{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature Selection for Uplift Trees by Zhao et al. (2020)\n",
    "    \n",
    "This notebook includes two sections:  \n",
    "- **Feature selection**: demonstrate how to use Filter methods to select the most important numeric features\n",
    "- **Performance evaluation**: evaluate the AUUC performance with top features dataset\n",
    "  \n",
    "*(Paper reference: [Zhao, Zhenyu, et al. \"Feature Selection Methods for Uplift Modeling.\" arXiv preprint arXiv:2005.03447 (2020).](https://arxiv.org/abs/2005.03447))*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:16.975121Z",
     "start_time": "2021-11-29T22:45:16.749881Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:18.094794Z",
     "start_time": "2021-11-29T22:45:16.976822Z"
    }
   },
   "outputs": [],
   "source": [
    "from causalml.dataset import make_uplift_classification"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Import FilterSelect class for Filter methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:18.103040Z",
     "start_time": "2021-11-29T22:45:18.097362Z"
    }
   },
   "outputs": [],
   "source": [
    "from causalml.feature_selection.filters import FilterSelect"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:18.109523Z",
     "start_time": "2021-11-29T22:45:18.105177Z"
    }
   },
   "outputs": [],
   "source": [
    "from causalml.inference.tree import UpliftRandomForestClassifier\n",
    "from causalml.inference.meta import BaseXRegressor, BaseRRegressor, BaseSRegressor, BaseTRegressor\n",
    "from causalml.metrics import plot_gain, auuc_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:18.112934Z",
     "start_time": "2021-11-29T22:45:18.111213Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:18.116303Z",
     "start_time": "2021-11-29T22:45:18.114315Z"
    }
   },
   "outputs": [],
   "source": [
    "import logging\n",
    "\n",
    "logger = logging.getLogger('causalml')\n",
    "logging.basicConfig(level=logging.INFO)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generate dataset\n",
    "\n",
    "Generate synthetic data using the built-in function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:18.120534Z",
     "start_time": "2021-11-29T22:45:18.117882Z"
    }
   },
   "outputs": [],
   "source": [
    "# define parameters for simulation\n",
    "\n",
    "y_name = 'conversion'\n",
    "treatment_group_keys = ['control', 'treatment1']\n",
    "n = 10000\n",
    "n_classification_features = 50\n",
    "n_classification_informative = 10\n",
    "n_classification_repeated = 0\n",
    "n_uplift_increase_dict = {'treatment1': 8}\n",
    "n_uplift_decrease_dict = {'treatment1': 4}\n",
    "delta_uplift_increase_dict = {'treatment1': 0.1}\n",
    "delta_uplift_decrease_dict = {'treatment1': -0.1}\n",
    "\n",
    "random_seed = 20200808"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:19.388449Z",
     "start_time": "2021-11-29T22:45:18.123271Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:numexpr.utils:Note: NumExpr detected 12 cores but \"NUMEXPR_MAX_THREADS\" not set, so enforcing safe limit of 8.\n",
      "INFO:numexpr.utils:NumExpr defaulting to 8 threads.\n"
     ]
    }
   ],
   "source": [
    "df, X_names = make_uplift_classification(\n",
    "    treatment_name=treatment_group_keys,\n",
    "    y_name=y_name,\n",
    "    n_samples=n,\n",
    "    n_classification_features=n_classification_features,\n",
    "    n_classification_informative=n_classification_informative,\n",
    "    n_classification_repeated=n_classification_repeated,\n",
    "    n_uplift_increase_dict=n_uplift_increase_dict,\n",
    "    n_uplift_decrease_dict=n_uplift_decrease_dict,\n",
    "    delta_uplift_increase_dict = delta_uplift_increase_dict, \n",
    "    delta_uplift_decrease_dict = delta_uplift_decrease_dict,\n",
    "    random_seed=random_seed\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:19.417996Z",
     "start_time": "2021-11-29T22:45:19.391936Z"
    }
   },
   "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>treatment_group_key</th>\n",
       "      <th>x1_informative</th>\n",
       "      <th>x2_informative</th>\n",
       "      <th>x3_informative</th>\n",
       "      <th>x4_informative</th>\n",
       "      <th>x5_informative</th>\n",
       "      <th>x6_informative</th>\n",
       "      <th>x7_informative</th>\n",
       "      <th>x8_informative</th>\n",
       "      <th>x9_informative</th>\n",
       "      <th>...</th>\n",
       "      <th>x56_uplift_increase</th>\n",
       "      <th>x57_uplift_increase</th>\n",
       "      <th>x58_uplift_increase</th>\n",
       "      <th>x59_increase_mix</th>\n",
       "      <th>x60_uplift_decrease</th>\n",
       "      <th>x61_uplift_decrease</th>\n",
       "      <th>x62_uplift_decrease</th>\n",
       "      <th>x63_uplift_decrease</th>\n",
       "      <th>conversion</th>\n",
       "      <th>treatment_effect</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>treatment1</td>\n",
       "      <td>-4.004496</td>\n",
       "      <td>-1.250351</td>\n",
       "      <td>-2.800557</td>\n",
       "      <td>-0.368288</td>\n",
       "      <td>-0.115549</td>\n",
       "      <td>-2.492826</td>\n",
       "      <td>0.369516</td>\n",
       "      <td>0.290526</td>\n",
       "      <td>0.465153</td>\n",
       "      <td>...</td>\n",
       "      <td>0.496144</td>\n",
       "      <td>1.847680</td>\n",
       "      <td>-0.337894</td>\n",
       "      <td>-0.672058</td>\n",
       "      <td>1.180352</td>\n",
       "      <td>0.778013</td>\n",
       "      <td>0.931000</td>\n",
       "      <td>2.947160</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>treatment1</td>\n",
       "      <td>-3.170028</td>\n",
       "      <td>-0.135293</td>\n",
       "      <td>1.484246</td>\n",
       "      <td>-2.131584</td>\n",
       "      <td>-0.760103</td>\n",
       "      <td>1.764765</td>\n",
       "      <td>0.972124</td>\n",
       "      <td>1.407131</td>\n",
       "      <td>-1.027603</td>\n",
       "      <td>...</td>\n",
       "      <td>0.574955</td>\n",
       "      <td>3.578138</td>\n",
       "      <td>0.678118</td>\n",
       "      <td>-0.545227</td>\n",
       "      <td>-0.143942</td>\n",
       "      <td>-0.015188</td>\n",
       "      <td>1.189643</td>\n",
       "      <td>1.943692</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>treatment1</td>\n",
       "      <td>-0.763386</td>\n",
       "      <td>-0.785612</td>\n",
       "      <td>1.218781</td>\n",
       "      <td>-0.725835</td>\n",
       "      <td>1.044489</td>\n",
       "      <td>-1.521071</td>\n",
       "      <td>-2.266684</td>\n",
       "      <td>-1.614818</td>\n",
       "      <td>-0.113647</td>\n",
       "      <td>...</td>\n",
       "      <td>0.985076</td>\n",
       "      <td>1.079181</td>\n",
       "      <td>0.578092</td>\n",
       "      <td>0.574370</td>\n",
       "      <td>-0.477429</td>\n",
       "      <td>0.679070</td>\n",
       "      <td>1.650897</td>\n",
       "      <td>2.768897</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>control</td>\n",
       "      <td>0.887727</td>\n",
       "      <td>0.049095</td>\n",
       "      <td>-2.242776</td>\n",
       "      <td>1.530966</td>\n",
       "      <td>0.392623</td>\n",
       "      <td>-0.203071</td>\n",
       "      <td>-0.549329</td>\n",
       "      <td>0.107296</td>\n",
       "      <td>-0.542277</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.175352</td>\n",
       "      <td>0.683330</td>\n",
       "      <td>0.567545</td>\n",
       "      <td>0.349622</td>\n",
       "      <td>-0.789203</td>\n",
       "      <td>2.315184</td>\n",
       "      <td>0.658607</td>\n",
       "      <td>1.734836</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>control</td>\n",
       "      <td>-1.672922</td>\n",
       "      <td>-1.156145</td>\n",
