Liver-Patient-Prediction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Importing the required Libraries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing the required Libraries.\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import sys\n",
    "import os\n",
    "import time\n",
    "#ignore warnings\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Common Model Helpers\n",
    "from sklearn.preprocessing import OneHotEncoder, LabelEncoder\n",
    "from sklearn import model_selection\n",
    "from sklearn import metrics\n",
    "\n",
    "#Visualization\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "#Configure Visualization Defaults\n",
    "#%matplotlib inline = show plots in Jupyter Notebook browser\n",
    "%matplotlib inline\n",
    "sns.set_style('white')\n",
    "\n",
    "from sklearn.model_selection import cross_val_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Total_Bilirubin</th>\n",
       "      <th>Direct_Bilirubin</th>\n",
       "      <th>Alkaline_Phosphotase</th>\n",
       "      <th>Alamine_Aminotransferase</th>\n",
       "      <th>Aspartate_Aminotransferase</th>\n",
       "      <th>Total_Protiens</th>\n",
       "      <th>Albumin</th>\n",
       "      <th>Albumin_and_Globulin_Ratio</th>\n",
       "      <th>Dataset</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>579.000000</td>\n",
       "      <td>583.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>44.746141</td>\n",
       "      <td>3.298799</td>\n",
       "      <td>1.486106</td>\n",
       "      <td>290.576329</td>\n",
       "      <td>80.713551</td>\n",
       "      <td>109.910806</td>\n",
       "      <td>6.483190</td>\n",
       "      <td>3.141852</td>\n",
       "      <td>0.947064</td>\n",
       "      <td>1.286449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>16.189833</td>\n",
       "      <td>6.209522</td>\n",
       "      <td>2.808498</td>\n",
       "      <td>242.937989</td>\n",
       "      <td>182.620356</td>\n",
       "      <td>288.918529</td>\n",
       "      <td>1.085451</td>\n",
       "      <td>0.795519</td>\n",
       "      <td>0.319592</td>\n",
       "      <td>0.452490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>63.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>2.700000</td>\n",
       "      <td>0.900000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>33.000000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>175.500000</td>\n",
       "      <td>23.000000</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>5.800000</td>\n",
       "      <td>2.600000</td>\n",
       "      <td>0.700000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>45.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>208.000000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>42.000000</td>\n",
       "      <td>6.600000</td>\n",
       "      <td>3.100000</td>\n",
       "      <td>0.930000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>58.000000</td>\n",
       "      <td>2.600000</td>\n",
       "      <td>1.300000</td>\n",
       "      <td>298.000000</td>\n",
       "      <td>60.500000</td>\n",
       "      <td>87.000000</td>\n",
       "      <td>7.200000</td>\n",
       "      <td>3.800000</td>\n",
       "      <td>1.100000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>90.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>19.700000</td>\n",
       "      <td>2110.000000</td>\n",
       "      <td>2000.000000</td>\n",
       "      <td>4929.000000</td>\n",
       "      <td>9.600000</td>\n",
       "      <td>5.500000</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Age  Total_Bilirubin  Direct_Bilirubin  Alkaline_Phosphotase  \\\n",
       "count  583.000000       583.000000        583.000000            583.000000   \n",
       "mean    44.746141         3.298799          1.486106            290.576329   \n",
       "std     16.189833         6.209522          2.808498            242.937989   \n",
       "min      4.000000         0.400000          0.100000             63.000000   \n",
       "25%     33.000000         0.800000          0.200000            175.500000   \n",
       "50%     45.000000         1.000000          0.300000            208.000000   \n",
       "75%     58.000000         2.600000          1.300000            298.000000   \n",
       "max     90.000000        75.000000         19.700000           2110.000000   \n",
       "\n",
       "       Alamine_Aminotransferase  Aspartate_Aminotransferase  Total_Protiens  \\\n",
       "count                583.000000                  583.000000      583.000000   \n",
       "mean                  80.713551                  109.910806        6.483190   \n",
       "std                  182.620356                  288.918529        1.085451   \n",
       "min                   10.000000                   10.000000        2.700000   \n",
       "25%                   23.000000                   25.000000        5.800000   \n",
       "50%                   35.000000                   42.000000        6.600000   \n",
       "75%                   60.500000                   87.000000        7.200000   \n",
       "max                 2000.000000                 4929.000000        9.600000   \n",
       "\n",
       "          Albumin  Albumin_and_Globulin_Ratio     Dataset  \n",
       "count  583.000000                  579.000000  583.000000  \n",
       "mean     3.141852                    0.947064    1.286449  \n",
       "std      0.795519                    0.319592    0.452490  \n",
       "min      0.900000                    0.300000    1.000000  \n",
       "25%      2.600000                    0.700000    1.000000  \n",
       "50%      3.100000                    0.930000    1.000000  \n",
       "75%      3.800000                    1.100000    2.000000  \n",
       "max      5.500000                    2.800000    2.000000  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('indian_liver_patient.csv')\n",
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['Age', 'Gender', 'Total_Bilirubin', 'Direct_Bilirubin',\n",
      "       'Alkaline_Phosphotase', 'Alamine_Aminotransferase',\n",
      "       'Aspartate_Aminotransferase', 'Total_Protiens', 'Albumin',\n",
      "       'Albumin_and_Globulin_Ratio', 'Dataset'],\n",
      "      dtype='object')\n",
      "**************************************************\n",
      "Age\n",
      "count    583.000000\n",
      "mean      44.746141\n",
      "std       16.189833\n",
      "min        4.000000\n",
      "25%       33.000000\n",
      "50%       45.000000\n",
      "75%       58.000000\n",
      "max       90.000000\n",
      "Name: Age, dtype: float64\n",
      "**************************************************\n",
      "Gender\n",
      "count      583\n",
      "unique       2\n",
      "top       Male\n",
      "freq       441\n",
      "Name: Gender, dtype: object\n",
      "**************************************************\n",
      "Total_Bilirubin\n",
      "count    583.000000\n",
      "mean       3.298799\n",
      "std        6.209522\n",
      "min        0.400000\n",
      "25%        0.800000\n",
      "50%        1.000000\n",
      "75%        2.600000\n",
      "max       75.000000\n",
      "Name: Total_Bilirubin, dtype: float64\n",
      "**************************************************\n",
      "Direct_Bilirubin\n",
      "count    583.000000\n",
      "mean       1.486106\n",
      "std        2.808498\n",
      "min        0.100000\n",
      "25%        0.200000\n",
      "50%        0.300000\n",
      "75%        1.300000\n",
      "max       19.700000\n",
      "Name: Direct_Bilirubin, dtype: float64\n",
      "**************************************************\n",
      "Alkaline_Phosphotase\n",
      "count     583.000000\n",
      "mean      290.576329\n",
      "std       242.937989\n",
      "min        63.000000\n",
      "25%       175.500000\n",
      "50%       208.000000\n",
      "75%       298.000000\n",
      "max      2110.000000\n",
      "Name: Alkaline_Phosphotase, dtype: float64\n",
      "**************************************************\n",
      "Alamine_Aminotransferase\n",
      "count     583.000000\n",
      "mean       80.713551\n",
      "std       182.620356\n",
      "min        10.000000\n",
      "25%        23.000000\n",
      "50%        35.000000\n",
      "75%        60.500000\n",
      "max      2000.000000\n",
      "Name: Alamine_Aminotransferase, dtype: float64\n",
      "**************************************************\n",
      "Aspartate_Aminotransferase\n",
      "count     583.000000\n",
      "mean      109.910806\n",
      "std       288.918529\n",
      "min        10.000000\n",
      "25%        25.000000\n",
      "50%        42.000000\n",
      "75%        87.000000\n",
      "max      4929.000000\n",
      "Name: Aspartate_Aminotransferase, dtype: float64\n",
      "**************************************************\n",
      "Total_Protiens\n",
      "count    583.000000\n",
      "mean       6.483190\n",
      "std        1.085451\n",
      "min        2.700000\n",
      "25%        5.800000\n",
      "50%        6.600000\n",
      "75%        7.200000\n",
      "max        9.600000\n",
      "Name: Total_Protiens, dtype: float64\n",
      "**************************************************\n",
      "Albumin\n",
      "count    583.000000\n",
      "mean       3.141852\n",
      "std        0.795519\n",
      "min        0.900000\n",
      "25%        2.600000\n",
      "50%        3.100000\n",
      "75%        3.800000\n",
      "max        5.500000\n",
      "Name: Albumin, dtype: float64\n",
      "**************************************************\n",
      "Albumin_and_Globulin_Ratio\n",
      "count    579.000000\n",
      "mean       0.947064\n",
      "std        0.319592\n",
      "min        0.300000\n",
      "25%        0.700000\n",
      "50%        0.930000\n",
      "75%        1.100000\n",
      "max        2.800000\n",
      "Name: Albumin_and_Globulin_Ratio, dtype: float64\n",
      "**************************************************\n",
      "Dataset\n",
      "count    583.000000\n",
      "mean       1.286449\n",
      "std        0.452490\n",
      "min        1.000000\n",
      "25%        1.000000\n",
      "50%        1.000000\n",
      "75%        2.000000\n",
      "max        2.000000\n",
      "Name: Dataset, dtype: float64\n",
      "**************************************************\n"
     ]
    }
   ],
   "source": [
    "print(df.columns)\n",
    "print('*'*50)\n",
    "for i in df.columns :\n",
    "    print(i)\n",
    "    print(df[i].describe())\n",
    "    print('*'*50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 583 entries, 0 to 582\n",
      "Data columns (total 11 columns):\n",
      " #   Column                      Non-Null Count  Dtype  \n",
      "---  ------                      --------------  -----  \n",
      " 0   Age                         583 non-null    int64  \n",
      " 1   Gender                      583 non-null    object \n",
      " 2   Total_Bilirubin             583 non-null    float64\n",
      " 3   Direct_Bilirubin            583 non-null    float64\n",
      " 4   Alkaline_Phosphotase        583 non-null    int64  \n",
      " 5   Alamine_Aminotransferase    583 non-null    int64  \n",
      " 6   Aspartate_Aminotransferase  583 non-null    int64  \n",
      " 7   Total_Protiens              583 non-null    float64\n",
      " 8   Albumin                     583 non-null    float64\n",
      " 9   Albumin_and_Globulin_Ratio  579 non-null    float64\n",
      " 10  Dataset                     583 non-null    int64  \n",
      "dtypes: float64(5), int64(5), object(1)\n",
      "memory usage: 50.2+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, column \"Albumin_and_Globulin_ratio\" has some data missing in it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>Gender</th>\n",
       "      <th>Total_Bilirubin</th>\n",
       "      <th>Direct_Bilirubin</th>\n",
       "      <th>Alkaline_Phosphotase</th>\n",
       "      <th>Alamine_Aminotransferase</th>\n",
       "      <th>Aspartate_Aminotransferase</th>\n",
       "      <th>Total_Protiens</th>\n",
       "      <th>Albumin</th>\n",
       "      <th>Albumin_and_Globulin_Ratio</th>\n",
       "      <th>Dataset</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>209</th>\n",
       "      <td>45</td>\n",
       "      <td>Female</td>\n",
       "      <td>0.9</td>\n",
       "      <td>0.3</td>\n",
       "      <td>189</td>\n",
       "      <td>23</td>\n",
       "      <td>33</td>\n",
       "      <td>6.6</td>\n",
       "      <td>3.9</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>241</th>\n",
       "      <td>51</td>\n",
       "      <td>Male</td>\n",
       "      <td>0.8</td>\n",
       "      <td>0.2</td>\n",
       "      <td>230</td>\n",
       "      <td>24</td>\n",
       "      <td>46</td>\n",
       "      <td>6.5</td>\n",
       "      <td>3.1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>253</th>\n",
       "      <td>35</td>\n",
       "      <td>Female</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.2</td>\n",
       "      <td>180</td>\n",
       "      <td>12</td>\n",
       "      <td>15</td>\n",
       "      <td>5.2</td>\n",
       "      <td>2.7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>312</th>\n",
       "      <td>27</td>\n",
       "      <td>Male</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.6</td>\n",
       "      <td>106</td>\n",
       "      <td>25</td>\n",
       "      <td>54</td>\n",
       "      <td>8.5</td>\n",
       "      <td>4.8</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Age  Gender  Total_Bilirubin  Direct_Bilirubin  Alkaline_Phosphotase  \\\n",
       "209   45  Female              0.9               0.3                   189   \n",
       "241   51    Male              0.8               0.2                   230   \n",
       "253   35  Female              0.6               0.2                   180   \n",
       "312   27    Male              1.3               0.6                   106   \n",
       "\n",
       "     Alamine_Aminotransferase  Aspartate_Aminotransferase  Total_Protiens  \\\n",
       "209                        23                          33             6.6   \n",
       "241                        24                          46             6.5   \n",
       "253                        12                          15             5.2   \n",
       "312                        25                          54             8.5   \n",
       "\n",
       "     Albumin  Albumin_and_Globulin_Ratio  Dataset  \n",
       "209      3.9                         NaN        1  \n",
       "241      3.1                         NaN        1  \n",
       "253      2.7                         NaN        2  \n",
       "312      4.8                         NaN        2  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df['Albumin_and_Globulin_Ratio'].isnull()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's do some research on the data set and try to understand what each column is telling us. Afterall, we data scientists love stories."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Column Name\n",
    "\n",
    "**Age** - Tells the person's age.\n",
    "        \n",
    ">         Google - This we all know....duh\n",
    "\n",
    "**Gender** - (Male or Female) Tells the person's gender. This is a very controversial column as we now know that there can be a spectrum of genders. But here we will only consider two genders.\n",
    "            \n",
    ">          ME - My sincere apologies to LGBTQIA+ community....I hope in near future we will have a dataset where the spectrum of genders are included.            \n",
    "         Google - Oh my! you knew that LGBTQ community has now become LGBTQIA+ community...Impressive\n",
    "         ME - Thank you google.\n",
    "\n",
    "**Total_Bilirubin** - Well, I'm a mechancial engineer and not a doctor. So obviously I have no clue what this means. Let's ask google baba.\n",
    "\n",
    ">         Google - A bilirubin test measures the amount of bilirubin in your blood. It’s used to help find the cause of health conditions like jaundice, anemia, and liver disease.\n",
    ">         Bilirubin is an orange-yellow pigment that occurs normally when part of your red blood cells break down. Your liver takes the bilirubin from your blood and changes its chemical make-up so that most of it is passed through your poop as bile.\n",
    ">         If your bilirubin levels are higher than normal, it’s a sign that either your red blood cells are breaking down at an unusual rate or that your liver isn’t breaking down waste properly and clearing the bilirubin from your blood.Another option is that there’s a problem somewhere along the pathway that gets the bilirubin out of your liver and into your stool. \n",
    "        \n",
    ">         ME - Thank you Google. So fellows, I think now you have some knowledge on this as I do. And if you knew it already, you are awesome.\n",
    "        \n",
    "**Direct_Bilirubin** - It's technically the same as \"Total_Bilirubin\". The difference will be given us by our own google.\n",
    "        \n",
    ">         Google - Bilirubin attached by the liver to glucuronic acid, a glucose-derived acid, is called direct, or conjugated, bilirubin. Bilirubin not attached to glucuronic acid is called indirect, or unconjugated, bilirubin. All the bilirubin in your blood together is called total bilirubin. \n",
    ">         \n",
    ">         ME - Damn you google, how much information do you have.....\n",
    "        \n",
    "**Alkaline_Phosphotase** - .......\n",
    "\n",
    ">         Google - Alkaline phosphatase (ALP) is an enzyme in a person's blood that helps break down proteins. The body uses ALP for a wide range of processes, and it plays a particularly important role in liver function and bone development.Using an ALP test, it is possible to measure how much of this enzyme is circulating in a person’s blood.\n",
    ">         \n",
    ">         ME - I knew this....\n",
    ">         Google - No, you don't\n",
    ">         Me - Yeah.....you know everything....\n",
    "\n",
    "**Alamine_Aminotransferase** - First of all it is \"Alanine\" and not \"Alamine\" . Rest our friend google will tell.\n",
    "\n",
    ">         Google - Alanine aminotransferase (ALT) is an enzyme found primarily in the liver and kidney. It was originally referred to as serum glutamic pyruvic transaminase (SGPT). Normally, a low level of ALT exists in the serum. ALT is increased with liver damage and is used to screen for and/or monitor liver disease. Alanine aminotransferase (ALT) is usually measured concurrently with AST as part of a liver function panel to determine the source of organ damage. \n",
    ">         \n",
    ">         ME - So, we need to change the column name to aviod confusion.\n",
    "        \n",
    "**Aspartate_Aminotransferase** - Help Google.....\n",
    "\n",
    ">         Google - AST (aspartate aminotransferase) is an enzyme that is found mostly in the liver, but also in muscles. When your liver is damaged, it releases AST into your bloodstream. An AST blood test measures the amount of AST in your blood. The test can help your health care provider diagnose liver damage or disease.\n",
    ">         \n",
    ">         ME - WOOAAAHHH.......\n",
    "        \n",
    "**Total_protein** - Albumin and globulin are two types of protein in your body. The total protein test measures the total amount albumin and globulin in your body. It's used as part of your routine health checkup. It may also be used if you have unexpected weight loss, fatigue, or the symptoms of a kidney or liver disease.\n",
    "        \n",
    ">         ME - Atlast, something I knew.        \n",
    ">         Google - You googled it. Don't play smart with me.\n",
    ">         ME - uughhhh........There's no pleasing you.\n",
    "        \n",
    "**Albumin** - I think it's related to the protein in our bodies....\n",
    "            \n",
    ">         Google - Albumin is a protein made by your liver. Albumin helps keep fluid in your bloodstream so it doesn't leak into other tissues. It is also carries various substances throughout your body, including hormones, vitamins, and enzymes. Low albumin levels can indicate a problem with your liver or kidneys.\n",
    ">         \n",
    ">         ME - Close enough!!...\n",
    ">         Google - *Face palms*\n",
    "\n",
    "**Albumin_and_Globulin_Ratio** - This one's pretty eas......\n",
    "\n",
    ">         Google - The Albumin to Globulin ratio (A:G) is the ratio of albumin present in serum in relation to the amount of globulin. The ratio can be interpreted only in light of the total protein concentration. Very generally speaking, the normal ratio in most species approximates 1:1.\n",
    ">         \n",
    ">         ME - I give up....\n",
    ">         Google - Who made you a data scientist\n",
    ">         ME - heyyy!! That's mean..\n",
    ">         Google - The Arithmetic Mean is the average of the numbe....\n",
    ">         ME - I know THAT...\n",
    "\n",
    "**Dataset** - This is labelled incorrectly. From my perspective it should be \"Liver_Disease\" indicating that the patient has liver disease or not "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, now we have a better understanding of the dataset let's first make the changes that are required to be made"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## A Little bit of cleaning is required"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Re-naming the columns\n",
