\"utf-8\")\n",
    "    )\n",
    "    predict_prob_batch = parse_response(query_response_batch)  # prediction probability per batch\n",
    "    predict_prob.append(predict_prob_batch)\n",
    "\n",
    "\n",
    "predict_prob = np.concatenate(predict_prob, axis=0)\n",
    "predict_label = np.argmax(predict_prob, axis=1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcwAAAHjCAYAAACuIS9lAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8/fFQqAAAACXBIWXMAAAsTAAALEwEAmpwYAAA4M0lEQVR4nO3deVxXVf7H8Teryu6K5gYuIIjgnoqauaSmaWWNS6KZS5aZ06SNWlmZv1JbbFLLykzLJXOr1CaXzBItc8EdwZXEHREQQUD4/v4wGb+xeFDwi/p6Ph7zmOGec+75XBh8c+89937tLBaLRQAAIF/2ti4AAIDbAYEJAIABAhMAAAMEJgAABghMAAAMEJgAABggMIFCEhkZqf79+6tJkyby9/fX1KlTi2SepUuXyt/fX5s3by6S/d9J/P39NXr0aFuXgTuEo60LAG5WamqqFi5cqNWrV+vgwYO6ePGiPD09VbduXXXu3FndunWTo2PR/l/98uXLGj58uC5fvqwRI0bI3d1d/v7+RTqnLcXGxqpdu3aSpDZt2uiTTz7J0ScjI0OtWrXS+fPnVblyZa1bt+6G5lq7dq0iIyM1fPjwm6oZuFkEJm5rMTExGjJkiI4ePaoWLVpoyJAhKl26tM6dO6fffvtNY8aM0cGDB/XSSy8VaR3Hjh3TsWPHNHr0aPXt27dI5+revbu6dOkiJyenIp3HRIkSJbRhwwadOXNGFSpUsGpbt26dzp8/rxIlStzUHGvXrtWyZctuKDB37dole3supKFwEJi4bV26dElPP/20YmNjNXXqVD3wwANW7UOGDNGuXbu0e/fuIq8lLi5OkuTp6Vnkczk4OMjBwaHI5zFx//33a+3atfruu+80ePBgq7YlS5bI399fWVlZSklJuWU1Xbp0SY6OjnJ0dLzpsAauxZ9euG0tWrRIR44c0YABA3KE5VXBwcF64oknrLatXbtWvXr1UoMGDdSgQQP16tVLa9euzTG2bdu2CgsL06FDhzRkyBA1aNBAjRo10vPPP6+zZ89m9wsLC8s+qxwzZoz8/f3l7++v2NjYfO83hoWFqW3btlbbtm/frkGDBik0NFT16tVTq1atNHjwYO3YsSO7T177jI+P1xtvvKH77rtPQUFBuu+++/TGG2/o/PnzVv2ujv/tt9/0+eefq3379goKClLHjh21bNmyXL+PeSlbtqxat26tpUuXWm0/c+aMwsPD9eijj+Y6bteuXRo9erQ6duyokJCQ7J/DmjVrcnyPrtZ09fvq7++fPd/o0aPl7++v+Ph4jRkzRi1atFD9+vV16tSp7DHX3sOcN2+e/P39NX36dKt5Tp8+rWbNmqlz585KTU0t0PcAdw/OMHHbWrVqlSSpZ8+exmPmzZun8ePHq0aNGnrmmWckScuWLdOwYcM0fvz4HPs6ffq0+vXrp/bt2+ull17S/v37tXDhQiUnJ2vWrFmSpKFDh6phw4aaMWOGevbsqUaNGkmSypQpU6DjOXz4sJ566imVK1dO/fr1U9myZRUXF6ft27dr//79ql+/fp5jL1y4oN69eysmJkY9evRQYGCgIiMjtWDBAv3+++9atGiR3NzcrMZMmTJFly5dUs+ePeXs7KwFCxZo9OjRqlatWvYxmOjRo4eGDRumiIgINWjQQJL07bffyt7eXt26ddPixYtzjFmzZo0OHz6sTp06qXLlykpISNCyZcv03HPP6d1339VDDz0k6cr3NisrS1u3btXkyZOzxzds2NBqfwMGDFC5cuX07LPPKiUlRS4uLrnW+sQTT+j333/X9OnTde+996px48bKysrSqFGjdPHiRc2ePVulSpUyPnbcXQhM3LYOHDggV1dXVa1a1ah/YmKi3n33XVWrVs0qQPr06aOHH35YEydOVOfOneXh4ZE9JiYmRlOmTNGDDz6Yvc3e3l7z58/XoUOHVLNmTYWGhsrR0VEzZsxQ/fr11b179xs6nvDwcKWmpur9999XcHBwgcbOnDlTR48e1bhx46zOqAMCAjR+/HjNnDlT//znP63GpKena/HixXJ2dpYkderUSe3atdO8efMKFJht2rRRuXLltHTp0uzAXLp0qdq2bZvnHw3PPPOMXnzxRattYWFhevjhh/Xxxx9nB2ZoaKiWL1+urVu35vt9rV27tt59912jeidMmKC9e/dq5MiR+u677zR37lxt3rxZr776qurUqWO0D9yduCSL21ZycnKOs6b8bNy4USkpKQoLC7Ma5+bmpr59+yolJUWbNm2yGlOhQgWrsJSkZs2aSZL+/PPPm6g+J3d3d0nSTz/9pLS0tAKNXbNmjcqUKZPjDLlnz54qXbp0rpec+/Tpkx2WkuTt7S1fX18dPXq0QHM7OjqqW7du+uGHH5Samqpt27bpyJEj6tGjR55jrj0DTE1N1fnz55WamqpmzZrp0KFDSk5OLlANAwcONO7r6empd999V2fPntXgwYM1ffp0tW3btsgXa+H2xxkmbltubm66ePGicf/Y2FhJV85G/s7Pz0/SldWu18rt7NXLy0uSlJCQYDy3iS5duuj777/XjBkzNHv2bIWEhKhly5bq0qWLKleunO/Y2NhYBQUF5Xh8xtHRUb6+vtq3b1+OMXkd2/Hjxwtc+2OPPaZZs2Zp9erV2rx5sypUqKCWLVvm2f/cuXP64IMP9NNPP+ncuXM52pOSkgr0x5CPj0+B6m3YsKEGDRqkGTNmqHz58