       "      <td>3.871476</td>\n",
       "      <td>-1.883713</td>\n",
       "      <td>-0.220122</td>\n",
       "      <td>-4.615669</td>\n",
       "      <td>0.141980</td>\n",
       "      <td>-0.933756</td>\n",
       "      <td>-0.765592</td>\n",
       "      <td>...</td>\n",
       "      <td>0.485798</td>\n",
       "      <td>-0.355315</td>\n",
       "      <td>0.982488</td>\n",
       "      <td>-0.007260</td>\n",
       "      <td>2.895155</td>\n",
       "      <td>0.261848</td>\n",
       "      <td>-1.337001</td>\n",
       "      <td>-0.639983</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 66 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  treatment_group_key  x1_informative  x2_informative  x3_informative  \\\n",
       "0          treatment1       -4.004496       -1.250351       -2.800557   \n",
       "1          treatment1       -3.170028       -0.135293        1.484246   \n",
       "2          treatment1       -0.763386       -0.785612        1.218781   \n",
       "3             control        0.887727        0.049095       -2.242776   \n",
       "4             control       -1.672922       -1.156145        3.871476   \n",
       "\n",
       "   x4_informative  x5_informative  x6_informative  x7_informative  \\\n",
       "0       -0.368288       -0.115549       -2.492826        0.369516   \n",
       "1       -2.131584       -0.760103        1.764765        0.972124   \n",
       "2       -0.725835        1.044489       -1.521071       -2.266684   \n",
       "3        1.530966        0.392623       -0.203071       -0.549329   \n",
       "4       -1.883713       -0.220122       -4.615669        0.141980   \n",
       "\n",
       "   x8_informative  x9_informative  ...  x56_uplift_increase  \\\n",
       "0        0.290526        0.465153  ...             0.496144   \n",
       "1        1.407131       -1.027603  ...             0.574955   \n",
       "2       -1.614818       -0.113647  ...             0.985076   \n",
       "3        0.107296       -0.542277  ...            -0.175352   \n",
       "4       -0.933756       -0.765592  ...             0.485798   \n",
       "\n",
       "   x57_uplift_increase  x58_uplift_increase  x59_increase_mix  \\\n",
       "0             1.847680            -0.337894         -0.672058   \n",
       "1             3.578138             0.678118         -0.545227   \n",
       "2             1.079181             0.578092          0.574370   \n",
       "3             0.683330             0.567545          0.349622   \n",
       "4            -0.355315             0.982488         -0.007260   \n",
       "\n",
       "   x60_uplift_decrease  x61_uplift_decrease  x62_uplift_decrease  \\\n",
       "0             1.180352             0.778013             0.931000   \n",
       "1            -0.143942            -0.015188             1.189643   \n",
       "2            -0.477429             0.679070             1.650897   \n",
       "3            -0.789203             2.315184             0.658607   \n",
       "4             2.895155             0.261848            -1.337001   \n",
       "\n",
       "   x63_uplift_decrease  conversion  treatment_effect  \n",
       "0             2.947160           0                 0  \n",
       "1             1.943692           1                 0  \n",
       "2             2.768897           1                 0  \n",
       "3             1.734836           0                 0  \n",
       "4            -0.639983           1                 0  \n",
       "\n",
       "[5 rows x 66 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:19.608438Z",
     "start_time": "2021-11-29T22:45:19.419633Z"
    }
   },
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>conversion</th>\n",
       "      <th>conversion</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>treatment_group_key</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>control</th>\n",
       "      <td>0.50180</td>\n",
       "      <td>10000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>treatment1</th>\n",
       "      <td>0.59750</td>\n",
       "      <td>10000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>0.54965</td>\n",
       "      <td>20000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          mean       size\n",
       "                    conversion conversion\n",
       "treatment_group_key                      \n",
       "control                0.50180      10000\n",
       "treatment1             0.59750      10000\n",
       "All                    0.54965      20000"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Look at the conversion rate and sample size in each group\n",
    "df.pivot_table(values='conversion',\n",
    "               index='treatment_group_key',\n",
    "               aggfunc=[np.mean, np.size],\n",
    "               margins=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:19.613010Z",
     "start_time": "2021-11-29T22:45:19.609718Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['x1_informative',\n",
       " 'x2_informative',\n",
       " 'x3_informative',\n",
       " 'x4_informative',\n",
       " 'x5_informative',\n",
       " 'x6_informative',\n",
       " 'x7_informative',\n",
       " 'x8_informative',\n",
       " 'x9_informative',\n",
       " 'x10_informative',\n",
       " 'x11_irrelevant',\n",
       " 'x12_irrelevant',\n",
       " 'x13_irrelevant',\n",
       " 'x14_irrelevant',\n",
       " 'x15_irrelevant',\n",
       " 'x16_irrelevant',\n",
       " 'x17_irrelevant',\n",
       " 'x18_irrelevant',\n",
       " 'x19_irrelevant',\n",
       " 'x20_irrelevant',\n",
       " 'x21_irrelevant',\n",
       " 'x22_irrelevant',\n",
       " 'x23_irrelevant',\n",
       " 'x24_irrelevant',\n",
       " 'x25_irrelevant',\n",
       " 'x26_irrelevant',\n",
       " 'x27_irrelevant',\n",
       " 'x28_irrelevant',\n",
       " 'x29_irrelevant',\n",
       " 'x30_irrelevant',\n",
       " 'x31_irrelevant',\n",
       " 'x32_irrelevant',\n",
       " 'x33_irrelevant',\n",
       " 'x34_irrelevant',\n",
       " 'x35_irrelevant',\n",
       " 'x36_irrelevant',\n",
       " 'x37_irrelevant',\n",
       " 'x38_irrelevant',\n",
       " 'x39_irrelevant',\n",
       " 'x40_irrelevant',\n",
       " 'x41_irrelevant',\n",
       " 'x42_irrelevant',\n",
       " 'x43_irrelevant',\n",
       " 'x44_irrelevant',\n",
       " 'x45_irrelevant',\n",
       " 'x46_irrelevant',\n",
       " 'x47_irrelevant',\n",
       " 'x48_irrelevant',\n",
       " 'x49_irrelevant',\n",
       " 'x50_irrelevant',\n",
       " 'x51_uplift_increase',\n",
       " 'x52_uplift_increase',\n",
       " 'x53_uplift_increase',\n",
       " 'x54_uplift_increase',\n",
       " 'x55_uplift_increase',\n",
       " 'x56_uplift_increase',\n",
       " 'x57_uplift_increase',\n",
       " 'x58_uplift_increase',\n",
       " 'x59_increase_mix',\n",
       " 'x60_uplift_decrease',\n",
       " 'x61_uplift_decrease',\n",
       " 'x62_uplift_decrease',\n",
       " 'x63_uplift_decrease']"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_names"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature selection with Filter methods"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### method = F (F Filter)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:19.616362Z",
     "start_time": "2021-11-29T22:45:19.614534Z"
    }
   },
   "outputs": [],
   "source": [
    "filter_method = FilterSelect() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:22.010814Z",
     "start_time": "2021-11-29T22:45:19.617574Z"
    }
   },
   "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>method</th>\n",
       "      <th>feature</th>\n",
       "      <th>rank</th>\n",
       "      <th>score</th>\n",
       "      <th>p_value</th>\n",
       "      <th>misc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x53_uplift_increase</td>\n",
       "      <td>1.0</td>\n",
       "      <td>190.321410</td>\n",
       "      <td>4.262512e-43</td>\n",