    "df =  df.rename(columns={'Dataset':'Liver_disease','Alamine_Aminotransferase':'Alanine_Aminotransferase'}, inplace=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>Total_Bilirubin</th>\n",
       "      <th>Direct_Bilirubin</th>\n",
       "      <th>Alkaline_Phosphotase</th>\n",
       "      <th>Alanine_Aminotransferase</th>\n",
       "      <th>Aspartate_Aminotransferase</th>\n",
       "      <th>Total_Protiens</th>\n",
       "      <th>Albumin</th>\n",
       "      <th>Albumin_and_Globulin_Ratio</th>\n",
       "      <th>Liver_disease</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>579.000000</td>\n",
       "      <td>583.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>44.746141</td>\n",
       "      <td>3.298799</td>\n",
       "      <td>1.486106</td>\n",
       "      <td>290.576329</td>\n",
       "      <td>80.713551</td>\n",
       "      <td>109.910806</td>\n",
       "      <td>6.483190</td>\n",
       "      <td>3.141852</td>\n",
       "      <td>0.947064</td>\n",
       "      <td>1.286449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>16.189833</td>\n",
       "      <td>6.209522</td>\n",
       "      <td>2.808498</td>\n",
       "      <td>242.937989</td>\n",
       "      <td>182.620356</td>\n",
       "      <td>288.918529</td>\n",
       "      <td>1.085451</td>\n",
       "      <td>0.795519</td>\n",
       "      <td>0.319592</td>\n",
       "      <td>0.452490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>63.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>2.700000</td>\n",
       "      <td>0.900000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>33.000000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>175.500000</td>\n",
       "      <td>23.000000</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>5.800000</td>\n",
       "      <td>2.600000</td>\n",
       "      <td>0.700000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>45.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>208.000000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>42.000000</td>\n",
       "      <td>6.600000</td>\n",
       "      <td>3.100000</td>\n",
       "      <td>0.930000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>58.000000</td>\n",
       "      <td>2.600000</td>\n",
       "      <td>1.300000</td>\n",
       "      <td>298.000000</td>\n",
       "      <td>60.500000</td>\n",
       "      <td>87.000000</td>\n",
       "      <td>7.200000</td>\n",
       "      <td>3.800000</td>\n",
       "      <td>1.100000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>90.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>19.700000</td>\n",
       "      <td>2110.000000</td>\n",
       "      <td>2000.000000</td>\n",
       "      <td>4929.000000</td>\n",
       "      <td>9.600000</td>\n",
       "      <td>5.500000</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Age  Total_Bilirubin  Direct_Bilirubin  Alkaline_Phosphotase  \\\n",
       "count  583.000000       583.000000        583.000000            583.000000   \n",
       "mean    44.746141         3.298799          1.486106            290.576329   \n",
       "std     16.189833         6.209522          2.808498            242.937989   \n",
       "min      4.000000         0.400000          0.100000             63.000000   \n",
       "25%     33.000000         0.800000          0.200000            175.500000   \n",
       "50%     45.000000         1.000000          0.300000            208.000000   \n",
       "75%     58.000000         2.600000          1.300000            298.000000   \n",
       "max     90.000000        75.000000         19.700000           2110.000000   \n",
       "\n",
       "       Alanine_Aminotransferase  Aspartate_Aminotransferase  Total_Protiens  \\\n",
       "count                583.000000                  583.000000      583.000000   \n",
       "mean                  80.713551                  109.910806        6.483190   \n",
       "std                  182.620356                  288.918529        1.085451   \n",
       "min                   10.000000                   10.000000        2.700000   \n",
       "25%                   23.000000                   25.000000        5.800000   \n",
       "50%                   35.000000                   42.000000        6.600000   \n",
       "75%                   60.500000                   87.000000        7.200000   \n",
       "max                 2000.000000                 4929.000000        9.600000   \n",
       "\n",
       "          Albumin  Albumin_and_Globulin_Ratio  Liver_disease  \n",
       "count  583.000000                  579.000000     583.000000  \n",
       "mean     3.141852                    0.947064       1.286449  \n",
       "std      0.795519                    0.319592       0.452490  \n",
       "min      0.900000                    0.300000       1.000000  \n",
       "25%      2.600000                    0.700000       1.000000  \n",
       "50%      3.100000                    0.930000       1.000000  \n",
       "75%      3.800000                    1.100000       2.000000  \n",
       "max      5.500000                    2.800000       2.000000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Renaming Done\n",
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dropping Null Values\n",
    "df = df.dropna()\n",
    "# Changing the values in \"Liver_Disease\" column \n",
    "df['Liver_disease'] = df['Liver_disease'] - 1 \n",
    "# Converting Gender column into categorical data \n",
    "LabelEncoder = LabelEncoder()\n",
    "df['Is_male'] = LabelEncoder.fit_transform(df['Gender'])\n",
    "df = df.drop(columns='Gender')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df[['Age', 'Total_Bilirubin', \n",
    "        'Direct_Bilirubin',\n",
    "        'Alkaline_Phosphotase',\n",
    "        'Alanine_Aminotransferase', 'Aspartate_Aminotransferase',\n",
    "       'Total_Protiens', 'Albumin', 'Albumin_and_Globulin_Ratio', 'Is_male']]\n",
    "y = df['Liver_disease']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total Unhealthy Livers :  414 and its percentage is 71.5 %\n",
      "Total Healthy Livers :  165 and its percentage is 28.5 %\n"
     ]
    }
   ],
   "source": [
    "# Validate each class to understand if the dataset is imbalanced.\n",
    "\n",
    "print ('Total Unhealthy Livers :  {} and its percentage is {} %'.format(df.Liver_disease.value_counts()[0], round(df.Liver_disease.value_counts()[0]/df.Liver_disease.value_counts().sum()*100,2)) )\n",
    "print ('Total Healthy Livers :  {} and its percentage is {} %'.format(df.Liver_disease.value_counts()[1], round(df.Liver_disease.value_counts()[1]/df.Liver_disease.value_counts().sum()*100,2)) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Age                           -0.033591\n",
       "Total_Bilirubin                4.890768\n",
       "Direct_Bilirubin               3.199163\n",
       "Alkaline_Phosphotase           3.753502\n",
       "Alanine_Aminotransferase       6.527575\n",
       "Aspartate_Aminotransferase    10.512251\n",
       "Total_Protiens                -0.292433\n",
       "Albumin                       -0.048516\n",
       "Albumin_and_Globulin_Ratio     0.992299\n",
       "Liver_disease                  0.955179\n",
       "Is_male                       -1.209212\n",
       "dtype: float64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.skew(axis = 0, skipna = True) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Here in column name \"Liver_disease\" **0** *indicate that the the person has some kind of Liver Disease or the liver of the patient is unhealthy* and **1** *represents that the person's liver is healthy.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Alanine Aminotransferase (units/L)')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA68AAALFCAYAAAAsgNVhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nOzdf3zP9f7/8dt7v2usERWmcI5pjJyIOG/T1EwiP4thyhZRFvk1zK9O1jJMWWlo4mBvv1rU4VQOi7Psi1OSH+voIxEV5ee2bHtve3//cNn7NNvY1uz1Nvfr5eLS+/36eX+/L3l6Pd6v5+v5NNlsNhsiIiIiIiIiDszJ6AAiIiIiIiIi16PiVURERERERByeilcRERERERFxeCpeRURERERExOGpeBURERERERGHp+JVREREREREHJ6KV6kUVqsVs9nMc889V6nHTU5Opk2bNvTq1Ysnn3yS7t278/zzz3PmzJlKPQ/A7t276dGjx3W3e+utt/jXv/5V6ecXkfKZPXs2vXr1olevXvj7+xMcHGx/n52dXep+Fy9e5Nlnn73u8devX88LL7xQbPmuXbto1aqV/Vw9evQgNDSU7777DoCtW7fy2muvARASEsK//vUvfvrpJwYNGlTuz1i4/9Uqcrz169cXa09HjhzJr7/+Wu5c17Nr1y569ep13e0WLlxISkpKpZ9fpDo4efIkfn5+9ramZ8+eDBw4kC1btti3efPNN9m4cWOlnTMsLIxz584VW3696zGLxcKSJUsA6NKlCwcOHODAgQO89NJL5c5QuP/VKnK8+Ph4Hn74YXvuxx9/nPHjx5OZmVnuXNeTnJzM888/f93tpk2bxsGDByv9/LcKF6MDSPWwdetW7r//fg4ePMjRo0f505/+VGnHbtu2LYsXL7a/nzVrFgsXLmT27NmVdo7y2L17N3/+858NObeI/M+0adPsr7t06cK8efNo2bLldfe7cOHCH75waNy4MZs2bbK/T0hI4LXXXuPdd98lKCiIoKCgItvXq1ePpKSkP3TOyjhe+/btWbRokf399OnTiY+P55VXXqm0bOWRlpZG8+bNDTm3yM3Aw8OjSFtz6tQpnn32WZydnQkODmbMmDGVer7PP/+81HXXuh4LCQkptn3Lli1ZuHBhpWWr6PG6d+/OjBkzAMjPz+fFF19k5cqVjBo1qtKylceuXbsYMGCAIeeuDnTnVSqFxWLh0UcfpXv37qxYscK+fMmSJXTt2pU+ffoQHR1Nly5dAMjNzeW1116jT58+PPnkk0yePLlMv4JZrVYyMzOpW7eu/f2rr75K9+7d6dmzJ1FRUWRmZvLrr79iNpvZsWMHAG+88QbDhg2joKCALl26MH/+fPr27UtQUFCJF4AZGRlMmDCBHj160LNnT2JjY8nLy2P16tUcPHiQ2NhYtm7dWhlfnYjcIHv27OGpp57iySefpF+/fqSmpgIwZcoUsrKy7HcG161bx1NPPUXv3r0JDAxk7dq15TqPzWbj4sWL9nappDu2x48fp23btgAsWLCA8PBwevbsSWRkJAsWLCA6Otq+7dXv//nPf9K3b1+6d+9uv7Nx9fGmTJlCWFgYwcHBDBs2rEx3U61WK1lZWfbcubm5vPLKKzzxxBP07NmT6dOnk5WVxZkzZ+jYsSP//ve/AZg/fz7h4eHYbDYCAgKIi4ujb9++dO3atcTv7tKlS4wfP54ePXrQo0cP5s+fT15eHn//+9/55ptveO2119i2bVuZv2+RW1mDBg146aWXSExMBGDy5Mn21/7+/owZM4bg4GAOHDjA0aNHCQsLo2/fvvTq1YsNGzbYj7Nhwwb73/WhQ4fy008/MWXKFACeeeYZfvrpp2vmuPp6LD4+nr/97W9Ftvl9j7bJkyczcuRInnjiCebOnVsk99WfAyApKYk+ffrwxBNP2HNffbzZs2cTGhpKUFAQo0ePJisr67rfX05ODr/99ps9d2nXe0ePHuWhhx4iPT0dgEmTJjF16lQAmjdvzoIFC+jbty/dunXj008/LXaen3/+mZEjR9KzZ0969OjBu+++C1xpr8+cOcOECRPYv3//dfNKcbrzKn/Y//3f/7Fv3z4WLlxIixYtCA0N5eWXX+bgwYMkJyezYcMGatasSVRUlH2fJUuW4OzsTHJyMiaTibi4OObNm8esWbOKHf8///kPvXr1wmazcfr0adzd3Xn55ZcBeOeddzhz5gybNm3C2dmZqKgoYmNj+dvf/sbrr7/O1KlTmT59Ohs3biQ5ORknpyu/11y8eJH333+f06dP07t3b9q0aVPknLNnz8bb25uPPvoIq9XKqFGjWLZsGSNGjODjjz9m8ODBxe6siIjjOHfuHGPHjmXx4sW0bNmS//73vwwdOpQPPviAmJgY+vXrx6ZNm8jMzOT9999n6dKleHt785///IdRo0Zd91fxY8eO2YvfX3/9lZycHFatWlXmfD///DMffvghzs7OLFiw4JrbZmdns27dOi5fvkz//v1p0aIFPj4+Rbb54osvSE5OpkaNGgwfPpy1a9fy4osvFjvW7t27i7Snt912GxMnTgTg7bff5vz582zatAmTycTUqVOZP38+M2bMICYmhunTpzN16lQ2b97Mhg0bMJlMwJXiNDk5mR9//JE+ffoUa09feeUV6taty/z588nJyeH5559nxYoVhIeH889//pPw8HAeffTRMn93Ire6+++/nyNHjhRbbrVaCQwM5M033yQvL49evXoRGxtLixYtyMjIYMCAAfz5z3/Gw8ODefPm8cEHH1CvXj2WL1/OO++8Q0xMDMnJyaxYsYLatWsXO/61rsfKIjs7m82bNwNXis9rcXd354MPPuD06dP06dOHBx54oNg2Bw8e5O9//zsmk4mnn36ajz/+mH79+hXbbsuWLXzxxRcUFBTw448/Ur9+fbp27Qpc+3pv4sSJREZGEhoayjfffMO6deuAK3dvb7vtNpKTk/nmm28YMmSI/cfEQhMmTODRRx9l2LBhZGRkMHjwYOrVq8fLL7/MRx99VOaeQlKc7rzKH2axWAgMDKRWrVq0atUKHx8f1q1bx44dO+jWrRteXl6YTCYGDx5s3+ezzz5j+/bt9O7dm169evGvf/2Lo0ePlnj8tm3bsmnTJj788EPS0tIYMGAAzz33HDabjZ07dzJw4EBcXV1xcnIiNDTUfnfAbDbTvXt3IiIimDdvXpGGeNCgQZhMJu655x46depUrJvMzp07GTJkCCaTCTc3NwYOHMjOnTtvwLcnIjfCvn37aNKkif3ioFmzZjzwwAPs2bOnyHY1atTgnXfeISUlhQULFrBkyRJ+++236x6/sNvwpk2b+Pzzz5k8eTJhYWFl2hegdevWODs7l2nbp556ChcXF2rWrEnXrl1L7Nb38MMPU6NGDeDKXYGLFy+WeKz27dsXaU/79evH8OHDi7SnLi4uODs7M3jwYHu717lzZx577DFeeumlYu3pkCFDAKhfvz5//etfi+VLTU21t//u7u4MGDBA7anIH2AymfDw8ChxXWER9f3333PixAmmTp1Kr169GDJkCNnZ2Rw+fJi0tDTMZjP16tUD4Nlnny1217S0Y5d2PVYWV/+wdS0DBw4E4O677+avf/0raWlpxbbp1KkTbm5uuLq64uvrW2q71717dzZt2sRHH33E//t//4/WrVvbi+5rXe89/fTTNGrUiNmzZ7Nw4cIi33lhu3f//ffj6+vL3r177et+++03vvzyS3u7V7NmTfr27at2r5KoeJU/5LfffmPTpk188cUXdOnShS5duvDLL7+watUqXFxcijRov79QKygoYOrUqfaLv/Xr1/Pmm29e93yFBep3333H2bNnKSgosP/6X3hcq9UKXOnKd/ToUerUqcNXX31V5DguLi5F9im8I/v7ZVcfNy8vr4zfiogY7eq/w4XLrv57fOrUKfr06cPPP/9M27ZtGTNmTJkvxH6vf//+XL582T5o0/Xcfvvt9tcmk6nIOQvbsEK/b58KCgpwdXUtdjx3d/ci78vyGZycnBg8eDDffvstFy5cID8/v8h3ZrPZ7N+XzWbju+++o06dOnz99ddFjnN12351UX6t44pI+R04cABfX98S1xW2Lfn5+dSsWdN+nbVp0ybWrVtHv379cHZ2LvJ3Mjs7u9QbCKW5+nqsLP5Iu/f767ZCvy8mrz5eaVxdXRk0aJC92LzW9V5ubi7Hjx+nZs2a9u7Dha7V7hUUFBTLouvIyqPiVf6Qjz76CG9vb/7973+zfft2tm/fzr/+9S9+++03WrRowaeffkpGRgZAkWctzGYzq1evJjc3l4KCAqZPn05cXFyZzvnZZ5/RoEEDateuTadOnbBYLFitVgoKCli9ejV//etfAVi+fDm//fYb77//PsuXLy9ywVU4Mt+PP/7I559/TkBAQJFzmM1mVq1ahc1mIzc3l3Xr1tGxY0fgSoOlBkjEsf3lL3/hyJEj9hEr//vf//Lll1/Svn17nJ2dyc/Px2azceDAAerWrcvIkSMxm82kpKRQUFBQ7vPt2bMHFxcXGjVqVO59a9euzaFDh7DZbGRmZhb7dX7Tpk3YbDYuXLjAJ598QqdOncp9jtLs2LGD++67D29vb3t7mpeXZ29PC9u9d999F6vVyvr161m6dCmHDh2yH6OwPT158iRpaWnF8hW2p3DlebPft6cuLi7FLlpFpHTHjh1j0aJFhIWFXXO7xo0bFxns6aeffqJHjx4cPHiQ9u3bk5aWZh8peM2aNcydOxco3zXO76/HyqtWrVr2gfNOnz5drFfMBx98AFy5TktLS6NDhw7lPkdpPvvsM1q1agVc+3ovNjaWpk2bkpiYyOzZszl16pT9GIXt3qFDhzh27BgPPfSQfV2NGjV44IEHWL16NXDludqNGzfqOrKS6JlX+UMsFgvDhg0r8ouTl5cXoaGhLF++nKeffpoBAwbg4eFB06ZNue222wB44YUXmDNnDn369CE/Px8/P79Sn38ofMbCZDKRl5eHt7c3b7/9Nk5OTowaNYo5c+bQu3dv8vLyaNWqFdOnT+fw4cMkJCSwYcMG7r77bqZOncr48ePtjeHJkyfp27cv2dnZTJs2jSZNmvDLL7/Yzzlt2jRmz55Nz549sVqtdOrUiZEjRwJXRjWNi4vDarXSp0+fG/XVisgfUKdOHd544w1mzZpFbm4uTk5OxMbG0rBhQ6xWK35+fvTo0YM1a9aQnJxMt27dMJlMtG/fnjvuuIMTJ05c8/i/f+a1oKAAd3d33nrrLXvX3fLo1asXqampdO3alXvuuafIRRBcuVtR2F4NGzaMtm3bcvz48XKfB/73zGthe1qrVi0WLlyIyWRi9OjRvP766/Tq1Yu8vDxat27NlClTOHDgAImJiSQnJ1OvXj0iIyMZN24cycnJAJw4cYI+ffqQm5vLzJkzue+++4pc5M2YMYNXX32VHj16YLVaCQgIYPjw4cCV9nTu3Lnk5uaWaWodkVtNdna2/e+Gk5MT7u7ujBs3jkceeeSa+7m5ubFo0SKio6N59913ycvLY8yYMfauuxMnTrRPb1i3bl379F7dunUjNDSU+Pj4Ynd3r3U9Vl6hoaFMmDCB4OBgfHx8ePjhh4usz8nJoU+fPlitVqZNm0bjxo0rPE1i4TOvJpOJnJwcGjZsyJw5c4DSr/c+++wztm7dykcffYSXlxfPPPMM48ePt/8Q9+WXX7Ju3ToKCgpYsGABd9xxR5Fzzps3j7/97W8kJyeTm5tLz5496du3LwBBQUFMnDiRWbNmYTabK/SZbmUmW0X6R4mUwYEDB9i3bx9Dhw4F4L333mP//v288cYbhubq0qULb775ph6UFxH5gwICAli8eDF+fn5GRxERqRLNmjUjLS2tQnec5Y/TnVe5YRo3bszSpUtZt24dJpOJevXq8eqrrxodS0REREREbkK68yoiIiIiIiIOTwM2iYiIiIiIiMNT8SoiIiIiIiIO76Z75rV9+/Y0aNDA6Bgi4kBOnTrF7t27jY5RqdTWiUhJqlt7p7ZOREpSWlt30xWvDRo0sA/PLyIC2Iefr07U1olISapbe6e2TkRKUlpbp27DIiIiIiIi4vBUvIqIiIiIiIjDU/EqIiIiIiIiDk/Fq4iIiIiIiDg8Fa8iIiIiIiLi8G660YZFRERE5MayWq1MnTqVU6dOkZuby6hRo/jzn//M5MmTMZlMNG3alJkzZ+Lk9L/7INnZ2UycOJGzZ8/i6enJnDlzqF27toGfQkSqG915FREREZEiPvzwQ7y9vUlKSmLp0qW8+uqrxMTEMHbsWJKSkrDZbGzbtq3IPhaLBV9fX5KSkujduzeLFi0yKL2IVFcqXkVERESkiG7dujFmzBj7e2dnZw4dOkS7du0ACAgIYNeuXUX2+eKLL+jUqZN9fVpaWtUFFpFbgopXERERESnC09OTGjVqkJmZyUsvvcTYsWOx2WyYTCb7+oyMjCL7ZGZmUrNmzVLXi4j8USpeRUREHJDFYsHf3x9nZ2f8/f2xWCxGR5JbzE8//cTQoUPp1asXPXv2LPJ8a1ZWFl5eXkW2r1GjBllZWaWuFymJ2jopDxWvIiIiDsZisRAVFUV8fDzZ2dnEx8cTFRWlizqpMr/++ithYWFMnDiR/v37A9C8eXN2794NwM6dO2nbtm2RfR588EF27NhhX9+mTZuqDS03HbV1Ul4qXkVERBxMdHQ0iYmJBAYG4urqSmBgIImJiURHRxsdTW4RCQkJXLp0iUWLFhEaGkpoaChjx44lPj6eAQMGYLVaCQ4OBiAsLIzc3FxCQkL49ttvCQkJYe3atYwePdrgTyGOTm2dlJemyhGH4+/vz6FDhyrteC1atODgwYOVdjwRkRstPT0ds9lcZJnZbCY9Pd2gRHKrmTZtGtOmTSu2fNWqVcWWLVu2zP564cKFNzSXVC9q66S8dOdVHM7Bgwex2WzX/QOUaTsVriJys/Hz8yM1NbXIstTUVPz8/AxKJCJS+dTWSXmpeBUREXEwUVFRhIeHk5KSgtVqJSUlhfDwcKKiooyOJiJSadTWSXmp27CIiIiDCQkJASAiIoL09HT8/PyIjo62LxcRqQ7U1kl5qXgVERFxQCEhIbqAE5FqT22dlIe6DYuIiIiIiIjDU/EqIiIiIiIiDk/Fq4iIiIiIiDg8Fa8iIiIiIiLi8FS8ioiIiIiIiMNT8SoiIiIiIiIOT8WriIiIiIiIODwVryIiIiIiIuLwVLyKiIiIiIiIw1PxKiIiIiIiIg7PxegAIiLVSXJyMh988AEAOTk5pKens3LlSqKjo3F2dsZsNjN69GiDU4qIiIjcfHTnVUSkEvXt25eVK1eycuVKWrRowbRp05g5cybz58/HYrGwf/9+Dh06ZHRMERERh2CxWPD398fZ2Rl/f38sFovRkcSBqXgVEbkBDhw4wP/93//xxBNPkJuby7333ovJZMJsNpOWlmZ0PBEREcNZLBaioqKIj48nOzub+Ph4oqKiVMBKqVS8iojcAIsXL+bFF18kMzOTGjVq2Jd7enqSkZFhYDIRERHHEB0dTWJiIoGBgbi6uhIYGEhiYiLR0dFGRxMHpeJVRKSSXbp0ie+++46HH36YGjVqkJWVZV+XlZWFl5eXgelEREQcQ3p6Omazucgys9lMenq6QYnE0al4FRGpZHv37qVjx44A1KhRA1dXV06cOIHNZiM1NZW2bdsanFBERMR4fn5+pKamFlmWmpqKn5+fQYnE0al4FRGpZMeOHcPHx8f+/pVXXmHChAn079+f5s2b88ADDxiYTkRExDFERUURHh5OSkoKVquVlJQUwsPDiYqKMjqaOChNlSMiUsmee+65Iu9bt27NunXrDEojIiLimEJCQgCIiIggPT0dPz8/oqOj7ctFrqbiVUREREREDBESEqJiVcpM3YZFRERERETE4al4FREREREREYen4lVEREREREQcnp55FREREZES7d+/n3nz5rFy5Upefvllfv31VwBOnTrFAw88wIIFC+zb2mw2AgICaNSoEXBlsLrx48cbEVtEqikVryIiIiJSzNKlS/nwww+57bbbAOyF6sWLFxk6dChTpkwpsv2JEydo0aIFCQkJVZ5VRG4NVdZt2Gq1Mn78eAYOHMigQYM4evQox48fJyQkhEGDBjFz5kwKCgqqKo6IiIiIXMO9995LfHx8seXx8fEMGTKEu+66q8jyQ4cOcfr0aUJDQxk+fDjfffddVUUVkVtElRWvO3bsIC8vjzVr1vDiiy/yxhtvEBMTw9ixY0lKSsJms7Ft27aqiiMiIiIi1xAcHIyLS9FOemfPniUtLY2+ffsW275u3bqMGDGClStX8vzzzzNx4sSqiioit4gqK14bN25Mfn4+BQUFZGZm4uLiwqFDh2jXrh0AAQEB7Nq1q6riiIiIiEg5ffzxx/To0QNnZ+di6/z9/Xn00UcBaNu2LadPn8Zms1V1RBGpxqqseL399ts5deoUjz/+ONOnTyc0NBSbzYbJZALA09OTjIyMqoojIiIiIuWUlpZGQEBAieveeustVqxYAcA333xD/fr17dd5IiKVocqK1+XLl2M2m/nkk0/YtGkTkydPxmq12tdnZWXh5eVVVXFEREREpJyOHTtGw4YNiywLCwsjNzeXESNGsHfvXoYMGUJMTAwxMTEGpZSbicViwd/fH2dnZ/z9/bFYLEZHEgdWZaMNe3l54erqCsAdd9xBXl4ezZs3Z/fu3bRv356dO3fy8MMPV1UcEREREbkOHx8f1q1bZ3+/efPmYtssW7YMADc3N5YsWVJl2eTmZ7FYiIqKIjExEbPZTGpqKuHh4QCEhIQYnE4cUZXdeX322Wc5dOgQgwYN4plnnuHll19mxowZxMfHM2DAAKxWK8HBwVUVR0REREREDBQdHU1iYiKBgYG4uroSGBhIYmIi0dHRRkcTB1Vld149PT158803iy1ftWpVVUUQEREREREHkZ6ejtlsLrLMbDaTnp5uUCJxdFV251VERERERKSQn58fqampRZalpqbi5+dnUCJxdCpeRURERESkykVFRREeHk5KSgpWq5WUlBTCw8OJiooyOpo4qCrrNiwiIiIiIlKocFCmiIgI0tPT8fPzIzo6WoM1SalUvIqIiIiIiCFCQkJUrEqZqduwiIiIiIiIODwVryIiIiIiIuLwVLyKiIiIiIiIw1PxKiIiIiIiIg5PxauIiIiIiIg4PBWvIiIiIiIi4vBUvIqIiIiIiIjDU/EqIiIiIiIiDk/Fq4iIiIiIiDg8Fa8iIiIiIiLi8FyMDiAiUt0sXryY7du3Y7VaCQkJoV27dkyePBmTyUTTpk2ZOXMmTk767VBERESkPHT1JCJSiXbv3s2+ffuwWCysXLmSn3/+mZiYGMaOHUtSUhI2m41t27YZHVNERETkpqPiVUSkEqWmpuLr68uLL77IyJEjeeSRRzh06BDt2rUDICAggF27dhmcUkREROTmo27DIiKV6Pz58/z4448kJCRw8uRJRo0ahc1mw2QyAeDp6UlGRobBKUVERERuPipeRUQqkbe3N02aNMHNzY0mTZrg7u7Ozz//bF+flZWFl5eXgQlFREREbk7qNiwiUonatGnDv//9b2w2G6dPn+by5ct06NCB3bt3A7Bz507atm1rcEoRERGRm4/uvIqIVKLAwED27t1L//79sdlszJgxAx8fH6ZPn05cXBxNmjQhODjY6JgiIiIiNx0VryIilWzSpEnFlq1atcqAJCIiIiLVh7oNi4iIiIiIiMNT8SoiIiIiIiIOT8WriIiIiIiIODwVryIiIiJSov379xMaGgrAoUOH6NSpE6GhoYSGhrJly5Yi22ZnZxMREcGgQYMYPnw4586dMyKyiFRjGrBJRERERIpZunQpH374IbfddhsAhw8fZtiwYYSFhZW4vcViwdfXl4iICDZv3syiRYuYNm1aVUYWkWpOd15FREREpJh7772X+Ph4+/uDBw/y2WefMXjwYKZOnUpmZmaR7b/44gs6deoEQEBAAGlpaVWaV0SqPxWvIiIiIlJMcHAwLi7/66TXqlUrJk2axOrVq2nYsCFvv/12ke0zMzOpWbMmAJ6enmRkZFRpXhGp/lS8ioiIiMh1BQUF4e/vb399+PDhIutr1KhBVlYWAFlZWXh5eVV5RhGp3lS8ioiIiMh1hYeH8/XXXwOQlpZGixYtiqx/8MEH2bFjBwA7d+6kTZs2VZ5RRKo3Fa8iIiIicl2zZs3itddeIzQ0lC+//JIXXngBgLCwMHJzcwkJCeHbb78lJCSEtWvXMnr0aIMTi0h1o9GGRURERKREPj4+rFu3DoAWLVqwZs2aYtssW7bM/nrhwoVVlk1Ebj268yoiIiIiIiIOT8WriIiIiIiIODwVryIiIiIiIuLwVLyKiIiIiIghLBYL/v7+ODs74+/vj8ViMTqSODAVryIiIiIiUuUsFgtjxowpMj/wmDFjVMBKqVS8ioiIiIhIlZs0aRIuLi4sW7aM7Oxsli1bhouLC5MmTTI6mjgoFa8iIiIiIlLlTp48yYoVKwgMDMTV1ZXAwEBWrFjByZMnjY4mDkrFq4iIiIiIiDg8Fa8iIiIiIlLlfHx8GDp0KCkpKVitVlJSUhg6dCg+Pj5GRxMHpeJVRERERESqXGxsLFlZWQQHB+Pm5kZwcDBZWVnExsYaHU0clIpXERERERExhIeHBw0aNMBkMtGgQQM8PDyMjiQOTMWriIiIiIhUuejoaNauXcuxY8coKCjg2LFjrF27lujoaKOjiYNS8SoiIiIiIlUuPT0ds9lcZJnZbCY9Pd2gROLoVLyKiIiIiEiV8/PzIzU1tciy1NRU/Pz8DEokjk7Fq4iIiIiIVLmoqCjCw8OLjDYcHh5OVFSU0dHEQbkYHUBERERERG49ISEhAERERJCeno6fnx/R0dH25SJXU/EqIiIiUo3ZbDY+++wz9uzZw4ULF6hduzYdOnTgr3/9KyaTyeh4cosLCQlRsSplpm7DIiIiItVUWloazzzzDHv27KFZs2b06NGDFi1akJqayrPPPsuuXbuMjigiUma68yoiUop9+/axYsUK/vOf/2AymXBzc6NNmzYMHjyYBx54oNT9evfuTc2aNQHw8efaL8QAACAASURBVPFhwIABREdH4+zsjNlsZvTo0VX1EUTkFnf8+HHee+89nJ2diyzv3r07+fn5rF27lo4dOxqUTkSkfFS8ioiUYPbs2bi7u/P8888zb948XFxcsNlspKen8+GHH7Jp0yZmzJhRbL+cnBwAVq5caV/Wq1cv4uPjadiwISNGjODQoUO0aNGiyj6LiNy6Bg4cWOLy//znP7Rt25ZBgwZVcSIRkYpT8SoiUoIRI0Zw1113FVlmMplo3rw5zZs358yZMyXu980333D58mXCwsLIy8sjIiKC3Nxc7r33XuDK/HVpaWkqXkXEUK+//jobNmwwOoaISLlUafG6ePFitm/fjtVqJSQkhHbt2jF58mRMJhNNmzZl5syZODnpMVwRMd7VhWuh+fPnM378+FLXe3h4EB4ezlNPPcX333/P8OHD8fLysq/39PTkhx9+uCGZRUTKymazGR1BRKTcqqxS3L17N/v27cNisbBy5Up+/vlnYmJiGDt2LElJSdhsNrZt21ZVcUREKiQtLe2a6xs3bsyTTz6JyWSicePG1KxZkwsXLtjXZ2VlFSlmRUSMoFGGxVFYLBb8/f1xdnbG398fi8VidCRxYFV25zU1NRVfX19efPFFMjMzmTRpEuvWraNdu3YABAQE8PnnnxMUFFRVkUREKt2GDRs4cuQIs2bN4vTp01y+fJnbb7+dEydO0LBhQ1JTUzVgk4hUmQEDBhQrVG02G999951BiUT+x2KxEBUVRWJiImazmdTUVMLDwwE0fY6UqMqK1/Pnz/Pjjz+SkJDAyZMnGTVqFDabzd6genp6kpGRUVVxRESuqaSuvTabjdzc3Gvu179/f6ZMmUJISAgmk4nXXnsNJycnJkyYQH5+Pmaz+ZojFYuIVKa4uDijI4iUKjo6msTERAIDAwEIDAwkMTGRiIgIFa9SoiorXr29vWnSpAlubm40adIEd3d3fv75Z/t6daUTEUcSGRlZ4vIaNWpccz83Nzfmz59fbPm6desqJZeISHns3bu31HUNGjSowiQixaWnp7N+/Xoef/xxcnJycHd3JywsjPT0dKOjiYOqsuK1TZs2/P3vf2fYsGGcOXOGy5cv06FDB3bv3k379u3ZuXMnDz/8cFXFERG5pqSkJKMjiIj8YUePHgXgq6++4rbbbuMvf/kLBw4cIC8vj969e193//379zNv3jxWrlxJeno6r776Ks7Ozri5uTFnzhzq1KlTZPur57mOiYmp/A8l1Ya3tzeLFy9m7ty5jBw5koSEBCZOnIi3t7fR0cRBVVnxGhgYyN69e+nfvz82m40ZM2bg4+PD9OnTiYuLo0mTJgQHB1dVHBGRaxo2bFipA5osW7asitOIiFTM+PHjAQgPD2fJkiX25WFhYdfdd+nSpXz44YfcdtttwJUuntOnT8fPz481a9awdOlSpkyZYt++pHmuRa7l0qVLeHt785e//AVXV1f+8pe/4O3tzaVLl4yOJg6qSqfKmTRpUrFlq1atqsoIIiJlMnXqVAAWLVpEly5dePDBB/n666/ZuXOnwclERMrv3LlzXLp0CS8vL86fP19kFPTS3HvvvcTHx9uv3+Li4uzThOXn5+Pu7l5k+6vnuR43bhytW7eu/A8j1UZeXh7z588nIiKC9PR0/Pz8mD9/PsOGDTM6mjioKi1eRURuFk2bNgWuDDbXs2dP4MrzYWvWrDEylohIhYwcOZJ+/fpRs2ZNMjIyytSdNzg4mJMnT9rfFxauX375JatWrWL16tVFti9pnuuPP/4YFxddbkrJ3N3dOXfuHAcPHrQvi4uLK/bDiEghtSZSZWrXrs358+cr9ZiVOU9drVq1OHfuXKUdT6qHgoICPvjgA1q1asWXX36Jm5ub0ZFERMrsgw8+oHPnzgQHBxMcHMzZs2fx9vbG2dm5QsfbsmUL77zzDkuWLKF27dpF1jVu3Jj77rvPPs+1t7c3v/zyC/Xq1auMjyLV0PDhw+0DJBY+8xoZGcnIkSMNTiaOSsWrVJnz589js9mMjlEqTdguJZk7dy6JiYl89NFHNGnShAULFhgdSUSkzDw9PZk/fz4XLlzA39+fwMBA7rzzzgoda9OmTaxdu5aVK1eWOKDO1fNcZ2ZmUrdu3T/6EaQai4+P58iRI0yYMIHx48djMpkICgoiPj7e6GjioFS8ioiUICIigrZt2xIYGGh//lVE5GbTtWtXunbtCsDXX3/NJ598whtvvME999zDrFmzynyc/Px8oqOjqVevHhEREQA89NBDvPTSS0yaNImxY8eWOM+1ugzLtVgsFr799lu2bduG2WwmNTWV8PBwLBaL5nmVEqlFEREpwYIFC9i7dy+rV6/mxIkTNGrUiMDAQNq0aVPh7nYiIkb58ccfqVOnDk899RQALi4uWK1WXF1dr7mfj4+PfZ7qPXv2lLhNbGys/XVJ81yLlCY6OprExEQCAwOBK7OTJCYmEhERoeJVSqTiVUSkBC4uLnTo0IEOHToA8P3335OSksLKlSvVnUlEbjrPP/88p0+fpkmTJhw7dozbbruNvLw8Jk6cSK9evYyOJ7eo9PR0zGZzkWVms5n09HSDEomjczI6gIiII0tISCAhIYGPP/6YnJwcWrVqxebNm8nLyzM6mohImfn4+PDxxx+zZs0aPv30U1q2bMk//vEPTVkohvLz8