nrrrbcKNB53JwITt63atWtry5YtOnbsmPFl2YLKbzWqyUfJ2tnZ5dl2+fJlq6+dnZ31xRdfaNeuXdqwYYO2bt2qDz/8UNOmTdN7772nDh06mBduoDAft6hZs6ZCQkI0f/58RUdHq2/fvnl+7ywWi5566ikdOnRIYWFhqlevntzd3eXg4KAlS5ZoxYoVysrKKtD8Bb3vmJ6ervDwcElX/vA5efKkSpcuXaB94O7DJVnctq6ujF20aJFR/6uheuDAgRxtBw8etOpTWK4+ZpKYmJij7eoZ798FBwdr2LBh+uKLL7RmzRqVKlVKH3zwQb7zVK1aVUeOHMkRwpcvX9bRo0eL7A+Ka/Xo0UM7duxQSkpKnqtjJSkqKkr79+/XkCFD9O9//1sPPvigWrVqpRYtWuQalPn90XGj3n//fe3Zs0ejRo2Sm5ubXnjhhVv66AtuTwQmbluPP/64fH19NWvWrFzv0UnSnj17NG/ePElXFpC4uLho7ty5VvfIkpOTNXfuXLm4uCg0NLRQa7x6qfDv90ZXrFihM2fOWG2Lj4/PMb5ixYoqU6ZMroF7rfbt2ys+Pj7HHw/ffPON4uPj1b59+xuovmC6dOmi5557Ti+//LJ8fX3z7Hf1zPbvZ+jR0dE5HiuR/ne/s7Augf/yyy+aPXu2HnnkEQ0aNEgTJ07U0aNH9eabbxbK/nHn4pIsblulSpXSJ598oiFDhmjYsGFq2bKlWrRoIS8vL8XHx2vz5s0KDw/XoEGDJEkeHh4aOXKkxo8fr3/84x965JFHJF15rCQmJkbjx4/PXnhTWGrUqKEWLVpo4cKFslgsCggIUGRkpNauXavq1atbnRF+/PHH2rhxo9q0aaMqVarIYrHo559/1uHDh7OPIS+DBg3Sjz/+qPHjx2vfvn3Z8yxevFi+vr7XHV8Y3NzcjN7GU7NmTdWuXVszZ87UpUuX5OvrqyNHjmjhwoXy8/PT3r17rfqHhIRo7ty52c+YOjk5KTg4+IbOms+cOaPRo0erevXqevXVVyVdWeXbr18/ffnll9n3jIHcEJi4rVWvXl3ffvutFi5cqFWrVmnGjBlKSUmRp6engoKCNHHixOxHFKQrz+FVqFBBn3/+efbD63Xq1NH06dOL7Cxs8uTJevPNN7V8+XJ9//33atSokb788ku9/vrrVgts2rdvr7Nnz+rHH39UXFycSpYsqerVq2vChAl67LHH8p3D3d1dCxYs0Icffqh169Zp6dKlKlu2rHr16qXhw4cXaAFNUXNwcNAnn3yiSZMmadmyZUpNTVXt2rU1adIk7d+/P0dgdu3aVZGRkVq5cqV+/PFHZWVl6e233y5wYGZlZemll15ScnKyPv/8c7m6uma3jRo1Slu3btW4ceNuOIxx57OzmKxcAADgLsc9TAAADBCYAAAYIDABADBAYAIAYIDABADAAIGJQnHkyBH17NlTHTt2VM+ePQv8Am/gbjFp0iS1bdtW/v7+io6OtnU5KAACE4XitddeU58+fbRq1Sr16dNH48aNs3VJQLF09SPUrvdCfRQ/BCZu2rlz57Rv3z517dpV0pUHzfft25frq96Au13jxo1VqVIlW5eBG0Bg4qadPHlS3t7e2Z9O4eDgoAoVKujkyZM2rgwACg+BCQCAAQITN61SpUo6ffq0MjMzJUmZmZk6c+YMl50A3FEITNy0smXLKiAgQCtWrJB05aOrAgICVKZMGRtXBgCFh5evo1AcOnRIo0ePVlJSkjw8PDRp0iTVqFHD1mUBxc6ECRO0evVqxcXFqXTp0vLy8tLKlSttXRYMEJgAABjgkiwAAAYITAAADBCYAAAYIDABADBAYKLQJCUlaerUqUpKSrJ1KUCxxu/K7YnARKFJSkrStGnT+EcAuA5+V25PBCYAAAYITAAADBCYAAAYcLR1ATciKytLFy9elJOTk+zs7GxdDv6SmZmpcuXKKTMzU2lpabYuByi2+F0pniwWizIyMuTq6ip7+5znk7flq/EuXLig6OhoW5cBALgD+fn5yd3dPcf22/IM08nJSZKUUtJbFvvb8hDuWK4px3XRpbKty8Df+FV0s3UJ+Jvjh6NUuYa/rcvANS5nZOj0scPZGfN3t2XaXL0Ma7F3JDCLIX4mxY+Tk7OtS0Au+LkUT3nd6mPRDwAABghMAAAMEJgAABggMAEAMEBgAgBggMAEAMAAgQkAgAECEwAAAwQmAAAGCEwAAAwQmAAAGCAwAQAwQGACAGCAwAQAwACBCQCAAQITAAADBCYAAAYITAAADBCYAAAYIDABADBAYAIAYIDABADAAIEJAIABAhMAAAMEJgAABghMAAAMEJgAABggMAEAMEBgAgBggMAEAMAAgQkAgAECEwAAAwQmAAAGCEwAAAwQmAAAGCAwAQAwQGACAGCAwAQAwACBCQCAAQITAAADBCYAAAYITAAADBCYAAAYIDABADBAYAIAYIDABADAAIEJAIABAhMAAAMEJgAABghMAAAMEJgAABggMAEAMEBgAgBggMAEAMAAgQkAgAECEwAAAwQmAAAGCEwAAAwQmAAAGCAwAQAwQGACAGCAwAQAwACBCQCAAQITAAADBCYAAAYITAAADBCYAAAYIDABADBAYAIAYIDABADAAIEJAIABAhMAAAMEJgAABghMAAAMEJgAABggMAEAMEBgAgBggMAEAMAAgQkAgAECEwAAAwQmAAAGCEwAAAwQmAAAGCAwAQAwQGACAGCAwAQAwACBCQCAAQITAAADBCYAAAYITAAADDjaugDcOjEH92vfji06Gr1PR6L3KSE+To5Ozvpoyc/G+/jPa//S3ojNkqR35nwvz9JlrdovXbqk37b8mD1H7JGDunw5Q4/0G6rOj4Vdv8ZDUVrz7QJF745QclKiXNzcdU/1GmrZ4SE1bd2+YAcMFKGdEds1/6vZ2rN7p5ISk+Tq5ir/OgF65LGeatPW+v+rLRrVu+7+GjZuqmmffG617VJqquZ/NVvr1q7W8dhYWSxZ8q5YSc1DW6nfgIEqU7ZcoR4T8kdg3kVWLJytnZs33PD4zb+s1t6IzbKzs5PFYsm1z6lTp/TFlDdvaP+//HeZFnwyRZLk6xeo2nXrKzHhnP48GKXtrm4EJoqNH39YrjfHvSyLxaLAuvXUsHFTnT19Wlv/2Kwtm3/XkwOHaMizw7P7P9i1W459JCclyM3DS5vCNygh4bzqN2ho1Z6amqJnBj6p6KhIeXmVVuOm98rBwUH79uzWNwvmat3a1fps9jx5V6xY5MeLKwjMu0jNOkGq6ltLPrUD5FMrQCP7P2Q89mJykr6Z+aEC6jfRmRPHdO7MqVz7lSpVSi07dJVP7UD51A7Q9t/W64dv5lx3//t2bNH8Ge+pTHlvDXtlsqr41Mxuy8hI1+nYP41rBYpSenq6pkx+W5L09rsf6L7722W3RWzbohHPDtGXX8xU90cfzw6zV974vxz7ORq1W2Xvqa61q3+UJHV8sKtV+5JvvlZ0VKTqN2ykd//zkVxcXCRJaWlpenXMSIX/sl6zPpuhMa++XhSHiVxwD/Mu0qlHX3XrM0jBTULlUbpMgcYunjVdqSkX1Wfoi/n28/b2Vr/hY9S6U3dVq+kne3sHo/1//ekUWSwWDXnpTauwlCQnJ2dV8a1VoHqBonLo4AFduHBBtfz8rcJSkho0aqJGTe5VVlaWoqMir7uvdWtWKz09XXXrBatqtepWbTu2b5Mk9XqiX3ZYSlKJEiXUb8AgSVJU5L6bPRwUAIGJ64raE6FNP61Upx595X1P1ULf/4G9O3QqNka1AoPl6xdY6PsHCpOzk5NRP09Pr+v2WfXfFZKkTg/mvNrj7Ox8/Tm8PI1qQeEgMJGvjIx0zZ0+WeUrVlbnx6+/aOdGRO2OkCTVCW6slOQLWv/DUs376B198/mH2rbxZ2VmXi6SeYEbUd3HVxW8vXUwOkq/rl9n1RaxbYu2bdmsatV9FBiU/0KfuLg47YzYLkdHR7V/oFOO9sZNm0mSvp73pVJSUrK3p6Wl6csvZkqSOnXJeW8URcem9zCPHDmi0aNHKyEhQV5eXpo0aZJ8fHxsWRL+5r+LvtTp43/qn29MkZPT9f/ivREnjx2VJGVmXtbrz/VVQnxcdtva7xaqcvWaGj7uHZUp710k8wMF4ejkpFden6DRI/+p0S+OUN2gYFW85x7FnTmjXTsjFFy/gV594//k6Jj/P68bN26UxWJRs9CW8vTyytHe7eFHtfm3jdrwy8967KFOCgoOkb29vfbt2a20tEt67p8j1bmL+ToE3DybnmG+9tpr6tOnj1atWqU+ffpo3LhxtiwHf3MyNkY/Lp6rxq3aKbBB0yKbJyX5giRp1dJ5sndw0IjX39d/vl6tcf+ZozohjXU85pBmTHw5z5W5wK3WuGkzTfvkc3l7V9TePbv00+oftXPHdrm4uKpBw8YqU6bsdfcRHh4uKffLsdKVYP6/ye/rsZ59lJBwXuG/rtev69cpLu6savn5K6Bu3UI9JlyfzQLz3Llz2rdvn7p2vbIyrGvXrtq3b5/i4+NtVRKuYbFYNHfaJDk6O+sfA58v0rmysjKv/Hdmpp4Z+7bqNrxXpVxcVcW3loa/OlleZcvr6IFIRe7YUqR1AKZ+/GG5nh4Qptp16mjOgsVaF/6H5i/+Tve376DZn3+q558drMsZGXmOj4rcp+PHj8vd3V0tW7fJtU9SYqKGPz1Qq39cqbGvjdfyVT/rx3XheuudKToee0zPDx2sX37+qYiOELmx2SXZkydPytvbWw4OV1ZROjg4qEKFCjp58qTKlDFbwemacrwoS7wr2Mkit+SYHNt//vlnHdi3U/3791cV54tS8sX/jcm6ck/R9WKs3JySc4y9dn/O6QmSpBLpCbnOI0luznaSpGrVqqmud0npb/1aNr9XK1as0NGIX9S0Npdlb8TRKFtXcOc4efKk/u/1V1W1alUNHTxI9pZ0nfrzoCSpT8/HdfzPo4qIiNCcz6arXbt2ue7jm3lfSZIaN26sE0dy/+F8/PHH2rlju1544QUF+dfShXMnJUk+lb313LBheu211/TexAmq4l02+99RFK3b+jnMiy6VZbG/rQ/B5iyyU7Jb9Rzb/9i1X5K0cUuEftu2y6otITFJkvTuhx/J0dFJ3fsOVu3AEElXwvLa/aU7e0mS0py9cp1Hkjwq+UraotIVq+Xax6NKHUkrdC4lK899IH9BlT1sXcId4+dfNyozM1MdOndVjYCQHO1dH3lcERER+jP2hHz8cy78yczM1JYtWyVJj/Xpl2ef33//Xc7Oznq0d/8cgejjX08zPvlUJ47HysnVS9Wq+xTOwd3lMjLSdfxw3n9d2ixtKlWqpNOnTyszM1MODg7KzMzUmTNnVKlSJVuVhFwc3Lcrz7bD+/dIkpKTEm9qjqq+tSX9717m3128cGX/ziVL3tQ8QGE4c+a0JMnNzS3X9qvbk5KScm3f+sdmxcWdVbly5VS/QaNc+5yPj1dGRoY8PD3zPHt0c3P/a56b+/2DOZsFZtmyZRUQEKAVK1aoe/fuWrFihQICAowvx6JoDXt5Yp5tYwb10Lkzp3J9l+yNCGrUXPb2Djp29KBSki/I5a9/CK6K2nPlsZNqNfxuei7gZpX96/2tkfv25tq+d89uSVKle+7Jtf3qs5ctW7aUnZ1drn08PD3l5OSkpMREHY89pspVrJ9/Tr5wQX/GHJUkVayY+zwofDZdJfv6669r7ty56tixo+bOnas33njDluXARjy8Sqt52866lHJRCz/7QJcv/++5y/A1K7R/51a5eXipQfP7bFglcEWr++6XJK3+78oci24itm/Vwr/uT97frkOOsZdSU/XrX2NCQ0PznMPZ2Vn3Nr/SPun/3rA6i7yUmqqJE17XpUupCg5poHLly9/cAcGYTW8A1qxZU4sWLbJlCXeVXVs2aeXCL6y2ZV7O0NsjB2d/3aXnAAU3aXFT83z01hgl/vUs5flzZyVJ639Yph2//ypJ8ixTTs+OfdtqzONPPafDUXv0288/KnrvDlWr6a9zZ07pz0NRcnJ21sB/jVPJUi4CbM0/IFB9wp7U/K9ma8