       "      <td>df_num: 1.0, df_denom: 19996.0, order:1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x57_uplift_increase</td>\n",
       "      <td>2.0</td>\n",
       "      <td>127.136380</td>\n",
       "      <td>2.127676e-29</td>\n",
       "      <td>df_num: 1.0, df_denom: 19996.0, order:1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x3_informative</td>\n",
       "      <td>3.0</td>\n",
       "      <td>66.273458</td>\n",
       "      <td>4.152970e-16</td>\n",
       "      <td>df_num: 1.0, df_denom: 19996.0, order:1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x4_informative</td>\n",
       "      <td>4.0</td>\n",
       "      <td>59.407590</td>\n",
       "      <td>1.341417e-14</td>\n",
       "      <td>df_num: 1.0, df_denom: 19996.0, order:1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x62_uplift_decrease</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3.957507</td>\n",
       "      <td>4.667636e-02</td>\n",
       "      <td>df_num: 1.0, df_denom: 19996.0, order:1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     method              feature  rank       score       p_value  \\\n",
       "0  F filter  x53_uplift_increase   1.0  190.321410  4.262512e-43   \n",
       "0  F filter  x57_uplift_increase   2.0  127.136380  2.127676e-29   \n",
       "0  F filter       x3_informative   3.0   66.273458  4.152970e-16   \n",
       "0  F filter       x4_informative   4.0   59.407590  1.341417e-14   \n",
       "0  F filter  x62_uplift_decrease   5.0    3.957507  4.667636e-02   \n",
       "\n",
       "                                      misc  \n",
       "0  df_num: 1.0, df_denom: 19996.0, order:1  \n",
       "0  df_num: 1.0, df_denom: 19996.0, order:1  \n",
       "0  df_num: 1.0, df_denom: 19996.0, order:1  \n",
       "0  df_num: 1.0, df_denom: 19996.0, order:1  \n",
       "0  df_num: 1.0, df_denom: 19996.0, order:1  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# F Filter with order 1\n",
    "method = 'F'\n",
    "f_imp = filter_method.get_importance(df, X_names, y_name, method, \n",
    "                      treatment_group = 'treatment1')\n",
    "f_imp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>method</th>\n",
       "      <th>feature</th>\n",
       "      <th>rank</th>\n",
       "      <th>score</th>\n",
       "      <th>p_value</th>\n",
       "      <th>misc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x53_uplift_increase</td>\n",
       "      <td>1.0</td>\n",
       "      <td>107.368286</td>\n",
       "      <td>4.160720e-47</td>\n",
       "      <td>df_num: 2.0, df_denom: 19994.0, order:2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x57_uplift_increase</td>\n",
       "      <td>2.0</td>\n",
       "      <td>70.138050</td>\n",
       "      <td>4.423736e-31</td>\n",
       "      <td>df_num: 2.0, df_denom: 19994.0, order:2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x3_informative</td>\n",
       "      <td>3.0</td>\n",
       "      <td>36.499465</td>\n",
       "      <td>1.504356e-16</td>\n",
       "      <td>df_num: 2.0, df_denom: 19994.0, order:2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x4_informative</td>\n",
       "      <td>4.0</td>\n",
       "      <td>31.780547</td>\n",
       "      <td>1.658731e-14</td>\n",
       "      <td>df_num: 2.0, df_denom: 19994.0, order:2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x55_uplift_increase</td>\n",
       "      <td>5.0</td>\n",
       "      <td>27.494904</td>\n",
       "      <td>1.189886e-12</td>\n",
       "      <td>df_num: 2.0, df_denom: 19994.0, order:2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     method              feature  rank       score       p_value  \\\n",
       "0  F filter  x53_uplift_increase   1.0  107.368286  4.160720e-47   \n",
       "0  F filter  x57_uplift_increase   2.0   70.138050  4.423736e-31   \n",
       "0  F filter       x3_informative   3.0   36.499465  1.504356e-16   \n",
       "0  F filter       x4_informative   4.0   31.780547  1.658731e-14   \n",
       "0  F filter  x55_uplift_increase   5.0   27.494904  1.189886e-12   \n",
       "\n",
       "                                      misc  \n",
       "0  df_num: 2.0, df_denom: 19994.0, order:2  \n",
       "0  df_num: 2.0, df_denom: 19994.0, order:2  \n",
       "0  df_num: 2.0, df_denom: 19994.0, order:2  \n",
       "0  df_num: 2.0, df_denom: 19994.0, order:2  \n",
       "0  df_num: 2.0, df_denom: 19994.0, order:2  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# F Filter with order 2\n",
    "method = 'F'\n",
    "f_imp = filter_method.get_importance(df, X_names, y_name, method, \n",
    "                      treatment_group = 'treatment1', order=2)\n",
    "f_imp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>method</th>\n",
       "      <th>feature</th>\n",
       "      <th>rank</th>\n",
       "      <th>score</th>\n",
       "      <th>p_value</th>\n",
       "      <th>misc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x53_uplift_increase</td>\n",
       "      <td>1.0</td>\n",
       "      <td>72.064224</td>\n",
       "      <td>2.373628e-46</td>\n",
       "      <td>df_num: 3.0, df_denom: 19992.0, order:3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x57_uplift_increase</td>\n",
       "      <td>2.0</td>\n",
       "      <td>46.841718</td>\n",
       "      <td>3.710784e-30</td>\n",
       "      <td>df_num: 3.0, df_denom: 19992.0, order:3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x3_informative</td>\n",
       "      <td>3.0</td>\n",
       "      <td>24.089980</td>\n",
       "      <td>1.484634e-15</td>\n",
       "      <td>df_num: 3.0, df_denom: 19992.0, order:3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x4_informative</td>\n",
       "      <td>4.0</td>\n",
       "      <td>23.097310</td>\n",
       "      <td>6.414267e-15</td>\n",
       "      <td>df_num: 3.0, df_denom: 19992.0, order:3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>F filter</td>\n",
       "      <td>x55_uplift_increase</td>\n",
       "      <td>5.0</td>\n",
       "      <td>18.072880</td>\n",
       "      <td>1.044117e-11</td>\n",
       "      <td>df_num: 3.0, df_denom: 19992.0, order:3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     method              feature  rank      score       p_value  \\\n",
       "0  F filter  x53_uplift_increase   1.0  72.064224  2.373628e-46   \n",
       "0  F filter  x57_uplift_increase   2.0  46.841718  3.710784e-30   \n",
       "0  F filter       x3_informative   3.0  24.089980  1.484634e-15   \n",
       "0  F filter       x4_informative   4.0  23.097310  6.414267e-15   \n",
       "0  F filter  x55_uplift_increase   5.0  18.072880  1.044117e-11   \n",
       "\n",
       "                                      misc  \n",
       "0  df_num: 3.0, df_denom: 19992.0, order:3  \n",
       "0  df_num: 3.0, df_denom: 19992.0, order:3  \n",
       "0  df_num: 3.0, df_denom: 19992.0, order:3  \n",
       "0  df_num: 3.0, df_denom: 19992.0, order:3  \n",
       "0  df_num: 3.0, df_denom: 19992.0, order:3  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# F Filter with order 3\n",
    "method = 'F'\n",
    "f_imp = filter_method.get_importance(df, X_names, y_name, method, \n",
    "                      treatment_group = 'treatment1', order=3)\n",
    "f_imp.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### method = LR (likelihood ratio test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:33.978540Z",
     "start_time": "2021-11-29T22:45:22.012629Z"
    }
   },
   "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>method</th>\n",
       "      <th>feature</th>\n",
       "      <th>rank</th>\n",
       "      <th>score</th>\n",
       "      <th>p_value</th>\n",
       "      <th>misc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x53_uplift_increase</td>\n",
       "      <td>1.0</td>\n",