yM1NbXIstTUVPz8/AxKJI5Od15FRK7h4MGDuLq60rZtW7766ivOnDnDnXfeyc6dO5kzZ47R8UREyuTs2bP20YHvuOMOfv31V7y9vXFy0n0MMU5UVBS9evUiOzvb3o3dw8ODxYsXGx1NHJSKVxGRa8jIyGDFihUADB48mLCwMOLi4hg4cKDByUREyq5FixaMGzeO1q1bs2/fPvz8/NiyZUuFRx4WqQy7du0iMzPT/iNKQUEBmZmZ7Nq1S8+8Son0c5uIyDVkZGRw4cIFAC5evMilS5fIy8sjJyfH4GQiImU3c+ZMnnjiCbKzs+nduzczZszg/vvv1wBLYqiEhARq1arF1q1byc3NZevWrdSqVYuEhASjo4mD0p1XEZFreOGFF+jXrx/e3t5cunSJqKgoEhMT6d27t9HRRETK7IcffuDUqVOYTCaOHDnCkSNHGD58uNGx5BaXl5fHqlWriow2vGrVKrp3725wMnFUFS5ejxw5wp49e7hw4QK1a9emQ4cONG7cuDKziYgY7rHHHuORRx7h7Nmz1KlTB2dnZx555BGjY4mIlMsLL7xA165d8fLyMjqKSBEHDx7k8ccfL/JepDTlLl6PHj1KbGws7u7u+Pr6ctddd3Hx4kXi4uLIy8tj3LhxNG3a9EZkFRGpcuvXr2ft2rVFugl/9NFHBiYSESm/evXqERERYXQMkSJq165NZGQk8+fP58yZM9x1112cOXPGPriYyNXKXbxu3ryZefPmUbNmzWLrLl68yPLlyxkzZkylhBMRMdp7773HO++8o7sVInJTCwwMZN68efz5z3+2L9PjD2K0QYMG8dZbb3H69GkA+38HDRpkZCxxYOUesOmll14qVrgWFBQAV4ZeV+EqItWJr68vDRs2pFatWvY/IlXBYrHg7++Ps7Mz/v7+WCwWoyPJTWzLli1kZGRw9OhRjh49ynfffWd0JBE2btzIHXfcQaNGjXBycqJRo0bccccdbNy40eho4qAq/MzrP//5TwoKCsjNzSU2NpbnnnuO8PDwyswmImK4jh07EhQUxH333YfNZsNkMrFs2TKjY0k1Z7FY7IODmc1mUlNT7f/GavoIqQg3NzdeeeUVo2OIFHHy5Ek+/fRTgoKC7Mu2bt1K165dDUwljqzCxeuyZctYsmQJ48aNY8eOHYSFhal4FZFqZ82aNcTGxqrbsFSp6OhoEhMTi4zAmZiYSEREhIpXqZD69euzePFimjdvjslkAsBsNhucSkSkfCpcvLq5uQHg6emJm5sbWVlZlRZKRMRR3H333Tz44IP2iz2RqpCenl6ssDCbzaSnpxuUSG52eXl5fP/993z//ff2ZSpexWg+Pj4888wzrF692t7L5JlnnsHHx8foaOKgKly8NmzYkH79+jF9+nTeeustWrVqVZm5REQcQl5eHr1798bX19dewMbGxhqcSqo7Pz8/UlNT7XdeAVJTU/Hz8zMwldyMTp8+zd13301MTMw114sYITY2ljFjxhAWFsaJEye49957ycvLY/78+UZHEwdV4eK1d+/eTJ8+HU9PT/z9/albt25l5hIRcQhhYWFGR5BbUFRUFOHh4cWeeY2OjjY6mtxk3n33XVxcXOjZsydNmzbF1dUVm83GoUOH2LRpEwUFBUyfPt3omHKLKnwMorBt8/T05LXXXtPjEVKqChev8fHxrF69GkCFq4hUO6tXr2bgwIF06NCh2Lr8/HzWrFnD4MGDDUgmt4LCC7eIiAjS09Px8/MjOjpaF3RSblFRUezfv5/ExET27t1LQUEBHh4ePPjggwwaNIjWrVsbHVFEpMwqXLyaTCZefPFFGjdujJPTlRl3xo0bV2nBRESM1LRpU8LDw7n//vtp1qwZd955JxkZGXz11Vd88803vPDCC0ZHlGouJCRExapUigceeEDdMMUhaWR1Ka8KF6/9+vWrzBwiIg6lXbt2PPTQQ+zcuZM9e/awd+9e7rzzTgICApgyZYr9RzsRERGpGI2sLuVV4eK1Z8+eHDhwgLy8PGw2G2fOnKnMXCIihjOZTHTu3JnOnTsbHUVERKTa0cjqUl4VLl5Hjx6N1WrlzJkz5Ofnc9ddd9GjR4/KzCYiIiIiItWURlaX8qpwv7fMzEwSExNp1aoVycnJ5OTkVGYuEZGb2tmzZ+ncuTNHjx7l+PHjhISEMGjQIGbOnElBQYHR8UTkFrNx40Yef/xxHn30Ubp06cKjjz5qdCQR+8jqKSkpWK1WUlJSCA8PJyoqyuho4qAqfOfVxeXKrpcvX8bDwwOr1VppoUREHMmhQ4eK/ED34IMPXnN7q9XKjBkz8PDwACAmJoaxY8fSvn17ZsyYwbZt2wgKCrqhmUVEfm/p0qUkJCRQr149o6OI2IWEhLBr1y4ef/xxcnJycHd3Z/jw4XreVUpV4eI1KCiIt956i/vvv5+nn36aGjVqVGYuERGH8NJLL3H69Gnuvvtu4MpzsNcrXufMmcPAgQNZsmQJcKX4bdeuHQABAQF8/vnnKl5FpEo1bNiQ++67z+gYIkVYLBY2b97MP//5zyKjDXfs2FEFrJSowsXr7+c37Ny5sxpEEamWTp8+zdq1a8u8fXJyMrVr16ZTp0724tVms2EymYArE7BnZGTckKwiIqXx8PDgueeew8/Pz94eaYpDMZpGG5byqnDx+u233zJz5kwyMjLo2bMnTZs2LfKwtYhIddCkSRN+/fVX6tSpU6bt33//fUwmE2lpaaSnpxMZGcm5c+fs67OysvDy8rpRcUVESqRR08URpaenc/LkSfz9/UlPT8fPz4/IyEiNNiylqnDxOnv2bGJiYpg2bRr9+/fnueeeU/EqItXOvn37eOSRR7jzzjuBK92GP/vss1K3X716tf11aGgos2bNYu7cuezevZv27duzc+dOHn744RsdW0QEgAMHDtCyZUvq1q1rdBSRYurXr8+kSZNISkqydxseNGgQ9evXNzqaOKgKF68A9913HyaTidq1a+Pp6VlZmUREHMbmzZv/8DEiIyOZPn06cXFxNGnShODg4EpIJiJyfWlpabRs2bLEtuzq+TVFjJCdnU1YWBgnTpzg3nvvJTs7W2PpSKkqXLzecccdrFmzhsuXL7N582Z1gxORamXx4sU8//zzTJ482f58WKHY2NgyHWPlypX216tWrarUfCIiZTFixAjgyqjn33zzDd9//z1NmzblT3/6U5n2379/P/PmzWPlypUcP37c3iY2bdqUmTNn4uT0v1kXs7OzmThxImfPnsXT05M5c+ZQu3btG/K5pHo4deqU/bEcm80GgKurK6dOnTIyljiwCs/z+tprr3Hy5Elq1arFwYMHiY6OrsxcIiKGKrwj0bdvX/r06VPkj4jIzeadd97hlVde4csvvyQqKorly5dfd5+lS5cybdo0+1RhhdN+JSUlYbPZ2LZtW5HtLRYLvr6+JCUl0bt3bxYtWnQjPopUI25ubkyePJljx45RUFDAsWPHmDx5Mm5ubkZHEwdV4eJ1wYIFPPHEEyxZsoTIyEi8vb0rM5eIiKFatGgBgJ+fH6mpqaxatYrdu3fTsmVLg5OJiJTfjh07WL16NVOnTmXVqlVs2bLluvvce++9xMfH299fPe3Xrl27imz/xRdf0KlTJ/v6tLS0SvwEUh3l5uYSHx9PSkoKVquVlJQU4uPjyc3NNTqaOKgKF6+dO3cmISGBgQMHkpSURGZmZmXmEhFxCJGRkdSrV49Ro0bh7e3N5MmTjY4kIlJutWvX5vLlywBYrdYydecNDg7GxeV/T5hdb9qvzMxMatasWep6kas1b96cwYMHExERgYeHBxEREQwePJjmzZsbHU0cVIWfeQ0ICCAgIIBz584RHR1NbGws3bp1IyIiggYNGlRmRhERw1y+fJkhQ4YA4O/vX6ybnIiIIxswYAAmk4mzZ88SHBxMs2bNOHr0aIV6zP3++daSpv2qUaMGWVlZpa4XuVpUVBRRUVEkJibaRxsODw/X44hSqgoXr0ePHiU5OZmUlBTat29PUlISeXl5REREkJycXJkZRUSq3A8//ADAnXfeydatW2nTpg0HDhygYcOGBicTESm7uLi4SjtW8+bNrznt14MPPsiOHTto1aoVO3fupE2bNpV2bqmeQkJCAIiIiLDP8xodHW1fLnK1ChevUVFRDBgwwH6bv1C/fv0qJZiIiJEiIyMxmUzYbDbee+893nvvPYBiIw+LiDiywt5wb731VrF1o0ePLtexSpv2KywsjISEBEJCQoiMjCQkJARXV1fmz5//xz+AVHvLly/n8OHD2Gw2Dh8+zPLly1W8SqnKXbwePnyY5s2bs2bNmhLXt27d+g+HEhExWlJSktERREQqze+nIzl8+DAFBQVl2s/Hx4d169YB0Lhx4xKn/Vq2bJn99cKFCyshrdwqgoOD+fTTTxk1ahQxMTFMmTKFd955h+DgYD755BOj44kDKnfx+u9//5vExES6detGs2bNuPPOO8nIyGD//v1s2bKF+++/3z5Kp4jIza5r165F7rbefvvtfPDBBwYmEhEpv4EDBxZ5/9xzzxmUROR/tm7dyqhRo+zTKhX+NyEhwchY4sDKXbw+//zz/Pzzz1gsFhITEzl//jx16tShXbt2REZGUr9+/RuRU0TEEBs3bgSu3K04dOiQBmwSkZvSsWPH7K9/+eUXfvrpJwPTiFxhs9mIiYkpsiwmJoZ33nnHoETi6Cr0zOs999zDyy+/XNlZpJo7MMoTZt1hdIxSHRjlaXQEcUC33367/XW7du3UJU5EbkozZsywv/bw8GDSpEkGphG5wmQyMWXKFPsdV4ApU6ZofAkpVYUHbBIpr0d/8uSu9nfZ36/pceW56YH/+F9XplEPjOKF1i/QZV0Xfrn8CwB+tf1Y13Mds3bN4v1v37dvu+2pbRw+e5iI7RH2ZTM6zOAp36douaKlfVlnn8689ehbjN42mh0nd9iXH3jmAOuPrOdvaX8D4Mz+M5yu5M8sN7833njD/o/oL7/8YnAaEZGK6dy5s7oKi8MJCgqy32X9/TOvXbt2NTiZOCqTzWazGR2iPPr27aupeG5ShSO3OipHzyelu5Htwrp16+zFq7u7O507d+aOO258DwK1dSJSkoq2DUOHDuW9997D2dn5BqSqOLV1EhwczNatW7HZbJhMJoKCgjRYk5TaNvyhO6/ff/89x48fp1mzZtx99926xS8i1c4nn3xCYmKi0TFERP6Q8+fP06lTJ3x8fDCZTJhMplJnjhCpSr6+vuzYsYOcnBzc3Nzw9fU1OpI4sAoXr6tWrWLr1q1cvHiR3r17c+LEiSLPU4iIVAdeXl7s2LGDRo0a4eTkBEDDhg0NTiUiUj4avVUcUUREBIsWLeKuu+7izJkz1KpVy/78a3x8vMHpxBFVuHjdvHkzSUlJDB06lGeffZZ+/fpVZi4REYdw+vRpFi9ebH9vMplYvXq1gYlERMpu/fr1PPXUU6xZs6ZYD7lx48YZlErkioSEBLy9vUlKSsJsNpOamkr//v1JSEhQ8SolqnDxWvhsYGFD6ObmVjmJREQcSFJSktERREQq7J577gGgSZMmBicRKS4vL4+OHTvy+OOPk5OTg7u7O0FBQfzjH/8wOpo4qAoXr0888QSDBw/mxx9/ZPjw4Tz22GOVmUtExFBjx47ljTfeoHPnzvYf6QoHk/jss8+MDSciUkYmk4nU1FTq1q1rdBSREm3ZsoW5c+cycuRIEhISmDhxotGRxIFVuHgNDQ2lY8eOHDlyhCZNmtCsWbPKzCUiYqi4uDgAtm/fbnASEZGK27x5c6nrzGZzFSYRKdnVMz1o5ge5lgoXr1OmTLG/3rlzJ66urtxzzz0MHjy4SqaREBG5kSIjI0sdQT02NraK08ityGKxEB0dTXp6On5+fkRFRRESEmJ0LLnJvPrqq7i4uJCbm2t0FJESeXp6MnnyZMaPH4+rqyuenp5kZmYaHUscVIWL15ycHBo2bEjbtm3Zv38/Bw4coHbt2kRGRmpEOxG56fXt29foCHILs1gsREVFkZiYaB/EJDw8HEAFrJRLZGQk8+fPp1u3bsUegdi2bZvB6eRW5+7uTv369fn222+BK8/ANm7cmOPHjxucTBxVhYvXc+fO2bvVderUibCwMMaOHcvgwYMrLZyIiFFyc3Pp3LkzGzZsMDqK3IKio6NJTEwkMDAQgMDAQBITE4mIiFDxKuUyf/58QI9AiGPy9fXlwIEDODk52X9UOXLkCC1btjQ6mjioChevmZmZHD16lD/96U8cPXqU3377jfPnz/Pbb79VZj4REUP88ssvAJw8edLgJHIrSk9PL/Y8otlsJj093aBEcrP69ddfSUhIoEGDBgQEBDBq1Cjy8vJ45ZVX6NSpk9Hx5BZ36NAh4H+zlxT+t3C5yNUqXLzOmDGDiRMncubMGTw8POjTpw9btmxh5MiR19zv7Nmz9O3bl2XLluHi4sLkyZMxmUw0bdqUmTNn4uTkVNFIIiKVpn///sCVUYcBvv76a3Jzc2nbtq2RseQW4efnR2pqqv3OK0Bqaip+fn4GppKb0aRJk+jWrRsXL15k8ODBLFiwgHr16hEZGaniVQxXUFBATEwMkydPti97/fXXi4ytI/J7Fa4UW7VqxaxZs+jYsSOXL1/m7NmzDB48mODg4FL3sVqtzJgxAw8PDwBiYmIYO3YsSUlJ2Gw2PXshIg5j+/btmM1m+vfvT2JiIpMnT2bevHnMmTPH6GhyC4iKiiI8PJyUlBSsVispKSmEh4cTFRVldDS5yeTk5PD0008zfPhwGjRoQIcOHWjUqBG333670dFEANi4cSMeHh6YTCY8PDzYuHGj0ZHEgZX7zmtubi6bN29m9erVuLm5kZmZybZt2+wF6bXMmTOHgQMHsmTJEuBKl4B27doBEBAQwOeff05QUFB5I4mIVLq3336bLVu2kJGRQZ8+fdi+fTuenp563lCqROH/ZxEREfbRhqOjo/X/n5Sbs7Oz/bWnp6f9dX5+vhFxRIowmUzs3r2bJ598ksTERMLDw/nwww9LHe1fpNzFa5cuXejRowfz5s2jUaNGPPfcc2UqXJOTk6lduzadOnWyF6+FD2bDlQY1IyOjvHFERG6I22+/HS8vL7y8vPD19aVGjRoAuLm5GZxMRKTsfvjhB+Li4rDZbEVe63l+cQQmkwmbzcbmzZupW7eu/ccWFa9SmnIXr0OHDuUf//gHp06don///mWeSPj999/HZDKRlpZGeno6kZGRnDt3zr4+KysLLy+v8sYREbkhfv8Pp57Fl6pmsVgYOXIkly9fpqCggCNHjtjHlNDdVymPl156qcTXERERRsQRKaKgoIARI0awYsUK8vPzcXFxITw83H6jS+Rq5S5eR4wYwYgRI9izZw/r16/n4MGDzJ07l169euHr61vqfqtXr7a/Dg0NZdasWcydO5fdu3fTvn17du7cycMPP1yxTyEiUsn27dvHI488gs1m49y5c/bX58+fNzqa3AJGjx5NZmYmc+fOZeTIkSQkJDBx4kRGjx6t4lXKpU+fPkZHECmVu7s7zZo1Izs7274sLi4Od3d3A1OJI6vwaMPt2rWjXbt2XLp0iU2bNjFp0qRyP2AdGRnJ9OnTiYuLo0mTJtcc7ElEpCp99dVXFdovPz+fadOmcezYMZydnYmJicFms2lkdSmXc+fOERsby7hx4wAYN24c+fn5TJo0yeBkIiKVZ/jw4URGRgLYf6iLjIy87uwlcuuqcPFayMvLi9DQUEJDQ8u8z8qVK+2vV61a9UcjiIhUut8PclIeKSkpAKxZs4bdu3fbi9exY8fSvn17ZsyYwbZt2zQ4nVyXv7//Nd+LiNzs4uPjOXLkCBMmTGD8+PGYTCaCgoKIj483Opo4KP30LyJSiR577DFeffVVAH788Ufq1KlTbGT1Xbt2GRlRbgIuLi4MGTKkyFQ5Q4YMwcXlD//mLLeY/Px8cnNzGT16NFarldzcXHJychg6dKjR0USwWCzs27eP++67DycnJ+677z727duHxWIxOpo4KBWvIiKVzMXFhcjISF599VWCg4M1srqU28iRI7lw4QKDBg3Cw8ODQYMGceHCBXWlk3J7//336datGzt37qRbt25069aNJ554gvr16xsdTYRJkyaRlZXFqVOnKCgo4NSpU2RlZekRCSmVfsIVESnBoEGDig3VX1iE/n4AutLMmTOHCRMm8PTTT5OTk2NfrpHVpSwKu8wtXbqUgoICzp8/zwsvvKCudFJuTz/9NE8//TQbNmz4/+zdeVhUdf8+8Htmp+QY6AAAIABJREFUYEAWRXzUUsTMJRA0NM1QXMDMFQFXINHUSk0xcAdRSYXcexQflwwrF1AKU/NpUVFTccmv5oKClhmKIpqg7Mww8/n94Y95JBY35Bzgfl3XXDhnzpy5Z3DenPdZPgdDhgwBAGg0Gl72i2QhJSUFKpUKS5YsMZzzOmPGDF7KicrE5pWIqBSLFi16puft3LkTaWlpGDduHGrVqgWFQgFHR0eOrE5PLSIigs0qVRidTofw8HAEBwdj/PjxGDhwIDw9PZ96OTt27MB3330HACgoKEBiYiLi4+MNG+UWLlyIM2fOwNzcHACwZs0aWFpaVtwboWpn7NixxQanu3z5Mi+VQ2Vi80pEVApbW1sAwI0bN/Dzzz+jsLAQQgjcuXMH8+bNK/N577zzDoKCgvDuu++isLAQwcHBaN68OUdWJyJJbdu2Ddu2bQMArF+/HiNGjHim5nXQoEEYNGgQAOCTTz7B4MGDix1NcvHiRXzxxRewtraumOBU7e3cuRPe3t5wcXHB0aNHn/rqJVSzsHklIirH9OnT0bVrV/zf//0f6tWrV+wQ4NKYmZlh5cqVJaZzZHUikpJSqTRcO9PY2LjEaRFP68KFC/jjjz+KbczT6/VITk7G3Llz8ffff2PIkCGGQ5WJSmNkZIScnByMGTMGycnJaNq0KXJycjg4HZWJAzYREZXD1NQUEydOxEsvvYRly5bh3r17UkeiGiI6OhqOjo5QqVRwdHTk6Jv0XHr27AlfX18sWrQIfn5+cHNze67lrV+/HhMnTiw2LTc3FyNGjMDSpUvxxRdfICoqCklJSc/1OlS9jR8/Hnl5ecjLywMAw785OB2VhZs1iIjKIYRAeno6cnNzkZ+fjwcPHkgdiWqA6OhozJ49G5GRkYZD6caOHQsA8PHxkTgdVUUfffQRXF1dce3aNXh6esLOzu6Zl5WZmYk///yzxPn7tWrVwsiRI1GrVi0AwFtvvYWkpKTnei2q3oqu87pv3z7DqTm8ziuVh3teiYjKMWHCBPzwww8YMGAAevTogc6dO0sdiWqAsLAw+Pr6wt/fH6ampvD394evry/CwsKkjkZVVFpaGiIjIxEbG4vz58/j3Llzz7ysU6dOlVoL//rrL/j6+kKn00Gr1eLMmTNwcHB4nthUzUVHR+P3339HXFwcNBoN4uLi8Pvvv/NIEyoTm1cionIUFBRgxIgR6NWrF06cOIEOHTpIHYlqgEuXLmHr1q2IiIhAfn4+IiIisHXrVly6dEnqaFRFzZkzB4MHD4ZGo0GHDh2ea0PItWvXYGNjY7j/5ZdfIi4uDs2bN4e7uzuGDRsGPz8/eHh4oGXLlhURn6qpsLAwREZGwtXVFcbGxnB1dUVkZCQ31FGZeNgwEVEpDh06hLNnz2L37t3w8PAA8HAwkr1793K0YHrh1Go1/P394erqCgBwdXWFv78/goODJU5GVVVBQQGcnZ2xdu1avPrqq4bBm57F+++/X+z+6NGjDf/+4IMP8MEHHzzzsqlmSUxMxDfffIO+ffuioKAAJiYmGDNmDBITE6WORjLFPa9ERKVo2bIlGjduDLVajUaNGqFRo0awtbXF0qVLpY5GNYBGo8GiRYvQrFkzKJVKNGvWDIsWLYJGo5E6GlVRarUaR44cgV6vx9mzZ6FWq6WORAQrKyusX78e4eHhyMnJQXh4ONavXw8rKyupo5FMsXklIipF48aNMXToUHz//fdo3749rKys0KZNGzg6OkodjWqAxo0bQ6vVAoDhkiZarRaNGzeWMhZVYQsWLMCOHTuQkZGBjRs3IjQ0VOpIRMjMzISVlRXatWsHY2NjtGvXDlZWVsjMzJQ6GskUDxsmIipHTEwMduzYgddffx1r167FwIED8d5770kdi2oAU1NTbNy40TDasK+vr9SRqAr78ssv8dlnn0kdg6iYwsJCLF++HP7+/khMTIS9vT2WL19e7FB0okexeSUiKsfu3buxbds2GBsbQ6PRwMfHh80rvXC3bt3CV199VWyFbsmSJfy/R8/s6tWryMzMRO3ataWOQmRgYmKC9PR0JCQkGKatWLHiuc7JpuqNzSsRUTmEEDA2Ngbw8Jyxon8TvUj29vawsbEptkJ38OBB2NvbS5iKqrKrV6+iU6dOsLa2NhyKfvToUYlTUU33wQcfYObMmQCA8ePHY926dZg5cybGjx8vcTKSKzavRETlcHJyQmBgIDp06IDTp0/j9ddflzoS1QCzZ8/G8OHDYW5ujuvXr8PW1hY5OTlYuXKl1NGoijp48KDUEYhKiIiIAAAEBwdj6tSpMDExwfjx4w3Tif6JzStVqqKtvXJUt25dqSOQDAUHB2P//v34888/0a9fP7z99ttSR6IaRgghdQSqwtasWYOPPvoIU6ZMKfE3ePny5RKlIvqfiIgINqv0xNi8UqWp6BUwhULBlTp6YQICAvDvf/8bANiwUqULCwvD9u3bDdd5BR7uOfP394ePj4+EyaiqcXNzAwB4e3tLnISodNHR0QgLCzOc3z979mzWOSoTm1ciolKkp6dLHYFqsMTERLi4uBSb5uLigsTERIkSUVVlZ2eHpKQkHD9+HBkZGXjppZfQp08fvPLKK1JHI0J0dDRmz56NyMhIw8jqY8eOBQA2sFQqNq9ERKW4ceMGVqxYUepjU6ZMqeQ0VNPY29vj6NGjxfa8Hj16lAM20VP78ccfsWHDBnh7e8PR0RG3bt3C5MmTMXnyZB5VQpILCwtDZGSkoda5uroiMjKSR5lQmdi8EhGVwtTUFM2aNZM6BtVQs2fPxtixY0vsjQgLC5M6GlUxmzZtwpYtW2BmZmaY5uXlhQkTJrB5JcklJiYiJSUFjo6OhsOGZ86cyaNMqExsXomISvGvf/0LXl5eUsegGsrHxwdfffUVevbsCSEEFAoFevXqxT0R9NSMjIyKNa4AYGFhAZVKJVEiov9p1KgR/P39DYNm5uTkwN/fH40aNZI4GcmVUuoARERy5OjoKHUEqsH8/f2xf/9+NGjQAADQoEED7N+/H/7+/hIno6qmrFH+9Xp9JSchKik3NxeZmZnIz88HAOTn5yMzMxO5ubkSJyO54p5XIqJSFF00nUgK69atg0KhQFpaGgAgLS0NKpUK69at4yUl6Kn88ccfmDp1arFpQghcvXpVokRE/5Oeno46derA1NQUQgiYmpqidu3aHDSRysTmlYiISGYKCwsBPLz+9IMHD1CnTh1kZGRInIqqoqJLfv0TL51DcjF79mxMnz7dcH/p0qWYMWOGhIlIzti8EhERyZCJiQliY2MNAzb17dsXBQUFUseiKubNN9+UOgJRuZYvX44OHToYat3y5culjkQyxnNeiYiIZEir1eK3334r9pOIqDqxsbFBfn4+xowZAxMTE4wZMwb5+fmwsbGROhrJFJtXIiIiGVIqlZg1axbMzc0xa9YsKJX8k01E1cuSJUugVquLTVOr1ViyZIlEiUju+JeQiIhIZszNzVFYWAgLCwsADy9tUlhYCHNzc4mTERFVHB8fHwwfPhypqakQQiA1NRXDhw/nZcGoTGxeiYiIZGbDhg2oVauWYZCmjIwM1KpVCxs2bJA4GRFRxYmOjsb27dvx8ssvQ6FQ4OWXX8b27dsRHR0tdTSSKTavREREMuPj44PIyEg4ODhAqVTCwcEBkZGR3BtBRNXKjBkzoFKpsHHjRhQUFGDjxo1QqVQcbZjKxNGGiYiIZMjHx4fNKhFVaykpKdi7dy9cXV0BAK6urti0aRPeeecdiZORXHHPKxERkQxFR0fD0dERKpUKjo6OPIyOiIhqPO55JSKqQFqtFsHBwbh58yY0Gg0mTJiAFi1aYNasWVAoFGjZsiXmzZvHkWOpXNHR0Rg/fjzy8vKg1+tx5coVjB8/HgC4N5aIqg0bGxsMGzYMVlZWuH79OmxtbXH//n1eKofKxLUnIqIKtHv3blhZWSEqKgobNmzAggUL8OmnnyIgIABRUVEQQiAuLk7qmCRzkyZNQlZWFurVqwelUol69eohKysLkyZNkjoaEVGF8fT0RGZmJvLy8iCEQF5eHjIzM+Hp6Sl1NJIpNq9ERBWoT58++Pjjjw33VSoVLl68iDfffBMA0K1bNxw7dkyqeFRFpKenGzaC5OfnIyoqClZWVkhPT5c6GhFRhTl48CAGDhyI+/fvQwiB+/fvY+DAgTh48KDU0Uim2LwSEVUgc3NzWFhYIDs7G5MnT0ZAQACEEFAoFIbHs7KyJE5JVcGMGTPg6uoKY2NjuLq6cvRNkg1PT0/4+fnBz88PQUFBxR6LiYnBoEGDMGzYMDYg9FiXLl1CXFwc9Ho9AECv1yMuLg6XLl2SOBnJFc95JSKqYKmpqZg4cSJ8fX3h7u6OpUuXGh7LyclB7dq1JUxHVcXSpUvRqVMnuLi44OjRo8X+HxFJpaCgAACwefPmEo/dvXsXmzdvRmxsLAoKCuDr64suXbpArVZXdkyqIpRKJbKystCwYUOkpaXB2toaaWlpUKlUUkcjmeKeVyKiCvT3339jzJgxmD59OoYMGQIAaN26NU6ePAkAOHz4MDp06CBlRKoCrK2tkZ6eDjc3N6jVari5uSE9PR3W1tZSR6MaLikpCXl5eRgzZgxGjhyJs2fPGh47f/482rVrB7VaDUtLS9ja2iIpKUnCtCR3Op0OSqUSM2bMQE5ODmbMmAGlUgmdTid1NJIpNq9ERBVo3bp1yMzMxJo1awyH1QUEBCAiIgLDhw+HVqtF7969pY5JMte4cWMAMBxuXvSzaDqRVExNTTF27FhERkbik08+wbRp01BYWAgAyM7OhqWlpWFec3NzZGdnSxWVqoihQ4di48aNsLS0xMaNGzF06FCpI5GM8bBhIqIKFBISgpCQkBLTt2zZIkEaqqoSEhLg4OCAP/74AwUFBVCr1WjRogUSEhKkjkY1XLNmzdC0aVMoFAo0a9YMVlZWuHv3Ll5++WVYWFggJyfHMG9OTk6xZpaoNPv27cO3335rOEWi6KglotJwzysREZHMCCFw+/ZtvPzyy1AqlXj55Zdx+/ZtCCGkjkY13LfffotFixYBANLS0pCdnY369esDANq2bYvTp0+joKAAWVlZuHr1Klq1aiVlXJI5a2trZGRkwNfXF6ampvD19UVGRgZPkaAysXklIiKSoczMzHLvE0lhyJAhyMrKgo+PDwIDAxEeHo7NmzcjLi4O9evXh5+fH3x9fTFq1CgEBgbCxMRE6sgkY6tXr4aJiQlu374NvV6P27dvw8TEBKtXr5Y6GskUDxsmIiKSIa1Wi7Zt2+LUqVMYO3Ys/vrrL6kjEUGtVmP58uXFprVv397w72HDhmHYsGGVHYuqsKJz+su6T/Qo7nklIiKSoZYtW+L7779H/fr18f3336Nly5ZSRyIiqlCTJk1CQUEBli9fjpycHCxfvhwFBQWYNGmS1NFIprjnlYiISIZSUlJgZGQErVYLIyMjpKSkSB2JiKhCpaenY8mSJZgyZQoAYMqUKdDpdJgxY4bEyUiuuOeViIhIZszNzZGXl2e41qFOp0NeXh7Mzc0lTkZEVLEOHz4MU1NTKBQKmJqa4vDhw1JHIhlj80pERCQz+fn5AID69etDoVAYRnMtmk5EVB0oFArs2bMHZmZmUCqVMDMzw549e3jeK5WJzSsREZHM6HQ6tGvXDnfu3IEQAnfu3EG7du0Me2KJiKobXgqMngSbVyIiIhn67bff0LBhQyiVSjRs2BC//fab1JGIiCqUEAJNmzZFRkYGhBDIyMhA06ZN2chSmdi8EhERyVRBQQH0ej0KCgqkjkJE9EIkJyeXe5/oUWxeiYiIZCojI6PYTyKi6qhWrVpQKBSoVauW1FFI5ti8EhERyVTDhg2hUCjQsGFDqaMQEb0wGo0GQghoNBqpo5DMsXklIiIiIiJJdO/eHXZ2dlAqlbCzs0P37t2ljkQyZlRZL6TVahEcHIybN29Co9FgwoQJaNGiBWbNmgWFQoGWLVti3rx5UCrZTxMREQFAWlpasZ9ERNXNL7/8ApVKBb1ej6SkJI6qTuWqtE5x9+7dsLKyQlRUFDZs2IAFCxbg008/RUBAAKKioiCEQFxcXGXFISIikr26detCqVSibt26UkchIqpwRkYP96MVNaxFP4umE/1TpTWvffr0wccff2y4r1KpcPHiRbz55psAgG7duuHYsWOVFYeIiEjWFAoF1Go1hBBQq9VQKBRSRyIiqlCFhYVPNZ2o0ppXc3NzWFhYIDs7G5MnT0ZAQACEEIY/xubm5sjKyqqsOERERLLWv39/3L9/H0II3L9/H/3795c6EhFRhXNycoKDgwOUSiUcHBzg5OQkdSSSsUo9wTQ1NRUjR46Eh4cH3N3di53fmpOTg9q1a1dmHCIiIlkyMjLCsWPH8OOPP0Kj0eDHH3/EsWPHeCgdEVU7Xl5eSEhIgE6nQ0JCAry8vKSORDJWac3r33//jTFjxmD69OkYMmQIAKB169Y4efIkAODw4cPo0KFDZcUhIiKSrfHjxyM9PR1ubm5Qq9Vwc3NDeno6xo8fL3U0IqIKFRoaihUrViA3NxcrVqxAaGio1JFIxiqteV23bh0yMzOxZs0a+Pn5wc/PDwEBAYiIiMDw4cOh1WrRu3fvyopDREQkawqFAiqVCsDDcSJ4zisRVTdt2rSBEAJTp06Fubk5pk6dCiEE2rRpI3U0kqlKO/4oJCQEISEhJaZv2bKlsiIQERFVCRs2bICvry/Onj2LxMRE2NnZwcnJCRs2bEBERITU8YiIKkRQUBB8fX1LnU5UGl5UlYiISGYKCgoQHx+PiIgI5OfnIyIiAvHx8SgoKJA6GhFRhRkxYgQAwMHBAcnJyXBwcCg2neif2LwSERHJjEKhQPPmzeHv7w9TU1P4+/ujefPmPHSYiKoVvV4Pe3t7JCQkwNbWFgkJCbC3t4der5c6GskUhy0kIiKSGSEE4uLiULduXej1ety6dQsXL16UOhYRUYVLTU0ttmHOyspKwjQkd9zzSkREJDNGRkZQKBTIyMgAAGRkZEChUPBSOURU7dy/f7/YYcP379+XOhLJGP8KEhERyUxhYSEAQKlUQq/XG34WTSciqk6uXLmCpk2bwtjYWOooJHPc80pERCRTRYfS8VxXIqqurKysoNVqAQBarZaHDVO52LwSERHJVP369aFUKlG/fn2poxARPRNHR0coFIpSbwBKHCZcdL+s5zg6Olb6eyD5YPNKRPQCnDt3Dn5+fgCA5ORk+Pj4wNfXF/PmzeMoivTEhBCGGxFRVZSQkFCslj16a9KkCQCgc+fOxX42adKkzOckJCRI9l5IemxeiYgq2IYNGxASEmK4Juenn36KgIAAREVFGUaRJXpSbFyJqLq6fv06mjRpgmPHjgEAjh07hiZNmuD69esSJyO5YvNKRFTBbG1tERERYbh/8eJFvPnmmwCAbt26Gf5IEz1OWlpasZ9EUtJqtZg+fTp8fX0xZMiQEhvivvzyS/Tv3x9+fn7w8/PDn3/+KVFSqkquX79u2EgnhGDjSuXiaMNERBWsd+/eSElJMdwXQhjO7TE3N0dWVpZU0aiKsLa2Rnp6OlQqFXQ6neGntbW11NGoBtu9ezesrKywdOlSZGRkwMvLCz179jQ8fvHiRSxevJjnJBLRC8PmlYjoBVMq/3eQS05ODmrXri1hGqoKzMzMkJmZabg0jk6ng5GREczMzCRORjVZnz590Lt3b8N9lUpV7PGLFy/i888/x927d9GjRw+MGzeusiMSUTXHw4aJiF6w1q1b4+TJkwCAw4cPo0OHDhInIrlLSUkpcU3XwsLCYnv0iSqbubk5LCwskJ2djcmTJyMgIKDY4/3790doaCi+/vprnD59GgcPHpQoKRFVV2xeiYhesJkzZyIiIgLDhw+HVqsttueCqDwvvfQSlEolXnrpJamjEAEAUlNTMXLkSHh4eMDd3d0wXQiBUaNGwdraGmq1Gt27d8elS5ckTEpE1REPGyYiegFsbGwQExMDAGjWrBm2bNkicSKqiu7evQu9Xo+7d+9KHYUIf//9N8aMGYO5c+fC2dm52GPZ2dkYMGAAfvjhB5iZmeHkyZMYPHiwREmJqLpi80pERCRTRQN9Ff0kktK6deuQmZmJNWvWYM2aNQCAoUOHIi8vD8OHD0dgYCBGjhwJtVoNZ2dndO/eXeLERFTdsHklIiKSKZ1OV+wnkZRCQkIQEhJS5uOenp7w9PSsxEREVNPwnFciIiKZevTah0RERDUdm1ciIiIiIiKSPTavREREREREJHtsXomIiIiIiEj22LwSERHJlFKpLPaTiIioJuNfQyIiIpnS6/XFfhIREdVkbF6JiIiIiIhI9ti8EhERERERkewZSR2AiIiISlKpVNDpdGXeJyKS2pqza7D23FrD/W0DtgEAvPd4G6ZNeH0CPnL6CG4xbribdxcAYG9tjxj3GIQeC0Xs77EAAMevHHEn9w4u3bsE/wP+hufPdZ6Loa2Gos3XbQzTutt0x+qeqzEpbhJ+SfnFMP3CqAv45so3mH98vuF1qXph80pERCRDOp0Or7zyCq5fvw5bW1v89ddfUkciIirmI6ePSm0QL4y6UGLagWEHSkwL7RyK0M6hAACFQoEGoxqggVmDUp9f2rTVPVeXmDa01VAMbTX0SeJTFcTDhomIiCTg6OgIhUJR6q3IX3/9Bb1eX6xxLes5jo6OErwLIiKiysPmlYiISAIJCQkQQpR6i4qKKtbEAg+b1qioqDKfk5CQINE7IaKaytrauswNak97A8reOPcsN2tra4k/HXoR2LwSERHJjI+PD7Zu3QoHBwcAgIODA7Zu3QofHx+JkxER/U9GRkaZG9SkvmVkZEj98dALwHNeiYiIZMjHxwc+Pj5QKBTcq0pERATueSUiIiIiIqIqgM0rERERERERyR6bVyIiIiIiIpI9nvNKRERUwdacXYO159Ya7m8bsA0A4L3H2zBtwusT8JHTR3CLccPdvLsAAHtre8S4xyD0WChif48FADh+5Yg7uXdw6d4l+B/wNzx/rvNcDG01FG2+bmOY1t2mO1b3XI1JcZPwS8ovhukXRl3AN1e+wfzj8w2vS0REVNWweSUiIqpgHzl9VGqDeGHUhRLTDgw7UGJaaOdQhHYOBfDw0hENRjVAA7MGpT6/tGmre64uMW1oq6EY2mrok8QnIiKSJR42TERERERERLLHPa9EREQVzNraukKvMahQKCpsWXXr1kV6enqFLY+IiKiysHklIiKqYBkZGRBCSB2jVBXZCBMREVUmHjZMREREREREssfmlYiIiIiIiGSPzSsRERERERHJHs95JSIiqmAXJpgDoXWkjlGqCxPMpY5ARNUEax1VNjavRESVQK/XIzQ0FJcvX4ZarcbChQvRtGlTqWPRC9JmbY7UEcpUt25dpK+ROgVVVY+rZTExMdi2bRuMjIwwYcIEuLq6SpiWXjTWOqpsbF6JiCrB/v37odFosH37dpw9exaLFi3C2rVrpY5FL0hFjjSsUChkO3Ix1Tzl1bK7d+9i8+bNiI2NRUFBAXx9fdGlSxeo1WqJU9OLwlpHlY3nvJLsODo6QqFQPPYG4Inmc3R0lPgdEQGnT59G165dAQBOTk5ISEiQOBFJjbWOqqLyatn58+fRrl07qNVqWFpawtbWFklJSVJFJZlgraOKxD2vJDtcqafqKDs7GxYWFob7KpUKhYWFMDJiGa6pWOuoKiqvlmVnZ8PS0tLwmLm5ObKzs6WISTLCWkcViXteiYgqgYWFBXJy/ndukF6vZ+NKRFVOebXsn4/l5OQUa2aJiJ4Xm1ciokrQvn17HD58GABw9uxZtGrVSuJERERPr7xa1rZtW5w+fRoFBQXIysrC1atXWeuIqEJxsz8RUSXo1asX4uPj4e3tDSEEwsPDpY5ERPTUSqtlX375JWxtbdGzZ0/4+fnB19cXQggEBgbCxMRE6shEVI2weSUiqgRKpRLz58+XOgYR0XMprZY1b97c8O9hw4Zh2LBhlR2LiGoIHjZMREREREREssfmlYiIiIiIiGSPzSsRERERERHJnuTnvOr1eoSGhuLy5ctQq9VYuHAhmjZtKnUsIiIiIiIikhHJ97zu378fGo0G27dvx9SpU7Fo0SKpIxEREREREZHMSN68nj59Gl27dgUAODk5ISEhQeJEREREREREJDeSN6/Z2dmwsLAw3FepVCgsLJQwEREREREREcmN5Oe8WlhYICcnx3Bfr9fDyKjsWDdv3sSgQYMqIxoRVRE3b96UOkKFY60jotJUt3rHWkdEpSmr1knevLZv3x4HDx5Ev379cPbsWbRq1arc+U+ePFlJyYiIpMNaR0Q1AWsdET0NhRBCSBmgaLThK1euQAiB8PBwNG/eXMpIREREREREJDOSN69EREREREREjyP5gE1EREREREREj8PmlYiIiIiIiGSPzSsRERERERHJHptXIiIiIiIikj02r1QlnTt3Dn5+flLHICJ64VjviKgmYK2jJyH5dV6JntaGDRuwe/du1KpVS+ooREQvFOsdEdUErHX0pLjnlaocW1tbRERESB2DiOiFY70jopqAtY6eFJtXqnJ69+4NIyMeNEBE1R/rHRHVBKx19KTYvBIREREREZHssXklIiIiIiIi2WPzSkRERERERLKnEEIIqUMQERERERERlYd7XomIiIiIiEj22LwSERERERGR7LF5JSIiIiIiItlj80pERERERESyx+aViIiIiIiIZI/NKxEREREREckem1ciIiIiIiKSPTavREREREREJHtsXomIiIiIiEj22LwSERERERGR7LF5JSIiIiIiItlj80pERERERESyx+aViIiIiIiIZI/NKxEREREREckem1ciIiIiIiKSPTavREREREREJHtsXomIiIiIiEj22LwSERERERGR7LF5JSIiIiIiItlj80pERERERESyx+aViIiIiIgikFveAAAgAElEQVSIZI/NKxEREREREckem1ciIiIiIiKSPTavREREREREJHtsXomIiIiIiEj22LwSERERERGR7LF5JSIiIiIiItlj80pERERERESyx+aViIiIiIiIZI/NKxEREREREckem1cZ+PbbbzF06FD069cPb7/9NkaPHo1z585V6GvMnz8fERERz7WMiIgIzJ8/v8R0Nzc3XLhwAWlpafD29i53GTdu3IC/v/9z5XjRit5PZTp//jzmzp0LALhw4QImT578XMsbM2YM0tPTKyIaUZWh1Wrh4uKC999/3zDt5MmTGDBgAABg1qxZiIyMfKplvvbaa0hPT0dcXBwWLlxYYVnd3NzQu3dveHh4wNPTE/3790d4eDj0ej1SUlLQrl27Cnutsjz62TyN1atXY//+/S8gEZH8Pa7OPIuKri9F/P390alTJ+Tl5T3T8z08PJCZmfnMrx8SEoKEhIRnfv6Tys7Ohre3N/r374+9e/e+8Nd7EhEREXjrrbfg4eGBgQMHom/fvpg6dSqys7Mr/LV27NiBcePGPXa+yvp9vGhsXiW2YsUKxMbG4t///jd++OEH7N+/Hx9++CHGjRuHW7duSR3vqTRs2BDbtm0rd55bt27h2rVrlZSo6vjjjz+QlpYGAGjTpg1WrVr1XMuLj4+viFhEVcq+fftgZ2eHhIQEXL16tUKX3bNnT4SEhFToMpctW4Zdu3Zh586d+O6773Du3DlERUVV6Gu8CCdPnkRhYaHUMYgk8SLqzIuoL2lpaTh16hScnJywc+fOZ1rGrl27ULt27WfOcOzYMQghnvn5TyoxMRH37t3Df//7X7zzzjsv/PWeVL9+/bBr1y7s3r0be/bsQU5ODjZv3ixZnsr6fbxoRlIHqMn+/vtvfP3119i3bx8aNGhgmO7s7IxZs2YZtpSlpaVh/vz5SE1NhVarRf/+/TF+/HikpKTgvffeQ/fu3XHu3DlkZmZi+vTp6NWrF7KzszF79mwkJSWhQYMGUKlUeOONNx67vHfffRfNmzfHzZs3sXnz5mK5HiclJQXu7u747bffcPXqVcyePRsajQZCCAwZMgTe3t4ICQlBWloaxo4di8jISOzfvx+rV6+GXq+Hubk5goKC0LZtW+Tl5WHevHk4d+4cLC0t0aJFCwDAokWL4ObmhrZt2+Ly5cuYMmUKjIyMsH79emg0GqSnp8PT0xMBAQE4efIkVqxYgZdffhnXrl1DrVq18OGHH2Lz5s24du0a3nnnHQQHB5f6XqKiopCUlASNRoPRo0djyJAhCAkJQb169RAYGAjgYVHfu3cv/vOf/xR7rpubG/r374/4+HhkZWVh9OjR8PX1hV6vR3h4OM6dO4ecnBwIIbBw4UI0atQIq1atQlZWFoKCguDp6YkFCxZgz5490Gg0WLZsGU6dOgWdTofWrVsjJCQEFhYWcHNzg5eXF44fP47U1FR4eHggICAAQUFBAIBRo0bh888/x8GDB7Ft2zYYGxvDxMQE8+fPN3yeRNVJdHQ0+vXrB1tbW3z99delHilSJDw8HJcvX8aaNWtw584dzJ8/Hzk5Obh79y7s7Ozw73//GyYmJob5d+zYgZ9//hnr16+Hn58fnJyccObMGaSmpsLZ2RkLFiyAUqnEmTNnsGzZMuTl5UGpVGLSpElwdXV9bHa1Wo033ngDf/75J3r06AGdToe5c+fiwoULyMrKwvTp09G7d29otVosWrQIx48fh0qlQtu2bREUFAQLCwtERUWV+l0vqyYBQG5uLgIDA/Hnn3+ioKAACxcuRIcOHZCVlYVPPvkESUlJUCgU6Nq1K6ZMmYLt27cjISEBS5YsgUqlQosWLcr87FatWoV9+/bB2NgYdevWxaeffooGDRrg6tWrCAsLw/3796HT6eDn54chQ4Y8/38AokrwuDpz7dq1Mr8Tbdq0wYcffoj4+HjcuXMH77//Pnx9fV9IfYmJiYGzszN69+6NlStXwtvbGwqFAsDDjeSjR4/GsWPHkJubi0mTJuGnn37ClStX0KBBA6xbtw5mZmZ47bXXcPz4cRw6dAj79u2DUqlEcnIyTE1NsXjxYjRv3hy3b99GaGgobt68CSEEPD098f777+Ozzz7DnTt3MG3aNCxZsgTLli1DnTp18Oeff8LHxwdt2rTB0qVLodFocPfuXXTu3Bnh4eHlrtuWtm7p7OyM4OBgpKWlwcPDA9u3b8elS5dK/Zx27NiBb7/9Fnl5ebCwsMD69esRGhqK5ORk3L9/H+bm5li2bBleffVV7N27F2vXroVCoYBKpcKMGTPQsWNHZGVlISwsDFeuXIFWq4WzszNmzJgBI6PyW6qCggLk5uaifv36AFBmjU1OToa3tzc2bdoEe3t7w7LDw8PRunVrfPDBBzhy5Ahyc3MxZcqUEs36k/4+Xn/99Wf9CkhPkGT27dsnvLy8Hjufn5+fiIuLE0IIkZ+fL/z8/MR///tfcePGDdGqVStx4MABIYQQP/30k+jRo4cQQoiwsDAxY8