CPL_approval_prediction-code.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import the required libraries\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read and have a look at the dataset \n",
    "df_train = pd.read_excel('Project - 4 - Train Data.xlsx')\n",
    "df_test = pd.read_excel('Project - 4 - Test Data.xlsx')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(614, 20)\n",
      "(367, 19)\n"
     ]
    }
   ],
   "source": [
    "print(df_train.shape)\n",
    "print(df_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Loanapp_ID</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Marital_Status</th>\n",
       "      <th>first_name</th>\n",
       "      <th>last_name</th>\n",
       "      <th>email</th>\n",
       "      <th>address</th>\n",
       "      <th>Dependents</th>\n",
       "      <th>Qual_var</th>\n",
       "      <th>SE</th>\n",
       "      <th>App_Income_1</th>\n",
       "      <th>App_Income_2</th>\n",
       "      <th>CPL_Amount</th>\n",
       "      <th>CPL_Term</th>\n",
       "      <th>Credit_His</th>\n",
       "      <th>Prop_Area</th>\n",
       "      <th>INT_ID</th>\n",
       "      <th>Prev_ID</th>\n",
       "      <th>AGT_ID</th>\n",
       "      <th>CPL_Status</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>CPL001002</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>Austin</td>\n",
       "      <td>Pennycook</td>\n",
       "      <td>apennycook1@mac.com</td>\n",
       "      <td>007 Miller Road</td>\n",
       "      <td>0</td>\n",
       "      <td>Grad</td>\n",
       "      <td>N</td>\n",
       "      <td>7018.8</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>9694176212</td>\n",
       "      <td>92-247-0058</td>\n",
       "      <td>Mozilla/5.0 (Windows; U; Windows NT 5.1; zh-TW...</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>CPL001003</td>\n",
       "      <td>M</td>\n",
       "      <td>Y</td>\n",
       "      <td>Mareah</td>\n",
       "      <td>Tomanek</td>\n",
       "      <td>mtomanek2@bloomberg.com</td>\n",
       "      <td>145 Charing Cross Junction</td>\n",
       "      <td>1</td>\n",
       "      <td>Grad</td>\n",
       "      <td>N</td>\n",
       "      <td>5499.6</td>\n",
       "      <td>1809.6</td>\n",
       "      <td>153.6</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Rural</td>\n",
       "      <td>152387684</td>\n",
       "      <td>36-236-3028</td>\n",
       "      <td>Mozilla/5.0 (Windows NT 5.1; rv:21.0) Gecko/20...</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>CPL001005</td>\n",
       "      <td>M</td>\n",
       "      <td>Y</td>\n",
       "      <td>Myrna</td>\n",
       "      <td>Rosenthaler</td>\n",
       "      <td>mrosenthaler3@dmoz.org</td>\n",
       "      <td>268 Algoma Hill</td>\n",
       "      <td>0</td>\n",
       "      <td>Grad</td>\n",
       "      <td>Y</td>\n",
       "      <td>3600.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>79.2</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>6510396556</td>\n",
       "      <td>34-339-0917</td>\n",
       "      <td>Mozilla/5.0 (Windows; U; Windows NT 6.1; cs-CZ...</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>CPL001006</td>\n",
       "      <td>M</td>\n",
       "      <td>Y</td>\n",
       "      <td>Danyette</td>\n",
       "      <td>Stempe</td>\n",
       "      <td>dstempe4@tumblr.com</td>\n",
       "      <td>571 Stone Corner Park</td>\n",
       "      <td>0</td>\n",
       "      <td>Non Grad</td>\n",
       "      <td>N</td>\n",
       "      <td>3099.6</td>\n",
       "      <td>2829.6</td>\n",
       "      <td>144.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>7141853006</td>\n",
       "      <td>94-098-8902</td>\n",
       "      <td>Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_2_1 li...</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>CPL001008</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>Patten</td>\n",
       "      <td>McCathy</td>\n",
       "      <td>pmccathy5@ted.com</td>\n",
       "      <td>118 Moulton Drive</td>\n",
       "      <td>0</td>\n",
       "      <td>Grad</td>\n",
       "      <td>N</td>\n",
       "      <td>7200.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>169.2</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>23293411</td>\n",
       "      <td>38-247-1410</td>\n",
       "      <td>Mozilla/5.0 (Windows NT 6.0; WOW64) AppleWebKi...</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Loanapp_ID Sex Marital_Status first_name    last_name  \\\n",
       "0  CPL001002   M              N     Austin    Pennycook   \n",
       "1  CPL001003   M              Y     Mareah      Tomanek   \n",
       "2  CPL001005   M              Y      Myrna  Rosenthaler   \n",
       "3  CPL001006   M              Y   Danyette       Stempe   \n",
       "4  CPL001008   M              N     Patten      McCathy   \n",
       "\n",
       "                     email                     address Dependents  Qual_var  \\\n",
       "0      apennycook1@mac.com             007 Miller Road          0      Grad   \n",
       "1  mtomanek2@bloomberg.com  145 Charing Cross Junction          1      Grad   \n",
       "2   mrosenthaler3@dmoz.org             268 Algoma Hill          0      Grad   \n",
       "3      dstempe4@tumblr.com       571 Stone Corner Park          0  Non Grad   \n",
       "4        pmccathy5@ted.com           118 Moulton Drive          0      Grad   \n",
       "\n",
       "  SE  App_Income_1  App_Income_2  CPL_Amount  CPL_Term  Credit_His Prop_Area  \\\n",
       "0  N        7018.8           0.0         NaN     360.0         1.0     Urban   \n",
       "1  N        5499.6        1809.6       153.6     360.0         1.0     Rural   \n",
       "2  Y        3600.0           0.0        79.2     360.0         1.0     Urban   \n",
       "3  N        3099.6        2829.6       144.0     360.0         1.0     Urban   \n",
       "4  N        7200.0           0.0       169.2     360.0         1.0     Urban   \n",
       "\n",
       "       INT_ID      Prev_ID                                             AGT_ID  \\\n",
       "0  9694176212  92-247-0058  Mozilla/5.0 (Windows; U; Windows NT 5.1; zh-TW...   \n",
       "1   152387684  36-236-3028  Mozilla/5.0 (Windows NT 5.1; rv:21.0) Gecko/20...   \n",
       "2  6510396556  34-339-0917  Mozilla/5.0 (Windows; U; Windows NT 6.1; cs-CZ...   \n",
       "3  7141853006  94-098-8902  Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_2_1 li...   \n",
       "4    23293411  38-247-1410  Mozilla/5.0 (Windows NT 6.0; WOW64) AppleWebKi...   \n",
       "\n",
       "  CPL_Status  \n",
       "0          Y  \n",
       "1          N  \n",
       "2          Y  \n",
       "3          Y  \n",
       "4          Y  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature Selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#drop those columns which have unique string and are not necessary for our final prediction\n",
    "\n",
    "df_train = df_train.drop(['Loanapp_ID', 'first_name', 'last_name', 'email', 'address', 'INT_ID', 'Prev_ID', 'AGT_ID'], axis = 1)\n",
    "df_test = df_test.drop(['Loanapp_ID', 'first_name', 'last_name', 'email', 'address', 'INT_ID', 'Prev_ID', 'AGT_ID'], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Sex</th>\n",
       "      <th>Marital_Status</th>\n",
       "      <th>Dependents</th>\n",
       "      <th>Qual_var</th>\n",
       "      <th>SE</th>\n",
       "      <th>App_Income_1</th>\n",
       "      <th>App_Income_2</th>\n",
       "      <th>CPL_Amount</th>\n",
       "      <th>CPL_Term</th>\n",
       "      <th>Credit_His</th>\n",
       "      <th>Prop_Area</th>\n",
       "      <th>CPL_Status</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>0</td>\n",
       "      <td>Grad</td>\n",
       "      <td>N</td>\n",
       "      <td>7018.8</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>M</td>\n",
       "      <td>Y</td>\n",
       "      <td>1</td>\n",
       "      <td>Grad</td>\n",
       "      <td>N</td>\n",
       "      <td>5499.6</td>\n",
       "      <td>1809.6</td>\n",
       "      <td>153.6</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Rural</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>M</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>Grad</td>\n",
       "      <td>Y</td>\n",
       "      <td>3600.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>79.2</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>M</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>Non Grad</td>\n",
       "      <td>N</td>\n",
       "      <td>3099.6</td>\n",
       "      <td>2829.6</td>\n",
       "      <td>144.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>0</td>\n",
       "      <td>Grad</td>\n",
       "      <td>N</td>\n",
       "      <td>7200.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>169.2</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>Urban</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Sex Marital_Status Dependents  Qual_var SE  App_Income_1  App_Income_2  \\\n",
       "0   M              N          0      Grad  N        7018.8           0.0   \n",
       "1   M              Y          1      Grad  N        5499.6        1809.6   \n",
       "2   M              Y          0      Grad  Y        3600.0           0.0   \n",
       "3   M              Y          0  Non Grad  N        3099.6        2829.6   \n",
       "4   M              N          0      Grad  N        7200.0           0.0   \n",
       "\n",
       "   CPL_Amount  CPL_Term  Credit_His Prop_Area CPL_Status  \n",
       "0         NaN     360.0         1.0     Urban          Y  \n",
       "1       153.6     360.0         1.0     Rural          N  \n",
       "2        79.2     360.0         1.0     Urban          Y  \n",
       "3       144.0     360.0         1.0     Urban          Y  \n",
       "4       169.2     360.0         1.0     Urban          Y  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 8 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "columns = ['Sex', 'Marital_Status', 'Dependents', 'Qual_var', 'SE', 'Credit_His', 'Prop_Area', 'CPL_Status']\n",
    "\n",
    "for i in range(len(columns)) :\n",
    "    plt.tight_layout()\n",
    "    plt.subplot(4, 2, i+1)\n",
    "    sns.countplot(data = df_train, x = columns[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAMcAAAD0CAYAAAAv6zLFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAS4klEQVR4nO3df7RVZZ3H8ffHy49rappeJfRKsBJLCcXAdNVqJB1/wCxFKw1n/DHpGqxUtDU26dhKa4a1asocaxhLlwqyUnRM1MxsCH+gS9K4dkPQIbEYvXkHEYWkQLn0nT/2c/GE94Fz4Z6zz4XPa62zzjnPffbej9f7YT97n32+WxGBmb3TLmUPwKxRORxmGQ6HWYbDYZbhcJhlOBxmGQNqtWJJzcB8YHDazl0RcZWkGcAxwJrU9e8jol2SgOuAicCfUvvTW9pGS0tLDB8+vEb/BbazaGtrezUi9t28vWbhAN4Ejo2ItZIGAo9L+mn62Zci4q7N+k8ARqbHUcD16Tlr+PDhLFy4sI+HbTsbSf/bU3vNplVRWJveDkyPLX3iOAm4NS33C2AvSUNrNT6zranpMYekJkntwCvA3Ih4Mv1omqRFkq6VNDi1HQC8VLF4R2ozK0VNwxERGyNiDNAKfETSh4ArgA8CRwJ7A19O3dXTKjZvkDRF0kJJC1euXFmjkZvV9phjk4hYLekR4KSI+HZqflPSLcBl6X0HcGDFYq3Ayz2s6wbgBoBx48b5wrASbdiwgY6ODtavX1/2UKrS3NxMa2srAwcOrKp/Lc9W7QtsSMHYFfhr4JuShkZEZzo7dSqwOC1yH3CRpNkUB+JrIqKzVuOz7dfR0cEee+zB8OHDKf53Nq6IYNWqVXR0dDBixIiqlqnlnmMoMFNSE8X07c6IuF/SQyk4AtqBz6X+D1Ccxl1GcSr3szUcm/WB9evX94tgAEhin332oTdT8ZqFIyIWAUf00H5spn8AF9ZqPFYb/SEY3Xo7Vn9CbpZRlwPyMoz90q2lbLftW+eUst3+Ztq0adx22200NTWxyy678IMf/ICjjtriZ751t8OGwxrXggULuP/++3n66acZPHgwr776Km+99VbZw3oHT6us7jo7O2lpaWHw4OLz35aWFvbff3/a2to45phjGDt2LCeeeCKdnZ10dXVx5JFH8sgjjwBwxRVXcOWVV9ZlnA6H1d0JJ5zASy+9xMEHH8wXvvAFHn30UTZs2MDFF1/MXXfdRVtbG+eddx5XXnklAwYMYMaMGXz+859n7ty5PPjgg1x11VV1GaenVVZ3u+++O21tbTz22GM8/PDDfOYzn+ErX/kKixcv5vjjjwdg48aNDB1aXFo3atQozj77bE4++WQWLFjAoEGD6jJOh8NK0dTUxPjx4xk/fjyjR49m+vTpjBo1igULFvTY/5lnnmGvvfZixYoVdRujp1VWd0uXLuX555/f9L69vZ1DDjmElStXbgrHhg0bWLJkCQB33303q1atYv78+UydOpXVq1fXZZzec1jdrV27losvvpjVq1czYMAADjroIG644QamTJnC1KlTWbNmDV1dXVx66aUMGTKEyy+/nHnz5nHggQdy0UUXcckllzBz5syaj9PhsLobO3YsTzzxxDvaW1pamD9//jvaf/Ob32x6PXXq1JqOrZKnVWYZDodZhsNhluFwmGU4HGYZDodZhk/lWp/p668JVHP5vyTOOussZs2aBUBXVxdDhw7lqKOO4v7779+u7XvPYf3abrvtxuLFi1m3bh0Ac+fO5YAD+qaik8Nh/d6ECRP4yU9+AsDtt9/OmWee2SfrrVk4JDVLekrSryUtkfS11D5C0pOSnpd0h6RBqX1wer8s/Xx4rcZmO5bJkycze/Zs1q9fz6JFi/rsG4W13HN018o9HBgDnCTpaOCbwLURMRJ4HTg/9T8feD0iDgKuTf3Mtuqwww5j+fLl3H777UycOLHP1ltGrdxjge4i0jMpaldBUSu3+2qyu4Dj1J9KW1ipTjnlFC677LI+m1JBjc9WpZpVbcBBwHTgBWB1RHSlLpX1cDfVyo2ILklrgH2AVzdb5xRgCsCwYcNqOXzrR8477zz23HNPRo8evekrtdurpuGIiI3AGEl7AXOAQ3rqlp6rqpXrcqCNq8zKK62trVxyySV9us66nK2KiNXAI8DRFLcW6A5lZT3cTbVy08/3BF6rx/is/1q7du072saPH7/dn3FAbc9W7Zv2GFTUyn0OeBj4dOp2LnBven1fek/6+UOpCqJZKcqolfssMFvSvwK/Am5K/W8CZklaRrHHmFzDsZltVRm1cn8LfKSH9vXA6bUaj1lv+RNyswyHwyzD4TDL8CXr1mde/ProPl3fsK8+s9U+TU1NjB799nbvuece+ure9A6H9Wu77ror7e3tNVm3p1VmGd5zWL+2bt06xowZA8CIESOYM2dOn63b4bB+zdMqsxI4HGYZnlZZn6nm1Gt/4j2H9Ws9XbLeVxwOswyHwyzD4bDt0p++j9bbsTocts2am5tZtWpVvwhIRLBq1Sqam5urXsZnq2ybtba20tHRwcqVK8seSlWam5tpbW2tur/DYdts4MCBjBgxouxh1IynVWYZtaw+cqCkhyU9l2rlXpLar5b0e0nt6TGxYpkrUq3cpZJOrNXYzKpRy2lVF/CPEfG0pD2ANklz08+ujYhvV3aWdChFxZFRwP7AzyUdnArDmdVdLWvldkbE0+n1GxQ1q7Z044RJwOyIeDMifgcso4cqJWb1UpdjjnQ7gSOAJ1PTRZIWSbpZ0ntS26ZauUllHd3KdU2RtFDSwv5ylsT6p5qHQ9LuwI+ASyPiD8D1wPspbkvQCVzT3bWHxXuslRsR4yJi3L777lujUZvVOBySBlIE44cRcTdARKyIiI0R8WfgRt6eOm2qlZtU1tE1q7tanq0SRYnP5yLiOxXtQyu6nQYsTq/vAyanOzyNAEYCT9VqfGZbU8uzVR8DzgaekdT9PcZ/Bs6UNIZiyrQcuAAgIpZIuhN4luJM14U+U2VlqmWt3Mfp+TjigS0sMw2YVqsxmfWGPyE3y3A4zDIcDrMMh8Msw+Ewy3A4zDIcDrMMh8Msw+Ewy3A4zDIcDrMMh8Msw+Ewy3A4zDIcDrMMh8Msw+Ewy3A4zDIcDrOMMmrl7i1prqTn0/N7UrskfTfVyl0k6cO1GptZNWq55+iulXsIcDRwYaqHezkwLyJGAvPSe4AJFOV4RgJTKIq/mZWmjFq5k4CZqdtM4NT0ehJwaxR+Aey1WY0rs7oqo1bukIjohCJAwH6pm2vlWkOpKhyS5lXTlll281q52a49tLlWrpVmi0XdJDUD7wJa0oFz9x/wuynuobFFPdXKBVZIGhoRnWna9Epqd61cayhb23NcALQBH0zP3Y97gelbWjBXK5eiJu656fW5aV3d7eeks1ZHA2u6p19mZdjiniMirgOuk3RxRHyvl+vO1cr9BnCnpPOBF4HT088eACZS3LTmT8Bne7k9sz5VVa3ciPiepI8CwyuXiYhbt7BMrlYuwHE99A/gwmrGY1YPVYVD0iyKG860A92VzwPIhsOsv6u2yvo44ND0r7vZTqHazzkWA++t5UDMGk21e44W4FlJTwFvdjdGxCk1GZVZA6g2HFfXchBmjajas1WP1nogZo2m2rNVb/D2pRyDgIHAHyPi3bUamFnZqt1z7FH5XtKpvH2LZLMd0jZdlRsR9wDH9vFYzBpKtdOqT1a83YXicw9/5mE7tGrPVp1c8bqL4v7hk/p8NGYNpNpjDl8EaDudar/s1CppjqRXJK2Q9CNJrbUenFmZqj0gv4Xi+xb7U3x19cepzWyHVW049o2IWyKiKz1mAP6Oqu3Qqg3Hq5LOktSUHmcBq2o5MLOyVRuO84AzgP8DOoFP42/q2Q6u2lO5/wKcGxGvQ1G1EPg2RWjMdkjV7jkO6w4GQES8RlGHKkvSzens1uKKtqsl/V5Se3pMrPjZFakU6FJJJ/b2P8Ssr1Ubjl26a9rCpj3H1vY6M4CTemi/NiLGpMcDaX2HApOBUWmZ/5TUVOXYzGqi2mnVNcATku6iuGzkDGDalhaIiPmp0mE1JgGzI+JN4HeSllFc2LigyuXN+lxVe45UZeRTwApgJfDJiJi1jdu8KFVRv7lib1RVKVCzeqp2z0FEPAs8u53bu57i4D7S8zUUB/VVlQKFolYuRRV2hg0btp3D2bmN/VL9i8e0feucum9zW9X15jURsSIiNkbEn4Ebefs7IVWXAnWtXKuXuoZjs1sKnEZR1QSKS1MmSxosaQTFPTqequfYzDZX9bSqtyTdDoynKELdAVwFjJc0hmLKtJyiFi8RsUTSnRTTti7gwojY2NN6zeqlZuGIiDN7aL5pC/2nsZUzYGb15BtmmmU4HGYZDodZhsNhluFwmGU4HGYZDodZhsNhluFwmGU4HGYZDodZhsNhluFwmGU4HGYZDodZhsNhluFwmGU4HGYZDodZRs3CkamVu7ekuZKeT8/vSe2S9N1UK3eRpA/Xalxm1arlnmMG76yVezkwLyJGAvPSe4AJFOV4RlIUbLu+huMyq0rNwhER84HXNmueBMxMr2cCp1a03xqFXwB7bVbjyqzu6n3MMSQiOgHS836pvepauZKmSFooaeHKlStrOljbuTXKAXnVtXJdDtTqpd7hWNE9XUrPr6T2qmvlmtVLvcNxH3Buen0ucG9F+znprNXRwJru6ZdZWepdK/cbwJ2SzgdeBE5P3R8AJgLLgD/hm3FaA6h3rVyA43roG8CFtRqL2bZolANys4bjcJhlOBxmGQ6HWYbDYZbhcJhlOBxmGQ6HWUbNPgTcWb349dF13+awrz5T923uDLznMMtwOMwyHA6zDIfDLMPhMMtwOMwyHA6zDIfDLMPhMMtwOMwySrl8RNJy4A1gI9AVEeMk7Q3cAQwHlgNnRMTrZYzPDMrdc3wiIsZExLj0PldH16wUjTStytXRNStFWeEI4L8ltUmaktpydXT/gmvlWr2Udcn6xyLiZUn7AXMl/U+1C0bEDcANAOPGjeuxnq5ZXyhlzxERL6fnV4A5wEfI19E1K0XdwyFpN0l7dL8GTgAWk6+ja1aKMqZVQ4A5krq3f1tEPCjpl/RcR9esFHUPR0T8Fji8h/ZV9FBH16wsjXQq16yhuMCC1VV/KkDhPYdZhsNhluFwmGU4HGYZDodZhsNhluFwmGU4HGYZDodZhsNhluFwmGU4HGYZDodZhsNhluFwmGU4HGYZDodZRsOFQ9JJkpZKWibJJUGtNA0VDklNwHRgAnAocKakQ8sdle2sGiocFMXdlkXEbyPiLWA2RQ1ds7prtAILBwAvVbzvAI6q7JBq63bX110raWmdxlaV90EL8GpdN3qV6rq57dGgv5/39dTYaOHo6b/iL+rhVtbKbUSSFlbcVsE2059+P402reoADqx43wq8XNJYbCfXaOH4JTBS0ghJg4DJFDV0zequoaZVEdEl6SLgZ0ATcHNELCl5WL3VsFO+BtFvfj+K8C0uzHrSaNMqs4bhcJhlOBx9QIXHJU2oaDtD0oNljquRSApJ11S8v0zS1SUOaascjj4QxYHb54DvSGpOd6yaBlxY7sgaypvAJyW1lD2QajkcfSQiFgM/Br4MXAXcGhEvlDuqhtJFcabqi2UPpFoNdSp3B/A14GngLaBffApcZ9OBRZL+reyBVMPh6EMR8UdJdwBrI+LNssfTaCLiD5JuBaYC68oez9Z4WtX3/pwe1rN/B84Hdit7IFvjcFhdRcRrwJ0UAWloDoeV4RqKS9cbmi8fMcvwnsMsw+Ewy3A4zDIcDrMMh8Msw+EoiaT3Spot6QVJz0p6QNLBktZJak9t35e0i6ThkhZXud6jJT2Z1vFc95WvksZL+mgVy1fVb2fgy0dKIEnAHGBmRExObWOAIcALETFG0gDgIeBUiuu1qjUTOCMifp2K5H0gtY8H1gJPbGX5avvt8LznKMcngA0R8f3uhohop6JmV0R0UfyBHtTLde8HdKZ1bIyIZyUNp7ik/otpj/JxSSenPcyvJP1c0pBMvxmSPt29cklr0/NQSfNTv8WSPt77X0NjczjK8SGgbUsdJL0LOA54ppfrvhZYKmmOpAskNUfEcuD7wLURMSYiHgMeB46OiCMoKkv+U6Zfzt8CP4uIMcDhQHsvx9nwPK1qPO+X1E5RzO7eiPhp+he9KhHxdUk/BE6g+AM+k2KqtLlW4A5JQ4FBwO96Oc5fAjdLGgjck/Z8OxTvOcqxBBib+dkL6V/tIyLi6m1ZeUS8EBHXU+x5Dpe0Tw/dvgf8R0SMBi4AmjOr6yL9naRjpUFpG/OBvwJ+D8ySdM62jLWRORzleAgYLOkfuhskHUmmZmtvSPqb9EcMMBLYCKwG3gD2qOi6J8UfNsC5Fe2b91vO20GeBAxM23kf8EpE3AjcBHx4e8feaByOEqTvnJ8GHJ9O5S4BrmbLpU8/IKmj4nF6pt/ZFMcc7cAs4O8iYiPFV3hP6z7QTtv7L0mP8ZeFnTfvdyNwjKSnKIp6/zH1Gw+0S/oV8Cngul7+Ghqer8o1y/CewyzDZ6v6KUnTgY9t1nxdRNxSxnh2RJ5WmWV4WmWW4XCYZTgcZhkOh1mGw2GW8f9va/FpGAqxcwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAMcAAADzCAYAAAAy7gJ9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAATYElEQVR4nO3dfXBV9Z3H8fcXImSKuDWAFIwY6mIr5SGkoUth2IVtwYJQaxUEt4JQizsFV1ut68MUWRxHugumWBgdHZGHUYT1oa2t0kVdFzra0mADRlhEaFYDDApWHkSEhO/+cU/iJeSXnGBuzk34vGbu3Ht/OefcL3fuh/Nwzz1fc3dE5FTtki5AJFspHCIBCodIgMIhEqBwiATkJF3AZ9G1a1cvKChIugxp5TZu3LjP3bvVHW/V4SgoKKC0tDTpMqSVM7P/q29cm1UiAQqHSIDCIRLQqvc5zkTHjx+nsrKSo0ePJl1Kq5Obm0t+fj5nnXVWrOkVjlamsrKSzp07U1BQgJklXU6r4e7s37+fyspKevfuHWsebVa1MkePHqVLly4KRhOZGV26dGnSGlfhaIUUjNPT1PdN4RAJ0D5HM3tnbv+MLr9q1M/5ZPeJk8Y69vxKRl/zTKU1RxtgZlx77bW1z6uqqujWrRvjxo1r0nJ2797NVVddBUBZWRnPP/98o/O88sorDb7O3r17GTduHAMHDqRv376MHTsWgIqKCp544olGlx93ukxQONqATp06UV5ezscffwzA2rVrOf/885u0jKqqKnr27MlTTz0FxA9HY2bPns2oUaPYtGkTW7ZsYd68eYDCIS1ozJgx/Pa3vwVg5cqVTJ48ufZvGzZsYOjQoQwaNIihQ4eybds2AJYuXcqECRMYP348o0ePpqKign79+nHs2DFmz57NqlWrKCwsZNWqVcFlNGbPnj3k5+fXPh8wYAAAt99+O+vXr6ewsJCSkhIqKioYPnw4RUVFFBUV8eqrr9Y73dKlS5k1a1bt8saNG8crr7xCdXU11113Hf369aN///6UlJR8tjcU7XO0GZMmTWLu3LmMGzeOzZs3M336dNavXw/Al7/8ZdatW0dOTg4vvvgid955J08//TQAr732Gps3byYvL4+KigoAOnTowNy5cyktLWXRokUAHDx4MLiMhsycOZOrr76aRYsW8c1vfpNp06bRs2dP5s2bx/z58/nNb34DwJEjR1i7di25ubls376dyZMnU1paesp0S5curfd1ysrK2LVrF+Xl5QB8+OGHp/1e1lA42ogBAwZQUVHBypUra7fraxw4cICpU6eyfft2zIzjx4/X/m3UqFHk5eU1uvyGltGQSy+9lJ07d7JmzRpeeOEFBg0aVPsBTnf8+HFmzZpFWVkZ7du356233oq1/Bpf/OIX2blzJzfeeCOXXXYZo0ePbtL89dFmVRvy7W9/m1tvvfWkTSqAn/70p4wcOZLy8nKee+65k74I69SpU6xlN7SMxuTl5XHNNdewYsUKBg8ezLp1606ZpqSkhO7du7Np0yZKS0s5duxYvcvKycnhxIlPj9bV1HHuueeyadMmRowYweLFi7n++utj1xeicLQh06dPZ/bs2fTvf/Lh5AMHDtTuoIc2S+rq3Lkzhw4d+kzLAHj55Zc5cuQIAIcOHWLHjh306tWr3uX36NGDdu3asWLFCqqrq+uto6CggLKyMk6cOMG7777Lhg0bANi3bx8nTpzgyiuv5J577uH111+PXWOIwtGG5Ofnc9NNN50yftttt3HHHXcwbNiw2g9dY0aOHMmWLVtqd8hPZxkAGzdupLi4mAEDBvD1r3+d66+/nsGDBzNgwABycnIYOHAgJSUl/PCHP2TZsmUMGTKEt956q3aNVne6YcOG0bt3b/r378+tt95KUVERALt27WLEiBEUFhZy3XXXcd9998WuMcRa80XdiouLPdt+CZjpLwEPjPo5F1/4hZPG9CVgfFu3buWSSy45aczMNrp7cd1pteYQCdDRKmkWjz32GAsXLjxpbNiwYSxevDihij47hUOaxbRp05g2bVrSZTQrbVaJBCgcIgEKh0iA9jkEgK/+ZHmzLm/jf0xpdBoz48c//jELFiwAYP78+Rw+fJg5c+Y0ay2nS2sOSUzHjh155pln2LdvX9Kl1EvhkMTk5OQwY8aMZjm9PBMUDknUzJkzefzxxzlw4EDSpZwiY+EwswvM7L/NbKuZvWlmN0Xjc8xsl5mVRbexafPcYWZvm9k2M7s0U7VJ9jjnnHOYMmUKDzzwQNKlnCKTO+RVwC3u/rqZdQY2mtna6G8l7j4/fWIz6wtMAr4C9AReNLOL3T3+WW7SKt18880UFRVl3ZeIGVtzuPsed389enwI2Ao09MPmy4En3f0Td/8L8DbwtUzVJ9kjLy+PiRMn8uijjyZdykla5FCumRUAg4A/AsOAWWY2BSgltXb5K6ng/CFttkrqCZOZzQBmAPTq1SujdZ9J4hx6zaRbbrml9ie52SLj4TCzs4GngZvd/aCZPQjcA3h0vwCYDtR3ObpTzqd394eBhyF1ynqm6pbMO3z4cO3j7t271/4oKltk9GiVmZ1FKhiPu/szAO6+192r3f0E8AifbjpVAhekzZ4P7M5kfSINyeTRKgMeBba6+/1p4z3SJrsCqPm1/a+BSWbW0cx6A32ADZmqT6QxmdysGgZcC7xhZmXR2J3AZDMrJLXJVAHcAODub5rZamALqSNdM3WkSpKUsXC4+++pfz8ieBk9d78XuDdTNYk0hb4hFwlQOEQCdMq6AM1/1ZRes99o8O/uzvDhw7nrrrsYM2YMAKtXr2bJkiWsWbOmWWs5XQqHJMLMeOihh5gwYQIjR46kurqau+66K2uCAQqHJKhfv36MHz+en/3sZ3z00UdMmTKFiy66KOmyaikckqi7776boqIiOnToQLZdoE/hkER16tSJq6++mrPPPpuOHTsmXc5JdLRKEteuXTvatcu+j2L2VSSSJbRZJUDjh17PRAqHJC5bLsVTlzarRAIUDpEAhaPVcVpzw6EkNfV9UzhamfYH3+XDj44pIE3k7uzfv5/c3NzY82iHvJX53J8f4QN+wPvnXEDNz2VyDuj/uDhyc3PJz8+PPb3C0cq0O3aIs/94/0ljOgybGfovRyRA4RAJUDhEAhQOkQCFQyRA4RAJUDhEApJoXpNnZmvNbHt0f240bmb2QNS8ZrOZFWWqNpE4MrnmqGlecwkwBJgZNai5HXjJ3fsAL0XPAcaQuj5uH1ItBh7MYG0ijUqiec3lwLJosmXAd6LHlwPLPeUPwOfrXHRapEW1yD5HneY13d19D6QCBJwXTXY+8G7abMHmNWZWamal77//fibLljNcxsNRt3lNQ5PWM1Zv8xp3L3b34m7dujVXmSKnaPHmNcDems2l6P69aFzNaySrtHjzGlJNaqZGj6cCv0obnxIdtRoCHKjZ/BJJQhLNa+YBq83s+8A7wITob88DY0l1kT0CZFffXTnjJNG8BuAb9UzvwMxM1SPSVPqGXCRA4RAJUDhEAhQOkQCFQyRA4RAJUDhEAhQOkQCFQyRA4RAJUDhEAhQOkQCFQyRA4RAJUDhEAhQOkYBY4TCzl+KMibQlDf4S0Mxygc8BXaMrE9b8su8coGeGaxNJVGM/k70BuJlUEDbyaTgOAoszWJdI4hoMh7svBBaa2Y3u/osWqkkkK8S6wIK7/8LMhgIF6fO4+/IM1SWSuFjhMLMVwEVAGVAdDTugcEibFffSPMVAX1dneDmDxP2eoxz4QiYLEck2ccPRFdhiZr8zs1/X3BqawcyWmNl7ZlaeNjbHzHaZWVl0G5v2tzuixjXbzOzS0/vniDSfuJtVc05j2UuBRZy6X1Li7vPTB6KmNpOAr5A6bPyimV3s7tWIJCTu0ar/aeqC3X1d1JcjjsuBJ939E+AvZvY28DXgtaa+rkhziXv6yCEzOxjdjppZtZk11GujIbOinn9LavoBErNxTVSLmtdIi4gVDnfv7O7nRLdc4EpSm0xN9SCpQ8KFwB5gQTQeq3FNVIua10iLOK2zct39l8A/nsZ8e9292t1PAI+Q2nQCNa6RLBT3S8Dvpj1tR+p7jyZ/52FmPdIa0lxB6hAxpBrXPGFm95PaIe8DbGjq8kWaU9yjVePTHlcBFaR2ooPMbCUwgtQZvZXA3cAIMyskFawKUic24u5vmtlqYEu0/Jk6UiVJi3u0qsldltx9cj3DjzYw/b3AvU19HZFMiXu0Kt/Mno2+1NtrZk+bWX6mixNJUtwd8sdI7Rf0JHWI9bloTKTNihuObu7+mLtXRbelgI6jSpsWNxz7zOx7ZtY+un0P2J/JwkSSFvdo1XRSX/qVkDrS9CpZ3gr5qz9J5qcmz3ZO5GUlA+KG4x5gqrv/FcDM8oD5pEIj0ibF3awaUBMMAHf/ABiUmZJEskPccLRLO0mwZs0Rd60j0irF/YAvAF41s6dI7XNMRF/YSRsX9xvy5WZWSupkQwO+6+5bMlqZSMJibxpFYVAg5IyhC0mLBCgcIgEKh0iAwiESoHCIBCgcIgEKh0iATgGRFvXO3P4t/pq9Zr9xWvNpzSESoHCIBCgcIgEKh0iAwiESkLFwBJrX5JnZWjPbHt2fG42bmT0QNa/ZbGZFmapLJK5MrjmWAt+qM3Y78JK79wFeip4DjCF1fdw+wAxSV2MXSVTGwuHu64AP6gxfDiyLHi8DvpM2vtxT/gB83sx6ZKo2kThaep+je81V1qP786JxNa+RrJMtO+RqXiNZp6XDsbdmcym6fy8aV/MayTotHY5fA1Ojx1OBX6WNT4mOWg0BDqQ1uRFJRMZOPAw0r5kHrDaz7wPvABOiyZ8HxgJvA0fI8kuNypkhY+EINK8B+EY90zowM1O1iJyObNkhF8k6CodIgMIhEqBwiAQoHCIBCodIgMIhEqBwiATo0jxnsCSairamhqJac4gEKBwiAQqHSIDCIRKgcIgEKBwiAQqHSIDCIRKgcIgEKBwiAQqHSIDCIRKgcIgEKBwiAQqHSEAiv+cwswrgEFANVLl7sZnlAauAAqACmOjuf02iPhFIds0x0t0L3b04eh5qbCOSiGzarAo1thFJRFLhcOC/zGyjmc2IxkKNbU6i5jXSUpL6Dfkwd99tZucBa83sf+PO6O4PAw8DFBcX19vgRqQ5JLLmcPfd0f17wLPA1wg3thFJRIuHw8w6mVnnmsfAaKCccGMbkUQksVnVHXjWzGpe/wl3X2Nmf6L+xjYiiWjxcLj7TmBgPeP7qaexjUhSsulQrkhWUThEAhQOkQCFQyRA4RAJUDhEAhQOkQCFQyRA4RAJUDhEAhQOkQCFQyRA4RAJUDhEAhQOkQCFQyRA4RAJUDhEAhQOkQCFQyRA4RAJUDhEAhQOkQCFQyQg68JhZt8ys21m9raZqUeHJCarwmFm7YHFwBigLzDZzPomW5WcqbIqHKSutv62u+9092PAk6Sa2oi0uKT6c4ScD7yb9rwS+Lv0CaJmNzUNbw6b2bYWqi2WC6ErsK9FX/Rua9GX+yyy9P25sL7BbAtHff+KkxrUpDevyUZmVprW51DqaE3vT7ZtVlUCF6Q9zwd2J1SLnOGyLRx/AvqYWW8z6wBMItXURqTFZdVmlbtXmdks4HdAe2CJu7+ZcFlNlbWbfFmi1bw/5q6ekyL1ybbNKpGsoXCIBCgczcBSfm9mY9LGJprZmiTryiZm5ma2IO35rWY2J8GSGqVwNANP7bj9M3C/meVGLaTvBWYmW1lW+QT4rpl1TbqQuBSOZuLu5cBzwL8CdwPL3X1HslVllSpSR6p+lHQhcWXVodw24N+A14FjQKv4FriFLQY2m9m/J11IHApHM3L3j8xsFXDY3T9Jup5s4+4HzWw58C/Ax0nX0xhtVjW/E9FN6vdz4PtAp6QLaYzCIS3K3T8AVpMKSFZTOCQJC0idup7VdPqISIDWHCIBCodIgMIhEqBwiAQoHCIBCodIgMKREDP7gpk9aWY7zGyLmT1vZheb2cdmVhaNPWRm7cyswMzKYy53iJn9MVrG1prTws1shJkNjTF/rOnOBDq3KgFmZsCzwDJ3nxSNFQLdgR3uXmhmOcDLwHdIncwY1zJgortviq4g+aVofARwGHi1kfnjTtfmac2RjJHAcXd/qGbA3ctIu6Cdu1eR+oD+bROXfR6wJ1pGtbtvMbMCUr83+VG0RhluZuOjNcyfzexFM+semG6pmV1Vs3AzOxzd9zCzddF05WY2vOlvQ3ZTOJLRD9jY0ARm9jngG8AbTVx2CbDNzJ41sxvMLNfdK4CHgBJ3L3T39cDvgSHuPojUZVdvC0wXcg3wO3cvBAYCZU2sM+tpsyr7XGRmZaSu9Pgrd38h+h89Fnefa2aPA6NJfYAnk9pUqisfWGVmPYAOwF+aWOefgCVmdhbwy2jN16ZozZGMN4GvBv62I/pfe5C7zzmdhbv7Dnd/kNSaZ6CZdalnsl8Ai9y9P3ADkBtYXBXR5yTaV+oQvcY64O+BXcAKM5tyOrVmM4UjGS8DHc3sBzUDZjaYwAWNm8LMLos+xAB9gGrgQ+AQ0Dlt0r8h9cEGmJo2Xne6Cj4N8uXAWdHrXAi85+6PAI8CRZ+19myjcCQguiDDFcCo6FDum8AcGr4u8JfMrDLtNiEw3bWk9jnKgBXAP7l7Nanft19Rs6Mdvd5/mtl6Tr7qed3pHgH+wcw2kLri/UfRdCOAMjP7M3AlsLCJb0PW0ynrIgFac4gE6GhVK2Vmi4FhdYYXuvtjSdTTFmmzSiRAm1UiAQqHSIDCIRKgcIgE/D+XlT8JMXKn8AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(len(columns)) :\n",
    "    plt.figure(figsize = (12,8))\n",
    "    plt.subplot(2, 4, i+1)\n",
    "    sns.countplot(data = df_train, x = 'CPL_Status', hue = columns[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cont_columns = ['App_Income_1', 'App_Income_2', 'CPL_Amount', 'CPL_Term']\n",
    "\n",
    "for i in range(len(cont_columns)) :\n",
    "    plt.tight_layout()\n",
    "    plt.subplot(2, 2, i+1)\n",
    "    sns.scatterplot(y = cont_columns[i], x = 'CPL_Status', data = df_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\AMANJEET PANI\\Anaconda3\\lib\\site-packages\\statsmodels\\nonparametric\\kde.py:447: RuntimeWarning: invalid value encountered in greater\n",
      "  X = X[np.logical_and(X > clip[0], X < clip[1])] # won't work for two columns.\n",
      "C:\\Users\\AMANJEET PANI\\Anaconda3\\lib\\site-packages\\statsmodels\\nonparametric\\kde.py:447: RuntimeWarning: invalid value encountered in less\n",