       "      <td>203.811674</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>df: 1, order: 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x57_uplift_increase</td>\n",
       "      <td>2.0</td>\n",
       "      <td>133.175328</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>df: 1, order: 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x3_informative</td>\n",
       "      <td>3.0</td>\n",
       "      <td>64.366711</td>\n",
       "      <td>9.992007e-16</td>\n",
       "      <td>df: 1, order: 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x4_informative</td>\n",
       "      <td>4.0</td>\n",
       "      <td>52.389798</td>\n",
       "      <td>4.550804e-13</td>\n",
       "      <td>df: 1, order: 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x62_uplift_decrease</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.064347</td>\n",
       "      <td>4.379760e-02</td>\n",
       "      <td>df: 1, order: 1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      method              feature  rank       score       p_value  \\\n",
       "0  LR filter  x53_uplift_increase   1.0  203.811674  0.000000e+00   \n",
       "0  LR filter  x57_uplift_increase   2.0  133.175328  0.000000e+00   \n",
       "0  LR filter       x3_informative   3.0   64.366711  9.992007e-16   \n",
       "0  LR filter       x4_informative   4.0   52.389798  4.550804e-13   \n",
       "0  LR filter  x62_uplift_decrease   5.0    4.064347  4.379760e-02   \n",
       "\n",
       "              misc  \n",
       "0  df: 1, order: 1  \n",
       "0  df: 1, order: 1  \n",
       "0  df: 1, order: 1  \n",
       "0  df: 1, order: 1  \n",
       "0  df: 1, order: 1  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# LR Filter with order 1\n",
    "method = 'LR'\n",
    "lr_imp = filter_method.get_importance(df, X_names, y_name, method, \n",
    "                      treatment_group = 'treatment1')\n",
    "lr_imp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>method</th>\n",
       "      <th>feature</th>\n",
       "      <th>rank</th>\n",
       "      <th>score</th>\n",
       "      <th>p_value</th>\n",
       "      <th>misc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x53_uplift_increase</td>\n",
       "      <td>1.0</td>\n",
       "      <td>277.639095</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>df: 2, order: 2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x57_uplift_increase</td>\n",
       "      <td>2.0</td>\n",
       "      <td>156.134112</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>df: 2, order: 2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x55_uplift_increase</td>\n",
       "      <td>3.0</td>\n",
       "      <td>71.478979</td>\n",
       "      <td>3.330669e-16</td>\n",
       "      <td>df: 2, order: 2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x3_informative</td>\n",
       "      <td>4.0</td>\n",
       "      <td>44.938973</td>\n",
       "      <td>1.744319e-10</td>\n",
       "      <td>df: 2, order: 2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x4_informative</td>\n",
       "      <td>5.0</td>\n",
       "      <td>29.179971</td>\n",
       "      <td>4.609458e-07</td>\n",
       "      <td>df: 2, order: 2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      method              feature  rank       score       p_value  \\\n",
       "0  LR filter  x53_uplift_increase   1.0  277.639095  0.000000e+00   \n",
       "0  LR filter  x57_uplift_increase   2.0  156.134112  0.000000e+00   \n",
       "0  LR filter  x55_uplift_increase   3.0   71.478979  3.330669e-16   \n",
       "0  LR filter       x3_informative   4.0   44.938973  1.744319e-10   \n",
       "0  LR filter       x4_informative   5.0   29.179971  4.609458e-07   \n",
       "\n",
       "              misc  \n",
       "0  df: 2, order: 2  \n",
       "0  df: 2, order: 2  \n",
       "0  df: 2, order: 2  \n",
       "0  df: 2, order: 2  \n",
       "0  df: 2, order: 2  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# LR Filter with order 2\n",
    "method = 'LR'\n",
    "lr_imp = filter_method.get_importance(df, X_names, y_name, method, \n",
    "                      treatment_group = 'treatment1',order=2)\n",
    "lr_imp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>method</th>\n",
       "      <th>feature</th>\n",
       "      <th>rank</th>\n",
       "      <th>score</th>\n",
       "      <th>p_value</th>\n",
       "      <th>misc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x53_uplift_increase</td>\n",
       "      <td>1.0</td>\n",
       "      <td>290.389201</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>df: 3, order: 3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x57_uplift_increase</td>\n",
       "      <td>2.0</td>\n",
       "      <td>153.942614</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>df: 3, order: 3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x55_uplift_increase</td>\n",
       "      <td>3.0</td>\n",
       "      <td>70.626667</td>\n",
       "      <td>3.108624e-15</td>\n",
       "      <td>df: 3, order: 3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x3_informative</td>\n",
       "      <td>4.0</td>\n",
       "      <td>45.477851</td>\n",
       "      <td>7.323235e-10</td>\n",
       "      <td>df: 3, order: 3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR filter</td>\n",
       "      <td>x4_informative</td>\n",
       "      <td>5.0</td>\n",
       "      <td>30.466528</td>\n",
       "      <td>1.100881e-06</td>\n",
       "      <td>df: 3, order: 3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      method              feature  rank       score       p_value  \\\n",
       "0  LR filter  x53_uplift_increase   1.0  290.389201  0.000000e+00   \n",
       "0  LR filter  x57_uplift_increase   2.0  153.942614  0.000000e+00   \n",
       "0  LR filter  x55_uplift_increase   3.0   70.626667  3.108624e-15   \n",
       "0  LR filter       x3_informative   4.0   45.477851  7.323235e-10   \n",
       "0  LR filter       x4_informative   5.0   30.466528  1.100881e-06   \n",
       "\n",
       "              misc  \n",
       "0  df: 3, order: 3  \n",
       "0  df: 3, order: 3  \n",
       "0  df: 3, order: 3  \n",
       "0  df: 3, order: 3  \n",
       "0  df: 3, order: 3  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# LR Filter with order 3\n",
    "method = 'LR'\n",
    "lr_imp = filter_method.get_importance(df, X_names, y_name, method, \n",
    "                      treatment_group = 'treatment1',order=3)\n",
    "lr_imp.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### method = KL (KL divergence)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:46.414304Z",
     "start_time": "2021-11-29T22:45:33.981158Z"
    }
   },
   "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>method</th>\n",
       "      <th>feature</th>\n",
       "      <th>rank</th>\n",
       "      <th>score</th>\n",
       "      <th>p_value</th>\n",
       "      <th>misc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KL filter</td>\n",
       "      <td>x53_uplift_increase</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.022997</td>\n",
       "      <td>None</td>\n",
       "      <td>number_of_bins: 10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KL filter</td>\n",
       "      <td>x57_uplift_increase</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.014884</td>\n",
       "      <td>None</td>\n",
       "      <td>number_of_bins: 10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KL filter</td>\n",
       "      <td>x4_informative</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.012103</td>\n",
       "      <td>None</td>\n",
       "      <td>number_of_bins: 10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KL filter</td>\n",
       "      <td>x3_informative</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.010179</td>\n",
       "      <td>None</td>\n",
       "      <td>number_of_bins: 10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KL filter</td>\n",
       "      <td>x55_uplift_increase</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.003836</td>\n",
       "      <td>None</td>\n",
       "      <td>number_of_bins: 10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      method              feature  rank     score p_value                misc\n",
       "0  KL filter  x53_uplift_increase   1.0  0.022997    None  number_of_bins: 10\n",
       "0  KL filter  x57_uplift_increase   2.0  0.014884    None  number_of_bins: 10\n",
       "0  KL filter       x4_informative   3.0  0.012103    None  number_of_bins: 10\n",
       "0  KL filter       x3_informative   4.0  0.010179    None  number_of_bins: 10\n",
       "0  KL filter  x55_uplift_increase   5.0  0.003836    None  number_of_bins: 10"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "method = 'KL'\n",
    "kl_imp = filter_method.get_importance(df, X_names, y_name, method, \n",
    "                      treatment_group = 'treatment1',\n",
    "                      n_bins=10)\n",
    "kl_imp.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We found all these 3 filter methods were able to rank most of the **informative** and **uplift increase** features on the top."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Performance evaluation  \n",
    "\n",
    "Evaluate the AUUC (Area Under the Uplift Curve) score with several uplift models when using top features dataset "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:46.497534Z",
     "start_time": "2021-11-29T22:45:46.415884Z"
    }
   },
   "outputs": [],
   "source": [
    "# train test split\n",
    "df_train, df_test = train_test_split(df, test_size=0.2, random_state=111)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:46.510551Z",
     "start_time": "2021-11-29T22:45:46.499845Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 0 1 1 0 1 1 0 0 0]\n",
      "18998       control\n",
      "11536       control\n",
      "8552     treatment1\n",
      "2652     treatment1\n",
      "19671       control\n",
      "13244    treatment1\n",
      "3075     treatment1\n",
      "8746        control\n",
      "18530       control\n",
      "5066        control\n",
      "Name: treatment_group_key, dtype: object\n"
     ]
    }
   ],
   "source": [
    "# convert treatment column to 1 (treatment1) and 0 (control)\n",
    "treatments = np.where((df_test['treatment_group_key']=='treatment1'), 1, 0)\n",
    "print(treatments[:10])\n",
    "print(df_test['treatment_group_key'][:10])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Uplift RandomForest Classfier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:45:46.514976Z",
     "start_time": "2021-11-29T22:45:46.512507Z"
    }
   },
   "outputs": [],
   "source": [
    "uplift_model = UpliftRandomForestClassifier(control_name='control', max_depth=8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:48:04.459656Z",
     "start_time": "2021-11-29T22:45:46.516815Z"
    }
   },
   "outputs": [],
   "source": [
    "# using all features\n",
    "features = X_names \n",
    "uplift_model.fit(X = df_train[features].values, \n",
    "                 treatment = df_train['treatment_group_key'].values,\n",
    "                 y = df_train[y_name].values)\n",
    "y_preds = uplift_model.predict(df_test[features].values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Select top N features based on KL filter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:48:04.466886Z",
     "start_time": "2021-11-29T22:48:04.461249Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    x53_uplift_increase\n",
      "0    x57_uplift_increase\n",
      "0         x4_informative\n",
      "0         x3_informative\n",
      "0    x55_uplift_increase\n",
      "0         x1_informative\n",
      "0    x56_uplift_increase\n",
      "0    x51_uplift_increase\n",
      "0         x38_irrelevant\n",
      "0    x58_uplift_increase\n",
      "Name: feature, dtype: object\n"
     ]
    }
   ],
   "source": [
    "top_n = 10\n",
    "top_10_features = kl_imp['feature'][:top_n]\n",
    "print(top_10_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:48:04.472658Z",
     "start_time": "2021-11-29T22:48:04.469577Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    x53_uplift_increase\n",
      "0    x57_uplift_increase\n",
      "0         x4_informative\n",
      "0         x3_informative\n",
      "0    x55_uplift_increase\n",
      "0         x1_informative\n",
      "0    x56_uplift_increase\n",
      "0    x51_uplift_increase\n",
      "0         x38_irrelevant\n",
      "0    x58_uplift_increase\n",
      "0         x48_irrelevant\n",
      "0         x15_irrelevant\n",
      "0         x27_irrelevant\n",
      "0    x62_uplift_decrease\n",
      "0         x23_irrelevant\n",
      "Name: feature, dtype: object\n"
     ]
    }
   ],
   "source": [
    "top_n = 15\n",
    "top_15_features = kl_imp['feature'][:top_n]\n",
    "print(top_15_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:48:04.477660Z",
     "start_time": "2021-11-29T22:48:04.474508Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    x53_uplift_increase\n",
      "0    x57_uplift_increase\n",
      "0         x4_informative\n",
      "0         x3_informative\n",
      "0    x55_uplift_increase\n",
      "0         x1_informative\n",
      "0    x56_uplift_increase\n",
      "0    x51_uplift_increase\n",
      "0         x38_irrelevant\n",
      "0    x58_uplift_increase\n",
      "0         x48_irrelevant\n",
      "0         x15_irrelevant\n",
      "0         x27_irrelevant\n",
      "0    x62_uplift_decrease\n",
      "0         x23_irrelevant\n",
      "0         x29_irrelevant\n",
      "0         x6_informative\n",
      "0         x45_irrelevant\n",
      "0         x40_irrelevant\n",
      "0         x25_irrelevant\n",
      "Name: feature, dtype: object\n"
     ]
    }
   ],
   "source": [
    "top_n = 20\n",
    "top_20_features = kl_imp['feature'][:top_n]\n",
    "print(top_20_features)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Train the Uplift model again with top N features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:49:14.380345Z",
     "start_time": "2021-11-29T22:48:04.483465Z"
    },
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# using top 10 features\n",
    "features = top_10_features \n",
    "\n",
    "uplift_model.fit(X = df_train[features].values, \n",
    "                 treatment = df_train['treatment_group_key'].values,\n",
    "                 y = df_train[y_name].values)\n",
    "y_preds_t10 = uplift_model.predict(df_test[features].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:50:37.407711Z",
     "start_time": "2021-11-29T22:49:14.383261Z"
    }
   },
   "outputs": [],
   "source": [
    "# using top 15 features\n",
    "features = top_15_features \n",
    "\n",
    "uplift_model.fit(X = df_train[features].values, \n",
    "                 treatment = df_train['treatment_group_key'].values,\n",
    "                 y = df_train[y_name].values)\n",
    "y_preds_t15 = uplift_model.predict(df_test[features].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:52:02.337981Z",
     "start_time": "2021-11-29T22:50:37.409572Z"
    }
   },
   "outputs": [],
   "source": [
    "# using top 20 features\n",
    "features = top_20_features\n",
    "\n",
    "uplift_model.fit(X = df_train[features].values, \n",
    "                 treatment = df_train['treatment_group_key'].values,\n",
    "                 y = df_train[y_name].values)\n",
    "y_preds_t20 = uplift_model.predict(df_test[features].values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Print results for Uplift model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:52:04.153825Z",
     "start_time": "2021-11-29T22:52:02.340008Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_preds = pd.DataFrame([y_preds.ravel(), \n",