YModfrxb1790S3bt3EqlWrnmh5p06dKjXHqlWrRKdOncTAgQOL3RwcHMT58+fFjRs3hJOTkxBCiKCgILF+/XohhBB37twRAQEBQqfTiRMnToj+/fsLIYT4448/ROfOncX169eFEEIcO3ZMdOnSRWRlZYlly5aJKVOmCJ1OJ7KysoS7u7uYOXOmEEIIV1dXsXr1aiGEEHq9XowYMUJcu3ZNCCHE7du3hb29vbh37544ceKEsLe3FxcvXhRCCDF27FgxfPhwUVBQIO7duyccHBzE7du3S7xPV1dXMW/ePMPynJ2dxZUrV8SlS5dEly5dhFarFUII4evrKw4fPlzq8+fMmSP0er1ITU0VnTp1EklJSeLMmTPC399f6HQ6IYQQ69evF+PGjRNCCBEbGys+/PBDIYQo9hlFRESIRYsWCb1eL4QQYvny5YZsrq6uYtGiRYacbdq0MXyWrVq1Evfu3ROFhYXCwcFBpKWlCSGE+O6778S2bdtK/f0SVWW///67cHBwEOnp6eLcuXOibdu2Ij09vdj3aebMmWLDhg3ik08+ERMnThQFBQVCCCEWLVokdu7cKYQQQqPRiAEDBoiffvpJCPG/79Kj39ERI0aIyZMnG+qTi4uLOH78uLh//7545513xI0bN4QQD7+X3bp1Ezdv3iyR19XVVZw/f95w//bt26JPnz7ip59+MtTiogx79+4VPXv2FEIIsXLlSjFp0iSh0WiETqcTs2bNEnPmzCn3u15WTSqqkWfPnhVCCPHll1+KkSNHCiGEmDFjhliwYIHQ6/WioKBAjBkzxlDTR4wYIX788cdyP7tbt26J9u3bGz7jyMhIsW/fPqHVakW/fv1EQkKCEEKIzMxM0bdvX/Hbb7898++eqLI8SZ15XD3ZvHmzEEKICxcuCEdHR5Gfn1/h9UWr1QoXFxdx4MABUVBQIDp27CgOHTpkeLxVq1bi66+/FkI8XBdp166duH37ttDpdMLLy0vs3r3bMF9R/XvjjTdEamqqEEKI+fPnixkzZgghhHj33XfFxo0bhRAPv8/u7u5iz549QojidW7EiBEiKCjIkCEwMFCcOHFCCCFEdna26NSpk7hw4UK567ZPsm5Z3ucUGxsrOnbsKLKysoQQQvz4449iwYIFhkxz5swR8+fPF0II0bNnT0NdOnLkiIiIiBBCCDFr1iyxadMmIYQQhYWFYtq0aeLzzz8v8Tt4dJ15wIABon379mLAgAHiwYMHQojya+z27duFu7u7iImJEe7u7iIvL8/w+1i7dq0QQojExETxxhtvlPj79KS/j6qMe14lJP6x6z47OxvvvvsugIdbw/v27Yvx48fj1KlTePDgAVauXGl4LCkpCW3btoWxsTG6d+8OAGjdujXu378PADh+/DiCg4OhUChgbW2NXr16GZ5b3vKMjIzg5ORUZuZ+/foZzs0s4ubmVmK+Xr16YebMmTh//jycnZ0REhICpbL4UeonTpzAW2+9hSZNmgB4uMfZ2toaCQkJ+OWXXxAUFASlUgkLCwt4eXnh8uXLhud26NABAKBQKLBu3TocOnQIe/bswdWrVyGEMOy1trGxQevWrQEAtra2sLS0hFqthrW1NczNzfHgwQM0bNiwRP6ic3cbNmyILl264Pjx4xg5ciRsbGxw6NAhNGvWDHfu3IGLi0upn5Ovry8UCgVeeukldO3aFfHx8RgzZgzq1KmDbdu24caNGzh58iTMzc3L/KwB4NChQ8jKysKxY8cAPDzXpl69eobHe/bsachZr149PHjwwPB5AoBKpUKfPn3g7e2NHj16wMXFxfD/hag6iY6OhqurK+rWrYu6devCxsYGMTExJerZV199hXv37mHnzp1Qq9UAgOnTpyM+Ph4bNmzAX3/9hTt37iA3N7fc13N1dTXUp6ZNm+LBgwc4e/Ys7t69i4kTJxrmUygUuHz5Mho1alRiGdOmTYOpqSn0ej2MjY0xdOhQ9O7dGykpKTA2Nkbv3r0BAHZ2drh37x4A4PDhwwgMDISxsTEAwM/PDxMnTnzsd720muTg4IAmTZoYtsDb2dkhNjbW8DrR0dFQKBRQq9Xw9vbG119/jQ8//LDYeyjrs2vYsCHs7Ozg5eWFbt26oVu3bnB2dsYff/yB69evFzvqJT8/H5cuXSr3bw+RHDxJnXlcPSn6u+3g4ACNRlNqrXne+hIXFwe9Xo+uXbvCyMgI/fr1w6ZNm4rVhKL6Ymtri1atWhnWhWxsbPDgwYMSmRwcHPDSSy8BeLi+uW/fPuTm5uLMmTPYuHEjAMDS0hKDBg3C4cOH0b9//xLLKFp3Ax4eSXf48GGsW7fOcORHbm4urKysyly3fZJ1y/I+J+DhOAYWFhYAgD59+qBJkybYvHkzkpOT8euvvxrGG+jfvz8mTZqE7t27o0uXLvjggw8APFwvu3DhAr799lsAD+tXWR5dZ9ZqtZg/fz4CAwMRGRlZbo0dNmwYjh49ioULF2LXrl0wNTU1LHPEiBEAHtbrVq1a4dSpU4bHnvb3UVWxeZVQ27Ztce3aNWRkZKBu3bqwsLDArl27ADw80TsjIwN6vR5CCGzbtg21atUCAKSnp8PExAQZGRkwNjY2fHGLDgcp8mhzrFKpAOCxy1Or1Y899OFJuLq64ueff8axY8dw/Phx/Oc//8GOHTuKzaPX60vNXFhYCCMjo2L5/1mczMzMADz8onp5eeHtt99Ghw4dMHjwYOzfv9/w3KKV0yJP+t4efT29Xm943rvvvovY2Fi88sorGDZsWIn8pb2OXq+HUqnEoUOHEBYWhtGjR6Nnz5549dVXsXv37nJz6PV6BAcHG4p4Tk4OCgoKDI8/elijQqEo9VyGZcuW4cqVKzh27Bg+//xz7Nq1y7Dhgqg6yM3Nxa5du6BWqw0b07Kzs7FlyxY4OjoWm7djx45o3749goKCsH37dhgbG2PKlCnQ6XTo27cvevTogdTU1MeeF/ToykTRd0+n06F58+b45ptvDI+lpaXB2tq61GUsW7YMbdq0KfWxoua0aPlF/lk39Xo9tFqtYXllfddLq0mlvU7R+y7tdUo7z7Wsz06pVGLLli24cOECjh8/jvDwcHTt2hUeHh6wtLQ0/K0DHp5CY2lpWernQCQXT1pnHldPiv5uF32/Sqs1z1tfoqKikJ+fbziktOjQ3N9//x0tW7YEUPy7/+i/y1JapqJ1ykeVVSuA/627AQ+bsNdeew1du3ZF3759ce7cOcOyylq3fZJ1y/I+p++//75YhqioKMTExODdd9+Fu7s7rKyskJKSAgAIDAzE4MGDER8fjx07dmDjxo349ttvodfrsXLlSjRv3hwAkJmZWea64KOMjY3h6+uL4cOHGz6nsmqsRqNBcnIyLC0tkZiYiFdeecUwX9H6fNFz/nn/aX4fVRUHbJJQw4YNMXLkSHz88cfFBme6efMmzpw5Y9jq5uTkhC+//BLAwy+Jj48P4uLiyl12165dDV+yBw8eGOZ/1uU9ralTp+KHH35A//79MW/ePFhYWOD69etQqVSGFS1nZ2ccPXoUN27cAADDuZuvv/46unfvjtjYWOj1euTl5WHPnj2lFofk5GRkZ2cjICAAbm5uOHnyJDQaDfR6/XPl/+677wA8HGDq+PHjcHZ2BvBwS2ViYiJ+/vlnDB48uMznFw2OcOvWLcTHx6Nbt26Ij4+Hq6srfH194ejoiP3790On0wF4WIxKKy4uLi7YunWr4T3NmTMHK1aseGz+ouWlp6eje/fusLKywnvvvYeAgIBKH0mZ6EX7/vvvYWVlhSNHjuDAgQM4cOAA9u/fj9zc3BKjbjs6OmLEiBGwtLTE6tWrAQBHjx7FxIkT0a9fPwDAuXPnDN/Np+Hk5ITk5GTDlvDExET07t3bMBhbRejatSuio6Oh1Wqh1+uxdetWdOnS5bHf9dJqUnlcXFywZcsWCCGg0WgQExODzp07Ayher8r67JKSkjBgwAA0b94c48aNw3vvvYcLFy6gWbNmMDU1NTSvqampGDBgQLUYAZOqtyetMxVVT/7pSevLtWvXcOrUKezYscOQ8+jRo+jYsSM2bdr03DkeZWFhgddffx1bt24F8PA8zp07d5ZaKx6VmZmJCxcuYNq0aXjnnXdw+/ZtXL9+/bHrbmWtWz7qaerw0aNH4eXlhaFDh6JZs2Y4cOAAdDodCgsL4ebmhry8PPj4+GDevHm4fPkyNBoNXFxc8NVXXxlq44QJE7Bly5Yn+rwOHTqEtm3bAii/xi5ZsgQtW7ZEZGQkFi5ciJs3bxqWUVTLL168iGvXrqFjx46Gx57191HVcM+rxAIDA7F7925MnToVeXl5yMrKQp06ddCvXz/DIcTLli3DggUL4O7uDo1GgwEDBmDgwIGGrUOl8ff3x7x589C3b19YW1ujVatWhseeZXlP66OPPsLs2bOxfft2qFQqvP322+jYsSMePHgAExMTDBkyBN988w3mzZuHSZMmQafTwdTUFOvWrYOlpSXGjRuH+fPnw93dHZaWlqhXr16xrX5FXnvtNfTo0QN9+/aFWq1Gq1at0KJFCyQnJ5fY6/o0CgoK4OXlBa1Wi5CQEDRr1gzAwz25vXv3xt9//13m3hTg4eBVgwYNQn5+PkJCQvDqq6/C29sbU6dOhbu7OwoLC9GlSxfs3bsXer0eTk5O+M9//oNJkybBz8+v2Oe4ePFieHl5QafTwd7eHrNmzXps/j59+sDPzw8RERGYMGEC3nvvPZiamkKlUr2Q4fiJpBQdHY3Ro0cX2wJdu3Zt+Pn54auvvioxv0KhQHh4ODw9PdG9e3cEBgZi4sSJMDMzg4WFBTp27FhihehJWFtbY9WqVViyZAkKCgoghMCSJUtgY2PzPG+vmAkTJmDx4sXw9PREYWEh2rZtizlz5qB27drlftdLq0l3794t83VCQkKwcOFCuLu7Q6vVomvXrhg/fjyAh6eKrFixAlqttszPbujQoejbty8GDx4MMzMzmJqaIiQkBGq1GmvWrEFYWBi++OILFBYW4uOPPzYMKEgkV09aZyqqnvzTk9aX6OhovP3222jatGmx6RMnTsS4ceMMg05WlGXLlmH+/PnYsWMHNBoN3N3dMWjQIAAPD/OdPn06QkNDiz2ndu3a+PDDD+Hl5QUzMzM0bNgQ7du3R3JycrFTn/6prHXLX3/91TBPeZ/To/MBDy8rOHfuXMMhwE5OTrhy5QqMjIwQHByMadOmwcjIyPA3Q61WY/bs2QgLCzPUxs6dOxe7bNKjfvjhB5w+fRoKhQIFBQVo0qQJFi9eDKDsGls0ONb333+P2rVrY9SoUZg6daqhQT5z5gxiYmKg1+vx2WefoU6dOs/0+yjrtLeqQCEed2wUkQT++9//wsLCAt27d4der4e/vz+6dOliGCFTKrm5uRgxYgTmzp1b5vlZbm5uWLlyZZmHAxIRVSbWJCKiqq9o9Ofydp7UBDxsmGSpZcuWWLt2LTw8PDBgwAA0aNAAQ4cOlTTTkSNH0KNHD3Tt2pUDixARERERVTLueSUiIiIiIiLZ455XIiIiIiIikj02r0RERERERCR7VW604U6dOqFx48ZSxyAiGbl58yZOnjwpdYwKxVpHRKWpbvWOtY6ISlNWratyzWvjxo1LXJCYiGq2omHgqxPWOiIqTXWrd6x1RFSasmodDxsmIiIiIiIi2WPzSkRERERERLLH5pWIiIiIiIhkr8qd81oarVaLlJQU5OfnSx1FNkxNTWFjYwNjY2OpoxARERERET23atG8pqSkwNLSEq+88goUCoXUcSQnhMC9e/eQkpKCZs2aSR2HiIiIiIjouVWLw4bz8/NRr149Nq7/n0KhQL169bgnmoiIiIiIqo1q0bwCYOP6D/w8iIiIiIioOqk2zeuj8gsrdo9jRS+PiIiIiIiInk61OOf1n0yNTKH4pOL2PIp54rHznDx5Etu2bcNnn31mmLZs2TK8+uqrT3xB8ZSUFEyZMgUxMTE4deoULC0tYWdnhy5duiA+Pv6Z8xMREREREVV11XLPa3UQGxuLO3fuSB2DiIgkEh0dDUdHR6hUKjg6OiI6OlrqSFTNabVaTJ8+Hb6+vhgyZAji4uKQnJwMHx8f+Pr6Yt68edDr9QCA1atXY8iQIfD29sb58+cBoMx5icrDWkdPo1rueZWb5cuX49SpUxBC4L333kPfvn3x66+/YvXq1QAeDji1ePFiw2VtEhIScOTIEVy8eBEtWrSARqPB1KlTcevWLVhZWWHVqlXw8/PDggUL0LJlS/zyyy84dOgQ5s2bJ+XbJCKiChIdHY3Zs2cjMjISLi4uOHr0KMaOHQsA8PHxkTgdVVe7d++GlZUVli5dioyMDHh5ecHOzg4BAQHo1KkT5s6di7i4ODRq1Ai//vorvvnmG6SmpsLf3x+xsbH49NNPS8zbq1cvqd8WyRhrHT0t7nmtQCdOnICfn5/htmfPHuTn5yMlJQXbtm3Dpk2bsG7dOmRmZuL333/H0qVLsWnTJri5ueGnn34yLMfR0RFdu3bF9OnT0ahRI+Tm5iIwMBDR0dHIzs5GYmIihg4diu+++w7Aw720Q4YMkeptExFRBQsLC0NkZCRcXV1hbGwMV1dXREZGIiwsTOpoVI316dMHH3/8seG+SqXCxYsX8eabbwIAunXrhmPHjuH06dNwcXGBQqFAo0aNoNPpkJ6eXuq8ROVhraOnxT2vFeitt94qcc5rTk4OLl68CD8/PwBAYWEhbt26hYYNGyIsLAxmZmZIS0tD+/bty1xunTp1YGNjAwD417/+hby8PPTr1w9eXl4YO3Ysbt++DQcHhxf75oj+KT8fMDWtessmqgISExPh4uJSbJqLiwsSExMlSkQ1gbm5OQAgOzsbkydPRkBAABYvXmy4goG5uTmysrKQnZ0NKyurYs/LysqCEKLEvETlYa2jp8U9ry+YiYkJOnXqhM2bN+Prr79G3759YWNjg5CQEISHh2PRokVo0KABhCg+KJRCoTBMK+2yN7Vq1UKnTp0QFhYGDw+PSnkvRMWYmgIKxYu5sXGlGs7e3h5Hjx4tNu3o0aOwt7eXKBHVFKmpqRg5ciQ8PDzg7u4OpfJ/q4o5OTmoXbs2LCwskJOTU2y6paVlqfMSlYe1jp5Wtdzzml+Y/0QjBD/N8kyNnm1l2tzcHGZmZvD19UVubi7efvttWFhYwMPDA8OGDUPt2rXxr3/9q8TgTK+//jqWLVtm2ONammHDhsHHxwehoaHPlI2IiORp9uzZGDt2bInzwHgoHb1If//9N8aMGYO5c+fC2dkZANC6dWucPHkSnTp1wuHDh/HWW2/B1tYWS5cuNRz9pdfrYW1tXeq8ROVhraOnVS2b12dtNJ9neZ06dUKnTp2KTZs2bVqZ8wcFBSEoKKjE9JiYGACAt7c3vL29AaDYZXIePSxZp9OhT58+3LJJRFTNFA1U4u/vj8TERNjb2yMsLIwDmNALVTQux5o1a7BmzRoAD5uLhQsXYsWKFXj11VfRu3dvqFQqdOjQAcOHD4der8fcuXMBADNnzsScOXOKzUtUHtY6elrVsnmtCbZs2YLY2FisWrVK6ihERPQC+Pj4cAWOKlVISAhCQkJKTN+yZUuJaf7+/vD39y82rVmzZqXOS1Qe1jp6Gmxeq6gRI0ZgxIgRUscgIiIiIiKqFBywiYiIiIiIiGSPzSsRERERERHJHptXIiIiIiIikr3q2bzm58t7eURERERERPRUqmfzamoKKBQVdzN9/KVyTp48CWdnZ/j5+cHPzw/Dhg3D5s2by5x/+/bt0Gq1SExMxOrVq5/6LZ46dQpJSUlP/TwiIiIiIqKqiKMNV6C33nrLcB1WjUaDPn36wMPDo9TrsK5fvx6enp6wt7eHvb39U79WbGws+vXrBzs7u+fOTUREREREJHdsXl+Q7OxsKJVKJCUlGfas5ufnY/Hixfi///s/3L17F4GBgRg1ahS2bduGzz77DD/++CO++uorKJVKvPHGG5g2bRoiIiKQkpKCe/fu4datWwgKCkLdunVx5MgRXLx4ES1atECjRo0kfrdE1ZtWq0VwcDBu3rwJjUaDCRMmoEWLFpg1axYUCgVatmyJefPmQalUYvXq1Th06BCMjIwQHByMtm3bIjk5udR5iYiIiOjJsXmtQCdOnICfnx8UCgWMjY0xZ84c/P7771i6dCkaNmyIdevW4aeffsKECROwdu1afPbZZzh79iwA4P79+4iIiEBsbCxq1aqF6dOnIz4+HgCgVqvxxRdfID4+Hhs3bkRkZCS6du2Kfv36sXElqgS7d++GlZUVli5dioyMDHh5ecHOzg4BAQHo1KkT5s6di7i4ODRq1Ai//vorvvnmG6SmpsLf3x+xsbH49NNPS8zbq1cvqd8WERH9P/buPq7q+v7/+OMAAgooorFExPBiBaJtxaIWWSwV81uzbIboZlPSqHm8qomC4v4jL4YAACAASURBVCUSXtDX4ZTlsBQVL6o1E1tNxRGauq+lW3iqaaZ5UZqIyREBOef3hz/PdqaIKPgBz/N+u3GT8+Z9Pj5PNzt8Xud9JSJNiorXevSf04Yv2bRpE2lpabRo0YJvv/2We+6554rPPXz4MCUlJYwcORIAq9XK119/DeCYVnz77bdTWVnZgK9ARK6kb9++xMbGOh67u7tTXFzMfffdB0DPnj3Ztm0boaGhREdHYzKZCAoKorq6mpKSkiv2VfEqIiIiUjeat9bAJk+ezOzZs3nllVcIDAzEbrcDYDKZsNlsjn7BwcG0a9eOpUuXkpubyy9/+UvuvvtuR9//ZjKZHNcSkYbl4+ODr68vZWVljB49mrFjx2K32x3/b/r4+HD27FnKysrw9fV1et7Zs2ev2FdERERE6ubWLF7Pnwe7vf6+buConP79+/PMM88waNAgrFYrJ06cACAyMpKRI0c6CtCAgAB+/etf86tf/YqBAwdSWFjIHXfcUeN17777bubNm8eBAweuO5uIXLvjx48zdOhQ+vfvzxNPPOG0ZtVqtdKyZUt8fX2xWq1O7X5+flfsKyIiIiJ1c2tOG76Go23q+3pRUVFERUVd1j5p0iQmTZp0WXtGRobj+/vvvx+4WOj279/fqZ/ZbHZ837lzZ8fxO4MGDWLQoEHXll9Ebsh3333H8OHDSU1N5YEHHgAgPDycnTt3EhUVRWFhIffffz8hISHMnTuXhIQEvvnmG2w2GwEBAVfsKyIiIiJ1c2sWryIi9Sg7O5vvv/+eRYsWsWjRIgBSUlKYNWsWmZmZdOrUidjYWNzd3YmMjCQuLg6bzUZqaioASUlJTJkyxamviIiIiNSNilcRkVpMnjyZyZMnX9a+YsWKy9rMZrPTjAmA0NDQK/YVERERkWt3y6x51eZFzvTfQ0REREREbiW3RPHq7e3NqVOnVLD9f3a7nVOnTuFd32t/RUREREREDHJLTBsODg7myJEjnDx50ugojYa3tzfBwcFGxxAREREREakXDVa8njp1igEDBrB06VI8PDyYOHEiJpOJrl27MnXqVNzc3Fi4cCFbt27Fw8OD5ORkevTocV1/V7NmzQgNDa3nVyAiIiIiIiKNRYNMG66qqiI1NdUxbTU9PZ2xY8eyatUq7HY7mzdvpri4mF27drFu3ToyMzOZPn16Q0QRERERERGRW0CDFK8ZGRkMGjSIwMBAAIqLi7nvvvsA6NmzJ9u3b2f37t1ER0djMpkICgqiurqakpKShogjIiLS5OTl5REREYG7uzsRERHk5eUZHUlERMRQ9V68vv322wQEBPDQQw852ux2OyaTCQAfHx/Onj1LWVkZvr6+jj6X2kVERFxdXl4eKSkpZGVlcf78ebKyskhJSVEBKyIiLq3e17y+9dZbmEwmPvroIywWC0lJSU4jqlarlZYtW+Lr64vVanVq9/Pzq+84IiIiTU5aWho5OTnExMQAEBMTQ05ODmazmfj4eIPTiYiIGKPeR15XrlzJihUryM3NJSwsjIyMDHr27MnOnTsBKCwsJDIyknvuuYeioiJsNhvHjh3DZrMREBBQ33FERESaHIvFQnR0tFNbdHQ0FovFoEQiIiLGuylH5SQlJTFlyhQyMzPp1KkTsbGxuLu7ExkZSVxcHDabjdTU1JsRRUREpNELCwujqKjIMfIKUFRURFhYmIGpREREjNWgxWtubq7j+xUrVlz2c7PZjNlsbsgIIiJOvv76a1auXMmuXbsoLS2lTZs2PPDAA8TFxdG+fXuj44kAkJKSQkJCAjk5OURHR1NUVERCQgJpaWlGRxMRETHMTRl5FRFpDBYuXMjXX39N3759GTp0KLfddhvff/89e/fu5dVXX6Vjx476QE0ahUvrWs1mMxaLhbCwMNLS0rTeVW6KvXv3Mm/ePHJzcxk3bhzfffcdAEePHuXuu+/m1VdfJTExkdLSUpo1a4aXlxd//OMfOXToEBMnTsRkMtG1a1emTp2Km1uDHGwhIi5KxauIuIw+ffrwwx/+0KmtTZs2/OxnP+NnP/sZn3/+uUHJRC4XHx+vYlVuuiVLlrB+/XqaN28OwKuvvgrAmTNnGDp0KJMmTQLg8OHD5OfnO06TAEhPT2fs2LFERUWRmprK5s2b6d27981/ESJyy9LHYSLiMv6zcP3oo49Yu3Ytn332GRUVFQDceeedRkUTEWkUQkJCyMrKuqw9KyuLX/7ylwQGBvLdd9/x/fffk5iYSHx8PAUFBQAUFxdz3333AdCzZ0+2b99+U7OLyK1PI68i4nIyMzP55ptvOHDgAM2aNeO1114jMzPT6FgiIoaLjY3lyJEjTm2nTp3io48+coy6VlVVMXz4cIYOHcqZM2eIj4+nR48e2O12x0isj48PZ8+even5ReTWppFXEXE5u3fvZs6cObRo0YKnnnrqshs1ERH5t7/85S88/vjjuLu7A9C2bVsGDRqEh4cHbdq0ISwsjIMHDzqtb7VarbRs2dKoyCJyi1LxKiIup7q6moqKCkwmE9XV1dpQRETkKj766CN69uzpeLx9+3bGjh0LXCxS//Wvf9GpUyfCw8PZuXMnAIWFhURGRhqSV0RuXZo2LCIu59e//jUDBgygpKSEgQMHMmzYMKMjiYg0WgcPHqRDhw6Oxw8//DBFRUU888wzuLm5MX78eAICAkhKSmLKlClkZmbSqVMnYmNjDUwtIrciFa8i4nL8/f1ZtWoVhw4dIjg4mICAAKMjiYg0GsHBwaxdu9bxOD8//7I+KSkpl7WFhoayYsWKBs0mIq5Nc+VExOVkZWXRqlUrevToocJVREREpIlQ8SoiLsdkMvGb3/yGefPmkZmZqZ2GRUREDJKXl0dERATu7u5ERESQl5dndCRpxDRtWERcztNPP+30+NLRDiIiInLz5OXlkZKSQk5ODtHR0RQVFZGQkABAfHy8wemkMdLIq4i4nH/+85889dRTjq/t27cbHUlERMTlpKWlkZOTQ0xMDM2aNSMmJoacnBzS0tKMjiaNlEZeRcRlrFy5ksWLF1NaWsoHH3zgaO/cubOBqURERFyTxWIhOjraqS06OhqLxWJQImnsNPIqIi5jyJAhFBUVMWrUKIqKihxfy5YtMzqayGW0DkxEbnVhYWEUFRU5tRUVFREWFmZQImnsNPIqIi6joKCAmJgY/P39WbNmjdPP4uLiDEolcjmtAxMRV5CSkkJCQsJl73WaNiw1UfEqIi6jtLQUgO+++87gJCJX95/rwADHOjCz2aziVURuGZfez8xmMxaLhbCwMNLS0vQ+JzVS8SoiLuOpp54CYNSoUU7tb7zxhgFpRGqmdWAiIiKX05pXEXF5GzZsMDqCiBOtAxMRV3BpiURWVhbnz58nKyuLlJQUrfGXGql4FRGXZ7fbjY4g4uTSOrCCggKqqqooKCggISGBlJQUo6OJiNQbHZUjdaVpwyLiMiorK42OIHJNtA5MRFyBlkhIXal4FRGX0bdvX0wmk2Ok9dL3JpPJ4GQil4uPj1exKiK3tEtLJC5tTgdaIiFXp+JVRFxGeno69957Lx4eeusTERExmo7KkbrSHZyIuIx9+/axYsUKWrRowYMPPkjPnj3x9/c3OpaIiIhL0hIJqSsVryLiMoYNG8awYcMoKyujsLCQjIwMzp49S48ePRg5cqTR8URERFyOlkhIXah4FRGX4+vrS79+/ejXrx92u509e/YYHUlEREREaqGjckTEpZlMJn784x8bHUNEREREaqHiVURERERERBo9Fa8i4pLKysqwWq288847nDlzxug4IiIiIlILrXkVEZczYcIEHnzwQT755BNsNht//etf+f3vf290LBGRevfFF19QWlpKmzZt6Ny5s9FxRERuiEZeRcTlHD16lP79+3PgwAFmzJhBWVmZ0ZFELpOXl0dERATu7u5ERESQl5dndCRpIiorK1m4cCGPP/44aWlprFmzhldeeYWf//znZGdnc/78eaMjiohcF428iojLqaqqYuPGjXTp0oWSkhJKS0uNjiTiJC8vjzFjxuDj44PdbsdqtTJmzBgAHSkhtUpNTeWJJ57gxRdfxM3t3+MUdrudwsJCUlNTmTNnjoEJRUSuj0ZeRcTljBgxgvfff5/nn3+e3Nxcxo0bZ3QkEScTJkzA3d2dpUuXUlFRwdKlS3F3d2fChAlGR5Mm4JVXXuHBBx90Klzh4u7qDz/8sApXEWmyVLyKiMspLS1lwYIF3H777YwZM4bDhw8bHUnEyZEjR1i+fDkxMTE0a9aMmJgYli9fzpEjR4yOJk3Ys88+a3QEEZEbomnDIuIyNmzYwJYtW9i5cyc7duwAwGaz8cUXXzB06FCD04mINCyt7xeRpk7Fq4i4jIceeojbbruN0tJS4uLiAHBzc6NDhw4GJxNxFhwczLPPPsvKlSuJjo6mqKiIZ599luDgYKOjSRNmMpmuqd/evXuZN28eubm5FBcXk5iYyB133AFcXHPdr18/Fi5cyNatW/Hw8CA5OZkePXpw6NAhJk6ciMlkomvXrkydOvWyqcsiIjdCxauIuIzy8nKioqIIDAx0aj937pxBiUSubM6cOYwZM4bhw4dz+PBhQkJCuHDhAvPnzzc6mjQBa9asuazNbrdTUlJS63OXLFnC+vXrad68OQD79u1j2LBhDB8+3NGnuLiYXbt2sW7dOo4fP47ZbOatt94iPT2dsWPHEhUVRWpqKps3b6Z3797198JExOWpeBURl/H6668zadIkUlNTndpNJhPLly83KJXI5eLj49m+fTtLlizBZrNx/PhxRowYoZ2G5ZqcPHnyiu0DBgyo9bkhISFkZWU5Ngf79NNPOXjwIJs3b6Zjx44kJyeze/duoqOjMZlMBAUFUV1dTUlJCcXFxdx3330A9OzZk23btql4FZF6peJVRFzGpEmTAMjNzTU4icjV5eXlkZ+fz3vvveeYNpyQkMBPf/pTFbBSq2PHjvHwww8THR2Nj49PnZ4bGxvrtDFYjx49GDhwIBERESxevJjf//73+Pn54e/v7+jj4+PD2bNnsdvtjqnJl9pEROqTFiKIiMt555136NevH48++qjjS6QxSUtLIycnx2m34ZycHNLS0oyOJk3Ab3/7W8rLy5k6dSqjR49m2bJl172reu/evYmIiHB8v2/fPnx9fbFarY4+VqsVPz8/p/WtVquVli1b3tgLERH5LypeRcTlLFmyhMWLF/Pee+85vkQaE4vFQnR0tFNbdHQ0FovFoETSlLRu3Zonn3ySefPmkZmZyQ9/+ENWrlxJYmJina+VkJDAP/7xDwA++ugjunXrxj333ENRURE2m41jx45hs9kICAggPDycnTt3AlBYWEhkZGS9vi4REU0bFhGX06FDBzp27Gh0DJEahYWFMX36dN555x0sFgthYWE8+eSThIWFGR1NmpBz587x/fff06VLF3bv3s2UKVPqfI1p06Yxc+ZMmjVrRtu2bZk5cya+vr5ERkYSFxeHzWZz7COQlJTElClTyMzMpFOnTsTGxtb3SxIRF6fiVURcjre3N8899xxhYWGO9Vnjx483OJXIv8XExJCRkUFGRgaJiYlkZ2eTlJR0XSNn4rpefvllBgwYwAcffECXLl1ITU0lJyen1ucFBwezdu1aALp168bq1asv62M2mzGbzU5toaGhrFixon7Ci4hcgaYNi4jLefjhh/mf//kfOnXqRGhoKKGhoUZHEnFSUFBAUlISS5cuxc/Pj6VLl5KUlERBQYHR0aQJ+f7773n00Uf55ptvGDlyJJWVlUZHEhG5IRp5FRGX8fe//x24OKpgMpnw8vKiS5cujvMMRRoLi8XCJ598wqxZsxxtVVVVpKenG5hKmpqqqiqWLl1Kt27d2L9/v9MmSyIiTZGKVxFxGXl5eU6Pz507x7/+9S+mTJnCI488YkwokSsICwujqKiImJgYR1tRUZHWvEqdJCUlsWnTJl544QXeffddpk2bZnQkEZEbouJVRFxGZmbmZW2lpaUkJiaqeJVGJSUlhYSEBHJycpzOedVROVIXO3bsYMKECQAMGTKE+fPn06NHD4NTiYhcPxWvIuLS/P398fC4trfCvXv3Mm/ePHJzcykuLiYxMZE77rgDgPj4ePr168fChQvZunUrHh4eJCcn06NHDw4dOsTEiRMxmUx07dqVqVOnOp2HKPLf4uPjgYub4lzabTgtLc3RLnI169at48033+TAgQMUFhYCUF1dzYULF3jppZcMTicicv1UvIqISzt37hxlZWW19luyZAnr1693rI/dt28fw4YNY/jw4Y4+xcXF7Nq1i3Xr1nH8+HHMZjNvvfUW6enpjB07lqioKFJTU9m8eTO9e/dusNckt4b4+HgVq3Jd+vfvzwMPPMAf/vAHxw7Vbm5utGnTxuBkIiI3RsWriLiM+fPnO47GAaisrGTbtm0MGTKk1ueGhISQlZXlmIL36aefcvDgQTZv3kzHjh1JTk5m9+7dREdHYzKZCAoKorq6mpKSEoqLi7nvvvsA6NmzJ9u2bVPxKiIN5vPPP6d79+706dOHgwcPOtoPHDhAdHS0gclERG6MilcRcRmdOnVyeuzl5UVcXNw1HZUTGxvLkSNHHI979OjBwIEDiYiIYPHixfz+97/Hz88Pf39/Rx8fHx/Onj2L3W53FM2X2kREGspHH31E9+7d2bhx42U/U/EqIk2ZilcRcRlPPfVUvV2rd+/etGzZ0vH9zJkzefTRR52OorBarfj5+Tmtb7VarY7niYg0hJEjRwLoaCURueVoxxARkeuQkJDAP/7xD+DiKEe3bt245557KCoqwmazcezYMWw2GwEBAYSHh7Nz504ACgsLiYyMNDK6iLiIP/zhD0RGRhIdHe34EhFpyjTyKiJyHaZNm8bMmTNp1qwZbdu2ZebMmfj6+hIZGUlcXBw2m43U1FTg4lmLU6ZMITMzk06dOhEbG2twehFxBRs3buTDDz90bDQnItLUqXgVEZexZs2aGn8WFxdX6/ODg4NZu3YtAN26dWP16tWX9TGbzZjNZqe20NBQVqxYUce0IiI3pn379nh7exsdQ0Sk3qh4FRGXcfLkSaMjiIjcNFVVVTzxxBP88Ic/BMBkMjF//nyDU4k4y8vLIy0tzXGmdUpKio4JkxqpeBURlzFq1CjH9ydOnODChQvY7XZOnDhhYCoRkYYxYsQIoyOIXFVeXh4pKSnk5OQQHR1NUVERCQkJACpg5YoapHitrq5m8uTJHDx4EHd3d9LT07Hb7UycOBGTyUTXrl2ZOnUqbm5uLFy4kK1bt+Lh4UFycjI9evRoiEgiIg7Jycns2bOH8vJyzp8/T4cOHRzTgUVEbhXHjh0zOoLIVaWlpZGTk0NMTAwAMTEx5OTkYDabVbzKFTXIbsMFBQUArF69mtGjR5Oenk56ejpjx45l1apV2O12Nm/eTHFxMbt27WLdunVkZmYyffr0hogjIuLkyy+/JD8/n+joaPLz8/Hy8jI6kshl8vLyiIiIwN3dnYiICPLy8oyOJE3MgQMHOHDgAPv37+fdd9/lww8/NDqSiBOLxXLZLtjR0dFYLBaDEklj1yAjr7169eKRRx4BLn7q17ZtW7Zu3cp9990HQM+ePdm2bRuhoaFER0djMpkICgqiurqakpISAgICGiKWiAgAPj4+mEwmzp07R0BAAFVVVUZHEnGiqXRSH1566SXH93a7neeff97ANCKXCwsLo6ioyDHyClBUVERYWJiBqaQxa7BzXj08PEhKSmLmzJnExsZit9sxmUzAxRvHs2fPUlZWhq+vr+M5l9pFRBpSt27dyMnJITAwkHHjxnHhwgWjI4k4+c+pdM2aNXNMpUtLSzM6mjQhlZWVjq9jx45x5MgRoyOJOElJSSEhIYGCggKqqqooKCggISGBlJQUo6NJI9WgGzZlZGTw8ssv88wzz1BRUeFot1qttGzZEl9fX6xWq1O7n59fQ0YSEWH8+PGUlZXh7e1NYWEhd999t9GRRJxoKp3Uh759+2IymbDb7Xh7eztG70Uai0szScxms2O34bS0NM0wkRo1SPH6zjvv8O233/L888/TvHlzTCYTERER7Ny5k6ioKAoLC7n//vsJCQlh7ty5JCQk8M0332Cz2TRlWEQa3PHjx9mwYYPjQ7V9+/Y57UQsYjRNpZP6sGXLFqMjiNQqPj5exapcswaZNtynTx/27dvHkCFDSEhIIDk5mdTUVLKysoiLi6OqqorY2FgiIiKIjIwkLi4Os9lMampqQ8QREXEyZswYysrKaNu2reNLpDHRVDq5ERMmTGDr1q1UV1c7tdtsNjZt2sTLL79sUDIRkRvTICOvLVq0YMGCBZe1r1ix4rI2s9mM2WxuiBgiIlfk4+PDuHHjjI4hUiNNpZMbMWvWLJYtW8b8+fPx8/Ojbdu2nDlzhpKSEp544gmtnRaRJqtB17yKiDRGXbt2JT8/n7CwMMdGcqGhoQanEnGmqXRyvTw9PRkxYgQjRozgq6++4vTp07Rp04aQkBCjo4mI3BAVryLiciwWi9PGNyaTieXLlxuYSESkYdxxxx3ccccdRscQEakXKl5FxOXk5uZy+vRpvv76a4KDg7VRnDRKZrOZJUuWUFFRgZeXFyNGjCArK8voWCIiIoZpsHNeRUQaq/fee49BgwaRnZ1NXFwcf/7zn42OJOLEbDaTnZ3N7NmzsVqtzJ49m+zsbO0RISK3HLPZjLe3NyaTCW9vb73PyVVp5FVEXM4bb7zB22+/jY+PD2VlZTz77LP079/f6FgiDkuWLCEjI4Px48cDOP5MTk7W6Ktcsy+++IJp06Zx9uxZnnjiCbp27ep0/FJN9u7dy7x588jNzcVisTBz5kzc3d3x9PQkIyODtm3bMmvWLD7++GN8fHwAWLRoEVVVVbz88sucP3+ewMBA0tPTad68eUO/TGnCLn1Ql5GRQWJiItnZ2SQlJQHovU6uSCOvIuJyTCaT44bL19cXLy8vgxOJOKuoqCAxMdGpLTEx0XE2sci1SEtLIz09HX9/f37xi19cUzGwZMkSJk+e7Pi3lpaWxpQpU8jNzaV3794sWbIEgOLiYv74xz+Sm5tLbm4ufn5+LFq0iMcff5xVq1YRHh7OmjVrGvT1SdP3nx/UtWjRgvHjx5ORkeH4dyby31S8iojLCQkJ4ZVXXmHTpk288sor2oFTGh0vLy+ys7Od2rKzs/VBi9RZx44dMZlMBAQEOD60u5qQkBCnIjczM5OwsDAAqqur8fLywmazcejQIVJTUxk0aBBvvvkmALt37+ahhx4CoGfPnmzfvr0BXpHcSvRBndSVilcRcTmzZ8+mQ4cObN++nQ4dOjBz5kyjI4k4GTFiBElJSWRmZnLu3DkyMzNJSkpixIgRRkeTJqRVq1asXr2a8vJy8vPzadmyZa3PiY2NxcPj36vKAgMDAfj4449ZsWIFv/71rzl37hy//OUvmTt3Ln/84x9ZtWoVn332GWVlZfj5+QEXz9M+e/Zsw7wwuWXogzqpK615FRGX8c9//pPu3buzY8cOOnbsSMeOHQHYuXMn0dHRBqcT+bdLI1/Jycm89NJLeHl5kZiYqDVgUieXNvpq3bo1n376KWlpadd1nY0bN7J48WJee+01AgICqK6uZujQoY71rPfffz+fffYZvr6+WK1WvL29sVqt11Qsi2u79EEd4LTm9b9HY0UuUfEqIi7jo48+onv37uTn51/2MxWv0thkZWWpWJUb4uvry/PPP4/JZGLTpk2YTKY6X+PPf/4za9asITc3F39/fwC++uorxo0bx5/+9CdsNhsff/wxTz31FPfccw9/+9vfGDBgAIWFhdx77731/ZLkFqMP6qSuVLyKiMsYOXIkAPfccw8DBw50tC9fvtyoSCIiDWbChAk8+OCDfPLJJ9hsNv7617/y+9///pqfX11dTVpaGu3atXMcX/KTn/yE0aNH88QTT/DMM8/QrFkz+vfvT9euXXnhhRdISkpi7dq1tG7dmvnz5zfUS5NbiD6ok7pQ8SoiLmPDhg1s2bKFnTt3smPHDgBsNhtffPEFQ4cONTidiEj9Onr0KP379+fNN98kNzeXZ5999pqeFxwczNq1awHYtWvXFfuMGDHisjXYbdu2JScn58ZCi4hchYpXEXEZDz30ELfddhulpaXExcUB4ObmRocOHQxOJiJS/6qqqti4cSNdunShpKSE0tJSoyOJiNwQFa8i4jJatWpFVFQUUVFRnDhxggsXLmC32zl27Bg/+MEPjI4nIlKvnnvuOfLz85k0aRK5ubmMHTvW6EgiIjdExauIuJzk5GT27NlDeXk55eXlhISEOKbIiYjcKvr06UOfPn0AGDNmDCdOnDA4kYjIjdE5ryLicr788kvy8/OJjo5m48aNOk9ORG5JCxYs4P777+fee++lW7duDBs2zOhIIiI35JqL1zNnzjRkDhGRm8bHxweTycS5c+cICAigqqrK6EgiIvXuww8/pLCwkCeeeIKNGzdqeYSINHm1ThvetWsXM2bMoLq6mr59+xIUFOR0xISISFPTrVs3cnJyCAwMZNy4cVy4cMHoSCIi9c7f3x9PT0+sVisdO3akvLzc6EgiIjek1uJ1wYIFrFixArPZTGJiIvHx8SpeRaRJGz9+PGVlZXh7e1NYWMjdd99tdCQRkXp3++238+abb9K8eXPmz59PWVmZ0ZFERG5IrcWrm5sb/v7+mEwmvLy88PHxuRm5REQazPHjx9mwYQMVFRUA7Nu3j1GjRhmcSkSkfv32t7+lrKyMvn378qc//YlXX33V6EgiIjek1jWvISEhzJ8/n9LSUl577TWCgoJuRi4RkQYzZswYysrKaNu2reNLRORWk5iYSPv27fH19eVXv/oVXbp0MTqSiMgNqXXkdfr06axbt457772XFi1aMHPmzJuRS0Skwfj4+DBu3DijY4iINKhWrVqxbNkyQkNDcXO7OF4RHR1tN08o9QAAIABJREFUcCoRketXY/FaVFTk+L5Dhw506NABuLiBk974RKQp69q1K/n5+YSFhWEymQAIDQ01OJWISP1q3bo1n332GZ999pmjTfdwItKU1Vi85ufn1/gkvfGJSFNmsViwWCyOxyaTieXLlxuYSESk/qWnp3Pw4EEOHz7MnXfeSWBgoNGRRERuSI3Fa3p6uuP7ffv2cfDgQbp06cKdd955U4KJiDSUhx9+mOeee87oGCIiDWrFihX89a9/5cyZMzz11FMcOnSI1NRUo2OJiFy3Wjds+t///V9mzJjB3r17mTp1Kn/84x9vRi4RkQZTWFhIdXW10TFERBpUfn4+b7zxBn5+fjz77LPs3bvX6EgiIjek1g2bCgsLefPNN3Fzc6O6upq4uDiNWIhIk3b69GkeeughgoODMZlMmEwmVq9ebXQsEZF6ZbfbARxr+z09PY2MIyJyw2otXm+//XasVit+fn5cuHBBR0qISJOXnZ1tdAQRkQb3+OOPM2TIEI4dO8aIESPo1auX0ZFERG5IrcXriRMniI2N5a677mL//v00a9aMQYMGAWikQkSaJA8PD+bOncvp06eJjY3lzjvvpH379kbHEhGpF++99x6PPfYYMTExPPDAA3zxxReEhoZy1113GR1NROSG1Fq8LliwALg45eTS9BMRkaZsypQpDBs2jEWLFhEZGcnEiRNZu3at0bFEROrFa6+9RpcuXUhJSWHOnDmOovXgwYM6FkxEmrRai1d3d3dmz57NgQMHuOOOO5g0aRLBwcE3I5uISIOoqKjggQceYPHixXTq1AkvLy+jI4mI1JuBAwcye/ZsDh48SGpqqtPaVx0LJiJNWa3F6+TJk4mPj+cnP/kJu3btIiUlhWXLlt2MbCIiDcLT05MPP/wQm83Gnj17tImJiNxSBg8ezODBg1m7di3PPPOM0XFEROpNrUflVFRU8Oijj9KyZUt69erFhQsXbkYuEZEGM3PmTN5++21Onz7N0qVLmT59utGRRC6Tl5dHREQE7u7uREREkJeXZ3QkaWJCQ0MpLCzkb3/7G7169eLdd981OpKIyA2ptXitrq7m888/B+Dzzz93bLcuItJUffjhh7z66qvk5+fzu9/9ji1bthgdScRJXl4eY8aMwWq1YrfbsVqtjBkzRgWs1MncuXO54447WL58OXl5edpoU0SavGuaNpycnMzJkycJDAxk5syZNyOXiEi927BhA1u2bGHnzp3s2LEDAJvNxhdffMHQoUMNTifybxMmTMDd3Z2lS5cSHR1NUVERgwcPZsKECcTHxxsdT5oILy8v2rRpg4eHB7fddhuVlZVGRxIRuSG1Fq/h4eG88cYbHD16lA4dOuDj43MzcomI1LuHHnqI2267jdLSUuLi4gBwc3OjQ4cOBicTcXbkyBEmTZqE2WzGYrEQFhbGsGHDSE9PNzqaNCG+vr4MGzaMwYMHs3LlStq1a2d0JBGRG1Jr8fr++++zePFiqqur6du3LyaTiRdffPFmZBMRqVetWrUiKiqKqKgoTp06RUVFBXBxeYRIY/P666+zatUqp5FXkbpYsGABhw8fpkuXLnzxxRcMHDjQ6EgiIjek1uL19ddfZ+3atSQkJPDiiy/y9NNPq3gVkSZt+vTp/O1vfyMwMBC73Y7JZNJaMGlUPDw8LpviWVlZiYdHrb+2RRxOnTpFQUEBf/nLXxxto0aNqvV5e/fuZd68eeTm5nLo0CEmTpyIyWSia9euTJ06FTc3NxYuXMjWrVvx8PAgOTmZHj161NhXRKS+1Ppb0M3NDU9PT0wmEyaTiebNm9+MXCIiDWbv3r1s2rRJN1XSaFVXV+Pm5sbw4cM5fPgwISEhuLm5aZaA1MmYMWN44IEH6jRdeMmSJaxfv95xv5eens7YsWOJiooiNTWVzZs3ExQUxK5du1i3bh3Hjx/HbDbz1ltvXbFv7969G+rliYgLqrV4jYyMZPz48Xz77bekpqbSvXv3m5FLRKTBdOzYkYqKCn0YJ41WeHg4Tz75JO+88w4APj4+DBkyxPFY5Fr4+Pgwbty4Oj0nJCSErKwsJkyYAEBxcTH33XcfAD179mTbtm2EhoYSHR2NyWQiKCiI6upqSkpKrthXxauI1Kdai9fx48dTWFhIeHg4nTt3JiYm5mbkEhFpMMePHycmJoaOHTsCaNqwNDopKSmkpKSQk5PjWPOakJBAWlqa0dGkCenatSv5+fmEhYU5jjoMDQ296nNiY2M5cuSI4/GlpRVwsRg+e/YsZWVl+Pv7O/pcar9SXxGR+lRr8Xrq1CkKCws5ePAgp06d4p577qFVq1Y3I5uISIOYP3++0RFErio+Pp7t27