      "  X = X[np.logical_and(X > clip[0], X < clip[1])] # won't work for two columns.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x1c3f28d7988>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA9QAAAN4CAYAAAAiCqARAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdeZxcVb3v/c/au4buru6kO6EThiQMCmhAjkoUNI9HFJlkCFPCnIAMIjid4zkX5Cr6AHrheo8iIrNAwhQSAQMCIiDguYhoIj6AQGROmiFpku6kh+quYa/nj72ru7q7qlM9VNfQ3/fr1a/qWrWraiW9VtX+7bXWbxlrLSIiIiIiIiIyMk6pKyAiIiIiIiJSiRRQi4iIiIiIiIyCAmoRERERERGRUVBALSIiIiIiIjIKCqhFRERERERERkEBdZZDDz3UAvrRz0h+Sk7tVj+j+Ck5tVv9jOKnLKjt6mcUPyWndqufUfxIgRRQZ/nggw9KXQWREVO7lUqkdiuVSm1XKpHarUjxKKAWERERERERGYWiBtTGmJuNMRuNMS9mlU0zxjxqjHk1uG0Kyo0x5ipjzGvGmOeNMZ/Mes6S4PhXjTFLssr3Nca8EDznKmOMGe49RERERERERMZLsUeobwUOHVR2IfC4tXZ34PHgPsBhwO7BzznAteAHx8APgP2ATwM/yAqQrw2OzTzv0G28h4iIiIiIiMi4KGpAba39I7B5UPECYGnw+1Lg6KzyZdb3Z6DRGLMDcAjwqLV2s7W2DXgUODR4bIq19hlrrQWWDXqtXO8hIiIiIiIiMi5KsYZ6prX2PYDgdkZQvhOwPuu4lqBsuPKWHOXDvccQxphzjDGrjTGrW1tbR/yPSXuWZNob8fNExmKs7VakFNRupVKp7UolUrsVmRjllJTM5CizoygfEWvtDdbaedbaec3NzSN9Oode+Uc+/5MnRvw8kbEYVbv1POjcAO3r/VtPF4JkYo3183ZcqB/IKJRF25WB1Je3qSLarf6OUgVCJXjPDcaYHay17wXTtjcG5S3A7KzjZgHvBuUHDCp/MiifleP44d5j3L26sROAnmSamrBbrLcRGRvPg40vwfKToH0dNM6BE++CGXPBKafraiJFpH4gUh3Ul6uD/o5SJUrRWu8HMpm6lwCrssoXB9m+9we2BNO1HwEONsY0BcnIDgYeCR7rMMbsH2T3XjzotXK9R9Gsebut2G8hMnrdrf1fWODfLj/JLxeZLNQPRKqD+nJ10N9RqkSxt826C3gG2NMY02KMORO4HDjIGPMqcFBwH+Ah4A3gNeBG4DwAa+1m4FLgr8HPJUEZwNeAm4LnvA48HJTne49x11QXBuCvbw3OvSZSRlKJ/i+sjPZ1frnIZKF+IFId1Jerg/6OUiWKOuXbWntSnocOzHGsBc7P8zo3AzfnKF8N7J2jfFOu9ygG1/GXcnf0pCbi7URGJxTxp1Jlf3E1zvHLRSYL9QOR6qC+XB30d5QqoQUKY9ST9ILbdIlrIjKMumZ/XVLjHP9+Zp1SXZkmKREpBvUDkeqgvlwd9HeUKlGKpGRVpTflB9KZwFqkLDmOn+TjrMf8qVShiP+FpaQfMpmoH4hUB/Xl6qC/o1QJBdRj4O9B7e/UpRFqKXuOA/UzS10LkdJSPxCpDurL1UF/R6kCugQ0BpnRaVBALSIiIiIiMtkooB6D7GnePSkF1CIiIiIiIpOJAuoxGDhCrTXUIiIiIiIik4kC6jHIDqK7E9o2S0REREREZDJRQD0GmXXTjtEItYiIiIiIyGSjgHoMMgF1fTREr5KSiYiIiIiITCoKqMegN+WPStdHQ/SkNEItIiIiIiIymSigHoOeZJpjnT9yR/Lf6ElqDbWIiIiIiMhkEip1BSpZT9Ljp5HrwIO6ZDvWWowxpa6WiIiIiIiITACNUI9BbyrNRtsIwBzeJ5m2Ja6RiIiIiIiITBQF1GPQm/RotVMB2NW8T09KiclEREREREQmCwXUY9CTSrPZNgCwi/N+X9ZvERERERERqX4KqMegJ5kmapJAMEKdUKZvERERERGRyUIB9Rj0JD3q6QFgF035FhERERERmVQUUI9BbypNzMSBIKDWlG8REREREZFJQwH1GGSPUNebHnp6EyWukYiIiIiIiEyUkgXUxph/M8b8wxjzojHmLmNMjTFmV2PMs8aYV40xdxtjIsGx0eD+a8Hju2S9zneD8rXGmEOyyg8Nyl4zxlxYjH9DMu1RRw8Wf+/pRLyzGG8jIiIiIiIiZagkAbUxZifgm8A8a+3egAucCFwB/MxauzvQBpwZPOVMoM1a+2HgZ8FxGGPmBs/bCzgUuMYY4xpjXOCXwGHAXOCk4NhxlU4lqTUJesL+XtSpno7xfgsREREREREpU6Wc8h0Cao0xIaAOeA/4IvDr4PGlwNHB7wuC+wSPH2iMMUH5cmttr7X2TeA14NPBz2vW2jestQlgeXDsuHKSXQAkIv5e1F6PRqhFREREREQmi5IE1Nbad4D/A6zDD6S3AGuAdmttKjisBdgp+H0nYH3w3FRw/PTs8kHPyVc+hDHmHGPMamPM6tbW1hH9O0KpbgASkSYAvET3iJ4vMlpjabcipaJ2K5VKbVcqkdqtyMQo1ZTvJvwR412BHYEY/vTswWzmKXkeG2n50EJrb7DWzrPWzmtubt5W1QfIjFCnghFq29s1oueLjNZY2q1IqajdSqVS25VKpHYrMjFCJXrfLwFvWmtbAYwx9wKfBRqNMaFgFHoW8G5wfAswG2gJpohPBTZnlWdkPydf+bgJp/wAOp0JqBMKqKV8eZ5lU1eCRCpNJOQyPRbBcXJdexKZnNRHZLJRmxeZeOp31adUAfU6YH9jTB0QBw4EVgNPAMfjr3leAqwKjr8/uP9M8PgfrLXWGHM/cKcx5qf4I927A3/BH6He3RizK/AOfuKyk8f7HxFKBwF1jZ+UjKQCailPnmdZu6GDs5etpqUtzqymWm5cPI89ZzboQ1wE9RGZfNTmRSae+l11KtUa6mfxk4v9DXghqMcNwAXAvxtjXsNfI/2r4Cm/AqYH5f8OXBi8zj+AFcBLwO+A86216WCE++vAI8DLwIrg2HEVSftrpm3UH6FGa6ilTG3qSvR9eAO0tMU5e9lqNnVp73QRUB+RyUdtXmTiqd9Vp1KNUGOt/QHwg0HFb+Bn6B58bA+wMM/r/Aj4UY7yh4CHxl7T/MJBUjIv6o9Qm5QCailPiVS678M7o6UtTiKVLlGNRMqL+ohMNmrzIhNP/a46lXLbrIoXDkao08EItZNUQC3lKRJymdVUO6BsVlMtkZBbohqJlBf1EZls1OZFJp76XXVSQD0GEc8PoFOZgDoVH+5wkZKZHotw4+J5fR/imTU702OREtdMpDyoj8hkozYvMvHU76pTyaZ8V4NouhsPg+fW0ksYR1O+pUw5jmHPmQ3cd958ZZUUyUF9RCYbtXmRiad+V51GHVAbYw6y1j46npWpNGGvhwQRMIY4UUIKqKWMOY6huSFa6mqIlC31EZls1OZFJp76XfUZy5TvX237kOrmegmSJgxADzWE0pryLSIiIiIiMlkMO0Id7POc8yH8ba0mNdcmSQf/hT1EFVCLiIiIiIhMItua8v054FSgc1C5Icf2VpONa5OkjP9f2GsUUIuIiIiIiEwm2wqo/wx0W2ufGvyAMWZtcapUOUI2SSqY8t1rokS8nhLXSERERERERCbKsAG1tfawYR771/GvTmUJ2SQpx/8vTJgodd7ggXwRERERERGpVuOyD7Ux5pnxeJ1KE7ZJ0vgj1AkTJaoRahERERERkUljXAJqoGacXqeihGySdDBCnXSiRK0CahERERERkclivAJqO06vU1HC9Gf5ThoF1CIiIiIiIpPJeAXUk1LIpkgHScmSTg019Ja4RiIiIiIiIjJRxiugNuP0OhXDWkuE/infKSdKlCSkUyWumYiIiIiIiEyEggNqY8zOxpgvBb/XGmMash4+bdxrVuZSnh9Qe8EIdcqJ+g8ku0pYKxEREREREZko29qHGgBjzNnAOcA04EPALOA64EAAa+2LxapguUqlLRFSdDohHCDtBnnZEt1QM7WkdRPJxfMsm7oSJFJpIiGX6bEIjjPpJpfIJKY+IFI+1B9lslLbrz4FBdTA+cCngWcBrLWvGmNmFK1WFSDpeURMkrQJEwLSrj9CbRNdk2/+u5Q9z7Os3dDB2ctW09IWZ1ZTLTcunseeMxv0IS6TgvqASPlQf5TJSm2/OhU65bvXWpvI3DHGhJikmb0z0sEItTX+NQkvmPKd6tGUbyk/m7oSfR/eAC1tcc5etppNXYltPFOkOqgPiJQP9UeZrNT2q1OhAfVTxpiLgFpjzEHASuCBsbyxMabRGPNrY8wrxpiXjTGfMcZMM8Y8aox5NbhtCo41xpirjDGvGWOeN8Z8Mut1lgTHv2qMWZJVvq8x5oXgOVcZY8b1sk/S8wiTwguSktlghDrZ0zGebyMyLhKpdN+Hd0ZLW5xEKl2iGolMLPUBkfKh/iiTldp+dSo0oL4QaAVeAL4KPAR8b4zv/XPgd9bajwD/ArwcvM/j1trdgceD+wCHAbsHP+cA1wIYY6YBPwD2w5+S/oNMEB4cc07W8w4dY30H8NdQJ7FBUjLPzYxQd47n24iMi0jIZVZT7YCyWU21REJuiWokMrHUB0TKh/qjTFZq+9WpoIDaWutZa2+01i601h4f/D7qKd/GmCnAvwK/Cl4/Ya1tBxYAS4PDlgJHB78vAJZZ35+BRmPMDsAhwKPW2s3W2jbgUeDQ4LEp1tpngnouy3qtcZFKpYmYdP8IdSgzQq0p31J+psci3Lh4Xt+HeGbNzvRYpMQ1E5kY6gMi5UP9USYrtf3qVGiW7yOAS4Gdg+cYwFprp4zyfXfDH/G+xRjzL8Aa4FvATGvte/gv/l5W4rOdgPVZz28JyoYrb8lRnuvfdg7+SDZz5swp+B+QSvYAYB1/hNqE/SzfGqGWiTDSdus4hj1nNnDfefOVVVJKZrSft+NBfUDGopRttxqpP04Mtdvyo7ZfnQqd8n0lsASYbq2dYq1tGEMwDX5Q/kngWmvtJ4Au+qd355KrldlRlA8ttPYGa+08a+285ubm4WudJZ3o9Z8fjFCbYMp3WgG1TIDRtFvHMTQ3RNmpqY7mhqg+vGXCjfbzdryoD8holbrtViP1x+JTuy1PavvVp9CAej3w4limeQ/SArRYa58N7v8aP8DeEEzXJrjdmHX87KznzwLe3Ub5rBzl4yYzQk0QULvhIKDuVUAtIiIiIiIyGRQaUP8P4CFjzHeNMf+e+Rntm1pr3wfWG2P2DIoOBF4C7scfCSe4XRX8fj+wOMj2vT+wJZga/ghwsDGmKUhGdjDwSPBYhzFm/yC79+Ks1xoXXspPb5+Z8u0EU75tb/d4vo2IiIiIiIiUqYLWUAM/AjqBGmC8Vs1/A7jDGBMB3gDOwA/wVxhjzgTWAQuDYx8Cvgy8BnQHx2Kt3WyMuRT4a3DcJdbazcHvXwNuBWqBh4OfceMlMmuo/f/CSMih14bxepWUTEREREREZDIoNKCeZq09eDzf2Fr7d2BejocOzHGsBc7P8zo3AzfnKF8N7D3Gaublpfw11Lj+CHXUhW6ikFRALSIiIiIiMhkUOuX7MWPMuAbUlc5LBgF1MEKdCahtQlO+RUREREREJoNCA+rzgd8ZY3qMMR3Bz9ZiVqzceSl/yrfJTPl2IW6jmKQCahERERERkcmgoCnf1tqGYlek0nipJAB20JTvxpQCahERERERkcmg0DXUGGOOAv41uPuktfa3xalSZbDBtlnGDaZ8OxAnyjSNUIuIiIiIiEwKBU35NsZcDnwLf2url4BvBWWTlg2Skplg2yzXgW5bg5uOl7JaIiIiIiIiMkEKHaH+MvBxa60HYIxZCjwHXFisipW7voA6mPIN0GuihNLtpaqSiIiIiIiITKBCk5IBNGb9PnW8K1JxUgmgPykZQMJECac15VtERERERGQyKHSE+n8BzxljngAM/lrq7xatVhXApjMj1P3/hb0mSlhTvqVMeZ5lU1eCRCpNJOQyPRbBcUypqyVSVtRPZDJRe5dSUxuUalBolu+7jDFPAp/CD6gvsNa+X8yKlb20P0LtZE35Tjo1RGxPqWokkpfnWdZu6ODsZatpaYszq6mWGxfPY8+ZDfriEgmon8hkovYupaY2KNWi0KRkxwDd1tr7rbWrgB5jzNHFrVqZy0z5zhqhTjhRwjYJ6VSpaiWS06auRN8XFkBLW5yzl61mU1di2Od5nqW1o5d32rpp7ejF8+xEVFdkXIy0/Y62n4hUokpv7/p+qnyV3gZHS223+hQ65fsH1tr7Mneste3GmB8AvylOtcqfCaZ8Z49QJ0xt8EsH1DaVoloiOSVS6b4vrIyWtjiJVDrvc3TlWCrZaNrvaPqJSKWq5Pau76fqUMltcLTUdqtToQF1rpHsgvewrkp9U76zRqjdWkgCvQqopbxEQi4Hz53BcfvOprE2THs8yT1r1hMJuXmfk+/K8X3nzae5ITpRVRcZldG039H0E5FKFQm5zGqqHRDQzGqqrYj2vqkrwc8eXcv3j5jb11d/9uhafnTMPvp+qiCV3AZHS223OhUaFK82xvwU+CVggW8Aa4pWqwqQGaF23f5On3SCEerejlJUSSSvptow3zxwD869fU3fFdHrTt2Xptpw3udMxivHUj1G035H009EKtX0WIQbF88bMlI2PRYpddW2yfM8lnx2Vy645/m+ul9x3D54nlfqqskIVHIbHC213epU6LZZ3wASwN3ASqAHOL9YlaoEJp2g14YHTM9IuZmAurNEtRLJrS2e7AsSwA8szr19DW3xZN7nZK4cZ6v2K8dSPUbTfkfTT0QqleMY9pzZwH3nzefpC77AfefNr5hpp2lLX0ACfl+94J7nSWspakWp5DY4Wmq71amggNpa22WtvdBaO89au6+19rvW2q5iV66cGS9JYtAAfzqkEWopT6MZrctcOc4EJZPhyrFUj9G0X83KkMnGcQzNDVF2aqqjuSFaMYGMtTZnX7VWUUmlqdQ2OFpqu9WpoCnfxpg9gP8Adsl+jrX2i8WpVvkz6QQJBk4DdMKZgHprCWokkt9o1illXznW/pBSaUbTfifjej6RSqS+KpVKbbc6FTrleyXwHPA94D+zfiYtJ50gOeh6hBvxA+pk95ZSVEkkr9GONk+2K8dSXUbafjUrQ6QyqK9KpVLbrU6FJiVLWWuvLWpNKozjDR2hzgTUPV1bUAobKSeOY9hzRownzp2LSfdi3ShufUwBskiWzKj2A1//LHXJNkI2iROKY6gH1FdEyoW+06RSqe1Wp0JHqB8wxpxnjNnBGDMt8zPWNzfGuMaY54wxvw3u72qMedYY86ox5m5jTCQojwb3Xwse3yXrNb4blK81xhySVX5oUPaaMebCsdZ1MNdLkDIDr0dEozUAJLrax/vtRMbG83BaXyZ8y0GErtqH8C0H4bS+DMoqKTKAg6Wp4zWitx6M+/OPYX71Jdj4kvqKSDnRd5pUKrXdqlRoQL0Ef4r3n/C3y1oDrB6H9/8W8HLW/SuAn1lrdwfagDOD8jOBNmvth4GfBcdhjJkLnAjsBRwKXBME6S7+Fl+HAXOBk4Jjx43xkkOmfMciDp22hlRca6ilzHS3wvKToH2df799nX+/u7W09RIpN+orIuVP/VQqldpuVSo0y/euOX52G8sbG2NmAYcDNwX3DfBF4NfBIUuBo4PfFwT3CR4/MDh+AbDcWttrrX0TeA34dPDzmrX2DWttAlgeHDtuXC9JctDE7lgYOqklHdcaaikzqUT/h3dG+zq/XET6qa+IlD/1U6lUartVadg11MaYY4d73Fp77xje+0rgfwANwf3pQLu1NhXcbwF2Cn7fCVgfvGfKGLMlOH4n4M9Zr5n9nPWDyvfLVQljzDnAOQBz5swpuPKuTdBrQgNC6oYwdNkaoj3ah1qKa8TtNhSBxjkDP8Qb5/jlIhNktJ+3E0p9RXKoiLY7maifFkTttgyp7ValbY1QHznMzxGjfVNjzBHARmvtmuziHIfabTw20vKhhdbeEOyvPa+5uXmYWg/keokhI9T1YeigVvtQS9GNuN3WNcOJd/kf2uDfnniXXy4yQUb7eTuh1Fckh4pou5OJ+mlB1G7LkNpuVRp2hNpae0YhL2KMWWKtXbrtI/vMB44yxnwZqAGm4I9YNxpjQsEo9Szg3eD4FmA20GKMCQFTgc1Z5RnZz8lXPi5CNknK1Awoi4Whw9YyI6GAWsqM48CMuXDWY/60olDE//B2Ck2jIDJJqK+IlD/1U6lUartVabz+et8aycHW2u9aa2dZa3fBTyr2B2vtKcATwPHBYUuAVcHv9wf3CR7/g7XWBuUnBlnAdwV2B/4C/BXYPcgaHgne4/5R/+tycL0E6UHXI0IOdJs6QklN+ZYy5DhQPxMaZ/u3+vAWyU19RaT8qZ9KpVLbrTqF7kO9LeO1edoFwHJjzGXAc8CvgvJfAbcZY17DH5k+EcBa+w9jzArgJSAFnG+tTQMYY74OPAK4wM3W2n+MUx0BcG2KlBm623SvU0s43TWebyUiIiIiIiJlaLwC6pzrkwt6orVPAk8Gv7+Bn6F78DE9wMI8z/8R8KMc5Q8BD422XtsStklSztD/vqRTQ0QBtYiIiIiISNUbrzkG4zVCXTFCNkk6xwh10q2l1usGO+prDCIiIiIiIlIBxiugfnqcXqdihEiSNjlGqN06HDxIaJRaRERERESkmhUUUBtjphtjfmGM+ZsxZo0x5ufGmOmZx621Xy9eFctTOM8IdToU83/paZ/gGomIiIiIiMhEKnSEejmwETgOP8t2K3B3sSpV9qwlnGeEui+gjiugFhERERERqWaFJiWbZq29NOv+ZcaYo4tRoYrgpXGweDkCahvxA+p0dxvuRNdLREREREREJkyhI9RPGGNONMY4wc8i4MFiVqyspXv9mxxTvk2kDoCejk0TWiURERERERGZWIUG1F8F7gQSwc9y4N+NMR3GmK3FqlzZSvkBtZdj2ywTjFDHtyqgFhERERERqWYFTfm21jYUuyIVJZ0AwMsxQh2q8QPqRKfWUEt58TzLpq4EiVSaSMhleiyC40y6He9kElMfEKke6s9SqdR2q0+ha6gxxhwL/D+ABf7bWvubotWq3A0zQh2J1uFZQ6pr80TXSiQvz7Os3dDB2ctW09IWZ1ZTLTcunseeMxv0IS6TgvqASPVQf5ZKpbZbnQrdNusa4FzgBeBF4FxjzC+LWbGy1jdCPTSgro84dFCL19020bUSyWtTV6LvwxugpS3O2ctWs6krUeKaiUwM9QGR6qH+LJVKbbc6FTpC/Xlgb2utBTDGLMUPrienIKC2ztAp3/Vh2GJj2odaykoile778M5oaYuTSKVLVCORiaU+IFI91J+lUqntVqdCk5KtBeZk3Z8NPD/+1akQwZRvm2PKd30YthDD9GyZ6FqJ5BUJucxqqh1QNquplkhIm7vJ5KA+IFI91J+lUqntVqdCR6inAy8bY/4S3P8U8Iwx5n4Aa+1RxahcuUone3EBL8cIddSFrcSYllBALeVjeizCbV/5FJ2b36cx4tGecKiftj3TY5FRvZ4SakilmR6LcOPieUPWreXtA54H3a2QSkAoAnXN4BR6Dbp8qK9KNZoei3D32ftRn24jalL02hCdbtOov9NEJsp4n49JeSg0oL4463eDn5zsJOC8ca9RBUgne3DJPeXbGOgyMcLJDRNfMZE8HCy7mA0Y500wdcx2urHGYIjhd+ltyAourBvh7Z46Trv5r0qoIRXDcQx7zmzgvvPmDx9cplPQ+b6/tCedgj/9HN78I5x4F8yYmzeoLsfAVclvpGpZj5l8QIhewFBDghhpsLOA4o/0lWN/l8rgYNkl1IaZZsE4zLYWG2or/HxMylKh22Y9ZYz5OHAysAh4E7jOWvtUMStXrtJJf8o3Tu4P7bgTI5rqmMAaiQzPxjdjOt6DB78D7eugcQ5mwTXY2kZMbLvhn+x5sPElWH4StK/DNM6hacFSmuvDtLTF+xJq3HfefJobohPzDxIZgcEnvztMrc198ptOwYYXYcVpff2Ehcv8x5afBGc9BvUzc75+OQau+ZLfqK9KpfN6thDq2QwrFvf11dCiZaSiDTixacV97zLt71Ihetox8U0D2q5ZtAyi9VBX3LYrxTPs/DVjzB7GmIuNMS8DVwPrAWOt/YK19hcTUsMylAmoc41QA/Q4MWrTCqiljCS7YdV5/oc3+LerzvPLt6W7tS+Yzjy3cdUSvndAc98hSqgh5Spz8nvMNU8z/4onOOaap1m7oQPPs0MP7ny/P5gG/3blYvjMN/zfU7mzsJZr1lYlv5Fq5SY7+wMS8G9XLPbLi6xc+7tUBpvoytl2baKrtBWTMdnWCPUrwH8DR1prXwMwxvxb0WtV5ry+EercAXXCjRFJJyEZh3BtzmNEJpTnQf0MOOTHUNsE8TZ4+kq/fFtSif4P/oz2dcyo678Sr4QaUq6GHaWNhbFdrXipXlImTMSAydHWcVx/tDqUe41buQaumeQ32XVTX5Wq4KXzfKcVv8+Va3+XCuGl8rTdVKlrJmOwrYD6OOBE4AljzO+A5WiCf19AnSvLN0AyFIMEEG9XQC1lIRWqI3zgD/tHqRvnwIJr/PJtPTkU8Y/PDjQa59Ce8Ce4bDO5k0gJ5Tv5TXtp7MZXMctPwm1fh9s4B2/R7Zg9D4e1D/Yf3DgHrPXXUNc1k0u5Bq4jTsQmUiFsqBaT4zvNhmqLfpJarv1dKkMp264Uz7ABtbX2PuA+Y0wMOBr4N2CmMeZa4D5r7e9H86bGmNnAMmB7wANusNb+3BgzDbgb2AV4C1hkrW0zxhjg58CXgW7gdGvt34LXWgJ8L3jpy6y1S4PyfYFbgVrgIeBbmX20x8pL9vi/5BmhTodj/i897TBlh/F4S5Gx8dK5p3yf/vttJ1epa/aDicy078Y52BPvYlbDbJ6+YCclZJGylu/kN5ZswwxayuCsOBW7+H7Mhhf6T3QW3Qax7aCmMW9CsumxCMu+8mne3tRNXcSlO5Fm5+l1AwLXUiQxKjgRm0iFSa2+yxQAACAASURBVFuLk+M7LX3G7wveD3a0dKFKxqKUbVeKp9CkZF3AHcAdQdC7ELgQGFVADaSA71hr/2aMaQDWGGMeBU4HHrfWXm6MuTB4jwuAw4Ddg5/9gGuB/YK6/ACYB9jgde631rYFx5wD/Bk/oD4UeHiU9R3AC/ahxs3932fDdf5tvE1Xm6QsOF4y57RtxyZ5+f2tfPW2NfmTqziOn934rMf6thAydc00OQ5NsYn/t4iMRK6T3yuO24conTn7RFvco/H0h/0+44ahfvshn/WDg+Om2jC9KY/vr3pxQD/KPr5USYwcxygBmVQdJ53nO81LFv+9daFqfFXJFoWFKmXbleIZcYu11m621l5vrf3iaN/UWvteZoTZWtsBvAzsBCwAlgaHLcUfFScoX2Z9fwYajTE7AIcAjwZ1agMeBQ4NHptirX0mGJVelvVaY5YORqjdPAE1ET/KSHRuHq+3FBkTzw37o23ZGufgOeG+YBqGSa7iOH5248bZ/m0Vf9lJdcmc/C4/Z3/uPmd/vn/EXB54roWQY3P2iXc6PTa5zTBtV5g6K2cwPSTJ2cYOfvbo2rz9SEmMRMbXcN9pEyFzoWqnpjqaG6IKpkcrs4vITV+CK/f2bze+VFh+lwqVdnK33bSZmLYrxVHys2JjzC7AJ4BngZnW2vfAD7qBGcFhO+FnGM9oCcqGK2/JUZ7r/c8xxqw2xqxubW0tqM7ppH8S5IZyN34TBNTxrW0FvZ7ISI203VrjwoJr+j/EM2t2jKvkKjJhRvN5OxKeZ2nt6OWdtm7aunqwHRugfT1O90YiDnxn5f/HV29bw8K5NTiP/E846uoBfSK18HZiTTNxHXinrZvWjl48zw543fe39gwJjr962xqO23f2gLpk9yMlMap8xW67MjLDfadJv7Jvtzl2EWH5SX55lWo3U2lfsHRA221fsJR2M7W0FZMxKWjKd7EYY+qBe4BvW2u3+kulcx+ao8yOonxoobU3ADcAzJs3r6A11pkR6lAo939fKOoH1L2dmwp5OZERG2m7dVJxePyHA7NKPv5DnGNv4ZC5zZyz7xRm1Bk2dltuWLNVyVWkKEbzeVuo7GnVzfVhbvlyDLNqSd9a6O1OuJNrT/44X7vz736G+rUPQteGgX2ivplGN8w7LetojHi8n3BINO/I1t50XxD963M/kzM4Hrx+MjtJkZIYVb5itt1SKsXa/vHgpOLw/HI4eaWfhd9LwzO/wPncBaWuWlkp+3abZxeRfFsUVoN4yvLT1fC9kx+gxvHo8Rwuf2oz532x/P48UriSBdTGmDB+MH2HtfbeoHiDMWYHa+17wbTtjUF5C5B9+X8W8G5QfsCg8ieD8lk5jh8XNtlL0rqE3dwD/OEaP6BOdWmEWsqD47jQuRHuPrW/sHEOxg1z9YE1hFcuhPZ1zG6cwy8X3YlTV9JrbSIjlj2t+udH7EjjqoUDE47dfTIfXvIIly7Ym8aGpD860LK6v080zqHluAfZyX2Vjz18Sl9/SCy8g0se7+kLhjd1JXIGxzMaon3lg5MUKYmRlKNSru0fKycUhX1OhDsXDsiU7OTZ2k7KlDE5dxEh/wBbxYuFHS6cB/V3Hgnt66hvnMOFC5aSCpd80rCMQUn+ekHW7l8BL1trf5r10P3AkuD3JcCqrPLFxrc/sCWYEv4IcLAxpskY0wQcDDwSPNZhjNk/eK/FWa81Zl6qlwQhonkGF+ojLh22llSX1lBLefCMM2R6K0ddjZdOEl55yoDAI7TiZJz4B6WrrMgoZE+rnlFnco561Jg0uzXHaEnESCy8Y0B/2HTkUtKpJOEVA/tDZOUpnLPvlL6Xue7J17niuH2Y1eRviTirqZbrTt2XHabUcN9583n6gi9w33nzBwQl2UmMcj0uUgqVvLbfeKmcO1cY7eVbWYyb89yEKp66P8VrpzEzewqgfR2Nq5YwxWsvbcVkTEo1DDUfOA14wRjz96DsIuByYIUx5kxgHX42cfCzdH8ZeA1/26wzwE+QZoy5FPhrcNwl1tpMFPs1+rfNephxyvANQKKbOFHyXUxqjMIWYqS7NUIt5cGzBp69fuD01mevJ3TIj3NPt0rG/aQg45V8bJJl8ZSJlz2temO3ZXaOUQ8TilBjHF55vwdn2s7svOT3dHZ18faWFDc8vZWrjwjl7A87T+3/qnxufTtL//Qmt5z+KTZ3JWiPJ7nq8X/yo2P2GTab9pBs257nzxrJ7hOgfiITpqLX9nup3N9dCqgri+PkPDfhyJ+VumZF4+bZdcVVlu+KVpKA2lr7f8m9zhngwBzHW+D8PK91M3BzjvLVwN5jqGZeJtlJl63JO0I9NQodto5oXFebpDxsdRqpPeBiYp3Bh3goStcBFxNxooRzTbf64J/Q2+lvlzXWE/pMFs+sfaw58a7xeW2RQPa06suebOWWBUv9UYD6GfD5C2Dah8BCJGS49Lcv0dIW5xOzGzn3gA8xPRbhosNnkzZbc/aHhlhswHTuJZ/dlf/x6+d5bn3/Z/ylCzxaO3oLW4uar0+EauD2Y7bZTyp13auUl0jI5eC5Mzhu39k01oZpjye5Z836iljbn3ajhHL01bQbLW1yIBmZumb4wkVDPwszFxirkM1kqB/Udq2rLN+VTJ87o+Akuuiglkie75ywA50mRm3v1omtmEhelqg7sMFGXZfO0FSmnHAnzt0n93+ZHXU1/OESf/TsrMf8bbLGIl8Wz/F4bZHA4L1hbcTBnv0EZuu7cPcpfe278YQ7eOzcvfnWqrd45KVWLv3tS1x36r68t6WH7ppaZhy5lOkP9Ccz23LcCmoNPHXOh0mZMFudRv7nb14cEEwfPHcGH3Qlht/PPVu+PnH4f22zn1TyulcpL021Yb554B6ce3t/u73u1H1pqi3/E3vHDfkJybasg3AdJLth6hy/XCqH40DzR+CMhyGdBDcM9dtX9cV2xw3BomWwYnH/edeiZWq7FU5/vVFwUt10ESUyTH+POzHCSWX5lvIwlU5C8Q/gwe/0fYCHjrmBWM1UvvZonGuXPIizZZ0/3eoPl/hPOuTHkOiGzg1jm3o6CbN4Smk4jqE5Fobudkgm/MQ2T14+IEg1d59CzeH/xTUH7cDGow7A4hCLOrR3p3jl/Q6uXJPinINWMqPO0OVF2MlrJ3Lrwf6UvMY5bHfiXXz3sI/w0nsdfUHI9w6fy8k3PTtkLep9583PPQ08X58I1w0tG9RP8q17zfteInm0xZP84vG1/PyIHft2efjF42u57Jh/Kfu2ZLwU9G4Z8J3GcTdharT1UEXxPGh7C9re6L8wkuyBabtVbVBtUnF44d6BGeqfuwOz39mlrpqMgQLqUQglO+mytTQOMyuq141Rk3pz4iolMoyQ1wv3nTNw9Ou+cwid/hDf+9xU4ilL7Jlr/K2EZs2DL14M9399fKZohyK5s3gqG6uMN8/DbnwJkz198Phb4dNnQ7QeIg2QikPNFNxHvk/skJ/SuqWLaQ0uM90oT7Z1cfERezHVtuN4SZqNQ83tiwcG5MtPYpczH+sbCY+E3JGvRc3XJ5LdA49rnEMvIbZ29PZN667oda9SVqyX5sfzQ0x/oH+Xhx8fuZS0V3hbKtnyAy8F95w18DvtnrPg9IeK/94yfuKbYfBszt6tfnlsu9LUqdiMAx8/yb/NdV8qkgLqUQiluulmBtsN852RdGPUpTonrlIiw7HpnCNixqaZvfTT/vqdRbf5iQ0+flJ/MB0cN9wU7W2eUNU1+wH5JFojJaVhu1r7g2nw108nu+GBbw5c0vDUT+Az5xNLbWHKquP7Hjt10Z10x9dRv3KRX/aVR3KPJKcTNDf2j+C1dvT2rbHOXpdtjMHz7NAAI1+fCNX0B9qNc2hfsJQz7niN1s5k37Ru7Wkt46WJrYQfGJhtePoDS0ie8ShQN+xzYezLD8YUjOf5TsN6hT1fykM6CYmugTMNFlzjl1crNwrpXrjrhAFTvnHLe1aIDE8B9SiE093ETe2w2+SlwzFqe3sh1QshdRIpMePAnof7wXImk+bf7+q/Itq+DrPiNNKnP4TjpTDt6/yR6vnfhobtIdKATSdJbXkft74ZJ1iPXdAJleP4o9tnPabsxVJUXqoXN/ske/63h26tc//X/eUMvzkXN7NmedY8OPAHuNE6GoyBI6+CJy7zs93nGElOEMJNeYRCfhvOJET72aNrWfLZXbngnueHDzDy9QmAsx4jnezlpY29XPzQ+zy33h+9yUzr1p7WMl5CXu6lByFb2HKcsSw/GHMugLzfacojUFG8JPz52oFZvv98LRx2ealrVjypuH9RN/vf/NRPqvvfPAkooB6FcLqLHlMz7DGpaBN0Qm/bO0Sbd5ugmonk4Ubg8/85JAkGbtZJePs62jp7mVIfI7Ln4bDfVwdM+zZHXU342etJff67MHMujusWfkLlOEpAJkWXMmHc7AC4tin3KFbzR/yTmVizH0wf+r/9kezbjh7YP6JT/NGSTFDeOIctRy/j2qc3c9L+Uwg5pm9kbc+ZDfzwqL1ZdP0zhQUY+fpE/Uzeb+vmyFueGFCcmdY9OPlaMafZKpt4dUuYMNGcF4zCFDIMMJblB5u6Elz56CsD1m9f+egrha/fdqN5vtM0gFFRjDvkXKPa96HGOPC570A8yLMUivr3NeW7oumvN1LWEvHi9G4joDax6QBsbHltImolMrx0qv/EA/zbFYv98ozGOdTU1vGN+9eTPOiyodO+7/86fPwkQitOJt3ZiudZEqk0/7XwX7j+tH35xOxGQOs5pXS6w020L1jqn5SBHyRnfs9onAOtr8AjFwEOHHuTP2IweCR7xWKwFp5f7gffpz8Ih/8X0akz+fZ+MbZ+8C5fv3MNx1zzNGs3dABgrR2X9c2Zad3Zsqd1Z/a03qmpjuaGaMFBrudZWjt6eaetm9aOXjzPDnvs2g0dHHPN08y/4om+f+dwz5HKspkpbDl6WX8fCS4YbWZKQc/fVjsdTmb99r6PLmT20k+z76ML+fH8ELbQ9dvpZJ7vtCqeKlyNrJf7XKOap+4bF9IJf5r7rYf7t+lEdV9EmAQ0Qj1SqR5cvG0G1FOm+skUPmh5ndmfmIiKiQzDS+YeqfOSfqCQ7KajbjZvddewsSPJ5niambmOD0b8nHRiyHS9X578SRwDH3QmqM23p5xIEU2tjfJ2/e6sP+xeGiP+5/Qui+4ktCLHtnDt62DFqf5WVdN2y90/uj+APQ6Fu0/tK675yiNw8yF8rHEONx25lLN+19U3Cj1e65uLMa17pFNslU184pRqJoBjHNa5O0PQX9oTDrjTmVHgSNlY2ulY128P+50mlcPLtxa+ii/Ke6mcSWKVUK+yKaAeqUSXf7ONgHr6NH+EunPjW8Wukci2Oe7A9WbWC6YX2b61ZzX/eiEPPv8e5x7wIVq2tjMz1/q0eBs0ziHlhIdM17vmiX9y7L5zuPS3L3Hj4nk01mp6qBRPziAEyy413XgzoqRNGOOGwAtjT38Ik07Axpf6t4U74Xa/bU/Z0b+fmfqayR0Qa4aaRn863uCyWfOgZTXTH1jC9w5ayXG3vU48kSIacrjzrP24889vcdAuLjvWO0xtqCfpWjZsiZO2/ih2vqAp+980vT7C/V+fTzwxPkHWSANkZROfGCXdV9wYfvHE6xy372y6w2HaU0nueeJ1Lj36YwU9fSzLD0I2d0AcsgUGxI4Ln/kmfOKUAVsP4ehibkVx3Nw7HlTzaK2Xgl3/FT7zjf62+8wv/HKpWAqoR6rXn9qXcGuHPSwUidDGFLz29RNRK5HhhevhwIthyzqIxPyprCuX9I/aLVxK+MWVXDD/bJLGoSNehz34Uszm1+GxH0DnRli4DF68l9SiO9lipuTcbuX9aFQjWVJ02UFIc32Ubx64Ox+ZGWNmzxs4d5+MG+wZzYJr/Cnbcxdgp++OCUVh90Ngr2OgexN0tcIz18CB34fFq6C30z+pye4bi5bBkb+ArS3+Y1vW+aPcz1wNz93OjDrDrKZaXm/tIhJy+O+1G/iPT3iEVpzY9xrdx97GW2YO59/5d1ra4hwyt5mfHTGLKCk8N4Jb3wzGyRlYzZwSJZ5Is6krMaageqQBsrKJT4zhLnRMj0WKOnLtWY/zv/AhnO5NNBqPupDD+V/4EN4Ipts6WJpNO5gEmAjQDGy7jikTJpwjkEqZMOFC3jg6FT52HNy5MKuv3uaXTwTPg+5WJdocq0i9/3dbcdrAv2OkvtQ1K55IDD519tC2G4mVumYyBur9IxWMUCe3MUIN0OZOJ9L5LtZqzZmUWLrXDx4e/A5saekPGMC/XbkE5i3BuGEiXe8wffnhmF980j/+ixf72w+tXIzd72y2NuxGbbKd6Tmm6+0Y8ftHzhN1z4PODdC+3r/1qniNlBSP55Hu2EBj4n1uOm4OPzzqo3x/1Yu8++56nLtPHtiuV53njyw/+B3MVR+H1bfCHofA7cfCzYf466g/9x2It8OyBdD+9tC+sWKxn7wve71bMg4HXAjfWEPztCYeO3M3dghtZdmf3uCrn5raP8U8eI26e0/D6d4UbKvlX4yqW3YI7lUfI3zLQXgbXqI93pszsPrnhk5efHcrb2/qoqWtm1RqdP1mpOtdM9N5M89RNvHiyHehw/O8oq9hd4GdU2/xsYePZfbST/Oxh49l59RbFHzJxPP8WR83fQmu3Nu/3fhSQZ/tbUwZmO8g2CaurcD12/Ru7Q/CIOirpw3d07gYPA82/mPQv/sf+k4bjUQHvHAPnLwSvr7av33hHr+8WqUTudtuurDs+lKeFFCPVMLfWzrpbjugTtVux/T0Rl7dqP2opcSyt6bIZDieNa//8fZ10PUBpNP+9g2DE4TM/7a/tVaqh7rUVmqdVM7peole/8RwyIn6GE68RPoE7Sh8y0HscMun+Mhvj2Hn1Fs014eZUWfyrIPelLU11sWA8U/aznrM7wdeon89W76s4N0fDL0A9e7f4bZjqOl6h5qnf8JHfnsMvziwhimh3GsC95jm8t9f/TC3LZzdfzFq1jw45MeEUl00Jj+guX7g2FxLW5zmhiiX/vYljr/uGU6+6VnWbhxZUJVJRJZIpbnzrP04eO4MYNsBcvZ03qcv+AL3nTd/YqYhV7iRJH6D/Bc60pacF1g2dY3fSfcUbwuNqwZeQGpctYSp3pbCXqC7tX8v9eD5LD/JL9+GnrTljIe6WHPQStYv+QtrDlrJGQ910VPoigIv5V/oPeF2Pw/ICbf79ydi2mzXRlg+6OLd8pP9chkRaxzY62h/5k/nBv92r6P98mqVzrP+Xwn1KpqmfI9Urx8cp51tB9R1U6bT2PECd7y8gT1mNhS7ZiL5GSf31hR/uARaVvv3k3E/EchB/y98+mx/H96W1f7xDdvDKSvBCWHSCd7qhA/nmK4Xi4Y5ZG4z3z7oIwNP1POdeJ31mLbTksLlaEeNq5bw6yUP+qPIp6yEcJ2/1v/pKyE201/zfM6TYIHHLxnaDxYt80/E29f15QgYsp6va1CAkB18r1jsB+jP3U5k5Sl4Sx7M+RrR9teYfcdC+Moj/cH0Fy/uq4vTOIdbFizljIfo23t6VlMt6zZ1DwiqvnrbmoKXU+SaGv/dL3+UHx65FzURd5t5DjLZxMdTNW/FNZr10PkSe1lraa4PD8hTcdmTreO6hj2cbx0zBZ7Yp3LvY01q20F/TdiluaGGD2gkZcO0k6S5oYOacIGBVKgGDvzhgC3tWHCNX15kNhnH5Ph322S8gMnuMoBx/d0YHvzOwL9jNa+hzrduXOv/K1oVXwIqkmCEOlXACLXbMJN608NTf1lDWluNSClZm3trivnf9j/Ij74OaqbC0iPg6nnwwDfhS5fAkgf8ACA2w58ue9UniN5+BM1spmPhigHT9Tjqauofv4hrD6plzxmxgSeQYzjxEumTpx258c24nRv6p2U/cpF/UnbY5eCE/OD3xXv9JHuD+8GKxfD5C/z7T1/pX2jKbtcLl/oJ+bI1zvGD78xrZE6E6mf47X7R7UNf46kr/PtdrX7Z/G8PqUvjqiX8ny/7SdJmNdVy3an7UhN2+rak+8TsKfz8iB1pSr5f0LKJzPrc5voo/3HIntz1l7f554ZO3t3SQ2dPasK3wKr2rbjyrYfe1JXwtyjc0gKb3/Rvgy0L880EqIs43HFkw4Btpe440i/Plkp5vNse5+1NXbzbHh/RkgDrRvxkldmjvHsejnUKm9Zv3UjObemsu+3nT6uNcMGhe7JjqIMdzQfsGOrggkP3ZFptgUsKbHroVnerzpuY7NCZgCibAqLR8ZK5/47VnK3djfjfT9nfEQuu8culYmmEeqSCgNor4Cpo5/R9ANhjy//lxv+ex7mf/1BRqyaSl82zNcWMuXDqvRCOwS2HDPxS+825/pZCS4/sH9Hu2gAtq5n6m8W8csR97HH6wzhbW/wgIRjtdja8gD3zMWjIGnkORXJfkQ3pC0RGIF87ijX7axiPvtYPdDe+7F9Ean3FH7FOdvvZgL3cSxWY9iH/dVpWw7PXw5LfQsd7frte/St/VHvDC/0jKEdfB49d3P/+bjgYBbdwy5f9Ee/D/wuadvVPslMJ/7WhP2gPRXPWZbcp8MwFB/DKhi6+/5sXae3s5Yrj9uGB51r4z096fYkAaZwDJ94FM+biYXKO+mbW537/iLks/dOb/QmorEf7Bw6p9Ex22W7ipnFX+1Zc+dZDu6Rhw8tDEy/N3Bvc3KdhU70tOPcOXGdZd+9p1Jz5GOCff6RSHq9s6ODc29f0jW5fd+q+fGRmA6FQAeMlobCfC+DuU/rrdcIdfnkB2pypOAuW9k8bD9ZBe85Upm3juVt7E2zPB8SmWTAOs62ly37A1t4I0woZZS7ltFknDMfdBPec1f//dtxNfrmMTL5tswrdj7xSRWL+d0Tm+0kJySqeAuqRCpKSpQsYoU7EdqAnthPHJp/nqIdf4VO7NLHvztv6mhEpAiecOxBxw/6I3tHX9j+W2SKotsmf6h1sEcT9X/fXnN59KtTPYLemMMZL+kHH01f2Bwzt60gle2nr6GG7WNQ/Wa9r9k/+M9N1M8FAXfO4/ROreSqpBHK1o+Nv9dc4Z08ZPPnXEN88sOyYG2Dq7Pz94JR7/Od0tfprIW8+pP+Y1lf68w+0vQnh2v7nLroNfneRP/r9yEX+a7evgzsW+o+fco9/IeCE2/v7ybPXkz70cj8T+aC6mM2v84Hnccat/+wrvuCe53ngjD1ouvOwIcsm7JmPsbazNuc040jI5ZC5zXx2ZprPH7Ejofh6Qg+f2peZv33BUrbE5tIUK3CabAGZjYfrh9W+FVe+zOiN6c25kxCd8TBew045p4nvHu3FyRFoeMnevqmFGzt7+4Jp8P8vz719DSu++hl2bBx+JxIAN9UDT17ut+3M9ohPXo775Z8U9O/t7E3zrYe6+N5BK/unpT/Uys9P8pi2jfigwXYRTm3xZ4gEfTS2aBkRO5XMBYNhhaJ5LtIW/8KMF6rBjTYODIiijX550d+9ypTw71gyybh/oXbwtlmf+49S10zGQFO+R8gG22aZ8La/rAA6mj/J3skX2DW8mRV/bSlm1UTyM07uKUaeN3DtaGZd5yMX+YH2bcf492fN6183OmseHPhDIrcd4WdOfuSi/mOC137+vW6OveZP/dM5HccfDT/rMfj2i/7tjLnjts2I51ne3tTJ+++uw2tfz/vvruPtTZ3lM5VUGc7HR9COek//Pd43/+6f0PZu9S/yZAcrW97uTzSWKbvvHMAOndJ91NX+yHImiH76Sog0DJzS2bLab+etr/iB8solcOwN/trpp/43rH0wf0Kz+Ga46hP9/eQz38Q75Ed0JMCecPvQujx1BY2Rge2jpS1OQ55kZ14qd3bwTV0JpteF+OWXamm4/VBqNvyN0MqB/0+Nq5ZQl2wr7P/e87CDEgvaQYkFtzWle6SZxitNvszoxss/mppv1D5lwjmnFadN/yhoMu3lvECRSo/g82W/r/Z/3j9ykX+/QK4xtHYmOe621/nc9a9x3G2v09qZxC3gOmYo1d0fTEPf8otQqrvAdze5v9MmYBWzk+yCO4/3PwtuPdy/vfN4v1xGqHR/x5IJ1cA+J/rbZl09z7/d58QJWf8vxaOAeoQS8Q6S1qU+UtjgftusL4Fx+V/1K/ntC++QGOWWJyJjkuqBx3/oj0Sc/qB/+/gP+9cpZaahfv6C4ddax9v8Ywaveco6ZtORS7nsydahWWkdx09A1jjbvx3HPTu3xHtp6nx1wPYvTZ2vsiXeO27vMWrKcD6uPAwb0lN4hxl0RWb4o8WDg5VwXe4AJtntT+nO7gfPXu+PGMTb/PWkX7wYnrvD33d9wDroZfDP3/W/VjoFPe1+MA39F6WyZSc0a18Hz16P3WchzrKjaLz2Y5gnr4BT7+vPOP6HS6BzI+2JgX1jVlMt1o3mDbDyjfo68Q/6t/DKE/CHbGFTZG1XK2ZQQjiz/CRsVsK2YdcQU/1bceVbD23c3MExbjjvqP1WZ2rObaW2uo19x4VdJ+cFipBb4GerTef+vC9wHXJtxOUnx+8z4O/5k+P3oTZSwAWSfMsvCs3Sne87LdVT2PPHQlmax08p/46lUsr1/1I0VT3l2xhzKPBz/O0Wb7LWXj7W10x0bCZJLVNrCrt6lqxt5oNdjmD/N+7lCq+H19fuxEf32mes1RAZGceFzo3+SF5G4xx/ZA78Ebg/XAILrs19ohBr9hMt4eUNVlLNe9Fx8sOctfLNvizFEzWdsy7ZRjTH9i+9p/8e2KHo7z+sYmY4L2AKbrXZ1JXgsgdf4vtH7MV7tokP1/UOnTKY7M49jXDru0OzfC9c5k+3a30Fjrnen5VxyI/hjz8ZOBX2jz/xp3U/F4wqOyGo267/fTIXpXJl0oe+bbtMT7v/uk9f6QfjG17wR9rvPtVP6HTiXdS72zOraf2AKcBufQx74l39QW0QYG1OPV710wAAIABJREFUxDh47gx+/1L/lj19o77ZSdzyZDB3Cpxa6aV6cfOMkGfCp21N6c4OOKt1aUbOzOj12/tLAwavoa7fnkh3Ouc08UjYZVPsw6w/7F4aIx7tCYe62Ex2rut/7bqIwzWnfJLz7vhbX1u55pRPDklcltcY16821kaYOaWGSxfsTV3EpTuRZuaUGhoLSSw21kzHTij3d5ozAae1ytI8fkr5dywVXZCpSlXbYo0xLvBL4CCgBfirMeZ+a+1LY3ldr30979npTB3BBfXW3Y6lMx3ioLfuw73nS7DdEzBzr7FUQ2Rk3IgfOKxcnHVCt8xfO5o5Mejc6E8Nz3WiUNMIqTjccIC/NVGOY3qdGi546J2+YBombjpnKN/2LwWOvhVVsTKcZ0a+B69LH8ep9OUokUrz+5c28rUDPsRH6y385mtDA9mps+H4W+DXZ/SXLbjGH/kAP6CNNcOUHeHZG/uDZJz+0dy1D/aPPmd85ryB27qk49gT7sDcfUrf2ujUaavwrCHkuji/u9AvzyyluHPh0GC7ZTV2+h6Yb78IoQimrpmdMTmDzraGD9MSBFiZ9aqtnau586z9eOm9jgEB+PRYBLojwwb89sS7MLHC8hikTDjnmu+UCfcF1PnWEGd/BhRjK66y54b8BGRnPOyfNLthP8h2Q0yPuTm3zZpSE6E+EmZjJEwq7THddZhRHx2QbKyrN83tz7zNLad/CtcxpD3LjX98g299aXca6wqp1zC5NQrgOIZdpsdoqAmP+AJJOlxPaNGyAWuoWbTMLy/kzcM1uS9ShCdg2qwTZGkevGVXgdnRJUukPvffMVJf6poVjy7IVKWqDaiBTwOvWWvfADDGLAcWAGMKqN2t63nHbkfjSM4HjEN8j6M59u3PcRffY8q958A5T+XN7iky7qznbxt08sr+JBjP3QHzv+FnNE73+sF0omvol9vR1/n79378JGicw5aa2XD0Mqb+pv9EyDvhTrzaaXzrwFjuE/sic/IkNil09K2oipXhfJLu7Z0J2i554GVuO3FX/0LQHy7xg+Sps6FuGpiQ354zSYPAH73v3Oj/P2W21fLSsN858Kmv9I+IZJY25PqbNWzvv2asGTrfw6tporVuN5LHPYBN9ZIgRJ0zk4QHtY5huwO+i7PhhZxbZPUl+XvkIlJuDeGp2/e9lQM5g87uhMeRt/xzSLnr5A7AByRxy2QwX3y//28NgvdCL750h5uI58jobMNNZGqab0/lapnSPSZuCKbOGlI83Ki945hhk4tFQi5/emMTK9b052eZ1VTLf4Y+UlidQtHcgeEIPjdHe4HEqZlKqncaoVPuAWPAWlJuFKdmamEvkOyBt//sb+1oPb+/v/I7+OjhI67LSMUjjdTGZmCykpLZ2AzikUYKuY4hgzRs70/39tL++Um1bx/l1uTudwUkO5byVc0R3U7A+qz7LcB+Y3pFa4l2vcs79kPsMor+Pm3adH7cdgaXb7gSXvw1/MuJY6qOSKFMXTN2n4WYrBEye8IdmFQPdLzrn5yEa/1ALNENp/3GP8np3AhPX0XqXy9gE1NoOWgll93fyoyGKFed/nuipCAUwalrpsFx+OgO0ZJM5zSx5iHTYUcy+lZUxcpwPkn39s4O2i77w0YuWXgHkZWn9E2ZTp9yL14kgolMwWmei7Uexjh+Isklv8VkLh51tfqBcyTm/3hpP7hcdLu/Z/TgUe+FS7GhWmj+KBiDjdTT7k6jJuSQrp9ByrPUOIZoxGVmbdDuG/byL3AkuvMupUgtuhO3vrC2MNwIcM6gJjsZ4BiXBUytjfJ2/e4DpiDX12/PzrX97zsZpnQXw2iD0rFewDDJrPWrmaUNj/8Qc/ytI67LSDmuC1NnkexsxaQTWDeCW9/slxeiZirsvL+/rWP2yGahAfkYtMU9Hn4rwjF7fBTjpbBOiPv+mWL/3T3q/n/27jxOrqrO///rc2vpfUvSCUsTAhKRDKBIUEZmFAUhLBpQwyoERFGB7zjzdcTl5+iM4Iw648gooiIoYQnbKLKLEYkzX0bRIA5KIBJEQ4CkO0l3p/da7vn9cW+lqzvVSXVVd9fS7+fj0Y/uOnWr+lT3uefWp845n6OYaFIsXo8bGhmTgsyF5dXLBe+1srPENy4IyqViVXNAnesKvltrNbNLgUsBFi5cuNsDxhjuIZYe5GU3jyMLCKjfMA+ufWUp/zR/ETX//VU44qyqnpop02dS7RYgGsPaD4OLHoLwDUA61kD0viuCkee6tiAxU/8WqJsDfa+Saj2Q7TaX/mOvwkXncdFNo3udXrnsMAZiTdQ0jH0TWLLpnJ6HjQscJjP6Nt11m6qgZowK3Nt70u02h/FB20DcI3bJT7F08LeN1LdjWfsymxkRgxrz2JGMUTOynX0bwJr3By+GDfcGMzO8KM6roa92H5pO+TI21B3M6Ej0Qd8W+K9/Y2TZV3lqRw3ff/xFvnjmkbvaestEg4iZRHz9W3P+r/zmDrymffMOIgoKoDJ1KJLnGQfObWR77UISqTT7TBAsV+uU7qlou1Ot6A8wovHc61dnqA/xIhG8rJkZk5EaGSRm0bEjm0M7g/La5imu6VjxaISvPfYiH//h2A+27jlsv2n9vYUox3Y7RrQGq2sL2iEAFtwuh9ll0ySZ9on/5hY46vzRGYO/uYXkmy+jfK/esjfmXHV+ImJmfwn8o3Pu5PD2pwGcc/8y0WOWLl3q1q1bN/GTvvo0fOev+VjqY1x6yuQHu7cOwgcehe+/7te8/U9fC9awLlk+6eeRslLyoZe9ttuJ5FiDmz7rVlJ18/hD5yCf++kWuvqTfPfCpTTWRNjYObAr8cyBc+tZNLdBI0+lVNwa6pL/4wput0XI3iM57TtW//JPvHNRhH0bPRobGrjuiR6Wv/EADp3fgNf17Ji/bc/yVVz80MCuc+LQBU35t/8pXO8+y/dbL4sXWoq2Oy0qOA9DKpXCup4lcud5o9evs1fj2g8jGp3esaLM9nC59n7fw7lY8rZbNe22wqWSKeh6dnQHhtaFpM5aDe2HEY2V3ThnydttpajmgDoK/AE4AXgZ+DVwnnPumYkes9fO5rmH4I5zudi+wN+feEhB9bpsLbTWwI+8T2D1c+DD/x1MrQ0553hx2wAHzWvATO24ApT8n1TURTIrS7SLxOnxWkikHGkXtMXMG3ZgNr+JL1+FZ/ku+T+v1G/usgPTzCi253mjbTvHuTGY8Atv/7MwI/s0KHm7hdK33SlVwe0ylUrh93fh+Ul8L4bX2D7twXRGAR9slbztVlW7rXCpZIr0wGjbjTS0l2MwDWXQbitFWf73poJzLmVmVwCPEGyb9b09BdN56Q2WZA/EC1/7eNoiuO538OLR7+HgZ74BLzzK8IFv577fvsJI2ufRZ7eydkMX/+cdh/Dxkw4tqroie5U1HdSAtj0cWo3TOCveFE3nnY32OjU5x7nR1lDUL9T/SspPBbfLaDQKraXZFrFalzbIzIjGokRL1HZlelRtQA3gnHsIeGjKnm/HiyRcjDnNha/POaED7nge/v7Pb+YH9bfD/7uGjz8xhwd/9yoANVGPQxc08Y2fbWTZ4fvwF/u1wKYn4GdXwWtPhmMvU2p9ERERERGRMlAZ83rKROIPP+Updwivm1v4DIjaKFxyGPxmm8c90VOxP/0373vub/l9w0f5Q9NHuOsvX+ITJx9KfTzC1x99HjY+Crcsh5efhJ98Fn594xS+IhERERERESmUAup8bXuemu7n+XH6GA7b07zYPBzfAecshk92vpPV6RN4W+Rp0vMOI9WwD6//9Sc4/Lf/xIWH+hz03HfxbzuLRP0+rDnqm/TPPQJ+/iUY7p2a1yQiIiIiIiIFq+op31PJ/XY1BqyLHcP7illHF7rgdbDikAhml7DBnY8frQU/xYKNd7LPH27jU9wGMXgktZRPbP0wO7fWcLidzX01n8N/+DNEzriWZ17Zyc6hJEctbKMuPolp4C88Fox293fC4e+Ft38aaltIpX3WbujiV3/awYvbBvj9y73s01LLp085jDcdNKeg1+n7jnV/7sYM3nBAK7FIYZ/h/GnbABs7+zlm0Rxa6mMFPcdeDffCz74IT98Jcw6Cd14FB/319PwuERERERGpeAqo9+a5B3n+sVtZvPUhHki/mROXzGWqkm/Xhn99n9rgBy/K1teeT89+b6Nhx+/oaVzMr7sP4RyDv5gDv9hyMN968XQu/99b+Z/f/4GnR+YzQpynohEOmddAY02EiGeAw3A4l52ez+G5NAv617Owdx3d8X3pqjmYxU98m8Hf3Mn6lrfxbI9Hz3CaJvN4c02ck+ti7NiW4PEbUnQuaKSlLkbEMzyyNvR2Lmtzb5cpAiCZdvxp+wC9Q0nA8b+xCAfPb6QuGhn3Nwyfw+36Ket5HF19I7zUHez3uDFivGZeI811UY7saKEuluODhAkz1+8ho/3wTnjufujvgkV/Bds3wqrT4eC3w/5HwzGXQHP57TEpIiIiIiKlU7XbZhXCzLqAP+dx6Dxg2zRXpxzpde9um3Nu2UxWZrxJtNuMSv4/qu5ToxLbbUY5/R3LwWz6e5S83cIe2261/S/0eqZOydtuEX3uTKm29paPcn/NJW+3lUIBdQHMbJ1zbmmp6zHT9LqrQyW/HtVd9HccS3+P8lFt/wu9HplJs/H/Mxtfc7VSUjIRERERERGRAiigFhERERERESmAAurCXF/qCpSIXnd1qOTXo7qL/o5j6e9RPqrtf6HXIzNpNv5/ZuNrrkpaQy0iIiIiIiJSAI1Qi4iIiIiIiBRAAbWIiIiIiIhIARRQi4iIiIiIiBRAAbWIiIiIiIhIARRQi4iIiIiIiBRAAbWIiIiIiIhIARRQi4iIiIiIiBRAAbWIiIiIiIhIARRQi4iIiIiIiBRAAbWIiIiIiIhIARRQi4iIiIiIiBRAAbWIiIiIiIhIARRQi4iIiIiIiBRAAbWIiIiIiIhIARRQi4iIiIiIiBRAAXWWZcuWOUBf+prMV8mp3eqrgK+SU7vVVwFfZUFtV18FfJWc2q2+CviSPCmgzrJt27ZSV0Fk0tRupRKp3UqlUtuVSqR2KzJ9FFCLiIiIiIiIFEABtYiIiIiIiEgBFFCLiIiIiIiIFEABtYiIiIiIiEgBFFCLzCJP/HE7L/cMlboaIiIiIiJVoWQBtZn9ycx+Z2a/NbN1YdkcM1tjZs+H39vCcjOzr5vZRjN72szemPU8K8PjnzezlVnlR4fPvzF8rBVdad+H/q3Q81Lw3feLfkqRmXT29b/k7f+6ttTVEJl9dP2QcqR2KSJStFKPUL/dOfcG59zS8PangEedc4uBR8PbAKcAi8OvS4FvQRCAA58H3gy8Cfh8JggPj7k063HLiqqp70PnerjhRLjm8OB753pdfKRiOBdsKZhIq82KzChdP6QcqV2KiEyJUgfU4y0HVoU/rwLOyCq/2QV+CbSa2b7AycAa59wO51w3sAZYFt7X7Jz7hQuiiJuznqswg11wx7nQsym43bMpuD3YVdTTisyUgUS61FUQmZ10/ZBypHYpIjIlShlQO+AnZvakmV0ali1wzr0KEH6fH5bvD7yU9djNYdmeyjfnKN+NmV1qZuvMbF1X1x4uIqnE6EUno2dTUC4yw/Jut1n6hhIcaS8QnHoiM6+QdlsVdP2oeFXZdtUuq15VtluRMlTKgPo459wbCaZzX25mb93DsbnWP7sCyncvdO5659xS59zS9vb2iWsQjUPrwrFlrQuDcpEZlne7zeL/4SfcV/MPnBN5jOGkRqtl5hXSbquCrh8Vryrbrtpl1avKditShkoWUDvnXgm/dwL3EKyB3hpO1yb83hkevhk4IOvhHcAreynvyFFeuPp2OOf20YtP68Lgdr06KKkMI/3dAPyV9zu6BzUCITJjdP2QcqR2KSIyJaKl+KVm1gB4zrm+8OeTgC8A9wErgS+F3+8NH3IfcIWZ3UGQgKzXOfeqmT0C/HNWIrKTgE8753aYWZ+ZHQs8AVwIfKOoSnsezF8CH/xpMB0qGg8uOl65LUMXya3f1QLQYdsYTirpjMiM0fVDypHapYjIlChJQA0sAO4Jd7KKAqudcz82s18Dd5nZJcAmYEV4/EPAqcBGYBC4GCAMnK8Cfh0e9wXn3I7w548CNwF1wMPhV3E8DxoXFP00IqUwMjQAwP7WxTZN+RaZWbp+SDlSuxQRKVpJAmrn3B+B1+co3w6ckKPcAZdP8FzfA76Xo3wdcHjRlRWpEomRQQDabScvpzRCLSIiIiJSLM3rEZklXGJo18/J/h17OFJERERERPKhgFpklnCpkV0/R3Y8X8KaiIiIiIhUBwXUIrOEpYZ3/ewGu0tYExERERGR6qCAWmSWyA6o01nTv0VEREREpDAKqEVmCUuPTvn2k8N7OFJERERERPKhgFpkloikR4NoP6kRahERERGRYimgFpklLD1CwkUAcBqhFhEREREpmgJqkVnCSyfotwZAAbWIiIiIyFRQQC0yS0T8YfoIA+qUAmoRERERkWIpoBaZJSLpEYapIeU8TCPUIiIiIiJFU0AtMktE/RGSFmeEOKQVUIuIiIiIFEsBtcgsEfUTJC1GwmJYamTvDxARERERkT1SQC0yS0RcgiRxEsTw/ESpqyMiIiIiUvEUUIvMEnF/hKTFSBIfsye1iIiIiIgURgG1yCwRdQlSXpyExYikNeVbRERERKRYCqhFZom4GyFtMZIWI+I05VtEREREpFjRUlegIvk+DHZBKgHRONS3g6fPJqS8xVySlMVJESeSVkAtkpP6d5HKoHNVKpXabtUp6X/PzCJm9pSZPRDePsjMnjCz583sTjOLh+U14e2N4f2Lsp7j02H5BjM7Oat8WVi20cw+NWWV9n3oXA83nAjXHB5871wflIuUsTgJ0l6MlMWIOa2hFtmN+neRyqBzVSqV2m5VKvXHIR8Dns26/WXga865xUA3cElYfgnQ7Zw7BPhaeBxmtgQ4B/gLYBlwXRikR4BvAqcAS4Bzw2OLN9gFd5wLPZuC2z2bgtuDXVPy9CLTwjlqSZC2IKCOKsu3yO7Uv4tUBp2rUqnUdqtSyQJqM+sATgNuCG8b8A7gP8NDVgFnhD8vD28T3n9CePxy4A7n3Ihz7kVgI/Cm8Gujc+6PzrkEcEd4bPFSidGTIKNnU1AuUq7CKd5pL07KixPTGmqR3al/F6kMOlelUqntVqVSjlBfA1wJZOY4zAV6nHOp8PZmYP/w5/2BlwDC+3vD43eVj3vMROW7MbNLzWydma3r6srj06FoHFoXji1rXRiUi8yQSbfb5BAAvsVIW4yYS05zDUV2N+l2O9PUv8sEyr7tzjY6V/OidluG1HarUkkCajM7Heh0zj2ZXZzjULeX+yZbvnuhc9c755Y655a2t7fvodah+nY45/bRk6F1YXC7Po/HikyRSbfbVLBNVjoSJ+3FiKNts2TmTbrdzjT17zKBsm+7s43O1byo3ZYhtd2qVKos38cB7zazU4FaoJlgxLrVzKLhKHQH8Ep4/GbgAGCzmUWBFmBHVnlG9mMmKi+O58H8JfDBnyo7n1SOVDBCnQ5HqOMaoRbZnfp3kcqgc1UqldpuVSpJQO2c+zTwaQAzOx74e+fc+WZ2N/A+gjXPK4F7w4fcF97+RXj/z5xzzszuA1ab2b8D+wGLgV8RjFAvNrODgJcJEpedN2UvwPOgccGUPZ3IdHPJYQzwvRi+F6cGBdQiOal/F6kMOlelUqntVp1y24f6k8AdZnY18BRwY1h+I3CLmW0kGJk+B8A594yZ3QWsB1LA5c65NICZXQE8AkSA7znnnpnRVyJSRlKJYWKA78XxvRg1lsT5PqZPREVEREREClbygNo5txZYG/78R4IM3eOPGQZWTPD4LwJfzFH+EPDQFFZVpGL5icHguxfDeUHii+TIEPG6hlJWS0RERESkoml4SmQWSCWHgx+8GH4kBsDIyGAJayQiIiIiUvkKDqjNrNnMXpOj/MjiqiQiU81PBvsbOi+K84KAOjk8VMoqiYiIiIhUvIICajM7C3gO+IGZPWNmx2TdfdNUVExEpk46FQTUeFEIp3ynEhqhFhEREREpRqEj1J8BjnbOvQG4mCBh2HvC+3LtAS0iJeTvCqgjEMmMUA+XsEYiIiIiIpWv0KRkEefcqwDOuV+Z2duBB8ysA3BTVjsRmRJ+Mtgmy7ICao1Qi4iIiIgUp9AR6r7s9dNhcH08sBz4iymol4hMIT89AoB50dGAekRrqEVEREREilHoCPVHGTe12znXZ2bLgLOKrpWITCk/FYxQ40UwgjXU6YQCahERERGRYhQ0Qu2c+1/n3MYc5Unn3G2Z22b2i2IqJyJTw08FI9R4USwcoU5ryreIiIiISFEKHaHOV+00P7+I5MGFI9ReJIJZDaARahERERGRYk13QK0EZSJlwE8HWb7Ni2LhtllOAbWIiIiISFEKTUomIhUkM0JtXhSLBiPUflJTvkVEREREijHdAbX2pBYpAy6dmfIdJRILRqjRGmoRERERkaIUHVCb2YFmdmL4c52ZNWXdfUGxzy8ixXPpBL4zPM8jEg0D6qSmfIuIiIiIFKOogNrMPgT8J/CdsKgD+FHmfufc74t5fhGZIukkSaJEPYhGoyRdBDTlW0RERESkKMWOUF8OHAfsBHDOPQ/ML7ZSIjLF0kmSRPAM4h4MEcdLaYRaRERERKQYxQbUI865ROaGmUVRZm+RsuOyRqhjHgxTg6WGS10tEREREZGKVmxA/XMz+wxQZ2bvBO4G7i++WiIypdIJUkSIGkTCEWoF1CIiIiIixSk2oP4U0AX8Dvgw8BDw2WIrJSJTzA+mfEfCM36EOJF0MOV753CS3qFkCSsnIiIiIlKZigqonXO+c+67zrkVzrn3hT/vdcq3mdWa2a/M7H/N7Bkz+6ew/CAze8LMnjezO80sHpbXhLc3hvcvynquT4flG8zs5KzyZWHZRjP7VDGvU6TSWTpJ0kWJhhvZjRAnEo5Qn/71/8cbr1pDHqeuiIiIiIhkKTbL9+lm9pSZ7TCznWbWZ2Y783joCPAO59zrgTcAy8zsWODLwNecc4uBbuCS8PhLgG7n3CHA18LjMLMlwDnAXwDLgOvMLGJmEeCbwCnAEuDc8FiR2clPkQqTkgGMWA0Rf4hk2mfTjkHSvuOZV/I5dUVEREREJKPYKd/XACuBuc65Zudck3OueW8PcoH+8GYs/HLAOwi24QJYBZwR/rw8vE14/wlmZmH5Hc65Eefci8BG4E3h10bn3B/DpGl3hMeKzEqWTpAIk5IBJIgTTQ+zYUvfrmNe6Oqf4NEiIiIiIpJLsQH1S8Dv85nmPV44kvxboBNYA7wA9DjnUuEhm4H9w5/3D38X4f29wNzs8nGPmag8Vz0uNbN1Zrauq6trsi9DpCQm224tHKHOTPlOWA1Rf5hNO0b3on61V0nKZHqpv5VKpbYrlUjtVmRmFBtQXwk8FK5j/r+Zr3we6JxLO+feAHQQjCgfluuw8LtNcN9ky3PV43rn3FLn3NL29va9V1ykDEy23ZqfIEl0V1KypMWJpYfZunM0iN6igFqmmfpbqVRqu1KJ1G5FZka0yMd/EegHaoF4IU/gnOsxs7XAsUCrmUXDUegO4JXwsM3AAcDmcK/rFmBHVnlG9mMmKheZdcxPkSRKXfhRU9qrIZoeYevOEaKesaC5VgG1iIiIiMgkFRtQz3HOnTTZB5lZO5AMg+k64ESCRGOPAe8jWPO8Erg3fMh94e1fhPf/zDnnzOw+YLWZ/TuwH7AY+BXBCPViMzsIeJkgcdl5hb9Mkcpmfoqki9AQBtR+JE48NULnzmFa62O01cfGjFaLiIiIiMjeFRtQ/9TMTnLO/WSSj9sXWBVm4/aAu5xzD5jZeuAOM7saeAq4MTz+RuAWM9tIMDJ9DoBz7hkzuwtYD6SAy51zaQAzuwJ4BIgA33POPVPUKxWpYOYnSVsMywTUXpwaRtjaO0RbfZzG2ihbd46UtpIiIiIiIhWm2ID6cuBKM0sAybDM7S3Tt3PuaeCoHOV/JFhPPb58GFgxwXN9kWDq+fjyh4CH9vYCRGYDz0+Som7XbRepIYrPtp39tDU20hCPsnNYWb5FRERERCajqIDaOdc0VRURkenjuSRpi+y67SJByoP+/j4OnN9GTdSjbyg10cNFRERERCSHYkeoMbN3A28Nb651zj1Q7HOKyNTy/BTp7NM9WgNAcniQtoYguE6kfYaTaWpjkVxPISIiIiIi4xS1bZaZfQn4GMEa5vXAx8IyESkj40eoLRyhrrMR2urjNMSD+3YOJ3M+XkREREREdlfsCPWpwBuccz6Ama0iSCb2qWIrJiJTJ+JSpLJOdy8aBNS1JGirj9E3HGQr6xtOMV8LOURERERE8lLUCHWoNevnlil4PhGZYhGXws8aoY7HwxFqErQ31lCfGaEe0gi1iIiIiEi+ih2h/hfgKTN7jGDv57cCny66ViIypbxxAXV7Y7CGus5GWNBcS99IkJBs57ASk4mIiIiI5KvYLN+3m9la4BiCgPqTzrktU1ExEZk6EZcibaOne2N9LQBNDOJ5phFqEREREZECFBVQm9mZwM+cc/eFt1vN7Azn3I+mpHblyvdxA134qRFSFmMw1kZLXQ2eZ6WumUhOEZfC90ZHqFN18wB4/+uCNlsfD7oCJSWTqub7MNgFqQRE41DfDp437hDH9oEEiVSaeDTC3Ia4+naRMuSn06T7u7D0CC5SQ6SxHS+iXSqk/KntVp9ip3x/3jl3T+aGc67HzD4PVG9A7fu4zvXYHecS6dlEpHUhQ8tX8efGxRw4t1FvvKT8OEeU9JgR6nSsiXS0ntfGungRaKjJjFBryrdUKd+HzvVwx7nQswlaF8I5t8P8JbuCat93bNjax4duXsfm7iE62ur47oVLOXRBk/p2kTLip9P4W9cTu+u8Xedz6qzVsGBkz/b2AAAgAElEQVSJAhMpa2q71anYpGS5Hl/03tZlbbALy7whA+jZROu9K+nfsYXtA4nS1k0kl3Qw6uxnBdSYkajfh9q+PwMQj3hEPdMItVSvwa7RYBqC73ecG5SHtg8kdgXTAJu7h/jQzevUt4uUmXR/F9FMQALQs4noXeeR7u/a8wNFSkxttzoVG1CvM7N/N7PXmNnBZvY14MmpqFjZSiVGT4KMnk20xn0SqXRp6iSyJ+kgGMhOSgaQqJtPU+c6Dn/4vey//rs0xE1rqKV6TdB3kxoNlhOp9K5gOmNz95D6dpEyY+mRnOezpfXhl5Q3td3qVGxA/X+ABHAncDcwDFxebKXKWjQeTBXM1rqQnoRHPKqpGlKG/Bwj1ECifh8iqUGatj3Fgb/5EidF/1dZvqV6TdB3E+7JDhCPRuhoqxtzSEdbnfp2kTLjIjU5z2cXied+gEiZUNutTkUF1M65Aefcp5xzS51zRzvnPu2cG5iqypWl+nbcObePngytC+lZvorGOfswt0Eng5ShdBAk7x5QLxhTfkjkFfo05VuqVX17sGY6q+/mnNuD8tDchjjfvXDprqA6s4ZafbtIeYk0tgfrTrPO59RZq4k0tu/5gSIlprZbnYrN8v1a4O+BRdnP5Zx7R3HVKmOeh81fgrvkp7uyfLtYGwcqy7eUq8w0Im/sKFuyNsj03d1xAi1bHqeDbZryLdXL84IEZB/86YRZvj3POHRBE/dcdpyyfIuUMS8SgQVLSF68BksncJG4MiVLRVDbrU7FJhC7G/g2cAMwexaZeR7WtIAIEAFqSl0fkT2ZYMr3QNsSXjri/9A3/xjqezawb6KTXgXUUs08DxoX7OUQo71JvbpIufMiEbyWfUpdDZFJU9utPsUG1Cnn3LempCYiMj3CLN9uXFIyzNi5z18CwWj1guGt9GkNtYiIiIhI3opNSna/mV1mZvua2ZzM15TUTESmRmbbLG/iz8+Sde3MS3XSN6IRahERERGRfBUbUK8EPgH8D8F2WU8C6/b2IDM7wMweM7NnzewZM/tYWD7HzNaY2fPh97aw3Mzs62a20cyeNrM3Zj3XyvD4581sZVb50Wb2u/AxXzczLYKT2Smzhnr8CHWWRN08atwwtYke0r6boYqJiIiIiFS2YrN8H5Tj6+A8HpoCPu6cOww4FrjczJYAnwIedc4tBh4NbwOcAiwOvy4FvgVBAA58Hngz8Cbg85kgPDzm0qzHLSvmtYpULD+Yxu32NEJdG2SX7LBt9Gvat4iIiIhIXgpaQ21m79nT/c65H+7l/leBV8Of+8zsWWB/YDlwfHjYKmAt8Mmw/GbnnAN+aWatZrZveOwa59yOsF5rgGVmthZods79Iiy/GTgDeHiyr1Wk4oUj1OZNPEKdrAsyfu9v29g5nKSlPjYjVRMRERERqWSFJiV71x7uc8AeA+psZrYIOAp4AlgQBts45141s/nhYfsDL2U9bHNYtqfyzTnKc/3+SwlGslm4cGGuQ0TKzqTabSYp2Z5GqOPBxI5261FiMpk26m+lUqntSiVSuxWZGQUF1M65i/M5zsxWOudW7eH+RuAHwN8653buYZlzrjtcAeW7Fzp3PXA9wNKlS7V4VCrCpNptOELtbOLTPR1vwsej3XroH1FALdND/a1UKrVdqURqtyIzo9ikZHvzsYnuMLMYQTB9W9YU8a3hVG7C751h+WbggKyHdwCv7KW8I0e5yKyTTg4D4CJ7mMZtHolYM/PopW9Ymb5FRERERPIx3QF1ziHnMOP2jcCzzrl/z7rrPoLM4YTf780qvzDM9n0s0BtODX8EOMnM2sJkZCcBj4T39ZnZseHvujDruURmlUxAzR5GqAGSsWbabaemfIuIiIiI5KnQNdT5mmh6yXHABcDvzOy3YdlngC8Bd5nZJcAmYEV430PAqcBGYBC4GMA5t8PMrgJ+HR73hUyCMuCjwE1AHUEyMiUkk1nJT44AexmhBlLxFtqthy2a8i0iIiIikpfpDqhzjlA75/7fRPcBJ+Q43gGXT/Bc3wO+l6N8HXB43jUVqVLpRBBQ2x6SkgH4NS3Ms1c15VtEREREJE/TPeX78Wl+fhHZCz8VrqH29jxC7de2BGuohxRQi4iIiIjko6iA2szmmtk3zOw3Zvakmf2Hmc3N3O+cu6L4KopIMTJTvvc2Qp2Ot1BrSRKDvTNRLRERERGRilfsCPUdBJm43wu8D+gC7iy2UiIydVwYUJPHGmoAb6Bzj8eJiIiIiEig2DXUc5xzV2XdvtrMzijyOUVkCvmp/EaoUzWtAEQGt017nUREREREqkGxI9SPmdk5ZuaFX2cBD05FxURkarjUCCMuSiQyUR7AQGaEOjbcNRPVEhERERGpeMUG1B8GVgOJ8OsO4P+aWZ+Z7Sy2ciJSPJcaIUGM6F7O9kxAXTuyfQZqJSIiIiJS+Yqa8u2ca5qqiojI9AgC6ijRPQ9Qk4434WPUJxRQi4iIiIjko+h9qM3sPcBfAQ74b+fcj4qulYhMnTxHqDGPfq+ZhlT3jFRLRERERKTSFRVQm9l1wCHA7WHRR8zsnc65y4uuWSXyfRjsglQConGobwdvurf6FtmL1AgJF2UvS6gBGIy20JbsZjiZpjYWmf66iZSK+muRyqZzuDro/yhVoNgR6rcBhzvnHICZrQJ+V3StKpHvQ+d6uONc6NkErQvhnNth/pLcHYM6EJkp6TxHqIHhaAvt1kvPYJJ9WhRQS5WabH89k/XSdUFk78r1HJbJma3/R/X1VafY/94GYGHW7QOAp4t8zso02DXaIUDw/Y5zg/LxMh3IDSfCNYcH3zvXB+UiU8zSCRLkN0KdiLUwz3rpHkxMf8VESmUy/fVM0XVBJH8DE5zDA9qloqKUY1883dTXV6ViA+q5wLNmttbM1gLrgXYzu8/M7iu6dpUklRjtEDJ6NgXl483GDkRKJ5XIe4Q6XdPCPHrp7h+Z/nqJlMpk+uuZouuCSP5SQxOcw0OlqY8Uphz74ummvr4qFTvl+3NZPxtBcrJzgcuKfN7KE40HU1WyO4bWhUH5eLOxA5GSMT/M8p3Px2e1LdRakr6d3UD7dFdNpDQm01/PFF0XRPJnkdznsGmpUkUpx754uqmvr0pFjVA7534O9AKnATcBJwDfds79PLxv9qhvD9Z9tIYz4DPrQOpzBCWZDiRbtXcgUjKWTjDiYnvdNgvAaoO9qId7Xp3mWomU0GT665mi64JI/mJ1sPy6sefw8uuCcqkc5dgXTzf19VWpoBFqM3stcA7BaPR24E7AnHNvn8K6VRbPC5IofPCne08ykOlAxidhqOYOREomWENdRySPj8+i9a0ApHq3TnOtREpoMv31TNF1QSR/dXOgaV847asQq4fkYHC7bk6payaTUY598XRTX1+VCp3y/Rzw38C7nHMbAczs76asVpXK86BxQX7HzbYORErGSydI0ExDHiPUhCPUrm/L9FZKpNTy7a9niq4LIvnzPJhzMNQ26XypdOXWF0839fVVqdCA+r0EI9SPmdmPgTsI1lDPGr7v2D6QIJFKE49GmNsQx/Mm8SeYbR2IlIwXrqFuzqOvTtbOBSA+8Mo010pEdlMF14Wir40iefIxtrtWEi5N3EWYixWdaVdm3qzsM6qgr5exCup7nHP3OOfOBl4HrAX+DlhgZt8ys5P29ngz+56ZdZrZ77PK5pjZGjN7PvzeFpabmX3dzDaa2dNm9sasx6wMj3/ezFZmlR9tZr8LH/N1M5vSM9P3HRu29nHmdY9z3Jcf48zrHmfD1j58303lrxGZEp6fJJHnGmo/1sBOGmgY3Dz9FRORqqJro8wUtbXqoP+jVItik5INOOduc86dDnQAvwU+lcdDbwKWjSv7FPCoc24x8GjW85wCLA6/LgW+BUEADnweeDPwJuDzmSA8PObSrMeN/11F2T6Q4EM3r2Nzd7A9w+buIT508zq2DYxuNeT7jq6+EV7uHqSrb0Sdg5SM5yfyz/INdHnttI5ohFpmpz313erX92yia+P2AWWvld0Vcz6prVWH2fp/1LWk+hS7bdYuzrkdwHfCr70d+19mtmhc8XLg+PDnVQQj358My292zjngl2bWamb7hseuCX8vZrYGWBbuh93snPtFWH4zcAbwcOGvbqxEKr3r5Ac46oBmPnt8O23Jrbi+Wlz9PDZ0DuzqJDra6vjOBUczryGO53mzYzqLlI2IH+5DnWeT2x6dz4Lky9NbKZES8n1H79AI9cluoi6JF63BGtrxMTZs7RvTd3/3wqUcuqAJYML71J8Hxl8bIXiDnEilS1QjKVeZkclCz6dEKk17Y4z/OH0/5tcbnYOOq9d2qa1VmNnYZ/i+49WeQRrT3TRZipFUlFeTbezbWq9rSQUrp+UmC5xzrwKE3+eH5fsDL2Udtzks21P55hzlUyYejdDRFmzNcNQBzdywrIGj16wg9vUjsRtPhM71XLPmuTGfuH34lif57eZeTWeRGef5SRLE8sryDdAXX8C+/lacX70XNJm9fN/x5+39WOd6am46ich/HIHdeCKucz29QyMTjpbM1pGUyci+NmZ0tNURj2pvYBlr20Ducy17pt+exCPG908N3nsdsOpNHL1mBd8/tYF4RAFJJZmNfUbf8Ajzh16g5bZTqL329bTcdgrzh16gbzi/ti/lacpGqKdRrt7RFVCe+8nNLiWYHs7ChQsnOmyMuQ1xvvP+o/nwrU/y2ePbmXv/itFN2hvn4/W/yjdOPYSBZfvxSqKBzb0jfHvtC7TWxXZdNO657Djam2ry+n0i4+Xdbp0j6pIkiZLvB58jte3E+1Ps7NpM84IDp6C2IoFC+tupkkl8M5RMMYdeWu5dOdpv92zC7jiX+ot+susNfmbm0fx6ozm9g15rmXUjKZM1tyHOdy9cutuo49yGyt9ftZRttxoNJ3OPTA4n/bwe3+L3EP/lv8HJ/wx1bTDUTesv/43EKV8FtBd1Rrm327kNcW75wDH079hCa9ynJ+HROGefqugzJlKb6CZ29/ljrj+xu8+n9qKfQP2+pa2cFKycAuqtZravc+7VcEp3Z1i+GTgg67gO4JWw/Phx5WvD8o4cx+fknLseuB5g6dKleQ0be56xb2stXzvrDRzZ0jd6UnQshXd8Du67gnjPJuKtC/HftYqvP5niymWH4rvg6fUmTIqVd7tNB6NnvsXyfm6/YT5sgx0v/0EBtUypQvrbqTB+eukvPrqYlky/ndGziahL0tFWR3tjjBuWNYx+WNq6kHlnr+bkJe08sr5r10OqfSRlsjzPOHRBE/dcdlzVZewtVdudbqXKsBwxo6OtbkxQ3dFWR74DzBED3vxhuO+K0b18331t3o+fLcq93Xo4FqX/jD0c7Ml8QOtC3Dm3YyyhWjcPiviJ0bgho2cTET9ZmgrJlCinKd/3AZlM3SuBe7PKLwyzfR8L9IZTwh8BTjKztjAZ2UnAI+F9fWZ2bJjd+8Ks55oyzTURlrQM49JpOP/uIJg+7m9HO3eAnk3MvX8llx7dzCf+82nS4TTvQt6EKYGBFCQVTCFKe/l/dhZpCrZyGNryh2mpkshMy0zXbm+M8YMLXsOc+uhov53RuhAvWsO33380XzhxH+beP3YE27vzPL52eseu6YmVMvo609cOzzPam2rYv62e9qaaqgimq1Vm+cOWVzbh97zEllc28eft/TPy/qIuHuFf33fkmPPpX993JHXx/N4bmUvv9n6L+64IyqVyDHZhd5y722whBrv2/LgKlvbicOhpcPatcNGDwfdDTyPt5T/wIeWnJCPUZnY7wejyPDPbTJCt+0vAXWZ2CbAJWBEe/hBwKrARGAQuhiAJmpldBfw6PO4LmQRlwEcJMonXESQjm7KEZAD4Pl7XszTced7oJ6PLr4OaxpyfOs2vNzZ3DxGLeAW9CSs2eYfMYuEIdWoSI9T1Le0Muhqs89npqpXIjMokMBo/6szy6+DRf4T+TvyzV9MXaeHrj/6Oa05pz9mX13npihp91bVD9qR3aIS2/uc56OGVu0YHe5avord+CW0NtdP6u1vr4ixoruWq5YdTH48wmEizoLmW1rr83hv5fhovxznq+35ZjRTJXqRyj9aSqt7cFP1eM7G3Xol39wW7rkX+ilvo95rRQtDKVZJ+xzl3rnNuX+dczDnX4Zy70Tm33Tl3gnNucfh9R3isc85d7px7jXPuCOfcuqzn+Z5z7pDw6/tZ5eucc4eHj7kizBA+dQa78DLBNATf770MVz83eJOWrXUhnYOOjrY6OtrquOey4yb9ZkbJcKRg4Qj1ZKZ8t9V6/MF1ULdDAbVUh3g0knPUmXsvI/3e7/Pc6ffw0TVDbO4eoasvwbOdwzn7covGK2r0VdcO2ZP6ZDet43IJtN67kvpk97T/bs8zFs1t4PD9W+hoq+Pw/VtYNLch73Mq7cVznqMa5asw0dz/R6LlPfOnGI3+ztFgGoIZUHdfQKO/s7QVk6Log7xCTPCJWiLlSJ+9erRzaF3I9net4vond/LdC5eyb0tdQW/CZuO2AjJF0mFAPZkp3x686C1k3sAfYIo/ixKZKpPZL7qtLsah8+I5++1XeoZYdsMGHlnfxYdvfZK/OWExV6/tYvu7Vo3pyznndqhvn/a6TyVdOypDqZZ0RV0y5zkRdTOzlrOY5QF9kRaSK24bc44mV9xGX6Rlmmor06K+Pehbp6mvLUdRv7TnnUyPckpKVhFSKR+fKPHWhWNPiNaFmMHzA3VEz3iAA5rAeXES1sxVZ0aY11DgaIbvM9/byS8+uphX+n2uXtvFUy/tVDIcyU84bcpN8lP7V2MLaUg+Bv1boWmf6aiZSMFyTWW+5QPHsKh2ENIJhv0In31gM4+s76KjrY6bP/Am9suMhIzrtzsHR4OXzd1DvHZ+A184cR9ScRg4/wHisRjRSIQer4XB3uGip3rP5DTszJY04xM/6dpRPko5Ld+L1uQ8J7xo+U88TaTgV/3tvOWih8BPgRflV1ujHFRf6prJpHgezF8CH/xp8H4lGg+Caa96x/vSXoxIjvMubTHUM1eu6m2x0yCV8nluax9fe3w7yRW3jv1Ebfl1xO/5AK974Eza0tv5mwe38tvuON0DSVrT3fg9m0j2bsFPT2JkwPehcz2x77+Tfb9/DEevWcENyxo4eUl7RSTDkTKQGaG2yX12tr0mbNtbfj/VNRIp2vipzO2NMdr6n8duPBG75nDqbz6Zfz4uylEHNLO5e4jOvhGu+Z8du41opc5aTcu8ffnOBUdz1AGtnLyknflDL3DEw+9hwY1LabjtdKJD20j7aZoTXcx13ezctrmoxE0zOQ07s41VpSVSm01KOS2/x2uh78ybx5wTfWfeTI9X/qO8TTUeb2nait10Kvb1N2A3ncpbmrbSVKO3tVLe+iMt+GfdMua888+6hX7NrqhoGqGehM7+Eb7x6Ab++bgosf8K9z9sPxS6/xQkttkcLO+ee/9Kvnre/YzU1tLS/0eiN40mL0udtRoWLMGL5PE51GAXjMt+OPf+lVx78RoiTdP36XWpttGQaZBJ7DHJEeqhpgPx+wxeeQpv8YnTUDGRwo2fyvzZ49tpvXfFbn3lv5x+D912GAuaazhq0VyueHQTl77zbubXG52Djut/upP3HD3EVQ+s51/fdyRHtCaI3XLy2Iyzd55P9OR/hkc+Q+Td1/KaJ77D4HFX0jv0Olrqatg+kMD3fdIOnHN77TMzCdL+4/T9dtXj6rVd0zINu5q3saoWpZyWP5jwueKBfj6bdU5c/UAX157n09Yw7b++KA2JbdhdY9eh2l0X0HDRw1DfsecHS/kIB452vdfNTPmev6RqR6kb0r14P//KmD3UvZ9/hYZTvgpoikWlUkA9Ccm0z6VHN49mid3wIFy6Nkj8dOI/wVA3PH4NbF5HY9Snwd+O/fxfxnT40bvOI3nxGryWPKbRTrBWO+aSMI3BtLLCVpHMCHVkcgH1/JZ6Nr68Hx1//CX1b5uOiokUbvxU5vn1lrOvXNQa5dVtPn/aNsjchjiPrO/ikfVdHHVAM589vp3P/nULrU1BgPv9x1/kuuX7j3mTk+nPqWvbtS0PF9xD/Y4/Evf3Z8PWBF9bs4GVbzmIT/7g6V195ncuOJp5DXE8z6OtLkb3UHJXQNtY4/H9Uxt2fQBwQOtCvr98FS4+PW8eM+tUpTyVclp+PBrh2EVtHNpeS63n09bgceyitopYEmATrEM17eVbWQa74H/vhPPuBi8Cfhqeug2OuwIaF5S6dtMi6hJB/LDhwbHly75UohrJVFBAPQmxiMd+jd5oJ96xNEja9MhnRj9Ze/e18MR3YOsz2COfgTO+Dcd+BMzb9QbN0nlO5Zpgzd90Zj+caPrZPZcdpzdllSg1HHyf5Aj1oiZ4yl/MolfXBW3c9GGKlI/MVOZMX9WT8DggR1/5/PYk9fEIP3hyM1eccAgdbXX81cGtXPWWCLG7R7fPuvXMW/BaO4gMvpy7Px8Ksx73bIK+LfDgx4mcfSvXrBnh4uMOZjjp89UVr6dnKMlj67fg93XSFIvxwo4kvW0L+MqPN/CT9Z10tNXxkw+9jvocmZXdJT8FpnerIik/49vyTE7Ln1sX4cqjUtjqd0HPJhpbF3LlWbfg6mYooPb9IKAqZO2sF8v9/khZviuMweHvgdVZ2xmuuDkor1KeRXLnLrDqHJGfLRRQT8L8xhpG/MZgQ/ZjPwKtB0Lfq8GIRmYk474rYOUDsOOP0DgffvQROO2rcNuKXWutvWg8uJDs7cKRyX44firMNGY/VFbYKjPSB0AqUjephx3QCHe7xZydWBu05bmvmYbKiRRm/FTm+riHO+d2LKuvTJ+9ms/9cAtfee98vnDifCJuG49++DCi5hPpfAbO+NauDzkb7rkAzv8B3Pn+sdtq3XcFvP+HQT8OQR881B1OBX8/f3f6PSRronziP3/D5u4hTl7SzrUn1BK7+z3Qs4kjwn19P3DcIn6yvpPN3UP09vVTn2tkLd8PWqWqlHJavjewBcZNPbWffwU79SvQMs3Tpoud6huJwopVMLgNYvWQHIT6eUH5DNDSuCmSGoG7Lxzb7959IVz0UGnrNZ2icbjwPkgng8EK5yASq+qtwmYDBdST4HnGQKSF+hM+hw10warTx45k/OwLQVDd+xLc/zejZbFwTUS476l3xreCi8DeLhwlyH6orLBVZjjY1zA5yYC6NgqbahaDD7z0KwXUUnZ2m8pcN7av3GktvHnRCxzkbyK6KsxjcehpcPwn4cGP7953W+5p44z0Bf169rHhfQe1xjjhxt/s6i//7i1ziN195m6jz3/x/h/vespX+n32neGZR1LeSjot/80fDj44yj4fZkKOHDHccW5wDucz1ddPQXJo7Ll8xreD8mmmpXFTyE/l7ndn4P9YUsM9cNeFo233rJshXuaJC2SPNL9gErYPJOjbsQXrDQLj3UYyjvvbMSMY3HcFvO2To9MFM8dCcOEY7Nr7L/W84OLSekDwfZqTNCgrbJUJR6jT0cknuvAb92OAOtj8q6mulcjUy+4r69tp9Xu58i8biA68Cge9Fc6+FU76Qu5R6Ld9Mli7l8m6mtG6ENcwD65YF8w0+t1dQT9/0YNw/t3EYxH+6uA5ABx1QCuLWqM53xzWWmpXJvHrn9yJf/bqvPddLdUexaX+3TL9nHOjwTTsOh+cm4H/8wQ5YnYl0twb5wczR7Lr/qOPBOXTrJSZ2auOF8nZ7+JV8SBOang0mIbg+10Xji7Rk4qkEepJcH6aRU0OEvNzXwjaDwumBz5+zWhZ20Gj0wVhbMCdGAz2+S2jPfeUFbbKhAG1i0x+beaiFo/f9BzCWzY9ob0RpayNmX4ZMeYMbCRy53lYZlT6bVfCXRcE07xz9d1tBwX99ruvHQ0wDj0Nd/IXsYFtMNwLzR1w5DmjH6a2LsTOvJ5/OPFgNnQO8JHjX8Pz2/s4Isfo89NbhrjqgRf41/cdyYLmWvy2WtIXryGaHg76/lh9zhWDuUbCbv7Am2isjZJM+dPaP2sUrjBTOhW4mDXG+XDp3OeDm/4lXi4Sx3KcKy4Sz2/1rPMnqPv0fxigpXFTqKYZzrol6J93jdbeEpRXK3+C885X+6lkCqjz5PuOFr+XyGAXNO2XOxlG17NBQpt3Xwtdz0F/J74Xw+vvHD0mM2WwdSFs/X1wfJltEaCssNXDDffS7+qIRibftg6fC795cTF/1fkjGOmHmsZpqKFIcTKB3zVrnuPSo5s5Yp86IneeF+SweNfXg1wXRrBmDTdBIqMovPhfQb992ldh3qEwuAO7+d2jb/LOvhV++a2xowr3XEr8wgf4mxMWUx+P8LmHt3DDu1Yx9/6Vux6XXHEbV/+oi83dQ3ziP5/mhx99Cy91D9HW/wqt944e5865HRt3Hdh9v+0atu4c5sLvPT3tQa4SVE7elH4IMRPbCZmX+3yYgeRIPV4LtnzVmHOgZ/kqnNdCWz5PYJHgw7I3nDualf+3t89I3ePRCCcvaefSo5tHt+B7cqeWxhVipA/+/EtYeX/wIYl58NyPoX4u1LWWunbTIzJBQr1J7sYi5UUBdZ62DySwgQFqksPBaMUZ3x6dbpQdKGemEJ7xLbAIPtD5vvtpr00T6X4hOKa/c+zxj/0znPoVcA4XidPjtTCYmN7RB5kd0sM76aeOmgKu84e1wTfdIRg+vPKbYNqsSBnxfceWncPgfL5xQi3xu1cEfW/jfDjxC2P76DOvh9aOIGlkagRGeoM34fXzIFITBNKZ5Ea43RPl3Pn+4LEnfA5+8Q146tZgK0R8/nr+ECmLc+yiNrpcLY3n3wsW4aV+2DZcz1MvvQQEQelwyqd/xxYOenhspm/LsX50/EjYR45/DZ/4z6cLDnInM3qqUbjJm9IPIYpdY5yHZKSB+Fk377aWMxVpYLrf2g8mfK54aGDsHtgPTWIP7HgDnPzF0cRODfOD2zOwDnVufXmyXO0AACAASURBVJRvnlhH9K7Rre++edZqvHq9pZ4sZ4Yd8o5gFoYZuBQc8o6gvNSVmy6ROCy/bsxsJ5ZfF5RLxdLZn6dEKk3fSIR5zfuH0//2Hf1EbeszownJIDhBmvaF5DDRm99F0wU/5v/7WQ/nH3kQh7/nBmzr70eP71gaJAX5/inBm6rWhdjyVVzx0ABd/UlNsZOipAeDEeq6As702ij0Nx8CQ8CmJxRQS1nxfceftg+wvT/BEa0jxFedH/S9Q93Bmujx6yvvuRTec33wxi07CdPZtwE+tL8O+rbCTz8PJ30x95S83pfgRx8Nt3UBXvwvbPvzRG9bQfTQ07jybVdid52567kPXHEbN/x6NIdGR1sdEYPWeDhdtWNpsCY7M8Lmj13/OT5JZGtdrOAg1/cdf97eT/+OLbTGfbYkPPrn7MOBcxtzXl+UoHLypvRDiGLXGOehnzpam/bFu+jBYLqpF8GPxOmjjjlT9ltyi0cjtDfVso1WUi5GD0nam/ryb1/ODz4Y6900+kFYy8IZWUPtDW3Du+u8Mf1L9K7zpvTDjlkjWhPMgBv/f4xW8SyYxCA8+o9jsuvz6D/Ce28sdc2kCOUxx7gC1Mc99o/2wNB2aJgHeNDzEmzfGEzbzgTTEE6ZMqhthuXfpC4Kl/z1QRw0r4GU78Z2FMf97W5JQVrvXclnj29Xogspmj+8kz4KC6gBDp7XyAb/AJJ/+p+prZhIkXqGEqTTaV7XNEQ0PTTahz5+TbAmOlcw0tC+exKmO88Ptobreg7q5wRb8TTvnztRTib/xd0XwluvDKaR923dlfDMMusAw+eO3X0+5x8ZjJh1tNXxnQuODpKJJ7xguuo7PhdcP246Lfg+0DUmqB6fJHIwkd71c0ZHWx2xqLfX5GG9QyO09T/PEQ+/hwNWvYkjHn4Pbf3P0zs0kvPvO1GCyohHUUnKqjnRWeZDiGwFfwgRjedug1OYDb7ZhvCSQ5AcDrNmD+Mlh2i2ob0/uEhtdTH+4fTDOHZ+itc393Hs/BT/cPphtNXlOTaeTgbny4MfD86fBz8e3E4np7figEslgg+YL3siSFh4WfCBs5vCDztmjXRqgv9jFWf59iLBTNU73x+85jvfH9w2fVhZyRRQ56nV76XpnguDC0//1iBgvvcy+PmXg+nb2Rlbl18HP/wQ3HQqYFhigEXeVhpvOZnYtUcFHcYJ/xiMTjS053zjN78+GDHQFDsphhvpK3iEGuCYBfBr/7Xw0hNKmCFlJZVKc2D6zzTduoxo1zOjffDmddC7OXcwkisZTON8qGkK+uVvvDGYLTTQFSzrye7X333t2ISTvS/Bze+GfY4I1m72bcnZl7+uPc69lx/HTRe/iZ1DSf7xvmfw6+eSeufVuYP7rN0fspNEPv7Jt/P6A1pyBrn9wynOvO5xjvvyY5x53eNs2Nq3W6Ban+weXa8a/r7We1dSn+wml/G/+4eXvYWaqMe7r93z79mTzBrjvdW1Uk3pLhn17cGa6TyzwRciQjrYvue298K1S4Pvwz1B+TQbSCTYZ+iPtNx2CrXXvp6W205hn6E/MpDIMyj1k7vvtnLvZUH5NEvHGuCYD8LqFcHfbfUKOOaDQblMTgn/jyUTiQWznLLP7RU3aw11hdOU7zxZOpx+FYkFex9msmP2bAqmb2embrQdDNs2wIn/FIxm/PJb2NKLgvVI4zqMxAUPEIl4RHIkJ+gcDN5gaIqdFMOGe+ljHoUu7XptK6yOvI5Y6tEgid6+r5/aCooUwPcdzX4v8bvDad7jM3Rv/BmcdSvc9f6xa9R2vrJ7Mpi3fXLsFiaN82FwWzDKfd7dUNMAnc+OXdaTPVp91wVB/z/UnTPRzHNdCbakhmmti+GZ8ZP1nXT1Jbj97P2J5jGld3ySyNa6+JhdGCIevPvax/e6bjfqkjkD/qib+I1r9u/u6hvhwu/9qqj1wdWe6GxKd8nwPPz2w0hfvAZLJ3CROJHGdrxxCcmKyio+0fY9Fz04+fpOUm2im1jm/A1/d+zu86m96CdQv+/en8BPB+dq9rTZx6+ZkQ9+I8mBnH+3yEUPwbRPlq8yszHjdToBv/9hcH3xIsFrfeo2ePOHSl0zKYIC6nxlpl8NdQdTvr3o6JunzeuCKRuHngZv/QTc/zdjk5XVNkNiYOzz9Wxi685hrv7vXr519mq8O88bk+ny6oe6tAe0FM1L9NPnFhY8Qu0ZuHmHwnZIvvg/xBRQS4llRjkPio2MvhHbvC4IeFc+AH2vQm0rPPqFIMt3834Qa4CBTvivfxsbeLcuHDs9vGNpMA07+/6zbgnW9OXarQGC4+ragrXX4547ueI2bvufAd53zNxdS3c62up46qUentk6l6NzZXqNxvcYJI0PsF/uHsxr3a4XrckZ8Ht5rlWcivXBsyHR2VTtkuH7jg2dA3zo5vUTZgwvOqt4CYOZiJ97jXgk35HJaG0w0298Yqfo5LeInDQ/NcHfrYqnKU+X7PfSGZmdF6qVRWDxicHMhuxYQVO+K5qmfOcrM/3qt7cH2STjTUEym+wpGydfvXtm2PuuCN7MJQfHPl84Cv3I+i621b+G5MVreOXiX/Pc6ffwnH8AnzplCVctP5wFzTVKSCYFiyT76S9iDTXAkQvnsdnNo/OZtVNWL5FCZUY5/9STGjute/O6IEnR904ORpg3PAi3nAHffBMkB+DhTwRb7NQ2ByMDl/4cLrw3WMKTeZ4cOS246wKI1+Muegj+5qkgG3iu0epMUH/aV0lc/hRPvvNurnh0mOVv7GBBcw3fXvsC3177Al9+75F0tNVx9douepav2m1Kr183b1LTovNdt+vq5+GfvXrM73Pn3I415DeFeCrWB0/pGuMqt/3/Z+/O49uqzoSP/869kmzJmxzHTiCO2crSlNJCwta8bSmUAmVJaZqwhK0Lgaa005V2OtBhph3eodApnVIKgZYmIYEEKG9YStOUFtqhDG1SKEsghDVxFlt2LNuyZS33nvePK9mSLSeSvGjx8/189LF1JV0dW4/uueeec57TF+XHG7dy/TlzWLv0JK4/Zw4/3rg1LafKaD3+2eZd0YY749QIbUz80FPbyDxH3M72vbWVeajwJKyhPdgITFXujcCJYroTKzCkHAfPX17ew5+1PbKeeeSaSUmoJyZOWX/7lVJnAj8BTOBurfV/5r0zw4CmOdhn/RDDMCEagtqZztAoZThfBDvuDD965ta0jN/a5YGGw1Ffe8W5AhUfQJsejjNcvP6d4wi7FDqmqayoYEatgdb92C4vSlsYkTZiMTcYJq54v1MOtw9dWU9nfzy/YV627czTi0ednndfozO8LKVXxO0ycBmKcHRsw9bGNBxtHPeRw5tl/N+UJNvCZfXTq314x3C++v4GeEUdwQm7/wpaO/kDhCgQbVv85JwDafIq7MsewdjwL9DXBh/9NtrlQX31JTDccPJX4NglzpA60wPvW+TswPSAu9I5YTNc0PheJ6nY2kucnuZMPU9WDMwKNArV8B6oSmTy9bdgL16F8fQPnfuhdno9jXQOeDh6hslPP1mBTT9uM8pDSw7CVgYWNn+46r2YOoZ2edGf34iKD4Ay0W4vnX1R+iJxrj9nDnc89SbP7why5cpNrLvqZGbWeDDCHeh4lLhy00UtGAYrP3fC4HDs4SObbMvCCgXQ8QjbI7XEz3mYWreNv6YaT20TriyPbw1VHlZ97vjBLOHBqEH1tJk5jaBKzjEe3qM6JUZhWXEI7UnEkhuqZ4LpnIJlquNs2+byDx3Ctx8aWnP8poXHYKckrYvGLT50aANXfuRQTENh2Zq7/vRW1j3+Pa5p1C1eNZRMz9+CXryKHtc0sl0BOB6LY/UFMO0oluHBrGrE5d7/qaVR3Yh12WOYViSxXJLGMiswqrOcI27FncRgJ395aNjssz+dlF7icGUj3gvuRa0dmlKiL7iXcGUjvgl/9zLjqoRphznn0nbcOSableOafK/ojDYyZDIuBokJU7YNaqWUCfwMOB1oBf6mlHpEa70l750aBmFPA77uN1AvroOjPw1/utlZ9ip1iGByOGDrJvC3oPo64O7ThoYkPXkDKtQOF9yL58UHcb9/Ieqlh5h2+Mfh/muceUEpQ5nMlNcRaocFt2NXzeS633axYUsgt2Fetg3tW4bWt0z2ijS+NzG8bOhE5+bPHMMPf7uVQCiS1/JdYx6ONk77yKHAGf83NM0pzUZ1pBdgzD3UhoLotCOp3/sXtm19mcOPev84FVCIHNk20/vfpGnj0BQZveRBVGwA1l2CSl0K6/2fSR9St3gVvPQQVNTCfRemrbvLnpfhkoedE/tMww+1Rv3qrJSltu6FM/8TdJwoHnae9H0OOuM/ieJCRQc4OPw23LssvU547k6MD38DlxV1lvBKacDw9A9h6+MofwvuBSv4j8SyiTctPIZbNmzl+R1BuvsjNPW/gbHuYlRwO25/C+a5K/juM3G+evpRPHLN/BEXQG3Lwm7bgjuxxM97/C10nruCLzzWRyDUypov1NFcb2Z1LDXQHGy9i3riosG1d/WF96GYA1muGDuuc4xLiRV3clCkNFxZvApmHI2tzIx1XJ3XPdiYBqf3+dsPvci6q04e3G1VhcklJx/EZ3/1t8HX3r7kOKoqsruCWukyoLIOljw02KjFdDvbsxCPxSHwKhWJ+HL5W4gvXkO88b1ZNKptjIGutP+JsXgVVm0W86fBSSSYTAyW+l321GT3+jEwYOTa9WaFDPnMg+WqxOx6Z8RFHavxqPJtoJieUYa5l/FFhCmgnL//JwBvaK3f0lpHgfuBBWPdqS/agVp7sdPz8cBlzhDCTEM35n91qAHtaxh6bP0y57HgdqdH5NglzoHk2CVD+5n/1cxDmZKvW78Ms/ttls6tBXIc5tUfGGowJvd9/0VYocCIoWPfevBFrj7lsLyX7xrrcLTx2kfWRvnfpGbdLSmRHgB68eadlCyp5dCjAHjmyfVjLZUQ+esPDOWbAAhuRwXfHUo+lthGaM/IbcOPs4PbL4NDPgz3nu80dIev2rBoBWz83rBs3Jc4J9ErF1DZuYVgf4xW3UAkblPd+8bI4/cj1zh1RbhzqDGdLP+6S53HEvdTl0389kPOMbi53stMV8hZ6zbltQ2PXs7SubVcuXITlg2z6n001gxNE7JCgYyvSe6/vTeS/bG0P4AadnxUeRwfk3OMh5e1rIX2DDUcYSgeQ3tGrePilp1xvrmVMvS/P2qzbPXf0167bPXf6Y9mN3S0YiCAWnke/Ox4J1v1z45HrTyPioHsPlOrb2R8udZdjNW3/9ebobYRy8ypdZdihtqyem9i/ZkTqg2fXjcBKgYCqDWLYPUiZ9mj1YtQaxZl/X8TQ8y+QOY4yCKGSpZSTttg+OpAMvqvpJXtBSBgFrAj5X4rcOLwJymllgJLAVpaWoY/PIJKJqMwzKFkNJmGbjQe5Qz/fvIGZ/mV1Me89UO/J/eT/Amj7zP1dW4fTa6hL1/WiV3imROBKCuasfL2J9aEzCdxTMklsRnlfzM8624xyCpuEz3UEeXDHOOlM5d/FiGjlto9z/LKrm7ed2Dd2HYopqRcj7cjZPqOun3ZbRt+nE3drm3nZ3C7M7ro0oedZbDCXc5JztbHR9+X28cBHgNLKaKR8OjvnXr8Hu2xxP3UZRMbqjzctPAYopGujK9t8qlRj4nKioz6muZ6Z3j5AXVZJnEqoePjRBhT7FqZM6xjxYjames4w3A+o9THmuu9pF5/iI3S6I5bWc7FHGNyrdESixnZJBYba2KvQiYGK6GkZGM+5k60Evpfjpv4gNM2SM1Q/+QN8Om7C10yMQbl3EOd6VLPiKwuWuvlWut5Wut5jY1ZzN1JJqOwraFkNJmSUwReG1qsPTVjZvI1yd+T+0n+hNH3mfq6WP/g0lqQQ2IXV+ZEINr0ZEwWEwzHctt/ipJLYjPK/6YY5/JkFbeJBjUeb+bHc6EMBqYfzUeMl/jJxq1j35+YknI+3g6X6Tsa689u2/DjbOp2ZaSvY93+Kvy/LzrH8H2taZ04FtdUVaHRtPZYo793uGvfj6XcT102sc7r5pYNW519Z3hte78e9ZiozYqMrwlGDW5aeAwPbd6R/bG0hI6PE2FMsWtmTv6F6R69jjMNbv7MMWlrWt/8mWPweoY+L7dpZHytK9srqGNMrmWNJbHYWBN7FTIxWAklJRvzMXeildD/ctwYLqdtsPYSZ4RDsq1Qzn/zFFDODepWYHbK/WZg15j3Wj0TfcFqZ824RSudrN/DhwguuN1JTJacJ/X86syPXXAvPL/amUP3/Oqh/Txza+bhIMnXLbgdq+4Qlm92hvTmlNglma18WGZZs7qRuy6bN6LyvuOpN/NOHJNMQJO6z1z3Mx77yNoo/xt8RVgJZSM5ZMpdPS67G2g8humqmx2vbeKl1u5x2acQOcnwHdV1LbDw7vTvbVWTc+xN3Tb8ODu4fSW8uC79mPvCfc7xfbTj8eKVzr4W3E687mCCRh3haJzlm3sIVbWMfP55tzn79DaMyGirF69yHkvcDy5YwQ+eCgzOh715w2s8vyPI8s09xBenZ+nuPHcFyzf3jHpMNKsbR7zGWrwG29vAir+8zddOPzL7Y2m5HR8nU/XMzPFYPXPUOm6az8OM2kq+v+Bo1i49KbHqRyV+79Dn1VRdwR2XzE177R2XzKWpOrtlu+KVTU78DYvHeGVTVq83q0bGV3zxGswsMsdb1TMyvrdVPSOr97YqGzO/vnLi49GqHuW9s02oJgZZVaP8L7NcfaAUFTJ2xcRRWmdeiqPUKaVcwOvAacBO4G/AxVrrV0Z7zbx58/SmTZv2v/N4DN0fcHo1rGgiy7d2MvQZrsFM3hjm0JJZdtxJRKAM5zHTA4aLmGUTdvvxxbtRaAxtYds2tqvSyfJtx7CVZPku4izfBZ/0Mmrc/ulm+MMP+Lz/F3z9+LH3UrsGujjyz1/iZ3oRfz/kKn5xxfFj3qcomOKN2/1JfEeTma77XXX4rF5cdtQ5BivDGaZtepxltOx44ljshdhA+qoMhgtcFRALO8+344l9mE6vohVz7puVzmN2LJGF1g1WlLhRSZeupq7SQygWJxyzcSlNg9GHyx4A20YbLrQyUdrCViYWzu+mjmG4KlC+BmdudTyKNj0EjTr6ozZul4HHVPRFLCwNlW6D6T73iCzfytj3MTGZ5VtZUWzTQ9jtJxSx8zuWFn4VhILHLeQZuzlm+TYMlVXdF4/btIcixC0bl2nQVF2BK8ukYgCxgQiugfbB70O8sgl3ZfbraCezfBt2DNtwZ53l2yl71JkznXhvq3oGrhxGPMQHBjAHAkOvr2zEVTkJ61AD8fgAZijlvasbce17DeyCx27ex9wJFo8NOHOmk//LqkZc7sn5HAulkLGbo4LHbako2/EFWuu4UuoaYAPOslm/3FdjOicuN6r2wBxe0DDqI57EDYYaPMOrwkwD8hTQWJPnsGfDgAxXgZPJYtJU5fcW+9xnAfaRw5tl/N+UpPZX2akb8VaOw5BvIF5ZT6jhGJZ0P8V/vXYe/9gR5AOzs11cRYhxkviOKsANOLP5xyfGc+UGkn15FRWp1enQgTP1bMRglEo3ccxRQD1Qn3Lc9Q9fhyflb8+mH9EwTYy6mYP3PUBdvmv7lNPxcbKZLqhrzvjQaHVcNnWfy2VwoD//+HdXVkDl0GC+XFf/dblduPxZZuYe/lqXB/xD753rCamrsjKt7JN5QutyVY6p7GKIyz31/peFjF0xMcr6M9Ra/wb4TaHLIUQh2Hu28JrdTP04XovY23waLZ0/5uKKv/DlNT6ufm+Ulnovh75vHgdOm/jlSoQQQgghhCgmZd2gFmLKsmKwdxuv67M4aBzbub3Tj6PPfwT/FryT/vAKqv/uLFGydWMz35p+LSd86FQWHtc8chhpbAD+dreTfOl9n4KZspa1EEIIIYQofeWclEyIqavzTQw7zut2M+8Zz1HZhsm7x32HvQedRbR5Pm+/94s8f/BSZrjD/Nvea3n016v41O3P8MKO4NBrulth+Snwu3+B//kx3HUq/PUuJ+/AeOtudbI0R0Ljv++x6noX3nkGBnrGd7+9bbD9OejfO777FUIIIYQQ+yU91EKUo843ANjtns30cc5zoc1K2o5YMnjfA7S1fJCW53/IPaFbuK2zlUt/9jHOnnsoX2t5ixl/+i5E++C0G2D6EfDMf8Fvvgnv/BnO/i+omj62AkX7GXjlMcLP/Qr/nr+g0MRMHwNHX0T1/CtZ31rD+2bVcfiMAgxJjw3AG7+HZ2+D7c862ww3HP1pOPFqOPBYJ4FWrqwYvPM/8NflsPUJQDvJto44E05YCgd/GCKJhrtv2rj9OUIIIYQQIl3ZZvnOR7FmQBRFreAZEEeLW73rBZ5/aw8+9+QMRDFi/Rzw0s+oaftr2vZtupkb3V/hHbMFpUBpm8XW43whfh8KTVDVEaaSeNr1PT3ydw0q8Xvy0QpiNNKJiaZVT+ch6yO8xsF8XP2V84xncSuLkK4k7qnFX1uXQ+M1l49VJ7L820O/o51Gb6jdyQxdPQPee66zTMbOzbBto5P931MNXj+4vCP3l/oThn63LejrACsClXVwxFnQ9F5o3wKvb4CBoJOlWltw6vXwkW+O9Q+cEHK8FXkoeNyCxK7IS8FjV+JW5KHgcVsqpEGdQikVAN7N4qnTgY4JLk4xkr97pA6t9ZmTWZjhcojbpFL+HKXs46MU4zapmP6PxWAq/T8KHrewz9gtt89C/p7xU/DYHcMxd7KUW7xlo9j/5oLHbamQBnUelFKbtNbzCl2OySZ/d3ko5b9Hyi7k/5hO/h/Fo9w+C/l7xGSaip/PVPyby5UkJRNCCCGEEEIIIfIgDWohhBBCCCGEECIP0qDOz/JCF6BA5O8uD6X890jZhfwf08n/o3iU22chf4+YTFPx85mKf3NZkjnUQgghhBBCCCFEHqSHWgghhBBCCCGEyIM0qIUQQgghhBBCiDxIg1oIIYQQQgghhMiDNKiFEEIIIYQQQog8SINaCCGEEEIIIYTIgzSohRBCCCGEEEKIPEiDWgghhBBCCCGEyIM0qIUQQgghhBBCiDxIg1oIIYQQQgghhMiDNKiFEEIIIYQQQog8SINaCCGEEEIIIYTIgzSohRBCCCGEEEKIPEiDWgghhBBCCCGEyIM0qIUQQgghhBBCiDxIg1oIIYQQQgghhMiDNKhTnHnmmRqQm9xyuRWcxK3c8rgVnMSt3PK4FQWJXbnlcSs4iVu55XETWZIGdYqOjo5CF0GInEncilIkcStKlcSuKEUSt0JMHGlQCyGEEEIIIYQQeZAGtRBCCCGEEEIIkQdpUAshhBBCCCGEEHmQBrUQQgghhBBCCJEHaVALIYSYcnZ3h/n8r/7GrmC40EURQgghRAlzFboA5cC2NZ19UaJxC4/LpKHKg2GoQhdLiH2SuBVT2SMv7OLJ19oJP/AP1lx5UqGLI4TIkdRholRJ7JYfaVCPkW1rtrb1cuXKTbR2hWmu93LXZfM4ckaNfDlE0ZK4FVPdi63dAPzlzU7eDIQ4rLG6wCUSQmRL6jBRqiR2y5MM+R6jzr7o4JcCoLUrzJUrN9HZFy1wyYQYncStmMq01vz17b3MrvcC8Nru3gKXSAiRC6nDRKmS2C1P0qAeo2jcGvxSJLV2hYnGrQKVSIj9k7gVU9m7nf0EQhE+ckQjAO909hW4REKIXEgdJkqVxG55kgb1GHlcJs2JXo6k5novHpdZoBIJsX8St2Iqe73N6ZE+amYtfq+b7Z39BS6RECIXUoeJUiWxW56kQT1GDVUe7rps3uCXIzkXoqHKU+CSCTE6iVsxlbX1RgCYVuVhRm0l7+6VHmohSonUYaJUSeyWp5JKSqaU+hrwBUADLwGfBQ4A7gemAX8HLtVaR5VSFcBKYC7QCVygtX5nvMtkGIojZ9Tw8LL5kq1PlAyJWzGVtXUPYCjwe9001VawrS1U6CIJIXIgdZgoVRK75alkGtRKqVnAV4A5WuuwUmodcCHwSeDHWuv7lVJ3AJ8Hfp742aW1fo9S6kLgJuCCiSibYSgaayomYtdCTBiJWzFVtfUM4Pc5JzAzayv587YOBmIWlW4ZcidEqZA6TJQqid3yU2pDvl2AVynlAnzAbuBU4MHE4yuATyV+X5C4T+Lx05RScvlHCCGmuD09A9T73ADMqK0EYMdemUcthBBCiNyVTINaa70TuAXYjtOQ7gY2A0GtdTzxtFZgVuL3WcCOxGvjiec3DN+vUmqpUmqTUmpTIBDIujy2rQn0RtjZ1U+gN4Jt6zz/MiFyl2/cTjT5Xoh9KZa4besZoN7nzFeblpi31tYTKVh5RPErltgVQ6S+2b9SiFv5HEU5KKUh3/U4vc6HAEHgAeCsDE9NfhMz9UaP+JZqrZcDywHmzZuX1bdYFmUXhZZP3E40+V6I/SmWuN3TM8BBDVUA1HmdnuqOkDSoxeiKJXaFQ+qb7BR73MrnKMpFyfRQAx8H3tZaB7TWMeDXwIcAf2IIOEAzsCvxeyswGyDxeB2wdzwKIouyCzGSfC9EKRiIWfSE40xL9FBLg1qI0iP1TXmQz1GUi1JqUG8HTlJK+RJzoU8DtgB/BD6TeM7lwPrE748k7pN4/A9a63G5OieLsgsxknwvRClo6xkAoL7KaUj7PCYuQxGQBrUQJUPqm/Ign6MoFyXToNZaP4eTXOzvOEtmGTjDWL4NfF0p9QbOHOlfJF7yC6Ahsf3rwHfGqyyyKLsQI8n3QpSC5Fzp5BxqpRR+n5uOXukREaJUSH1THuRzFOWiZBrUAFrrf9VaH6W1PlprfanWOqK1fktrfYLW+j1a60Va60jiuQOJ++9JPP7WeJVDFmUXYiT5XohSEOyPorA5+5Wvc8hz38OI9VPndcuQbyFKiNQ35UE+R1EuSiYpWTGRRdmFGEm+F6IUBMMxTjJepbn9KWiHH0uGcAAAIABJREFUqG8Gdd7TCPRKg1qIUiH1TXmQz1GUC2lQ50kWZRdiJPleiGLXE46x0PwzcdNHvLIe/86nqfOdyY6unkIXTQiRA6lvyoN8jqIclNSQbyGEEGIsusMxTjFeoLfxOHqb5lHT8TwzPBH2hqKy/qkQQgghciYNaiGEEFNGrLud6aqHgdpDCE07BqUt3m+9gqU1wXCs0MUTQgghRImRId95sG1NdziCL9aFS8cwXBWoqkYw5PqEKGK2Df0BiEfB5cH2TqezPy7zlsSUUtX9OgCR6mYGag8GoCX+LjCbjlCEaZIMR4iSYFsWViiAsiJoswKzuhHDlOzQJWfYuQm+8j+ftm1NZ19Uzr/KiDSoc2Tbmnc7Q9SHtlGx/nIIbgd/C/rC+1BNc8r+ICBKlG1D+xa4/6LBmLUXr+G634fZsCUwmFnzyBk1clAXZa2+7w0AItWzsV1eopXTmRl9F/g/dIaiMKOw5RNC7J9tWdhtW3Cvu3iwTosvXgMz5kijupRkODfhwvugjM+nbVuzta2XK1duorUrLOdfZaI8o3UCdfZFCe3dgz/ZmAYIbkfdf5FzhU2IYtQfGKqwAILbca27mKVzawFo7Qpz5cpNdPbJWryivE0Pv0WvqibuqQMgUnUgDf3Oqopd/RL/QpQCKxTAlWxMw2CdZoXkPKykZDg3oczPpzv7ooONaZDzr3IhDeocReMWfo899OVPCm53hqsIUYzi0Ywx2+Qbuhra2hUmGrcmuWBCTK4Do++y02wG5cR+pGoWdX1vobDZKyc0QpQEZUUy1mnKku9wSRnl3KScz6ejcWuwMZ0k51+lTxrUOfK4TIJRwxmWksrf4sz9EKIYuTwZY7a9fyircXO9F49LhsqJ8tZkt9Plahq8H6lqxrQiNKuANKiFKBHarMhYp2lTzsNKyijnJuV8Pu1xmTTXe9O2yflX6ZM51DlqqPIQmjaT4IIVQ8O+k3OofY1pz5WkA6Jo+BqdeUkp85Tii9ew/PfO2rvJOTz1XjeB3ojErChLOh6lUe/lZXcDtYlt0aqZABzllga1EKXCrG7EWrwGM2UOtbV4DWZ14/5fLIpHhnMTLrzP2V6mGqo8rPzcCbzb2Y/PY9IftTiowUeDJMQsadKgzpFhKA5qqKbbN4fIFb8bNcu3JB0QRcUwnCQfX/j9YCZNwzudH5wf53vnOo3neq+bbYGQxKwoW32drVQrTbhi+uC2WKXz+6HuLvbIHGohSoMy2O4+mNBZv8bvsQlGDardMzlIycDLkpLh3GQqZPmOxG2uX/9y2rmWKG3SoM6DYSjqqyqBA0Z9zmhJBx5eNp/GmopJKqkQKQwDqodSGBtAY83QEKNAb0RiVpS1/sDbVAORypQGdcU0tDJoMTvZIj3UQpSEzr4ol/7yb2lzUZvrd0h9VYqGnZuUO2kflKfyvgRUQJJ0QJQaiVlR7iIdTvIb29swtNEwiVVMY5bMoRaiZEh9JUqVxG55kgb1BJGkA6LUSMyKcmd3JbLJ+hrStscqp3MAAWcdaiFE0ZP6SpQqid3yJA3qCdJQ5eGuy+YNfmmScyQk6YAoVhKzotypnlY6dC3eivRhdTHvdBqtdrr6o2itR3m1EKJYSH0lSpXEbnmSOdQTxDAUR86o4eFl8yVjsigJErOi3JmhPbTpemqGnbfEKqczLf4X4vEY4ZiFzyNVoxDFTOorUaokdsuTnDVMIMNQkmBAlBSJWVHO3OEA7dpPjTt9e6xyOiY2M9lLZyiKb5pUjUIUO6mvRKmS2C0/MuRbCCHElFAZ6aSTOiqHTVWLep2s37NUB12ydJYQQgghciANaiGEEOVPa3yxvfQYdahhI+uSa1HPUh10SqZvIYQQQuRAGtRCCCHKX7gLF3F6Tf+Ih1Ib1F3SoBZCCCFEDqRBLYQQovz1BQAIm3UjHtKmh6i7jlmqQ9aiFkIIIUROpEEthBCi/IXaAAi7RzaoAeLe6TQb0qAWQgghRG6kQS2EEKL8hdoBiI3SoI5VNjDbkKRkQgghhMhNSTWolVJ+pdSDSqnXlFKvKqVOVkpNU0ptVEptS/ysTzxXKaX+Wyn1hlLqRaXUcRNRJtvWBHoj7OzqJ9Abwbb1RLyNEBNGYlhMCYkGdbxi5BxqgJi3kQPoZG8oMpmlEkLkSeouUaokdstPSTWogZ8Av9VaHwV8AHgV+A7wpNb6cODJxH2As4DDE7elwM/HuzC2ZTEQ3E1877vs2bWd6x7+B1vbege/GPKFEcUkUzzuL4aFKBd2qJ2oNjE8VRkfj1U2UEGUWG9gkksmhMiVbWve7QyxZ9d27OAO9uzazrudIam7SpFtO1Nygjucn7Zd6BJNKNvWbG3r5fzbn2H+TX/k/NufkfOuMjDpDWql1PxstmV4Ti3wEeAXAFrrqNY6CCwAViSetgL4VOL3BcBK7fhfwK+UOmAc/gSHbUP7Fnwrz+CAe45n7sZF3Djfxa0bX6OjLyJfGFFUbFvzTmcfL+/sprUrzMs7u9kd7B81hmXpIFFuYt276aCOmgqV+fHKBgDcfbsms1hCiDwEwxHqQ9t4/xOfZvaKE3j/E5+mPrSNYFhGmJSUxLk0d38cbj3a+dm+pawb1R19Ea5cuYnWrjAArV1hrly5iY4+id1SVoge6p9muW24Q4EAcI9S6nml1N1KqSpghtZ6N0DiZ1Pi+bOAHSmvb01sS6OUWqqU2qSU2hQI5NAz0R/AWHsxBLc794PbaXj0cpbOraU/YhEMRzN+YaShIsZDrnEbDEdp6xng+vUvc8Hy/+X69S9TawdHjeFo3Jrgv0BMRXkfb8eB3dtOh66j2p358WSD2hfePYmlEqWikLErRqqKdeFff3la/eVffzlVsa7CFqzIFH3c9gfg/ovSPkfuv8jZXqYiMWuwbZDU2hUmEivfiwhTgWuy3kgpdTLwIaBRKfX1lIdqATOLXbiA44Ava62fU0r9hKHh3RnfMsO2Ed3DWuvlwHKAefPmZd99HI8OHQCSgts5oNpga0cfR86ozviFybehYtuazr4o0biFx2XSUOXBMDL3tIjyl2vchqMWv3rmLX5yzoE0+RTt/ZrIQJiaUWLY7dr3V1LiUeQj7+PteLx37x4C2r/fBrU/2o5la0yJZ5GikLErRjLtzOdgph0rTIGKVNHH7Sjn0sTLt/PJUIoz5jSydG7t4PnY8s09SJVT2iazh9oDVOM0jGtSbj3AZ7J4fSvQqrV+LnH/QZwGdltyKHfiZ3vK82envL4ZGJ+xfLYNdhz8Lenb/S1UV1Xx309uI2JpPjGnKe3h5novnv00VDK/nQwfF2PjMjQ3zncxd+MiZq84gbkbF1FXaWSMYX9NNQ1VnlH3JfEoSpHZ30GHrqNmlAa15a4lrtzMVJ10h+WkXIhipk1PxvpLm6PXXaIIuTJ/jrjK93OscMFtp1WmnY/ddlolFS5pUZeySWtQa62f1lr/G3CS1vrfUm7/pbXelsXr9wA7lFJHJjadBmwBHgEuT2y7HFif+P0R4LJEtu+TgO7k0PAx6w/Ahn+B824bOhD4W7AXr+bGPwYIhCK809HHdWfPobneCziN6bsum7fPhspoOvtk+LgYm3rdQ8Oj6cPj3BuvQ1+wOj2GL1hDZd2MffY2SzyKkmPbuCOddFBL9WiHYKUIuRuYpTrplEzfQhQ1rUxYcHta/cWC29Gq1HLtTnG+RrjwvvTP8cL7nO1lqtbqxv3AkvTzsQeWUGsFC1swMSaTNuQ7RYVSajlwcOr7a61PzeK1XwZWK6U8wFvAZ3EuCqxTSn0e2A4sSjz3N8AngTeA/sRzx0c8Clsfh742OONG8NZDuIuAruF/3nqbmxYewy0btnLbxcfy8LL5Yx4WG41nnm8h81xFtkw7NnJY1dbHsT/5I8wv/N6JaZcHw9cIxr5PSCQeRckJd2Foi4D28/5ReqgBBjwNHDDQSUcoyuEzJq94QojcGPEwPHlD2jkYT96A8el7Cl00kQvDgKY5kHIeQhbnIaUs4/mYTFcoeYVoUD8A3AHcDeR0Bq61fgGYl+Gh0zI8VwNfyqeA+5UcotK6CdZe4mzzt+C75Ldcf84cbtmwlUAogsdl0lhTMea387hMmuu9aY2YfIePi6nJNj2Y/pb0g7i/BRuFWZ1by0HiUZScPmcm0L6SkgHEKxs4sPdl/iY91EIUNdvwOGvLJ8/BwKnTjH18wUVxMgzI8TyklI16PmZ6skooJYpTIS4BxbXWP9da/1VrvTl5K0A58pdhiEp88Rq++ZudXLVqM4FQJO/h3Zk0VHm467J54zJ8XExNZnUj8cVrRsSsWZ37sCqJR1FyQm0AdCk/Ffs6Y/E10EQXHd19k1MuIURejFHqNCOPOk2IyTSe52OieBSih/pRpdQy4GFgsBtAa723AGXJT4YhKoZ3Oj84P873zh3/rMeGoThyRs24DB8XU5NhmjBjDrHPbkRZUbTpwaxudLbnui+JR1FqQs4SLGF37b6fV9WAqTQDXTuBwye+XEKIvLhcLuJN7yV6xe8w7Bi24caobsTlKsRprRDZG8/zMVE8CnHkSSYQ+1bKNo2zznTpGDZExQAaaybuy2AYalyGj4upyzBNjLqZ47MviUdRShI91BGXf59PiyeWzrK7dkx4kYQQY+NyucB/QKGLIUTOxvN8TBSHSW9Qa60Pmez3nCiyFq8oZRK/YsroayeGC9y+fT4t2aA2Qzsno1RCiDGQOkyUKond8jPpDWql1GWZtmutV052WcYiuRZvcvmg5DzSI2fUyJdCFD2JXzGlhALsVX6qPPuO7ViiQV3ZNz4rLAohJobUYaJUSeyWp0IkJTs+5fZh4AbgvAKUY0xkLV5RyiR+xZQSaqNT11GznwTAtstLn6qiKrJncsolhMiL1GGiVEnslqdCDPn+cup9pVQdsGqyyzFWshavKGUSv2JKCbWzx973kllJPa5p1A8EsG0tvQVCFCmpw0SpktgtT8WQDrGfEkynOpa1eGXuhCiE1LhTSvGJOU38bkv74OOylrQoVzrURpv9PqqzWNmtzz2dWZEOuvqjNFRL4j0hipHHZfKJOU0snDsbv9dNMBzjoc07pA4TRU9itzwVYg71ozhZvQFM4L3Auskux1gl1+IdPgdif2vxytwJUQiZ4u6OS+YC8Lst7bKWtChftgX9nXSQXQ91uLKJg/teo7VnQBrUQhSpeq+br5x2BFffuzmtTqv3ZvElF6KAJHbLUyF6qG9J+T0OvKu1bi1AOfKS2svXUO3hkWvmE45m39M82tyJh5fNl2WIxITJFHdX37uZdVedzL+eq/G4TOq9bhk5IcpP/16UtgjoOpqyOF+xfE3U7A3TGdgFB9ZNfPmEEDnrCscGGyQwVKfJuZQodhK75akQc6ifVkrNwElKBrBtssuQr0y9fKs+dzwHV/ajrCj0e8DX6KxRPQqZOyEKIRq3aKx285NzDqTJp2jv1/zgqQBaa2bV+2TkhChffc60hg5dx6FZNKjNmiYA+ve8AR9470SWTAiRp9HqNDmXEsVOYrc8FWLI92LgZuApQAE/VUp9S2v94GSXJVfDe/kaq93Uh7ahVl8Owe3gb4EL74OmOaM2qscy91qIfPk8Bvd8sgr/+kUQ3M5sfwv3LFiB9jhxKiMnRNkKtQEQ0H5qspjR4K6dAUC88+2JLJUQYgw8pspYp8VMuQAsipvEbnkqxLJZ/wIcr7W+XGt9GXACcH0BypGz4b3L153SiH99ojENzs/7L4L+wKj7SM69bq73AsjcVTEp/Hb3iFj1r78cv90NyMgJUcZCzvG4gzpqszjM2lVOD7UZfGcCCyWEGIs6O5ixTquzg4UtmBD7IbFbngoxh9rQWren3O+kMA37nA3vXW7yqaEvRFJwO8RHX0vOMBRHzqjh4WXzZa6qmDTKimaMVWU5sSojJ0TZSvRQt2v/ftehBtCmhw7q8YW27//JQoiCMHUsY51m6lhhCiREliR2y1MhGrK/VUptUEpdoZS6AngceKIA5cjZ8N7lYNRwhnmn8rcQU25sW2fYg8MwFI01Fcyq99FYUyGNaTHxXJ6MsYrL6bLLNHLizksl66QoA6E2oqqCMJVUZRnOAdcM6iI7J7ZcQoi8Ga6KjHWa4ZIpSqK4SeyWp0lvUGutvwUsB44BPgAs11pfO9nlyEdq7/Iz3/4Yzc2z0RfeN/TF8LfQee4Krlm/nVf39BCP24UtsBBJvsYRsaovvM9JoocT24c3VrPmCyfy4NUnc/05c/jJ719nWyC0z4tDQhS9UDu9pp8ajyLba5fd7iZmxHdPbLmE2A/b1gR6I+zs6ifQG5FjcQrtm461eE1anWYtXoP2TS9swYTYD4nd8lSIId9orR9SSm1Mvr9SaprWem8hypKrZO/yIO8cYp/dSCDYw+6QzQ9+G+D5HT28sjvEmi+cSHO9D8NQacttyTBvMdlsFO+aBxE669f4PTbBqEG1OZODUINX1brCMS6++7m0Yd9bdvdKYjJR2kJtBFV2CcmSwhUzmBH+E729PdTU1E5c2YQYhay8sG+d/XGu+32Ypac/MJgpefnve/jB+XEaa2SqkiheErvlqRBZvq8C/h0IAzZOpm8NHDrZZRkXhkG7Xcv8n29O29zaFaa9N4LX46KhyiMVoyiojr4Il/7yb8PmSO/g18s+RFNNJSCJyUSZCrXRQV1W86eT4lVNEISOHa9TM2fexJVNiFHIygv7Fo1bbNgSYMOW9CSw3ztX6itR3CR2y1Mheqi/CbxPa91RgPceV8leZ0tr7rnieP77yW08v8PJ0tdc76WzL8oBdZVSMYqCG4g5jeVjZ/u5+pTD8HvdBMMxrJQhhJKYTJSlUBvtuiWnHmpXjbN0VnDXNpAGtSgAucC5b0opPjGniYVzZw/WZw9t3oFS0kkhiptSKuO5lsRuaStEUrI3gf4CvO+4Sg7HOv/2Z/jID5/i+vUvc+2ZR3LsbD/N9V5uWngMD23egcdlSsUoCs5MnHx884wj+f5jW7hg+f/y/ce2sDcUHZyXJ0u6ibITj0K4i11WdktmJVXVO0tnRdrfmKCCCbFvyQucqeQC5xCPqbjm1MPT6rNrTj0cj6zlK4qcqeCmhceknWvdtPAYyiF0lVIzlVL3K6XeVEptUUr9Ril1hFIqrJR6IbHtDqWUoZQ6WCn1cpb7PUkp9VxiH68qpW5IbD9FKfWhLF6f1fPGohA91P8M/EUp9RwQSW7UWn+lAGXJW6Ze5289+CL3Lz2RcNSmP2px3dlzqPe66Qoz4mrUGXMaaTJ6INjpZFr2NYJREquHiRLk9Zhcf84c3mjv40eLPkAwHOOOp97kqns3D46UmNAl3WzbWZ89HpV4F5Onz1mhcUcstyHfPl8NvdqL2vvWBBVMiH1LXuAcPlVMLnA6opbmN//YyaOfPQKvESdsu7jzbzu5bH5pzh4UU4dhGLyxJ8jGK4/CRYw4btZtCXLEjJpCF21MlNPF/jCwQmt9YWLbB4EZwJta6w8qpVzAH4BPAX/PYfcrgMVa638opUzgyMT2U4AQ8Jf9vD7b5+WtEA3qO3H+mS/hzKEuSZl6nRurKwj2x7n63s1pFeDhjdVpFeMZcxr52ce9uO453VmLzt8CF94HTXOkkSEmRG2Fm13BAa5f//JgbN608Bhu2bA1baTEiKR748G2oX0L3H+RxLuYXIk1qHfbdTTn0A5RhqLVOJCakDSoRWFM6AXOMlBpwjc+aOFecxYEt1Ppb+Ebi1bTKx34osjVe02WHBrGtfJ8CG7H7W9hyeI14C354P0YENNa35HcoLV+QSl1cMr9uFLqL8B7yK1B3QTsTuzDArYk9ns1YCmlLgG+DPiB6wAP0AksAbwZnvd54DGt9YMASqmQ1rpaKXUAsBaoxWkjf1Fr/edsCliIs9m41vrrWut7tNYrkrdsXqiUMpVSzyulHkvcPyQxBGCbUmqtUsqT2F6RuP9G4vGDx/uPyDQc6yunHT7YmIahudJd4Vjaclu3LWjBte7ioYXdg9udxkZ/YPjbCDEuusKxEbH57Yde5CunHT7xQwj7A0ONaZB4F5Mn5PRQB3RuWb4B2tzNzIy8M/5lEiJLyQucs+p9g6OIhKPaCuJ+YElaveJ+YAnVVrCwBRNiP3Rfx4g2gGvdxei+kk8tdTSweV9PUEr5gNNwOlVz8WNgq1LqYaXUVUqpSq31O8AdwI+11h9MNHz/BzhJa30scD9w7SjPG83FwAat9QdxlnZ+IdsCFqKH+o9KqaXAo6QP+c5m2ax/Al7FuXIAcBPOP+h+pdQdOFccfp742aW1fo9S6sLE8y4Yx79hcDjWrRtfY+ncWg6sNqivtWisdqf1XCfnSqf1/AU7h75IScHtznBYISZANO7E5k/OOXBwmYYfPBXgkOlVEz+EMB6VeBeFkeihDmg/tTkM+QboqWxmWvRp7FAnRnXDBBROCJEvw45CdROccSN46yHcBc/cimHHCl00IfZJWZGMsaussj4nOkwp9QLOqk7rtdZP5NLZqbX+d6XUauATOI3ei3CGcQ/XDKxN9DR7gLdzLOffgF8qpdzA/9NaZ92gLkQP9cUk5lHjXMnYDGza34uUUs3A2cDdifsKOBV4MPGUFThj8gEWJO6TePw0Nc7p8wxDcWRTFT8/3cvcjYs4YOMXqex4iQcvOIDffuFIjp3ttPkzJhFxeYYWdE/ytzjbhZgAPo/BPZ+sYu7GRcxecQJzNy7ink9WUe8zMfrbIbjDaXzYEzALQ+JdFEqih7qDOvw5zmSIVjcDsPedf4x3qYQQY+X2wVk3gyvxxXZVOPfd3n2/TohCK9/YfQWYO8pjbyZ6h4/VWt+Qz8611m9qrX+O08P9AaVUpivdPwVu01q/H7gKqBxld3ESbeBE+9CTeI8/AR8BdgKrlFKXZVu+Se+h1lofMnxbcqj2ftwKXAskZ+03AEGtdTxxvxWYlfh9FrAj8X5xpVR34vnjOp7CCHfA2oudK02nfg8euQYzuJ2j/C3cfe4KvvtMBV89/aiRPYDeBrjsEacB0xeAF+6Dj33XSdQkxATw292o9ZenDTHy/+8t6FO+A2uXDM1tvmA1VCUSho1X4jBfozNnevgcaol3MdFCbURctcRw5dygNvzNsAt6d7zE9KNPnZjyCSHyYhoKon3w+DeG6pUFtzvbhShiZRy7fwBuVEpdqbW+C0ApdTzgG+uOlVJnA7/RWmvgcMACgkAvQ6OWAepwGsMAl6dsH/68d3Aa/+twOmHdifc5CNiptb5LKVUFHAeszKaMBcsIpBynKqXuJtH43cdzzwHatdapY/MzRZ7O4rHh+16qlNqklNoUCOQ4pzM5lHX+V+GRa9IaKw2PXs5PF7Rw5Iya9HlPtg2B12DlefDLM2DDd+GU70DjUZKgSWQt17hVVoZh1x+8CLU2fQ4aa5fAzk1w98edRGIpPda2rQn0RtjZ1U+gNzK43NZ+GYaTgOwLv4evvuz8lIRkU9KYjrf5CLXR5/IDUJ9jg7q2bho92ou1Z8sEFEyUmkmPXbFPKhaG9cvS66/1y5ztYpDEbfEp19hNNHbPB05PLJv1CnADsGsfLztSKdWacls0yvMuxZlD/QKwCliSSE72KHB+YjmtDyfe7wGl1J9J70Qd/ry7gI8qpf4KnAj0JZ53CvCCUup5YCHwk2z//knvoVZKnYgz7Pt8YBrwJeBb+3nZfOA8pdQncbrva3F6rP1KKVeil7qZoQ+tFZgNtCZStNcBGedoa62XA8sB5s2bl2ULISE5lNVbn3GOqLIiI1+TKUHT2iVOI6N6Rk5vL6aunOM2GaupcVrVmHluczKe778IrvwjaAsdjzJgm1z3WCsbtgQGM9iPuGA0GsOQ+BZjO97mo7eNbqMOtwHeHGu7Rp9im26mqev1iSmbKCnZxq5tazr7opKZe4JpO47KUH9pO56xR2WqmvRjrtivco5drfUuYHGGh47O8Nx3SPQMZ7HfC0fZ/jpwzLDN67N83kkpv/9z4nkrGJoynJNJ6yJSSv2HUmobcCNOdrdjgUAiy3fXvl6rtf5nrXWz1vpg4ELgD1rrJcAfgc8knnY5Q//ERxjq6v9M4vnjdiCxLYtY9x6s6AD6skecjRnmiCpt0R0e1qiWBE2iEHyNcMnDsOQBuOJx52fNAZnnNocTX8fqJnTPLrj746hbj8a38gxunO/i2Nm1gxnsO/skbkURC+1hL37qKyDXLBouA1rNZur73pyYsomyY9uarW29nH/7M8y/6Y+cf/szbG3rzX40j8ie6c5cf5k5Zh8UhWfbzhTIiczlUkwkdsvSZI65XAq04WThvldr3ckow7Bz8G3g60qpN3DmSP8isf0XQENi+9eB74zxfQbZloXdtgX3Padj/vcxqJXnoaua0BesHvqC+FvgvNtw/e46fLFh1wokQZMoEB0fcObs/OpsePwbaCuKtXjNiLjlmVud+x/99ogh4Q2PXs51pzhzn5MZ7IUoSlpDqJ02nXtCsqQOdzPVVjeEZKik2L/OvihXrtw0YulMufA4/rRZAYtXptdfi1c620XpsG1netndH4dbj8443azcaOWC85enx+75y53tU5xS6meJYdmpt88WulzZmMxPbyZOuvOLgFuVUn8EvClDtrOitX4KeCrx+1vACRmeMwCMNg5/TKxQAPew9ePU6oVEL9+AJzUF/h/+HVo34TrzP9N3IAmaRAHovgBq2FQDde+n2b3oN7Sd/gDvn+lFaQv3xuugdRP4W7DqD8XMMJqiyed09WXMYC9EsYj0QnyAXaoOf54pUUK+ZogCgVehWo7RYt+icStt2UyQC48TRcXD8PTN6UsPPX0z6qz/3P+LRfHINA3y/ovKehqktqKw8br02N14HXrhPYUuWsFprb9U6DLka9Ia1InJ408ATyilKoFzcDK/7VRKPam1vniyypI328alo/Cpnw+uG0frpsSQ7QEnwVhqA8TfQkS7qLT10Bxr1DVAAAAgAElEQVSq1ARN8ajTM52STVnmX4mJYMcjGRvH0yss/s+qN7nz0rn8evN2ls79Hk0n/Svt/Rp/r+Kw4fOu/S209+vBOdQTvoa12Dfbdk5IMhxLprzEGtTbY/n3UMdrmiEIAztfpvKQj4xj4UQ58rhMmuu9aY3qor/wWKrHENuGrY87t1Rn3FiY8oj8TMFpkJZyY4baYe0lQxv9Lc72whVLjFFBjppa6wGt9YNa64U46c83JB9TSl0++isLKDEsRf3qbGfI7IbvOktlNc8Dfwum24O1+N60IRzdn1rJ1x5rHTncK5mgyT/b+ZnSmJb5V2IiWCrznB2VWAfxjqfe5LIPHco/PbaLD9/5Bv/02C6M6ib0hfelxbR9wRpmN7fw8LL52SckExNjCg6Vy0mPk6PyzWg9/jyv+9TU+unS1fS3vjSOBRPlqqHKw12XzaO53llPtugvPJbwMSRuVmSs0+Iy5Lu0TMFpkL2mn/ji9Gmi8cWr6TX9hS2YGJOCD9jXWveQnlHtn8gzw9qEyjQs5ZFr4OwfYfuaMJ+4Fvra4OwfoacdxvaQwVcfa+X5HT1879zshnuNNv/q4WXzaayRSkLkL2TW4V+8CmPdpYNTDezFq+gz62iu9/L8jiAr/vI2q79wIi5D4XGZ1HvdBCPvwXfF73DpGIarAqOqkaZS6L2YCqbgULmc9O4GYKeextF5Hj5nVSte180c3v7qOBZMlCvDUBw5o4aHl80vjVFmJXwM6TXq8C++F2PdJSl12r30GnVMK3ThRPam4DRIy9aEtZuas38Ebh/E+glrN5Z0npW0gjeoMyjOmmeUYSm64XCM0B6nMd26CVYvQvlb6Dj9AZ7f0UNzvRe3K7sGiMy/EhOlKt6N8dJDcPEDYJhgWxjPr8Z3wjKuP2cOfq+bYDhGlcdEo4jGLXZ1x/nB41v43Zb2oWWyqlThFq8X6abgULmc9OwEYI+elveQ75k+eNmezQe7/+IkOcs1VbiYcgxDlc4F8BI+hjh12oMj6rSqE5bhzCYUJWE/0yDLUZ3dTcWfvg8fvMhpUMcj1Pzp+3jO/BESu+NPKaWAPwP/obV+IrFtMfA5rfWZ4/U+xdigLs5LNJnW8fW3oNpedoZ/n3fbYCIygttprnXmUt38mWNwZXl1uiTnX4mSYBrA4R+HNYuGrgKfdxsuA65atRmAT8xporney1WrNtPaFaa53stNC48h0Bvl+R1BGS1RbEY5JpXzULmc9Owi5qpmgAoaKvPbRaULdrlmU2H1QXerM01HiHJRwseQ0eo0s3zbYeUrOQ1yijAN4MSrnFGuErsjROLWyR290R/FbfsAl2Hsnl7j+UaFy3w23/1prbVS6mrggURCbBP4D2DcGtNQoDnU+1GcXQDJYSmZlhhKDv+e/9XBx+qqq7n+nDn88LdbCUez62EuuflXomQY2ho6eMNgzBraic3mei/XnT1nsDENzuiIbz/0IlefctjgfRktUUQyHZPKfKhcTnp20+duAGC6N//ddFUmGtHtW8ahUEIUkRI+huyvThOiWEnsji4St05+vS30yAXLnz35ozc/dfAFy589+fW20CORuHXyWPartX4ZeBRnueV/BVZqrd8cjzInFUUPtVJqodb6ocTdZwpamNGkDkuJ9kPby0M90uB8IZreB1/6G9pdiduOclBFjKYad9Y9zCU3/0qUDNu2MKqb0pdpeOZWtG3z3D+fSsRyktAkG9PHzq7lulMaafIpGv2a+688gV8+8w5KhrwWjyk4VC4nPTsJms5syml59lADRKubYQCnQX3EGeNTNiGKQQkfQ0ar02zbLsqeIrEPVhxCe8CKgemG6plgFkXzZEJI7I6uozf6oy/eu3l6asfOF+/dPH3t0pN/NKve+6Ex7v7fgL/jLIY5b4z7GqFYIvbHwEMAWutrClyW0SWHpYTaMi6RRV8bmJWo1Qsxg9s5yt/C7ResQXtNAr2RrBrJJTX/SpQOtw9OuwHWLxsaYrTgdnaHFXt1BAV4PS4+MaeJz80/mLmVu3E/MDSU7qgFK7j2jCPwmNKgLipTbKhcTnp2EVBHU18B7jGcpdTXVLEz0EDj7peRsUKi3NgoOrWfqLbwaJMGSiRPhsubsU7DNYarZ2LyWXGngyolYSqLV8GMo8u2Ua1HiV0tsUvctg/IlEsqbtsHjHXfWus+pdRaIKS1jox1f8MVy3GzpM7Sbe90rAvWOF+C5nmw5AG45Nfgmw4PXJY2jMNcezHRnvasl8KybU2gN8LOrn4CvRFZMkuMC8OKDB28wfm5fhnTKjQ/+8PrNHtCTLfa+PE5s5hu9uN+YEnac/3rLyccbCdqSTyKEmDFoC/ATnsa08d4jnJgFbxuN2PtfmV8yiZEkSjlpToNO5qxTjPs4k+oJlKE9sDTP3R6a6943Pn59A+d7WVKaTtj7Cpd/MvVTTSXYexOTntNaq734jKM3eP0FnbiNu6KpUFd/EfvFJ39cdoqDyV2xQb45C3w+DfgtnnQ35kxY2awNzRiKazd3eERDebRKrd43JZGthgTZccyxmaFsrhxvov6NWfhv/M4qlaewSGeIFQ3jXiu32OjtcSeKAG9ewDNO/H6vBOSJR1YBVt1C57gG05DXYgyMdpSnZ19xd8oHa1OU3Z8Ut5fOj/G0YlXOaM+f3W28/PEqwpdoollRTJn17eK/3s30abXeL7x80vmdqTmkvr5JXM7ptd4vlHgou3XpDWolVIvKaVezHB7CSipMYvRuEU4polEo7AupUe6L5BxgfrdofSLIa1dYVq7wiOuBo9Wue3qDpfkFWRRPLThzhibGCYNj16ePqpi3SXw0W+PeG5tdRVej2ScFyWgZxcArw/UjykhGcABVbDVbsa0Y7D3rXEonBDFoaSX6jRco9ZpE62Ue/aLjtYZE3RRxhfv48qTMXYt5S5MgYpIhct89ogZ1eetXXrys09/65R31i49+dkjZlSfN5Ys35NlMnuozwHOzXA7BzhiEssxZl6PiduliFtW+lWmZ251Mn+nZMyML17DxnfSK6fmei/BcGzE1eDRKrf23khJXkEWxaPHNQ29eFVabNqL70WjMl4pteoPS3tu96dWcu0Tu2jrkSvxogT0Og3qt2LTxtxDXWFCp+dA507nG2MsmBDFQynFGXMaeejSw/jzVe/hoUsP44w5jSWRfNJWrhHnW5x3m7N9gpVyz37x0Zl7a0tr4GpOeo06ggtWpMVucMEKeoy6whasSFS4zGdn1Xs/dFBD1SGz6r0fGs/GtNb6Bq31LeO1v1STNuNfa/2uUupTwHuAl7TWGybrvcebRhOJxqmtMNLXcGzdBM/dib7icWwb+m2T13sruXy+l8debktb2/eWDVudl6RcDR5tHerhB+mSuYIsikZfDB55q5qLLv8NLh1HG27saBhXx6sZ1yHdFtTEzvo1hze42dlrETLrae99VdaiFqUh0UO9W499DjVAzDcT+oGObWPfmRBFotKE206rHExAOdvfwm2LVtM7SQORbFvT2RfNa1WTmA3mc3emZ0p+7k7iZ/6IiS5+SffsFxuzIvNa6Gb5nmPENPT6DsO95DEqlEVEm+y1/XjL9xrClDCZQ75vB74GNADfV0pdP1nvPd4GojYD3e0YG/5l5BXSj3wT9cR3sMJBLlv7Fv+07h/s7YvxyDXz+dO1H+P7C47mlg1beX5HEHAazMlltTKtQ33npXN5aPOOtPdPfY0Q2VAKDpvp52PLt3HID7fwSlsY15pPw9M3jYhhvXgV0xoPYFe8hovW7uS0O1/jS2te4OpTDpOTBlEaenZhGRV0U8UM39h3V1ddRSe10kMtykq13T0iAaX7gSVU290T/t5jHTYddtVhf/TatLm39kevpd818b18yc6PVHJelifDdLKzp55HL7h9UobuF4rbgMaBt6hafQ6u246lavU5NA68hatYslqJvExmTvqPAB/QWltKKR/wZ+D7k/j+4yZua/weG7Y+7iyVlXqFVCnY+jietpe47vQHWLjqTa66dzMPL5tPs99LXyROIORka082mG3bSTpW73XTUO1hzZUnYiqF12NSW+Hma6cfyZbdvYM93HddNo+GKlnARWTPZRj8aWsb91xxPKahmG0kEugFtzvrqSdjuGYm6nffY+/x/85Vq7YOvr61K4zf66a53lsSwwHFFNeziz73NEAxs2rsuzuwGt7YcyB1gW1Fs9akEGNl2tGMw23NSciUPdqw6WxHQFVZ3RjJ7NCJ8y/j6R9SddaPgHG4irYPyc6PZPnlvGwMYmF48ob08+gnb4DP/KrABZs4dXY3nofTVwSqefgyKq74HRMdu2LiTOa5QVRrbQForftVCZ+Vuw2F7UoMU2ndBGsvcR7wt8ClD8MF98Izt9Jca3LnpXPxe92DvXpHzqjh4WXzicYtLFvzg8e38Lst7XxiThNfOe0Irr53c9oB2j/Dk/aaXIdFCQHgr3RxyckHUWsFqVBxlGEODbNKxrC/xanUtj7OIaf/X+68dC53PPUmz+8I0lzvpT9qcdPCY5ClqEXR69lFlzENjwH14zBy8MAqeNueyXEdL499Z0IUibhyY2YYbhtX7qIfNm3qmNOhkaqvzdk+wQxDyXnZONGmBxVqHzqPBmeknOkprfV0c2DqzBnqJyN2xcSZzAb1UUqpFxO/K+CwlPtorY+ZxLKMibfCIGI00nv+SmqSV5mO/P/s3Xl4W9WZ+PHvuVeSLS+xHMdLiGPWEBqWQpOytqUtUEihBEhjIBAShnUYOqVDocu0HabTdkopv2FahhKWtlkIkBRooOzQgXbYSlIo0JAAISEJgdhxbMe7pHvP748jWZItJfIiyZLfz/P4kXUlXV1J5y7vWd5zOpzyQ1O75imC2TdRVjKB/1jx94QAeXptOdXlRTR39HH2bS/0n1DmzpzaH0zD4NpaGbMqRqIzGKau5/1Y977pp6Pnr0K1bwFvCYS6wV8FT1wPgQY2tYX4jz9s5Ma5R7DkxU187YvT6OwL85sXNvHjs/NmVxXj1e7tfKwbqCuF0bjG3acU/qYn4+19DnrboViSx4j81+2tpGfOEgKrIzM9RJIjaW8lmb7iSJUzJt1u067tx559E3TvNAsi112u7c94ZQCYoFquy0auw66gbP4qrLhrEbeigU67ggm53rgMcVJUZDlZqMgSmZPNgPqTmOmxtg5Yvi+wPYvbMWKhMPgsKKusQy96DJRCdbfA8rP7T0rMuQ1Picu9jVNo6tb86LnmhAB5YO1swO+VJBciY0rDrXj/9NNYtyrtovrazRzq0TJ77nI4+ip0/aeYDjx1yYH88pWPuf60T3Ddqr/R3NmXslvbSJLLCDGqXAc6PuID+0hqRzhlVtTkEvi9rjN3WjbClE+NzoqFyKEKfxEfVxyckByp266izp/5QLGq1Meyf/g0nbs+JuBzaQtalE2sG0K3aQ1uKPEcNvcuCjk7dCHyh3djBTsTfkfrq7/BH94NjEJGyTGoTVVQftFj+JUL2gFl06MtOlQFNbneODFs2RwC/1/Abq31B/F/mNyp/5XF7RixItulsuNd1KPXona9j2paZ7qrxM+jt/oqitxupi45mplPz+Ou00qpLvMOyugd1dYTkiQXImNspeGYK2IJXMK98MCliWX2/gvR9Z9CLT0T9YsjKVl+BtcdFabSb3Hr/KN46KoTmF5bPihQljk5xZjS8TG4Idb3TRqV8dMAPhvafZPNHUlMJgqEhWZycHNCcqTJwc1YWQhKLTT7OR9w+OPnMHXJ0Rz++Dns53yQ9nt7dHDwOeyBS81ykTc8Ogi/uzjxd/zdxQX9O/psTXFvCyw5A35xFCw5g+LeFopsuWbKFKWUVkrdHHf/m0qpG0bzPbIZUO+ntX5j4EKt9Rpgvyxux4hNCLWgVi6AI883E9B7S1LMo6egfha0baHqkYX88OQ6wq4mHHYHZfR+YO1Wbr9wZkKGb0lyIUaLrR1TVqPlNEWZVZ07Ek5s1soFTHR2MsXaRbXehdXdBK6b8DKZk1OMKe2mE9T74UnUjWJ+l3BpLS5KAmpROLqbUf/7E9NzadGjcOpPzP3u5uy8933nJ5xv1H3np//eTvJxqDgyDjWvjMPfsSK8y8QQ8WV/5QImhHfldsPGinDfcbRtfZFdmzbRtvVFwn3HjcJa+4BzlFKTRmFdSWWzy/ee+m6MUse8zHMdB+WGoKwGJn/SJCHDggtWmSmIAE64BkqrwfLA3F/Dn34Gry1nWpWXf350Hd87fQb1lSX9SS1c18XRYCtYecVxaK2l26wYXa4D+38OjvuamY7C9ppx/xseNZU+0TJbXgdHXQivLTeva9uCatsCv/9HM73WK4txP/8dqJmBZZveEzInpxhTIhcpW3U1nx3FgLqmzMv2zmrqJaAWhcJ1Tc+laGVroMEc5wdUmu51Hd3NEA6Cxwcl1WCl0VYTTp5hnHCaFbGWbc5hR54fyw79+r3Zm25puJ9bJMr175gLbjjxesx14KVfmuXjXbjvOJrefpiVCyZFjkn70bjsYWo+cSaeopdGsmbgDsz0zf86OhubKJsB9atKqcu01nfGL1RKXQKszeJ2DJvraoK7myi2i2H2TdD6Aay+KnYimrfETAHw+ysTxlJz3NXQ3crHPRZXzCynjp04HUV4y2uoKvWxYUfHoOkXptf6JZgWo8dXCp++FFbMi5XNxmUw8UCYdnLiBdW8peY1ry0393tazWMPXw2n/gTr/vn0LnwKX0UdlqVGnFxGiFEVuUjfrqtGtYV6n1J4z51MnUydJQqE1g7qlcWJUxa9shj95Z+ll2HZdaFpHURbmgMNcN69UDNjr8Gltn2oJImZ0s7u7CuHE6+HaEtf9JzmK0/n1SMzgs8tBiiakPx3LCrUlGSkuB5bapaPd51NN/cH02C+n5ULJrHosZsJTD1+hGv/H+ANpdTPRrqZyWRzz78GuFgp9ZxS6ubI3/PApcDXs7gdw9bSFUQ5fWaqhu6dsWC6fpY5IRVXxIJp6B9LTftW9OyfUm+1MvPpefh+eQTe35wCTeto7+mT7rIi88K9sDJx3kNWLkAfc1ksmI6WY6cPPnedqTU+81Z45wmTsOysX0HNJ6CsBh3u6y+jA4cvyHAFkVPtW+n2VNBD8agG1FNKYZOuQ7W8B1rGuokscF3o3AFtW83tUFqO06CxE3NrPPldOOYKszwd3c2xoBLMbZrdttusCtrmLDHBBPRnGG+z0sygH+yMBWHR9165wCzPtBF8bpFI9XUk/R1VX0duNyyTwn1JrscuMsvHOzc8OWnPFTc8eaSr1lrvBpYC/zzSdSWTtYp2rfUO4Hil1BeAwyKLH9Va/zGd1yulpmK+iDrABe7QWv+3UmoicD9mHPZmoFFr3RqZ5/q/gS9jEp8t0lr/dUSfwXXwejxgV0FRuQkwtAvKNoH0Wb9K3oXJW4IK9eBZdeGgA3DJoqfS6y4r3YvESLjh5GXTdWLB9Bd/kNhSfe5yeOdJOLwxcfmc23A9frTrQOcOrHCQQ8p8PHL18XQHXRmuIHKrbQst1iQqi6B4FM9w+5TCi7oOO9wNXc1m2I8QoyDpLAnotFpBRzLDgtLhxNwakZ5IatFj6W34CLptdwddrn6si++dsoqaEmVmQ3msmVvnu1Sm01Dnhs0+GN+6/sIt2ek2O9Lu6iIm5bVJAXd/Ho+fOV2W5yMCDfsN7LmC5flolN7hFuCvwG9GaX39sh6Raa3/V2v9y8hfWsF0RBi4Vmv9CeBY4J+UUjOAbwPPaq2nAc9G7gPMBqZF/i4HfjWiDXddJnVvxHr8etM6vezsWLbkaKt0T2ustjUq0GDm+LXspDuQR6eR3Tvaveiuk+GWw8xt07rh11ZnuNZbjEGWN3nZjC4/4ZrBF1b3Xwgzzhq8fPVVKO0wqXtjf5lUd59MZcd7TKkoprq8SIJpkTttW9iuRzchGUBdKWyJTp216/3RXbkYt1LNkqC79t4KOuIZFvZU0ZoOjy/5ecWz995JPo9Nc2eIucs28tnF7zF32UaaO0PpDxXyFMNJNyS2rp90g1meaSP43GIAy05xbVLAQ8YsT4rPLIOJKKu5lsZlO+N7rtC4bCdlNdeOxuq11ruAlcAlo7G+eHnz62mtPwI+ivzfoZR6G5gCzAE+H3naEuA54FuR5Uu11hp4WSkVUEpNjqxn6Lqbse6fb2pDVy1Mni35hVtMF9kBrXl4S2D3dnN/QK1LWHn53RXH0NX6MT7CyediTNW96NJnoKx2aJ9jD2N/XFR/TbfXY+GxFD1BmVe4IFgeUxbjx/zPuQ1l2Wa8Urg3+YWV7Um6vJig2R8GlMm+RU+x256Yk/Iic2ELtIb2bbzvTKd2lIfgeS3oKq41/aNaNkLDsaP7BmJcSjVLwvOXH4S9l1bQVK996KoTqC7f+1zS2vKikiSE0pYnvXHMJdXm+mHg9URJ9V5fWlXqY+k/HM0HLd2U+Gy6gw77VpWkP1RI69j5DGJD7C5+Mr3Xj8QIPrcYwFtqrkEGjqH2FvB4YmUNjhXOvNUsH+88RS9R84kzWfTYzbjhyViejyiruXaECckGuhm4ehTXB+RRQB1PKbUfcBTwClAbDZK11h8ppaL98KYAW+Neti2ybHgBdbSLj78yMcCItkq3bYFta+DNlXDBAyYQsTzm8Uf/xTx3wA7UcfZSfvz0R3x71kfUrTZB+tRAA07jCtAzANsEwOEQXPA7U2O3ezs8++/mvYbTvShFcK4veYYNnf6E5Gg3ffUIfvbEBpo7+yKJ0gbPQZwNEiiNAqcXnr0h1j1Ou6Z1ItgFxRPArkla4aMtj0kcU1ZjWrH9lRDqRnmKkgbaTa27OX/l21kvL9GWmsHJ/XJTZpNsoAzZyIauZgj3sj40+i3UAKp0Ek6HhS0t1GKUpJolIay82EmOyfGtoCOdYaG3aBL+E6+PTeETaEA3LqO3aBJp7T6WZbqgX/rMsI5tfWGX769+K+GYnTanL3mXbycL41AtC3fSIbDocZQbQlteKKvDkmP6kGnMcZWFfwDtmCGUlm2W53rjMkU78O4zMH9VLMv3a/fAxP1zvWVjg6fopVFIQJZAa10W9/8OSO8QNxR5t/crpcqAB4BrIgPMUz41ybJB/aCUUpcrpdYopdY0N+8hoUS0i48TSuyq8cItpuUv0GDGoR5xHtwz10zW/tvTzXO++lvzHF+pOWhc+gycfjMfhgPMm1FCYPXChADXXjkfp7M51pr829nwP0ebbuYoOO1nJmHUcLoXpRj744YHJ0e77ndvcOXnD8xporQRd2krUGmX2/4X2NDZZLpxP/Nvpnb/kX+GW2fC0jnQuslkqY/rZqPPvQfH9qMvWg1nLTZTWfz2dHj0WhO4TD898T0CDTR165yUlzE9F/ZoD9nIY0Mut0MVObZty0CXb4C6Mg8f6mq0BNTjTqbKbnSWhHj1lX66vZXo8+5NPCYPaAVN9dp0u017g22o53+WOA/18z/DG2xL/wNYlukpF5hqbtMMKkd8zM5hl2/XcXCb3sb67WzUL47E+u1s3Ka3cZ2xN1Vkxo+5IxXqhpcXmxlyXMfcvrzYLC9U3hI4fK7J8n3rLHN7+FyzXOStvAqolVJeTDB9j9b6wcjiHUqpyZHHJwNNkeXbgKlxL68Htg9cp9b6Dq31LK31rOrqPXTXiXbx8RbHAmgwQYqvFC58CM6+Y3AXpPsvhOa3TUD8u4vNmOUnvg33zKPUClJTopIGuMoJJm9NXn0V9LTAqT8eXveiFGN/wsqbtKY74Pf2/5+LeYXHdKCUQ2mX2yi7yHSjSjVe+vdXghM0FT6LHjV/nmI8d38R9YujYPnZJhts/SxTPu+/gPApP0q42Gv5yhJ+9Jw5YWe7vIzpubAlI2y/IZfboeoPqKupL9vLc4dhn1LY5NYS3rlx9FcuxrRMld1UsySUF/nYbO/Lm7MfZOvCv/Dm7AfZbO+LG9dWMNIZFhRu0izfanDbw6gb8THbdZJ3+U53/PcIOJ3NeJ7/z4SKCM/z/2kaQsaYjB9zR8r2wmHnJAaXh51jlheqUE/yDPWhnj2/ToxpedPlO5K1+27gba31/4t76GFgIfDTyO3quOVXK6XuA44B2oc9fhr6uzbpti2oBy9J7Gb0+HVw8r9Dim6w+Ctj/z9wCXzlF/DIP9PUbU5aU1PMxZgyk6S3xHQnH073ohRjf7q9lUnnEm7rCfX/n4t5hcd0oJRPnF6ItkRUH5K8XJVUmSRloW5ww6h75g7K/sqpPzGVRG1baOlx2HbKKj5RU8zGXSF+8MTHvLbVdBrJdnkZ03NhS0bY7Gk3o3w+1JMyElBPKYXNupYTdr1senmogu2UKEZbimEflqWYVl3GyiuOI+S4eG2LmrIiWntCLPj1qwOOaVsTxkdblmJ6bTkPXXXCsIZEWdpNmuXbWvT4qH/8gUZ8zHZDKRKqhUZxK5Prr4gYMAY2GxURBccJwqoBU0itushUVBQqyfJdkPKphfoEYAHwRaXU65G/L2MC6VOUUu8Cp0TuAzwGvA+8B9wJXDXSDXBRaMsT6zr729PNbWeTCax95ckz9/nKY/fbtkDFFJzGFdyxdjc/eq550FyM4cYVqNJJhFSKzMyh7uFnk4wf83TNW+a2ZgYV/qJBNd03ffUIbn9uY07nFR5plzYRoV3Y8Gikx8T65OWqdRPosOnSDXuuHAo00NQNc5dt5IKVH7LbrqS5M1b5ku3yMqbnwpaMsNnTtoVuqwxvUQnlGfh69ymDD3QdnlAHdO8a/TcQhWkPwz5cV/NucyeNi1/ixJueo3HxS7zb3Jl2ZbJlKarLi5hSWTLkGRa06yY9zmud+eEoIz5m5zA7dMqKCC0V/UMWnboz3lAyzeej8ZjZfBzImxZqrfX/kTpHwUlJnq+Bfxqt94+O5Z1WYmM1Lo1Nyh7NzvfHH8KXf548c1847qQYaEB7ivlQ13HOzG4Cfi/rXZfqOb9n31nQZGAAACAASURBVAovIeUl6KtkW1MXv3hmCz87aykVv497rzm3QfnkkWWTjI55il8ECTXd0Szft84/KqeJwKIn3YHJpsZEoJRPlB1LOvbCLZCqDM+92yR60TppkrJoEr5Q4z1MLN6HL81o46l1TSx5cRMrLj0G21I5KS8jbanJKMkImz1tW/mYzLROA1T7YRuRY+eu96G0KjNvJArLHmbqaNGBpMOaVl5xXOZ73UQr+wYe5+3Mn1/jW+bDjosn0jKf9jHbUzz4PNa4NCtjqBXJKyKkhXoYolNIDZ53OHfblGm2D+YtjbXMBxrM/SzsdyJzCrjEjq6WriC3PL2e28+cDGV1sOgxk6BMu/D0D0zW7a5mWPPbxO7gryyGWYvMSgINcPYdqN0fMsXbzo/X9vDkumbqK/3cdsGn+PlLH3LW9CIOnBikKAjfOH4ivbbGe+Ef8FoWHkujvCXgn5iRDMHRmu4EOZ65YEwHSvlE2XDW7Was9LY1Zpqs0282wwd6Wk0w3dlkAulTfwq7Pxw8zda8JeiyWt6a/SA/eOhjmjtfYfGFM/np2YdR7rbj0btQ/V0Zk/w+Gc50nbT8jgUjzIQr0qd3bWSjU5OxgNpW0O2vhTAmoJ766cy8kSgsexj2EdTJW6JtRcYrkzVW0ukUdRY6L7quZmtrF527Pibgc2kJWvSF6ti3qiy987t2oThgZlVRypy7bK9ZnmHKLkoaBCoJiIZMeYuTTpulvFmYTzxX3DC89eDgLN/HXpHrLRMjIAF1moJhh28cPxEV6jFjpZWCZXNi0wkdd5XZMU68bnCNackk+Npr0LHd1EA9cT12ZxO3LHyKd75wMJWlPla8tInLDumj6pHzoKyGg066Af4QO8m1fGUJzqRDqCkdf1kAx2yglE+UBV5/LIgG00shMh66v5X6yX+F0/4TnvgOnPT92PND3bi+MnYS4AfPbO4fK/2LZzfwq1P8sTmp4+Y1TwgY9zD/+bgILJP0ChGjzAlD62bedWYwNUMBNQAlNbi7FZZk+hbpStUS7PHh08nHEluWlVZl8kimlbTD3YnTKfa0wrM3YJ9z92h86j1q7+mjsvNd9n88NmVo25wltJfMoLI0jWDKCZvz1YA5tJn9s4xvu2t7B/dUbFxqlmf83QtMqAfefGBwcHnMZbnessyxvHB4JBFbfKxgFXAitnFA9v00+Tw2+1Z4oK/dXBxrNzb39Au3mIO565jg+R+egn9+zWT+fvNB6NllEi/0tMIT15vXtG3Bq0PYlqJpdy+n7GdT9Uhk+qwTrhmUvbLqkYVUsqdZwsSocF2Tib1tq7ktlKmN3DD86ecQjszR2bUTPP6ELKX88YdmnDXaBNUAkw6GiqlQ0UCHKuOc21/hm6dO56ipAQAunzkhFkxD6gzWkulaZFr7FpQbZpOuy1gLNUBtuZftugpXAmqRruiwj7hcKdFhHyMZSzziaSUtK3lOmCxUcpaEWgdNGRpYvZCSUGt6K3BD0LUjcVnXjuwkJQt2wvM3JZ4/n7/JLBdDo13Y+iK0vGuuuVreNfd1AXefd/pMbDB/FVy9xty++WB25lAXGSMt1GmqKvURDBdB2AfBbjNOZ/rp8JlvmFbnlQNa+l5ZDMf+Ixw1H15bAQd/yQTe29aYFQYaCCkPv3j2HebOnMrMCk/sxOKvTNo9zKMzf6IY1wq5FVUpOO6fTJfv+BrR1+81FyEnXGMy1Ye6zdil38weVJ5LP/ctqsu8fOuBN