    "                         y_preds_t10.ravel(),\n",
    "                         y_preds_t15.ravel(),\n",
    "                         y_preds_t20.ravel(),\n",
    "                         treatments,\n",
    "                         df_test[y_name].ravel()],\n",
    "                        index=['All', 'Top 10', 'Top 15', 'Top 20', 'is_treated', y_name]).T\n",
    "\n",
    "plot_gain(df_preds, outcome_col=y_name, treatment_col='is_treated')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:52:04.355368Z",
     "start_time": "2021-11-29T22:52:04.155514Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "All       0.773405\n",
       "Top 10    0.841204\n",
       "Top 15    0.816100\n",
       "Top 20    0.816252\n",
       "Random    0.506801\n",
       "dtype: float64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auuc_score(df_preds, outcome_col=y_name, treatment_col='is_treated')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### R Learner as base and feed in Random Forest Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T22:52:04.359998Z",
     "start_time": "2021-11-29T22:52:04.356729Z"
    }
   },
   "outputs": [],
   "source": [
    "r_rf_learner = BaseRRegressor(\n",
    "    RandomForestRegressor(\n",
    "        n_estimators = 100,\n",
    "        max_depth = 8,\n",
    "        min_samples_leaf = 100\n",
    "    ), \n",
    "control_name='control') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T23:07:56.770197Z",
     "start_time": "2021-11-29T22:52:04.361299Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:causalml:Generating propensity score\n",
      "INFO:causalml:Calibrating propensity scores.\n",
      "INFO:causalml:generating out-of-fold CV outcome estimates\n",
      "INFO:causalml:training the treatment effect model for treatment1 with R-loss\n"
     ]
    }
   ],
   "source": [
    "# using all features\n",
    "features = X_names \n",
    "r_rf_learner.fit(X = df_train[features].values, \n",
    "                 treatment = df_train['treatment_group_key'].values,\n",
    "                 y = df_train[y_name].values)\n",
    "y_preds = r_rf_learner.predict(df_test[features].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T23:10:22.727884Z",
     "start_time": "2021-11-29T23:07:56.771842Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:causalml:Generating propensity score\n",
      "INFO:causalml:Calibrating propensity scores.\n",
      "INFO:causalml:generating out-of-fold CV outcome estimates\n",
      "INFO:causalml:training the treatment effect model for treatment1 with R-loss\n"
     ]
    }
   ],
   "source": [
    "# using top 10 features\n",
    "features = top_10_features \n",
    "r_rf_learner.fit(X = df_train[features].values, \n",
    "                 treatment = df_train['treatment_group_key'].values,\n",
    "                 y = df_train[y_name].values)\n",
    "y_preds_t10 = r_rf_learner.predict(df_test[features].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T23:14:02.389101Z",
     "start_time": "2021-11-29T23:10:22.729730Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:causalml:Generating propensity score\n",
      "INFO:causalml:Calibrating propensity scores.\n",
      "INFO:causalml:generating out-of-fold CV outcome estimates\n",
      "INFO:causalml:training the treatment effect model for treatment1 with R-loss\n"
     ]
    }
   ],
   "source": [
    "# using top 15 features\n",
    "features = top_15_features \n",
    "r_rf_learner.fit(X = df_train[features].values, \n",
    "                 treatment = df_train['treatment_group_key'].values,\n",
    "                 y = df_train[y_name].values)\n",
    "y_preds_t15 = r_rf_learner.predict(df_test[features].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T23:18:54.190818Z",
     "start_time": "2021-11-29T23:14:02.390709Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:causalml:Generating propensity score\n",
      "INFO:causalml:Calibrating propensity scores.\n",
      "INFO:causalml:generating out-of-fold CV outcome estimates\n",
      "INFO:causalml:training the treatment effect model for treatment1 with R-loss\n"
     ]
    }
   ],
   "source": [
    "# using top 20 features\n",
    "features = top_20_features \n",
    "r_rf_learner.fit(X = df_train[features].values, \n",
    "                 treatment = df_train['treatment_group_key'].values,\n",
    "                 y = df_train[y_name].values)\n",
    "y_preds_t20 = r_rf_learner.predict(df_test[features].values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Print results for R Learner"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T23:18:56.044272Z",
     "start_time": "2021-11-29T23:18:54.192388Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_preds = pd.DataFrame([y_preds.ravel(), \n",
    "                         y_preds_t10.ravel(),\n",
    "                         y_preds_t15.ravel(),\n",
    "                         y_preds_t20.ravel(),\n",
    "                         treatments,\n",
    "                         df_test[y_name].ravel()],\n",
    "                        index=['All', 'Top 10', 'Top 15', 'Top 20', 'is_treated', y_name]).T\n",
    "\n",
    "plot_gain(df_preds, outcome_col=y_name, treatment_col='is_treated')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T23:18:56.211023Z",
     "start_time": "2021-11-29T23:18:56.045507Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "All       0.859891\n",
       "Top 10    0.865159\n",
       "Top 15    0.872650\n",
       "Top 20    0.870669\n",
       "Random    0.506801\n",
       "dtype: float64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# print out AUUC score\n",
    "auuc_score(df_preds, outcome_col=y_name, treatment_col='is_treated')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "(a relatively smaller enhancement on the AUUC is observed in this R Learner case)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### S Learner as base and feed in Random Forest Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T23:18:56.214863Z",
     "start_time": "2021-11-29T23:18:56.212530Z"
    }
   },
   "outputs": [],
   "source": [
    "slearner_rf = BaseSRegressor(\n",
    "    RandomForestRegressor(\n",
    "        n_estimators = 100,\n",
    "        max_depth = 8,\n",
    "        min_samples_leaf = 100\n",
    "    ), \n",
    "    control_name='control')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T23:26:06.103012Z",
     "start_time": "2021-11-29T23:18:56.216262Z"
    }
   },
   "outputs": [],
   "source": [
    "# using all features\n",
    "features = X_names \n",
    "slearner_rf.fit(X = df_train[features].values, \n",
    "                treatment = df_train['treatment_group_key'].values,\n",
    "                y = df_train[y_name].values)\n",
    "y_preds = slearner_rf.predict(df_test[features].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T23:27:14.666207Z",
     "start_time": "2021-11-29T23:26:06.104715Z"
    }
   },
   "outputs": [],
   "source": [
    "# using top 10 features\n",
    "features = top_10_features \n",
    "slearner_rf.fit(X = df_train[features].values, \n",
    "                treatment = df_train['treatment_group_key'].values,\n",
    "                y = df_train[y_name].values)\n",
    "y_preds_t10 = slearner_rf.predict(df_test[features].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T23:28:56.526315Z",
     "start_time": "2021-11-29T23:27:14.667927Z"
    }
   },
   "outputs": [],
   "source": [
    "# using top 15 features\n",
    "features = top_15_features \n",
    "slearner_rf.fit(X = df_train[features].values, \n",
    "                treatment = df_train['treatment_group_key'].values,\n",
    "                y = df_train[y_name].values)\n",
    "y_preds_t15 = slearner_rf.predict(df_test[features].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T23:31:13.640303Z",
     "start_time": "2021-11-29T23:28:56.527765Z"
    }
   },
   "outputs": [],