fz2GOPUVFRgZeXFyNGjNAxOVInFosFi8XieGwymVi+fHmdrvGfyyusVistW7bE19cXq9Xq1O7n53fFviIi9anW4nXs2LH069ePX/ziF+zevZsJEybwhz/84WZkExGpV2vWrKnxZ5eOzhFpDPLy8sjPz+e9995zGnn96U9/qgJWrllubq7T4+s55zU8PJydO3cSFRVFYWEh999/PyEhIcydO5eEhAS++eYbbDYbAQEBV+wrIlKfrmnbwku/KO+66y6nHetERJqSkydPGh1B5JqkpaWRk5PjWKoTExNDTk4OZrNZxatcs9WrV/P6669z4cIF7HY7zZo14/3336/TNZKSkpgyZQqZmZl06tSJ2NhY3N3diYyMJC4uDpvNRmpqao19RUTqU63Fa6dOnVi/fj1RUVEUFxfj7+/PwYMHgdrXTYiINCYJCQlX3aSpvLxcmzhJo2CxWIiOjnZqi46OdpoCKlKbtWvXkpuby+LFi+nbty/Lli27pucFBwezdu1a4OK93ooVKy7rYzabMZvNTm019RW5mry8PNLS0rBYLISFhZGSkqIP6aRGtRavX375JV9++SXr1q1ztKWmpl7XugkRESPNmDGDiIgI+vXrR+vWrR3tJSUlrF+/HovFQkZGhoEJRS4KCwtj+vTpvPPOO44buieffJKwsDCjo0kT0rp1awIDA7FarURFRfG73/3O6EgiTvLy8q64OR2gAlauqNbi9b/XS4iINFXp6els3LiR3/zmN3zzzTf4+/tjtVq57bbbGDx4ML/+9a+NjigCXJwmnJGRQUZGBomJiWRnZ5OUlERiYqLR0aQJ8fPzY9OmTY4d1UtKSoyOJOJESySkrq5pzauIyK2iX79+9OvXj4qKCs6cOYO/vz+enp5GxxJxUlBQQFJSEkuXLuW3v/0tYWFhJCUl6ZxXqZNZs2Zx+PBhXnrpJZYuXcq0adOMjiTiREskpK7cavqBzuYSkVuZl5cXgYGBdSpc9+7dy69+9SsADh06RHx8PIMHD2bq1KnYbDYAFi5cyC9+8QsGDRrEP/7xj6v2FamJxWJh6tSpfPrpp1RXV/Ppp58ydepU3dBJnYwePZrw8HACAwOZOHEiUVFRRkcScRIWFkZRUZFTW1FRkZZISI1qLF4vTU2aOnXqTQsjItJYLVmyhMmTJ1NRUQFcnII8duxYVq1ahd1uZ/PmzRQXF7Nr1y7WrVtHZmYm06dPr7GvyNXohk7qw6VpwwcOHODgwYOODTdFGouUlBQSEhIoKCigqqqKgoICEhISSElJMTqaNFI1Thv29vbm6aef5tChQ3z++ecA2O12x7oJEZGm7syZM7Rq1eqa+oaEhJCVlcWECRMAKC4u5r777gOgZ8+ebNu2jdDQUKKjozGZTAQFBVFdXU1JSckV+/bu3bthXpTcEi7d0P33JiZpaWlGR5MmpKSkxGmHYW22KY3NpXWtZrPZsTldWlqa1rtKjWosXpcsWcKJEydITU1l2rRp2O32m5lLRKTB7Nq1ixkzZlBdXU3fvn0JCgpi4MCBV31ObGwsR44ccTy+9GEegI+PD2fPnqWsrAx/f39Hn0vtV+orcjW6oZP6MHz4cMdGOAAbN240MI3IlcXHx+u9Ta5ZjcWrm5sbt99+O4sWLWLNmjXs37+fO+64Q/+4RKTJW7BgAStWrMBsNpOYmEh8fHytxet/c3P796oLq9VKy5Yt8fX1xWq1OrX7+fldsa9IbXRDJ9eroKCAjz/+mPz8fD755BMAbDYbmzdvpl+/fganExG5fjWueb0kNTWVw4cP8+CDD3L06FEmT558M3KJiDQYNzc3/P39MZlMeHl54ePjU+drhIeHs3PnTgAKCwuJjIzknnvuoaioCJvNxrFjx7DZbAQEBFyxr0ht8vLyiIiIwN3dnYiICPLy8oyOJE3EXXfdRadOnfDy8iI0NJTQ0FC6dOlCZmam0dFERG5IrcXroUOHmDRpEr169SI5OZnDhw/fjFwiIg0mJCSE+fPnU1paymuvvUZQUFCdr5GUlERWVhZxcXFUVVURGxtLREQEkZGRxMXFYTabSU1NrbGvyNXk5eWRkpJCVlYW58+fJysri5SUFBWwck3atWvHU089RX5+PhEREXh7e3PnnXdqwy9plMxmM97e3phMJry9vTGbzUZHkkas1nNeKyoqKC8vp3nz5pw/f57q6uqbkUtEpMFMnz6ddevWce+999KiRQtmzpx5Tc8LDg5m7dq1AISGhrJixYrL+pjN5st+8dbUV6QmaWlpDB482GnN6+DBg7XuVepk5cqVvPvuu9x9993k5OTw2GOPkZCQYHQsEQez2Ux2djYZGRkkJiaSnZ1NUlISAFlZWQank8ao1uJ16NCh9O/fn65du7J//35Gjx591f5VVVUkJydz9OhRKisreeGFF+jSpQsTJ07EZDLRtWtXpk6dipubGwsXLmTr1q14eHiQnJxMjx496u2FiYjUpLKykpiYGHr16sXatWs5efIk7du3NzqWiMO+ffs4d+7cZbsNf/XVV0ZHkyZkw4YNrFq1Cg8PD6qqqhg0aJCKV2lUlixZQkZGBuPHjwdw/JmcnKziVa6o1uL15z//OT179uTrr78mODiY1q1bX7X/+vXr8ff3Z+7cuZw+fZqnnnqKu+66i7FjxxIVFUVqaiqbN28mKCjIcR7i8ePHMZvNvPXWW/X2wkREavLyyy/z9NNP8/7779OlSxdSU1PJyckxOpaIg6enJ6NGjXLsFBsTE8OoUaNITk42OJk0JXa7HQ+Pi7d6zZo1o1mzZgYnEnFWUVFBYmKiU1tiYiIvvfSSQYmksat1zSuAv78/3bt3r7VwBejbty9jxoxxPHZ3d7/sjMPt27eze/fuK56HKCLS0L7//nt+9rOf8c033zBy5EgqKyuNjiTipLKykqysLAoKCqiqqqKgoICsrCz9W5U6uffeexk9ejTLli1j9OjR/PjHPzY6kogTLy8vsrOzndqys7Px8vIyKJE0drWOvNbVpV07y8rKGD16NGPHjiUjI+Oaz0MMCAio70giIk6qqqpYunQp3bp1Y//+/U7H24g0BuHh4XTt2pXHHnuMiooKvLy8eOyxx65rZ2xxXUlJSWzdupUDBw4wYMAAHnnkEaMjiTgZMWKEY43rf655/e/RWJFLah15/eCDD5z+vBbHjx93rJV94okn6nQeoohIQ5swYQKnTp3ihRdeYOfOnUybNs3oSCJOYmJi2LBhA7Nnz8ZqtTJ79mw2bNjgmEYsci3Kyso4d+4cbdq0obS0lHfeecfoSCJOsrKySExMJDk5GR8fH5KTk0lMTNR6V6lRjSOvw4YNw8fHhwMHDtCmTRuWL19Onz59ar3gd999x/Dhw0lNTeWBBx4A/n0eYlRUFIWFhdx///2EhIQwd+5cEhIS+OabbxznIYqINLR7772XDh06UFZWxiOPPMKJEyeMjiTipKCggMcff5zk5GReeuklvLy8ePzxxykoKDA6mjQhL774IoGBgbRr1w7AMQtOpDHJyspSsSrXrMbi9fXXX6e8vJyBAweyY8cO9u/fz/DhwwkODmbGjBk1XjA7O5vvv/+eRYsWsWjRIgBSUlKYNWsWmZmZdOrUidjYWNzd3R3nIdpsNsd5iCIiDS05OZk9e/ZQXl5OeXk5ISEhjiNwRBqDS7sNv/fee9ptWK6b3W5n3rx5RscQEak3NRavycnJdO/eHV9fXxITE9mxYwdLly7l6NGjV73g5MmTmTx58mXt13oeoohIQ/vyyy/Jz88nNTWVcePGOW0yJ9IYaLdhqQ933nkne/fuJSwszNHm6elpYCIRkRtTY/E6fvx49uzZw9GjR3nxxRfZv38/r776KuHh4ToPUUSaNB8fH0wmE+fOnSMgIICqqiqjI4k4ubTb8I9//GPHyKt2G5a62rVrF1u2bHE8NplMbN682cBEIiI3psbitW3btvTq1Yt3332XBQsWMHr0aHr37k1xcfHNzCciUu+6detGTk4OgYGBjBs3jgsXLhgdScRJeHg4Tz75JGazGYvFQlhYGEOGDNGGO1In69evNzqCiEi9qvWonAULFgDwu9/9DoCIiIiGTSQi0sBGjx7N+fPn8fb2prCwkO7duxsdScRJSkoKY8aMcRyNY7Vaee211xy/k0WuZsaMGaSmphIXF3fZJk2rV682KJXIleXl5ZGWlub4oC4lJYX4+HijY0kjVe/nvIqINFYnT56krKyMpKQk5syZg91up2PHjrzwwgu8+eabRscTcVJRUUFpaSk2m42jR4/SvHlzoyNJE/Hiiy8CkJmZaXASkavLy8u77IO6S/tQqICVK1HxKiIuY+/evSxbtoyDBw8yZcoUANzc3IiOjjY4mYizCRMmYDKZaN++PYcPH6Z9+/acPn2aCRMm6IZOatW2bVsATp06RX5+PhUVFY6f6VxraUwmTJiAh4cHS5cudazvHzJkiN7rpEYqXkXEZfTq1YtevXrxt7/9jYcfftjoOCI1OnLkCLfffrvTDd3gwYM5cuSI0dGkCUlKSmLEiBG0bNnS6CgiV3TkyBE++OADp53Vly1bRp8+fQxOJo2VilcRcTk/+MEPePrpp/n2229p27Yts2fPJjw83OhYIk7Gjx/vdEM3fvx4JkyYYHAqaUo6duzIgAEDjI4hIlJvVLyKiMtJS0sjLS2Nu+66C4vFwvTp07WJiTQ6mZmZREZGOkZetX5R6io2NpZx48bRuXNnR9uoUaPqfJ23336bP/3pT8DFtdgWi4X58+czZ84c2rVrB4DZbCYyMpJp06bx+eef4+npyaxZs+jYsWP9vBi5JQUHBzN06FBWrVrleK8bOnQowcHBRkeTRkrFq4i4HLvdzl133QVAWFgYHh56K5TGJTg4mLNnzzJ8+HAOHz5MSEgI5eXluqGTOlm1ahW9e/e+4WnDAwYMcIzgTp8+naeffpri4mJ++9vfEhsb6+j3wQcfUFlZyZo1a9izZw+vvPIKixcvvqG/W25tc+bMYcyYMQwfPpxDhw7RsWNHqqur9WGd1Eh3bCLicjw8PCgoKCAyMpK///3veHp6Gh1JxMmlGzq4+GELgKenJ3PmzDEyljQxrVq1YuTIkfV2vX/+85/s37+fqVOn8txzz2GxWFi2bBk9evTg5ZdfZvfu3Tz00EMA/OhHP+LTTz+tt79bbk2XNmVKS0vDZDLh4+PD7NmztVmT1EjFq4i4nLS0NDIyMpg/fz6dO3dm5syZRkcScRIfH88bb7zBX//6V+x2O4cOHaJ37966oZM6ad26NampqYSHhzvOe42Li7vu6/3hD3/gN7/5DQAPPvggvXr1Ijg4mKlTp7J69WrKysrw9fV19Hd3d+fChQua3SJXFR8fr/c2uWZ6NxERl9O+fXumTZtGeXm544ZOpDExm81s2bKFefPmkZiYSHZ2NklJSZjNZrKysoyOJ03EpfWm33333Q1f6/vvv+fLL7/k/vvvB+Dpp592TEd+9NFHef/99/Hz88NqtTqeY7PZVLiKSL3SO4qIuJwpU6awY8cO2rRpg91ux2QyacMmaVSWLFlCRkYG48ePB3D8mZycrOJVrtl/bs60ZcsWVq5ced3X+vvf/85Pf/pT4OJU9p///OesXr2a22+/nY8++ohu3brRtm1bCgoK6NevH3v27OGHP/zhDb8GEZH/pOJVRFzO559/zgcffKBRV2m0KioqSExMdGpLTEzkpZdeMiiRNEWlpaWsW7eOtWvXEhISwi9+8YvrvtbBgwcdG4aZTCZmzZrFqFGj8Pb2pnPnzjzzzDO4u7uzbds2Bg0ahN1uZ/bs2fX1UkREABWvIuKCAgMDsVqtTmuzRBoTLy8vRo4cyZ49e7BYLISFhfGjH/0ILy8vo6NJE/Dpp5+ycuVKPv74Yx577DFuv/12cnJybuiazz33nNPj6OhooqOjL+s3Y8aMG/p7RESuRsWriLiMuLg4TCYTp06dok+fPnTo0AFA04al0Xn44YdZuXIlL7zwAtu2bWPSpEksXryYPn36GB1NmoBBgwaRkJDAu+++i6enJyNGjDA6kohIvVDxKiIuQ+fGSVNx9OhRnnzySZYuXcrixYvx8vLiySef5F//+pfR0aQJWLlyJW+++SaPP/44vXv35ty5c0ZHEhGpFypeRcRltG/fHri4dmvOnDl89dVXdO3alaSkJIOTiTizWCy0a9eOyspKACorKykrK8NisRicTJqCu+++m7vvvptz586Rn5/P//3f/zFw4ED69+/PL3/5S6PjiYhcNzejA4iI3GxJSUkMGjSItWvXMmDAACZOnGh0JBEnzZs3Z9OmTSQmJlJaWkpiYiKbNm2iefPmRkeTJqRFixYMHDiQNWvWMGvWLL766iujI4mI3BAVryLicpo3b87DDz+Mn58fjzzyCG5ueiuUxsVqteLn58fAgQMdBch/n6EpUhd33nknkydPNjqGiMgN0R2biLicdu3asWjRIj7++GNyc3Px9PSkqKiIoqIio6OJOLz66quYzWa8vb0xm828+uqrRkcSERExlNa8iojLMZlMfP3113z99dcAtG3blvz8fIArHv0gcrNdaQfs1atX62xiuS5nzpyhVatWRscQEblhKl5FxOWkp6dTXV2N3W5nz5499OjRA09PT6NjiThERESwadMmfv7zn7N161YSEhJYv3493bt3NzqaNCG7du1ixowZVFdX07dvX4KCghg4cKDRsURErpuKVxFxOXPnzqVDhw4cO3aM4uJibrvtNl555RWjY4k42Gw2IiMjeffdd7ntttswmUxERkZSXl5udDRpQhYsWMCKFSswm80kJiYSHx+v4lVEmjSteRURl7N7924GDRrEJ598Qk5ODsePHzc6kogTi8XC6NGjCQ8Px83NjfDwcEaPHq2jcqRO3Nzc8Pf3x2Qy4eXlhY+Pj9GRRC6Tl5dHREQE7u7uREREkJeXZ3QkacQ08ioiLsdms/GPf/yD4OBgKisrKSkpMTqSiJOgoCCef/55Lly4gM1m44svvuD5558nKCjI6GjShISEhDB//nxKS0t57bXX9O9HGp28vDxSUlLIyckhOjqaoqIiEhISAIiPjzc4nTRGGnkVEZfTv39/Zs6cyfDhw5k7dy5Dhw41OpKIk9OnT1NeXo6vry8Avr6+lJeXc/r0aYOTSVMyffp0goKCuPfee2nRogUzZ840OpKIk7S0NHJycoiJiaFZs2bExMSQk5NDWlqa0dGkkdLIq4i4nCFDhtCvXz++/vprXnjhBQICAoyOJOLEarXi6elJWVkZAGVlZXh6euqcV6kTd3d3unXrRpcuXQDYu3cvP/nJTwxOJfJvFovlsl3+o6OjtURCaqTiVURczsaNG1mwYAGdO3fmX//6F6NGjaJ///5GxxJx4u7uzg9+8AMOHz5M+/bt+fbbb42OJE3MqFGjOH36NO3atcNut2MymVS8SqMSFhZGUVERMTExjraioiLCwsIMTCWNmYpXEXE5y5Yt4+2338bHx4eysjKeffZZFa/S6JSXl1NeXo7dbnd8L1IXp06duuy8YJHGJCUlhYSEhMvWvGrasNRExauIuByTyeTYddPX1xcvLy+DE4lc2cmTJ7Hb7Zw8edLoKNIEhYaG8u233/KDH/zA6CgiV3RpUyaz2YzFYiEsLIy0tDRt1iQ1UvEqIi4nJCSEV155hcjISP7v//6PkJAQoyOJXJGbmxs2m83xp0hd7N69m5iYGKd1/UVFRQYmErlcfHy8ilW5ZipeRcTlzJ49mzVr1rB9+3Y6d+7Myy+/bHQkkcu0aNGCwMBADh8+THBwMCdOnODcuXNGx5Im5IMPPjA6gohIvVLxKiIup7y8nDZt2jimDm/YsIEnn3zS4FQizux2O0ePHsVms3H06FE8PPQrW67NokWLePHFFxk/fjwmk8npZ/PnzzcolYjIjbv1fxOePw/e3k33+iJS737zm9/Qvn172rZtC3DZzZ2I0QICAigpKXEUrJc2bdKxTnItfvaznwEwaNAgg5OIiNSvW7949faGhrwxtdsb7toi0iDsdjvp6elGxxCp1YULF5z+FLkWd911FwDh4eEUFhZSWVlpcCIRkfpx6xevIiL/36UbuA4dOvDJJ5/QrVs3x888PT2NiiVymZKSEgBat25NaWkp/v7+nD592tEuci1efPFFAgMDadeuHaBZJiLS9Kl4FRGX0bdvX0wmE3a7nR07dji+N5lMbN682eh4Ik5+9KMfUVVVxZkzZwgKCqJjx47s2bPH6FjShNjtdubNm2d0DBGReqPiVURcxpYtW4yOIHLN9uzZg4eHBzabjc8//1xTh6XO7rzzTvbu3UtYWJijTbNMRKQpU/EqIi7jL3/5C+np6Xh7ezN37lx69OhhdCSRq7p0tqvOeJXrsWvXLqcP7TTLRESaOjejA4iI3CxvvPEG69ev/3/t3XtcVXW+//HXZguiXFROmhfEwWloKDAlH5ZJdbTMyspLgsIjzMRo7IjdfGgDaqjheMk6HU0s00gcEawZR3McnUGTJEPHkwlq6YPJC+SFM2AKItvNXr8/+rknBgFRcG3h/Xw89mOzvov93e+F7C2fvdb3++XDDz9k6dKlZscRqZeKV7keGzZUvPm0AAAgAElEQVRsYNu2bc6bCldxRenp6YSEhGC1WgkJCSE9Pd3sSOLCdOZVRFoMDw8P2rVrR7t27aioqDA7johIk8rKymLNmjVcunQJwzA4e/YsGzduvKa+hg8fjo+PDwD+/v6MHj2a5ORkrFYr4eHhTJo0CYfDQVJSEt999x0eHh68+eab9OjRozEPSZqZ9PR0EhMTWbFiBeHh4ezcuZPY2FgAoqKiTE4nrkjFq4i0SIaWuRIXd3lCsdq2Rerz3nvvMWPGDNauXcs999xDTk7ONfVTWVkJQFpamrNt2LBhLF68mO7duxMXF8eBAwcoKirCZrORkZHBvn37mDdvHikpKY1yLNI8JScns2LFCgYOHAjAwIEDWbFiBfHx8Spe5YpUvIpIi3HixAnefvttDMNwfn3Zq6++amIykZr+vVBV4SoN1aFDB/r06cPatWsZOXIkf/jDH66pn2+//ZaKigrGjx+P3W4nPj4em81GQEAAAOHh4ezatYvi4mLuv/9+4KfZsvPz8xvtWKR5OnToEIWFhYSEhHDo0CGCg4OZNm0ahw4dMjuauCgVryLSYkyePPmKX4uINEfu7u7s2bMHu93OF198QXFx8TX14+npSWxsLBERERw9epTnn38eX19f534vLy9OnDhBWVkZ3t7eznar1YrdbqdVK/25KVfWtWtXpk6dypo1a5yXDUdHR9O1a1ezo4mL0ruJiLQYI0aMMDuCSIO4ubnhcDic9yINMWvWLP7xj38wceJE3n333Wv+0C4wMJAePXpgsVgIDAzEx8eHs2fPOveXl5fj6+vLxYsXKS8vd7Y7HA4VrlIvi8VS57bIz2m2YRERERel2YblWnz//fd8//33XLhwgc6dO2O1WnnllVe44447rqm/Tz75hHnz5gFw+vRpKioqaNu2LcePH8cwDHbu3Enfvn0JCwsjOzsb+Gmd4qCgoEY7JmmefvjhB+bPn098fDyenp7Ex8czf/58fvjhB7OjiYvSx2EiIiIuytvb23kpZllZmdlx5CYxc+ZM59eXz2KVlpZy9OhR8vLyGtzfqFGj+O1vf0tUVBQWi4W5c+fi5ubGlClTqKqqIjw8nLvuuovQ0FBycnIYM2YMhmEwd+7cRjsmaZ6Cg4Px9/evNj56+/btBAcHm5hKXJmKVxFpccrKyli+fDnFxcX853/+J7fffruWcxCXdLlgVeEqDfHzWYH379/P6tWrKSgoYNSoUdfUn4eHB4sWLarRnpmZWW3bzc2N2bNnX9NzSMuUmJhIbGxsjaVykpOTzY4mLkrFq4hcE6OiAksTzX5qVFRgadOmSfoGSEhI4IEHHmDPnj3ccsstJCYmsnr16iZ7PhGRG8lms7Fp0ybWrFmDu7s7ZWVlZGVl4enpaXY0kWqioqL48ssveeyxx6isrKR169Y8//zzWiZHaqXiVUSuiaVNGyyzmmZSBeONpl0S5OzZs4waNYoNGzYQFhZ2XUuQDB8+HB8fHwD8/f0ZPXo0ycnJWK1WwsPDmTRpEg6Hg6SkJL777js8PDx48803daZXRJrMoEGDeOKJJ1i4cCG/+MUvmDBhggpXcUnp6els2rSJzZs3Vzvzet9996mAlStS8SoiLVJBQQEAp06dws3t2uauq6ysBKpfojds2DAWL15M9+7diYuL48CBAxQVFWGz2cjIyGDfvn3MmzePlJSU6z8IEZErGDt2LJ999hlFRUWMGjVKawSLy0pOTiY6Opr4+HjnOq/R0dEkJyereJUrUvEqIi3O9OnTSUhIoKCggMmTJ/PGG29cUz/ffvstFRUVjB8/HrvdTnx8PDabjYCAAADCw8PZtWsXxcXF3H///QD07t272sQUIiKNLS4ujri4OHbv3s26devIz89n4cKFDBs2TDMAi0s5ePAgFy5cqDHm9ejRo2ZHExel4lVEWpygoCAyMjKuux9PT09iY2OJiIjg6NGjPP/88/j6+jr3e3l5ceLECedssZdZrVbsdrvWP5R6ubu7c+nSJee9SEP069ePfv36ce7cOf70pz8xdepU1q9fb3YsEScPDw+6du1abcxr3759tVSO1Ep/OYlIi7N+/Xo++OAD52W/AFlZWQ3uJzAwkB49emCxWAgMDMTHx4ezZ88695eXl+Pr68vFixcpLy93tjscDhWuclW0zqs0Bl9fX2JiYoiJiTE7ikg1lZWV5OTkOIfvXLp0iZycHJNTiSu7toFeIiI3seXLl5OSksLmzZudt2vxySefMG/ePABOnz5NRUUFbdu25fjx4xiGwc6dO+nbty9hYWFkZ2cDsG/fPl22JwCEhIRgsViueLusqqqq2j1Q62NCQkJu+DGIiDSGTp064ebmRqdOncyOIi5OH/2LSIvTvXv3Rpntd9SoUfz2t78lKioKi8XC3LlzcXNzY8qUKVRVVREeHs5dd91FaGgoOTk5jBkzBsMwmDt3biMchdzs6hr7fLmAtVqtVFVVOe8BTb4jIs2Kl5cXnp6eGIaBp6cnXl5e1a5WEvm5Jitev/nmG9566y3S0tI4duwYr7/+OhaLhV/96le88cYbuLm5sWTJEj7//HNatWpFQkICvXr1aqo4IiJOnp6eTJgwgeDgYGeR8Oqrrza4Hw8PDxYtWlSjPTMzs9q2m5sbs2fPvraw0iKFhoaSl5fnLFQv34eGhpoZS0Sk0VVWVjonaDp69KiG1UidmuSy4eXLlzN9+nTneLLf/e53vPzyy6xZswbDMMjKyuLAgQPOWfDefvttZs2a1RRRRERqePDBBxk6dCg9e/YkMDCQwMBAsyOJVLN//35CQ0OrjXkNDQ1l//79JicTEWlcdrudp556iuLiYp566insdrvZkcSFNclHGwEBASxevJipU6cCcODAAfr16wfAAw88QE5ODoGBgYSHh2OxWOjatStVVVWUlJTg5+fXFJFERMjLyyM0NJSOHTuaHUWkXpcLVYvFokuFRaRZ27RpEx07dsRqtZodRVxckxSvQ4YMobCw0LltGIbz0jwvLy/Onz9PWVkZ7du3d37P5XYVryLSVHbt2kVoaCibNm2qsS88PNyERCIiIi3bE088wV//+leqqqpo1aoVjz32GJ999pnZscRF3ZCLyi9Pfw3/WjrC29u72mDs8vJyfHx8bkQcEWmhxo0bh81m0zAFERERF9CqVSu2b99Oly5dOH78OF26dGH79u0a9yq1uiG/GXfccQe5ubncc889ZGdnc++99xIQEMDChQuJjY3l1KlTOBwOnXUVkSb16KOPVluGBP51Zci1rPMqIiIi127QoEFs3boVDw8PHA4HP/74I+Xl5TzyyCNmRxMXdUOK12nTpjFjxgzefvttevbsyZAhQ7BarfTt25fRo0fjcDiYOXPmjYgiIi3Ytm3bzI4gIiIi/19RURHDhw93rrd+4cIFhg8fzpEjR0xOJq6qyYpXf39/53IRgYGBrF69usb3xMfHEx8f31QRRESuKCsrizVr1nDp0iUMw+Ds2bNs3LjR7FgiIiItyqFDh/j6669xd3d3tl26dAlPT08TU4kra5KlckREXNl7773HpEmT6NKlCyNGjCAoKMjsSCIiIi1OcHAwkZGReHp6YrFY8PT0JDIykuDgYLOjiYtS8SoiLU6HDh3o06cPACNHjuT06dMmJxIREWl5unXrxvr16xk/fjxnz55l/PjxrF+/nm7dupkdTVyUpvISkRbH3d2dPXv2YLfb+eKLLyguLjY7koiISIuzY8cOBgwYwMqVK0lJSaF169YMGDCAHTt2mB1NXJTOvIpIizNr1izsdjsTJ04kMzOTl156yexIIiIiLU5lZSWFhYVs3rwZm83G5s2bKSwspLKy0uxo4qJUvIpIi/Pjjz/Sv39/brvtNt59910KCwvNjiQiItLiWCwWHn/8cQYOHIi7uzsDBw7k8ccfr7GsnchlKl5FpMVJSEjgxIkTFBYW8swzz1BUVGR2JBERkRbHMAzef/99OnfujJubG507d+b999/HMAyzo4mL0phXEWlx3n77bV599VUuXrxIQkIC/fv3NzuSiIhIi+Pv709JSQklJSUYhkFJSQmenp74+fmZHU1clM68ikiLkZGRQUZGBrt27SIsLAy73c7x48fJyMgwO5qIiEiL5Ovry5YtW7DZbGzZsgVfX1+zI4kL05lXEWkxfj6rsI+PD0OHDtVMwyIiIib54YcfSE1NJT4+nkOHDhEcHMyCBQsYN26c2dHERal4FZEWY+jQoWZHEBERkf8vODgYf39/8vPznW3bt28nODjYxFTiylS8ikiLMXPmTCwWC4ZhOGcyLC0t5ejRo+Tl5ZmcTkTENV26dImEhASKioqw2WxMnDiRzp0785vf/IZf/OIXAERFRfH444+zZMkSPv/8c1q1akVCQgK9evUyN7y4tMTERGJjY1mxYgXh4eHs3LmT2NhYkpOTzY4mLkrFq4i0GGlpac6v9+/fz+rVqykoKGDUqFEmphIRcW0bNmygffv2LFy4kNLSUkaMGMF//dd/8dxzzzF+/Hjn9x04cIDdu3ezbt06Tp48SXx8PJ9++qmJycXVRUVFAVS7bDg5OdnZLvLvVLyKSIths9nYtGkTa9aswd3dnbKyMrKysvD09DQ7moiIy3r00UcZMmSIc9tqtZKfn8/3339PVlYWPXr0ICEhgb179xIeHo7FYqFr165UVVVRUlKimWOlTqmpqRw8eBDDMDh48CCpqakqXqVWmm1YRFqMQYMG8d1337Fw4ULWrFlDp06dVLiKiNTDy8sLb29vysrKmDx5Mi+//DK9evVi6tSp/P73v6d79+689957lJWV4e3tXe1x58+fNzG5uLohQ4awdetW2rdvj8VioX379mzdurXahyUiP6fiVURajLFjx/Lll1+yaNEiduzYoUXQpcn4+flhsVga5QY0Wl8Wi0VnweSanDx5krFjxzJs2DCefPJJBg8eTEhICACDBw/m4MGDeHt7U15e7nxMeXk5Pj4+ZkWWm8DWrVvx9PSkXbt2ALRr1w5PT0+2bt1qcjJxVbpsWERajLi4OOLi4pxjsvLz81m4cCHDhg0jKCjI7HjSjLg/6E7I8BDn9ton1gIw5rMxzraJd03kxd4vMihzEMUVPy3ZFOwXTOaTmSR9mcSnR/41VjArIouD/zxI/LZ4Z9vM/jOJCIog9ONQZ9uD/g+y5KElTMqaxI7CHc72vGfzWHd4HbN3zebM+jONf8DSrP3f//0f48ePZ+bMmfTv3x+A2NhYZsyYQa9evdi1axd33nknYWFhLFy4kNjYWE6dOoXD4dCHJVIvDw8PAOeHdR4eHly8eNHMSOLCVLyKSIvTr18/+vXrx7lz5/jTn/7E1KlTWb9+vdmxpBk5s/4Mp/94ukZ73rM1Z7XeFrmtRlvSfUkk3ZdUra1T205XfPyV2pY8tKRGW0RQBBFBEVjGWeqKLlLDsmXLOHfuHEuXLmXp0qUAvP7668ydOxd3d3duueUW5syZg7e3N3379mX06NE4HA5mzpxpcnK5GVwuVC9fDaXCVeqi4lVEWixfX19iYmKIiYkxO4qIiMuaPn0606dPr9G+du3aGm3x8fHEx8fXaBepjc1m4+jRowDOe5HaaMyriIiIiIiIuDwVryIiIiIiIuLyVLyKiIiIiIiIy1PxKiIiIiIiIi5PxauIiIiIiIi4PBWvIiIiIiIi4vJUvIqIiIiIiIjLU/EqIiIiIiIiLq+V2QFERESam7yJXpDUzuwYV5Q30cvsCCIiItek2RevRkUFFsNo2v7btGmy/kVE5OYTmlKO0YT/91yPUIsFY6nZKURERBqu2RevljZtsMyyNFn/xhuu+ceJiIiIiIhIc6IxryIiIiIiIuLyVLyKiIiIiIiIy1PxKiIi0gQsFotL3jp06GD2j0ZEWpCQkJBa34/qUttjQkJCblBycUXNfsyriIjIjdaYkzVZLBaXnfxJRKQ++fn5te6rq4DV+55cic68ioiIiIjIDffII480qF1ExauIiIiIiNxwW7Zs4ZFHHnGegbVYLDzyyCNs2bLF5GTiqnTZsIiIiIiImOJyoWqxWHA4HCanEVen4lVERERERBps6b6lpHyT4txe+8RaAMZ8NsbZNvGuibzY+0UGZQ6iuKIYgGC/YDKfzCTpyyQ+PfIpACGpIZy5cIaD/zxI/LZ45+Nn9p9JRFAEoR+HOtse9H+QJQ8tYVLWJHYU7nC25z2bx7rD65i9a7bzeaV5UfEqIiIiIiIN9mLvF69YIOY9m1ejbVvkthptSfclkXRfEvDTmddOz3aiU9tOV3z8ldqWPLSkRltEUAQRQRFXE19uQipeRURERESkwfz8/CgtLW20/upbPqchOnToQElJSaP1J65BxauIiIiIiDRYaWmpyy5p05iFsLgOzTYsIiIiIiIiLk/Fq4iIiIiIiLg8XTYsIiIiIiINljfRC5LamR3jivImepkdQZqAilcREREREWmw0JRylx3zGmqxYCw1O4U0NhWvIiIiItIoHA4HSUlJfPfdd3h4ePDmm2/So0cPs2NJE3LViZE6dOhgdgRpAhrzKiIiYoKQkBAsFku9N+Cqvi8kJMTkIxKBv/3tb9hsNjIyMnjttdeYN2+e2ZGkCRmGUe/tzjvvbNTnvPPOO6/qebVMTvOkM68iIiImyM/PNzuCSKPbu3cv999/PwC9e/fW77nod0Aalc68ioiIiEijKCsrw9vb27lttVqx2+0mJhKR5kTFq4iIiIg0Cm9vb8rLy53bDoeDVq10oZ+INA4VryIiIiLSKMLCwsjOzgZg3759BAUFmZxIRJoTfRQmIiIiIo1i8ODB5OTkMGbMGAzDYO7cuWZHEpFmRMWriMgNoOUjRKQlcHNzY/bs2WbHEJFmSpcNi4jcAFo+QkREROT6qHgVEbkBtHyEiIiIyPVR8SoicgNo+QgRERGR66PiVUTkBtDyESIiIiLXx/Ti1eFwMHPmTEaPHk1MTAzHjh0zO5KISKPT8hEiIiIi18f0j/1/PonJvn37mDdvHikpKWbHEhFpVFo+QkREROT6mF68ahITEWkJtHyEiIiIyPUxvXitbRKT2saCFRUVMXLkyAY9xwhGXFfGujQ0i0hz0lSvrYa+roqKipokh5mu5b1ORJq/5vZ+p/c6EbmS2t7rTC9eGzqJSW5u7o2IJSJiKr3XiUhLoPc6EWkI0yds0iQmIiIiIiIiUh+LYRiGmQEcDgdJSUkcPnzYOYnJL3/5SzMjiYiIiIiIiIsxvXgVERERERERqY/plw2LiIiIiIiI1EfFq4iIiIiIiLg8Fa8iIiIiIiLi8ppF8epwOJg5cyajR48mJiaGY8eOVdufmZnJyJEjiYyMZPv27SalrF19+VNTU4mIiCAiIoIlS5aYlLJ29eW//D0TJkwgPT3dhIS1qy/7jh07iIyMJDIykqSkJFxtiHh9+VesWMHIkSN5+umn+etf/2pSyvp98803xMTE1Gjftm0bTz/9NKNHjyYzM9OEZCLmq+31ISLSnOi9Tq6G6eu8Noa//e1v2Gw2MjIy2LdvH/PmzSMlJQWA4uJi0tLS+PTTT6msrCQ6OpoBAwbg4eFhcup/qSv/iRMn2LBhA+vWrcNisRAdHc3DDz/Mr3/9a5NT/0td+S/77//+b3788UeTEtauruxlZWUsXLiQVatW4efnx/LlyyktLcXPz8/k1P9SV/5z586RlpbG1q1bqaioYPjw4QwePNjkxDUtX76cDRs20KZNm2rtly5d4ne/+x2ffPIJbdq0ISoqioEDB9KxY0eTkorceLW9PkREmhO918nVahZnXvfu3cv9998PQO/evcnPz3fu279/P3369MHDwwMfHx8CAgL49ttvzYp6RXXl79y5Mx9++CFWqxU3NzfsdjutW7c2K+oV1ZUf4C9/+QsWi4UHHnjAjHh1qiv7119/TVBQEPPnzyc6OppbbrnFpQpXqDt/mzZt6Nq1KxUVFVRUVGCxWMyKWaeAgAAWL15co72goICAgADatWuHh4cHd999N3//+99NSChintpeHyIizYne6+RqNYvitaysDG9vb+e21WrFbrc79/n4+Dj3eXl5UVZWdsMz1qWu/O7u7vj5+WEYBvPnz+eOO+4gMDDQrKhXVFf+w4cP89lnn/HSSy+ZFa9OdWUvLS0lNzeXKVOmsHz5cj7++GO+//57s6JeUV35Abp06cLQoUMZMWIEY8eONSNivYYMGUKrVjUvArkZXrsiTa2214eISHOi9zq5Ws3it8Tb25vy8nLntsPhcL4A/n1feXl5tT+IXUFd+QEqKytJSEjAy8uLN954w4yIdaor//r16zl9+jTPPvssRUVFuLu7061bN5c5C1tX9vbt2xMaGuq8TLVv374cOnTIpT48qCt/dnY2Z86cISsrC4DY2FjCwsLo1auXKVkb6mZ47YqIiIjIjdMszryGhYWRnZ0NwL59+wgKCnLu69WrF3v37qWyspLz589TUFBQbb8rqCu/YRi8+OKL3H777cyePRur1WpWzFrVlX/q1KmsW7eOtLQ0RowYwbhx41ymcIW6s4eEhHD48GFKSkqw2+1888033HbbbWZFvaK68rdr1w5PT088PDxo3bo1Pj4+nDt3zqyoDfbLX/6SY8eOcfbsWWw2G3//+9/p06eP2bFERERExCTN4szr4MGDycnJYcyYMRiGwdy5c/noo48ICAjgoYceIiYmhujoaAzD4JVXXnG5MaN15Xc4HOzevRubzcYXX3wBwKuvvupSf8TX9/N3ZfVlf+2115gwYQIAjz76qMt98FFf/i+//JLIyEjc3NwICwtjwIABZkeu18aNG7lw4QKjR4/m9ddfJzY2FsMwePrpp7n11lvNjiciIiIiJrEYrrb2h4iIiIiIiMi/aRaXDYuIiIiIiEjzpuJVREREREREXJ6KVxEREREREXF5Kl5FRERERETE5al4FREREREREZen4vUmduLECSZPnkxkZCRjx44lLi6OI0eOXFefBQUFxMTENPhxubm59O/fnwULFnDx4kXi4+OJjo7m+eefp6Sk5Kr7eeeddxg5ciS5ubkNzlCXmJgYCgoKrrufuXPnkp6e3giJ6pabm8srr7xS6/4XXniB0NBQKisrmzyLiIiIiIgrUPF6k6qoqGDixIk899xzZGZmsmrVKiZNmsTs2bNNy3TvvfcydepU0tPTCQoKYs2aNQwfPpylS5dedR9//vOfWbVqFffcc08TJm24kpISJkyYwLZt28yOAsD7779Px44dzY4hIiIiInLDtDI7gFyb7du3c++999KnTx9nW69evVi1ahUAJ0+eZMaMGVRWVtK6dWvmzJlDVVUVr732Gp07d+bEiROEhoYya9Yszpw5w5QpUzAMo1pBtHv3bt555x2sVivdu3dn9uzZbNy4kU8//RSHw8HkyZPp379/jWx79+5lwoQJADzwwANXXbwuWbKEU6dO8cILL7BixQree+899uzZg2EYjBs3jscee4yYmBhuv/12jhw5Qtu2benbty87d+7k3LlzrFy5EqvVSmJiIufPn6e0tJSIiAiio6Odz3H+/HkSExMpLS0FYPr06dx+++3VcowfP55ly5bh4eHhbCsvLyc+Pp7s7OyrOpbLcnNz+eCDD3B3d+fUqVOMGTOGr776im+//ZaxY8cSHR3NX/7yF37/+987H/Puu+9W62Pz5s2kpqbi5ubG3XffzZQpUxqUQURERESkOdCZ15tUYWEhAQEBzu2JEycSExPDo48+yqlTp5g/fz4xMTGkpaURGxvLW2+9BcDRo0dJTk5m3bp1ZGdnU1xczEcffcQTTzxBWloaDz/8MACGYTBjxgyWLFnC6tWrufXWW/njH/8IgK+vL+np6VcsXAHKysrw8fEBwMvLi/Pnz1/VMU2aNImOHTuycuVKcnNzKSwsZO3ataxatYply5Zx7tw54Kci/eOPP8Zms+Hp6clHH33Ebbfdxp49ezh27BhDhw5l5cqVLFu2jNTU1GrPsWzZMu69917S0tKYM2cOSUlJNXKsXLmyWuEK0L17d+66666rOo5/d+rUKRYvXkxSUhIpKSksWLCA5cuXk5GRAfz0b/LBBx+QlpZGYGAgO3fudD727NmzLF68mNTUVNLT0zl9+jQ5OTnXlENERERE5GamM683qc6dO5Ofn+/cTklJASAyMhK73c7hw4d5//33+fDDDzEMA3d3dwACAgLw9vYGoGPHjlRWVnLkyBGGDRsGQFhYGOnp6ZSUlHDmzBlefvllAC5evMiAAQMICAggMDCwzmze3t6Ul5cDP52x9PX1rbZ/9erVbNmyBYC33nqLW2+9tUYfhw8f5sCBA87xt3a7nR9++AGAO++8E/ipiL7tttucX1dWVnLLLbfw8ccfs3XrVry9vbHb7TX6/eqrr9i8eTOAsyC+Xu+88w7/+7//C0BqaipWq9W571e/+hXu7u74+PgQEBCAh4cH7dq1c45X/Y//+A+mTZuGl5cX//jHP+jdu7fzscePH6ekpIS4uDjgp5/niRMnGiWziIiIiMjNRMXrTeqhhx5i+fLl7Nu3z1nsHDt2jFOnTmGxWOjZsyfjx48nLCyMgoIC9uzZA4DFYqnRV8+ePfn666/59a9/TV5eHgAdOnSgc+fOLF26FB8fH7Kysmjbti0nT57Eza3uE/ZhYWHs2LGDXr16kZ2dzd13311t/zPPPMMzzzxTZx89e/bknnvuYc6cOTgcDpYuXYq/v3+9P5eVK1fSu3dvoqOj+eqrr9ixY0eNfp966imefPJJ/vnPf7Ju3bp6+7wadU2udKWf+WXnz5/nf/7nf/j8888BeO655zAMw7nf39+fLl26sHLlStzd3fnDH/5AcHBwo2QWEREREbmZqHi9SXl5eZGSksKiRYt46623sNvttGrVijlz5tCtWzemTZtGUlISlZWVXLx4kcTExFr7eumll3jllVf485//7CwQ3dzcSExMJHBD1XMAAAEvSURBVC4uDsMw8PLyYsGCBZw8ebLebFFRUUybNo2oqCjc3d1ZtGhRg49v0KBB7N69m+joaC5cuMDDDz/sPGNcl4EDB5KUlMTGjRtp3749VqsVm83m3P+b3/yGxMREMjMzKSsrY9KkSTX6uNKY16bi7e1NWFgYI0aMoG3btvj6+nLmzBnnv4Ofnx/jxo0jJiaGqqoqunXrxmOPPdbkuUREREREXI3F+PlpHpFrlJuby9q1a3nnnXfMjtJiDBo0iM2bN9O6dWuzo4iIiIiINDlN2CSN5quvvmLBggVmx2gRXnjhBYqLi82OISIiIiJyw+jMq4iIiIiIiLg8nXkVERERERERl6fiVURERERERFyeilcRERERERFxeSpeRURERERExOWpeBURERERERGX9/8Aeu9PDLn5m3UAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1152x864 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting the box plots \n",
    "plt.figure(figsize=[16,12])\n",
    "\n",
    "plt.subplot(231)\n",
    "plt.boxplot(x = X['Age'], showmeans = True, meanline = True)\n",
    "plt.title('Age Boxplot')\n",
    "plt.ylabel('Age (years)')\n",
    "\n",
    "plt.subplot(232)\n",
    "plt.boxplot(X['Total_Bilirubin'], showmeans = True, meanline = True)\n",
    "plt.title('Total Bilirubin Boxplot')\n",
    "plt.ylabel('Total Bilirubin (mg/dL)')\n",
    "\n",
    "plt.subplot(233)\n",
    "plt.boxplot(X['Direct_Bilirubin'], showmeans = True, meanline = True)\n",
    "plt.title('Direct Bilirubin Boxplot')\n",
    "plt.ylabel('Direct Bilirubin (mg/dL)')\n",
    "\n",
    "plt.subplot(234)\n",
    "plt.hist(x = [X[y==1]['Is_male'], X[y ==0]['Is_male']], \n",
    "         stacked=True, color = ['g','r'],label = ['Healthy','Patient'])\n",
    "plt.title('Gender Histogram by patients')\n",
    "plt.xlabel('Gender [0 - female : 1 - male]')\n",
    "plt.ylabel('# of people')\n",
    "plt.legend()\n",
    "\n",
    "plt.subplot(235)\n",
    "plt.boxplot(x = X['Alkaline_Phosphotase'], showmeans = True, meanline = True)\n",
    "plt.title('Alkaline Phosphotase')\n",
    "plt.ylabel('Alkaline Phosphotase (International Units /Litre)')\n",
    "\n",
    "plt.subplot(236)\n",
    "plt.boxplot(X['Alanine_Aminotransferase'], showmeans = True, meanline = True)\n",
    "plt.title('Alanine Aminotransferase Boxplot')\n",
    "plt.ylabel('Alanine Aminotransferase (units/L)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see from Gender histogram Number of Males having liver diseases are way more than the females. Ladies you can relax a little bit. \n",