/j+GTO4YtlaU8mULKlNOJiY8CAbma4z3KVcjHEtJsDd5Nbx5QwG1FPLYbNby6SmdyngToliNO1h2MfA/CXRVmaADTs6BnX5nl5b3t8CnWpayfhM4HveLm/SLt/ZaCnz6uRZur3pXud4IxXCD1wa2/a5d5nlmeZGknxuGJCJ+tSfZP69C4z2FKNO+RE8dHnsdzz7DrSnKGXSpLynbDj0LGjf0t8DkEPPMstF3pKrzTRZlqLbV4kurYnU4HrgxG9Bb2ssmIbY9EJHnm9OUu1bYdZCk5jsSz8yz4l0bVrXVsTcmVO5/bmNlJWWxmqvI4mfEgQazPjUCNfVNHf08WFrN80dfenXSGdZvmwnUNitqFrHgmmIzHt4Ecz+KZx0Q2we0kevNa3zZTWx50XKs2fVhSw99wC+f8YMDpzk59krDqGqxAMXrDLzU0dF5jVPkOlM13vIoivGiV1mbuiPrMlMyuA19UEVsFnXoXZtytybiMITHfYRmGpu4yr7kmXqThUst3TFKiFHZVrJ0klmHPLVa8xt6aQhfaxw2GV7Ww8ftHSxva2HcDjNY66yTKPEuctNK++5y819leZlqevEgmmITEt6aVayQ+toIq14gQazXAyNG44F02BuH7q8sKeQUpYJoh+9NnbdFepOv+yLMUl+vSGYUOQz3Yle/pXZ2VdeaGqXkrW8ldfF5ozu+BgevRbtLcG55i2chY/iltZwcHkvDZXFvLa1jQ96SwjNu8ccpF+4xdQSJ+keBqPQzStL8mU7+xXyfMEp5+x0Bg0v4KHLTYt1/PMivSbaOjp5cO0W9nO2cODqs1C/+KQ5GZx0gwmqI+P9W5mQ+F576PI4Kgq5MkSkp+U9epSf4tIJSXPijZapZbCVOopCbabyU4gMSCdYHum0kmHLZ4YB3TMXbp1lbsN9Znk6rw+7rN/RkTDl1/odHWkF1a7tg89dF6vMffK78LnrzPJ0OCnO107mz9fK44PG5Ynns8blCY0eIk1OimsTp4B7ZLoheOepxC7f7zyVleEKInMkoB6C1p6QCUCOvTJ2ME/RmkxxwNS2hrrNc9q2oO6/EPuj17GXnM7Ezo2UP30dB6utnDqjms27emguOZA3Zz/I1pNuY71bz/ozHmLnpWvpW/QUTf4DaO4K9Scg2VvN9ViQL9vZr5DnC05Ro46yU1cIxT8vUs6bujWXz5yAZ+WAcdOrryJ49q9pnf84330hjBo4n2KK+c9HrUt2IVeGiPS0bOQDXcfU8sx2FPRY0FMcnTpLWqlFZqQTLI94LudwMJZdGSI9lxakfdxs6uzjyuVrE87xVy5fS1Pn3seCWk4wNm1Q9L1XXWSWpyOXc/lqF7RjknYuetTcaicrGcYLznick9n2wWGRpGS3zjK3h50j02blOQmohyAYdtjRrWDCFGjdFGtNPvPWxJrKM2+Fp/8NTvkh+KvMc+pnmfE11YeY21cWw5HnY98/n1vnNHBkfQUhF77ym3f47OL3OO3OdXznyR1s6A1w0uK3Ofo/41t43ZF388qCUemOlk2ZbkXNJcszuNfDnNvMNCPJTmZFFbH/z7wVXr+X9rOW8qPnmqkpUUmD1x27e9nYXcI1pxyS/IJuD10eR6yQK0NEWtyW93jHyWxCsihrgqlw0pKYTGRIOsFy/LSSL3zrCzx01QkJY6z3RrnJKyJVmi1lISf5tUjY2XtgqaOJXQe8t043KI2O/x54TstGpmQnCKsWwj3zTOv6PfPM/Sy0jhcc25fi2qSAz91OKGllUkG3yo8DMuBjCHwem+/9sYnbz6hGPX+jCTQevtpkRz79Zph4oNkx/vhDk3zs1B/D2l+bF3/xB+a58YmeJu4Pp/4Er9LUTvDTtLs3YfqMKz9/IN964I1BLbwrrzgu6TQb6XbzypZoDftY385+hTxfsNOXdHoUvvqbWDmOL5ueIrh6rQm4Q93oU3/MPX/t47Wt22nqrmZqkilg2oIW+9T6qZtQnP15wqOVIQMTyhVCZYjYu3AQ1b6VTfpTNGQhoJ44sQZ2wa6tb1N1eObfT4w/8cHynqbEGtG0kinmU043mPHaVtJzvMfe+znTtXzYSd7btXykdYWgHTOkLm5qR7wlZnmmuckrAyRnx9ApNwzFFYm/Y3GFWV6o3HCK8lPAn3kcKIBIIXuqSn18/eTpaMtrEpO9udKMfTjrdgjsB73tsOwsE0wHGkz6/yPOgy98LxawQCzRU7DbjBvqagbXZVJZUUKNdFWpL2ntr9Z6ZN289mK0EomNuDtaLmSyFTWXlA2lA+ZBLq01ZfCVxYnTf7yy2NS0BztgyRnQ8h5q6Zlc8EkTqdyxdndsvD/0J9krm1iXm2AaMt+lXIxtrZtR2mWzW8f+FZl/uxk1PrbribRtXZ/5NxPjVrJkZaPJ9iTvuWR70mtrqSkr4vYLZyac42+/cCY1ZXsP8O2yasKNKxLeO9y4ArsszUpQreHvv4eKBnOurmgw97Mw3VLYLk7aIypsS97/IdMa3liV+Du+saqwp81KNQRPktrlNfn1hsCyFLUTilBOB5x7jwmg4ydmP3c5fOFf4bVlMG8JPH6dCbwvejh5bVSww9zefwFc/DiW1hxS5uORq4+nO+iilErZwju91r/XmuvhiCYS29NUHUP5vtKpYRdZYHvhxOtMZu/4abPefthM7/byr0xm+tJqOOXfYe1SOOqChIRkJbbL7648jpauIDe/3sI1C5/CS4iw8qK9lezrH/0LviGJVoaI8ac/w3cddSWZf7vJpbBVTaGuZUPm30wUhjE4rZ8b7MZO0nPJnfsb7NK9v97jsTiktpyVVxxH2HHx2BY1ZUV4PHv/XJZtQ+0MQhc/jXKCaNuHXVZtlqfDVwaHz028BmtcZpZnmF1WjdO4AjuaSyTQYO6nWxkgYrzFcPhXB/yOy83yQuUtMZ8xfrrdxuVmuchbElAPUVfQodJjo4rKTCAc3+p8/4UmeD7oZDPvb2TOaW3ZqGTdqjo+jr22fRv8+lRUoIHK8+6lsmYGLoo7L5o1KLiNBqXD7ua1ByOe13KATG2nGCInHAumITZt1kWPmIP4566LjemJdvvWTkJCMkd5Cbuag2vLOWJKBZ6yImyPhQ3ILyxyqsUE1E7Z5Ixm+I7X7p/KJ3ueJhQK4fVmYdymyF/Raf0GDknJcS+asPJidzaZa5eoyLSH6Q7M8ngs9gkMb546y7axKur2/sRkgp3JE6otegxKKoe3zjRZto07ksoA0U+H+1ADp55deSH64scLdx7qcK+JEeK7uStllou8Jf0hh8ijFK7GtDwna3Xu3GFqSNu3mWWBBrTlxW1cltitat4SeOeJ2P2u5tg6ItP9jDThyHDkXSIxkZ5U02Y5fdCza3CCjIevNsldGpfClr8QblxBs1tGe0+Izt4Q8xa/xLvNnWN3CjQxrujm9ezS5VRXZGEAdURx1VSKCfHq2jVZe0+Rp8botH69vspBw3dC8+6h15d+QDqiIWKua66Z2raa26GMQc7xOFRLKby2wmOZW0sVbPiXWeNx2iztmga5+KR2918gWeLznLRQD5HHtlDhoAmAk7U6dzVDWQ1U1MMFq8BfhavhI9/+1F/0MKpzh3nOn34Ox1wB3a3m9o8/jK0nbrqfbLfw5l0iMZGe6NQUA8urZfdP65agbYupMX3+JkKn3cj3n2nhvrV/or7Sz/fPmDHingtCjKbg9rd4223ggAl7f+5oqZs8FbbBW6+9xPHHHpe9Nxb5Z4xO69fnwPXP9nL5KauoKVE0dWvueHY3/3F2eq8f0RAx10U3rUPFtdrr8+5Fpdtqv6dzWqaN0R4HeSmXv2OuaCf58SAbCfVExsieP0QhxzVz7L5+rxnzMHC6rNfvNbVMd3weHr0WnCBa2bS1NKGWngm/PtV0r9rwqGkFnH2jSQK1La6VI4fT/eRlIjGxd1aKqSl2bzdDD5IlyGheDxsepXl3D/et/ZD6Sj83zj2C258z3Wul54IYE1wXe+fbbNBTmRbI3tuGy+txUfRtf5ONzZ3Ze2ORf8botH6hsMuT65qZu2wjn138HnOXbeTJdc2Ewum1lKUaItbStfeKAt3VHAumwUzXdd/56K40W+1zOW3WGO1xkJdy+Tvmikox97Yq4EqEcUBaqIfI57FxXQvr89+GzS/CRatN9++uZhMYn3g9vPWgeXLbFnjocvSipwj4Usy5CDgnfgfPjjf7azrDjSuw/JNyUtshicQKlQZfaeKYHV+pSZxXWmu6dscnLDvzVtNrItBAdWACL3/nC/iDu+jtbeV7n6/iR8+5NHeGpOeCyL3WTXicXt5lKjOz2EKtbR89pfUc2bmRH6x+iyUXH53WdEFiHBqj0/p5PcmnvfKmkVQMRjZEzA33YSe5JnLDfemN37Zsk0Qz/pxWWp2dls0x2uMgL+Xyd8wVZSWfrlTJ+SOfSUA9RFWlPpwOC0omwZRPmjT3ZbVQPhlOPQxe/TVMOxnqZ5lW57Yt2DpEW9BKOndvWHm5+pmexC5Xz+zmR2eHqS7PzQFFEokVoHAvrLkbjvta7ETV1wHn3AWtm8FTDGf9ypTl1k0mmO5swjl3Bd6ySdQ2r0fddz4VbVuoDTTwmzlLaC2bJj0XRO41rQOgu7SBNOOAUdMXOJhP97zMRe81c96dL3PyJ2qpKvVxzP5VNFRJxlYRET+t3xjK8u2xFDd99Qiu+90b/V22b/rqEXjSrEAfyRCxsPImnYc67YRoTiR4DexnhidpbXoHOlkIaqM9DgZ2U85xj4N8pIJdZraRIxrN76cseGMl6vB5ud60zNFObLrSaHb9VxbD7J/mesvECEhAPUSWpWi3JhBwd5mx1L89PbGGaeuL8PbvzY5y/4UQaMDyFFE2sZK2OUsIrF6YMF6olQk8ua6ZJ9cldhX6wVekK60YRZ5iMyd6/NQUc26DsskmYdmKRrO8fhaccA186cd0+SfTXVRDdU/LoK55gdULqbjkGZRVntvPJcY998PXcbWFt7I+6+/dHZjGxA+f5bszFbev7+Knm8281B5Lce2XpvOPnz8w69skxqgxOK1fT9Dhob9+yG8WfRrbUjiu5s4/vc/XT54GaUybFR0ilmwmkr3p9lbSM+CaqG3OErS3Mr1ZI5RtegauvirxnFachYnox2iPg7zkKYaG42DJVxJ/R08BT5tlec10pQPLbiF3cx8HJKAeKtdlQudGlM8f664BsczI0UDaXxlLslFazb6livaSGfQtegqPDmF5ilCl1aiukCQBE5mnndjBG8zt6qvQix6Hqmmo6PJta/qnUNm18C9YHg1u8u5tKhstAULsRdemv7BVT+WAidnvVdMTOBiAk0o3cuj5F9ATcmjtDrFqzVZufGI9E/weLjhm36xvlxDp8Ptszv7UFC7+7asJLdR+X3rXHyMZIlbhL+KDsmlsnf0gAZ9LW9CirKyOff1p7sduKOk5jUWPpvf6kRijPQ7yknaT/44XP5Hb7cok7Zru7fHd3L0lkuU7z8neP1Tdzdj3z0+dGTkaSAcaeHP2g7SWHwSWhWUpKkuLKQpMxq5sQJXXgmVJEjCRHSmmGFFdTagdbyVNkNEWtEzFToqEOtqWMipyTGt8O17jdfcADp2Y/bcP+mvp89dSue0ZlFKU+DxMCfj55y9O48ipAW54+O+8u6Mj+xsmRBrCru7v7g1m/PN1v3uD8BCmvooOEZtSWUJ1eVHa+VYsS7FvVRl1+zRgBaZSt08D+1aVpZ+vxU2RKdnNUu++aI+DwFRzK8H08Ojk+YUKObjUAH++GcJ9ZkG4D/58MzIJaX6TI8BQRZNRpMqMHOqGxqX0qhK+8pt32N3j7HFexlzMNS3GIWUlL6+7P4QXbjHDFeKybDrnrqBsYp2p2CmpRp93b8LjbXOWsLm3ROahFrm1632Kwh1s9h7EJH8O3l8pdtceTcVHL+Lpa+tfbFmKK088EL/X5pur/kbYKdyLQ5FbI5kHOhR2kyYVSzfL90gNNxgHwPYmP6fZ0m02r4zRDPiZpJRlunw/+V0zbPTJ75r7kpQsr8mvN0Tajuz8SYIQzl0OVdPg1bvQwU7qK/1s2tm11ykkRnRSESINYat48NQUZ95qyvG2NSYJ2ak/IXz1a7w5+0FaSg6MtRZYFm3lB/Hm7AfZuvAvrD1lFRc/1sWCX7+a1vQoQmSKu+VlAIKB3I1V3l17DJYOU73xdwnLK/xeFh2/H3/b1s5d/7cpR1snCll0Huizb3uBE278X86+7QU27OhIO6hWSvX3jouqr/SjVD5cgyg4587Ec9o5d5rlIn9Ex6PH/46FPh493AfP3mCGiC561Nw+ewMq2mIt8pKMoR6iNquC8sYVeFbON0HIBQ9Azy4I9UBvu8mcPGMOu4I2N849gp8/uYFb5x+V680W41yHKmdCxVQ8FzxgMqLaPvjLXbH5z7etgSe/y9uzH6TTU0mtshIqdrqDLl/5zTuD1ivzUItcav/7M7i6nIk12U9IFtU74QA6qw6n/s1badnvDIIldf2PHXtAFS+/v4v/9/Q7nPyJGg6qkSR+YvSkmgf6oatOSGumDlvBjXOP4FsPxLJ83zj3COw8iEnDWHjffhTmrzIzV7gOvHYP4WOvQtqo84hlQfUhcPHj4IRMD4OyuoLuQu/YxXg6m/rz1QCmZ6BdLEFZHivo304pdRrw34AN3KW1HnFO+u6gy4p1PhZd9CTtHZ3s3u1lv5IAxU4oIWNf7bx7uPW1Fpo7+yTBmMg5ZSmCnW14HlzQX0bdxmVYuzbChkch0EBo3j0EgxOpLvUxsSSxu9VIpkcRIiO0xrflT/yveyifqs3txdfH0y5g/1dv4Ig/nEHHpCOxnF48oQ6CxdXcUHMs3/5oEtfc+xr3X3k8pT7b7IO7NppENJOPBG8BZ7QVGTOSeaABLMtiyYub+P4ZMwj4vbT1hFjy4iZ+fPYRmdjcURUsqsRz+FxU3MwVunEZwaJKCajzietC8/rBGdNrZhRsUN3pCVB+7gqTjynymZ1zV9DpCRDI9caJYSvYgFopZQP/A5wCbANeVUo9rLVeN5L1+jw29679kA9aq7ji8wfitxTK3g33z0vIUuhddQEXzH6QhZ85QBKMiZwLuO2oaDAN0LYFa+UCui/8A72f+w/8JSW0qQlMLrWoKSvCM2BC35FMjyJERjRvoDS4k3eKzuLU7Cf4TtBX3sCmT99AzfsP4G9/D20X43iKKW19mxkf/pGHFWzfNZGmn09kP7UDFYxLVFY+Gc74L5g+O3cfQOSlkVZ0VpX6+MYp0/PyuF4SaketTDynqZULKLnkGSiWCqq80d0cC6bB3N53vsmgPsammRstLopt3n2pWfgUHh0krHw0OaWUy3CFvFawATVwNPCe1vp9AKXUfcAcYEQBdXxgsXLtNr40o4bbzqhJmqVwRk0RqkISjIncU07yqa9aOno5f+WH3HnRLKbXlqQsqyOZHkWITOj524MUaUVP9ZG53hTABNVbP/mNQcu9Pc2UtbxBx7Z1bG/r5oPy4/nckQdgVe4Lfbvhjfvgvvlw9mI4ojEHWy7y1UgrOvP5uJ7qnCbTOeaZcPLfkXDh/o4Bv4+27hAv77Ap8ZXRHXTYt8oi4B/7FVkitUIOqKcAW+PubwOOGfgkpdTlwOUADQ0NAx8eJNkJyKbNdFOJPygEGrC9RZAHJyaRf4ZabvszaQ4oozWVE3joqv3SuoiKJs8TYriGXG73IPi3Vbyhp3NYfQ7myxqCkL+a1vqToP4kXnwf7vw7nNfk5T8PKTbJn6Z8Cp79d1j9TzDxAKifletNFkmMZtkdLaMREOftcT3FOa2Qs0MPx1gstwnG4e9oWYr9qkopL/bmXUWWSK0wBygYyUrmoNSXWus7tNaztNazqqvTyyo4MCu3Kh2HWQpFTg253KbIpFk0oVYyy4usGc7xNqkP/0pF5/v82T6WAytGb/sy7awDoPEguG99iJtfjWR09RTDid8B/0TTUt3xcW43UiQ1amV3lI3bWULGY3boYRir5bbfOP0dx+1+W8AKuYV6GzA17n49sD0j72RZJoHCpc+YbioeM3dvoSZUEHlIyqgoID1/+m/C2k/7Pp8hL2b4iXPRIdAWhFtfC3LYJJvTDvBC8QT4wr/C49fByoWw6A8yn64QeyLntMIgv6MoEIUcUL8KTFNK7Q98CJwHzM/Yu1lWwSZQEAVCyqgoBDv+TtGGR1junMYX9i/J9dYMmVJw1WGweTdc+1wPB0+0OCBgw8T94bivwZ9vgqe+D7N/ytZd3bR1h2ioKqHCP/wAW2vN0+t28Ma2dg6ZXM6ph9bhteWCVeQ5OacVBvkdRQEo2IBaax1WSl0NPImZNuvXWuu/53izhBBCDJcTIvzw1+nAz18mnsnx+RdPA+C14Tsz4et/hkuf6GHVnBKq/BYccCJ65wbUK7/izjeC/Lj1JABspfjMtElceOy+fGF6NZ4hBMO7e0P84N4/E3r3f5mk2lml67i77jh+ueAY6ivz9AsUQgghxpCCDagBtNaPAY/lejuEEEKMkOvC49fj+fBV/j14FadPL8/1Fo1ITQn86yz4/ssuX13dxXeOLcZx4dcfnssi52Mu67mbI+pbeX3a11jXqvjzuzu5bOkayoo8HLrPBA6oLuWASWXsP6mUI+orqJkweKqg9Zu38+I9/86Pgr+nzNfbv/zDlmqW/vIczrnk20yfMraTugkhhBBjXUEH1EIIIQpAZxM8+i/w9iP8KvwVglM/w7RArjdq5A6rgh8eAz9/XXP5k2Yu4apixfpDvsbRvSs4eutDfHrXH+gOHAzlDq6nFU9wN9ZHYd75aF9eCU/jDmcma/R0plaVcfR+E/n0/hOp9gYJ/+XXHLVtKf+gdrO18miapp1OsKSGkrZ3Kd/4MN/pXMzGO//A3z7zPT75xXPBGjx38cftvaz7qJ3dHV2U7fgLDc3PM3nn/+Hva8H1V2Ht/xk8h8yGg04GXwlaa9q6QzR39lFa5KFuQjH2gGQ7rqtp6QqitTZJPfNtELwQQggxgATUQgghxo7tr0OwC/o62LH5LaztrzFx69No1+GnoQt4ueLL/PDQXG/k6Dl8EtzxBXi3zcyyOK0CvLZNEwvYPeWzVGz/M0Vd29EeH+7ESfR4SkApDty9mcPan+Iyz2N02RW8FT6UrW/5Cbyxi1nWespVD2/7DuXDQxvxVE/rf7+Omll0VM9k27a1VG64hwP/7x9pfunf+Hjyasi3dwAAIABJREFUSXRXfoKPQqV8uDtMU9NHlPZsZ6b1Lidbb1OmeunVXl5wD+MDPYO6vl2c8LffU/HGCnpVEX+1P8ma8AG8Haxll55ADz4+smqZMLGWuopieoIOH7X30tzRR9g1E25Ulng5dJ8Kjqiv4ODaciaVFVFe7KG0yMNBNWW5+kmEEEKIIZGAWgghxNhx9yngBAGoBZp0gHudz3CPdTqfOuxg/t+RE/B7Cy+hVvU+SRbWNtBz0OfoSfJQN6BC3RRvfxn/1v/jU63vMEt10OstZ9eEL9B8yFmUTZ6R+g3r9qXviDNY/fJTTN76GIdvfQD/tmDic7zQWdpAV9XJ7NjneNonfYoyq5j9Qi5b2kL8d2svxU1/4/DO/+NI5y2O5y8QN33syvrv8ojaj11dQXwei8OmVFBdVsSkMh8u8H5zJ+/u6OSOP7X0B9kAn5g8gce//tn0vzwhhBAih5TWg6ZmHreUUs3AB2k8dRKwM8ObMxbJ5x5sp9b6tGxuzEBDKLdR+fw7yraPjnwst1Fj6XscC8bT95Hzcgt7LLuF9lvI5xk9OS+7IzjmZkuhlbd0jPXPnPNymy8koB4GpdQarfWsXG9HtsnnLgz5/Hlk24V8j4nk+xg7Cu23kM8jsmk8/j7j8TMXqsLrNyeEEEIIIYQQQmSBBNRCCCGEEEIIIcQwSEA9PHfkegNyRD53YcjnzyPbLuR7TCTfx9hRaL+FfB6RTePx9xmPn7kgyRhqIYQQQgghhBBiGKSFWgghhBBCCCGEGAYJqIUQQgghhBBCiGGQgFoIIYQQQgghhBgGCaiFEEIIIYQQQohhkIBaCCGEEEIIIYQYBgmohRBCCCGEEEKIYZCAWgghhBBCCCGEGAYJqIUQQgghhBBCiGGQgFoIIYQQQgghhBgGCaiFEEIIIYQQQohhkIBaCCGEEEIIIYQYBgmohRBCCCGEEEKIYZCAWgghhBBCCCGEGAYJqIUQQgghhBBCiGGQgDrOaaedpgH5k7+h/OWclFv5G8Zfzkm5lb9h/I0JUnblbxh/OSflVv6G8SfSlFcBtVJqs1LqTaXU60qpNZFlE5VSTyul3o3cVkaWK6XUL5RS7yml3lBKfWpv69+5c2emP4IQo07KrchHUm5FvpKyK/KRlFshMievAuqIL2itj9Raz4rc/zbwrNZ6GvBs5D7AbGBa5O9y4FdZ31IhhBBCCCGEEAUrHwPqgeYASyL/LwHOilu+VBsvAwGl1ORcbKAQQgghhBBCiMKTbwG1Bp5SSq1VSl0eWVartf4IIHJbE1k+Bdga99ptkWVCCCGEEEIIIcSIeXK9AUN0gtZ6u1KqBnhaKbV+D89VSZYNGmAfCcwvB2hoaBidrRQiw6Tcinwk5VbkKym7Ih9JuRUiO/KqhVprvT1y2wQ8BBwN7Ih25Y7cNkWevg2YGvfyemB7knXeobWepbWeVV1dncnNF2LUSLkV+UjKrchXUnbHKK0h1JvrrRizpNwKkR1500KtlCoFLK11R+T/LwE/BB4GFgI/jdyujrzkYeBqpdR9wDFAe7Rr+Ei4rqalK4jfdikNt4MbBNcB2wueYgj3gXZAuyiPH0qrwRpZvUX0PYNhB5/HpqrUh2Ula4AXIrVwuBe7sxncMFgenOJq7N64+95S7FAX2D7arAq6g66UN5F/XBe6myEcBI8PSswx2HUcnM5mPE6vOSZ7isEJmT9PEdp1UTiElQ+tLGynF8tThCqpgp6WQesb/LZynBYi6956AB77Jly7ATxFud4aMRzhIHTu6L8WoazWHGsLWLivF7sn7vrLX42nqDjXmyVGIG8CaqAWeEgpBWa7V2itn1BKvQqsVEpdAmwB5kWe/xjwZeA9oBu4eKQb4LqaDTs62LSjjdn79KC6mmD1VdC2BQINMH8V9LbBg5fFlp13L9TMGHZQHX3Py5auYVtrD/WVfu68aBbTa8vlYk2kLRzuxW5aj1q5wJTN6adjn3h97H6gAU/jUnj1Ltj0J9ScJVz9WBfNnSEpbyJ/uC40rYP7zk84BruTDkE3vY135fz+8s+J18PKBVBWAyfdgIocy72BBphzGzx7A3Q2oRuXoZ7/GWx4NOUxXY7TQuTI7u3Q0wrhXgmo81E4aI7ZcdciNC4zx9gCDarDfb3YLesTrr/sxmWEqw6RoDqP5U2Xb631+1rrT0b+DtVa/ziyvEVrfZLWelrkdldkudZa/5PW+kCt9eFa6zUj3YaWriCXLV3DafspVPsHsWAazG37llgwHV123/mmtWSE77mttQeAba09XLZ0DS1dwZF+HDGO2J3NsYM3wJHnJ95v2wIrL4LjvgZtWwisXsj3Pl8t5U3kl+7mWDANsWNw58fY0WAa4MjzYxdwJ1wz+Fi++iqzvG2L2U+OPD9xfQOO6XKcFiJHtJt4K/JL547YsRgi1yILzPICZfc0D7r+UisXmBZrkbfyJqAeC4Jhh22tPSg3DN6S2M4QlWxZ2xZTAzfC94y3rbWHYNgZ9jrFOOSGE8umvzJ5WbXs/v9rSkzLmpQ3kTfCwaTlWrmh1OU/1b7grxz8f/T+gGO6HKeFyJH+gHpQzlmRDwZem4C574Zzsz3ZMB4/8zggAfUQ+Dw29ZV+tOWBULfpmhIv2bJAw4i6rUTfM159pR+fxx72OsU4ZHkSy2ZPa/Ky6jr9/zd1mwsUKW8ib3h8Scu1trypy3+qfaGndfD/0fsDjulynBYiRySgzm8Dr03A3LfyaUTqEI3HzzwOSEA9BFWlPu68aBZPbNboin3NOLvoThFogIoGOOfOxGXn3WuS2IzwPaMXa9GxeVWlhTm2RGSGU1aNblwWK5uv35t4P9AAjUvhpV9CoIG2OUv40XPNUt5EfimpNsfcgcfgsjqcxhUJ5Z9o+X/hlsHH8jm3meWBBrOfvH5v4voGHNPlOC1EjkQDaenynZ/KamPHYoiNoS6rze12ZZDjrx50/aUbl+H4JQt7PlNaavX6zZo1S69Zs+eh1pLlWwyQ8x8inXILkuVbJMj5D5puuR0yyfJdyMbEF5qxsiuG7rkb4bmfwDffNQkGx66cl90xW24ly/dYzvKd83KbL6R/wRBZlqK6PJpJ0r/H52bmPYUYHo+nGAKxqdk9AMUD7jMRgEqgsjSLGyfEaLGspK0blm1jVdSlfFn0qsGb7ME0WkvkOC1EDkhSsvzn8SVcm4wHnqJiKBp4/SXymXT5FkIIIYQQ+UfGUAshxgAJqIUQQgghRP6RFmohxBggAbUQQgghhMg/OjIzhQTUQogckoBaCCGEEELkn/5AWrp8CyFyRwJqIYQQQgiRf6TLtxBiDJCAWgghhBBC5B+Zh1oIMQZIQC2EEEIIIfKPZPkWQowBElALIYQQQoj8IwG1EGIMkIBaCCGEEELkH0lKJoQYAySgFkIIIYQQ+UeSkgkhxgAJqIUQQgghRP6RgFoIMQZIQC2EEEIIIfKPjKEWQowBElALIYQQQoj8Iy3UQogxQAJqIYQQQgiRfySgFkKMARJQCyGEEEKI/NPf1Vu6fAshcievAmqllK2Uek0p9YfI/f2VUq8opd5VSt2vlPJFlhdF7r8XeXy/XG63EEIIIYQYZdJCLYQYA/IqoAa+Drwdd/9G4L+01tOAVuCSyPJLgFat9UHAf0WeJ4QQQgghCoUkJRNCjAF5E1ArpeqB04G7IvcV8EXgd5GnLAHOivw/J3KfyOMnRZ4vhBBCCCEKgQTUQogxIG8CauAW4Hog2q+nCmjTWocj97cBUyL/TwG2AkQeb488fxCl1OVKqTVKqTXNzc2Z2nYhRpWUW5GPpNyKfCVld4ySLt97JOVWiOzIi4BaKXUG0KS1Xhu/OMlTdRqPJS7U+g6t9Syt9azq6uoRbqkQ2SHlVuQjKbciX0nZHaP6A2lpoU5Gyq0Q2eHJ9Qak6QTgTKXUl4FiYAKmxTqglPJEWqHrge2R528DpgLblFIeoALYlf3NFkIIIYQQGSEt1EKIMSAvWqi11t/RWtdrrfcDzgP+qLW+APhf4KuRpy0EVkf+fzhyn8jjf9RaBtgIIYQQQhQMCaiFEGNAXgTUe/At4F+UUu9hxkjfHVl+N1AVWf4vwLdztH1CCCGEECITom0l0mYihMihfOny3U9r/RzwXOT/94GjkzynF5iX1Q0TQgghhBDZIy3UQogxIN9bqIUQQgghxHgkScmEEGOABNRCCCGEECL/SAu1EGIMyLsu37nkOg5uZzN2UQn0toO3GEI9oCwzfke7YNmgbAj3mv99ZRDsAjcMtjf2mO0zjzthHF85drDDPMeycZSNqzwo7WC5IZSyUEqBdsDy4igP+Ctp7XEIhh18HpuqUh+WlWy2sGQfxIXuZggHweODkmqwLFxX09IVJBh2UEphK7Asa2jrzncpvptCEO7txe5tjpQzD3hLINhhyqRSkfLrNWXTDUEoUoaVAuUxr3NDaMtLp28SYRdKw63YbhDLstHKxtHQbgUIOjqxXLou9Owy+4t2wOOH0tH9bl3HwelsRjl9aLsIu6way7ZHbf0jUsDlKusi36UOB3EsLygL2w0BGpyQKbOWFzzFEOzsP65SVAF9u81x2AnF7Qel0Ndunh85xuIEY69TNqDMfhDujewjHlOWbR9huxg71EVYeem0Kyh32vEo16xLu2jlwVUWlnZxlY2DjYWDx+kDT5F5nhNCWR6014/jagh141o+dtsVBMPgsS1CjkuJzyLgtuOG+wgrL61qApZlM9Hvo7UnlPR8EA6HcTubsdwg2uMH7aCcIK7lwyqrxuNJfRkQf07weWwq/Ta6a+ce97GBrxl4/tjb40IMiQTUeS/c14vdE7s2cfzVeIqKc71ZGTXweswprsZTXNifudBJQJ0m13HQTevwoKArDB+8DPseC8//DI65Ah6+Gtq2QKAB5twGz94AnU3QuAzefABe+sXgx85dDptfwrPvsbByQf/rPV/9rbmge+jy2DrPvBVeWQzH/iMeXynhnjbufjXI7X/eTH2lnzsvmsX02vK9X5i4LjStg/vOj637vHtxqz/BhqYuLlu6hm2tPdRX+rlx7hEseXET3zhlenrrzncpvhtqZuR98BPu7cXetR4VV85oXAqv3gWb/pRQvkyga8PvrzTl9KzbweuHVQuhbQsq0EB54zKc4gCepV9JKKP2K4vxHvtNLn+si+bOkCmXNaVYrZug4yNYfVVGvlvXcXB3rMO7cn7/+sONK6B2Ru6D6gIuV1kX912qti14Ag0wb4m5KHngktj3e+5y0wN05YVx5X2ZOW43HAOrLkpc/vGbULkfvPMUHHZO4uPx+0b02B13HPc2LodX78DX3UrgxOuxBpwTVKABK7IO67PX4tFheOBSKKuBk25I2CfUnNvw+Erh8eugs4mKefdw8+s2Jx5Sy+q/buNfj9aohy7CbtuCHWjA+5UlfPeFMF87aTq/ePYdnlrXlHA+cF0Hmt7Gt3J+0vcLN64gXPOJpEG162o27OjoPyecOqOa/znZv8d9bOBrBp6b9va4EEPWH1BLl+98FO7rxW5JvDaxG5cRrjqkYIPqZNdjduMywhMPkaA6j8nVXJqczmbs++eDf4IJfg85zdweeX4smAZzu/oqOOEa8//KBXDUBckfu//C2HriX9/TEgumo8sevtq81+qroHsnnvZNXPHpCgC2tfZw2dI1tHQF9/5BuptjF/bRdd93Pk5nc/9FTnSd33rgDebOnJr+uvNdiu+G7ubcbtcosHubYwdviJTNi+C4rw0uX+1bzMVJtJz+/kro3pnwWrVyAR4nmLSMBlYv5Hufr+4vl05nM7S+H7uQjz5/FL9bp7MZT/RCP7J+z8r55r1zrYDLVdYl+y67d8aC6eiyzh2xYDq6LHrcjgbL8csPONGUz6MuGPx4/L4R3ScSjvEXmv3oyPOxUp0TouvoaTHBdNsW8/qB+0Tk+B5dt3fVBVzx6Qqu+90bfOvESZQ/lLhtVY8s5PKZE7hy+VrmzpwKJJ4P3Pj9Isn7eVbOx02xj7R0BRPOCZfPnLDXfWzgawaem/b2uBBDobXm47bu6J3cbowYFrtn8LWJWrnAtFgXqGTXY2rlAtNiLfKWtFCnSTl9pvC7YXOrXXPrr4ztFFHR5dH/LTv1Y9H1xPOWpF5n2xbzOOC3nP6Ht7X2EAw77FU4mHTdygn2X+TErzPg96a/7nyX4rshXAAXe9FyGy++bA4sX0olltNImUt4rVKDl0XWUVNiHtvW2mP2nVRlepS+2/79c8D6lTMGfrtCLlfZluy7TFa2UpW3ZMfbti3gOrH9YU/H3vh9YuAxPv45ezsnQOrneUti+1vbFvyWw7bWHvxW8n24pkT1H6ujosdsSwf3+n6WGyKZYNhJOCfUlKi97mMDXxO/Lek8LsRQvPBeC6qpgzobJClZnkp1beKGc7M92TAeP/M4IC3UadJ2ken+Z3nMrbLMbU+ruY0XXR7933VSPxZdT7xQd+p1BhrM46FuetxYoF5f6cfnSaNrq8eXdN3a9lFf6U9YXF/pp60nlP66812K7waPLzfbM5qi5TZefNkcWL60Tiynoe7Brx3YIhC3jqZu81h9pd/sO6nK9Ch9t/3754D1a3sM/HaFXK6yLdl3maxspSpvyY63gQYTEEf3hz0de+P3iYHH+PjnpFpH/Halel6oO2HdPa5NfaWfHjf5PtzUrfuP1VHRY7Zr+fb6fq7lJRmfx044JzR1673uYwNfE78t6TwuxFB0B8NY0UBaxlDnp1TXJlYBt/eNx888DkhAnSa7rBrn3BXQs9uMuVv/hLl9/V4zxi66c0THSb9wS2x83mv3JH/s3OWx9cS/3l8FZ9+RuOzMW817zbkNSiYRrtifxa+2A/SPQ6sqTeMCvaTajN+MX/d592KXVXPnRbP6L3aiY6gfWLs1/XXnuxTfDSXVud2uUeAUV6MHlrPGpfDSLweXr4oG0/ocLadn3Q4lkxJeqxuXEbZ9Scto25wl/Oi55v5yaZdVQ+UBZt0Z+m7tsmoznjNu/eHGFea9c62Ay1XWJfsuSybB3LsTl5XVQuPyAeU9ctyet3Tw8vefN+XztXsGPx6/b0T3iYRj/HKzH71+L26qc0J0Hf4qmHuXWfbCLYP3icjxPbru0Lx7WPxqOzd99QhufH4nHWcnblvLV5Zwx9rd3H7hTB5YuxVIPB9Y8ftFkvcLN67ASrGPVJX6Es4Jd6zdvdd9bOBrBp6b9va4EEPhsRWWio6hloA6Hzn+wdcmunEZjr9wz4/Jrsd04zKc4sL9zOOB0jLupN+sWbP0mjVrUj4ey/JdCr1tGczy7cFVtmT5zoWhZ2PO+Rezt3IbNeQs3+He2GN7zPIdwrKsMZTlO4i2fZLle8/yptwOkjLLNyaZ45jK8q3/P3t3HifHVd/9/nOqqrtnlWY0mpGMx2OzOJYdIIAmLPEr4EC4AbMI41hYtiyTGySM7RBCrglPLjchCU9e4UkIhIDtyAnB8ooMGJsnZnUAs5g8SBgMyDZ2vMiDbc14NCNp1u6qOveP6nWmW+rZurp7vu/Xq1/dXVVddarrV6fr13XqFNa4hMbFscECevmeJnQSHHW7SPv2OL18r8VxnCp7+c5gs+toggyhk1hCL9+V97EV7uU79riFJcSuLKvv/HKE1hvfxMudh6I/2jadG3eRjif22K3XuFUv33Xdy3fscdso1L5gARzXxVm7MXrT0ln9B9vWHXe0B9C2tvT9iaYHejsXufkcJzqDM2+wobcztbh5NosK300z8FpaoOWU0oHtPQuejwHW5N+dVDLOBcr+x+o40L5+wctaiJL9s940cVzVXPa7NFTxA9bWXfq+dW356dqPX0cfT9RguockEM2ltL8BQ6EpmMPxyzx3ncrvMS24RPtacbRXqrs9z4Ouk8qOO5Gyvwkn2MdO9Dui3xlZLq4xavLdBLxUC6QKxyarITGZezy2Gta52anJt4iIiIg0FMcBBzX5FpH4KaEWERERkYbiOU7hDLV6+RaRGCmhFhEREZGG4hgwavItInVACbWIiIiINJTQFjf51hlqEYmPEmoRERERaSihteqUTETqghJqEREREWkopQm1zlCLSHxq3lO7McYB3gCcVrx8a+0na10WEREREWk81hZdQ61OyUQkRnHc+uwOoprvZ4Da6IiIiIjIggSh1W2zRKQuxJFQn2atfVEMyxURERGRJqBrqEWkXsSRUH/NGPNaa+1/LuRDxpgW4B4gRVTuz1tr/9IY81zgVmAd8GPgEmtt2hiTAvYAm4FR4B3W2seXayVC34epZzFhGsIA6yY56q6joyWF4xienZwlDAO67VE8myF0koyxBuO49LQncRyzXEWpa2FoGZ1Mk/YDkt7qWvd64/tp3IlDEPrgeARtfYzPhHQE43g2jXVTTCW6mJgNaUs6dIVHMEEavCS09YJTvsuFMLSMT6eZTgcE1tKScFnfntJ2lpoLQ8uR6VnaMmN4NoPjpTDthdgNg4BwYgQ3mAHHxToJbJAmMAkmE92sbU0BzKuzcsNsGNBNVKcbL0nYup7RKV/1m0gMSpp86xrqxuVnYOKZ/LEJHRvBS8RdqhWVmU3jTg9jwgzWSRC09pFIJeMulixBHAn1d4EvG2MskAYMYK21607wuVngtdbaCWNMAvieMeYrwPuBj1trbzXGXAv8IXBN9nnMWvsCY8yFwEeBdyzHCoS+jzn8CGZyBO64HMYPYroGWLP1Rp5OP49Z3/DRrx7gb8/2SH35Uhg/iNs1QOpte/jAPRne9/pNnLGhs+kPvMLQ8tChY+zcs4+hsWn6u1u5bsfgqlj3euP7adzhA5i9l8D4QegawN16A+2JtbTc9Nb8sPatN3PTgQQXPX8Kc8el+eFceAv0nTUvqQ5Dy+Ojkxw6OsNVn79f21liE4aWJ0Yn6J54mFRR7NoLb8H0nUVoLXb4AN7nLsqPM1uuhrs/jDsxzOx5e3i689c4OhvMq7NSnsNHvxLV6YkvF+Ydbr2ZD31zmq8dGFHci9SYmnw3AT8Dw7+AomMTtt4Afb/etEl1ZjaNN1o4HjNdA5itN5DpOUtJdQOLo5fvjwO/DXQDvcD67PNx2chE9m0i+7DAa4HPZ4dfD7wt+3pL9j3Z8a8zxizPUc7EM5gjB/PJNADjB3H2bqcrOMwTh6fYtXkNPbkDr+z4tV/awa7Na9i5Zx+jk+llKUo9G51M5w9MAYbGplfNutcbd+JQIZmGqBLfewmtZEqGeXsvYufmdrruKI1dbt0GUyPz5js6meaJ0al8Mg3azhKP0ck0E4efmRe7Jhu7wcQIbi6Zzo7jjsvh7PfB+EE6b99BRzBWts56YrR8ne7tvYhdm9eUTKu4F6mNkibf6pSsMU08U0imIXree0k0vEm508Nlj8fc6eF4CyZLEkdC/TBwn7U2Y60Nco9qPmiMcY0xPwGGgW8A/w2MW2v97CRDwMnZ1ycDTwJkxx8BesrMc5cxZp8xZt/IyPyEoWw5wgwk2go7Q874QVImoC3p0tdmyo7vazMMjU2T9qta5YaW9oP8gWnOaln3lbbguA39svHI3P+Yxg/i2grT+vMThbQfxbu2s1RjMfVttdJ+QFcyrBi7JpgtP661O/86ZcrXWSeq04unVdw3p5WMXVmc0KJrqE+g7uO20rFJ6JefvgmYMFN2nU2YiadAsiziSKifAv7TGHOVMea9uUc1H8wm3y8B+oGXA2eWmyz7XO5s9Ly/MK21u621g9bawd7eE54ojz7jJCAzFTVNKdY1wKx1mUoHDE/ZsuOHpyz93a0kPbeqZTWypOfS391aMmy1rPtKW3DcOl7ZeJx33VnXAIGpMK03vylS0oviXdtZqrGY+rZaSc9lPO1UjF3rpsqPmx7Lv5615eusE9XpxdMq7pvTSsauLI61lujqQZRQV1D3cVvp2MSJ44rU2rBOouw6W6c5m7ivFnEk1EPA94A1RE29c4+qWWvHgW8DrwS6jDG5Pa+fKGHPLecUgOz4tcDhJZY90rERu3YAtlxd2Cm6Bgi33si4u45T17Wxe/9RRt9yfcn4I2/bw+79R7lux2C+o5tm1tOe5Lodg/kD1Nw1hqth3etN0LEBu/WGkni0W29gmkTJMH/rzVy3f5LxLaWxy4W3RB2TzdHTnuTUnjb+/vdfrO0sseppT9KxbuO82LXZ2HU7egnecXNpXG+5Gr7/Cega4Nh5e5hwu8vWWaf2lK/T/a03s3v/0ZJpFfcitRHY4muo1eS7IXVsjK6ZLq6Xt94QDW9SQWtf2eOxoLUv3oLJkhhbw0rIGOMC/9Na+8FFfLYXyFhrx40xrcDXiToauxT4QlGnZPdba682xlwBvMhae1m2U7K3W2u3Hm8Zg4ODdt++fVWVp7pevkO67RH18t3cvXzHvjLVxm2lXr47g3Fcm8G6yWXo5RtaEo56+a5/sW+chdS31VpYL98e1vHUy3djqYsvdyViVxbuzp8+xcu/+FtsNGNw7j/Ay3fGXaTjiT126zZu1ct3PffyHXvcNoqatqmw1gbGmN9c5MdPAq7PJuUOsNda+7+NMQeAW40xHwHuA/4tO/2/ATcYYx4hOjN94RKLX8LxPFhT+AfNAF1F4/s6W7Kv2gBwgdX435PjGHo7U3EXQwDPS0LXKYX3wPokQKGJ61pgbVvuXQvVcBzDuvYUtC9TQUUWyXEM3e0tRD8XZca7Ls7a0nobovq5+FCmXJ1VGJbfQXCA3k418RaJgy25D7XOUDcsL1FybLIaJFJJSPUD0e9QHM2FZXnFcZHCfcaYLwK3AZO5gdbaO4/3IWvt/cBLywx/lOh66rnDZ4ALllxaEREREakrobWYXJNv9fItIjGKI6HeQJRIn1s0zALHTahFRERERACCUL18i0h9qHlCba29pNbLFBEREZHmEarJt4jUiZo32zfGvMAY8zVjzE+z719sjPkftS6HiIiIiDQmay1uvpdvnaEWkfjEcR38vwKbUJoFAAAgAElEQVR/BfkLX34GbI+hHCIiIiLSgIIQjJp8i0gdiCOhbrfW/iD3xkb37crEUA4RERERaUBh8X2o1SmZiMQojoR61BjzXLK1nzHmbcAzMZRDRERERBpQ6W2zdIZaROITRy/fVxLdI3qTMeYJ4GmW+R7RIiIiItK8Qlvo5TsMre7lKyKxqVlCbYx5u7X2i9baR4DXGmPWAsZaO16rMoiIiIhI4wvCwn2owzBQQi0isanlGeoPAV/MvbHWHqnhsldMGFrGp9NMpwMCa+lIuXSHRzBBGrwktPWC4xCGltHJNGk/IOm59LQncRwTd/FlFfEzPsHkCG6YJnCSuO29eIk4GqmIrJzj1bW+HzI8MUsmCEm4Dn0dKTyvisPwMISpEfBL63URiU/xbbOCMIilyaXIYvi+TzgxghOmCZ0kTkcvnqcIbmTaeksQhpbHRyc5dHSGqz5/P70dCf793HbMHZfC+EHoGoALbyHsPZOHhifZuWcfQ2PT9He3ct2OQc7Y0KmkWmrCz/gw8gCpvRfB+EG8rgH8rTfj956ppFqaRhhaHjp0rGxdG4aWBw8d47Ib9+fHXbt9M5s2dB4/qQ5DGD4At24rqdfpO0tJtUiMbHGT70Cdkklj8H0fhh8gmT0eI3c81nemkuoGVsujgU3GmPvLPH5mjLm/huVYNqOTaZ4YneKqz9/P0Ng0Hzqnl65cMg3R863bCCZG8gd4AENj0+zcs4/RyXSMpZfVJJgcwctV3hAl1XsvIpgcibdgIstodDJdsa4dnpjNJ9O5cZfduJ/hidnjz3RqpJBMQ75eZ0r7jkicgjDEMYUz1CKNIJwofzwWTug3pZHV8q+Qx4C31HB5Ky7tB7Ql3fwBWl+bKewgOeMHMUE6P03O0Ng0aV8/AFIbbpguG5tOqDvWSfNI+0HFutYPbdlxfnCC3oH98vsOvv4QFYlTWNSzdxjoeEoag6PjsaZUyzPUaWvtE5UeNSzHskl6LlPpgP7uVgCGp2zUHLBY1wDWTeanyenvbiXpubUqqqxygZMsG5uhk4inQCIrIOm5FevahOuUHee5J/gZ9MrvO3jJ5SiyiCxWUUId6LZZ0iBCHY81pVom1N+vZiJjzKUrXZDl0tOe5NSeNv7+919Mf3crH/n2CONbri/sKNlr7dyOXq7bMZg/mMtd19fTrgMyqQ23vRd/680lselvvRm3vTfegokso572ZMW6tq8jxbXbN5eMu3b7Zvo6UsefaVtvdM30nHqdNu07InGyQfEZaiXU0hicjvLHY06HflMaWc2afFtrr6xy0j8Grl/JsiwXxzGc1tNOV1uCz+16JYGFMOVg//CbJb18O47DGRs6uf3ys9XLt8TCS3j4vWcy+86v44QZQiehXr6l6TiOqVjXOo5h04ZO9r77VfhBiFdtL9+OE3VA9q5vqpdvkToS2kIz7zBUQi2NwfM8/L4zSRcdj6mX78ZXj1uvobJMxzGsa09Be/HQlrLT9Xae4EyIyAryEh5e10lxF0NkRR2vrvU8h+d0tZYdd4KZQseGJZZMRJaTDQs9eyuhlkbieR7oeKyp1ONf7Lr3gYiIiIhUVnKGWp2SiUh86jGhbqgz1CIiIiJSW8U9e+sMtYjEqS4SamPM+UVvq+q8TERERERWJ2vV5FtE6kNdJNTAx3MvKnVeZow5xRjzLWPMA8aYXxhj/jg7fJ0x5hvGmIezz93Z4cYY80ljzCPGmPuNMS+rzaqIiIiIyIpSk28RqRP1klBX08zbB/7UWnsm8ErgCmPMWcAHgbuttacDd2ffA7wROD372AVcs+ylFhEREZGas0VnpYs7KBMRqbV6SahPWBNaa5+21v44+/oY8ABwMrCFwm22rgfeln29BdhjIz8Euowx6lJPREREpMEVN/O2Vk2+RSQ+NbttljHmZ5RPnA2woPuRGGNOA14K/BewwVr7NERJtzGmLzvZycCTRR8byg57es68dhGdwWZgYGAhxRCJjeJWGpHiVhqVYrcO2eKEWmeoy1HcitRGLe9D/eblmIkxpgP4AvA+a+1RYyq2Fi83Yl6Na63dDewGGBwcVI0sDUFxK41IcSuNSrFbh4quobZW11CXo7gVqY2aJdTW2ieMMW8DXgD8zFr7tYXOwxiTIEqmb7LWfjE7+JAx5qTs2emTgOHs8CHglKKP9wNPLX4NRERERKQelDT5Vi/fIhKjWjb5vhr4deAHwN8YY15urf2bBXzeAP8GPGCt/ceiUXcClwJ/l32+o2j4lcaYW4FXAEdyTcMXLQxhagT8NDbRAv4sBBlwXDAuAIHbEg0LZgmcJMfctQTWIWEsbf44nk2Dl+KwXYNxXLpbE4xNZ0j7AUnPpac9iePEfyvuMLSMTqbrolz1VJZGlpmZxZsZhtAHxyPT0os38ywmzMaw10JgPKy1eP5U9O+/cTAm6mohbN/I6HRA2g9oSzp0hUcI/Vl8k2Aq0c3a1pS2S60V1Ul4SWjrBadeusZYWbl6IQxDAguhtTjG4BpIeA5+aMn4Ia1Jh05/HBPMQqINbIDxZzGOS5hoxQ3SEGYgDAgS7RCGOMEMxnEI3FYA3MwkOC7WSWCCGYxxIdFK2NLNkZkMbZkxPJshdJKMsYaZwNKScFnfXrRPrOJtJbISrJp8NwXfn8GdGMkfmwQdvXheS9zFWlH+zAzuTNE6t/TitTT3Oje7Wjb5fjXwG9bawBjTBnwXqDqhBs4GLgF+Zoz5SXbYnxMl0nuNMX8IHAQuyI67CzgXeASYAv5gSaUPQxg+ALdug44+zOs+DHdcDuMHoWsA3vop+K9/wTvng/Dtv4OH/gOvawDngpu45bE2tj13isRtF+en7zz/Rt73rVn+6HVn8Mm7f8nXDwzT393KdTsGOWNDZ6yJSRhaHjp0jJ179jE0Nh1rueqpLI0sMzOLd/gBzN5Lohh81XtJvOj8wvuuAdi6B6+9F6ZG4XPb58W2ec0H+Mx9Hj98fIw9b+7A3L4Dd/wgbtcA01uu54mO0zm1p0PbpVaK66TctrrwFug7q+kTtVy98PFvPMSlv/Vc/uwL9+frh49d8Bu0JByuuPk+ejsS/Pu57STuuBQ6+mBOve1svTH64+i2aLz3+o/A7bvy470tV0OyHb5yFUwMY7ZcDXd/GCaGYcvVhGsH8I4dJnX7DsjuC8kt1/PuuyYZmcgU6irsqt1WIiumuCMydUrWkHx/Bnf4wZJjEXfrDfh9m5o2qfZnZnAPl1nndZuUVDewWv6Sp232Ihdr7RTV3Sorz1r7PWutsda+2Fr7kuzjLmvtqLX2ddba07PPh7PTW2vtFdba51trX2St3bek0k+NFA6Gzn5f4aAMouc7r4SXbIPPXRw9Z4cnbruYbb+eKiTT2eGtX9jOrs1ruOzG/Zy/OWqZPjQ2zc49+xidTC+pqEs1OpnOJ7Bxl6ueytLIvJnhQuUN8NKLS9+PH4S9OyAMCsl0bng2ts3eS7jiNzv40Dm9dGYTiNw0XXdcysThZ7Rdaqm4ToLo+dZt0fAml6sXzt98Sj6Zhqh++NPbfsrhyQxDY9N86Jxeuu64tHK9vXc7TD1bGJ9LpnPj77g8Gn/2+wrvi157wWzZfeFD5/SW1lWreFuJrJjiW2WpyXdDcidG5h2LmL2XRGesm5Q7U2GdZ5p3nVeDWp6h3mSMuT/72gDPL3qPtfbFNSzLwvnpQvC3dhde54wfLAxv7S4Z7lq/7PR9bYahsWm6WhP5wUNj06T9eDvXSPtB/gA1J65y1VNZGlo4JwYdt3wM2+C4sd3ihPS1mbLTdCVDbZdaKq6TcsYPRsObXK5e6GpNlK0f2pLRJTglsVqp3k60nXh8bpri+n38IJjy+0Jfm8mXJe0HYFbvthJZKWFY+L0J1eS7Mc09NoHofejHU55aWI3rvArU8gz1bwCXA28h6vH7zOzrK4G317Aci+Mlo2Z6ANNjhdc5XQOF4dNjJcMD45WdfnjK0t/dyvh0Jj+4v7uVpOeu0EpUJ+m59He3lgyLq1z1VJaG5syJwTAoH8PGPW5sz4QOw1O27DTjaUfbpZaK66ScroFoeJPL1Qvj05my9cNUOjrQLonVSvV2ZurE43N1enH93jUAtvy+MDxl82VJeu6q3lYiK0ZNvhvf3GMTiN47tTzfV2OrcZ1XgVom1B8Hjlprnyh+EF3f/PEalmNx2nqja966BuD7n4AtVxd2iNx1pj+5Bd5xU/ScHZ654CZu+cUsmQtuKpl++vwb2b3/KNdu38wX9ke3y85dH9zTHu9BVk97kut2DOYPVOMsVz2VpZH5LX3YrTcUYvC+m0rfZ6+hxnHhHTeWjW279QY+/aMJPvLtEY6dt6dkmvEt19OxbqO2Sy0V10lQuC63rTfectVArl74wv4n+ej5Ly6pHz52wW+wrj1Bf3crH/n2CONbrq9cb2+9EdrWF8aft7t0/Jaro/Hf/0ThfdFr302V3Rc+8u2R0rpqFW8rkRVT0imZEupGFHT0zjsWsVtvIOho3roxaKmwzi3Nu86rgalVz4jGmJ9ba19YYdzPrLUvqklBjmNwcNDu23ecS60X1Mt3msBJlOnlOwNeUr18N2hZyoi9ICeM26yF9fI9DdZXL9/1bvE9R8e+oaqN20rm9vJtrcUct5fvNCRas718pzGOU7GXbzeYgXm9fHtYx6uyl29oSTjq5Xv5xR63sPTYleXxt5+9nT9//J0APLjhTWx6z83xFuj4Yo/deo1b9fJd1718xx63jaKW7QuOFymtxxlXPxwHOjYAlSPMm/M6VTK2sJp9RUN7O0unqgeOY+qmXPVUlkaWaElBS+HW7EmAlv6SaY5XIThAb2fxFC24gMvcOJeaKaqTVpuF1Qsby89jzvu58V94H505mFvvO0B3uwucBET7Qh8VrOJtJbISSm6bpU7JGpbntUBX4dhkNTR89lpaSo7HVsM6N7ta/j3+I2PMzrkDs7e72l/DcoiIiIhIIyu5hlqdkolIfGr5p8j7gNuNMRdTSKAHiU6UnVfDcoiIiIhII9M11CJSJ2qWUFtrDwG/ZYz5HSB3LfV/WGv/s1ZlEBEREZHGZ8Pi2zTqDLWIxKfmzfattd8CvlXr5YqIiIhIcyjuVFfXUItInNTFqIiIiIg0FKP7UItInVBCLSIiIiINxVo1+RaR+qCEWkREREQaSnGTb52hFpE4KaEWERERkYZS0uRb11CLSIyUUIuIiIhIYym5bZaafItIfGrey3czCEPL6GSatB/QnnKZSodkgpCE69DXkcLz9D+F1B/fDxmemM3HalvSYXI2IOm59LQncRwTdxFFqlJcBy8mfqv5/FKXISIrq/i2WQadoZbGkckEDE/M4ocWzzH0daRIJNy4iyVLoIR6gcLQ8tChY+zcs4/fel4P2191Kpff9GOGxqbp727l2u2b2bShU0m11BXfD3nw0DEuu3F/Plavvvhl3HjvE/zg0VGu2zHIGRs6lTBI3Suug3OxvJD4rebzS12GiKy8kttm6Qy1NIhMJuDB4QneU3Q8ds32zWzq61BS3cCU9S3Q6GQ6f5C189XPyyfTAENj01x2436GJ2ZjLqVIqeGJ2XwyDVGsXn7Tj9n56udFsbxnH6OT6ZhLKXJixXUwsOD4rebzS12GiNSAbpslDWh4YjafTEP0+/Ie5Q4NTwn1AqX9IL8TuI7Jv84ZGpvGD1SxS33JBGHZWHWzZ9uGxqZJ+