zIf8rPP0BVqlbV6dOntG/PblksFnV/9DE1ubd5jrG/rl+nlJQUBdQNUuXKlfOd5/l/vaR9e3dr6x+b9Y+Hu6huUD05ODoqcs8enTsXJw9PT40a+2pRHSZywYqZu0hy0nkdibZ+96TFYrHalpx0/qbnOXY4OsfL2c/Hndb5uCv3fspWyPnpCi5u7hrzzqdasXC2tm9ar91bNqmUq6satrhfXXs+ybtkUaw8988XVS+4vpYt+UZRkft06OABubi6qEGjJur28KN6oHOXXMddvRzbsXPXXNuvVaVqVX25YLG+mjNLv28M17atWySLRRW8K6pHu14Ke3KgKnjzSSW3kp3lNnwaPC0tTXv27GGVbDH091WyKB5YJVv8HI3anesKWdjO1VWyQUFBKlGiRI523iULAIABAhMAAAMEJgAABghMAAAMEJgAABggMAEAMEBgAgBggMAEAMAAgQkAgAECEwAAAwQmAAAGCEwAAAwQmAAAGCAwAQAwQGACAGCAwAQAwACBCQCAAQITAAADBCYAAAYITAAADBCYAAAYIDABADBAYAIAYIDABADAAIEJAIABAhMAAAMEJgAABghMAAAMEJgAABggMAEAMEBgAgBggMAEAMAAgQkAgAECEwAAAwQmAAAGCEwAAAwQmAAAGCAwAQAwQGACAGCAwAQAwACBCQCAAQITAAADBCYAAAYITAAADBCYAAAYIDABADBAYAIAYIDABADAAIEJAIABAhMAAAMEJgAABghMAAAMEJgAABggMAEAMEBgAgBggMAEAMAAgQkAgAECEwAAAwQmAAAGCEwAAAwQmAAAGCAwAQAwQGACAGCAwAQAwIBxYO7atUvffPON1ba1a9fqoYceUqtWrfT+++8XenEAABQXxoE5bdo0rVu3LvvrEydO6MUXX9TZs2fl7u6uzz77TEuWLCmSIgEAsDXjwNy/f78aNmyY/fXKlStlsVj03Xff6YcfflBoaGiOM1AAAO4UxoGZkJCgcuXKZX8dHh6uJk2ayNvbW5LUtm1bHT16tNALBACgODAOTA8PD8XFxUmS0tPTtXPnTjVu3Di73c7OTmlpaYVfIQAAxYCjacc6depo8eLFatGihdasWaO0tDS1bNkyuz02NlZly5YtkiIBALA148B89tlnNXDgQD3++OOyWCwKDQ1VvXr1stvXr1+vkJCQIikSAABbMw7Mhg0baunSpQoPD5e7u7sefPDB7Lbz588rNDRUHTp0KJIiAQCwNePAlCRfX1/5+vrm2F66dGmNHTu20IoCAKC44U0/AAAYyPMMs1+/fgXemZ2dnebMmXNTBQEAUBzlGZixsbG3sg4AAIq1PAPz2tfgAQBwt+MeJgAABghMAAAMFOixksTERC1evFg7d+5UUlKSsrKyrNpZ9AMAuFMZB+bx48fVu3dvnTlzRu7u7kpOTpanp2d2cJYuXVqlSpUqyloBALAZ40uyH3zwgS5cuKDZs2dr1apVslgsmjJlirZt26ann35arq6umj9/flHWCgCAzRgH5m+//abHH39czZo1k52dXfb2UqVK6YUXXpCfn5/eeeedIikSAABbK9DnYdauXVuS5OTkJEm6dOlSdntoaKg2bdpUyOUBAFA8GAdmmTJllJiYKElydXVViRIldPz48ez2jIwMqwAFAOBOYhyYtWvX1v79+yVdWQ0bHBys+fPn68SJE4qNjdXChQtVo0aNIisUAABbMg7Mtm3baseOHdlnkc8++6xiYmLUrl07dejQQTExMXr22WeLrFAAAGzJ+LGSJ554Qk888UT2182bN9fXX3+tFStWyN7eXh06dFDDhg2LpEgAAGytQC8u+Lt69eqpXr16hVULAADFFq/GAwDAgPEZ5pgxY67bx87OTm+99dZNFQQAQHFkHJjLli27bh8CEwBwpzIOzKuPlFwrMzNTx44d06xZsxQdHa2ZM2cWanEAABQXN3UP08HBQT4+Pho/fry8vLx4NR4A4I51U6tkr9WqVStNmzZNb7zxRmHt8roa+5RWiRIlbtl8uL4dETFqWbucrcvA36wK323rEvA33q7S5u1Rti4D17C3s6i8Sz7thTVRQkKCUlJSCmt3AAAUKzd9hpmUlKRNmzZpzpw5qlu3bmHUBABAsWMcmHXq1LH6WK9rWSwWeXp6avTo0YVWGAAAxYlxYD788MO5BqaXl5d8fHzUpUsXubm5FWpxAAAUF8aBOXHixKKsAwCAYs140c+0adMUHR2dZ/uBAwc0bdq0QikKAIDipkCBGRWV9xLoAwcOaPr06YVSFAAAxU2hPVaSlpYmBweHwtodAADFSr73MJOTk5WUlJT9dUJCgk6cOJGjX2JiopYvX65KlSoVfoUAABQD+Qbm7Nmzsy+zXn2xel4vV7dYLBo1alThVwgAQDGQb2A2bdpU0pUwnD59ujp06CB/f/8c/VxdXRUSEqKGDRsWTZUAANjYdQPzamieOHFCvXr1UkhIyC0pDACA4sT4Ocy33367KOsAAKBYM14lO2/ePD355JN5tj/11FP6+uuvC6MmAACKHePAXLp0qapXr55nu4+Pj5YsWVIoRQEAUNwYB2ZMTIz8/PzybK9Vq5ZiYmIKpSgAAIob48C8fPmy0tPT82xPT09XWlpaoRQFAEBxYxyYPj4+2rhxY57t4eHhqlatWqEUBQBAcWMcmF26dNHGjRv1wQcfWJ1pZmRk6MMPP9TGjRvVtWvXIikSAABbM36s5Mknn9Svv/6qGTNmaMGCBapRo4bs7Ox06NAhJSYmqnHjxhowYEBR1goAgM0YB6aTk5NmzZql2bNna8WKFYqMjJR05VLtkCFD1L9/f2VlZRVZoQAA2JJxYEpXQnPw4MEaPHiw1fY9e/ZowoQJ+u9//6vNmzcXaoEAABQHBQrMayUkJOj777/X4sWLdeDAAVksFvn4+BRiaQAAFB8FDswNGzZoyZIlWrdunTIyMuTj46Nhw4apY8eOql27dlHUCACAzRkF5rFjx7R06VJ9++23OnXqlMqUKaOOHTtqxYoVeuGFF/TAAw8UdZ0AANhUvoG5fPlyLV68WFu2bJGDg4PatGmjV155RW3atFFsbKyWL19+q+oEAMCm8g3MUaNGqWrVqho7dqy6du0qLy+v7DY7O7uirg0AgGIj3xcXODk56fjx4/rpp5/066+/6tKlS7eqLgAAipV8A3Pjxo0aO3asEhIS9NJLL6lFixYaO3astmzZIovFcqtqBADA5vK9JOvh4aG+ffuqb9++2rt3rxYvXqwffvhBy5YtU5kyZWRnZ6cLFy7cqloBALAZ43fJ1q1bV6+99po2bNigyZMnq1atWpKkV155Rd27d9dHH32kAwcOFFmhAADYknFgXuXs7KyHHnpIc+bM0Zo1azR06FAlJSXpww8/VPfu3YuiRgAAbK7AgXmtKlWqaMSIEVq3bp0+/fRTdejQobDqAgCgWLnhV+Ndy87OTq1bt1br1q0LY3cAABQ7N3WGCQDA3YLABADAAIEJAIABAhMAAAMEJgAABghMAAAMEJgAABggMAEAMEBgAgBggMAEAMAAgQkAgAECEwAAAwQmAAAGCEwAAAwQmAAAGCAwAQAwQGACAGCAwAQAwACBCQCAAQITAAADBCYAAAYITAAADBCYAAAYIDABADBAYAIAYIDABADAAIEJAIABAhMAAAMEJgAABghMAAAMEJgAABggMAEAMEBgAgBggMAEAMAAgQkAgAECEwAAAwQmAAAGCEwAAAwQmAAAGCAwAQAwQGACAGCAwAQAwACBCQCAAQITAAADBCYAAAYITAAADBCYAAAYIDABADBAYAIAYIDABADAAIEJAIABAhMAAAMEJgAABghMAAAMEJgAABggMAEAMEBgAgBgwNHWBaB46N8vTFu2/JFn+yeffqZWrVrnu48hgwcqPDxckvTLr+EqX768Vfsff2zWk/375Tk+ODhEXy/8pgBVA0XnWMwhRe3bpT+PHNCfRw4qMSFejo5OeufjBbn2f2HwY9fdZy3/IA0b+brVtrRLqVq36nvt2LpJ8XFnVKJkSVWv4acOD/aQT02/HPv48fuFWrV80XXnem7UeNX0C7xuP5gjMGHlgQc6ysXFJcd2b2/vfMetWLFc4eHhsrOzk8Viybdv1WrV1Khho5zbq1YtWLFAEVq9YrH27Nhi3L9J8zZ5tu3bvV0Xk5NUo3aA1fZLqSmaOnmcTsQelbunl+oE1VfyhURF7o7Q/j0R6jfkXwpp1MxqTOWqvnnOlZgQr+jIXXJ2LqEq1XyNa4cZAhNWRr30kipXrlKgMYmJiZo08W21aBGq6OgoxcXF5du/UcNGeuvtiTdTJlDkfGr46Z4qPqrmU1PVfGtp3IuD8u3f56nnct2emnJREVs2SpIaN7O+SrN8yVydiD0q/8AQDXhmpEqULCXpSsB+Pn2SFsyerpp+gXJz98geU69BU9Vr0DTXuZYv/krRkbtUr0HT7H2h8HAPEzftncmTdOHCBb3y6jhblwIUmnadH1Hn7j1VN6Sx3D28bng/O7Zu0uXLGapew0/lvStlb8/IyNCW39bL3t5e/wh72irgAus1VJNm9yntUqo2rl9lNI/FYtH27GC+74brRd4ITNyULX/8oWXLlmrQ4CHy8fGxdTlAsbNt8wZJOc8uT5w4oYz0dJUpV0FlylXIMa5WnSBJMr4sfDBqrxLi4+Tu6SW/wHo3WTVywyVZWFmyeLESEhJkb2+v6j4+ateuve65555c+6anp+v118epatVqGjLkaeM5YmKO6v3331NiQoK8SpdWw4aN1KpVK9nb8/cb7iznz53V4QORcnBwVIMmoVZtaWlpkiQXV7dcx7r+tf3kiT+VlZUpe3uHfOfa9vuvkqSGTVpety9ujM0Cc9KkSVq1apWOHz+u5cuXy88v52ow3HozZnxs9fU770zWM0Of0TPPDsvR95NPZujIkSOaOXOWnJ2djeeIiIhQRESE1TY/Pz998J+pnKXijrJt8wZZLBYFBNWXq5u7VZu7+5Wvz5/L/Z5//LmzkqTMy5eVlJAgrzJl85wnIyNdO7f/Lklq3Dz/1ey4cTb7k75du3aaN2+eKleubKsScI3GjRtr4qTJWrV6rbZH7NQPP/yoEf98QY4ODpo69UN99eUcq/6HDx/SzM8+VefOD6pFaGgee7Xm5uaup54aqAVff6NNv23Wpt82a9YXsxUSUl/R0dEaNPApXbhwoSgOD7CJrX+d9TVqnvOeYsWKFeXhWVoXkhK0b/f2HO1/bPw5+3+npaXmO8+eHVt1KTVFFe+pqirVatxk1ciLzQKzcePGqlSp0vU74pYY/vwIdevWXVWrVlXJkiXl4+urp58eqqnTpkuSpk2bqkuXLkm6srjgtXHjVKJECY0ePcZ4jsDAQI0c9ZJCQkLk5eUlLy8vNWvWXHPnzVejRo114sRxLVgwv0iOD7jVjsUc1umTsSrl4qqg4MY52u3s7NS2U3dJ0vxZ07Rz2+9KTbmos6dPau7M/+jPowezb1PY2dnlO9e2zVeC+e/3SVG4uGmEfIWGtlRQUJAuXLigXbt2Srpyn3Pbtq16fsQ/Vb5CzsUKBeXg4KBBgwZLkjb+9eID4HZ39Z5iSKPmcnRyyrVP63ZdFNqmoy4mJ2n2jHc1dkR/vfXKcG3/Y6O6PPqEnJyu3Ooo5ZL7fU5Juph8Qfv37JCdnb0a3tuq8A8E2W7rRT+R+/bYuoS7gofHlWfAtm75Q85Ojvruu2WSpCWLF2nZ0iVWfRMTEyVJgwYOkKOjox5//HHVqVPnunOkpCRLkmJjj2lHxLbCLB+SvF1tXcGdwc5O8nZNv26/rKws7dx65Y+/Dm2a5zmmoluGhg3pry4P3Kdt27YpPj5enp6eatq0qTw9PbVy6Ty5u7urVqVSknLfx5pNvyoz87ICAwNVp6pHnv1w827rwAwIDFKJEiVsXcYdz97hyoq7OgGBqt+gkTy9vCRJUVFReY45cOCAJKlceW/Vb5DzrT5/Z2d/5ezVy6u0UX8UzKrw3bYu4Y5gsUinL15/gdv+vTuUkJCg0mXLq3TVYJ2+mPOSqrdreva+SpX3U8tO1gsfN25aL0mq4Vc33zl//vXKs5fBTdsY1Ya82dtZVN4lI8/22zowUfTi4+O1bduVM77AwLqSpGnTPsqzf+tWoYqLi8v1XbL5WbN61V9z8O5L3P6uXo5tfG/r695/zE1WVpZ++WmlJKl56w559os7e1pHD0XLydlZIQ2b5dkPhcNm9zAnTJig1q1b69SpUxowYIC6dOliq1LuehER27V2zRplZmZabT9+PFbPDx+m1JQU3d+2rSpWrHhT83w5Z7ZOnDhhtc1isWjhwq/15ZdzZGdnp169e9/UHICtpaelaXfElQ8yaHSdRTiJCeeVmBCfY/zCOR/r+J9HVK9+U/kHhuQ5/mowB4U0UclSOd8BjcJlszPMV155Ra+88oqtpsc1jh49qpfHjlHZcuVUN7Cu3D3cdfLECe3du1dpaWmqVau2xo+fcNPzfPnll3rnnckKDAxU5cpVlJaepgPR0YqNjZW9vb3GjH1ZdesGFcIRATdv765tWrNisdW2zMzL+uCt/60M79D1MdUNtr6FsHvHH0pLu6RqPrXkXSn/x+ZiYw5p1keTVaV6DZUuU16ZlzN0+OB+pVxMlm+tOuozcHi+47PfIpTLYysofFyShYKDQ9SrV2/t2rVLe/bsVlJSkkqVKqU6dQLUsVMn9erVWyVLlrzpeZ4cMECbNm7UwYMHdPDQIV3OyFD58hX00EPd1DcsTPXqBRfC0QCF4+KFJMUcOWC1zWKxWG27eCEpx7irZ32Nml1/xap3pSpq0LSljh6K0snjf8rBwVHeFSurcfP71OK+Dvm+sSfmyAGdPX1Cbu4e+Z6FovDYWa73WUzFUFpamvbs2cOin2JoR8Q2Fu0UQyz6KX6uXfSD4uHqop+goNyzhecwAQAwQGACAGCAwAQAwACBCQCAAQITAAADBCYAAAYITAAADBCYAAAYIDABADBAYAIAYIDABADAAIEJAIABAhMAAAMEJgAABghMAAAMEJgAABggMAEAMEBgAgBggMAEAMAAgQkAgAECEwAAAwQmAAAGCEwAAAwQmAAAGCAwAQAwQGACAGCAwAQAwACBCQCAAQITAAADBCYAAAYITAAADBCYAAAYIDABADBAYAIAYIDABADAAIEJAIABAhMAAAMEJgAABghMAAAMEJgAABggMAEAMEBgAgBggMAEAMAAgQkAgAECEwAAAwQmAAAGCEwAAAwQmAAAGCAwAQAwQGACAGCAwAQAwACBCQCAAQITAAADBCYAAAYITAAADBCYAAAYIDABADBAYAIAYIDABADAAIEJAIABAhMAAAMEJgAABghMAAAMEJgAABggMAEAMEBgAgBggMAEAMAAgQkAgAECEwAAAwQmAAAGCEwAAAwQmAAAGCAwAQAwQGACAGCAwAQAwACBCQCAAQITAAADBCYAAAYITAAADBCYAAAYIDABADBAYAIAYIDABADAAIEJAIABAhMAAAMEJgAABghMAAAMEJgAABggMAEAMEBgAgBggMAEAMAAgQkAgAECEwAAAwQmAAAGHG1dwI2wWCySpPT0dBtXgtykpaXZugT8jb2dxdYlIBf8XIqXqz+Pqxnzd3aWvFqKsQsXLig6OtrWZQAA7kB+fn5yd3fPsf22DMysrCxdvHhRTk5OsrOzs3U5+MupU6fUp08fzZ8/XxUrVrR1OUCxxe9K8WSxWJSRkSFXV1fZ2+e8Y3lbXpK1t7fPNf1hWw4ODoqLi5ODg4NKlChh63KAYovfleKrZMmSebax6AcAAAMEJnAbiY2Nlb+/v6ZOnZrvtqKaC7ib3ZaXZIFbbfPmzerXr5/VNhcXF/n6+qp79+7q27evHBwcbFTdjYuNjdWyZcvUvn17BQQE2LocoFgjMFFoPDw89Nxzz8nDw8PWpRSZrl27qnXr1rJYLDpz5oyWLVumt956SwcPHtSbb75pk5oqV66sXbt23VBgHz9+XNOmTVPlypVzBObN7Bf5uxt+V+5EBCYKjYeHh4YPH27rMopUYGCgunfvnv11nz591LlzZy1atEgjRoxQuXLlcoxJTk6Wm5tbkdVkZ2dXJAtHimq/uDt+V+5E3MMEboKbm5saNGggi8WiY8eOqW3btgoLC9O+ffs0cOBANWrUSN26dcvuf/ToUY0aNUotW7ZUUFCQ2rZtq0mTJiklJSXHvrdu3apevXopODhYLVq00Pjx43Ptl9+9xlWrViksLEyNGzdWSEiIOnbsqAkTJig9PV1Lly7Nvsw8ZswY+fv7y9/fX2FhYfnu9/Lly/r000/14IMPql69err33ns1bNgwRUVF5VnXzz//rB49eqhevXpq2bKlJk2apMuXL1v1P3DggJ5//nm1atVKQUFBCg0NVVhYmNavX2/2wwCKGGeYwE2wWCyKiYmRJJUuXVqSdOLECfXv31+dOnXSAw88kB1ye/bsUf/+/eXh4aGePXvK29tb+/fv11dffaWIiAh99dVXcnJykiTt3LlTAwYMkKurqwYPHix3d3f98MMP+ve//21c25QpUzRjxgzVqlVLTz75pMqXL68///xTq1ev1vPPP68mTZpo6NChmjFjhnr27KlGjRpJUq5nydcaOXKk/vvf/yo0NFS9e/dWXFyc5s2bp169emnevHkKDAy06v/LL79o/vz56tWrl3r06KGffvpJs2bNkqenp4YOHSpJOn/+vPr37y9J6tWrl+655x6dP39ee/bs0c6dO9WmTRvj4waKjAXAdf3+++8WPz8/y9SpUy3nzp2znDt3zhIZGWl5+eWXLX5+fpZ//OMfFovFYrn//vstfn5+lm+++SbHPh566CFLx44dLRcuXLDavnr1aoufn59lyZIl2dt69uxpqVu3ruXw4cPZ29LS0iw9evSw+Pn5WT788MPs7ceOHcuxbefOnRY/Pz9LWFiY5dKlS1bzZWVlWbKysqyO69q589tveHi4xc/PzzJixIjsfVgsFktkZKQlICDA0rt37xzjQ0JCLMeOHbOav0uXLpbQ0NDsbWvXrrX4+flZVq5cmaMOoLjgkixQAFOnTlXz5s3VvHlzde/eXUuWLFHbtm01ffr07D5eXl569NFHrcZFRUUpKipKXbt2VXp6uuLj47P/06hRI7m4uGjjxo2SpHPnzikiIkJt27aVr69v9j6cnZ315JNPGtX5/fffS5JefPHFHPch7ezsbvgNWWvWrJEkDR061GofderUUZs2bbRt2zbFx8dbjWnXrp2qVKliNf+9996rs2fP6uLFi5KU/SKSDRs2KDk5+YZqA4oal2SBAujZs6c6deokOzs7lSpVSj4+PvLy8rLqU7Vq1RwrSw8dOiTpSuDm9VxjXFycJOnYsWOSpBo1auToU6tWLaM6Y2JiZGdnpzp16hj1NxUbGyt7e3vVrFkzR1vt2rX1008/KTY2VmXKlMneXrVq1Rx9r37PEhIS5OrqqqZNm+rhhx/W0qVLtXz5cgUFBalFixZ68MEHjY8ZKGoEJlAA1atXV4sWLfLtU6pUqTzbnnrqKbVq1SrXtquPGFj+er1zbmeBFsNXP1ssliJ5z7Lp/NfK77GUa/c3adIkDRw4UL/88ou2bdumL774QjNmzNDYsWPVt2/fG6oXKEwEJnALVK9eXdKV9yBfL3CrVasm6X9npdfKbVtufH19tWHDBkVFRSk4ODjPfgUN1WrVqik8PFyHDh3KcfZ6tbZrL78WlJ+fn/z8/DR48GAlJSXp8ccf13vvvacnnniCD1qAzXEPE7gFAgMD5efnp6+//jr7kuu1Ll++rISEBElS2bJlVb9+fa1bt05HjhzJ7pOenq7Zs2cbzffQQw9Jkt5///1cPzf26pmdi4uLJCkxMdFov+3bt5ckffrpp1Znh9HR0Vq3bp0aNWpkdTnWVEJCgrKysqy2eXh4qEqVKkpNTeUzVlEscIYJ3AJ2dnaaPHmy+vfvr27duqlHjx6qVauWLl26pJiYGK1Zs0b/+te/shcLjR49WmFhYerdu7eeeOKJ7MdKMjMzjeYLDg7W4MGD9dlnn+nRRx9V586dVb58ecXGxmrVqlVatGiRPDw8VKtWLbm6umr+/PkqWbKkPDw8VKZMGTVv3jzX/YaGhqpz585auXKlEhMTdf/99+vs2bOaP3++SpQooVdeeeWGvj/ffvut5syZo/bt26t69epydHTUli1bFB4ers6dO+f7CRLArUJgArdIQECAli1bpk8++UTr1q3T119/LVdXV1WuXFmPPPKIVUg1aNBAX3zxhd577z19+umncnNzU6dOndS7d+/ss8frGTlypOrUqaO5c+dq5syZslgsqlixolq3bp0dQCVLltSUKVP0wQcf6K233lJ6erqaNm2aZ2BK0rvvvqvAwEAtW7ZMEydOlIuLi5o0aaIRI0bI39//hr439957ryIjI7V+/XqdPXtW9vb2qlKliv79739z/xLFxm35AdIAANxq3MMEAMAAgQkAgAECEwAAAwQmAAAGCEwAAAwQmAAAGCAwAQAwQGACAGCAwAQAwACBCQCAgf8Hlnwx5zPKG6QAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 540x540 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize the predictions results by plotting the confusion matrix.\n",
    "conf_matrix = confusion_matrix(y_true=ground_truth_label.values, y_pred=predict_label)\n",
    "fig, ax = plt.subplots(figsize=(7.5, 7.5))\n",
    "ax.matshow(conf_matrix, cmap=plt.cm.Blues, alpha=0.3)\n",
    "for i in range(conf_matrix.shape[0]):\n",
    "    for j in range(conf_matrix.shape[1]):\n",
    "        ax.text(x=j, y=i, s=conf_matrix[i, j], va=\"center\", ha=\"center\", size=\"xx-large\")\n",
    "\n",
    "plt.xlabel(\"Predictions\", fontsize=18)\n",
    "plt.ylabel(\"Actuals\", fontsize=18)\n",
    "plt.title(\"Confusion Matrix\", fontsize=18)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mEvaluation result on test data\u001b[0m:\n",
      "\u001b[1maccuracy_score\u001b[0m: 69.30543572044866 % \n",
      "\u001b[1mF1 \u001b[0m: 71.63643611720151 %\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Measure the prediction results quantitatively.\n",
    "eval_accuracy = accuracy_score(ground_truth_label.values, predict_label)\n",
    "eval_f1 = f1_score(ground_truth_label.values, predict_label)\n",
    "\n",
    "print(\n",
    "    f\"{bold}Evaluation result on test data{unbold}:{newline}\"\n",
    "    f\"{bold}{accuracy_score.__name__}{unbold}: {eval_accuracy * 100} % {newline}\"\n",
    "    f\"{bold}F1 {unbold}: {eval_f1 * 100} %{newline}\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[['M'], [18], [5000], [2017], [11], [5], [5], [20], [10], ['discount'], [True], [True], [True], [True]]\n"