    "        \n",
    "        ME - Wanna grab a coffee sometime? \n",
    "        Google - Coffee is a brewed drink prep....\n",
    "        ME - I did not ask you!!!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Albumin (g/dL)')"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x864 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=[16,12])\n",
    "plt.subplot(231)\n",
    "plt.boxplot(X['Aspartate_Aminotransferase'], showmeans = True, meanline = True)\n",
    "plt.title('Aspartate Aminotransferase Boxplot')\n",
    "plt.ylabel('Aspartate_Aminotransferase (units/L)')\n",
    "\n",
    "\n",
    "plt.subplot(232)\n",
    "plt.boxplot(X['Total_Protiens'], showmeans = True, meanline = True)\n",
    "plt.title('Total Protiens Boxplot')\n",
    "plt.ylabel('Total Protiens (g/dL)')\n",
    "\n",
    "plt.subplot(233)\n",
    "plt.boxplot(X['Albumin'], showmeans = True, meanline = True)\n",
    "plt.title('Albumin Boxplot')\n",
    "plt.ylabel('Albumin (g/dL)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see, many boxplots tells us that there are many outliers present. But these cannot be ignored as they are still possible. A person can has high levels of Alanine Aminotransferase, this clearly indicates that the person has liver problems.\n",
    "\n",
    "This will be pretty much clear in the following plots."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1d9da9a52c8>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x864 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, saxis = plt.subplots(2, 3,figsize=(16,12))\n",
    "\n",
    "sns.barplot(y = 'Alanine_Aminotransferase', x = 'Liver_disease', data=df, ax = saxis[0,0])\n",
    "sns.pointplot(y = 'Total_Bilirubin', x = 'Liver_disease', data=df, ax = saxis[0,1])\n",
    "sns.pointplot(y = 'Direct_Bilirubin', x = 'Liver_disease', data=df, ax = saxis[0,2])\n",
    "\n",
    "\n",
    "sns.barplot(y = 'Alkaline_Phosphotase', x = 'Liver_disease', data=df, ax = saxis[1,0])\n",
    "sns.barplot(y = 'Aspartate_Aminotransferase', x = 'Liver_disease', data=df, ax = saxis[1,1])\n",
    "sns.boxplot(y = 'Total_Protiens', x = 'Liver_disease', data=df, ax = saxis[1,2])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see higher the values of individual test, the more risk you have of having one or more liver related diseases. So eat healthy guys!!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x864 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def correlation_heatmap(df):\n",
    "    _ , ax = plt.subplots(figsize =(14, 12))\n",
    "    colormap = sns.diverging_palette(220, 10, as_cmap = True)\n",
    "    \n",
    "    _ = sns.heatmap(\n",
    "        df.corr(), \n",
    "        cmap = colormap,\n",
    "        square=True, \n",
    "        cbar_kws={'shrink':.9 }, \n",
    "        ax=ax,\n",
    "        annot=True, \n",
    "        linewidths=0.1,vmax=1.0, linecolor='white',\n",
    "        annot_kws={'fontsize':12 }\n",
    "    )\n",
    "    \n",
    "    plt.title('Pearson Correlation of Features', y=1.05, size=15)\n",
    "\n",
    "correlation_heatmap(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now seeing all those graphs I'm sure you are pretty bored. Don't worry we will now coming to the most intersting part.\n",
    "\n",
    "<centre>THE MODELLING!!!!</centre>\n",
    "\n",
    "No? not interested? You don't want to see how can this dataset can help us? \n",
    "If yes, continue to read..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model The Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Little Note** - When it comes to data modeling, the beginner’s question is always, \"what is the best machine learning algorithm?\" To this the beginner must learn, the [No Free Lunch Theorem (NFLT)](http://robertmarks.org/Classes/ENGR5358/Papers/NFL_4_Dummies.pdf) of Machine Learning. In short, NFLT states, there is no super algorithm, that works best in all situations, for all datasets. So the best approach is to try multiple MLAs, tune them, and compare them for your specific scenario."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Before Modelling let us split the data into train and test data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import preprocessing\n",
    "X_scaler = preprocessing.normalize(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Shape: (434, 10)\n",
      "Test Shape: (145, 10)\n"
     ]
    }
   ],
   "source": [
    "# Splitting the data \n",
    "X_train, X_test, y_train, y_test = model_selection.train_test_split(X_scaler, y, random_state = 0)\n",
    "\n",
    "print(\"Train Shape: {}\".format(X_train.shape))\n",
    "print(\"Test Shape: {}\".format(X_test.shape))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Logistic Regression\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "lr = LogisticRegression()\n",
    "lr.fit(X_train,y_train)\n",
    "y_pred = lr.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Score of the model is -  0.7172413793103448\n",
      "Report card of this model - \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.717     1.000     0.835       104\n",
      "           1      0.000     0.000     0.000        41\n",
      "\n",
      "    accuracy                          0.717       145\n",
      "   macro avg      0.359     0.500     0.418       145\n",
      "weighted avg      0.514     0.717     0.599       145\n",
      "\n",
      "Accuracy score -  0.7172413793103448\n"
     ]
    }
   ],
   "source": [
    "# Use score method to get accuracy of model\n",
    "score = lr.score(X_test, y_test)\n",
    "print(\"Score of the model is - \",score)\n",
    "print(\"Report card of this model - \")\n",
    "print(metrics.classification_report(y_test, y_pred, digits=3))\n",
    "print(\"Accuracy score - \", metrics.accuracy_score(y_test,y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set ROC AUC score: 0.500\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "test_roc_auc = roc_auc_score(y_test, y_pred)\n",
    "\n",
    "# Print test_roc_auc\n",
    "print('Test set ROC AUC score: {:.3f}'.format(test_roc_auc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1, 'Accuracy Score: 0.7172413793103448')"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x648 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm1 = metrics.confusion_matrix(y_test, y_pred)\n",
    "plt.figure(figsize=(9,9))\n",
    "sns.heatmap(cm1, annot=True, fmt=\".3f\", linewidths=.5, square = True, cmap = 'Blues');\n",
    "plt.ylabel('Actual label');\n",
    "plt.xlabel('Predicted label');\n",
    "all_sample_title = 'Accuracy Score: {0}'.format(score)\n",
    "plt.title(all_sample_title, size = 15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Naives Bayes Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Naives Bayes\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "nb = GaussianNB()\n",
    "nb.fit(X_train,y_train)\n",
    "y_pred_nb = nb.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Score of the model is -  0.5655172413793104\n",
      "Report card of this model - \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.887     0.452     0.599       104\n",
      "           1      0.380     0.854     0.526        41\n",
      "\n",
      "    accuracy                          0.566       145\n",
      "   macro avg      0.634     0.653     0.563       145\n",
      "weighted avg      0.744     0.566     0.578       145\n",
      "\n",
      "Accuracy score -  0.5655172413793104\n"
     ]
    }
   ],
   "source": [
    "score = nb.score(X_test, y_test)\n",
    "print(\"Score of the model is - \",score)\n",
    "print(\"Report card of this model - \")\n",
    "print(metrics.classification_report(y_test, y_pred_nb, digits=3))\n",
    "print(\"Accuracy score - \", metrics.accuracy_score(y_test,y_pred_nb))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set ROC AUC score: 0.653\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "test_roc_auc = roc_auc_score(y_test, y_pred_nb)\n",
    "\n",
    "# Print test_roc_auc\n",
    "print('Test set ROC AUC score: {:.3f}'.format(test_roc_auc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1, 'Accuracy Score: 0.5655172413793104')"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x648 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm2 = metrics.confusion_matrix(y_test, y_pred_nb)\n",
    "plt.figure(figsize=(9,9))\n",
    "sns.heatmap(cm2, annot=True, fmt=\".3f\", linewidths=.5, square = True, cmap = 'Wistia');\n",
    "plt.ylabel('Actual label');\n",
    "plt.xlabel('Predicted label');\n",
    "all_sample_title = 'Accuracy Score: {0}'.format(score)\n",
    "plt.title(all_sample_title, size = 15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Stochastic Gradient Descent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Stochastic Gradient Descent\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "sg = SGDClassifier()\n",
    "sg.fit(X_train,y_train)\n",
    "y_pred_sg = sg.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Score of the model is -  0.7172413793103448\n",
      "Report card of this model - \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.717     1.000     0.835       104\n",
      "           1      0.000     0.000     0.000        41\n",
      "\n",
      "    accuracy                          0.717       145\n",
      "   macro avg      0.359     0.500     0.418       145\n",
      "weighted avg      0.514     0.717     0.599       145\n",
      "\n",
      "Accuracy score -  0.7172413793103448\n"
     ]
    }
   ],
   "source": [
    "score = sg.score(X_test, y_test)\n",
    "print(\"Score of the model is - \",score)\n",
    "print(\"Report card of this model - \")\n",
    "print(metrics.classification_report(y_test, y_pred_sg, digits=3))\n",
    "print(\"Accuracy score - \", metrics.accuracy_score(y_test,y_pred_sg))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set ROC AUC score: 0.500\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "test_roc_auc = roc_auc_score(y_test, y_pred_sg)\n",
    "\n",
    "# Print test_roc_auc\n",
    "print('Test set ROC AUC score: {:.3f}'.format(test_roc_auc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1, 'Accuracy Score: 0.7172413793103448')"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x648 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm3 = metrics.confusion_matrix(y_test, y_pred_sg)\n",
    "plt.figure(figsize=(9,9))\n",
    "sns.heatmap(cm3, annot=True, fmt=\".3f\", linewidths=.5, square = True, cmap = 'Greens');\n",
    "plt.ylabel('Actual label');\n",
    "plt.xlabel('Predicted label');\n",
    "all_sample_title = 'Accuracy Score: {0}'.format(score)\n",
    "plt.title(all_sample_title, size = 15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " ### K-Nearest Neighbours"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# KNN Model\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "hist = []\n",
    "for i in range(1,10):\n",
    "    clf = KNeighborsClassifier(n_neighbors=i)\n",
    "    cross_val = cross_val_score(clf, X_scaler, y, cv=5)\n",
    "    hist.append(np.mean(cross_val))\n",
    "plt.plot(hist)\n",
    "plt.title('Cross Validations score for KNeighborsClassifier')\n",
    "plt.xlabel('n_neighbors')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "knn = KNeighborsClassifier(n_neighbors = 7)\n",
    "knn.fit(X_train,y_train)\n",
    "y_pred_knn = knn.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Score of the model is -  0.6620689655172414\n",
      "Report card of this model - \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.731     0.837     0.780       104\n",
      "           1      0.346     0.220     0.269        41\n",
      "\n",
      "    accuracy                          0.662       145\n",
      "   macro avg      0.539     0.528     0.524       145\n",
      "weighted avg      0.622     0.662     0.636       145\n",
      "\n",
      "Accuracy score -  0.6620689655172414\n"
     ]
    }
   ],
   "source": [
    "score = knn.score(X_test, y_test)\n",
    "print(\"Score of the model is - \",score)\n",
    "print(\"Report card of this model - \")\n",
    "print(metrics.classification_report(y_test, y_pred_knn, digits=3))\n",
    "print(\"Accuracy score - \", metrics.accuracy_score(y_test,y_pred_knn))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set ROC AUC score: 0.528\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "test_roc_auc = roc_auc_score(y_test, y_pred_knn)\n",
    "# Print test_roc_auc\n",
    "print('Test set ROC AUC score: {:.3f}'.format(test_roc_auc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1, 'Accuracy Score: 0.6620689655172414')"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgAAAAH3CAYAAADNB+fGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO3deXwO5/7/8fedTXZid2wVWy21NaKIWFqiDuooTalQam2V2HdiOVKNlkoPtfVQmia0tRxUW6laSqMcpYlqagmKErtE9ty/P/zMaUiCfiWRzuv5eOTxyD1zzXV97jvLvOeauee2WK1WqwAAgKnYFHQBAAAg/xEAAAAwIQIAAAAmRAAAAMCECAAAAJgQAQAAABOyK+gCAAAobAYNGpTnYyxevDhP+2cGAAAAEyIAAABgQgQAAABMiAAAAIAJEQAAADAhAgAAACZEAAAAwIQIAAAAmBABAAAAEyIAAABgQgQAAABMiAAAAIAJEQAAADAhAgAAACZEAAAAwIQIAAAAmBABAAAAEyIAAABgQgQAAABMiAAAAIAJEQAAADAhAgAAACZEAAAAwIQIAAAAmBABAAAAEyIAAABgQgSAQiw+Pl61a9dWhw4dCrqUfJeSkqLQ0FD5+fmpbt26atKkiYYMGaLo6OiCLu2BrFixQq1bt1b9+vXVt29fxcXF5dr+xo0bmjhxory9veXt7a1Ro0bpypUrWdqsXbtWfn5+qlevnrp27aq9e/dmWb969Wq1b99eDRo0UIcOHbR27dos60+dOqXXXntNDRs2VMuWLbVs2bJ76rjfGF988YU6depkjPHZZ59lWR8XF6cBAwbIy8tLvr6+WrBggdLT07O0iYyMVOfOnVWvXj09//zz2rJlS5b1w4YNU82aNbN8vfrqq8b6b7/99p71NWvW1O+//37P89m0aZPatm2bZdn48eOz3b5mzZpav369JOnSpUsaN26cfHx85OXlpddee02xsbH39C/d/tm1bNlSoaGh2a6/du2afHx8tH///mzXS9KPP/6o2rVrKyoqKsc2wMMiABRiGzZsUMWKFXX8+PFc/3n8FU2cOFFbt27VpEmTtHXrVi1fvlxOTk7q1auXjh8/XtDl5Wrt2rVasGCBxo0bpzVr1qhIkSLq37+/UlNTc9xm6NChio6O1tKlS7Vs2TLFxsZqwoQJxvp169Zp+vTpGjBggP7zn/+ocePGev311/Xbb79JksLCwvTOO+9oyJAh2rhxo/r27avp06cbO7TU1FT1799fLi4uWrt2rUaPHq33339fa9aseeAx9u/frzFjxuiVV17Rf/7zH/Xu3VtTpkzRt99+K0m6fv26XnnlFaWkpOijjz7Su+++qy+++EJTp041xti7d6/efPNNdezYUZs2bdKLL76o0aNH69ChQ0abX3/9VaNGjdLu3buNr/fee89YHxsbq9q1a2dZv3v3bpUuXTrLa7p9+3ZNmjTpntd60qRJ92zbunVreXp66rnnnlNmZqaGDh2quLg4LVy4UOHh4XJ1ddWrr76qq1ev3tNfUFBQtuFDuh3iX3vtNcXHx2f/g5d069YtjR07VhkZGTm2Af4MAkAhtn79enXo0EG1a9dWREREQZeTbxISErR582aNGjVKvr6+qlChgurWrauQkBCVKFEiy07rcbRs2TL17dtX7du3V82aNfXOO+/o8uXL+vLLL7Nt//3332v//v1asGCB6tevr3r16mn8+PE6ceKEbt26JavVqtDQUA0YMEDdunVT5cqVNW7cOFWqVEkHDx6UJIWHh6tnz5564YUXVKlSJXXv3l2dO3fW559/Lkn66quvdOnSJQUHB6tatWrq1KmT+vfvr+XLl0vSA40RGRmpGjVq6OWXX1bFihX18ssvGzti6XaASEpK0oIFC1S7dm15eXlp1qxZ+uyzz4wQ8a9//UsdO3bUwIEDValSJfXv31/NmjXTvn37JN0OKqdPn1a9evVUqlQp46to0aLG6/Xrr7+qRo0aWdaXKlVKNja3/90lJydrypQpevPNN1W5cuV7Xm83N7cs2/3www/atWuX5s+fL1dXVx09elQHDx7U7NmzVa9ePVWrVk0hISG6deuWduzYkaWvzZs3KyYmRmXKlLlnnM2bN+uFF16Q1WrN9fflrbfeynZ74P+KAFBIHT58WL/++quaNWumdu3a6csvv9T169eN9QkJCZo+fbqaNWumhg0b6rXXXtOJEyeM9Tt27FD37t1Vv359tWnTJst0b82aNbVhw4Ys4/1x2fjx4xUYGKiAgAA9/fTTCgsLU0pKioKDg9W6dWvVrVtXzzzzjCZMmKCkpKQsNQcEBKhBgwby8fHR22+/rfT0dK1YsULe3t5ZjoATExPVoEEDbdu2Ldvnb2Njo927d2c5KrK1tdXKlSs1cOBAY1lcXJwGDx6sRo0a6ZlnntGkSZOUmJgoSUpPT9fSpUvVrl07PfXUU+rUqVOW6ebQ0FAFBARo2LBhatSokebNmydJ2rZtmzp37qynnnpK7du31/Lly5WZmWlsFxAQoICAgGzrvnz5suLi4uTt7W0sc3FxUd26dXOcxdm9e7dq1aqlJ554wljWvHlzff3113J2dtaJEyd09uzZLKeCbGxstGHDBnXq1EmSNHnyZL388sv3vIY3btyQdPvovW7dunJxcTHWe3t7Ky4uTpcuXXqgMTw8PPTrr7/q+++/l9Vq1Q8//KBff/1VdevWlXT7FEP16tVVrFgxo4/atWsb49+6dUsHDhy455TWsmXLNGDAAEnSiRMnlJ6erqpVq2b7Wkm3A0Bu6y9fvqzTp08rPDz8nun/uyUnJ+vtt99Wnz59VLNmTUlSuXLltHjxYlWpUsVoZ7FYZLVas/wNXrhwQTNnztRbb72lIkWK3NP39u3bNXTo0CyzF3fbsWOHvv32W02ePDnXOoE/gwBQSK1bt04lS5bU008/reeff14pKSnGdK4kBQYGau/evXrnnXf02WefydnZWf3791daWpoOHjyowYMHq3nz5lq/fr0mTJigf/3rXw915PzFF1+obdu2WrNmjdq2bas5c+Zo+/btCgkJ0datWzV16lRt3rzZmJk4c+aMevfurcqVK+vTTz9VSEiINm7cqNDQUHXu3Pmeo6evvvpKTk5Oatmy5T1ju7q6qmfPnvr444/l6+ursWPHau3atTp37pwqVKigEiVKSLp97rVXr16Sbk+BL168WAcPHjSmnN966y0tX75cI0eO1MaNG/X3v/9dI0eOzHIkvm/fPlWsWFHr1q1Tt27dtGPHDo0ePVq9e/fW5s2bNWbMGH300UdauHChsU1oaGiO53vvTAXffURXunTpHKeJ4+LiVKlSJa1cuVLt27dXy5YtNWnSJGNnc+f6gRs3bqh3795q2rSpXnnlFf33v/81+vD29lbFihWNx+fOndPmzZvVokULo667p8jvPD5//vwDjfHKK6+oefPm6tOnj+rUqaNevXqpZ8+e6tKli9HfhQsXsoSls2fPSvrfTjkzM1NWq1WDBw9W06ZN9eKLLyoyMtJoHxsbK3t7e4WGhqpVq1by8/PTvHnzlJKSIknKyMjQiRMnFB0drc6dO8vHx0dDhgzJEn7Lly+vlStXGsEkNxEREbp586aGDBliLPPw8FCrVq2MGQVJWrVqlVJSUuTj4yPp9ozJhAkT1K1bNzVs2DDbvufOnauePXvKYrFku/7KlSuaNGmSZs2alWWGA3hUCACFUGpqqrZs2aJ27drJxsZGTzzxhOrUqWPswE+cOKFdu3Zp+vTpatq0qTw9PTVjxgy1a9dO165d06pVq+Tl5aXAwEBVqVJFbdu21bRp0+Tk5PTANZQqVUq9e/dW1apVVapUKdWvX1/BwcHy8vJShQoV1KFDB9WrV8+4MGrNmjUqWbKkpk+frmrVqqlp06aaOXOmSpcureLFi6tFixbauHGj0f+GDRvUsWNH2dvbZzv+5MmTFRISoqpVq2rLli2aPHmy2rRpo2HDhunmzZuSpC1btujWrVuaO3eunnzySdWvX1+zZs1SlSpVlJCQoE8++UQjRoxQ+/btVaVKFQ0ePFjt27fXkiVLjHEsFosxVVyxYkV98MEH6tGjh7p166ZKlSrp2Wef1ahRo7R06VJjx1asWLEsR7l/dGdG5O4jQgcHB2MndreEhATt3r1b3377rYKDgxUcHKxDhw5p6NChslqtSkhIkHR7ZqZ79+5atmyZqlevrj59+mR7PcSVK1c0aNAglSxZ0pgtSU5OzrYm6fYFlw8yxpUrV3T58mWNGTNGn332mSZPnqywsDB9+umnkqTnn39ely9fVkhIiJKSknTp0iXNmjVLdnZ2SktLM8aYMmWKWrZsqQ8//FCtW7fWG2+8YVxseOzYMUlSlSpVtHjxYg0dOlSffvqpEepOnz6tlJQUpaamatasWZo/f75SU1P1yiuv6PLly9m+vjnJzMzUqlWr1LNnT7m5ueXYLjIyUu+++6769u1rzDysWrVK8fHxGjZs2EON+UfTpk1TmzZt5Ovr+6f7AHJjV9AF4OFFRkbq2rVrat++vbHs+eef19y5c7V//35dunRJklSvXj1jvYeHh8aPHy/p9lHU3f9U7hylPagKFSpkefzCCy9o9+7devvttxUXF6djx47p9OnTRrvY2FjVqVNHtra2xjatW7c2vu/atatGjhypGzduKCkpSVFRURo7dmyuNXTu3NmYPThw4IC++OILrVu3TjY2Npo/f75iY2Pl6ekpV1dXY5tGjRqpUaNGOnz4sNLT09WoUaMsfTZu3FjffPON8bhUqVJydHQ0Hv/888/66aefFB4ebizLzMxUcnKyzp49m+UoOzt3+rr7gr/U1NQcA5idnZ3S09MVGhpqPJfg4GB169ZNR44cMULS4MGDjen42rVr68CBA/rkk0+yTB+fOXNG/fv3V3JyslavXm3s2BwdHbOtSZKcnZ0faIzJkyerVq1a6t+/vySpVq1aunLlikJCQvTiiy/qiSee0HvvvaepU6dqxYoVcnZ21rBhw/TLL7/Izc3NGOOll15Sjx49jD5++uknrVy5Uk2bNlVgYKD69etnBKyaNWvK1tZWI0aM0Pjx41WlShV9//33Klq0qHGE/v7776tVq1basGGD+vXrl+vP54/++9//6syZM3rppZdybPP5559rypQp6tChg8aMGSNJOn78uN577z2tXr3aCFEPa926dTpy5EiWUAw8agSAQmjdunWSpL59+xrL7lxItGbNGrVr1y7X7e3sHu7HfvfbtCRl2SlKt4/It23bpn/84x9q166dRowYoRkzZjzwmK1atZKLi4txLUP16tWN88N3i4qK0rfffqtx48ZJur2DatGihVq0aKGSJUtq1apV9x0zp3/MGRkZWba7+3na29urf//+xk7wjx7kQq1y5cpJun319x8vQLt48WKO563LlCmj8uXLZwky1apVkyT99ttvxlR9jRo1jPUWi0Wenp7GxXWSdOTIEQ0YMEDu7u4KDw83apGksmXL6uTJk1nGvXjxojH+rVu37jvGoUOH1LFjxyx91K9fXwsXLtSNGzdUtGhRtWnTRm3atNHFixdVrFgxpaamavbs2apYsaLx+v1xDEmqWrWqdu3aJen2dQd3z67caf/777/Lw8NDHh4eWdY7OTmpYsWKOn/+/N0vba4iIyNVr169HEPdokWLNH/+fPXq1UuTJ082pvK/+OIL3bp1Sz179jTaJiUlafHixdq6das2b95837E///xzXbhwIcspBUkaMGCAunTpkuVvC/izOAVQyMTHx2v37t3q2bOn1q9fb3xt2LBBPj4+2rp1q0qVKiVJWd4Tn5CQoKZNm2r//v2qWrXqPe+Xnzdvnl5//XVJt3dyd6ZjpdsXb+UmISFBn332mWbMmKFx48apS5cuqlKlis6cOWP846pataqOHDmS5fxvRESEunbtaozZsWNHbdu2zQgSuY334YcfKiYm5p51bm5uxjUAVatW1cmTJ42L/iRp165datWqlSpVqiR7e3sdOHAgy/YHDhwwdq7ZqVatmuLi4lS5cmXjKzY21rhA8H5KlCihJ554wriqXbp9wWN0dLQaN26c7TZeXl46ffp0lgvMfv31V0lSpUqVVKdOHTk7O+unn34y1lutVh0/ftzYeR0/flx9+/bV3/72N4WFhWXZ+UvS008/rejo6CwXbUZFRalKlSoqUaLEA41RpkwZ/fLLL1n6jY2NVbFixVS0aFHt379fffr0UUZGhkqXLi0HBwdt27ZNzs7OatSokcqWLavy5ctnGePOc70zxvDhw/XGG29kWR8dHS0HBwdVqlRJ27ZtU8OGDbPcIyEhIUFxcXGqXr16tq9vTg4cOKBnnnkm23VLly7V/PnzNWzYME2ZMiXLefxevXpp69atWf4+y5Urp5dffjnL6aXczJ07V5s3bza2v3OR7qxZszR8+PCHeh5ATggAhcyGDRtktVrVv39/1ahRI8tX//79lZKSokOHDunZZ5/V9OnTtX//fh0/flwTJkyQm5ub6tWrp379+umHH37QwoULderUKX355Zf66KOP1KZNG0lSgwYNtGbNGh09elQxMTGaNm1arlOZRYoUkbOzsyIjI3X69GkdOXJEo0aN0vnz541p5FdeeUWXLl3SzJkzdfz4cX333XcKDQ3NcpFf165dtWfPHkVHR2d7hH1H69at1bhxYw0aNEhr167VqVOn9Msvvyg8PFyLFy82dhCdOnWSi4uLJkyYoNjYWB08eFDBwcHy9vaWs7Oz+vbtq/nz52vr1q2Ki4vTkiVL9NVXX2WZWbnbkCFDtHnzZi1ZskRxcXH69ttvNXXqVDk6Ohqv0bVr13Tt2rUc+3j11Ve1dOlSbd68WbGxsRo1apRKly5tXJGekZGh+Ph4JScnS7p9eqdcuXIaPny4jh49qkOHDmny5Mlq0qSJatWqJScnJ/Xp00fz58/XV199pbi4OAUHB+v06dPGVPq4cePk4OBgvPMiPj5e8fHxxo6ybdu2Klq0qEaNGqXY2Fht2rRJy5cvN64ReJAxevfurfDwcH3yySc6c+aM/vOf/2jx4sUaNGiQJMnT01NHjhzR3LlzdebMGX399deaOXOmBg0aZMxuDBkyRB999JHWrl2r06dPa8mSJdq9e7f69OkjSfLz81NkZKT+/e9/6/Tp09q6davmzJmjfv36ycXFRY0bN5arq6vGjBlj/P4OHz5cHh4eeuGFF3L8mdzNarXq6NGj98xGSNLRo0c1b948vfjii3rppZeM1zI+Pl63bt1SsWLFsgTEypUry87OTkWLFlX58uUfaPwyZcpk2f7OqbQyZcoYARf4v+IUQCGzfv16tWrVKtt/JE2bNtWTTz6pNWvWKCwsTMHBwXr99deVkZGhxo0ba9myZXJwcFCdOnUUGhqqBQsWaOHChSpbtqxGjBihbt26Sbp945KgoCB1795dpUuX1vDhw3XhwoUca7K3t9f8+fM1Z84cdezYUcWLF5evr6/69etnvI2vTJkyWrp0qebOnasuXbqoePHi6tatm4YOHWr0U7t2bT3xxBMqX768SpYsmeN4NjY2WrJkiZYtW6YVK1Zo1qxZslgsevLJJzV79mzjFIizs7OWL1+u4OBgde/eXS4uLvLz8zPO1Q4fPlw2NjaaPXu2rl69qqpVq+rdd9/V888/n+PYvr6+evvtt7VkyRItWLBAxYsXV5cuXTRixAijzZtvvilJxqmIu/Xo0UM3b95UcHCwEhMT1ahRI+NnI92+6v7ZZ59VcHCwunbtKkdHR61cuVKzZ89Wjx49ZGdnp+eeey7LjYCGDx8uJycnzZ49W5cvX1atWrX04YcfytPTUydPnjSOqv943Yh0ewbh66+/lqOjo5YtW6agoCB169ZNJUqU0IgRI4wZmvuNId0OeQ4ODlq5cqXmzJmj8uXLa+TIkcZUePHixbVo0SLNmTNHYWFhKl26tN58880sd/Hr3r27pNtv/Zs+fbqqVKmi0NBQNWnSRJLUoUMHpaamavny5Zo3b55KlCih3r17GyGjaNGiWrFihUJCQtS7d2+lp6erefPmWrlyZbZvxcvJ9evXlZKSku3FnFu2bFFGRoY+++yze+50OHz4cGMmDXjcWaz3uwsFkE/S09PVqlUrTZ069b7XMQBAQboTOvPS4sWL87R/ZgBQ4FJTU/XNN99o165dcnBwyPLuAABA3iAAoMDZ2dlp5syZcnBwUEhISI7v/QcAPDoEABQ4GxsbfffddwVdBgAUGmlpaRo/frzOnj0rGxsbzZw5U3Z2dho/frwsFouqV6+uadOmZblj5d0IAAAAFDI7duxQenq6wsPD9d1332n+/PlKS0tTYGCgmjRpoqlTpyoyMjLXz7vgbYAAADyG7twr5c7XHz/1tUqVKsrIyFBmZqYSEhJkZ2enmJgY44PGfH19tWfPnlz7fyxnAPLj6kqgMLpzVXDQvqCCLQR4TAV5BxV0CY+Mv7+//P39s13n7Oyss2fP6vnnn9fVq1f1wQcf6IcffjBuSuXi4mJ8LkpOHssAAAAAcrZixQr5+PgYN13r06eP0tLSjPWJiYlyd3fPtQ9OAQAAUMi4u7sbH+ZVtGhRpaenq3bt2oqKipIk7dy5U15eXrn2wQwAAACFzKuvvqqJEyeqZ8+eSktL04gRI1S3bl1NmTJF7777rjw9PeXn55drHwQAAAAKGRcXF7333nv3LF+9evUD98EpAAAATIgAAACACREAAAAwIQIAAAAmRAAAAMCECAAAAJgQAQAAABMiAAAAYEIEAAAATIgAAACACREAAAAwIQIAAAAmxIcBAQDwkHokbsiHURbnae/MAAAAYEIEAAAATIgAAACACREAAAAwIQIAAAAmRAAAAMCECAAAAJgQAQAAABMiAAAAYEIEAAAATIgAAACACREAAAAwIQIAAAAmRAAAAMCECAAAAJgQAQAAABMiAAAAYEIEAAAATIgAAACACREAAAAwIQIAAAAmRAAAAMCECAAAAJgQAQAAABMiAAAAYEIEAAAATIgAAACACREAAAAwIQIAAAAmRAAAAMCECAAAAJgQAQAAABMiAAAAYEIEAAAATIgAAACACREAAAAwIQIAAAAmRAAAAMCECAAAAJgQAQAAABMiAAAAYEJ2BV0AAAB4OJ9//rnWrVsnSUpJSdHPP/+sVatW6Z///KdsbW3l4+OjoUOH5toHAQAAgEKma9eu6tq1qyRp+vTpevHFFzVt2jSFhoaqYsWKGjhwoGJiYlSnTp0c++AUAAAAhdRPP/2kY8eO6e9//7tSU1NVqVIlWSwW+fj4aO/evbluywwAAACPoYiICEVERBiP/f395e/vn6XN4sWL9cYbbyghIUGurq7GchcXF505cybX/gkAAAA8hrLb4f/RjRs3dOLECT3zzDNKSEhQYmKisS4xMVHu7u659s8pAAAACqEffvhBzZo1kyS5urrK3t5ep0+fltVq1e7du+Xl5ZXr9swAAABQCJ08eVIVKlQwHk+fPl2jR49WRkaGfHx8VL9+/Vy3JwAAAFAI9e/fP8vjBg0aaM2aNQ+8PacAAAAwIQIAAAAmRAAAAMCECAAAAJgQAQAAABMiAAAAYEIEAAAATIgAAACACREAAAAwIQIAAAAmRAAAAMCE+CwAAAAeUsNPBuf9IKvztntmAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCG7gi4A+cPGxkZ9+/ZViRIllJmZqVWrVqlTp05yd3eXJJUoUUInT57UsmXLjG3s7e3Vr18/ubm5KTk5WStWrFBCQoLq1aunv//978rMzNR3332n3bt359gWKEwuHbukQxGH9OykZ/Xd+98p+XqyJCnxUqJKVC2h5kObG23TU9O1d9FepdxIkZ2jnZ4Z9Iwc3R119r9nFb0+WhZbizx9PVWtdbUc2wIFKU8DQGZmpmxsmGR4HDz11FOysbHR22+/rVq1aqlLly5avHixJMnZ2VkjR47UmjVrsmzTsmVLnT17Vps2bZKXl5c6dOigTz/9VN27d1dwcLBSUlI0duxYHT58WN7e3ve0vbs/4HF2ZNMRxX0XJ7sit/8t3tnZpyamKnJ2pBr1apSl/bHIYypWsZie6vqUTu09pZgNMWrYo6H++/F/5TfDT7ZFbLVtxjaVb1hep/aeuqft0wFP5/tzBP7oke+dz5w5o9dff12+vr567rnn1KpVKw0cOFAnT5581EPhIVy4cEG2trayWCxydHRURkaGsa5Tp07avn27bty4kWWbatWqKSYmRpIUHR2tWrVqqVy5coqPj9etW7eUkZGhY8eOqVq1atm2BQoTtzJuajG8xT3Lf/rsJ9VoW0NOxZyyLI+PjVe5euUkSeXql9PvMb/r+rnrci3jKgcXB9na2apUjVKK/yU+27ZAQXvkMwCTJk3SqFGjVL9+fWPZjz/+qAkTJig8PPxRD4cHlJKSohIlSmj69OlydXXV+++/L0lyc3PTk08+me3RuqOjo5KSkoztnZycsiyTpOTk5HuW32kLFCYVG1dUQnzW01bJ15P1+5Hf1bBXw3vapyWlyd7JXpJk72ivtFtpSk9Kl4Ozg9HGzslOaUlp2bYFCtojDwCpqalZdv6S1KBBg0c9DB7Sc889p5iYGK1fv14eHh4aMWKEZsyYoUaNGmnfvn2yWq33bJOcnCxHx9vnKYsUKaJbt24pOTlZRYoUMdrc2fFn1xYo7M78cEZPNH0i21OZ9k72Sk9OlySlJafJwdnB2OHfkZ6ULntn+2zbAgXtkZ8CqFmzpiZMmKAtW7Zo165d2rp1qyZMmKCaNWs+6qHwEBITE40j9MTERNna2srGxka1atVSdHR0ttscO3ZMdevWlSTVrVtXx44d0/nz51W6dGk5OzvL1tZW1atX14kTJ7JtCxR2v8f8rnL1y2W7rmT1kjp36Jwk6fyh8ypVs5SK/q2obl64qZSEFGWkZ+jiLxdVslrJbNsCBe2RzwAEBQVp27ZtOnDggBISEuTq6qrWrVurbdu2j3ooPITIyAqIubEAABw7SURBVEj17t1bo0ePlp2dndavX6/U1FSVKVNGly5dytJ2+PDhev/997Vjxw717dtXY8aMUXp6upYvX67MzEx9+umnGj58uCwWi/bs2aNr165l2xYo7G6cvyHXUq5Zlm2fs12+o3xV/dnq+n7x9/p65teysbVRs9ebycbORo16NtK3b38rq9UqT19PORd3zrYtUNAs1uzmfgvYoEGDCroE4LF0550bQfuCCrYQ4DEV5B2UL+Nct837cYpm5O0YvEcPAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCdgVdAAAAeHiLFy/WN998o7S0NPXo0UPe3t4aP368LBaLqlevrmnTpsnGJufjfGYAAAAoZKKionTw4EF98sknWrVqlX7//XcFBwcrMDBQYWFhslqtioyMzLUPAgAAAIXM7t27VaNGDb3xxhsaPHiwWrVqpZiYGHl7e0uSfH19tWfPnlz74BQAAACPoYiICEVERBiP/f395e/vL0m6evWqzp07pw8++EC//fabhgwZIqvVKovFIklycXHRzZs3c+2fAAAAwGPojzv8uxUrVkyenp5ycHCQp6enihQpot9//91Yn5iYKHd391z75xQAAACFzNNPP61du3bJarXqwoULSkpKUtOmTRUVFSVJ2rlzp7y8vHLtgxkAAAAKmdatW+uHH35Qt27dZLVaNXXqVFWoUEFTpkzRu+++K09PT/n5+eXaBwEAAIBCaOzYsfcsW7169QNvzykAAABMiAAAAIAJEQAAADAhAgAAACZEAAAAwIQIAAAAmBABAAAAEyIAAABgQgQAAABMiAAAAIAJcStgAAAeUvvXG+b5GHvzuH9mAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhOxyWnHy5MkcN6pSpUqeFAMAAPJHjgFg6tSp2S63WCz66KOP8qwgAACQ93IMAKtWrTK+v3nzps6ePauKFSvKxcUlXwoDAAB5J8cAcMeXX36pRYsWKSMjQ+3bt5fFYtHrr7+eH7UBAIA8ct+LAP/9739rzZo1KlasmF5//XVt27YtP+oCAAB56L4BwMbGRg4ODrJYLLJYLHJycsqPugAAQB66bwDw8vLSyJEjdeHCBU2dOlVPPfVUftQFAADy0H2vARg5cqR27typ2rVrq2rVqmrdunV+1AUAAPLQfQPA1atXtWfPHp08eVLXr1+Xl5eX3Nzc8qM2AACQR+4bAMaNG6fWrVurS5cu2r9/v8aNG6eFCxfmR20AACAHXbp0MQ7IK1SoIH9/f/3zn/+Ura2tfHx8NHTo0Fy3v28ASElJUY8ePSRJTz75pL788stHUDYAAPizUlJSJGW9Z88LL7yg0NBQVaxYUQMHDlRMTIzq1KmTYx/3vRWwh4eHvvjiC3l5eenw4cOqUKHCo6ofAAD8CUePHlVSUpL69eun9PR0vfnmm0pNTVWlSpUkST4+Ptq7d++fCwB/vBVwWFiYwsLCJN2+FTAAAMhbERERioiIMB77+/vL399fkuTo6KjXXntN3bt3V1xcnAYMGCB3d3ejrYuLi86cOZNr/w90K+A/SktLe6gnAAAAHt4fd/h3q1KliipXriyLxaIqVarIzc1N165dM9YnJiZmCQTZue99AMLDw+Xn56dnn31Wbdq0UceOHR/yKQAAgEfp008/1VtvvSVJunDhgpKSkuTs7KzTp0/LarVq9+7d8vLyyrWP+14EuGbNGq1atUqLFi1S+/bttXLlykdTPQAA+FO6deumCRMmqEePHrJYLJo9e7ZsbGw0evRoZWRkyMfHR/Xr18+1j/sGAA8PD5UuXVqJiYlq0qSJFixY8MieAAAAeHgODg5655137lm+Zs2aB+7jvqcA3NzctG3bNlksFoWHh+vKlSsPVyUAAHjs3DcAzJo1S3/72980atQoxcXFKSgoKB/KAgAAeSnHUwC7d+/O8vjKlSvy8fHhXQAAAPwF5BgANm/enONGPj4+eVIMAADIHzkGgODg4PysAwAA5KP7XgMAAAD+eixWq9Va0EUAAFCYNH1zQ56PsTf0hTzt/4EvAvwjrgEAAKBweywvAszcwfUHQHZsWk6QJAXtCyrYQoDHVJB3UL6Mszf0Qr6Mk5ce+iLAixcv5lkxAAAgf9z3VsALFixQWFiY0tLSlJycrCeeeCLX2QEAAPD4u++7AHbu3KmdO3eqU6dO2rJli8qUKZMfdQEAgDx03wBQrFgxOTg4KDExUZUrV1ZSUlJ+1AUAAPLQfQNA2bJl9emnn8rJyUnvvPOOEhIS8qMuAACQh+57DcCMGTN0/vx5tW/fXuvWrdO8efPyoy4AAJCH7hsANm7caHzv5uam6OhoVatWLU+LAgAAeeu+AeD48eOSJKvVqp9//lnFihVTly5d8rwwAACQd+4bAEaNGmV8b7VaNWjQoDwtCAAA5L37BoDU1FTj+/j4eP322295WhAAAMh79w0A7du3l8VikdVqlaOjo/r3758fdQEAgDx03wAwf/581atXz3i8b9++PC0IAADkvRwDwP79+3Xs2DGtWLFCffv2lSRlZmbq448/1qZNm/KtQAAA8OjlGADc3d116dIlpaamKj4+XpJksVg0ZsyYfCsOAADkjRwDQI0aNVSjRg11795dV65cUa1atbRt2zY1a9YsP+sDAAB54L63Av7nP/+pH3/8UZJ08uRJjR8/Ps+LAgAAeeu+AeDChQvq0aOHJGnAgAG6ePFinhcFAADy1n0DgHT7yF+STp06pczMzDwtCAAA5L37vg1w4sSJCgwM1OXLl+Xo6Kh//OMf+VEXAADIQ/edAahfv75mzpypZs2aKSkpSZcvX86PugAAQB7KcQYgNTVVmzdv1scffywHBwclJCQoMjJSjo6O+VkfAADIAznOALRp00a//PKL5s6dq7CwMJUuXZqdPwAAfxE5zgD07t1bmzZt0tmzZ9WtWzdZrdb8rAsAAOShHGcABg4cqI0bNyogIECbNm1SdHS0QkJCFBsbm5/1AQCAPHDfiwC9vb0VEhKir7/+WmXLltXYsWPzoy4AAJCHHug+ANLtzwYICAjQ+vXr87IeAACQDx44AAAAgL8OAgAAACZEAAAAwIQIAAAAmBABAAAAEyIAAABgQgQAAABMiAAAAIAJEQAAADAhAgAAACZEAAAAwIQIAAAAmBABAAAAEyIAAABgQgQAAABMiAAAAIAJEQAAADAhAgAAACZEAAAAwIQIAAAAFFKXL19Wy5Ytdfz4cZ06dUo9evRQz549NW3aNGVmZua6LQEAAIBCKC0tTVOnTpWjo6MkKTg4WIGBgQoLC5PValVkZGSu2xMAAAAohObMmaOXX35ZpUuXliTFxMTI29tbkuTr66s9e/bkuj0BAACAx1BERIS6du1qfEVERBjrPv/8cxUvXlwtWrQwllmtVlksFkmSi4uLbt68mWv/dnlTNgAA+L/w9/eXv79/tus+++wzWSwW7d27Vz///LPGjRunK1euGOsTExPl7u6ea/8EAAAACpmPP/7Y+D4gIEBBQUEKCQlRVFSUmjRpop07d+qZZ57JtQ9OAQAA8Bcwbtw4hYaGyt/fX2lpafLz88u1PTMAAAAUYqtWrTK+X7169QNvxwwAAAAmRAAAAMCECAAAAJgQAQAAABMiAAAAYEIEAAAATIgAAACACXEfAAAAHlLmjst5PoZNyzzuP2+7BwAAjyMCAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJmRX0AUgf2RkZmrqR3t08sIN2dhYNLtPcyWmpOmfn0TJxsYiBztbvdWvhUq6OxnbZGZaNSNsr47+dlUOdjaa2bu5Kpd2148nLio4fJ9sbW3UvPbf9EanBjm2BQqjjLQMRS2NUsLFBNk72curj5fcyroZ649tP6bj24/LYmNRnRfqqHzD8kq5maI9C/coIzVDTh5OajKgieyK2GXbFngcEABMYvuhM5KksHEdtO+X85qz9gfduJWqST2aqFbFEorY8YuWbf1J41/yNrbZ9uNppaRlKHz83/XjiYt6e+0P+tcbz2r66r16b3BrVSzlpkGh2xRz6rLOXk7Iti1QGB3/9rjsitipXVA73Th/Q/s/2q/WY1tLkpKuJSn2q1j5zfBTRlqGts3cprJ1yyp6XbQqN60sT19PHfnPER375pgqN62cbVtbe9sCfoYAAcA0nmtYWa3qVZQknbucqBLuTpr2SlOVLuYsSUrPzFSRu/4p/ffYBfnUuX200sCztKJPXVZCUqpS0zNV6f8f3fvUKa/vj55X/PVb97QFCqvrZ6+rXP1ykiT3cu66ce6Gse7yicsqVaOUbO1tZWtvK7cybrp25priY+NVu3NtSVK5euV0eO1huZZxzbZtCc8SBfK8gD/iGgATsbO10fh/79Ks8Cj5Naps7PwPHr+osO1H1ee5OlnaJySnyc3JwXhsa7EoITlNro72xjKXIva6mZSabdv0jMw8fkZA3vCo5KFzB8/JarXq0rFLSrqSpMzM27/PaUlpsnf639+AnaOd0m6l3V7ufHu5vZO9UpNSc2wLPA4e+QxAQECA0tKy/oJbrVZZLBaFh4c/6uHwkN7q20LxXW/JP3izNk3vom8P/6bFWw7rgzefU3E3xyxtXR3tlZjyv59lptV6z7LElDS5OzsoOTX9nrZ2tuRLFE6eLT1149wNfTP7G5WsUVIeVTxkY3P799neyV5pyf/7XU9PTpe9s73sneyVnpQuOwc7pSWlycHZIce2wOPgkQeA0aNHa/LkyfrXv/4lW1vOcz0uNuw9rgvXEjXw+XpycrCTjcWibQdPa83OX7RydHsVcylyzzaNqpbW9sNn9LxXFf144qJqlPeQq5OD7G1tdPriDVUs5abdMWf1RscGunA18Z62QGF15cQVlapZSo16NdLlE5eVcCHBWFfCs4QOrz2sjNQMZaRn6Pq56ypWoZhK1iipc4fOydPXU+cPn1epmqVybAs8DixWq9X6qDtdtmyZKleurLZt2/6p7TN3BD/iinArJU2TVnyn+BtJSs/I1ID2T2nSit0qV9xFbs63p+4b1yirNzs31LgPd2n4Cw1V1sNFM8L26pffrsoqaXaf5vIsV+z2uwAi9ikz06rmtcsr8B+NjHcB3N0Wj5ZNywmSpKB9QQVbyF9cys0Uffev75Seki4HZwd59/fW6e9Py7WMqyo0qmBc2W+1WlWncx1VbFxRSdeT9P3i75WenK4ibkXUbEgz2TnaZdsWeSfIOyhfxsmP/dSdv/e8kicB4P+KAABkjwAA5I4A8BD952nvAADgsUQAAADAhAgAAACYEAEAAAATIgAAAGBC3AoYAIBCJiMjQ5MnT9bJkydla2ur4OBgWa1WjR8/XhaLRdWrV9e0adOMG1hlhwAAAEAhs337dklSeHi4oqKijAAQGBioJk2aaOrUqYqMjMz1fjycAgAAoJB57rnnNHPmTEnSuXPnVLJkScXExMjb+/Ynuvr6+mrPnj259sEMAAAAj6GIiAhFREQYj/39/eXv7288trOz07hx4/T1119rwYIF2r59uywWiyTJxcVFN2/ezLV/AgAAAI+hu3f42ZkzZ45Gjx6tl156SSkpKcbyxMREubu757otpwAAAChk1q9fr8WLF0uSnJycZLFYVLduXUVFRUmSdu7cKS8vr1z7YAYAAIBCpl27dpowYYJeeeUVpaena+LEiapataqmTJmid999V56envLz88u1DwIAAACFjLOzs9577717lq9evfqB++AUAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCvAsAAICHtONK5zwfo3Ue988MAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJmRX0AUAAICHk5aWpokTJ+rs2bNKTU3VkCFDVK1aNY0fP14Wi0XVq1fXtGnTZGOT83E+AQAAgEJm48aNKlasmEJCQnT16lX94x//0JNPPqnAwEA1adJEU6dOVWRkpNq2bZtjH5wCAACgkGnfvr2GDx9uPLa1tVVMTIy8vb0lSb6+vtqzZ0+ufRAAAAB4DEVERKhr167GV0REhLHOxcVFrq6uSkhI0LBhwxQYGCir1SqLxWKsv3nzZq79cwoAAIDHkL+/v/z9/XNcf/78eb3xxhvq2bOnOnXqpJCQEGNdYmKi3N3dc+2fGQAAAAqZS5cuqV+/fhozZoy6desmSapdu7aioqIkSTt37pSXl1eufRAAAAAoZD744APduHFDCxcuVEBAgAICAhQYGKjQ0FD5+/srLS1Nfn5+ufbBKQAAAAqZyZMna/LkyfcsX7169QP3wQwAAAAmRAAAAMCECAAAAJgQ1wAAAPCQdpRfm+djtFadPO2fGQAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAEyIAAAAgAkRAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATsivoArJj03JCQZcAPNaCvIMKugQAhRwzAAAAmBABAAAAEyIAAABgQgQAAABMiAAAAIAJEQAAADAhAgAAACZEAAAAwIQIAMhWZmampk6dKn9/fwUEBOjUqVMFXRLwWDp06JACAgIKugzgoT2WdwJEwdu2bZtSU1MVERGhH3/8UW+99ZYWLVpU0GUBj5WlS5dq48aNcnJyKuhSgIfGDACydeDAAbVo0UKS1KBBA0VHRxdwRcDjp1KlSgoNDS3oMoA/hQCAbCUkJMjV1dV4bGtrq/T09AKsCHj8+Pn5yc6OiVQUTgQAZMvV1VWJiYnG48zMTP7RAcBfCAEA2WrUqJF27twpSfrxxx9Vo0aNAq4IAPAocUiHbLVt21bfffedXn75ZVmtVs2ePbugSwIAPEIWq9VqLegiAAAoTIL2BeX9GN55OwanAAAAMCECAAAAJkQAAADAhAgAAACYEAEAAAATIgAADyEqKkpNmzZVQECAAgIC9NJLL2nVqlV/qq+5c+fq888/188//6z3338/x3Zff/21Lly48EB97ty5U+PHj7+n5hEjRuS4zeeff665c+c+UP8P0xbA4437AAAP6ZlnntG8efMkSampqWrfvr1eeOEFubu7/6n+atWqpVq1auW4/qOPPlJQUJDKlCnzp/oHgOwQAID/g4SEBNnY2MjW1lYBAQHy8PDQjRs3tGTJEgUFBenUqVPKzMxUYGCgmjRpoi+//FKLFi1S8eLFlZaWJk9PT0VFRSk8PFzz5s3T2rVr9cknnygzM1PPPvusnnrqKf38888aN26cwsLCFBERoU2bNslisahDhw7q3bu3jh8/rokTJ8rJyUlOTk4qWrRojvWuXr1aX331ldLT0+Xm5mZ8kM2PP/6oPn36KCEhQW+++aZatWqlffv2ad68ebK1tVXFihU1Y8aM/HpZAeQDAgDwkL7//nsFBATIYrHI3t5eU6ZMkYuLiySpU6dOatu2rcLCwuTh4aHZs2fr6tWr6tWrlzZv3qyQkBCtXbtWxYoV08CBA7P0e/nyZePjZR0cHPTWW2+pcePGqlWrloKCgnT69Glt2bJFYWFhslgsevXVV+Xj46P33ntPw4YNU/PmzbVkyRKdOHEi27ozMzN17do1rVixQjY2Nnrttdf0008/SZKcnJy0ZMkSXblyRd27d1eLFi00ZcoUhYWFqUSJEpo/f77WrVvH50EAfyH8NQMP6Y+nAO5WpUoVSVJsbKwOHDigw4cPS5LS09N16dIlubq6ysPDQ5LUsGHDLNueOXNG1atXl6OjoyRp4sSJWdbHxsbq3LlzevXVVyVJ169f1+nTp/Xrr7+qXr16km5/hkNOAcDGxkb29vYaOXKknJ2d9fvvvxuf8Pj000/LYrGoRIkScnNz09WrV3Xx4kUFBgZKkpKTk9W8eXNVqlTpoV4rAI8vAgDwCFksFkmSp6enypYtq8GDBys5OVmLFi2Su7u7bt68qStXrqh48eL66aefVLZsWWPbSpUq6cSJE0pNTZWDg4OGDRumSZMmyWKxyGq1ytPTU9WqVdOyZctksVi0YsUK1ahRQ56enjp48KB8fX0VHR2dY21Hjx7Vtm3btHbtWiUlJalr1666cyfwOzMB8fHxunXrljw8PFS2bFktXLhQbm5uioyMlLOzs86fP5+Hrx6A/EQAAPLAyy+/rMmTJ6tXr15KSEhQz5495eDgoODgYL322msqWrToPdPpxYsX14ABA9SrVy9ZLBa1bt1aZcqUUcOGDTV27Fh9+OGHatq0qXr06KHU1FTVq1dPZcqU0bRp0zRixAgtX75cxYsXV5EiRbKtqXLlynJyclLXrl3l4OCgUqVK6eLFi5JuH+H37t1bt27d0owZM2Rra6tJkyZp4MCBslqtcnFx0dtvv00AAP5C+DAgAAAeEh8GBAAACiUCAAAAJkQAAADAhAgAAACYEAEAAAATIgAAAGBCBAAAAAqpQ4cOKSAgQJJ06tQp9ejRQz179tS0adOUmZmZ67YEAAAACqGlS5dq8uTJSklJkSQFBwcrMDBQYWFhslqtioyMzHV7AgAAAIVQpUqVjE/0lKSYmBh5e3tLknx9fbVnz55ct+dWwAAAPIYiIiIUERFhPPb395e/v7/x2M/PT7/99pvx2Gq1Gp9H4uLiops3b+baPwEAAIDH0N07/PuxsfnfpH5iYqLc3d1zb/+nKwMAAI+N2rVrKyoqSpK0c+dOeXl55dqeAAAAwF/AuHHjFBoaKn9/f6WlpcnPzy/X9nwaIAAAD4lPAwQAAIUSMwAAAJgQMwAAAJgQAQAAABMiAAAAYEIEAAAATIgAAACACREAAAAwof8HqMWNJZxU86gAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 648x648 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm4 = metrics.confusion_matrix(y_test, y_pred_knn)\n",
    "plt.figure(figsize=(9,9))\n",
    "sns.heatmap(cm4, annot=True, fmt=\".3f\", linewidths=.5, square = True, cmap = 'Accent');\n",
    "plt.ylabel('Actual label');\n",
    "plt.xlabel('Predicted label');\n",
    "all_sample_title = 'Accuracy Score: {0}'.format(score)\n",
    "plt.title(all_sample_title, size = 15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Decision Tree Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "dtree = DecisionTreeClassifier(max_depth = None , random_state = 1 , max_features = None, min_samples_leaf =20)\n",
    "dtree.fit(X_train,y_train)\n",
    "y_pred_dtree = dtree.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Score of the model is -  0.7517241379310344\n",
      "Report card of this model - \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.798     0.875     0.835       104\n",
      "           1      0.581     0.439     0.500        41\n",
      "\n",
      "    accuracy                          0.752       145\n",
      "   macro avg      0.689     0.657     0.667       145\n",
      "weighted avg      0.737     0.752     0.740       145\n",
      "\n",
      "Accuracy score -  0.7517241379310344\n"
     ]
    }
   ],
   "source": [
    "score = dtree.score(X_test, y_test)\n",
    "print(\"Score of the model is - \",score)\n",
    "print(\"Report card of this model - \")\n",
    "print(metrics.classification_report(y_test, y_pred_dtree, digits=3))\n",
    "print(\"Accuracy score - \", metrics.accuracy_score(y_test,y_pred_dtree))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set ROC AUC score: 0.657\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "test_roc_auc = roc_auc_score(y_test, y_pred_dtree)\n",
    "\n",
    "# Print test_roc_auc\n",
    "print('Test set ROC AUC score: {:.3f}'.format(test_roc_auc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1, 'Accuracy Score: 0.7517241379310344')"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x648 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm5 = metrics.confusion_matrix(y_test, y_pred_dtree)\n",
    "plt.figure(figsize=(9,9))\n",
    "sns.heatmap(cm5, annot=True, fmt=\".3f\", linewidths=.5, square = True, cmap = 'viridis');\n",
    "plt.ylabel('Actual label');\n",
    "plt.xlabel('Predicted label');\n",
    "all_sample_title = 'Accuracy Score: {0}'.format(score)\n",
    "plt.title(all_sample_title, size = 15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Random Forest \n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "hist1 = []\n",
    "for i in range(1,10):\n",
    "    clf = RandomForestClassifier(n_estimators=80, max_depth=i, random_state=0)\n",
    "    cross_val = cross_val_score(clf, X_train, y_train, cv=5)\n",
    "    hist1.append(np.mean(cross_val))\n",
    "plt.plot(hist1)\n",
    "plt.title('Cross Validations score for RandomForestClassifier')\n",
    "plt.xlabel('Max_depth')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.grid()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "ran_for = RandomForestClassifier(n_estimators=80, max_depth=8, random_state=0)\n",
    "ran_for.fit(X_train,y_train)\n",
    "y_pred_ran = ran_for.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Score of the model is -  0.696551724137931\n",
      "Report card of this model - \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.759     0.846     0.800       104\n",
      "           1      0.448     0.317     0.371        41\n",
      "\n",
      "    accuracy                          0.697       145\n",
      "   macro avg      0.603     0.582     0.586       145\n",
      "weighted avg      0.671     0.697     0.679       145\n",
      "\n",
      "Accuracy score -  0.696551724137931\n"
     ]
    }
   ],
   "source": [
    "score = ran_for.score(X_test, y_test)\n",
    "print(\"Score of the model is - \",score)\n",
    "print(\"Report card of this model - \")\n",
    "print(metrics.classification_report(y_test, y_pred_ran, digits=3))\n",
    "print(\"Accuracy score - \", metrics.accuracy_score(y_test,y_pred_ran))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set ROC AUC score: 0.582\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "test_roc_auc = roc_auc_score(y_test, y_pred_ran)\n",
    "\n",
    "# Print test_roc_auc\n",
    "print('Test set ROC AUC score: {:.3f}'.format(test_roc_auc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1, 'Accuracy Score: 0.696551724137931')"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x648 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm6 = metrics.confusion_matrix(y_test, y_pred_ran)\n",
    "plt.figure(figsize=(9,9))\n",
    "sns.heatmap(cm6, annot=True, fmt=\".3f\", linewidths=.5, square = True, cmap = 'viridis');\n",
    "plt.ylabel('Actual label');\n",
    "plt.xlabel('Predicted label');\n",
    "all_sample_title = 'Accuracy Score: {0}'.format(score)\n",
    "plt.title(all_sample_title, size = 15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Support Vector Machine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAESCAYAAADjS5I+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO3deVxU5f7A8c8MMCyyiQoiiimK4oKIS6VAqWnmiqCCJmVqZb/q3kpLLTUlF1xarubSioWVIJpJXTO1krS8iYqm4k4KLoAb+zDAnN8fXqdLyirDsHzfr1evmDnnPOc7j6NfznnO831UiqIoCCGEEKVQmzoAIYQQtZskCiGEEGWSRCGEEKJMkiiEEEKUSRKFEEKIMkmiEEIIUSZJFKJUxcXFREZGEhQUxMiRIxkyZAjLli1Dp9PVWAwpKSl06dKFtLS0O7YNHz6cHTt2lHl8hw4duH79Ort27WLBggV33WfYsGH85z//KTeOF198EYC0tDRCQ0Mr+AnqtjVr1vDwww8za9asKrdRWFjI0qVLGT58OCNGjGD48OGsXbsWRVH4/fff6dWrF1qttsQxOp2O+++/n+PHjwNw9uxZXnzxRUMbEyZMICEh4Z4+m6gERYhSzJ49W3nxxReVrKwsRVEUJTc3V3nuueeU6dOn12gcTz/9tLJ69eoS7x08eFB56KGHlKKiojKP9fT0VK5du1bmPkOHDlX27dtX5j779u1Thg4dWrGA65H+/fsr+/fvv6c2PvroI+XFF19UCgsLFUVRlKysLGXUqFHKhg0bFEW51f/ffPNNiWPi4uKUkJAQRVEU5ezZs0rfvn2V+Ph4w/Zff/1V6dGjh3Lq1Kl7ik1UjLmpE5WonVJTU4mLi2PPnj3Y2toCYGNjw/z58zl48CAAM2fO5ObNm6SkpPDwww8zdepU5s+fz4kTJ1CpVPj7+/PKK69gbm7OihUr2LFjBxYWFjRu3JjFixfj7Oxc6vv/6/HHH2fBggVMnToVlUoFQExMDKGhoZiZmZGcnEx4eDi5ublkZGTQsWNH3nvvPSwtLQ1tbN68me3bt/PBBx9w5swZXn/9dfLz82nbti15eXmG/dauXcuuXbvQarXk5+czY8YM+vfvz+zZs0lLS2Py5MnMnz+f4cOHc+jQIQoLC4mIiOC3337DzMwMb29vZs2aha2tLf3792fUqFH89ttvXL58mZEjR/LSSy+Rm5vLrFmzOH/+PGq1ms6dOxMeHo5aXfIC/8svv2TDhg1YWFhgaWlJeHg47dq1Izk5mblz53L9+nXUajXPPfccQ4YM4fTp04SHh3Pz5k1UKhWTJk0iMDCQ//znPyxcuBAbGxtyc3PZtGkTe/bsYc2aNRQWFmJlZcWMGTPo3r17ifO/9NJLpKWl8cYbb/DPf/4TX19f5s2bx8WLF1EUhcDAQKZMmUJqaiqPP/44Hh4eXLx4kaioqBJ/hhkZGRQWFqLT6TA3N8fOzo6lS5ei1+sBGDduHJs2bWLEiBGGY6Kjo3n88ccB+OijjwgODsbf39+w/cEHH+Ttt9/Gysqqkt9sUSWmzlSidvr++++V4ODgMveZMWOG8uSTTxpev/baa8pbb72l6PV6paCgQJk0aZLywQcfKJcuXVJ8fX2VgoICRVEU5ZNPPlF27NhR6vt/V1xcrAwYMMDwW39WVpbSq1cv5erVq4qiKEpERISyZcsWRVEURafTKcOGDVO+//57RVH+uqLYtGmT8swzzyiKoigjR45UYmJiFEVRlISEBKVDhw7Kvn37lNTUVCUsLEzJz89XFEVRvv32W2XYsGGKopS8okhJSVF8fHwURVGUf/3rX8oLL7yg6HQ6pbi4WJk5c6YyZ84cRVEUpV+/fkpERISiKIpy5coVpWvXrsqFCxeUr7/+Wpk0aZKiKIpSVFSkvPHGG8qff/5Z4jMXFRUpnTt3VtLS0hRFUZSvv/7a8Bt4YGCgsn79ekVRFOXSpUvKgAEDlOzsbGXAgAHK9u3bDefz9/dXDh48qOzbt0/p2LGjkpqaqiiKoiQnJyvDhg1Trl+/riiKopw6dUrp27evkpube0ff9+vXTzly5IiiKIry+OOPK59++qnhz2D48OHKt99+q6SkpCienp6lXnlcvnxZGTVqlNK1a1dlwoQJyjvvvKMcO3bMsD07O1vx9fVVLly4YIivb9++hu/FsGHDlJ9//vmubYuaIWMU4q7UarXhN76y9OjRw/BzfHw8EyZMQKVSodFoCA0NJT4+HhcXFzp27MioUaNYsmQJXl5ePPLII6W+f7dYQkND2bRpEwBbt27loYceokmTJgC8+uqrODk58dFHHzFv3jzS09NLXCX8rxs3bnDy5EkCAwMN8bdv3x4ANzc3li5dSlxcHMuXL2fDhg3k5uaW+fnj4+MJDQ3FwsICtVpNWFgYv/zyi2H7gAEDAHBxcaFJkyZkZmbSo0cPzpw5Q1hYGB9++CFPPvkkrVu3LtGumZkZgwcPJjQ0lPDwcOzt7Rk9ejQ3b97kxIkTjBkzBgBXV1d27tzJlStXKCgoYNCgQYbzDRo0yBCLq6srbm5uAOzdu5f09HQmTpzIyJEjmT59OiqVigsXLpT6OfPy8jh48KDht3w7OzuCgoKIj48HwNzcHB8fn7se27x5czZv3szmzZsZPHgwycnJhISE8MUXXwBga2vLiBEj2Lx5M3DramL06NFoNBoAVCpVhb6LwngkUYi78vb25ty5c+Tk5JR4Py0tjWeeecYw+GhjY2PYptfrDbeGbr8uKipCrVazfv16Fi9ejKOjI4sWLWLp0qWlvn83wcHB7N69m5ycHGJiYgz/YAG88sorxMTE4ObmxsSJE+ncuTNKOSXM/ne7ufmtO7DHjh0jJCSEnJwc+vbty5QpU8rtp7t95sLCQsPr/739pVKpUBSFVq1asWPHDp555hlycnJ46qmn+PHHH+9oe/ny5axduxZ3d3c+/PBDw228223ddu7cOYqLi0u8d/szFhUVAXf+OT344IN88803hv9iYmIMCbO0z/n3Pr395wug0WgMsf3d0qVLSU5Opl27djz++OOsWLGCBQsW8NVXXxn2GT9+PF9//TUFBQXExcWVeFjAx8eHxMTEO9p9//332bp1a6kxi+ojiULclYuLC8OHD+f11183JIucnBzmzZuHo6PjXe8N+/n5sX79ehRFQafTERMTQ58+fThx4gTDhg3Dw8ODZ599lokTJ/LHH3+U+v7dNG7cmH79+rFixQrMzMxK/Pa6Z88enn/+eYYMGQLA4cOHKS4uLrWdzp07s3HjRuBWcjh16hQA+/fvp0uXLjz11FP07t2bXbt2GdoxMzMrkQBu8/f356uvvqKwsBC9Xs8XX3xB3759y+zbL7/8klmzZuHn58err76Kn5+f4eme265fv85DDz2Eo6MjEydO5KWXXuKPP/7A1taWzp07s2XLFgAuX77MuHHjsLe3x9zcnB9++AG4ldC3b99Onz597jj/gw8+yN69ezl79iwAu3fvZsSIEXc8efS/bG1t6datm+EqIDs7my1btty1/b+7fv06//rXv8jPzwduJbDTp0/TqVMnwz7t27enVatWvP322/j6+tK8eXPDtsmTJ7Nx40b27NljeC8+Pp6oqCg6duxY7vnFvZPBbFGqN998k9WrVxsGjXU6HY888ojhMdG/mz17NgsWLGD48OEUFhbi7+/P1KlT0Wg0PPbYYwQHB2NjY4OVlRWzZ8+mY8eOd32/NOPHj2fs2LEsXLiwxPsvv/wyzz//PDY2Ntja2tKrV68yb6O88847zJo1iw0bNuDu7k7btm2BW4/J/vDDDzz22GPo9Xr69etHZmYmOTk5tGvXDktLS0aPHs27775raOu5555jyZIlBAYGUlRUhLe3N3PmzCmzXwMDA/n9998ZMmQI1tbWuLq6EhYWVmIfJycnnnvuOSZOnIiVlRVmZmaGx3vffvtt5s+fT1RUFCqVioULF+Lq6srq1atZsGABK1eupLi4mOeff54HHnjgjkd/27VrR3h4OK+88gqKomBubs6aNWto1KhRmXEvX76c8PBwNm/ejE6nY/jw4QQFBXHx4sUyj3vzzTd59913GTFiBBqNhqKiIh544AHmzp1bYr/x48fz8ssvs27duhLvt27dmrVr1/Lee++xZMkS9Ho9Tk5OrFmzBk9PzzLPLaqHSinvGl0IIUSDJreehBBClEkShRBCiDJJohBCCFEmSRRCCCHKVO+eerr//vsNE4sqS6fTGSb5iIqRPqsc6a/Kkf6qnHvpr4sXL5ZaHLPeJQo3NzfDDM/KSkpKwsvLq5ojqt+kzypH+qtypL8q5176KygoqNRtcutJCCFEmSRRCCGEKJMkCiGEEGWSRCGEEKJMkiiEEEKUSRKFEEKIMkmiEEIIUaZ6N49CCFH3KYrCd39cxkFXZOpQBJIohBC1UGLKTV748hBu9hbEeXXAqZHMzjYlufUkhKh1YhJSsTRXk55TxJTP9pOvu/uKhaJmGOWKQq/XM2/ePE6ePIlGo2HBggWGxeOTkpJYtGiRYd/ExERWrVpFu3bteP311ykuLkZRFMLDw2nbti0//vgjq1atwtzcnODgYMaOHWuMkIUQtUSeroi4w5cY5t0CL/tCFu5O458bDrFmQg/M1KryGxDVziiJYufOneh0OqKjo0lMTCQiIoI1a9YA4OXlRVRUFADbtm3D2dmZgIAAZsyYwYQJE3jkkUf45ZdfeOedd3j33XdZvHgxsbGxWFtbM27cOPr160ezZs2MEbYQohbY9scVcgqKGNuzJfYF6bw5rBPz4o4zb+sxwkd2RqWSZFHTjJIoDhw4gL+/PwA+Pj4cPXr0jn3y8vJYuXIl69evB2DGjBnY2dkBUFxcjKWlJWfPnsXd3R0HBwcAevToQUJCAo899lip59bpdCQlJVUpbq1WW+VjGyrps8qR/ipfZPwlWtiZY6dNQ1tQwP1OMLqLA1H7zmOuy2ZsV0dTh1hrGev7ZZREkZOTg62treG1mZkZRUVFmJv/dbrY2FgGDx6Mk5MTgOH/586dY8mSJaxatYrr168bkgdAo0aNyMnJKfPcGo2mytUTpVJl5UmfVY70V9mSr+ZyNO0crz7agU6d2hn6a2kHhaKYRCIPXqJru1YE+bY0dai1krG+X0YZzLa1tSU3N9fwWq/Xl0gSAHFxcYwZM6bEe/v27eP5559n6dKltG3b9o52cnNzSyQOIUT9sjEhBbUKRvcomQjUahVLR3ejj0cTXos9wi+nM0wUYcNklETh6+tLfHw8cGuw2tPTs8T27OxsdDodrq6uhvf27dvHwoUL+fjjj+natSsAHh4enD9/nps3b6LT6UhISKB79+7GCFkIYWJFxXpiD6TycAdnXOyt7tiuMVezNqwH7ZxtmRp1gKMXM00QZcNklFtPAwcOZO/evYSGhqIoCosWLSIyMhJ3d3cGDBhAcnLyHavQLVq0iMLCQmbOnAlAmzZtCA8PZ+bMmUyePBlFUQgODsbFxcUYIQshTCz+dAbp2QWM7dmq1H3srSz4bFJvRq3ay1Pr9rP5uT60crKpwSgbJqMkCrVaTXh4eIn3PDw8DD97e3uzevXqEtu3bt1617b69+9P//79qz9IIUStEr0/hSaNNPTv6Fzmfi72Vnw2qTfBa35lYuTvbHquD442MiHPmGTCnRDC5K7mFLArKZ0gXzc05uX/s9TexY6Pn+xFyo18pnyWgLZQJuQZkyQKIYTJfX3wIkV6pczbTn/Xu40T