0G5j4rUleI6OGch8VvN55e6DBFZeaX3odYZamkMfmjL5w6hYriRKaFeoKTn0t/dCkAQ2vzrnP7uVjxXX6vUl4TrlI3VIFuB93e3kvTU1EjqX3EdnLOQ+K3m80tdhoisPFtyVlonMqQxeI4pnzvocqKG1jCZnzHmM8aYYWPMz4uGrTPGfMMY83D2uTs73BhjPmmMecQYc78x5mXLVY6e9iTX7RiMrqm751Guvvhl+R0jdw11X0dquRYnsiz6OlJcu31zSaxeffHLuO6eR/PXh/a0J2MupciJFdfBwILjt5rPL3UZIlID2YTax9UZamkYfR0prplzPHaNcoeG10idkn0W+BSwp2jYB4G7rbV/Z4z5YPb9nwFvBE7PPl4BXJN9XjLHMZyxoZPbLz8738v33ne/Cj8I8dTLt9Qpz3PYtKGzJFbbkg5//Lunc5W3ST0YS8OYWwcvtAfuaj6/1GWISA1kE+oQp7T5t0gdSyRcNvV18Lldr1Qv302kYRJqa+09xpjT5gzeApyTfX098G2ihHoLsMdG3T7+0BjTZYw5yVr79HKUxXEMvZ2Ff5K62pZjriIry/McntNV2sxIsSuNaG4dvBKfX+oyRGRl2TBKogMcnaGWhpJIuJzcrQOwZtLop1I35JLk7HNfdvjJwJNF0w1lh4mIiIhIgzNFZ6h1DbWIxKnRE+pKyrXLK/v3pTFmlzFmnzFm38jIyAoXS2R5KG6lESlupVEpdutR7gy1rqGuRHErUhsN0+S7gkO5ptzGmJOA4ezwIeCUoun6gafKzcBauxvYDTA4OLigGjkMLc9OzhKGAd32KJ7NEDpJxliDcXTNnaycxcRtGFrGp9PMpAMy2et2WpMuXa3ZOA1DmBoBPw1eEtp6wWnW/9wkDkupb5dLGFpGJ9Pzro2eN7zNw5l+Ntb9oVJZpfbqIXZljmwSHRo1+a5EcVuffN8nnBjBCdOEThKnoxfPa/SUbHVr9K13J3Ap8HfZ5zuKhl9pjLmVqDOyI8t1/XROGFoeOnSMT3zjQf72bI/Uly+F8YO4XQOk3raHD9yT4X2v38QZGzp1ACSxC0PL46OTjE7M8id7f8rQ2DT93a38/e+/mA1rWjhtXSvOyANw6zYYPwhdA3DhLdB3lpJqaRq5envnnn35feC6HYOc3tvBwyMT+eG/d1Yvn/7dVpy9F8W2P1Qqq35TRCImjO4LH+Jg1ORbGoTv+zD8AMmi3xd/6834fWcqqW5gDXOkbIy5BbgXOMMYM2SM+UOiRPr1xpiHgddn3wPcBTwKPAJcB1y+3OUZnUyzc88+dm1eQ082mQZg/CBrv7SDXZvXsHPPPkYn08u9aJEFG51M88ToVD6ZBhgam+aqz9/PE6NTBBMjhWQaoudbt0VnrEWaRK7eLt4Hdu7Zx/DEbMnwXZvX4OUOdiCW/aFSWfWbIpKTu4ZaTb6lcYQTI/N+X7y9FxFO6HirkTXMXyHW2m0VRr2uzLQWuGIly5P2A4bGpulrM4WdImf8IH1thqGxadJ+sJLFEKlK2g9oS7r5g/OcobFp2pIuJpguG8f4OniX5pGrt4sNjU3jB2HJ8Er1ei33h0pl1W+KSMQUNfnWGWppFE6YLvv74oSZeAoky6JhzlDXm6Tn0t/dyvCUjZoDFusaYHjK0t/dStLTfeUkfknPZSod0N9detus/u5WptIB1k2VjWO8ZA1LKbKycvV2sf7uVjzXKRleqV6v5f5Qqaz6TRGJ5BIQn4TOUEvDCJ1k2d+X0EnEUyBZFkqoF6mnPcl1OwbZvf8oo2+5vrBzdA1w5G172L3/KNftGKSnXQmJxK+nPcmpPW18fOtv5A/Sc9dQn9rThtvRG10jWhTHXHhL1BGTSJPI1dvF+8B1Owbp60iVDN+9/yj+1ptj3R8qlVW/KSKRFLMAzJgWKtzIRaTuOB29835f/K0343ToeKuRNUyT73rjOIYzNnTyN+e9mEwYMvvOr+d7+Z5lDR85Tz2ySv1wHMNpPe10tSXYu+uV+KHFndvLd99Z8K5vqpdvaVq5evv2y8+e13P23OFOmxfr/nC8sooIJG2UUKdNChPOxlwakep4noffdybpd34dJ8wQOgn18t0EtPWWyGAIQjjqraOnPUnCMazP3urk6SPTOgiSuhKE0f/4rUlvflw6DnRsqPhZ3cJHmoHjGHo7UyXDysU2wIjtIm0DktalB1PzJl3lyioikVQ2oc44SZLBTMylEVkIh2fpImNDEjj0qcFww1NCvUjV3n5FtzqRerDUW/DoFj7SrCrFdspz2PGZ/6N4F6lTSZsmg0eAp2uopWH4fsiDh45x2Y37878v127fzKYNnXieEutGpS23SNXefkW3OpF6sNRb8OgWPtKsKsX2E6NTineROpZiloxJgjHq5VsaxvDEbD6Zhuj35bIb9zM8ocsWGpkS6kWq9vYrueG61YnEaam34NEtfKRZVYrttqQ7b5jiXaR+pOwsaZPKJtQ6Qy2NIVMhT/AD/SnUyJRQL1K1t1/JDdetTiROS70Fj27hI82qUmxPpYN5wxTvIvUjOkOdAEz+ntQi9S5RIU/wXKVkjUxbb5Gqvf2KbnUi9WCpt+DRLXykWVWK7VN72hTvInWsxabJZM9Qv8AMwZ3vjbtIIifU15Hi2u2bS35frt2+mb4OdUDZyIzVv3p5g4ODdt++fVVPX6nXY/WGvKrEvmGrjdulxqXiuqnEvuEWWt+upEq9fCve605dbIB6it1lMTECqQ5ItJ542jry3b98DQPJY4x5vbxk+ofQth4+8N9xF6uS2GO36eK2gfl+yPDELH4Q4rkOfR2peu2QLPa4bRTq5XsJHMfQ2+ZGP0aZDBxLQMdGHNfTrU6k7jg2oDccAZuBMAF2IwupAnQLH2koYQhTIye+j3QY4kyN0BukIZGbLjqGULzLqvBvr4cXvh1e9xdxl2RBWoo6JQMgMxlvgUSq5JmQ55hRMBkwCTAbUaPhxqaEeikCHw79HPZeAuMHoWsAtt4AG14Irr5aqSOKVVlNwhCGD8Ct2wrxfuEt0HdWaVJd7XQizezor+DIr+IuxYK12FkyppP8SbTMdLRPa9+VeqbjsaakWmcpJp4p7BAQPe+9JBouUk8Uq7KaTI0UkmSInm/dFg1fzHQizcpPQ5CG9ETcJVmwFGkyTgpM0aFsZiq+AolUQ8djTUkJ9VIEmcIOkTN+MBouUk8Uq7Ka+Ony8e6nFzedSLPKNZNON15z6RYzi2+S+RbfQG0T6se/B3dcCVOHa7dMaXw6HmtKSqiXwk1ETTWKdQ1Ew0XqiWJVVhMvWT7eveTiphNpVrMTpc8NpIUMGZPEFh/K1vJM+zM/h/tuqN3ypDnoeKwpKaFeio6N0XUPuR0jdx1Ex8Z4yyUyl2JVVpO23uha6OJ4v/CWaPhiphNpVrkz0w3W5NtaSwuz+E6q9Ax1uoZnqGePRs+pztotUxqfjseakq5+XwrXizoR+IOvYIMM1kkw7q4jmAroaXd1ixWpH0WxSpCJ/gnt2FjSAYZuiyVNw3GijsXe9c3j9/Jd7XTLTPua1I1cIt1ACbW1lo9/82H+iDS+SYAtPkNdw6brM0cg0aYzi7IwVRyPSePR1lsq1yPsPJmHDh1j5559DI1N09/dynU7BjljQ6cOkqR+uB6s7S87KgytYliai+NAx4blm26ZaF+TutKACfWvxqe5+u4HeH9LQMakMNjCyFreOmvmiM5Oy+Ic53hMGpOafC+D0cl0/uAIYGhsmp179jE6qY5tpDEohkVqQ/ua1JXctdMN1CnZ2GSGFqL9JXCSJG3RvlPrM9SpNbVbnojULSXUyyDtB/mDo5yhsWnSfhBTiUQWRjEsUhva16Su5BLQIM34scY4Sz06OUtrNqH2nSSJ4oT6B5+C+2+rTUFmj0KqozbLEpG61tQJtTHmDcaYh4wxjxhjPrhSy0l6Lv3drSXD+rtbSXruSi1SZFkphkVqQ/ua1JX0sfzLx58ejrEg1RubSpMyURKdcVIkwtnCyCd/CPv+rTYF0RlqEclq2oTaGOMCnwbeCJwFbDPGnLUSy+ppT3LdjsH8QVLumriedt16RRqDYlikNrSvSV0paiJ9eGwsxoJU7/BkJn+GOjBJPDvncoljT9emILqGWkSymrlTspcDj1hrHwUwxtwKbAEOLPeCHMdwxoZObr/8bPXaKg1JMSxSG9rXpJ7Y2WPkIu/I+OFYy1Ktw5OzJddQJ+Ym1BOHwFpK76e1AnSGWkSymvYMNXAy8GTR+6HssBLGmF3GmH3GmH0jIyOLXpjjGHo7U5zc3UZvZ0oHR7KilituiymGZaWtRNw2Iu1rjadZY3fi6JH866NFr2tlaGyK7/yy+u/ztn1Pkn7kHrpN1FTdn9vkGyAzDbdeBA99ZTmLWspamD1W99dQN2vcitSbZk6oyx2h2HkDrN1trR201g729vbWoFgiS6e4lUakuJVG1ayxOzZeaOY9dWy85sv/5Dd/ybuu/z/MVtEp37MTs/zTF77J/ztyFVd4dwAw7bTNb/IN8NBd8LPPL3dxC/wZCNJ85bGA4WMzK7ecJWrWuBWpN82cUA8BpxS97weeiqksIiIistrNHIGHvxF3KSL338bA44Uesacnj9a8CJc8eDl/ba7j4UPH72H8sWcnydyygw971wPwCudBfOtw0JyMN/cMdc7Be+Fjm+Cx7y5voQMffngNAD94KqQt2cxXT4pINZo5of4RcLox5rnGmCRwIXBnzGUSERGR1ep7H4ebfh+efRjCMJ4yWAv+LHztzwGYogWA94/+FQztX/DsvnHgEE8enlrw52YPP8mLgl+wzfsWIz8p0zx7eoxDj/6UqS+9n0f/eQsn/eqr/K57X370o/Ykxv0kXljhHu5HfxV1UPaNv4B/OAN+9WN47J5onJ3XYLHKQh+DB+6Eu/8KgPY13XSklFCLrHZNWwtYa31jzJXA1wAX+Iy19hcxF0tERESa3EwmYNYPWduaKB3xwJej508NwoYXwjkfZNy287ffO8Lbf/tlvPL5vZBsA8Bay0OPPc5U4PLSF5yCqdDJ1uf3DzE2meZdv/3c+dOkpyDRCsMPQMua6LrikQfhR/+an6SNQpPloZvew8beXoJXf5BUZhye/1qYnYC2dWAccEpv7/bwoWP86Z57OPvUNq55z5uO/6WkpwhwOHToKY4dHmbom5/mddlR/T/6n1ztbuLVM//JUWcNJz18Mw4hfccO4Fmf15n5TcIfsKdyLEP5Jt/Fnvpx9DgmdxQAACAASURBVPzv54I/DWecG63TwCuhayBKkk87G575OTzvNVHi/ZyXwMgvo168M5PRHw3JdvjPv4FM4c+DjevXHX/ZIrIqNG1CDWCtvQu4K+5yiIiIyOpw73+P8v/c9lOOTGd45fPWYYyhuy3BxvSTvH/0kcKEh34On9tOOx5/a0Oe/Vw3AWM83baJzvQwz5p1bMr8kmmb5N7O19DjTBA6KTr8w0wle1g78xRjiY38+thjPG3X8cR3J3Bb2nGCNKngGBm3nfXTjzKZWE9X+mkyJkXCVmgendU//RAcfAjvxjcDMGHaSeAzaTrxnSRgGXV6WWvH+SWnsj7zNPel/huegfs//ALaPLBhAIlWOtIjTCT72Dj7GM8mTuKUzKOMmPWsD0fpwuMME5Xlvekr+GTy07Teez795tn5hcr+R5CxLgkTcNS2ssZM80A4wNET5NI+Lh7ZZNyfjp4fyh4WPn5PfroQg4MlMB6u9fFNAs9mCHBxqXx992nr67tTMhGpDWMX2+ylCRljRoAnqph0PVCm1m96Wu/5nrXWvqGWhZlrAXGb08jbUWVfHo0Ytzn19D3Wg9X0fcQet3Dc2G22baH1WT6xx+4S6txaabZ4q0a9r3PscdsolFAvgjFmn7V2MO5y1JrWuzk08vqo7KLvsZS+j/rRbNtC6yO1tBq3z2pc52bVzJ2SiYiIiIiIiKwYJdQiIiIiIiIii6CEenF2x12AmGi9m0Mjr4/KLvoeS+n7qB/Nti20PlJLq3H7rMZ1bkq6hlpERERERERkEXSGWkRERERERGQRlFCLiIiIiIiILIISahEREREREZFFUEItIiIiIiIisghKqEVEREREREQWQQm1iIiIiIiIyCIooRYRERERERFZBCXUIiIiIiIiIoughFpERERERERkEZRQi4iIiIiIiCyCEmoRERERERGRRVBCLSIiIiIiIrIISqhFREREREREFkEJtYiIiIiIiMgiKKEWERERERERWQQl1EXe8IY3WEAPPRbyiJ3iVo9FPGKnuNVjEY+6oNjVYxGP2Clu9VjEQ6qkhLrIs88+G3cRRBZMcSuNSHErjUqxK41IcSuycpRQi4iIiIiIiCyCEmoRERERERGRRVBCLSIiIiIiIrIISqhFREREREREFkEJtYiIiIiISA1Ya/nnux/mvx4djbsosky8uAuwGMaYzwBvBoattS8sM94A/wScC0wB77TW/nipyw2DgHBiBNcxgIFgBsIA3AQYF/wZSHVCZhpCHxwPkh0wMx69djywAdgQvBYI0mBDgkQn+LOYIE3gJBhlDa4xdNkjuDaDMQ7GmOizTgLjJghbuhmd8kn7AUnPpac9ieOYKlckhKkR8NPgJaGtFxyHMLSMTqZJ+wHGGFwDjuMsbN7zFlWY54LLuYzzEPBnZnBnRqLY9FoAC/4sOC44CTAGggwYJ4rRMIjGuQmwFmxIGPr4Jsmku5bO8CgmSBM6CXBcHH+awEkyk+ymsyVVuo0qxNxyCoOAYGIEE8xi3RRuRy+O6y7rMhatBuu/moShZXw6jYOlwx/HDWbASwEGsBBmojrZZl87LiQ7IT2RrZtdcJNRvCdaov3AhtEj0Ra9z01nso8wkx2WyC9mItFNMnMUz2YITIIJt4sp3+IZSzfRcGMcQuPg2JDQOPihYcqL9h+PIF9G43jYRCtBaCEzReimmE50MZOxBDY6AGtLOnSFRwj9WXyTYMyswXFc1rUmGZvOlK0jc/uFE6QJnCTjZg3TviXpOvR2pPC8ynE4t+7tbk1UXM5Ctt2qrM8DHyaeiWLOTUDHRnCjQzDfDxmemCUThCRch77sdvF9n3BiBCdMEzpJnI5ePK/0sC2TCRiemMUPLZ5j6OtIkUhUX++V/C44HkFLL15LS/Wfr6KMlWRm07jTw5gwg3USBK19JFLJmpV9KfzZGdzpomW39uKlarNsaXyhhY9945f8ye/+Gq94Xk/cxZFl0JAJNfBZ4FPAngrj3wicnn28Argm+7xoYRBghw/g3b8XXnYJTByCOy6H8YPQNQBbroaD98Lpr4e9OwrDt94Az/wM7vlfcN7u6CDuux+D11wFbgp+cgvei86HvZfA+EHcrgHazt+LE86QvL1oPm/9FPzXv8Ar3wPJdsLEGB/66hhfOzBCf3cr1+0Y5IwNnSc+MAlDGD4At24rzPvCWwh7z+Sh4Ul27tnH0Ng0/d2tfPT8F3P9Dx7jT15/RnXznrcoy0OHjpXMs+pyLuM8JHvgcfhBzN5LoKMPXvfh0vg9bze0dsPdfw2veDfceWVRDO+JEvCbt+KOH8Q9400kXvMBnGzM5uP/7g/jTQzjXHATT7c+n5O62qJtVCHm6Dtr2ZLKMAgIDx0gsfei/DL8rTfDhrPiT6prsP6rSRhaHh+dxDUh/ZkncD93URTTb/z7aIL0JPzwGnjVFfCly6Lv/Iw3wWs+kK9n6RqAC/bAz78IL704+tPzizvhua+G33xXaR3+tmsh0Qq3XVo67N5P0/GaP8M8cS987c9wuwbovOAmrvuJw7s2pUl9uTC9k62/nVe8G+/hb5J80fk43/lfpWXsGsBsuRov2Q5fuQomhjFbb+KQcyr/9/X76e1IsOfNHZjbd0T7YdcAibdcz59/3+ePXncGn7z7l3z9wHBJHYkNS/YLt2uA5Jbrefddk4xMZLh2+2Y2begsm1SXq3uv3b657HJUn59A4MOhn5fG39YbYMML8a3Dg4eOcdmN+0u+5zN62zDPPkhyTp3m952ZT1gzmYAHhyd4T9Fnr9m+mU19HVUl1SW/C9lluFtvwF+3qarE1Pd9GH7guGWsJDObxhs9kF+26RrAbL2BTM9ZVSXVSy37UvizM7ijZZbds0lJtVTF2ugWz6aJq73VpiGP5qy19wCHjzPJFmCPjfwQ6DLGnLSUZQYTI9GB20svhvHHC8kIRM93XA4v3lo4EMsN33sJPO810evbd8H0KLxkWzSd40Xzy1XK2c+snXmSztvnzOfOK6PP3XE5TD2Ld+Qxdm1eA8DQ2DQ79+xjdDJ94hWZGikc2Ofmfes2gomR/EFObp5/9oX7OX/zKdXPe47RyfS8eS50XssxDwF3ZqTw43/2++bH7+274MjBKMZyyXRu3N4d0Rm63LCXbCsk07lp7rg8mu/4QRK3XUxHMFbYRhVijqmRZVu/YGIEL3dQl12Gt/cigonlW8ai1WD9V5PRyTRPjE7R505GdXIupqeejR53XB7FcS5RhWydOydmb9sR1b9HDkbJ9PhBeNUfza/Dv3RZNN+5w16yDbN3O2x6Q3544raLueI3O+jJJdO56XP1951XwksvjvafuWXM7UdTzxb2pb0Xs8E9xtDYNB86p3fe70LPly9l1+Y1XHbjfs7ffApQWkeW2y+67riUD53Ty9DYNJfduJ/hidmK3/PcurfSchay7VZlfT7xzPz423sJTDzD8MRsPpmGwvccTj1btk4Li+q04YnZfDKd++x7jrNN5yr5Xcguw+y9JDrrW4WwQr0bVlHvutPD5Zc9PVyTsi+FO11h2dOq02VhlE83j4ZMqKtwMvBk0fuh7LB5jDG7jDH7jDH7RkYqV4YmmI0qT8eNmgTmKtKc8YNRc8Fyw8Og8DrRFp0JHD8Y/TXluPM/U2n+uc8l2iDRRl9bYVccGpsm7QcVy5/np8vO2wTp/I9y8Ty7WhPVz3uOtB+UnedC5rUc82hG1cZtXugXtnsujorNjc2544r/Rq00TWt3/nXKBIVtVCHm8JfvIDq/f85Zhgnq4EC9BuvfKBYct2Wk/YC2pItn06Uxna0XS+rKnEoxO7c+L1cf5/aNucNy87RhyfAWp8LvQG763DJOtB9m36dMtB/1tZmy0/e1mXxdnZOrIyvtF7nfjqGxafwgpJxKdW+55VSrkevzJcVukCm/rYMMmSAs+524Yfl6wwkz+bd+aMt+1g9tdeUq/l0oLlfoV/Vxp4oyVmLC8t+JqeKzwJLLviRxLnuBlqPOleVX5R4qDaRZE+pyf/qUjV9r7W5r7aC1drC3t7fiDK2bippphQFkpqLXxboGomtPyw133MLrzBRMj0WvrY3mN/czleaf+1xmCjJTDE8VVqm/u5WkV0XTVi9Zdt7WTdLf3VoyuL+7lfHpTPXzniPpuWXnuZB5Lcc8mlG1cZvneIXtnoujYnNjc+44W7T7VJpmeiz/eta6hW1UIebwqr9W7kTy++ecZVh3+ZaxaDVY/0ax4LgtI+m5TKUDfJMsjelsvVhSV+ZUitm59Xm5+ji3b8wdlpuncUqGz4QVfgdy0+eWcaL9MPt+1kb70fCULTv98JTN19U5uTqy0n6R++3o727Fc8sfBlSqe8stp1qNXJ8vKXbdRPlt7SZIuE7Z7yRwytcboVP4Q8NzTNnPetU2ny/+XSgul1Pd1YBhFWWsxDrlvxNbxWeBJZd9SeJc9gItR50ryy93SKUm382jWRPqIeCUovf9wFNLmaHb0Uvwjpvhvpug67TomtFchZq7hvT+vdH1psXDt94Aj36n6DrVHvjJLdF0oR/Nb+sNJZ850nIKx86bM5+3fir63JaroW09/trnsnv/0Wjlsteh9bRXcYDe1htdv1k87wtvwe3o5bodg/kf59w11F/Y/2T1856jpz05b54LnddyzEMgaOnF5uLs+5+YH7/n7Ya1A1GMvfVTc2J4T9QhXm7YT24hnBOzbLk6mm/XAJkLbmLC7S5sowoxR9vy/bi7Hb3RNdNFy/C33ozbUQcHEDVY/9Wkpz3JqT1tDAftUZ2ci+m29dFjy9VRHL/t2pKYnVvPcsGeqP5dOwBvvy4adu8/z6/D33ZtNN+5w35yC3brjfDgV/PDMxfcxKd/NMHoW64vX3+/9VNw303R/jO3jLn9qG19YV/aehOHgk76u1v5yLdH5v0ujL7lenbvP8q12zfzhf1Ro6ziOrLcfjG+5Xo+8u2R/LW6fR2pit/z3Lq30nIWsu1WZX3esXF+/G29ATo20teR4trtm+d9z07b+rJ1mlNUp/V1pLhmzmevOc42navkdyG7DLv1BoKW6uomp0K961RR7watfeWX3dpXk7IvRdBaYdmtqtOlOpbcNdTKqJuFsbYxGx4YY04D/neFXr7fBFxJ1Mv3K4BPWmtffqJ5Dg4O2n379lUcv+K9fIdpApPgMGtw1Mv3ss5jBcVekBPFbc7y9PId4JtEhV6+ZwicRB308p3Gukn18n18DRO35ZTv5TvXGVA99vLt4tiA0Lj4IVX08j1N6Car6OV7LY7jLLiX7xnfkmjMXr5jj1tYZOxW0cu3H4R4ZXv5zhA6iTrv5btyGStZZb18xx67S6lzZXnNZAI2/X9f5arfO4MrfucFcRfneGKP20ZRf+1TqmCMuQU4B1hvjBkC/hJIAFhrrwXuIkqmHyG6bdYfLMdyHdfFWbtx4R9s6z7u6OKN4AKFJbTNmzZfFqC3c5HJguNAx4Yygw29ndX9s139opY+z5Uo12rktbRAyyknnvA43Owj2hrz49PLj5ujQswtp0Xvn7VQg/VfTRzHsK49F2kL2OYnqIsXqhOAdiDaL9ZlHxQNh0JTMIdoH4kOu+fvP4bS34MksHbeVC35/bB4zSvVkcX7hQtUd/4v+9kyda/q80VyPVjbX3aU5zk8p6u1zHAPuo7fn2oi4XJyd+VjhROZ+7uw0IPCaspYSSKVhFT0nRgW3mRyqWVfCi/VAql4li3NQyeom0dD1gHW2m0nGG+BK2pUHBERERERkRPKX0OtE8BNo1mvoRYREREREalLOkPdPJRQi4iIiIiI1IDVjbOajhJqERERERGRGig0+ZZmoYRaRERERESkBnLnp9Xku3kooRYREREREamB3C2L1SlZ81BCLSIiIiIiUgM6Q918lFCLiIiIiIiILIISahERERERkRqw6uS76SihFhERERERqYVcL99q8900lFCLiIiIiIjUQO4+1Eqnm4cSahERERERkRrI34daGXXTUEItIiIiIiJSA/levmMthSwnJdQiIiIiIiI1kL8PtU5RNw0l1CIiIiIiIjWkfLp5KKEWERERERGpAd01q/kooRYREREREamBfKdk8RZDlpESahERERERkRqwqJvvZtOwCbUx5g3GmIeMMY8YYz5YZvyAMeZbxpj7jDH3G2POjaOcIiIiIiIiQL7Nt9Lp5tGQCbUxxgU+DbwROAvYZow5a85kHwL2WmtfClwIXF3bUoqIiIiIiBTkb5uljLppNGRCDbwceMRa+6i1Ng3cCmyZM40F1mRfrwWeqmH5REREREREyjI6R900GjWhPhl4suj9UHZYsQ8D240xQ8BdwB+Vm5ExZpcxZp8xZt/IyMhKlFVk2SlupREpbqVRKXalESlu65NVN99Np1ET6nJ/6cwNz23AZ621/cC5wA3GmHnra63dba0dtNYO9vb2rkBRRZaf4lYakeJWGpViVxqR4rY+5TolU5Pv5tGoCfUQcErR+37mN+n+Q2AvgLX2XqAFWF+T0omIiIiIiMyh22Y1n0ZNqH8EnG6Mea4xJknU6didc6Y5CLwOwBhzJlFCrfYuIiIiIiISC3VK1nwaMqG21vrAlcDXgAeIevP+hTHmr40xb81O9qfATmPMT4FbgHdaq6sWREREREQkHrl0RJ2SNQ8v7gIslrX2LqLOxoqH/UXR6wPA2bUul4iIiIiISDn503vKp5tGQ56hFhERERERaVTKp5uHEmoRERERERGRRVBCLSIiIiIiUgP5Xr7VK1nTUEItIiIiIiJSA/n7UMdcDlk+SqhFRERERERqoHCGOt5yyPJRQi0iIiIiIlIDug9181FCLSIiIiIiUkO6D3XzUEItIiIiIiJSAzZ/I2ppFkqoRUREREREakBNvpuPEmoREREREZEa0Anq5qOEWkREREREpCayt83SKeqmEWtCbYxpN8Y42de/Zox5qzEmEWeZREREREREVkL+tlnxFkOWUdxnqO8BWowxJwN3A38AfDbWEomIiIiIiKwAXUPdfOJOqI21dgp4O/DP1trzgLNiLpOIiIiIiMiK0W2zmkfsCbUx5lXAxcB/ZId5MZZHRERERERkRahTsuYTd0L9PuB/ALdba39hjHke8K2YyyQiIiIiIrLsbL5TspgLIssm1rPB1trvAN8pev8o8N74SiQiIiIiIrIy1ClZ84kloTbGfMJa+z5jzJcpXJufZ619awzFEhERERERWTH5hFoZddOI6wz1Ddnnf1jMh40xbwD+CXCBf7XW/l2ZabYCHyZK2H9qrb1ocUUVERERERFZOlvo5zvWcsjyiSWhttbuzz5/50TTzmWMcYFPA68HhoAfGWPutNYeKJrmdKJrs8+21o4ZY/qWp+QiIiIiIiJLozPUzSOuJt8/o0xT7xxr7YuP8/GXA49kr7fGGHMrsAU4UDTNTuDT1tqx7PyGl1xoERERERGRJVAv380nribfb84+G6LbZZ27gM+eDDxZ9H4IeMWcaX4NwBjzfaJm4R+21n613MyMMbuAXQADAwMLKIZIfBS30ogUt9KoFLvSiBS39U0nqJtHLLfNstY+kX08DswWvX/CWvvECT5eLv7m/tfjAacD5wDbgH81xnRVKMtua+2gtXawt7d3YSsiEhPFrTQixa00KsWuNCLFbX0qdEqmlLpZxH0f6sUYAk4pet8PPFVmmjustRlr7WPAQ0QJtoiIiIiISCzy96GOuRyyfOK6hvplRW9bjTEvpSiurLU/Ps7HfwScbox5LvAr4EJgbg/eXyI6M/1ZY8x6oibgjy5H2UVERERERBZDt81qPnFdQ/2xotfPAP9Y9N4Cr630QWutb4y5Evga0fXRn7HW/sIY89fAPmvtndlx/5cx5gAQAFdZa0eXeyVERERERESqlb9plhLqphHXbbN+p5rpjDGvt9Z+o8zn7wLumjPsL4peW+D92YeIiIiIiEjdMGr03TTq/Rrqj8ZdABERERERkeVgdd+splPvCbX+uhERERERkaaQT6eV5TSNek+o9ReOiIiIiIg0hXynZPEWQ5ZRvSfUIiIiIiIiTSJ72yz1StY0Yk2ojTGpEwx7vHalERERERERWTk6Q9184j5Dfe/xhllr317DsoiIiIiIiKw4naBuHrHcNssYsxE4GWg1xryUwp80a4C2OMokIiIiIiKyktRBVPOJJaEGfg94J9AP/GPR8GPAn8dRIBERERERkZVUaPKtU9TNIpaE2lp7PXC9MeZ8a+0X4iiDiIiIiIhILeXuQ60m380jribf2621NwKnGWPeP3e8tfYfy3xMRERERESkYeWafCufbh5xNfluzz53xLR8ERERERGRmrLKqJtOXE2+/yX7/FdxLF9ERERERKTWbO4+1Mqom0ZcTb4/ebzx1tr31qosIiIiIiIitaRrqJtHXPeh3p99tAAvAx7OPl4CBDGVSUREREREZOXovllNJ85evjHGvBP4HWttJvv+WuDrcZRJRERERERkJekS6uYT1xnqnOcAnUXvO7LDREREREREmkr+PtRq89004urlO+f/b+/ew+So63yPvz/Tc8kNSQgBVO6I64l4g4C4qKBBRHcfEBdDIoJ6eGRdZT2e1fXgurrouq7gs+vlgIus1yASQeUhR1BECOiyCgS5hmtAhIiQcEnIZZKZnv6eP6p6pmemezLT09PVXfN5PU8/Xf2r26+qv1Vd366qX30RuF3SqvTz0cA52VXHzMzMzMxsagw2SuZ8OjcyPUMdEd8BXgtckb5eV74cfGckHS/pAUlrJZ09xnAnSwpJixpTazMzMzMzs4kbPEOdbTWsgTJNqJVc63As8KqIuBLolnTEOMYrABcAbwMWAsskLawy3C7AR4CbG1pxMzMzMzOzOvkMdX5kfQ/114HXAcvSz5tJEuWdOQJYGxGPREQfsAI4scpw/wycB2xvQF3NzMzMzMzq5ka+8yfrhPq1EfFh0oQ3Ip4Duscx3ouBxys+r0vLBkl6DbBPRPx0rAlJOlPSakmrN2zYMKHKm2XFcWvtyHFr7cqxa+3IcduaItzOd95knVD3p5dvB4CkBUBpHONVi8DBP3wkdQBfBj62swlFxEURsSgiFi1YsGB8tTbLmOPW2pHj1tqVY9fakeO2NQ2m086ncyPrhPprJI2R7SHpX4D/Ar4wjvHWAftUfN4beKLi8y7AIcANkh4FjgRWumEyMzMzMzPLjBsly51MH5sVEZdIug1YTBJX74iI+8Yx6q3AwZIOAP4ILAXeXTHdTcDu5c+SbgA+HhGrG1h9MzMzMzOzcRt6bJZT6rzILKFOL8u+KyIOAe6fyLgRUZR0FnANUAC+HRFrJH0OWB0RKxtfYzMzMzMzs/r5sVn5k1lCHRElSXdK2jciHqtj/KuBq0eUfabGsMfUV0szMzMzM7PG8gnq/Mj0km/ghcAaSbcAW8uFEXFCdlUyMzMzMzNrvPBzs3Ink4Ra0kuAPYHPjuh1NMk90WZmZmZmZrky9NAsn6LOi6zOUH8F+IeIuKuyUNJW4J+Ab2VSKzMzMzMzsylSfg61L/nOj6wem7X/yGQaIG2Fe//mV8fMzMzMzGxq+Yrv/MkqoZ4xRr+ZTauFmZmZmZlZkwy28u0z1LmRVUJ9q6QPjCyUdAZwWwb1MTMzMzMzawrfQ50fWd1D/VHgCkmnMpRALwK6gZMyqpOZmZmZmdkU8kXfeZNJQh0RTwF/LulNwCFp8VURcX0W9TEzMzMzM5tqvuQ7fzJ9DnVErAJWZVkHMzMzMzOzZhh8bJYT6tzI6h5qMzMzMzOzaWXwDLXvoc4NJ9RmZmZmZmZNEPg51HnjhNrMzMzMzKwJhs5QW144oTYzMzMzM2sin6HODyfUZmZmZmZmTeCHZuWPE2ozMzMzM7MmiBhs5zvTeljjOKE2MzMzMzNrIl/ynR9OqM3MzMzMzJrAjZLlT1sm1JKOl/SApLWSzq7S/+8k3SvpLknXSdovi3qamZmZmZmVDT02yyl1XrRdQi2pAFwAvA1YCCyTtHDEYLcDiyLilcCPgPOaW0szMzMzM7PqnE7nR9sl1MARwNqIeCQi+oAVwImVA0TEqojYln78LbB3k+toZmZmZmY2TLiZ79xpx4T6xcDjFZ/XpWW1nAH8bEprZGZmZmZmthOD91D7FHVutGNCXS38qv7XI+k9wCLgSzUnJp0pabWk1Rs2bGhQFc2mluPW2pHj1tqVY9fakeO2NQ09NMsZdV60Y0K9Dtin4vPewBMjB5J0LPAp4ISI2FFrYhFxUUQsiohFCxYsaHhlzaaC49bakePW2pVj19qR47Y1lZ9D7TPU+dGOCfWtwMGSDpDUDSwFVlYOIOk1wDdIkun1GdTRzMzMzMxsGN9CnT9tl1BHRBE4C7gGuA+4LCLWSPqcpBPSwb4EzAEul3SHpJU1JmdmZmZmZtYcvoc6dzqzrkA9IuJq4OoRZZ+p6D626ZUyMzMzMzMbBz+HOj/a7gy1mZmZmZlZOwpf9J07TqjNzMzMzMyaYPCxWdlWwxrICbWZmZmZmVkTDD42yxl1bjihNjMzMzMza4KhM9TOqPPCCbWZmZmZmWXm4Q1beHZrX9bVaIoHn9oM+Ax1njihNjMzMzOzzCz+txs57su/yroaTfHd/34U8D3UeeKE2szMzMzMMvX0lh1ZV8GsLk6ozczMzMzMplhExSOzfIo6N5xQm5mZmZmZTbFiaSihdqNk+eGE2szMzMzMbIoVByoSaufTueGE2szMzMzMMjHsMuicK5ZKg93Op/PDCbWZmZmZmWVioDSNEuphZ6idUueFE2ozMzMzM8tEcRol1P0+Q51LTqjNzMzMzCwT/QOlnQ+UE5VnqC0/nFCbmZmZmVkmplOS6UbJ8skJtZmZmZmZZWI6XfI9vFEyZ9R54YTazMzMzMwyUZlk5t2wPw+cT+eGE2ozMzMzM8tE5WXQpZyfrd7ePzDY7Uu+86Mz6wrUS9LxwFeBAvDNG9O+jwAAGGFJREFUiPjiiP49wHLgMOAZ4JSIeLQhMy/uIPq2gTpgx/PJFhEBUUq6O7qBQMXt0DULSv0wUISOQjJOlKCzBwiYtSCZ5rYNUCpBDCTT6uxO+nVk859HqRQ8s7WPvuIA3Z0F5s/upqNjGmz5pVLyXRT7Mv8OGq3Yv53C1g1QKkJHJ8zYDfo3w0B/WlaAji4odEJxByp0Q3HH0PCVMdvRUXVdlVDtuMnxurUmqhZHkJbtAAQS0TkT+rcNxW/PLmjH80ShuyLmk7hW/zYo9EAMEAP9Q9sCQEdnctDTvx1KRaK8jfT3EoVutvfMZ2b/JjTQB10zk+n29ybTKHTDrN2T4ceqf7O2g8nOexzjT9vfjjZVLG6nsGXod2FgzgI6O2c0aeb9sOXJoW1xzl7Q2TX+0Xdsp9BbUfeZC+jsaVbd+2DLUxV13zPZJqwulY2SfezyO/nyKa/OsDZT64Tzbxrs9p4xP9oyoZZUAC4A3gKsA26VtDIi7q0Y7AzguYh4iaSlwLnAKZOeeXEH0fsc6ijAxj/CjefBa/8aVp4FGx+DufvCiV+H7tmw+lvwyqVw5YeG+p1wPtz8DXjj38M9P4HD3gvF7bDqC6Ons/RS2GNh05OOUil44KnNfGD5atY918ve82byn6cv4s/23CXfB0alEqy/F1Ysy/w7aLRi/3YKG+5Hl502tGxLLoYZu8LyE4bH7uwFSQKwYwv84F1D/d61PInZV50CC14GG+4ftq5i6aX8obAfp3371tFxQ+R23VoTVdtG33NFsg+tLDvlEhQBl71neLw//TDa/UC47PSK8uXw0LWw/xvgijPRyP34C14Ezz8J6bajcr/rzkFb1jNzycXoxvNg61Ow+Jzh+/sTvw6zN8L8lyTbVJb7mMnOexzjT9vfjjZVLG6nsH7470JhycUU93jZ1CfVxX5Yv2ZwuxrcRvd4+biS6uKO7RSeqVL3+S+b+qS62JdsC6PqvtBJdZ0qL4O+4vY/5jqhtnxq1yPZI4C1EfFIRPQBK4ATRwxzIvC9tPtHwGI14gnqW9YnZyL6e5Od6auXDSXBkLxf+SHY9jS87m+HDq7K/VaelYxz+enwmlPhuUeSA5Rq01mxLDkb0GTPbO0bPCACWPdcLx9YvppntvY1vS5NtW3D0MEiZPodNFph64ahAw9I3i87LbliYmTsbnosOXje9NjwfuWYXbEsOaswYl1pxTK2PPtk9bjJ8bq1JqoWR+V9aGXZlieHkuly2WWnwb6HDyXTg+WnwyuXwBVnVt+PD/QNHThX9jvqo0ncl38Hjvro6P19eXva8mTt+jdrO5jsvMcx/rT97WhThS2jfxd02WnJGeuptuXJ0dvVZacNbSs7UeitUffeZtT9qRp1f2rq551T06WV74jhy9mItMRaQ7sm1C8GHq/4vC4tqzpMRBSBTcD8kROSdKak1ZJWb9gwjh1xqTj02vgYzJw3tFMt2/hYcql3R6F6v/I4HYVkuLGmU2z+gUhfcWDwgKhs3XO99BUHaoyRE8W+lvkOdqauuK22bDEwuqxrVnLrQtes0f3KMT3QX3V6c7uHNywyGDdttG5t6kw4bkeqFkflfejOyjY+NsZ2UKq9H681zsx5w7vH+i0Y6K9d/2ZtB5Od9zjGz/Nvx6RjtxXViu1SsfXn3c51b6J2idvp0ihZ/4g/DpxO50e7JtTVYnDk31vjGYaIuCgiFkXEogULFux8zh2dQ6+5+0Lvc8l7pbn7pvfuDVTvVx6nNJAMN9Z0Mrh8qLuzwN7zZg4r23veTLo7C02vS1N1drfMd7AzdcVttWVTYXRZ/7bkPv7+baP7lWO60FV1ehv7hu9SBuOmjdatTZ0Jx+1I1eKovA/dWdncfcfYDjpq78drjdP73PDusX4LCl2169+s7WCy8x7H+Hn+7Zh07LaiWrHd0YS7ASc773auexO1S9yOTDQHctow2cZtw//A9Anq/GjXhHodsE/F572BJ2oNI6kT2BV4dtJznrNH0qhN18zknpk7Lk3uiy7vXMv3zc3aHX7zf5Puyn4nnJ+M867lcPslMO/A5D60atNZeulQgztNNH92N/95+qLBA6PyfXDzZ+c8+Zm1IFnnLfAdNNrA7AXEkouHL9uSi5MzcyNjd9d9k3sid913eL9yzC69NGk8ZsS6iqWXMme3varHTY7XrTVRtTgq70Mry+bsBUu+PzreH7s1uWd6WPlyuOsyOOmi6vvxQncy7sh+N30lifvy78BNXxm9vy9vT3P2ql3/Zm0Hk533OMaftr8dbWpgzujfhVhyMQNzmhCPc/YavV0tuXhoW9mJgZk16j6zGXXfs0bd95z6eedUsaJRMoBPX3lPRjWZWkd84bphn/0c6vzQyOv520GaID8ILAb+CNwKvDsi1lQM82HgFRHxwbRRsndGxJKxprto0aJYvXr1zitQTyvfpWJyNtCtfLe2ibeCm/lKGW/cupVvq9A2cTuKW/muX/u38p153MIkYrcFuZXvOk28le/MY7eV4/amtU9z6jdvHlb26Bf/IqPaTJ39z75q2Of7//l4ZnS19BU8mcdtu2i961PGISKKks4CriF5bNa3I2KNpM8BqyNiJfAt4GJJa0nOTC9tWAU6e1BnT9I9c9fGTLPF/tns6BALdunJuhrN19HRct9Fo3R2zYC5+wwvnDG7/glWWVcdUDtucrxurYlqxdGIsuQoYESzGTPn7vToYCL9BcwCmDGBg/gst4PJznsc40/b34421dk5/HehqQeFnV2jf5MmMnrPDOjJqu7dk6q7DVescon3g09t5qV77pJBbabGpbc8tvOBrG21ZUINEBFXA1ePKPtMRfd24F3NrpeZmZmZmY3PyEu+AY778q946F/eRleh/a9ke3LTdj75k7uzroZNofaPUjMzMzMza0sjGyUrO/q8VWzvb++nBDz41GaO/NfrqvZzo2T50bZnqM3MzMzMrL3VemzWE5u287JP/5wPHXMQ7zlyP140d2bV4VpNX7HE3X/cxH/c8DC/vK/288ndKFl+OKE2MzMzM7PxGyhCcXvSGGRxOwzsGOoe9r4DBvqSxiAH+pJXqTjUPdDPwU88y8c71zNABwNRoEgHJTooUqBEB8//upOv/rqT/ihQ7OjmwD3nsmDuLszbZQ6zZ81ixowZzOiZQdeMmXR3z6Czu4dCVw8dXd10ds5AXd10FLoodHQws7tQ88xwRPLIrt7+AUoRg58Hiv2U+rZT7Oulf0cvxb5tbO/tpbd3G729W3l+8xY2Pv88Tz6ziac3Ps8M9dFDPwfRx8JCPz3qYwb99NDHDPXTRZESonPlyqQBS3Ukr47OpHHL8quz3D2r4n3GUHehO3l19iSPZyz0JN0dLd3QWS45oTYzMzMzm24e/AVc/bH0KTOVr4Hhn2v1b5CDVeDAAnRpnJd3P5O+JmhHdNJHF/0UGKCDDoIOAg17LzEL6KCECDoZoKAJPBGpSmPvfVFgB91sp4sddNMfhWR+jzycrsdI1mmpNPRnBJN4CpM6kuS60JUm1xpK2lXRnT4Vgzf+PRz23vrnZ+352KypImkD8IdxDLo78PQUV6cVeblHezoijm9mZUaaQNyWtfP36Lo3RjvGbVkrrcdWMJ3WR+ZxC2PGbt6+Cy9P42Qeu5PY5zZL3uJtPFp9mTOP23bhhLoOklZHxKKs69FsXu58aOflcd3N63E4r4/WkbfvwstjzTQdv5/puMx55Va+zczMzMzMzOrghNrMzMzMzMysDk6o63NR1hXIiJc7H9p5eVx383oczuujdeTtu/DyWDNNx+9nOi5zLvkeajMzMzMzM7M6+Ay1mZmZmZmZWR2cUJuZmZmZmZnVwQn1BEg6XtIDktZKOjvr+tRL0qOS7pZ0h6TVadlukq6V9FD6Pi8tl6Svpct8l6RDK6bz3nT4hyS9t6L8sHT6a9Nx1fylHKzLtyWtl3RPRdmUL2uteWSpVeJX0j6SVkm6T9IaSf8rLW+bGJRUkHS7pJ+mnw+QdHNajx9K6k7Le9LPa9P++1dM45Np+QOS3lpR3hLfUyuajuumUftrmzrtGpeN3Be3kkbsn6252nUbqpeqHJtam4sIv8bxAgrAw8CBQDdwJ7Aw63rVuSyPAruPKDsPODvtPhs4N+1+O/AzQMCRwM1p+W7AI+n7vLR7XtrvFuB16Tg/A96W4bK+ETgUuKeZy1prHo7fAHghcGjavQvwILCwnWIQ+DvgB8BP08+XAUvT7guBv0m7PwRcmHYvBX6Ydi9Mv4Me4ID0uym00vfUaq/pum5owP7aryn9fto2Lhu1L26112T3z341/ftq221oEss86tjUr/Z++Qz1+B0BrI2IRyKiD1gBnJhxnRrpROB7aff3gHdUlC+PxG+BuZJeCLwVuDYino2I54BrgePTfi+IiN9EstdYXjGtpouIXwHPjihuxrLWmkdWWiZ+I+JPEfG7tHszcB/wYtokBiXtDfwF8M30s4A3Az+qUffyMv0IWJwOfyKwIiJ2RMTvgbUk31HLfE8tyOtmyES3FZs6bRuXDdwXt4wG7Z+tudp2G6pXjWNTa2NOqMfvxcDjFZ/XpWXtKIBfSLpN0plp2Z4R8SdIfmSBPdLyWss9Vvm6KuWtpBnLWmseWWnJ+E0vsXsNcDPtE4NfAT4BlNLP84GNEVGsMr/BOqb9N6XDT3SZbPqum0bsr23q5GKdT3Jf3EoasX+25mqHuDIbU2fWFWgj1f61bNdnjh0VEU9I2gO4VtL9Ywxba7knWt4O8rysLVdXSXOAHwMfjYjnxzgx0DLfi6S/BNZHxG2SjtlJ/cbqV6u82p+crRpTzdZyMdwkjdhf29Rp+3XegH1xS2jg/tmay9+DtT2foR6/dcA+FZ/3Bp7IqC6TEhFPpO/rgStILrd5qnzpVvq+Ph281nKPVb53lfJW0oxlrTWPrLRU/ErqIjmAuyQifpIWt0MMHgWcIOlRksvS3kxyRmSupPIflJXzG6xj2n9Xksu8JrpMNk3XTYP21zZ12nqdN2hf3CoatX+25mr1uDLbKSfU43crcHDaWmQ3SQMWKzOu04RJmi1pl3I3cBxwD8mylFtJfi9wZdq9Ejg9bd3zSGBTegnYNcBxkualLYAeB1yT9tss6cj0XqTTK6bVKpqxrLXmkZWWid90XX0LuC8i/r2iV8vHYER8MiL2joj9Sdbh9RFxKrAKOLlG3cvLdHI6fKTlS9NWZg8ADiZpSK1lvqcWNO3WTQP31zZ12jYuG7gvbgkN3D9bc7XtNmQ2aLKtmk2nF0kLlw+StEb4qazrU+cyHEjSguKdwJrycpDcN3Qd8FD6vltaLuCCdJnvBhZVTOt/kjSmtBZ4f0X5IpKDvoeB8wFluLyXAn8C+kn+BT2jGctaax6O3wB4PcnlXHcBd6Svt7dbDALHMNSK7IEkCfFa4HKgJy2fkX5em/Y/sGL8T6X1e4CKVshb5Xtqxdd0WzeN3F/7NaXfU1vGZSP3xa32muz+2a+mf19tuQ1NYnlHHZtmXSe/JvcqH/ybmZmZmZmZ2QT4km8zMzMzMzOzOjihNjMzMzMzM6uDE2ozMzMzMzOzOjihNjMzMzMzM6uDE2ozMzMzMzOzOjihNjMzMzMzM6uDE+oWJukkSSHpZQ2e7nclndzIaU4FST+XtFHST7Oui03MdI5dSa+W9BtJayTdJemUrOtk9ZG0l6QVkh6WdK+kqyW9VFKvpDvSsgsldUjaX9I9E5h2p6SnJf3rVC7DOOpxjKQ/z7IO1jxTEdOSLqgYtzydO1p9X23to1bc1jmt90k6P+3+oKTTK8pftJNxb5C0qOLz4DYiaZGkr9VTJ2t/Tqhb2zLgv4ClWVckI18CTsu6ElaX6Ry724DTI+LlwPHAVyTNzbhONkGSBFwB3BARB0XEQuAfgD2BhyPi1cArgYXAO+qYxXHAA8CSdF5ZOQZwQj0NTFVMR8SH03HfXp5O+vrROOvVOdFlseljJ3FbHqZQz7Qj4sKIWJ5+fB8wZkK9k2mtjoiP1Du+tTcn1C1K0hzgKOAM0qQkPZPwK0lXVP6LnPbbIunfJP1O0nWSFoxzPo9K+mw63t3lM4qS5kj6Tlp2l6S/SsuXpWX3SDq3YjpbJJ0r6TZJv5R0RPpP3iOSTkiHKUj6kqRb02n+9Vh1i4jrgM0TX3uWpekeuxHxYEQ8lHY/AawHxrVM1lLeBPRHxIXlgoi4A3i84nMR+G/gJXVMfxnwVeAx4MhyYRrXX1BylcNqSYdKuiY9M/PBdBil8XhPGtOnpOXHqOKKHknnS3pfxXSHbS+S9gc+CPxvJWcU31DHclj7mOqYHkXSwWn83pb+Brw0Lf9+ut9fBXxB0ueVXIH0izRW35H2v0fSVXLSPZ3VituCpFWSfgDcDSDpPZJuSfdn31CaaEt6v6QHJd1IcnxCWn6OpI8ruZpiEXBJOu7MiVaycv8r6WgNXalxu6RdJrMCrPU5oW5d7wB+HhEPAs9KOjQtPwL4GPAK4CDgnWn5bOB3EXEocCPwTxOY19PpeP8BfDwt+zSwKSJeERGvBK5XcinMucCbgVcDh0sq/4s9m+Tfw8NIkuDPA28BTgI+lw5zRjrNw4HDgQ9IOmAC9bT24NhNSToC6AYensAyWWs4BLhtrAEkzQIWkx7MjVd6sLYY+ClwKUlyXenxiHgd8Gvgu8DJJEl3OR7fSRLHrwKOBb4k6YXjmPWw7SUiHgUuBL6cnlH89USWw9rOlMX0GC4CPpTuXz8JnF/R7yBgcUR8Iv18AMlZ7r8CfkDyO3IIUCK52semp7Hi9gjgUxGxUNL/AE4BjkqvmBgATk33jZ8lSaTfQnIFxjDp1RSrgVPTfWHvGPUpJ913AFfXGObjQPnKjTcAY03PcsAJdetaBqxIu1cwdMB1S0Q8EhEDJAdir0/LS8AP0+7vV5SPx0/S99uA/dPuY4ELygNExHMkicQNEbEh/Rf7EuCN6SB9wM/T7ruBGyOiP+0uT/M44PR0J3QzMB84eAL1tPbg2AXSH/GLgfdHRGkCy2St76A0Fm4CroqIn01w/L8EVkXENuDHwEkafsniyvT9buDmiNgcERuA7UpuH3g9cGlEDETEUyR/RB0+jvlW217MYPIxPUoaq0cCP06nfQHDL6m9fMS+8ep0/3w3QERcm5ZX7ovNKt0SEb9PuxcDhwG3pvG2GDgQeC1Dv/99DB1v1KucdJdvc6jmJuDfJX0EmJvGteWYL6FpQZLmk5xJO0RSAAUgSP4JixGDj/y8s/JqdqTvAwzFhKpMY6z7/Pojojx8qTzNiChVXKol4G8j4poJ1M3aiGM3HVh6AXAV8I8R8dvxjGMtZw3JmeFqyveb1msZcJSkR9PP80kua/xl+rkc16WK7vLnTmrHc5Hhf5TPGNG/2vZi08dUxnQ1IrkqotZ0t474XBn3fRXl5bi36WmsuK2MIQHfi4hPVg6QXo02keOKSYuIL0q6iiTh/q2kYyPi/mbWwZrLZ6hb08nA8ojYLyL2j4h9gN+TnJU4QtIBSu4/PYWk4SdIvsvyDufdFeX1+gVwVvmDpHkkZ+aOlrR7ejZlGcmZkfG6BvgbSV3pNF8qafYk62mtZdrHrqRukgZUlkfE5XUug2XveqBH0gfKBZIOB/abzETTP1teD+ybbiP7Ax9m9GXfY/kVcIqSe/sXkFxtcQvwB2ChpB5Ju5KcodmZzYDv75sepiSma0mvDvqTpJPSeXVIetVUzMtyrVbcHj1iuOuAkyXtkQ6zm6T9SH7/j5E0P/0Nf1eN+TRsXyjpoIi4OyLOJbmUvKFPPLHW44S6NS0jOSCv9GOSZOM3wBeBe0gSlfJwW4GXS7qN5Azh55iczwPz0gZB7gTeFBF/IrkHahVwJ8l9r1dOYJrfBO4FfqfkMQPfYIx/nSX9GrgcWCxpnaS31rks1jyOXVhCkuC8r6JRkkaf+bEpll61cBLwFiUNgq0BzgGeGGO0P0v3VeVXtQO3dwLXR0TlmecrgRMk9YyzelcAd5HE8vXAJyLiyYh4HLgs7XcJcPs4pvX/SC45d6NkOTeFMT2WpcAH033xGpLbHczGbbxxGxH3Av8I/ELSXcC1wAvT3/9zSI5Bfgn8rsasvgtcWG+jZCN8tOIYpBeY9C0U1to0dKWjtTpJx5A0JDPqB0nSloiY0/xame2cY9fMzMzM8shnqM3MzMzMzMzq4DPUOSbpAiqet5f6akR8J4v6VCPpFSQtIVfaERGvzaI+1hocu5YH7RDHZhPhmLbpQNIVJI9xq/R/3Kiu1eKE2szMzMzMzKwOvuTbzMzMzMzMrA5OqM3MzMzMzMzq4ITazMzMzMzMrA5OqM3MzMzMzMzq8P8B6edBbM4WN5oAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 963.125x900 with 30 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(df_train, hue = 'CPL_Status')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature Engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# to check the no of null values in each column\n",
    "\n",
    "def no_of_null(dataset, name) :\n",
    "    print(name)\n",
    "    print(dataset.isnull().sum())\n",
    "    print('\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TRAIN SET\n",
      "Sex               13\n",
      "Marital_Status     3\n",
      "Dependents        15\n",
      "Qual_var           0\n",
      "SE                32\n",
      "App_Income_1       0\n",
      "App_Income_2       0\n",
      "CPL_Amount         2\n",
      "CPL_Term          14\n",
      "Credit_His        50\n",
      "Prop_Area          0\n",
      "CPL_Status         0\n",
      "dtype: int64\n",
      "\n",
      "\n",
      "TEST SET\n",
      "Sex               11\n",
      "Marital_Status     0\n",
      "Dependents        10\n",
      "Qual_var           0\n",
      "SE                23\n",
      "App_Income_1       0\n",
      "App_Income_2       0\n",
      "CPL_Amount         0\n",
      "CPL_Term           6\n",
      "Credit_His        29\n",