     ]
    }
   ],
   "source": [
    "query = {\n",
    "    'gender' : ['M'],\n",
    "    'age' : [18],\n",
    "    'income' : [5000],\n",
    "    'became_member_on_year' : [2017],\n",
    "    'became_member_on_month' : [11],\n",
    "    'became_member_on_day' : [5],\n",
    "    'reward': [5],\n",
    "    'difficulty' : [20],\n",
    "    'duration' : [10],\n",
    "    'offer_type': ['discount'],\n",
    "    'email': [True],\n",
    "    'mobile': [True],\n",
    "    'web' : [True],\n",
    "    'social' : [True]    \n",
    "}\n",
    "\n",
    "query_keys = list(query.keys())\n",
    "query_list = list(query.values())\n",
    "print(query_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mPerson Information\u001b[0m\n",
      "\n",
      "\u001b[1mgender : \u001b[0mM\n",
      "\u001b[1mage : \u001b[0m18\n",
      "\u001b[1mincome : \u001b[0m5000\n",
      "\u001b[1mbecame_member_on_year : \u001b[0m2017\n",
      "\u001b[1mbecame_member_on_month : \u001b[0m11\n",
      "\u001b[1mbecame_member_on_day : \u001b[0m5\n",
      "\n",
      "\u001b[1mOffer Information\u001b[0m\n",
      "\n",
      "\u001b[1mreward : \u001b[0m5\n",
      "\u001b[1mdifficulty : \u001b[0m20\n",
      "\u001b[1mduration : \u001b[0m10\n",
      "\u001b[1moffer_type : \u001b[0mdiscount\n",
      "\u001b[1memail : \u001b[0mTrue\n",
      "\u001b[1mmobile : \u001b[0mTrue\n",
      "\u001b[1mweb : \u001b[0mTrue\n",
      "\u001b[1msocial : \u001b[0mTrue\n",
      "\n",
      "-*--*--*--*--*--*--*--*--*--*--*--*--*--*--*-\n",
      "\u001b[1mIs the Offer expected to be Completed ? : \u001b[0m Yes \u001b[1m with Probability : \u001b[0m 0.6588537096977234\n"
     ]
    }
   ],
   "source": [
    "query_response_batch = query_endpoint(\n",
    "    pd.DataFrame(query).to_csv(header=False, index=False).encode(\"utf-8\")\n",
    ")\n",
    "predict_prob = parse_response(query_response_batch)[0]\n",
    "\n",
    "print(f'{bold}Person Information{unbold}{newline}')\n",
    "for i in range(6):\n",
    "    print(f'{bold}{query_keys[i]} : {unbold}{query_list[i][0]}')\n",
    "\n",
    "print()\n",
    "print(f'{bold}Offer Information{unbold}{newline}')\n",
    "for i in range(6,14):\n",
    "    print(f'{bold}{query_keys[i]} : {unbold}{query_list[i][0]}')\n",
    "\n",
    "print()\n",
    "print('-*-'*15)\n",
    "print(f'{bold}Is the Offer expected to be Completed ? : {unbold} {\"Yes\" if label else \"No\"} {bold} with Probability : {unbold} {predict_prob[label]}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 🚮 Cleaning Environment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Delete the SageMaker endpoint and the attached resources\n",
    "predictor.delete_model()\n",
    "predictor.delete_endpoint()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  },
  "vscode": {
   "interpreter": {
    "hash": "27b35c700d8f8a1d1732cbeb4bc7e442283f2f53043886dc0fbd6eda7510d193"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}