74X4cODCDf654RDFesWIETZskiiEECalKAoxCSl0d3ekvUvlnmoc0tWVOUM7sf1YGuFxx1AUSRbGIEUBhRAmdSjlJqfTc4gI6lql4yf5teFyZj4f/ZKMq6M1Ux/yKP8gUSmSKIQQJrUxIQVrCzOGeruWv3MpZj3mxZWsAiK2naC5vRWB3d3KP0hUmCQKIYTJ3CoAeJmh3q7YWVlUuR21WsXyMd5czS7g1djDNLOzpG+7ptUYacMmYxRCCJP5t6EAYMUHsUtjaW7G2rAetG1qy7NRBzh+KasaIhQgiUIIYUIx+1No07QRve5rXC3tOVhbsG5SL+yszJkY+TupN/Kqpd2GThKFEMIkzmXk8Puf1xnTs2W1lg53dbDms0m9yS8sZmLkfm7m6aqt7YZKEoUQwiQ2HkjFTK1itBEqwXq62PHREz25cC2PZz4/IBPy7pEkCiFEjSsq1rPpQCoPezbD+S4FAKvDA22b8E5IN37/8zqvxCTKhLx7IE89CSFq3O5T/y0A2OveB7HLMsy7BVcytSz4Lglnu+O8ObyTrJBXBZIohBA1Lnp/Ck1tyy8AWB2m+LflcqaWT/Yk08LRimcCZEJeZUmiEELUqIzsAn48kc4kvzZYmNXM3e83hnhxJUvLon+fwMXeipE+MiGvMiRRCCFq1NeHUv9bALDmljNVq1W8PaYbV7MLmL7xMM1sLekjE/IqTAazhRA15lYBwFR83R1p51yzyxpbWZjx4RM9adO0Ec9GHSDpskzIqyhJFEKIGnPwwk3OpOcQYuRB7NI4WFuw7qneNLI056nI/Vy6mW+SOOoaSRRCiBqzMSEFG40ZQ71bmCyGFo7WrJvUi9yCIiZG/k5mXqHJYqkrJFEIIWpEbkERcYcvMbSrK7aWph0e7djcng+e6MGfV/N4OkpWyCuPJAohRI347o/L5OqKjT53oqL6eDRl+dhu/J58nWkbD6OXCXmlkqeehBA1YmNCCm2bNqJn6+opAFgdRnRrQVqmloX/TqK5vRVzhnUydUi1kiQKIYTRnc3IYf+fN5gxuGOtmxk9xb8NlzLz+WRPMq4OVkzxb2vqkGodSRRCCKPbmHCrAGCwb+2b6KZSqZgztBNpWbdKfbjYWzG8m+kG22sjGaMQQhhVUbGeTQdT6dfBeAUA75VareKdsT70vs+JaTGH+e3sNVOHVKtIohBCGNXPJzPIyC6ollXsjMnKwoyPnuhJ6yY2PBOVwIkrMiHvNqMkCr1ez9y5cwkJCSEsLIzz588btiUlJREWFmb4r2vXrsTHxxu2r1u3juXLlxteb926lVGjRhEcHMyXX35pjHCFEEYUnZBCU1tL+tVAAcB75WBjwbpJvbHRmDHx0/1czpQJeWCkRLFz5050Oh3R0dFMmzaNiIgIwzYvLy+ioqKIiopi/PjxDBo0iICAALRaLdOnT78jGSxdupTIyEi++uorIiMjyczMNEbIQggjSM/W8uOJdIJ93WqsAOC9cnO0JnJib3IKipj46X4y82VCnlEGsw8cOIC/vz8APj4+HD169I598vLyWLlyJevXrwegoKCAwMBA+vTpw7lz5wz7dejQgezsbMzNzVEUpdwnJnQ6HUlJSVWKW6vVVvnYhkr6rHIaWn/FHr1JsV6hh1PV/l6aqr9UwOsPNWPuzstMWBvPgoGuaMxq19Nad2Os/jJKosjJycHW1tbw2szMjKKiIszN/zpdbGwsgwcPxsnJCQAHBwf8/PzYvHlzibbat29PcHAw1tbWDBw4EHt7+zLPrdFo8PLyqlLcSUlJVT62oZI+q5yG1F+KovDzv3fTo3VjBj3QrUptmLK/vLygkdNF/rkhkY+OaFkR2h21unYnC2P1l1GuBW1tbcnNzTW81uv1JZIEQFxcHGPGjCmznRMnTvDzzz+za9cufvzxR65fv862bduMEbIQopodvHCDsxm5hNTyQeyyjPRxY+ZjHfn2yGUWb2s4V4J/Z5RE4evraxigTkxMxNPTs8T27OxsdDodrq6uZbZjZ2eHlZUVlpaWmJmZ4eTkRFaWPIkgRF0Qvf9WAcAh3mX/Pa/tng1oy8Q+9/HRL8l8sifZ1OGYhFFuPQ0cOJC9e/cSGhqKoigsWrSIyMhI3N3dGTBgAMnJybi5lT/xxs3NjZCQEMaPH4+FhQXu7u6MGjXKGCELIapRbkER3x65zDBv0xcAvFcqlYo5wzr9d+3t4zS3t2JoHU9+lWWUP0G1Wk14eHiJ9zw8/lqn1tvbm9WrV9/12KCgoBKvx40bx7hx46o/SCGE0Xx35DJ5uuJaP3eioszUKt4L9WHCx//h5ehEmtpquL9tE1OHVWPqxvNqQog6JSYhhbbNGtGjFhUAvFdWFmZ8/GRPWjlZ8/TnCZxKyzZ1SDVGEoUQolqdSc8h4fwNxvZsVesKAN4rRxsNn03qjZWFGU9++jtXMrWmDqlGSKIQQlSrjQdSMFOrCKqFBQCrQ8vGNkQ+1Yts7a0V8rK09X9CniQKIUS1KSzWs+nARfp1cMbZrnYWAKwOnVs4sGaCL2fSc3j28wMUFNXvFfIkUQghqs3PJzO4mlNASC1Zxc6Y/Ns3Y+lob347d41XNx6p1yvk1e3n1oQQtUr0/hSa2VnSr0MzU4dSI4J8W3IlS8vS70/i6mDFrCH1c9a9JAohRLVIz9by08l0pvi3wbyOFACsDs895MHlm1o+iD9HcwcrnurbxtQhVTtJFEKIarH54EWK9Uq9mTtRUSqVinkjOpOWpSX821sT8h7rWr8m5DWctC+EMBpFUYjZn0LP1o3xaGZb/gH1jJlaxYpx3fF1b8w/oxP5Pfm6qUOqVpIohBD37MD5G5y7msvYBjCIXRorCzM+fqInLRvfmpB3Jr3+TMiTRCGEuGfR+1NopDFjaD275VJZjRtp+Oyp3mjM1Tz56X7SsurHhDxJFEKIe5JTUMR3f1xmmHcLGtXxAoDVoZWTDZETe3EzT8eTn/5Odj2YkCeJQghxT747culWAcBeLU0dSq3Rxc2BNRN6cCY9h6nrD6Ar0ps6pHsiiUIIcU9iElLxaNYIX/f6UwCwOgR4NiMi2Ju9Z67xWuzhOj0hTxKFEKLKzqRnc6CeFgCsDqN7tOTVRzuwJfESS7efNHU4VSY3FIUQVbYxIRVztYogX7ntVJr/e9iDSzfzWbv7LC0crXjiwftMHVKlSaIQQlRJYbGeTQdT6d/RmWZ2lqYOp9ZSqVSEj+xCWlYBb249hrOdFYO7NDd1WJUit56EEFXy04l0ruboGtxM7KowU6tYOa47Pq0c+eeGQyT8Wbcm5EmiEEJUSUzCrQKADzeQAoD3ylpjxidP9qKFozWTP0vgTHqOqUOqMEkUQohKS8/S8tPJDIJ9WzaoAoD3yum/E/IszFQ8+envpNeRCXnyJyyEqLRNhgKAMohdWe5NbIic2JsbeTomRu6vExPyJFEIISpFURQ2JqTQ677GtG2ABQCrQ9eWDqx+3JeTadn83xcHa/2EPEkUQohKSbhdAFAGse/Jwx2ciQjqyi+nrzJz8xEUpfZOyJPHY4UQlXK7AOCQBl4AsDqM6dmKy5la3tlxClcHK159tKOpQ7oroyQKvV7PvHnzOHnyJBqNhgULFtC6dWsAkpKSWLRokWHfxMREVq1aRUBAAADr1q3j6tWrTJ8+HYAjR44QERGBoig0a9aMZcuWYWkpz2wLYQo5BUV8d+QyI32kAGB1ebF/Oy5naln101maO1gT9kBrU4d0B6P8Se/cuROdTkd0dDSJiYlERESwZs0aALy8vIiKigJg27ZtODs7ExAQgFarZfbs2Rw5coRBgwYBt+6FzpkzhxUrVtC6dWs2btzIxYsXadu2rTHCFkKU49vDl8gvLGaM3HaqNiqVirdGdiYjW8ub3xzFxc6SQZ1r14Q8o4xRHDhwAH9/fwB8fHw4evToHfvk5eWxcuVK3njjDQAKCgoIDAxk6tSphn2Sk5NxdHTks88+Y8KECdy8eVOShBAmFJOQQjtnW3zdHU0dSr1ibqZmxbjudG3pyItfHeLA+RumDqkEo1xR5OTkYGv719MQZmZmFBUVYW7+1+liY2MZPHgwTk5OADg4OODn58fmzZsN+9y4cYNDhw4xZ84cWrduzdSpU+nSpQsPPvhgqefW6XQkJSVVKW6tVlvlYxsq6bPKqcv9deGmjoMXbjKlpxMnTpyokXPW5f6qipl9HJj27xwmfrqPdx5rQUsHTaWON1Z/GSVR2Nrakpuba3it1+tLJAmAuLg4VqxYUWY7jo6OtG7dmnbt2gHg7+/P0aNHy0wUGo0GLy+vKsWdlJRU5WMbKumzyqnL/fX1v5MwV6uYOtiXprY1M05Yl/urqr5yb0Pwml+Zv/sam/+vD852VhU+1lj9ZZRbT76+vsTHxwO3Bqs9PT1LbM/Ozkan0+HqWvZTE61atSI3N5fz588DkJCQQPv27Y0RshCiDIXFejYfTGWAl3ONJYmGqnWTRnzyZC+u5eiYtG4/OQVFpg7JOIli4MCBaDQaQkNDWbx4MbNmzSIyMpJdu3YBt8Ye3Nzcym1Ho9GwcOFCpk2bRnBwMM2bN+fhhx82RshCiDLsSpICgDWpWytHVj/uS9LlWxPyCotNOyHPKLee1Go14eHhJd7z8PAw/Ozt7c3q1avvemxQUFCJ1w8++CCxsbHVH6QQosI2JqTgbGfJQ55SALCm9OvozOJRXXlt0xFmbvqD5WO8TbY4lDwILYQoU1qWlp9OpvPsQx5SALCGje3VikuZ+by38zQtHK2YNqiDSeKQRCGEKNOmg6noFeS2k4n8c0B7rmRqWfnjGZo7WPH4/TU/IU8ShRCiVLcKAKbS+z4n2jRtZOpwGiSVSsWCwC6kZWmZs+UoznZWDOzkUqMxyHWkEKJU+/+8QfLVXMb2kqsJUzI3U7PqcV+6ujnw4lcHOXihZifkSaIQQpQqen8KtpbmDOlau0pKNEQ2GnM+mdgLF3srpnyWwLmMmlshr9xEUVhY+xfVEEJUv2xtIf/+4zLDu7lio5G71LVBU1tLPnuqNyrgycjfycguqJHzlpsogoKCWLhwIadOnaqJeIQQtcS3Ry5LAcBa6L6mjfhkYi+uZt+akJdbAxPyyk0U33zzDX5+frz//vuEhYWxcePGEuU5hBD1U0xCCu2dbeneSgoA1jY+rRx5f3x3jl3K5PkvjT8hr9xEoVarCQgIIDg4GEdHR6Kiopg8eTLR0dFGDUwIYTqn07I5dOEmIb1amWySlyjbAC8XFo7qys8nM3h98x9GXSGv3BuPS5cuZdeuXfTu3Zunn34ab29v9Ho9QUFBhISEGC0wIYTpRO9PwVytIrB7+aV2hOmM6+3O5UwtK0icImMAACAASURBVHadxtXRmsdaGuc85SaK++67j6+//hobGxvDwLZareb99983TkQmcvFmPtO3XUL54aqpQ6kz7K3NmdKtEQ2rtmf9pyvS8/Whizzi5SIFAOuAlx9pz5XMfFbsOk2TAc0xRrHdchOFoii89957vP766zz77LOMGDGCwMBAWrY0UuoyEY2ZmmaNzDG3sjZ1KHXG78nXWJaVy8M99VLaoR758UQa13J1jO1Vv/6O11cqlYqFo7riYm+Fk3W+Uc5RbqLYsGEDGzZsAOCDDz5gwoQJBAYGGiUYU2pmZ8mMAOcGV/v+XsQdvsSLXx3ig/hzPN+vnanDEdUkJiEVF3tLAtpLAcC6wsJMzbRBHYy2yFOFBrMtLW9dflpYWMjAljAY3q0FAfc14r2dp0i6nGXqcEQ1uJKp5eeT6QT7tpSrRGFQ7hXFgAEDGD9+PN7e3hw7doz+/fvXRFyijvi/+5ty/OoVXok5zDfP90VjLv+41GVSAFDcTbmJ4v/+7//o168fycnJBAYG0rFjx5qIS9QRDlZmLA7qytOfJ7Dyx9MmK4Ms7t2tAoAp9G7jxH1SAFD8j3J//Tt//jzx8fGcO3eOnTt3Mnfu3JqIS9QhAzu5MLpHS1b/fJbElJumDkdU0e/J1/nzWh4hcjUh/qbcRDFjxgwADh48SGpqKjdvyj8E4k5zh3fCxc6SaTGJaAuLTR2OqILohFsFAB+TAoDib8pNFFZWVjz77LO4uLgQERHB1asyz0Dcyd7KgqWju3E2I5fl20+aOhxRSX8VAGwhBQDFHcpNFIqikJGRQV5eHnl5eWRmZtZEXKIO8mvflLAHWvPJ3mT+c+6aqcMRlRB3+DLaQj0hsu6EuItyE8ULL7zAzp07GTFiBAMGDCAgIKAm4hJ11MzHOuLuZMP02MM1UtVSVI/ohBQ8XWzp1tLB1KGIWqjcRHHkyBHGjRvHgAED+O233wxjFkLcTSNLc5aP6UbqjXwW/ds4k39E9Tp5JZvDKTcZ21MKAIq7KzdR7N69m+JiGZwUFdfrPiee9m/LF/+5wO5TGaYOR5QjJiEFCzMVo6QAoChFuaNWN27cwN/fn5YtW6JSqVCpVIaSHkKU5pWBnvx4Ip0ZsUfY/nIADtYWpg5J3MX/FgBsIgUARSnKTRRr166tiThEPWNlYcY7Y7sxavWvzI87xjtjfUwdkriLXUlpXM/VyUxsUaZyE8XXX399x3svvPBCmcfo9XrmzZvHyZMn0Wg0LFiwgNatWwOQlJTEokWLDPsmJiayatUqwyD5unXruHr1KtOnTy/R5pw5c3BwcLjjfVF7ebd05Pl+7Vix6zSPdm7Oo53l+fzaJiYhheb2VgR4SgFAUbpyE0XTpk2BW4/JHj9+HL2+/CX3du7ciU6nIzo6msTERCIiIlizZg0AXl5eREVFAbBt2zacnZ0JCAhAq9Uye/Zsjhw5wqBBg0q0t2HDBk6dOkWvXr0q/QGFab3Qrx27ktJ44+s/6HWfE06NNKYOSfzXlUwtu09l8NzDHpipZRBblK7cRBEaGlri9ZQpU8pt9MCBA/j7+wPg4+PD0aNH79gnLy+PlStXsn79egAKCgoIDAykT58+nDt3zrDfoUOHOHz4MCEhISXeF3WDxlzN22O7MXzlHmZv+YNV433lyZpa4nYBwDE95LaTKFu5iSI5Odnwc0ZGBpcvXy630ZycHGxtbQ2vzczMKCoqwtz8r9PFxsYyePBgnJycAHBwcMDPz4/Nmzcb9klPT+f999/n/fffZ9u2bRX6QDqdrso12bVardHquddXFe2zCd0aE3nwCmu3JfBwG9ty96+vast3TK8orP81ha4uVuRnXCCplj6cVlv6q64wVn+Vmyjmzp2LSqVCURSsrKx47bXXym3U1taW3Nxcw2u9Xl8iSQDExcWxYsWKMtv5/vvvuXHjBs888wwZGRlotVratm1LUFBQqcdoNJoqLz6UlJQkCxdVUkX77A1PPYlXf2Pt/hsE+3XF2d6qBqKrfWrLd2zfuWtczk7m1cc64+VVe1eyqy39VVcYq7/KTRQff/wxZ8+epVOnTuzcuZM+ffqU26ivry8//fQTQ4YMITExEU9PzxLbs7Oz0el0uLq6ltnOE088wRNPPAHA5s2bOXfuXJlJQtRe5mZq3h7TjSErfmHm5j/45MmecgvKhGL2p2Bnac5jXcr+OygEVGDC3auvvsrhw4eBW7ehZs6cWW6jAwcORKPREBoayuLFi5k1axaRkZHs2rXL0I6bm0zuaWjaNrNlxuCO/HginY0JqaYOp8HK0hby76OXGe7TAmuNmanDEXVAuVcUaWlpjBs3DoCnn36asLCwchtVq9WEh4eXeM/Dw8Pws7e3N6tXr77rsaVdMciVRP3w5IP3sf3YFcK/PU6fdk1o2djG1CE1OHGHL90qAChzJ0QFVWjdytsD2hcuXKjQ47FClEatVrFsdDcUReG12CPo9YqpQ2pwYvan0MHFDm8pACgqqNxE8frrr/PSSy/h5+fHSy+9VKFbT0KUpZWTDXOGdeLXs9eI2nfe1OE0KCeuZHE4NZOxvaQAoKi4cm89eXl5sXjxYsNgtqyZLapDSK9WfH/sCou3JRHg2Yw2skZzjYjZnyoFAEWllXtFMX369EoPZgtRHpVKxZJgbyzNzZgWk0ix3IIyulsFAFMZ2MlFZsiLSik3Ufx9MDs9Pd3oQYmGwcXeivCRnTl44SYf/SKz7o1tZ1IaN/IKGSOD2KKSKjWYff78eRnMFtVqRLcWPNalOe/8cIqTV7JNHU69ZigA2F4KAIrKqdRg9uTJkys04U6IilKpVCwI7IKdlTmvxCRSWCy/iBjD5cx84k9lMLpHSykAKCqt3ETRrVs33nrrLfr06UN+fj7Xrl2ribhEA9LE1pJFQV05dimL9388Y+pw6qVNB24VAJR1J0RVlPrUk06n47vvvuOLL75Ao9GQk5PDrl27sLJqmDV6hHE92rk5Qd3deP+nMzzi5UJXeca/2uj1CjEJqTzYtgnuTWSCo6i8Uq8o+vfvz8mTJ1m+fDlffvklzs7OkiSEUb05vDPNbC15JSYRbaGs015d9iVf48L1PMb2qr3F/0TtVmqieOKJJ/j11195++232b17N4oijy8K43KwsWDJaG9Op+fw7o5Tpg6n3tiYkIqdlRQAFFVXaqJ45pln2Lp1K2FhYXz77bccPXqUZcuWceqU/AUWxvOQZzPG3+/Oh7+cI+HP66YOp87LzC/k339cZkS3FlhZSAFAUTXlDmb37t2bZcuWsWPHDpo3b16h9SiEuBevD/GiZWNrpm08TJ6uyNTh1Glxhy9RUKQnpJcMYouqq9A8CgB7e3vCwsLYsmWLMeMRAltLc5aN7saF63lEbDth6nDqtJiEFDo2t6OrmzwcIKquwolCiJr0QNsmTOrbhs9/O8+e01dNHU6dlHQ5iyOpmYztKQUAxb2RRCFqrVcf7UDbZo14LfYwWdpCU4dT58QkpGBhpiJQCgCKeySJQtRaVhZmvDPWhytZWt6KO27qcOqUgqJithy6yKBOzaUAoLhnkihErebTypHnHvZg44FUdh5PM3U4dcbO4+n/LQAocyfEvZNEIWq9fwxoT8fmdszc/Ac3cnWmDqdOiElIwdXBCn8pACiqgSQKUetZmt+6BZWZr2Pu1mOmDqfWu3Qzn/jTUgBQVB9JFKJO6NTCnn8OaE/c4Ut8e+SSqcOp1WIPpKIoMKaHzJ0Q1UMShagzpj7kQbeWDszZcpT0bK2pw6mV9HqFjQdS6OMhBQBF9ZFEIeoMczM1b4/1IU9XzOubj0r9sbvYd+4aKdfzpZy4qFaSKESd0s7Zllcf7cDOpDQ2Hbxo6nBqnZiEFOyszBncpbmpQxH1iCQKUedM6tuG3m2cmL/1GJdu5ps6nFojM7+QbUevMNJHCgCK6mWURKHX65k7dy4hISGEhYVx/vx5w7akpCTCwsIM/3Xt2pX4+HjD9nXr1rF8+XLD62+//ZYxY8YQGhrK3LlzZc1ugVqtYvnobhQrCq/FHpFbUP+19XYBwJ7upg5F1DNGSRQ7d+5Ep9MRHR3NtGnTiIiIMGzz8vIiKiqKqKgoxo8fz6BBgwgICECr1TJ9+nS+/PJLw75arZb33nuPzz//nA0bNpCTk8NPP/1kjJBFHePexIY3hnqx58xV1v/ngqnDqRVi9t8qANjFzd7UoYh6ptSlUO/FgQMH8Pf3B8DHx4ejR4/esU9eXh4rV65k/fr1ABQUFBAYGEifPn04d+4cABqNhg0bNmBtbQ1AUVERlpaWZZ5bp9ORlJRUpbi1Wm2Vj22oTNln3e0UfFtYs+DbY7RQZdLC3sIkcVSGsfrr3PUC/riYydTeTThxov5U3JW/k5VjrP4ySqLIycnB1tbW8NrMzIyioiLMzf86XWxsLIMHD8bJyQkABwcH/Pz82Lx5s2EftVpN06ZNAYiKiiIvL4++ffuWeW6NRoOXl1eV4k5KSqrysQ2VqftslVsbBr0bz9pDOWx45sFaP8HMWP0VvfUYGjM1zz7qS+N6VNvJ1N+vusZY/WWUW0+2trbk5uYaXuv1+hJJAiAuLo4xY8aU25Zer2fJkiXs3buXlStXSrlkUYKrgzXzR3Rm/583+HRPsqnDMYmComK2JF5kYGeXepUkRO1hlETh6+trGKBOTEzE09OzxPbs7Gx0Oh2uruWv4Tt37lwKCgpYvXq14RaUEP9rVHc3BnVyYdkPJzmdlm3qcGrcjuNp3MwrlLkTwmiMkigGDhyIRqMhNDSUxYsXM2vWLCIjI9m1axcAycnJuLmVXyP/2LFjxMbGcurUKZ588knCwsLYsWOHMUIWdZhKpWLhqK7YWpozbeNhCosb1pNx0ftTaOFghV+7pqYORdRTRhmjUKvVhIeHl3jPw8PD8LO3tzerV6++67FBQUGGnzt37lyvBuaE8TSzs2RBYBf+74uDrPn5LP8Y0N7UIdWIizfz2XPmKi/2b1/rx2dE3SUT7kS9MaSrKyN9WrBi12mOXsw0dTg1IjbhdgFAWXdCGI8kClGvzB/RGadGGqbFHKagqNjU4RjV7QKAfds1oZWTFAAUxiOJQtQrjjYalgR7czItm/d2njZ1OEb127lrpN6QAoDC+CRRiHqnX0dnQnu14oPdZzlw/oapwzGamIQU7K3MebSzFAAUxiWJQtRLbwz1wtXBmukbD5Ovq3+3oDLzbhcAdJMCgMLoJFGIesnOyoJlY7xJvprLku/r35NzWw9fRFekJ6SX3HYSxieJQtRbfTyaMrHPfaz79U9+PXvV1OFUq+iEFLxc7encQgoACuOTRCHqtRmDO9KmaSNe3XiEbG2hqcOpFscuZXL0YhYhPVtKSRtRIyRRiHrNWmPG8jHduJyZz8Lv6kcV0o0JqWjM1Iz0Kb+6gRDVQRKFqPd6tG7MMwEebNifwk8n0k0dzj3RFhbz9aGLDJICgKIGSaIQDcLLA9vj6WLLjE1HuJmnM3U4VbbjeBqZ+YUyiC1qlCQK0SBYmpvxzlgfrufqeHPrMVOHU2UxCSm4OVrT10MKAIqaI4lCNBhd3Bx4sX97vkm8xLY/Lps6nEpLvZHHnjNXGd2jJWopAChqkCQK0aD8Xz8Puro58MaWo1zNKTB1OJUSeyAVgNFSAFDUMEkUokGxMFPz9thu5BQU8frmP1AUxdQhVYher7AxIZW+Hk2lAKCocZIoRIPj6WLH9EGe/HA8jS2JF00dToX8evYaF2/mM6anXE2ImieJQjRIk/3a0rN1Y+Z+c4zLmfmmDqdcUgBQmJIkCtEgmalVLB/TjaJihRmbavctqMy8Qr4/doXA7lIAUJiGJArRYN3XtBGvD+lI/KkMvvo9xdThlOqb/xYAlHUnhKlIohAN2uP3t8avXVMWfHecC9fyTB3OXUXvT6GTqz1d3BxMHYpooCRRiAZNrVaxZLQ3ZioV02MPo9fXrltQRy9mcuxSlszEFiYliUI0eG6O1swd3onfk68T+eufpg6nhI0JKWjM1Yz0aWHqUEQDJolCCG5NYnvEy5ml35/gTHqOqcMBbhUA3JJ4iUc7N8fRRgoACtORRCEEoFKpWBTUFWuNGdM2HqaoWG/qkPjhdgFAGcQWJiaJQoj/crazYkFgFw6n3OSD+HOmDoeY/bcKAPbxaGLqUEQDZ5REodfrmTt3LiEhIYSFhXH+/HnDtqSkJMLCwgz/de3alfj4eMP2devWsXz5csPrH3/8keDgYEJCQoiJiTFGuEIYDPNuwTBvV97beYrjl7JMFkfK9Tz2nr3KmJ5SAFCYnrkxGt25cyc6nY7o6GgSExOJiIhgzZo1AHh5eREVFQXAtm3bcHZ2JiAgAK1Wy+zZszly5AiDBg0CoLCwkMWLFxMbG4u1tTXjxo2jX79+NGvWzBhhCwHAWyO7sO/cdV6JSWTrC35ozGv+wlsKAIraxCiJ4sCBA/j7+wPg4+PD0aNH79gnLy+PlStXsn79egAKCgoIDAykT58+nDt367L/7NmzuLu74+Bw6/nxHj16kJCQwGOPPVbquXU6HUlJVVvyUqvVVvnYhqq+9tnzvR2Z/2Mac6N/40lfp2prtyL9pVcUvtqXgk9za7KvnCfpSrWdvs6pr98vYzFWfxklUeTk5GBra2t4bWZmRlFREebmf50uNjaWwYMH4+R06y+hg4MDfn5+bN68uUQ7dnZ2hteNGjUiJ6fsJ1I0Gg1eXl5VijspKanKxzZU9bXPvLzgeOZhYg6mEuLfie7ujaul3Yr01y+nM0jPTWb2iK54eTXsx2Lr6/fLWIzVX0a5pra1tSU3N9fwWq/Xl0gSAHFxcYwZM6ZS7eTm5pZIHEIY05zhnWhub8W0jYfRFhbX2HljElJxsLZgUCeXGjunEGUxSqLw9fU1DFAnJibi6elZYnt2djY6nQ5XV9cy2/Hw8OD8+fPcvHkTnU5HQkIC3bt3N0bIQtzB3sqCpaO7cS4jl2XbT9bIOW/m6dh+7AqBPi2kAKCoNYxy62ngwIHs3buX0NBQFEVh0aJFREZG4u7uzoABA0hOTsbNza3cdiwsLJg5cyaTJ09GURSCg4NxcZHfskTN8WvflLAHWvPp3mQGdnLhgbbGfVR1y6H/FgCUkh2iFjFKolCr1YSHh5d4z8PDw/Czt7c3q1evvuuxQUFBJV7379+f/v37V3+QQlTQzMc6En86g1djD7PtnwHYWhrlrw1w67ZT5xb2dG4hBQBF7SET7oQoRyNLc5aP6UbqjXwW/dt4T+AcvZjJ8ctSAFDUPpIohKiAXvc58bR/W778zwV2n8owyjlibhcA7Fb+bVkhapIkCiEq6JWBnrRztmVG7BEy8wqrtW1tYTFbDl1kcOfmONhYVGvbQtwrSRRCVJCVhRnvjO1GRk4B8+OOVWvb249dIUtbJLedRK0kiUKISvBu6cjz/dqx+dBFth+rvinTMQkptGxszYNGfqpKiKqQRCFEJb3Qrx2dW9jzxtd/cC2n4J7bS7mex94z1xjTo5UUABS1kiQKISpJY67m7bHdyMovYvaWoyjKvS2fuvFAKioVjO4pBQBF7SSJQogq6NjcnpcHerLt6BW2Hr5U5XaK9QqxCSn4tWuKm6N1NUYoRPWRRCFEFT0T0Jbu7o7M/eYYaVnaKrWx98xVLmVqGSur2IlaTBKFEFVkplbx9phuFBQVM3PTkSrdgopOSMHRxoJBnaU0jai9JFEIcQ/aNrNl5uCO/HQyg5iElEodeyNXx45jaQT6uGFpLgUARe0liUKIe/TEg/fxYNsmhMcdJ+V6XoWP25J4EV2xXm47iVpPEoUQ90itVrF0tDcqlYrXYo+g15d/C0pRFKL3p9DVzYFOLexrIEohqk4ShRDVoJWTDXOGefHbuWt8/tuf5e5/9GIWJ65kM1YeiRV1gCQKIarJ2J6t6NehGRHfn+BcRtlL9sYkpGBprmaEjxQAFLWfJAohqolKpSIi2BtLczOmbTxMcSm3oLSFxWxJvMjgLs1xsJYCgKL2k0QhRDVysbcifGRnDl24yYfx5+66z/ZjV8jWFhEig9iijpBEIUQ1G9GtBUO6NufdHac4cSXrju3R+1No5WRt9GVVhagukiiEqGYqlYq3RnbB3tqcaTGH0RXpDduuZBfy61kpACjqFkkUQhhBE1tLFo7qyrFLWbz/0xnD+z+cyUalguAe8rSTqDskUQhhJI92bk6QrxurfjrDkdSbFOsVdpzJxr99MykAKOoUSRRCGNGbwzvTzNaSV2IOsyspjat5xTJ3QtQ5kiiEMCIHawuWjPbmTHoOL0UnYmepZmAnKQAo6hZJFEIY2UOezRh/vzt5umL6t7WVAoCizjE3RqN6vZ558+Zx8uRJNBoNCxYsoHXr1gAkJSWxaNEiw76JiYmsWrWKLl26MH36dLRaLc7OzixevBhra2s++eQTvvvuO1QqFVOnTmXgwIHGCFkIo3pjiBeW5mr6uerL31mIWsYoiWLnzp3odDqio6NJTEwkIiKCNWvWAODl5UVUVBQA27Ztw9nZmYCAABYsWMCwYcMICgriww8/JDo6mqCgIKKiovjhhx/Iz88nMDBQEoWokxpZmvPm8M4kJSWZOhQhKs0ot54OHDiAv78/AD4+Phw9evSOffLy8li5ciVvvPHGHccEBATw66+/Ym1tTYsWLcjPzyc/Px+VSp47F0KImmaUK4qcnBxsbW0Nr83MzCgqKsLc/K/TxcbGMnjwYJycnAzH2NnZAdCoUSOys7MBcHV1ZejQoRQXF/Pss8+We26dTlfl39q0Wq38xldJ0meVI/1VOdJflWOs/jJKorC1tSU3N9fwWq/Xl0gSAHFxcaxYseKOY6ysrMjNzcXe3p74+HjS09PZtWsXAJMnT8bX1xdvb+9Sz63RaPDy8qpS3ElJSVU+tqGSPqsc6a/Kkf6qHGP1l1FuPfn6+hIfHw/cGqz29PQssT07OxudToerq2uJY3bv3g1AfHw8PXr0wMHBASsrKzQaDZaWltjZ2ZGVdWftHCGEEMZjlCuKgQMHsnfvXkJDQ1EUhUWLFhEZGYm7uzsDBgwgOTkZN7eSdfife+45ZsyYQUxMDI0bN+btt9/GxsaGX3/9lbFjx6JWq/H19aVv377GCFkIIUQpjJIo1Go14eHhJd7z8PAw/Ozt7c3q1atLbG/atCmffPLJHW394x//4B//+IcxwhRCCFEBMuFOCCFEmSRRCCGEKJNRbj2Z0sWLFwkKCjJ1GEIIUadcvHix1G0qRVHuvrCvEEIIgdx6EkIIUQ5JFEIIIcokiUIIIUSZJFEIIYQokyQKIYQQZZJEIYQQokwNIlEcPnyYsLCwCu+fmJjImDFjCA0N5f333ze8HxgYSFhYGGFhYcyaNcsYoZqEXq9n7ty5hISEEBYWxvnz50tsj4mJISgoiLFjx/LTTz8BcP36dSZNmsT48eN56aWXyM/PL3Xf29atW8fy5ctr5kPVkOrsu9vbBg0aREFBQY1+jppWlX67rT5+jyqjvL4DI3yPlHruww8/VIYNG6aMGTOmwseMGDFCOX/+vKLX65UpU6YoR48eVbRarTJy5EgjRmo627dvV2bMmKEoiqIcOnRImTp1qmFbenq6MmzYMKWgoEDJysoy/PzWW28pmzZtUhRFUT744AMlMjKy1H3z8/OVadOmKQMHDlSWLVtmks9oLNXVd4qiKPHx8crIkSOV7t27K1qttsY/S02qSr/V5+9RZZTVd4pinO9Rvb+icHd3Z+XKlYbXJ0+eNFwVvPjii4YFkm7LyclBp9Ph7u6OSqXCz8+P3377jRMnTpCfn8+kSZN44oknSExMrOmPYjRlrUh45MgRunfvjkajwc7ODnd3d06cOHHXFQlL27egoIDAwECmTp1qks9nTNXVd3CrmGZkZCSOjo41/0FqWFX6rT5/jyqjvBVEjfE9qveJ4tFHHy2xaNKcOXN48803iYqKIiAggI8//rjE/n9fne/2antWVlZMnjyZTz75hPnz5zN9+nSKiopq7HMYU2krEt7ednvlQbjVHzk5OXddkbC0fR0cHPDz86uhT1OzqqvvAPr27Uvjxo1rMHrTqUq/1efvUWWU1XdgnO9Rvav1VJ6zZ88yf/58AAoLC2nTpg3r169n+/btAERERJRYne/2antt2rShdevWqFQq2rRpg6OjIxkZGSUWX6qrylqR8O/bcnNzsbOzu+uKhKXtW59VV981NFXpN3FLRVYQrW71/ori79q0acOSJUuIiori1Vdf5aGHHmLChAlERUURFRWFm5sbFhYWXLhwAUVR2LNnDz179iQ2NpaIiAgA0tLSyMnJoVmzZib+NNWjrBUJvb29OXDgAAUFBWRnZ3P27Fk8PT3vuiJhafvWZ9XVdw1NVfpN3FLeCqLG0OCuKObNm8eMGTMoLi4GYOHChXfsc/vWUnFxMX5+fnTr1g0vLy9mzZrFuHHjUKlULFq0yOhZvKaUtyJhWFgY48ePR1EUXn75ZSwtLUtdkfBu+9Zn1dV3DU1V+k3cUl7fGYNUjxVCCFGmBnfrSQghROVIohBCCFEmSRRCCCHKJIlCCCFEmSRRCCGEKJMkCiGEEGWSRCGEEKJM9WPGmBA1QKvV8tprr5Geno6rqyv79+/nnXfeMZSi12q1LFmyBAsLC15++WVcXV1JTU1l6NChnD59muPHj/Pwww/zyiuvEBYWRocOHTh9+jQ2Njb07NmTPXv2kJWVxaeffoqZmRlvvPEG2dnZ3LhxgzFjxjB+/HgT94BoqOSKQogKio6OpmXLlmzYsIEXXniBa9eucfr0aZYtW8bnn39O//79+f777wFISUlh4cKFfPDBB/zrX/9i5syZbNy4kdjYWEN73t7efPbZZ+h0OqysrIiMjKRdu3bs37+f8+fPM3ToUD799FPWrl3LunXrTPSphZArCiEq7OzZswQEBADg4eGBk5MTLi4uLFy4EBsbHhhhDgAAATJJREFUG9LS0vD19QWgVatW2NnZodFoaNq0qaHks0qlMrTXuXNnAOzt7WnXrp3h54KCApo2bcpnn33GDz/8gK2tbb2pVCzqJrmiEKKCPD09OXToEAAXLlzgxo0bzJ49m0WLFhEREYGzszO3K+L8b0Koik8//RQfHx+WL1/O4MGDkUo7wpTkikKICho9ejQzZ87k8ccfp0WLFlhaWjJy5EjGjh2Lvb09TZs2JT09vVrO1a9fP+bNm0dcXByOjo6YmZmh0+nQaDTV0r4QlSFFAYWooIMHD5KXl4efnx9//vknU6ZMYefOnaYOSwijk0QhRAVlZGTwyiuvUFhYSFFREf/4xz8MYxZC1GeSKIQQQpRJBrOFEEKUSRKFEEKIMkmiEEIIUSZJFEIIIcokiUIIIUSZ/h/A9b7SSnDavAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Support Vector machine Model\n",
    "from sklearn.svm import SVC\n",
    "grid = [0.00001, 0.0001, 0.001, 0.01, 0.1]\n",
    "hist = []\n",
    "for val in grid:\n",
    "    clf = SVC(gamma=val)\n",
    "    cross_val = cross_val_score(clf, X, y, cv=5)\n",
    "    hist.append(np.mean(cross_val))\n",
    "plt.plot([str(i) for i in grid], hist)\n",
    "plt.title('Cross Validations score for SVC')\n",
    "plt.xlabel('gamma')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.grid()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "svm = SVC(kernel= \"linear\",C=0.025, random_state = 0 , gamma=0.01)\n",
    "svm.fit(X_train,y_train)\n",
    "y_pred_svm = svm.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Score of the model is -  0.7172413793103448\n",
      "Report card of this model - \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.717     1.000     0.835       104\n",
      "           1      0.000     0.000     0.000        41\n",
      "\n",
      "    accuracy                          0.717       145\n",
      "   macro avg      0.359     0.500     0.418       145\n",
      "weighted avg      0.514     0.717     0.599       145\n",
      "\n",
      "Accuracy score -  0.7172413793103448\n"
     ]
    }
   ],
   "source": [
    "score = svm.score(X_test, y_test)\n",
    "print(\"Score of the model is - \",score)\n",
    "print(\"Report card of this model - \")\n",
    "print(metrics.classification_report(y_test, y_pred_svm, digits=3))\n",
    "print(\"Accuracy score - \", metrics.accuracy_score(y_test,y_pred_svm))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set ROC AUC score: 0.500\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "test_roc_auc = roc_auc_score(y_test, y_pred_svm)\n",
    "\n",
    "# Print test_roc_auc\n",
    "print('Test set ROC AUC score: {:.3f}'.format(test_roc_auc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1, 'Accuracy Score: 0.7172413793103448')"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x648 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm7 = metrics.confusion_matrix(y_test, y_pred_ran)\n",
    "plt.figure(figsize=(9,9))\n",
    "sns.heatmap(cm7, annot=True, fmt=\".3f\", linewidths=.5, square = True, cmap = 'Accent_r');\n",
    "plt.ylabel('Actual label');\n",
    "plt.xlabel('Predicted label');\n",
    "all_sample_title = 'Accuracy Score: {0}'.format(score)\n",
    "plt.title(all_sample_title, size = 15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Conclusion** - Maximum accuracy of 75.17% can be achieved. This accuracy has been achieved by Decision Tree Model "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Decision Tree Model will be used as it has the highest accuracies among the other models that were used.**"
   ]
  },
  {
   "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>Actual values</th>\n",
       "      <th>Predicted values</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>499</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>403</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>485</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>103</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>122</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>172</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>301</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>232</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>178</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>134</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>145 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Actual values  Predicted values\n",
       "499              0                 0\n",
       "403              1                 1\n",
       "485              0                 0\n",
       "103              1                 0\n",
       "122              1                 1\n",
       "..             ...               ...\n",
       "172              0                 0\n",
       "301              0                 0\n",
       "232              0                 0\n",
       "178              0                 0\n",
       "134              0                 0\n",
       "\n",
       "[145 rows x 2 columns]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#print the true and predicted values\n",
    "dictionary = {'Actual values': y_test, 'Predicted values': y_pred_dtree}\n",
    "pd.DataFrame.from_dict(dictionary)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, after a long journey of data visulaisation, data cleaning, data modelling etc., we have finally got our model that we can use.\n",
    "\n",
    ">     So, the next question is - Is this the end?\n",
    ">     The answer is - I don't know. I'm no expert guys as I'm also learning. So, if anyone reading this knows what can be done more, kindly help me out here.\n",
    "    \n",
    "Till then, Have a good day!!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}