      "Prop_Area          0\n",
      "dtype: int64\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "no_of_null(df_train, \"TRAIN SET\")\n",
    "no_of_null(df_test, \"TEST SET\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# to check the %age of null values in each column\n",
    "\n",
    "def percent_of_null(dataset, name) :\n",
    "    print(name)\n",
    "    print(dataset.isnull().sum() / len(dataset) * 100)\n",
    "    print('\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TRAIN SET\n",
      "Sex               2.117264\n",
      "Marital_Status    0.488599\n",
      "Dependents        2.442997\n",
      "Qual_var          0.000000\n",
      "SE                5.211726\n",
      "App_Income_1      0.000000\n",
      "App_Income_2      0.000000\n",
      "CPL_Amount        0.325733\n",
      "CPL_Term          2.280130\n",
      "Credit_His        8.143322\n",
      "Prop_Area         0.000000\n",
      "CPL_Status        0.000000\n",
      "dtype: float64\n",
      "\n",
      "\n",
      "TEST SET\n",
      "Sex               2.997275\n",
      "Marital_Status    0.000000\n",
      "Dependents        2.724796\n",
      "Qual_var          0.000000\n",
      "SE                6.267030\n",
      "App_Income_1      0.000000\n",
      "App_Income_2      0.000000\n",
      "CPL_Amount        0.000000\n",
      "CPL_Term          1.634877\n",
      "Credit_His        7.901907\n",
      "Prop_Area         0.000000\n",
      "dtype: float64\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "percent_of_null(df_train, \"TRAIN SET\")\n",
    "percent_of_null(df_test, \"TEST SET\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# drop those columns which are string and have < 5% missing values\n",
    "\n",
    "df_train = df_train.dropna(subset = ['Sex', 'Marital_Status', 'Dependents', 'CPL_Amount', 'CPL_Term'])\n",
    "df_test = df_test.dropna(subset = ['Sex', 'Dependents', 'CPL_Term'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\AMANJEET PANI\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:3: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n"
     ]
    }
   ],
   "source": [
    "# fill the missing values of SE column with the previous valid value of that column\n",
    "\n",
    "df_train['SE'] = df_train['SE'].fillna(method = 'ffill', axis = 0)\n",
    "df_test['SE'] = df_test['SE'].fillna(method = 'ffill', axis = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# fill the  missing values of Credit_His column with the mean value for that column\n",
    "\n",
    "def fill_missing_values(dataset) :\n",
    "    if dataset['Credit_His'].mean() > 0.5 :\n",
    "        dataset['Credit_His'] = dataset['Credit_His'].fillna(value = 1, axis = 0)\n",
    "    else :\n",
    "        dataset['Credit_His'] = dataset['Credit_His'].fillna(value = 0, axis = 0)\n",
    "    return dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\AMANJEET PANI\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:5: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \"\"\"\n"
     ]
    }
   ],
   "source": [
    "df_train = fill_missing_values(df_train)\n",
    "df_test = fill_missing_values(df_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TRAIN SET\n",
      "Sex               0\n",
      "Marital_Status    0\n",
      "Dependents        0\n",
      "Qual_var          0\n",
      "SE                0\n",
      "App_Income_1      0\n",
      "App_Income_2      0\n",
      "CPL_Amount        0\n",
      "CPL_Term          0\n",
      "Credit_His        0\n",
      "Prop_Area         0\n",
      "CPL_Status        0\n",
      "dtype: int64\n",
      "\n",
      "\n",
      "TEST SET\n",
      "Sex               0\n",
      "Marital_Status    0\n",
      "Dependents        0\n",
      "Qual_var          0\n",
      "SE                0\n",
      "App_Income_1      0\n",
      "App_Income_2      0\n",
      "CPL_Amount        0\n",
      "CPL_Term          0\n",
      "Credit_His        0\n",
      "Prop_Area         0\n",
      "dtype: int64\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "no_of_null(df_train, 'TRAIN SET')\n",
    "no_of_null(df_test, 'TEST SET')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_dtypes(dataset, name) :\n",
    "    print(name)\n",
    "    print(dataset.dtypes)\n",
    "    print('\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TRAIN SET\n",
      "Sex                object\n",
      "Marital_Status     object\n",
      "Dependents         object\n",
      "Qual_var           object\n",
      "SE                 object\n",
      "App_Income_1      float64\n",
      "App_Income_2      float64\n",
      "CPL_Amount        float64\n",
      "CPL_Term          float64\n",
      "Credit_His        float64\n",
      "Prop_Area          object\n",
      "CPL_Status         object\n",
      "dtype: object\n",
      "\n",
      "\n",
      "TEST SET\n",
      "Sex                object\n",
      "Marital_Status     object\n",
      "Dependents         object\n",
      "Qual_var           object\n",
      "SE                 object\n",
      "App_Income_1      float64\n",
      "App_Income_2      float64\n",
      "CPL_Amount        float64\n",
      "CPL_Term          float64\n",
      "Credit_His        float64\n",
      "Prop_Area          object\n",
      "dtype: object\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#to check the datatypes of each column\n",
    "\n",
    "get_dtypes(df_train, \"TRAIN SET\")\n",
    "get_dtypes(df_test, \"TEST SET\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create dummy variable and drop the original column for each categorical column\n",
    "\n",
    "def to_categorical(dataset) :\n",
    "    dummies = pd.get_dummies(dataset, columns = ['Sex', 'Marital_Status', 'Dependents', 'Qual_var', 'SE', 'Prop_Area'], drop_first = True)\n",
    "    return dummies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train = to_categorical(df_train)\n",
    "df_test = to_categorical(df_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "dummy = pd.get_dummies(df_train['CPL_Status'], drop_first = True)\n",
    "df_train = df_train.drop('CPL_Status', axis = 1)\n",
    "df_train = pd.concat([df_train, dummy], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>App_Income_1</th>\n",
       "      <th>App_Income_2</th>\n",
       "      <th>CPL_Amount</th>\n",
       "      <th>CPL_Term</th>\n",
       "      <th>Credit_His</th>\n",
       "      <th>Sex_M</th>\n",
       "      <th>Marital_Status_Y</th>\n",
       "      <th>Dependents_1</th>\n",
       "      <th>Dependents_2</th>\n",
       "      <th>Dependents_3+</th>\n",
       "      <th>Qual_var_Non Grad</th>\n",
       "      <th>SE_Y</th>\n",
       "      <th>Prop_Area_Semi U</th>\n",
       "      <th>Prop_Area_Urban</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>5499.6</td>\n",
       "      <td>1809.6</td>\n",
       "      <td>153.6</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>3600.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>79.2</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>3099.6</td>\n",
       "      <td>2829.6</td>\n",
       "      <td>144.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>7200.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>169.2</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>6500.4</td>\n",
       "      <td>5035.2</td>\n",
       "      <td>320.4</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   App_Income_1  App_Income_2  CPL_Amount  CPL_Term  Credit_His  Sex_M  \\\n",
       "1        5499.6        1809.6       153.6     360.0         1.0      1   \n",
       "2        3600.0           0.0        79.2     360.0         1.0      1   \n",
       "3        3099.6        2829.6       144.0     360.0         1.0      1   \n",
       "4        7200.0           0.0       169.2     360.0         1.0      1   \n",
       "5        6500.4        5035.2       320.4     360.0         1.0      1   \n",
       "\n",
       "   Marital_Status_Y  Dependents_1  Dependents_2  Dependents_3+  \\\n",
       "1                 1             1             0              0   \n",
       "2                 1             0             0              0   \n",
       "3                 1             0             0              0   \n",
       "4                 0             0             0              0   \n",
       "5                 1             0             1              0   \n",
       "\n",
       "   Qual_var_Non Grad  SE_Y  Prop_Area_Semi U  Prop_Area_Urban  Y  \n",
       "1                  0     0                 0                0  0  \n",
       "2                  0     1                 0                1  1  \n",
       "3                  1     0                 0                1  1  \n",
       "4                  0     0                 0                1  1  \n",
       "5                  0     1                 0                1  1  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# we need to rename the target column to a more convenient name, since it was changed \n",
    "\n",
    "df_train = df_train.rename(columns = {'Y' : 'CPL_Status'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TRAIN SET\n",
      "App_Income_1         float64\n",
      "App_Income_2         float64\n",
      "CPL_Amount           float64\n",
      "CPL_Term             float64\n",
      "Credit_His           float64\n",
      "Sex_M                  uint8\n",
      "Marital_Status_Y       uint8\n",
      "Dependents_1           uint8\n",
      "Dependents_2           uint8\n",
      "Dependents_3+          uint8\n",
      "Qual_var_Non Grad      uint8\n",
      "SE_Y                   uint8\n",
      "Prop_Area_Semi U       uint8\n",
      "Prop_Area_Urban        uint8\n",
      "CPL_Status             uint8\n",
      "dtype: object\n",
      "\n",
      "\n",
      "TEST SET\n",
      "App_Income_1         float64\n",
      "App_Income_2         float64\n",
      "CPL_Amount           float64\n",
      "CPL_Term             float64\n",
      "Credit_His           float64\n",
      "Sex_M                  uint8\n",
      "Marital_Status_Y       uint8\n",
      "Dependents_1           uint8\n",
      "Dependents_2           uint8\n",
      "Dependents_3+          uint8\n",
      "Qual_var_Non Grad      uint8\n",
      "SE_Y                   uint8\n",
      "Prop_Area_Semi U       uint8\n",
      "Prop_Area_Urban        uint8\n",
      "dtype: object\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "get_dtypes(df_train, \"TRAIN SET\")\n",
    "get_dtypes(df_test, \"TEST SET\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>App_Income_1</th>\n",
       "      <th>App_Income_2</th>\n",
       "      <th>CPL_Amount</th>\n",
       "      <th>CPL_Term</th>\n",
       "      <th>Credit_His</th>\n",
       "      <th>Sex_M</th>\n",
       "      <th>Marital_Status_Y</th>\n",
       "      <th>Dependents_1</th>\n",
       "      <th>Dependents_2</th>\n",
       "      <th>Dependents_3+</th>\n",
       "      <th>Qual_var_Non Grad</th>\n",
       "      <th>SE_Y</th>\n",
       "      <th>Prop_Area_Semi U</th>\n",
       "      <th>Prop_Area_Urban</th>\n",
       "      <th>CPL_Status</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>App_Income_1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.108449</td>\n",
       "      <td>0.031098</td>\n",
       "      <td>-0.039125</td>\n",
       "      <td>-0.044563</td>\n",
       "      <td>0.055236</td>\n",
       "      <td>0.038984</td>\n",
       "      <td>0.045036</td>\n",
       "      <td>-0.028612</td>\n",
       "      <td>0.104880</td>\n",
       "      <td>-0.133853</td>\n",
       "      <td>0.143373</td>\n",
       "      <td>-0.009252</td>\n",
       "      <td>-0.014382</td>\n",
       "      <td>-0.032088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>App_Income_2</td>\n",
       "      <td>-0.108449</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.088806</td>\n",
       "      <td>-0.057810</td>\n",
       "      <td>0.009277</td>\n",
       "      <td>0.078265</td>\n",
       "      <td>0.064365</td>\n",
       "      <td>-0.033878</td>\n",
       "      <td>0.006501</td>\n",
       "      <td>0.055490</td>\n",
       "      <td>-0.072101</td>\n",
       "      <td>-0.031439</td>\n",
       "      <td>-0.026738</td>\n",
       "      <td>0.037258</td>\n",
       "      <td>-0.063209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>CPL_Amount</td>\n",
       "      <td>0.031098</td>\n",
       "      <td>0.088806</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.008742</td>\n",
       "      <td>0.006679</td>\n",
       "      <td>-0.031797</td>\n",
       "      <td>0.072036</td>\n",
       "      <td>0.081990</td>\n",
       "      <td>0.086258</td>\n",
       "      <td>0.062270</td>\n",
       "      <td>-0.106363</td>\n",
       "      <td>-0.043891</td>\n",
       "      <td>0.008238</td>\n",
       "      <td>0.016092</td>\n",
       "      <td>0.042352</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>CPL_Term</td>\n",
       "      <td>-0.039125</td>\n",
       "      <td>-0.057810</td>\n",
       "      <td>-0.008742</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.021134</td>\n",
       "      <td>-0.087174</td>\n",
       "      <td>-0.103351</td>\n",
       "      <td>-0.078728</td>\n",
       "      <td>-0.010175</td>\n",
       "      <td>-0.081563</td>\n",
       "      <td>-0.077637</td>\n",
       "      <td>-0.024070</td>\n",
       "      <td>0.055181</td>\n",
       "      <td>-0.087712</td>\n",
       "      <td>-0.014120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Credit_His</td>\n",
       "      <td>-0.044563</td>\n",
       "      <td>0.009277</td>\n",
       "      <td>0.006679</td>\n",
       "      <td>0.021134</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.014362</td>\n",
       "      <td>0.018805</td>\n",
       "      <td>0.001840</td>\n",
       "      <td>0.013743</td>\n",
       "      <td>-0.072188</td>\n",
       "      <td>-0.050286</td>\n",
       "      <td>-0.026493</td>\n",
       "      <td>0.040149</td>\n",
       "      <td>-0.021812</td>\n",
       "      <td>0.508520</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Sex_M</td>\n",
       "      <td>0.055236</td>\n",
       "      <td>0.078265</td>\n",
       "      <td>-0.031797</td>\n",
       "      <td>-0.087174</td>\n",
       "      <td>0.014362</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.357581</td>\n",
       "      <td>-0.005168</td>\n",
       "      <td>0.135377</td>\n",
       "      <td>0.093826</td>\n",
       "      <td>0.051076</td>\n",
       "      <td>-0.006822</td>\n",
       "      <td>-0.123618</td>\n",
       "      <td>0.057641</td>\n",
       "      <td>0.026409</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Marital_Status_Y</td>\n",
       "      <td>0.038984</td>\n",
       "      <td>0.064365</td>\n",
       "      <td>0.072036</td>\n",
       "      <td>-0.103351</td>\n",
       "      <td>0.018805</td>\n",
       "      <td>0.357581</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.116159</td>\n",
       "      <td>0.258682</td>\n",
       "      <td>0.139749</td>\n",
       "      <td>0.006556</td>\n",
       "      <td>-0.017953</td>\n",
       "      <td>-0.002208</td>\n",
       "      <td>0.016505</td>\n",
       "      <td>0.104116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Dependents_1</td>\n",
       "      <td>0.045036</td>\n",
       "      <td>-0.033878</td>\n",
       "      <td>0.081990</td>\n",
       "      <td>-0.078728</td>\n",
       "      <td>0.001840</td>\n",
       "      <td>-0.005168</td>\n",
       "      <td>0.116159</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.211026</td>\n",
       "      <td>-0.137998</td>\n",
       "      <td>-0.004115</td>\n",
       "      <td>0.071643</td>\n",
       "      <td>0.015600</td>\n",
       "      <td>0.068632</td>\n",
       "      <td>-0.045319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Dependents_2</td>\n",
       "      <td>-0.028612</td>\n",
       "      <td>0.006501</td>\n",
       "      <td>0.086258</td>\n",
       "      <td>-0.010175</td>\n",
       "      <td>0.013743</td>\n",
       "      <td>0.135377</td>\n",
       "      <td>0.258682</td>\n",
       "      <td>-0.211026</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.137161</td>\n",
       "      <td>0.020853</td>\n",
       "      <td>0.033357</td>\n",
       "      <td>-0.018744</td>\n",
       "      <td>0.032430</td>\n",
       "      <td>0.072042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Dependents_3+</td>\n",
       "      <td>0.104880</td>\n",
       "      <td>0.055490</td>\n",
       "      <td>0.062270</td>\n",
       "      <td>-0.081563</td>\n",
       "      <td>-0.072188</td>\n",
       "      <td>0.093826</td>\n",
       "      <td>0.139749</td>\n",
       "      <td>-0.137998</td>\n",
       "      <td>-0.137161</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.061531</td>\n",
       "      <td>-0.007799</td>\n",
       "      <td>-0.000344</td>\n",
       "      <td>-0.041825</td>\n",
       "      <td>-0.037070</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Qual_var_Non Grad</td>\n",
       "      <td>-0.133853</td>\n",
       "      <td>-0.072101</td>\n",
       "      <td>-0.106363</td>\n",
       "      <td>-0.077637</td>\n",
       "      <td>-0.050286</td>\n",
       "      <td>0.051076</td>\n",
       "      <td>0.006556</td>\n",
       "      <td>-0.004115</td>\n",
       "      <td>0.020853</td>\n",
       "      <td>0.061531</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.008280</td>\n",
       "      <td>-0.033317</td>\n",
       "      <td>-0.037537</td>\n",
       "      <td>-0.063335</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>SE_Y</td>\n",
       "      <td>0.143373</td>\n",
       "      <td>-0.031439</td>\n",
       "      <td>-0.043891</td>\n",
       "      <td>-0.024070</td>\n",
       "      <td>-0.026493</td>\n",
       "      <td>-0.006822</td>\n",
       "      <td>-0.017953</td>\n",
       "      <td>0.071643</td>\n",
       "      <td>0.033357</td>\n",
       "      <td>-0.007799</td>\n",
       "      <td>-0.008280</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000882</td>\n",
       "      <td>-0.032764</td>\n",
       "      <td>-0.013640</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Prop_Area_Semi U</td>\n",
       "      <td>-0.009252</td>\n",
       "      <td>-0.026738</td>\n",
       "      <td>0.008238</td>\n",
       "      <td>0.055181</td>\n",
       "      <td>0.040149</td>\n",
       "      <td>-0.123618</td>\n",
       "      <td>-0.002208</td>\n",
       "      <td>0.015600</td>\n",
       "      <td>-0.018744</td>\n",
       "      <td>-0.000344</td>\n",
       "      <td>-0.033317</td>\n",
       "      <td>0.000882</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.541702</td>\n",
       "      <td>0.138777</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Prop_Area_Urban</td>\n",
       "      <td>-0.014382</td>\n",
       "      <td>0.037258</td>\n",
       "      <td>0.016092</td>\n",
       "      <td>-0.087712</td>\n",
       "      <td>-0.021812</td>\n",
       "      <td>0.057641</td>\n",
       "      <td>0.016505</td>\n",
       "      <td>0.068632</td>\n",
       "      <td>0.032430</td>\n",
       "      <td>-0.041825</td>\n",
       "      <td>-0.037537</td>\n",
       "      <td>-0.032764</td>\n",
       "      <td>-0.541702</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.042679</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>CPL_Status</td>\n",
       "      <td>-0.032088</td>\n",
       "      <td>-0.063209</td>\n",
       "      <td>0.042352</td>\n",
       "      <td>-0.014120</td>\n",
       "      <td>0.508520</td>\n",
       "      <td>0.026409</td>\n",
       "      <td>0.104116</td>\n",
       "      <td>-0.045319</td>\n",
       "      <td>0.072042</td>\n",
       "      <td>-0.037070</td>\n",
       "      <td>-0.063335</td>\n",
       "      <td>-0.013640</td>\n",
       "      <td>0.138777</td>\n",
       "      <td>-0.042679</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   App_Income_1  App_Income_2  CPL_Amount  CPL_Term  \\\n",
       "App_Income_1           1.000000     -0.108449    0.031098 -0.039125   \n",
       "App_Income_2          -0.108449      1.000000    0.088806 -0.057810   \n",
       "CPL_Amount             0.031098      0.088806    1.000000 -0.008742   \n",
       "CPL_Term              -0.039125     -0.057810   -0.008742  1.000000   \n",
       "Credit_His            -0.044563      0.009277    0.006679  0.021134   \n",
       "Sex_M                  0.055236      0.078265   -0.031797 -0.087174   \n",
       "Marital_Status_Y       0.038984      0.064365    0.072036 -0.103351   \n",
       "Dependents_1           0.045036     -0.033878    0.081990 -0.078728   \n",
       "Dependents_2          -0.028612      0.006501    0.086258 -0.010175   \n",
       "Dependents_3+          0.104880      0.055490    0.062270 -0.081563   \n",
       "Qual_var_Non Grad     -0.133853     -0.072101   -0.106363 -0.077637   \n",
       "SE_Y                   0.143373     -0.031439   -0.043891 -0.024070   \n",
       "Prop_Area_Semi U      -0.009252     -0.026738    0.008238  0.055181   \n",
       "Prop_Area_Urban       -0.014382      0.037258    0.016092 -0.087712   \n",
       "CPL_Status            -0.032088     -0.063209    0.042352 -0.014120   \n",
       "\n",
       "                   Credit_His     Sex_M  Marital_Status_Y  Dependents_1  \\\n",
       "App_Income_1        -0.044563  0.055236          0.038984      0.045036   \n",
       "App_Income_2         0.009277  0.078265          0.064365     -0.033878   \n",
       "CPL_Amount           0.006679 -0.031797          0.072036      0.081990   \n",
       "CPL_Term             0.021134 -0.087174         -0.103351     -0.078728   \n",
       "Credit_His           1.000000  0.014362          0.018805      0.001840   \n",
       "Sex_M                0.014362  1.000000          0.357581     -0.005168   \n",
       "Marital_Status_Y     0.018805  0.357581          1.000000      0.116159   \n",
       "Dependents_1         0.001840 -0.005168          0.116159      1.000000   \n",
       "Dependents_2         0.013743  0.135377          0.258682     -0.211026   \n",
       "Dependents_3+       -0.072188  0.093826          0.139749     -0.137998   \n",
       "Qual_var_Non Grad   -0.050286  0.051076          0.006556     -0.004115   \n",
       "SE_Y                -0.026493 -0.006822         -0.017953      0.071643   \n",
       "Prop_Area_Semi U     0.040149 -0.123618         -0.002208      0.015600   \n",
       "Prop_Area_Urban     -0.021812  0.057641          0.016505      0.068632   \n",
       "CPL_Status           0.508520  0.026409          0.104116     -0.045319   \n",
       "\n",
       "                   Dependents_2  Dependents_3+  Qual_var_Non Grad      SE_Y  \\\n",
       "App_Income_1          -0.028612       0.104880          -0.133853  0.143373   \n",
       "App_Income_2           0.006501       0.055490          -0.072101 -0.031439   \n",
       "CPL_Amount             0.086258       0.062270          -0.106363 -0.043891   \n",
       "CPL_Term              -0.010175      -0.081563          -0.077637 -0.024070   \n",
       "Credit_His             0.013743      -0.072188          -0.050286 -0.026493   \n",
       "Sex_M                  0.135377       0.093826           0.051076 -0.006822   \n",
       "Marital_Status_Y       0.258682       0.139749           0.006556 -0.017953   \n",
       "Dependents_1          -0.211026      -0.137998          -0.004115  0.071643   \n",
       "Dependents_2           1.000000      -0.137161           0.020853  0.033357   \n",
       "Dependents_3+         -0.137161       1.000000           0.061531 -0.007799   \n",
       "Qual_var_Non Grad      0.020853       0.061531           1.000000 -0.008280   \n",
       "SE_Y                   0.033357      -0.007799          -0.008280  1.000000   \n",
       "Prop_Area_Semi U      -0.018744      -0.000344          -0.033317  0.000882   \n",
       "Prop_Area_Urban        0.032430      -0.041825          -0.037537 -0.032764   \n",
       "CPL_Status             0.072042      -0.037070          -0.063335 -0.013640   \n",
       "\n",
       "                   Prop_Area_Semi U  Prop_Area_Urban  CPL_Status  \n",
       "App_Income_1              -0.009252        -0.014382   -0.032088  \n",
       "App_Income_2              -0.026738         0.037258   -0.063209  \n",
       "CPL_Amount                 0.008238         0.016092    0.042352  \n",
       "CPL_Term                   0.055181        -0.087712   -0.014120  \n",
       "Credit_His                 0.040149        -0.021812    0.508520  \n",
       "Sex_M                     -0.123618         0.057641    0.026409  \n",
       "Marital_Status_Y          -0.002208         0.016505    0.104116  \n",
       "Dependents_1               0.015600         0.068632   -0.045319  \n",
       "Dependents_2              -0.018744         0.032430    0.072042  \n",
       "Dependents_3+             -0.000344        -0.041825   -0.037070  \n",
       "Qual_var_Non Grad         -0.033317        -0.037537   -0.063335  \n",
       "SE_Y                       0.000882        -0.032764   -0.013640  \n",
       "Prop_Area_Semi U           1.000000        -0.541702    0.138777  \n",
       "Prop_Area_Urban           -0.541702         1.000000   -0.042679  \n",
       "CPL_Status                 0.138777        -0.042679    1.000000  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# to check the correlation of each column of the dataset with each other\n",
    "df_train.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1c3f28b6c88>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# to plot the heatmap of the correlation dataframe\n",
    "\n",
    "plt.figure(figsize = (12, 8))\n",
    "sns.heatmap(df_train.corr())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1c3f2590548>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# to plot a bar chart for the correlation of each column with the target column\n",
    "\n",
    "plt.ylim(-0.15, 0.6)\n",
    "df_train.corr()['CPL_Status'].sort_values()[:-1].plot(kind = 'bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# drop the CPL_Term column since it is the least related and doesn't make a difference for our prediction\n",
    "\n",
    "df_train = df_train.drop('CPL_Term', axis = 1)\n",
    "df_test = df_test.drop('CPL_Term', axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# separate the independent and the dependent features for training of model\n",
    "\n",
    "X = df_train.drop('CPL_Status', axis = 1)\n",
    "y = df_train['CPL_Status']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Train - Test Split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(428, 13) (143, 13) (428,) (143,)\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape, X_test.shape, y_train.shape, y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature Scaling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "sc = MinMaxScaler()\n",
    "X_train = sc.fit_transform(X_train)\n",
    "X_test = sc.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model Creation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import all the model classes from their respective classes and create an object for each\n",
    "\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "model1 = LogisticRegression()\n",
    "\n",
    "from sklearn.svm import SVC\n",
    "model2 = SVC()\n",
    "\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "#model3 = KNeighborsClassifier()\n",
    "\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "model4 = DecisionTreeClassifier()\n",
    "\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "model5 = RandomForestClassifier()\n",
    "\n",
    "from xgboost import XGBClassifier\n",
    "model6 = XGBClassifier()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Hyperparameter Optimization/Tuning\n",
    "\n",
    "# For SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 5 candidates, totalling 25 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  19 out of  25 | elapsed:    5.5s remaining:    1.7s\n",
      "[Parallel(n_jobs=-1)]: Done  25 out of  25 | elapsed:    5.6s finished\n",
      "C:\\Users\\AMANJEET PANI\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:814: DeprecationWarning: The default of the `iid` parameter will change from True to False in version 0.22 and will be removed in 0.24. This will change numeric results when test-set sizes are unequal.\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, error_score='raise-deprecating',\n",
       "                   estimator=SVC(C=1.0, cache_size=200, class_weight=None,\n",
       "                                 coef0=0.0, decision_function_shape='ovr',\n",
       "                                 degree=3, gamma='auto_deprecated',\n",
       "                                 kernel='rbf', max_iter=-1, probability=False,\n",
       "                                 random_state=None, shrinking=True, tol=0.001,\n",
       "                                 verbose=False),\n",
       "                   iid='warn', n_iter=5, n_jobs=-1,\n",
       "                   param_distributions={'C': [0.1, 1, 10, 100, 1000],\n",
       "                                        'gamma': [1, 0.1, 0.01, 0.001, 0.0001],\n",
       "                                        'kernel': ['rbf']},\n",
       "                   pre_dispatch='2*n_jobs', random_state=None, refit=True,\n",
       "                   return_train_score=False, scoring='roc_auc', verbose=3)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "\n",
    "# hyperparameters to tune :\n",
    "params2 = {'C': [0.1, 1, 10, 100, 1000],  \n",
    "           'gamma': [1, 0.1, 0.01, 0.001, 0.0001], \n",
    "           'kernel': ['rbf']}\n",
    "r_search2 = RandomizedSearchCV(model2, param_distributions = params2, n_iter = 5,\n",
    "                                   scoring = 'roc_auc', n_jobs = -1, cv = 5, verbose = 3)\n",
    "r_search2.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'kernel': 'rbf', 'gamma': 0.0001, 'C': 1000}"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r_search2.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=1000, cache_size=200, class_weight=None, coef0=0.0,\n",
       "    decision_function_shape='ovr', degree=3, gamma=0.0001, kernel='rbf',\n",
       "    max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
       "    tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r_search2.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# rebuilding the model with the optimized hyperparameters\n",
    "model2 = SVC(C=10, cache_size=200, class_weight=None, coef0=0.0,\n",
    "            decision_function_shape='ovr', degree=3, gamma=0.001, kernel='rbf',\n",
    "            max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
    "            tol=0.001, verbose=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# For KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "# for selecting the best value of k\n",
    "losses = []\n",
    "\n",
    "for i in range(1,40):\n",
    "    \n",
    "    knn = KNeighborsClassifier(n_neighbors=i)\n",
    "    knn.fit(X_train,y_train)\n",
    "    ith_pred = knn.predict(X_test)\n",
    "    # losses obtained from each n_neighbors value\n",
    "    losses.append(np.mean(ith_pred != y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Losses')"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the losses vs k-value graph to get the best k-value with min loss\n",
    "plt.figure(figsize=(10,6))\n",
    "plt.plot(range(1,40), losses, color = 'blue', marker='o', markerfacecolor = 'red', markersize = 5)\n",
    "plt.xlabel('K-value')\n",
    "plt.ylabel('Losses')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "#from graph we can see for k-value = 14 the loss is min\n",
    "model3 = KNeighborsClassifier(n_neighbors = 14)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# For RandomForest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 5 candidates, totalling 25 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  19 out of  25 | elapsed:   10.4s remaining:    3.2s\n",
      "[Parallel(n_jobs=-1)]: Done  25 out of  25 | elapsed:   13.1s finished\n",
      "C:\\Users\\AMANJEET PANI\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:814: DeprecationWarning: The default of the `iid` parameter will change from True to False in version 0.22 and will be removed in 0.24. This will change numeric results when test-set sizes are unequal.\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, error_score='raise-deprecating',\n",
       "                   estimator=RandomForestClassifier(bootstrap=True,\n",
       "                                                    class_weight=None,\n",
       "                                                    criterion='gini',\n",
       "                                                    max_depth=None,\n",
       "                                                    max_features='auto',\n",
       "                                                    max_leaf_nodes=None,\n",
       "                                                    min_impurity_decrease=0.0,\n",
       "                                                    min_impurity_split=None,\n",
       "                                                    min_samples_leaf=1,\n",
       "                                                    min_samples_split=2,\n",
       "                                                    min_weight_fraction_leaf=0.0,\n",
       "                                                    n_estimators='warn',\n",
       "                                                    n_jobs=None,\n",
       "                                                    oob_sc...\n",
       "                   iid='warn', n_iter=5, n_jobs=-1,\n",
       "                   param_distributions={'max_depth': [10, 20, 30, 40, 50, 60,\n",
       "                                                      70, 80, 90, 100, None],\n",
       "                                        'max_features': ['auto', 'sqrt'],\n",
       "                                        'min_samples_leaf': [1, 2, 4],\n",
       "                                        'min_samples_split': [2, 5, 10],\n",
       "                                        'n_estimators': [200, 400, 600, 800,\n",
       "                                                         1000, 1200, 1400, 1600,\n",
       "                                                         1800, 2000]},\n",
       "                   pre_dispatch='2*n_jobs', random_state=None, refit=True,\n",
       "                   return_train_score=False, scoring='roc_auc', verbose=3)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# hyperparameters to tune :\n",
    "params5 = {'max_depth': [10, 20, 30, 40, 50, 60, 70, 80, 90, 100, None],\n",
    "           'max_features': ['auto', 'sqrt'],\n",
    "           'min_samples_leaf': [1, 2, 4],\n",
    "           'min_samples_split': [2, 5, 10],\n",
    "           'n_estimators': [200, 400, 600, 800, 1000, 1200, 1400, 1600, 1800, 2000]}\n",
    "r_search5 = RandomizedSearchCV(model5, param_distributions = params5, n_iter = 5,\n",
    "                                   scoring = 'roc_auc', n_jobs = -1, cv = 5, verbose = 3)\n",
    "r_search5.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'n_estimators': 1800,\n",
       " 'min_samples_split': 10,\n",
       " 'min_samples_leaf': 2,\n",
       " 'max_features': 'sqrt',\n",
       " 'max_depth': 60}"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r_search5.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "                       max_depth=60, max_features='sqrt', max_leaf_nodes=None,\n",
       "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                       min_samples_leaf=2, min_samples_split=10,\n",
       "                       min_weight_fraction_leaf=0.0, n_estimators=1800,\n",
       "                       n_jobs=None, oob_score=False, random_state=None,\n",
       "                       verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r_search5.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "# rebuilding the model with the optimized hyperparameters\n",
    "model5 = RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
    "                                max_depth=60, max_features='sqrt', max_leaf_nodes=None,\n",
    "                                min_impurity_decrease=0.0, min_impurity_split=None,\n",
    "                                min_samples_leaf=4, min_samples_split=10,\n",
    "                                min_weight_fraction_leaf=0.0, n_estimators=800,\n",
    "                                n_jobs=None, oob_score=False, random_state=None,\n",
    "                                verbose=0, warm_start=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# For XGBoost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 5 candidates, totalling 25 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  19 out of  25 | elapsed:    2.9s remaining:    0.9s\n",
      "[Parallel(n_jobs=-1)]: Done  25 out of  25 | elapsed:    2.9s finished\n",
      "C:\\Users\\AMANJEET PANI\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:814: DeprecationWarning: The default of the `iid` parameter will change from True to False in version 0.22 and will be removed in 0.24. This will change numeric results when test-set sizes are unequal.\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, error_score='raise-deprecating',\n",
       "                   estimator=XGBClassifier(base_score=None, booster=None,\n",
       "                                           colsample_bylevel=None,\n",
       "                                           colsample_bynode=None,\n",
       "                                           colsample_bytree=None, gamma=None,\n",
       "                                           gpu_id=None, importance_type='gain',\n",
       "                                           interaction_constraints=None,\n",
       "                                           learning_rate=None,\n",
       "                                           max_delta_step=None, max_depth=None,\n",
       "                                           min_child_weight=None, missing=nan,\n",
       "                                           monotone_co...\n",
       "                   iid='warn', n_iter=5, n_jobs=-1,\n",
       "                   param_distributions={'colsample_bylevel': [0.5, 0.7, 0.9, 1],\n",
       "                                        'colsample_bytree': [0.3, 0.4, 0.5,\n",
       "                                                             0.7],\n",
       "                                        'gamma': [0.0, 0.1, 0.2, 0.3, 0.4],\n",
       "                                        'learning_rate': [0.05, 0.1, 0.15, 0.2,\n",
       "                                                          0.25, 0.3],\n",
       "                                        'max_depth': [3, 4, 5, 6, 8, 10],\n",
       "                                        'min_child_weight': [1, 3, 5, 7]},\n",
       "                   pre_dispatch='2*n_jobs', random_state=None, refit=True,\n",
       "                   return_train_score=False, scoring='roc_auc', verbose=3)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# hyperparameters to tune :\n",
    "params6 = {\"learning_rate\" : [0.05, 0.10, 0.15, 0.20, 0.25, 0.30 ] ,\n",
    "          \"max_depth\" : [ 3, 4, 5, 6, 8, 10 ],\n",
    "          \"min_child_weight\" : [ 1, 3, 5, 7 ],\n",
    "          \"gamma\" : [ 0.0, 0.1, 0.2 , 0.3, 0.4 ],\n",
    "          \"colsample_bytree\" : [ 0.3, 0.4, 0.5 , 0.7 ],\n",
    "          \"colsample_bylevel\" : [ 0.5, 0.7, 0.9, 1 ]}\n",
    "r_search6 = RandomizedSearchCV(model6, param_distributions = params6, n_iter = 5,\n",
    "                                   scoring = 'roc_auc', n_jobs = -1, cv = 5, verbose = 3)\n",
    "r_search6.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'min_child_weight': 5,\n",
       " 'max_depth': 8,\n",
       " 'learning_rate': 0.2,\n",
       " 'gamma': 0.4,\n",
       " 'colsample_bytree': 0.3,\n",
       " 'colsample_bylevel': 0.5}"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r_search6.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=0.5,\n",
       "              colsample_bynode=1, colsample_bytree=0.3, gamma=0.4, gpu_id=-1,\n",
       "              importance_type='gain', interaction_constraints='',\n",
       "              learning_rate=0.2, max_delta_step=0, max_depth=8,\n",
       "              min_child_weight=5, missing=nan, monotone_constraints='()',\n",
       "              n_estimators=100, n_jobs=0, num_parallel_tree=1,\n",
       "              objective='binary:logistic', random_state=0, reg_alpha=0,\n",
       "              reg_lambda=1, scale_pos_weight=1, subsample=1,\n",
       "              tree_method='exact', validate_parameters=1, verbosity=None)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r_search6.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "# rebuilding the model with the optimized hyperparameters\n",
    "model6 = XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=0.9,\n",
    "                          colsample_bynode=1, colsample_bytree=0.4, gamma=0.3, gpu_id=-1,\n",
    "                          importance_type='gain', interaction_constraints='',\n",
    "                          learning_rate=0.05, max_delta_step=0, max_depth=5,\n",
    "                          min_child_weight=7, missing=np.nan, monotone_constraints='()',\n",
    "                          n_estimators=100, n_jobs=0, num_parallel_tree=1,\n",
    "                          objective='binary:logistic', random_state=0, reg_alpha=0,\n",
    "                          reg_lambda=1, scale_pos_weight=1, subsample=1,\n",
    "                          tree_method='exact', validate_parameters=1, verbosity=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "models = [model1, model2, model3, model4, model5, model6]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "# for calculating the accuracy for each model\n",
    "\n",
    "from sklearn.metrics import accuracy_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\AMANJEET PANI\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ACCURACY SCORE for model1 = 82.517483 percent\n",
      "ACCURACY SCORE for model2 = 76.223776 percent\n",
      "ACCURACY SCORE for model3 = 80.419580 percent\n",
      "ACCURACY SCORE for model4 = 66.433566 percent\n",
      "ACCURACY SCORE for model5 = 83.916084 percent\n",
      "ACCURACY SCORE for model6 = 83.916084 percent\n"
     ]
    }
   ],
   "source": [
    "# loop through each model\n",
    "\n",
    "i = 1\n",
    "for model in models :\n",
    "    # fit the train sets\n",
    "    model.fit(X_train, y_train)\n",
    "    \n",
    "    # predict the target column for the test set\n",
    "    y_pred = model.predict(X_test)\n",
    "    \n",
    "    # calculate the accuracy obtained\n",
    "    print(\"ACCURACY SCORE for model%d = %f percent\" %(i ,accuracy_score(y_test, y_pred)*100))\n",
    "    i +=1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# New Prediction for the Test Dataset File"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "# make similar changes for the test set (which we have already did in terms of df_test)\n",
    "# scale the test set and predict the target column with model6, i.e., XGBoost Classifier which gave the maximum accuracy\n",
    "\n",
    "df_test = sc.transform(df_test)\n",
    "test_pred = model6.predict(df_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "# replace 1 with Y and 0 with N to match the format of the test dataset file\n",
    "\n",
    "CPL_Status = [\"\" for x in range(len(test_pred))]\n",
    "for i in range(len(test_pred)) :\n",
    "    if test_pred[i] == 1 :\n",
    "        CPL_Status[i] = 'Y'\n",
    "    else :\n",
    "        CPL_Status[i] = 'N'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "# to make changes in the original test dataset file\n",
    "\n",
    "test_data = pd.read_excel('Project - 4 - Test Data.xlsx')\n",
    "\n",
    "# we just need to handle the missing values in our test dataset to maintain similar shape of dataset and the predicted column\n",
    "test_data = test_data.dropna(subset = ['Sex', 'Dependents', 'CPL_Term'])\n",
    "test_data['SE'] = test_data['SE'].fillna(method = 'ffill', axis = 0)\n",
    "test_data = fill_missing_values(test_data)\n",
    "\n",
    "# now create a column and add the model predictions\n",
    "test_data['CPL_Status'] = CPL_Status"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Loanapp_ID</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Marital_Status</th>\n",
       "      <th>first_name</th>\n",
       "      <th>last_name</th>\n",
       "      <th>email</th>\n",
       "      <th>address</th>\n",
       "      <th>Dependents</th>\n",
       "      <th>Qual_var</th>\n",
       "      <th>SE</th>\n",
       "      <th>App_Income_1</th>\n",
       "      <th>App_Income_2</th>\n",
       "      <th>CPL_Amount</th>\n",
       "      <th>CPL_Term</th>\n",
       "      <th>Credit_His</th>\n",
       "      <th>INT_ID</th>\n",
       "      <th>Prev_ID</th>\n",
       "      <th>AGT_ID</th>\n",
       "      <th>Prop_Area</th>\n",
       "      <th>CPL_Status</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>CPL001015</td>\n",
       "      <td>M</td>\n",
       "      <td>Y</td>\n",
       "      <td>Dollie</td>\n",
       "      <td>Scafe</td>\n",
       "      <td>dscafe0@is.gd</td>\n",
       "      <td>02 Garrison Pass</td>\n",
       "      <td>0</td>\n",
       "      <td>Grad</td>\n",
       "      <td>N</td>\n",
       "      <td>6864.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>132.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6869736235</td>\n",
       "      <td>79-489-1539</td>\n",
       "      <td>Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10_6...</td>\n",
       "      <td>Urban</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>CPL001022</td>\n",
       "      <td>M</td>\n",
       "      <td>Y</td>\n",
       "      <td>Kim</td>\n",
       "      <td>Arne</td>\n",
       "      <td>karne9@infoseek.co.jp</td>\n",
       "      <td>6568 Dwight Point</td>\n",
       "      <td>1</td>\n",
       "      <td>Grad</td>\n",
       "      <td>N</td>\n",
       "      <td>3691.2</td>\n",
       "      <td>1800.0</td>\n",
       "      <td>151.2</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3383749711</td>\n",
       "      <td>75-757-9174</td>\n",
       "      <td>Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_2)...</td>\n",
       "      <td>Urban</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>CPL001031</td>\n",
       "      <td>M</td>\n",
       "      <td>Y</td>\n",
       "      <td>Niccolo</td>\n",
       "      <td>Ramble</td>\n",
       "      <td>nramblef@gov.uk</td>\n",
       "      <td>3504 Lukken Park</td>\n",
       "      <td>2</td>\n",
       "      <td>Grad</td>\n",
       "      <td>N</td>\n",
       "      <td>6000.0</td>\n",
       "      <td>2160.0</td>\n",
       "      <td>249.6</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>7247894032</td>\n",
       "      <td>42-085-4345</td>\n",
       "      <td>Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKi...</td>\n",
       "      <td>Urban</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>CPL001035</td>\n",
       "      <td>M</td>\n",
       "      <td>Y</td>\n",
       "      <td>Rabbi</td>\n",
       "      <td>Haig</td>\n",
       "      <td>rhaigg@tiny.cc</td>\n",
       "      <td>1334 Florence Way</td>\n",
       "      <td>2</td>\n",
       "      <td>Grad</td>\n",
       "      <td>N</td>\n",
       "      <td>2808.0</td>\n",
       "      <td>3055.2</td>\n",
       "      <td>120.0</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3240492342</td>\n",
       "      <td>10-813-9695</td>\n",
       "      <td>Mozilla/5.0 (X11; Linux i686) AppleWebKit/534....</td>\n",
       "      <td>Urban</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>CPL001051</td>\n",
       "      <td>M</td>\n",
       "      <td>N</td>\n",
       "      <td>Gracie</td>\n",
       "      <td>Winear</td>\n",
       "      <td>gwinearj@devhub.com</td>\n",
       "      <td>9800 Ridge Oak Crossing</td>\n",
       "      <td>0</td>\n",
       "      <td>Non Grad</td>\n",
       "      <td>N</td>\n",
       "      <td>3931.2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>93.6</td>\n",
       "      <td>360.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1670564037</td>\n",
       "      <td>20-195-4948</td>\n",
       "      <td>Mozilla/5.0 (Windows NT 6.0) AppleWebKit/534.2...</td>\n",
       "      <td>Urban</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Loanapp_ID Sex Marital_Status first_name last_name                  email  \\\n",
       "0  CPL001015   M              Y     Dollie     Scafe          dscafe0@is.gd   \n",
       "1  CPL001022   M              Y        Kim      Arne  karne9@infoseek.co.jp   \n",
       "2  CPL001031   M              Y    Niccolo    Ramble        nramblef@gov.uk   \n",
       "3  CPL001035   M              Y      Rabbi      Haig         rhaigg@tiny.cc   \n",
       "4  CPL001051   M              N     Gracie    Winear    gwinearj@devhub.com   \n",
       "\n",
       "                   address Dependents  Qual_var SE  App_Income_1  \\\n",
       "0         02 Garrison Pass          0      Grad  N        6864.0   \n",
       "1        6568 Dwight Point          1      Grad  N        3691.2   \n",
       "2         3504 Lukken Park          2      Grad  N        6000.0   \n",
       "3        1334 Florence Way          2      Grad  N        2808.0   \n",
       "4  9800 Ridge Oak Crossing          0  Non Grad  N        3931.2   \n",
       "\n",
       "   App_Income_2  CPL_Amount  CPL_Term  Credit_His      INT_ID      Prev_ID  \\\n",
       "0           0.0       132.0     360.0         1.0  6869736235  79-489-1539   \n",
       "1        1800.0       151.2     360.0         1.0  3383749711  75-757-9174   \n",
       "2        2160.0       249.6     360.0         1.0  7247894032  42-085-4345   \n",
       "3        3055.2       120.0     360.0         1.0  3240492342  10-813-9695   \n",
       "4           0.0        93.6     360.0         1.0  1670564037  20-195-4948   \n",
       "\n",
       "                                              AGT_ID Prop_Area CPL_Status  \n",
       "0  Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10_6...     Urban          Y  \n",
       "1  Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_2)...     Urban          Y  \n",
       "2  Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKi...     Urban          Y  \n",
       "3  Mozilla/5.0 (X11; Linux i686) AppleWebKit/534....     Urban          Y  \n",
       "4  Mozilla/5.0 (Windows NT 6.0) AppleWebKit/534.2...     Urban          N  "
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data.head()"
   ]
  },
  {
   "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}