   "source": [
    "# using top 20 features\n",
    "features = top_20_features \n",
    "slearner_rf.fit(X = df_train[features].values, \n",
    "                treatment = df_train['treatment_group_key'].values,\n",
    "                y = df_train[y_name].values)\n",
    "y_preds_t20 = slearner_rf.predict(df_test[features].values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Print results for S Learner"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T23:31:15.343770Z",
     "start_time": "2021-11-29T23:31:13.641802Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjcAAAINCAYAAADCyV5lAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAADQE0lEQVR4nOydd3gVVfrHP3N7ei+kAklI6KH3Ll3sil1xXVfWuqv+lLXs2l3brq6KXbFXLCi9907ohNACCSG959aZ+f0RuDeTm0BAICGcz/P4yLznzJkzyc3c75zzFqmsrExFIBAIBAKBoJWga+4JCAQCgUAgEJxNhLgRCAQCgUDQqhDiRiAQCAQCQatCiBuBQCAQCAStCiFuBAKBQCAQtCqEuBEIBAKBQNCqEOJGIBAIBAJBq6LZxM2LL75IcHCw5r8OHTq421VV5cUXXyQtLY3o6GgmTpzI7t27NWPY7XYeeeQR2rdvT0xMDNdffz25ubnn+1YEAoFAIBC0IJp15SYlJYXMzEz3f6tXr3a3vfHGG7z99tv8+9//ZvHixURERHDllVdSWVnp7jNt2jRmzZrFRx99xOzZs6msrGTy5MnIstwctyMQCAQCgaAF0KzixmAwEBUV5f4vPDwcqF21mT59Og8++CCXX345nTp1Yvr06VRVVfHDDz8AUF5ezueff84zzzzDiBEjSE9P57333mPnzp0sXbq0Ge9KIBAIBAJBc9Ks4ubQoUN07NiRbt26cccdd3Do0CEAsrOzyc/PZ+TIke6+Pj4+DBw4kHXr1gGQkZGB0+nU9ImLiyM1NdXdRyAQCAQCwcWHobku3Lt3b9555x1SUlIoKirilVdeYcyYMaxdu5b8/HwAIiIiNOdERESQl5cHQEFBAXq9nrCwMK8+BQUF5+cmBAKBQCAQtDiaTdyMHj1ac9y7d2/S09P56quv6NOnDwCSJGn6qKrqZatPU/oIBAKBQCBovbSYUHB/f3/S0tI4cOAAUVFRAF4rMEVFRe7VnMjISGRZpri4uNE+55OsrKzzfs3zjbjH1oG4x9aBuMfWgbjHc0OLETc2m42srCyioqJITEwkKiqKJUuWaNrXrFlDv379AEhPT8doNGr65ObmkpmZ6e4jEAgEAoHg4qPZtqWeeOIJxo0bR1xcnNvnpqamhhtuuAFJkpg6dSqvvfYaKSkpJCcn8+qrr+Ln58c111wDQFBQELfccgtPPfUUERERhISE8Pjjj9O5c2eGDx/eXLclEAgEAoGgmWk2cXP06FHuvPNOiouLCQ8Pp3fv3ixYsICEhAQAHnjgAaxWK4888ghlZWX06tWLmTNnEhAQ4B7jhRdeQK/XM2XKFGw2G0OHDuXdd99Fr9c3120JBAKBQCBoZppN3Hz88ccnbZckiWnTpjFt2rRG+1gsFl555RVeeeWVsz09gUAgEAgEFyjNJm5aIi6Xi+rq6jM612KxUF5efpZn1LJoSffo5+eHwSA+vgKBQCDwRnw7HMflclFZWUlwcPAZhZKbzWYsFss5mFnLoaXco6qqlJWVERAQIASOQCAQCLxoMdFSzU11dfUZCxvB+UWSJIKDg894lU0gEAgErRshbuoghM2Fg/hdCQQCgaAxhLgRCAQCgUDQqhDiRiAQCAQCQatCiJuLgODgYH755ZdGjwUCgUAgaE0IcdNK2Lp1K6GhoYwdO7a5pyIQCAQCQbMixE0r4bPPPuNPf/oTu3fvJjMzs7mnIxAIBAJBsyGShJyC4E9yz+v1yqbEnvY5VquV77//ntmzZ2O1Wvn888957rnnzsHsBAKBQCBo+YiVm1bAL7/8Qnx8PF26dGHy5Ml88803OJ3O5p6WQCAQCATNghA3rYDPPvuM66+/HoDBgwfj4+PD7Nmzm3lWAoFAcJZQVVCU5p6F4AJCiJsLnAMHDrBu3TquueYaoDa53XXXXcfnn3/ezDMTCASCP44u5wC+j92K39SJmN97AelYTnNPSXABIHxuTkFTfWBsNluz1F367LPPkGWZLl26uG2qqgKQk5NDXFzceZ+TQCAQnC3Mn7yG7tgRAIyr52NYuxDXwNE4LrsVNer0fRQFFwdi5eYCxuVy8fXXX/PPf/6TFStWuP9buXIlnTt35ssvv2zuKQoEglaIqqocq5HdL1LnCktBDvp9OzU2SVEwrpyH72O3YJ7xH3C5zukcBBcmYuXmAmbevHkUFxdz2223ERoaqmm7+uqr+eijj3jkkUeaaXYCgaA1UulUuHxuEZuLnHQJNfLz2DDCLfpzcq2wjJWNtkmKgnHxL6Ao2Kc8dMqxVFXl9W1VzD5sJciko0e4kfRwEz3CjMT66UW9ulaGWLm5gPn8888ZMmSIl7ABuOKKKzhy5AhLly49/xMTCAStlje3V7G5qDYac0eJkyc3VJybCznshG5fqzEpAcFe3QzLfkeXc+CUw322t4ZnN1ewqcjJ4qN2XttWxS2LS+jyfT5dvsvng91V53wlSnD+ECs3FzDffPNNo21t27alrKwMwP3/E9Q/FggEgqbgUlS+yKrW2L7eV8NfOvqRHm46q9cybF6Jweq5luoXSM1r32DYtALTt++hKysCQFIVTN+9j+3vLzU6ltWl8lJG4yIst0bmkbXlrD7m4H+Dg/E3ivf+Cx3xGxQIBAJBk5h7xEZejXdI9uMbys/6qodh2e+aY+eg0WC24Bo4GvttD2r7bl2Lbk9Go2N9tKeqwXnX56dDVi75rZB95do8YVJxPvotq5GK85s8f0HzIsSNQCAQCJrEjMzqBu2rjjn4/bDNy+6QT1/w1LgUdmdmY9i1WWN3DZ3o/rfcYxByShdNu/nb92rz4dSjwqHw+rYqAHxkO12qDnNrpJW7O/nRP9JEfXehPWUuRs4q5PdsKwD6nRvx/cft+Pz3H/g+dAOWt/6Jrp6Ts6DlIbalBAKBQHBKsitdLMy1N9r+1IZyxsRZMOklyuwKj60rY+ZBK/H+er69JIzkIOMpr3Go0sUV84qYsu0n+tSxy0kdUeLbewyShH3y3fg+d6/bpD+wG/2GZch9h2vGnL6rihK7Qoy9hLlbX6RTzVHYCEqbeOS0dPISu3F7fgLLrf7ucyqcKjctLmFGFzs3fvIMkq1W6EiqgmHDMgwbliEndcQxbjJy76GgE+sELQ3xGxEIBALBKfk8q4a66yIJ/np0dQKMDlTKfLinmmVH7Qz6uYBv9ltxKLC/QuYvy0tPuW1V7lC4fmExR8od3Ja3TNO2t8dYr/5KShdcvYZobOYfPtCEhpfYZN7aUQWqyjt7P64VNsfR5R3BuGQWCZ8+z6K5U5mb9ykW2eFut8gOOs14Bqm6YV8d/f7d+Lz9Lyz//QfUVJ303gTnHyFuBAKBQABAtVPhL8tL6P79Mf61sdy9reRUVL7Yq92S+nu3AG7v4KexPbOpnCvmFZFbI2vsm4qc/JrtvW11Apei8qelJewpczG2dBtxjlJ3W5XOzIBjXbhnZSklNu249mv/jFpn1USXn4tx6Sz38Rvbq6h0qlxduJ5Li7c0en1JVbgkcwH7sp6nnb0IVJW3sj6hR+WhRs85gWHrWnyfvQcpX2RObkkIcSMQCAQCAB5eW863+61kV8n8d3sV1y8spsqpMPeIjWNWj0Ouv0Hi6vY+TOsRQIDRs3xjk6Gx9ZnnNlfgUhpufXx9uXvL6468pZq2byMHUGXw4cusGq6cX4y9jh+P2iYB17CJmv7Gn2cglRaRVyPz/u5qgp3VvJE1o0n3H31sH9u2PsV/9n3G7ceWa9ocPYdg/b9XcXXv73We7mg2vk9PRb9z08kv4HSg37oOKfcQADW2KvLKDrLtwBoqa8qaNEdB0xA+NwKBQCBgQY6Nr/fVaGyLj9q5bG4RZr02wd21ST4EGHUEGOGhbgH8a5P31s2JM05IkaxyF19m1XBbqna156M9Vby3u5oQZxU3569kYr0Vlk/aDHf/e2uxk8/3VnNnR49/jOOK2zGsWoDkqF0Z0lWW4fPsPXw16gmscij/OfAV0c5yd39Vb8D6+P+QqivQ78nAsH4pusI8d7tPTTn31czXzGGPTxump0/lroRIPh2awlr/vby+bTp9Kj35daTqSiyvPoLjhr/ivOQq0OmwOmWOlReiVyqpOpKJbf43VDiqyPM3kRsRTLni8WEyGcyM7T2ZPqkjRULBs4AQNwKBQHCRU+5QeGBVaYNtJxL21eW2OttRd3fy5+PMag5XebaM4vz0TB8Swnf7a/g8q1YwGRQXPyzfxU01MmYdgMTmIgcrN5cwo2A9Vxeux6Jqr1UaGkNZQhqUe8b+z7Yqbung5xZcanAYzvGTMf3iWZ3RFedz/8xHsSZcxp31VoKck25CSeoIgNytH45Lb8Ly3gsYMlY3eP9VOjPXdnmQ3QdU3jmQXyvWjDGMSH+S9zM/4MYCz3mSomD+8i1Yv4QPe/Zle8F6fKjjj9NGDwTV/lvROmc7XHZmrf2MndmbuHLQnwj2D2twPoKmIcSNQCAQXOQ8uaGco03IAwO4yxacwGKQ+GhYKNcvLKbcoXBNex9e6hdMiLWU3oVzmbB7B2lVR0itycOkylCnosKg4/81RlWfofw8LoL0H45xwt0mt0bmy6wa7kjzCCzHZTejy96LIWON2xbiquGlA9pEp0pMIo5Lb9JexNcf2wPPYfzlM8w/f+o1hz+l/YXdfrUFiOtuqtn0Jm7t+Fe2+yfw/IFv0aGiAtvDfZgVWkpRwWJ8TnJvjXEgbyevzvwHXTtdx7U9R6ATkVhnhBA3AoFAcBGzJNfGZ3u121F/6ehHZrmLpUe9Q7+n1NtWAugTaSLrhmiqXSoBRh1UV+LzzF/xK87n2uN9qg069gdaKPIxoki1QkGVQEUi0CHTsdiKn6tWYKk6Ha7B4yjqOYwUXz23dfDjvd0eh+bXt1Vyc4ovphPbZQYjtvufxfzZGxqH4vrYpjwExgYyKet0OK+8HSUxBct7zyPZan8euaOuZzb9obF8PZLEKwmT2O0by8uH32Nekh8Hgi2NXr8+ekUlqtrJMV8zit5zDUmxsWPHZ6zJPsRzl00RGZPPACFuLmCCg4NP2n7DDTcwffr0s3rN3bt38+KLL7J161ays7N59NFHmTZtmle/Dz/8kDfffJP8/HzS0tJ48cUXGThw4Fmdi0Ag+GNUOhXuW1WmsaUGGXi6dxA6CaauKOXHg1Z3W4BR4qp2Da9HWO2VFJfnkee0oSybhWKsoiY+gBx/E4cDzRT6njzPjU5RSa1S6RbTjQ7DbsQUEQtZWQA82C2AT/dWYz++epNTLfPVvhpuryu09Abst/+dHVIwPZZ87jW+c8RlKB26nXQOcs9B1Lz4KYa1i1GiYgnqOZi3D1q5a3kpruPao1uokTs7+hHvp2fywmIcCqwJj+HN8DYYJYfXmCZZIbLaSaBDxs+hst0nmTEFe+lQXklUjRODCiVmPV90jORgiFZ4BVQu5+af4vnv+JG0DRBf16eD+GldwGRmZrr/PW/ePO6//36NzWJp+htEU7FarSQkJDBp0iSee+65BvvMnDmTxx57jNdee43+/fvz4Ycfcu2117J27Vri4+PP+pwEAsGZ8dSGcnKqPf4sOgneGhyCxVC7IvLBsBDi/fW8s7MKRYX/Dmy47tKmrOXMWjMDWfHkmKFj+GnNRdFJ7A6U2F21A8Pcf9KtXX/aB/cEoI2vnls7+PFBndWb17ZVcmNyndUbAEni3ojLSE314b29H2FUa+9NCQ7Hft1dTZqHGhqJc8L17uOr2vuSEmxkc6GDzqFGeoUb3Q6/bw4K4e4VpQxijpewkVSVAUerGHewjACnQrY5jOs6P8imwPbMNhxhbtFLGNQyAELtMvdm5LE6xp9ZySE49J6fcYfqH5j0awxvjUhmWMzZf6a3VoS4OQX+tw1vWr+zdL2qGUub3DcqKsr976CgIC/bJ598wptvvklOTg5xcXE8+OCD3Hbbbe724OBgXn75ZRYuXMiKFSsICwvjySefZPLkyY1es2fPnvTsWfvAef311xvs8/bbb3PjjTe6r/XKK6+waNEiPv74Y/75z382+f4EAsG5QVVVntlUwSeZ2u2ov3byp0+kZ/VAJ0n8q3cQ93fxx2KQ8DV4C5vVO+cxZ8NXZ3V+LtnJ5n0r2KZbS5lyOQM7jeXBrgHMyKzGcdw16EiVzDf7a7i1jnNzRpGDdQUO1rUZRrYlgn8e+pH0SAv6KQ+C75k/pbuGGuka6r3ydH2yL3tzd+M8mKWxH1ZS6H9Mz9UHFqNzKZR06seNCXeyyeYLwE7/eIanP8nynS8RVV0I1OZlGXy0isQKB2/0ikY+niHRLNkZ4Piaa+bfxTN9wpja+Wx929RyIG8Xew5vISmmM6nx6Wd17OZEiJtWyqxZs3jkkUd44YUXGDlyJIsWLeKhhx4iMjKS8ePHu/u9+OKLPPXUU7zwwgv8/PPP3H333XTo0IEePXqc0XUdDgcZGRncd999GvvIkSNZt27dH7ongUDwx5EVlYfWlPFpPT+bpEA9j/cMbPCc0PoFmAC1ppplWQtZtGVmE68sUahGU0gMDkx0DzPTP8qMhMK+3B0UVeR5neFSnCzc/AObspYxptdkbksM4PuDdhQkVHS8mqHjhmRfjMeFQF2/nGUhnXi2cw9+GhveaO6dP4qqqkRXzeZIHVuO2o4dgX/i6WsjqDH9DamqHFNYFD+7VF7YUsGXWTUk+Ot5dFQn/K5/m6rXp+Gfs999fnyVg8v3lTKzQ6jbFiXl0Fedz7T1Ewiz6LguyfeUc5MKjmL++m10WTuR+wzDftN9YPB85cuKzLyN37BmV23Y+5rd85ky9lHat+n0x38wLQAhblopb731FpMnT+auu2qXYpOTk8nIyOCNN97QiJtJkyYxZcoUAB5++GFWrFjB9OnTef/998/ousXFxciyTEREhMYeERFBQUHBGd6NQHBx4FRUvsqqId8qMybOoolKagxVVVmQY+f3w1ZMOonHegQQ1oAYgdpClnctL+XnQ1aN3d8g8f7QUHwMTcivUlOF5Y1/MMdxmMWJQZomvQrJpVbMLgWTrGL0DSBwxDXERSYTE9aO+UcV1uY7mJBgYWC0WXMPx0qPsP3gWrYfWEdZdZFm3NLKQr5d+hYW4JY6U6ysDuJvS/7Ks4M74FJUfjygFWx3dfR2fj6bZOZkcKRwn8YmRU1i3shIwi16wIBqrt1K8jFIPNsniGd6B9bJY+ND1u2PkZIQBw47ktOJVFXOoOfvI6uwhu0RHhHTU1pJjprEY+s6MjLWfHz8BlBVDMt+x/zVW0j247l/Fv+C6heA45o7gdrkgd8ue5sDebs0p67fs1iIG0HLJjMzk5tu0oY8DhgwgDlz5mhsffr08TqeP1+bwOpMqJ+ESlVVkZhKIDgFz26q4M0dtXlRXthSSc9wI3ek+XFVOx+vLSGnAl9mVfPWjip2l3l8XZbl2VkyKQK/er4x1U6FWxaXsLheBFSoWcePY8Lo0QQhBWD66i1+lnNYVU/YGCU9f8o4Smqpp8yC9f+mIXfu7T6+NBEuTfR2SJYkiTahCbQJTWBUj6vZtHcpC7f8iNXecBXyEwRI5eQc/oZeP/6F3hFG95YVQOdAOyPlX3Bk6zHGTkAyNE3oKDVHcR1bhOTTBkP0SCSp4UglRVVYuPkHjS01Pp1nR/U66fgNPgfNPmD2qY0gC43AOf56Js+awZEAE2UWz9f0aL7nZ/uf+Md6M+8PDfUaRiorxvzxKxi2rvVqM/72Ja4ufTgaFcpXi96gtKrQq8++ozuQFRd63YUvDS78OzjHNNUHxmaznRMH3j9CQ39E51pghIWFodfrvVZpioqKvFZzBAKBh2KbzPu7tQUYNxc52byyjCfWl9M30oROktBJtf4Z645ZKHSU1emtoENhbzk8uq6ctwaHuFscssrNi0tYUk/YxPrqmTk2jNTgU1fsBtDv2Mj6/StZlapNMGdxKfx52zHal3vGd3XtqxE2TUWv09M3bRRd2vVj5pJP2Ju/6aRFN+Okgyj2fObneJ4vgVIN34b/G9fB2npPrrx5WLo/h84n+qTXVh1l2DY/jOooqTU4yzAmXNNg3+0H15Ff6qknJSFxSY+rm3qbJ8U57lp8F//MLbuKeDs9CuX4tpuPVMO16nQO7hhE6cYjxBRngyTVViXX6dDl5yI1UsRTUlUyv32Jb9r745C9o7oA7E4rhwv20S467azcR3MigudbKampqaxdq1Xva9asIS1N+6HduHGj13FqauoZX9dkMpGens6SJUs09iVLltCvX78zHlcgaMnMPWLl6vlFjJ9dyKQ5hVw9v4jrFxZz9/ISfs+2nrIiNsCnmTXUqwvppsyhMj/HztwjNmYftvHbYRuFDs/jO4Wt3MVzTOHfpLKFL7Jq3Fs0iqoydUWpl7BJDjQwd2K4RtjYnTZ+Wvkh/535KIu2zERR6iyF2K1Uf/Eqs5JCNOP4OWT+uiVfI2xUSYfj+rtPec8nw9fsT7+kcfz1smfp1n4AkcGxhAe2ISwwCoNB+yLZmQ3ufxtx8XHU24TIHuGh1uRg2/Q35Eqt4299HNnfeoQN4Dz8E6rqndxQVlwsrudr1LV9f6JDE07rHhvF4ovj8ttpX25n3KEyTZNBcpHiu4x1NVtRcw+gP7IffXYW+oOZXsJGNdX+nGQJfk0K5rNEk5ewqb9Kszdn69m5h2ZGrNy0Uu677z5uv/120tPTGTlyJAsXLuT777/n88+1+R9mzZpFz549GTx4ML/88gvLli1j0aJFjY7rcDjYs2cPULtaVVBQwLZt2/D396d9+/YA3HPPPfzlL3+hV69e9OvXj48//phjx465fXsEgtbE4lwbNy4qoZGakHyz38roWDOvDQwmwb/hR65TUfloT8Nv3KfCjJVR0kyM1JYuGMP36FSZB1f3oVeEiXd2Vmly1UBt9M/MMWFE+OihpgpMFlS9npkrP2BXdu0Lz9Ktv2C1VzGx3y1IkoTx+w/5PFLBUWd7zCyr3LMlnzY12rIJriHjUOLan9H91Cc6JJ5rh2qFUsb+Vfy4wuMX2JHNrGEMCnpeDfuUAeY9XuOojlJsm/8Pc5fHMYR5rygptkJcub/VO6cYpWwH+hBtfpyNe5dRUulZndZJekalX3lG99cYrmETUeZ9z6jsHMpNBlbFBWjaV8UFkBNg4vadhQTbvVWxnNIF25+nYV/0A9/kryIrxHs7cIAaRrusQ3yV5Bl7b842xvZuPGL2QkGIm1bKpZdeyssvv8z//vc/pk2bRnx8PK+99prGmRjgscce49dff+XRRx8lPDyct99+2x3q3RB5eXkMHTrUfXzw4EE++eQTBg0axO+//w7AVVddRUlJCa+88gr5+fl07NiR7777joSEs/RWIxC0EA5VurhjaePC5gQLcu0M+KmAx3sG8peOfuh12u3hXw9ZNeUPfA0Ss8eHM/OglS+yaiixN1wawdcgcWt4JsZ8rbgYxU8sdOq45LfeFNm05yYF6vlpbBjhJjB/+G+MK+agBIezeOIl7DqqXcldt2cRAb4hDLfEs3nXPPZ20G5HjR1wK6GxJagzP3Y7r6o+fjiuuuPkP5A/SOfEPsxe9yVWR61Pjo9UQ5K6k8uCDnKN/5rGT5St2Lf9EzXtAYxtxmianIe+AsW7jparYJlb3Kiqysods718bXp1GEpoYJTXuX8IgwH7tXfi89a/uCarhIRKOz90CMVZJwdOdpCZN3pGc++WfMJstX5XqsmM44rbcY6/jtySw3xtOER5PWGjV1Suyiph4NFsagw6pPb+qMddFgrKciirKr7ga1tJZWVl5ypK7oKivLzcnSvmTGiJPjenIjg4mBkzZnD55Zc3qX9Lu8c/+jtriKysLFJSUs7qmC0NcY9nh2qnwpjfC9lZ6jp15zr0DDfy4bBQ2gfWcRT9rYANhZ4v1j+l+fHagGAAbC6VTUUOKp0KilpbCUBRoST/KFemt+O7hc97RewAqCq0zw1ls74fX0QNBkki2kfHvIkRJAYYMP7+Nebv3gPgUKCJ//WIdvt21GdYYCVrK/2wq54v1naRqdw+/jF0kg6ppADj3O+RyopxjrsOpf3Z8dk42e9x9rovWbPbE/wQHRjJ7cHbNX0kn1j0Yb1w5fzqdb6pwz0Y4yYBtU7E1nV3QgNbUBiD8B30JTannZkrP2DPEW3VcoPeyN+ueplAP28H36Zw0s+qquLzzF/RH9gNQK6/kY+7RFDio/WRCjIHcVf3mwg2B6DEJ6EaTWzcu5TZ677EVU+wBdld3L6jkLYVnu2pN3tEcbBO2YjLet1In65jtXNxOUGWwXz63wHN8cwRKzcCgUBwmqiqygOry7yEzd+7+TOsjQWnolLjUpm+q4o1+Vofh81FTibNKWLuxHDi/Q1sLHRohA3U1nY6gcUgMahO2PQJspwKLltBg8IGav1MD8UWc/+uz0i25vOf1Gv5cUw4iQEGdNlZmH78CIAqo44ZnSMaFTYAyyq0WyJGnYErBv8J3fFIIjU0EseN9zR6/rmgV4dhGnFzrKKAEj89ocbjWzTGQCzdn0XnG4POJwZH1nvULX3p2Ps2IGGMuxTHwc8bFjYAznJyDs7n+y2LG4wwGt/nxkaFjVy2E9VeiD6sH5LhDMpoShL2G+/B58UHkGSZ2Conf8428s92Ufia893dyu3lfLT7R+4c/w/MOvh15fts3e9d5TypzMZtOwoJcGrvtVOxVSNu9q3+gT4dR4Fej37HRky/fYEucxsgoSQkIad0QUnugpzSGTX85E7azYUQNwKBQHCavLOrmh8OaP1YLm9r4cmegZqIxEsTLczIrOGfG8upcHq+WHNrZK6eX8ycCeG8u0vrazMq1kyHJkYvZexfqTkO8gujoqbU7QSrShKfd44gpXQV3xr0dAq5Bxx2LO8+hyS7UIAvO4Zrwo0Bhip2VulMyDSykhOho4lTPGdEBrchPjicI2WenDgZVT6MDKkCnRFLt3+h840BwBh/BZI5DPuulzVbT469b6FY85Dzl2rGlixRqLZ8bIrE6nI/Nh751mvr0cfkx9VD7mo0q6/zyM84st6tHc8cjrnLE+iDTn9FS0npgvWx/2DIWIPcviP+PQdTubyEIwc/I1XyOP+WVObzybx/o9PpNFFcJ+gTInGZnIC/s1YUqUYTrp6DcA0YTYflP/I7np/jPoMNw/+exFhegv5gZp1R1Frn5ewsWPgTAK70AdjufbrhgqTNiBA3FzFlZWXNPQWB4IJj5TE7T20o19g6Bht4e3CIV6oFnSQxJc2PcQkW/r66jDlHPDlg9pa7uGJeMbtLtas2d3dqWnp9VVXJqPd2PjL9CowGM98vfQe1zlSyQnzIsm5g6zd/p78zAFkqoSwpmFx/E3tDtSsK/Y1VDGxTTViNmZ+KgqCewIk1Oehpyse68X4sXZ5AH9K1SfNtDMVegqQzIBkbzo5cH1W2IRevx3noW7rpcjmCZ2t6e7UPQ4Or8Ov0CFJgGrlFB4/n0EnEEDkE9D7Ytz+tETiuIz9qxpf82mJM/jMbVj3LsjJ/ahTvZHmxYe2YPPweQgIaTm+h1BzFsf8jz5ztRdg2P4wp5S8YYi897ZQcSoduOOoU/XxtUChXVNzAvmIXydJOt72w/KjXuUZJYXxoJZ38bFQHFuHqfBkmY1+UjunukhQhad0J/PKvVBxXBA69juzsLZqcRY1hyFiDccFMHOMnc6RwPzZHNW1CEwnwDT6tezzbCHEjEAgETURVVf6xrhy5zlt8oEnii5FhDRaUPEEbXz1fjAxlytISfs32fGFsL9EKm+RAA6NivbegGuJY+SHKq4vdx0aDic5t+2BWwLinmK/SQt1OoifIsZfwAyXQoeFtlHiXnSHxtU66qb52xoRUMr/UIzr0qEwIq0AnAc5ybBnTsKS/eMYCx3n4Rxz7PgIUdCHpGGMnoA8fiFQvPFlVnMglm3HlL0UuWgty7apZmi8sLA1w+wJZFR27gq5FKaxhw8rHKK6oXaUI8Y8gPWkQ6UkDCez6T+zbnqbaJXPUbiTPYaBSrhUwKqBX4ilY9Qt5JQ378/VLG8W4Pjdg0De8dKWqau2WV33nZNWFY+/byOW7Mafdj6Q/c//FAKOO78dEMnH2TejLP6Od5B0dBhBmcHFlRBnhJ7bqVBf2mvm4QkswOkLQ+XREknRIPn6kJPdn0yFP+pBdYT5NEjcAxrnfMSvUxao9C9y2kIAIEsISiY9MIUjf9ozv9UwR4kYgEAiayJYiJ9vqCBIJ+GBoKElBp36U6nUSHwwLpWxBMcvz7A32+UsnP3RNfKvfX7BNc9wpoTdmow+6PRn0PlZFWI2TOe2DGgwBbghfsz9XhFVSt8h2b50VvS6WBWoNSBKTEsMIk+sk6FRd2He+gKXPW+jMpxddozorcOz/BKjdQlNKM7CXZiCZQjC0GQ3oCCnaTU1JKWrNUVC9HbeNOujiZ2NTladMwewdy736lVYVsmTrzyzZ+jMxYW2prkmk3FrZ8MSq9zZoDjQZmDBoKp0TT56YUC5chVyyqfH2/MVYqw5gSX8Onfn0KqfXJdSi58ex0Uz4/Wb01Z+SIGl9rzr5WhkXWonJYAJFGyoul2xELtmIZApFHzEQQ8RAkqPasOmQp8+uaB8uP1KKzg6unoNxTLoZJSIa/b5dVOzYSvDSnzG7aj/HuXKVRtgAlFeUsquihr2527mmz8NnfJ9nihA3AoFA0EQ+ydSWAxgTZ2ZsfNPfwM16iS9HhTJpThEZxdo3+0CTxA3Jpy6ICLXJ9g4Xa9/WeyQPBkC/r3abol2Fnb9mFHCwXQKL/a3sCG98bJPBwnWDbsFv37/qWCUswz9mYEAbejlt6HUG9DodzkNf4zzoyZelOkqx73geS4+XvVZcToarYEWDgkV1lOLM/g4AHzhl0cueiZ3ZtPNgk697tPhQk/sCGCSVfgHV9A92ERTb+aR9VVeN28/mBJJfAqqt0L3aBKBWH8K+7WksvV47rbnUJ8ZPz8xxbZj4+60Y7J8SwwEMksqI4Ep6+luRdEYsPV5EtZdg3/NfcNVL8ucowZX7G67c34hRJHREoBzfhiwyGtk3KYzY7s+jJtT6CuVUuXi5piNfyom8FF3B33LmIPvBb12D6/0gIFROxkcNRZIO41IacdY+hwhxIxAIBE2gwqF4JcO7PfX0CzMGGHX8MCaMcb8Xsa/C8+V+S4rfSbe26rIre6MmxDfQN5R20R0B0O/XFkNsM+RqbtLrKf3uDda08afYx0CgQyawXVf8+4wm2C+MmPC26IpWUDeuSxeYihTQBgCz0SPgTO1uAlWuzQtzHKV8F459H2Lu0PSsxK5jS07dqREkUxj6qKEYokeREJBMQuGzHC7wjhrT6/RI6LzCoZtKmr/K8MAigg21X85y0VoM0SMa7e84+CWqvU7RT0mPpfM/QNJj2/4sas1hd5NSmYUj6wPQjWlgpKbTPtDAD6NDKduQR7yuCF+dgo++VhIaU+9DH1RbCFMX2IFDG14gytnwFpZZpxJvdpBt92yLZikKP2ZuZe/h2hw+3+6vOV6/S+VYV38Ke5k57GNib752ddBfjcJXPb4q5WzLzjm/YTIFkNw+9g/d6+kgxI1AIBA0gR8OWKlxedYRYnx1jI47M7+JcEttTadbF5eQUeyka6iR/0sPOPWJx9myTxsllZ40EJ1OB6qKbp9W3CjJnVASUwi11XDlt+8iKQqubv2wXfskGDxfAbaSzZrz9KGNF4A0trsJpWIPcp1zXDk/ow9KwxA1/JTzV6z5KOU7NDZD9GhchatArmnwHMkUhj68H4ao4eiCOyNJHkffoV0n8cWi/7iPg/3C6Z06nF4pw9Dr9OzM3kDGvlVkF3i2nPQ6PdGhCcSFtSMyJAaD/sSXuoQkScSEtSWkeC7Ow56Efa6CZY2KG6XqIK6cnzQ2Y/zV6PzbAuDT+w3sO55HLvEkSnTlzsInNBQ48xwwqqqQlPcGsvmQxv5u+RiCKwbyp9qAMT466Muj+x5ijE8Gk/w2cInPVvx12u3R9j5acXPAamaoZQ5PZQ1ErVOt6Rb/Zfw9bBYKEivytQI/zmjAaG1H3Y0wRS8RFRl8xvd4JghxIxAIBKdAVVWvLambO/hhOElumFOR4G9g8aQICqwKUT66JkfQlFUVcfDYbo0t/fiWlFR0DF1FqWfeZgtKXDsAnOOuw5U+EKmqvDbJns4jDlRVQS7RJqfThzUubiRJj7nzY1jX34tq9/jg2Pf8FyQDOv92SJZIJF3DTreugmWaY11gGuZOD2FyTcWVvwylMgvJ6E9BpYk27Xqj841DMjYeRZYan85tox8muyCLuPD2pMR2qxV7x+ndYTi9OwynpLKA/NIc/C2BRIcmYDScPHxZNgzTiBu5eCOuY4vQRw5B0tWeq6oqStV+HHve1OTKkcwRGNvd6Dk2+NT+zDbci2o75rYHlX6NUj0Ind+ZZXB37v8EuVArdhdZu/J82bX4bqhgTJyF1fkOHllbDuiYa+3JXGtPzDgZ4rOLK/w3MSzwCPk2PYWSBfDU1jpkM5FkK2ekZSuLbD0ACJBqeCT4p+PtRo0YQpUIsHekWvVIC4PsRI3vRoD/6a9y/hGEuBEIBIJTsKXIqYls0klwS0rT/GNOhk6SiPb1DjU+6Vz2r9Icx4UnERFUu310wt/mBHK7NNB7HvNqdBwqcV5jKpX7wVknvN3ghy7g5AV0JWMg5q5PYNv0EKjHfzayDfuO54730CFZItCHdMeU8hckg+fLzXVssWYsQ/TI2jENvhhjxwO1ZWKsWVnog5q2qpEc25Xk2JNHbYUGRBIaENmk8QB0/slIPrGo1txag+rCvusVyHofY8xY0Pviyl+MWn3Y61xTh6leEVGS0R9z18exbfq7O5pKpzqwbX8On95vnHaiP2fubJyHv9fYMp2x3FN4Fwo6qlwqkxcUk1nu7dtkx8hCa3cWWruDOzehyq28SpBUK5AVJBaUBtLXMJNtRFJILI+FzyNMX4WqwooyreBsZ4uj2qCNMtM5DPTs3Pa07utsIKqCX8AEBwef9L+pU6ee9Wvu3r2bW2+9le7duxMcHMyLL77o1efFF1/0mkuHDh3O+lwEgvPFp3u1qzajY83EN1IE81wiKzIbM5dqbOnJg9z/1tUTN0pSp6aNWy+6Rx+SjqQ7tejSB3bA1OGvjbQqqLZ8XHnzse98yV0ZXak6iFp9yNNN0mGIHNrwEM2MJEkYokd5NzjLcWZ/h/PApw0KG31YX/ThAxocUx+QgilF+2xWaw5j3fQAtq1PYtvxPPbdr+M49C2qq+EtOgBX8SYce9/SztcUQmbsE1SqHuG9q8ylSV1g1MGAqMZWrCR20sfLWuWSuV56h2eiZjJItwSrLHHAZiLX4RnHooQgGxI154WVFdHj+ls4zbQ+ZwWxcnMBk5npyRw5b9487r//fo3tXNSBslqtJCQkMGnSJJ577rlG+6WkpPDbb54Ku3r96b2dCgQthQqHwo/1shHfdgaOxGeDvTlbqajxbBsYDSa6tevvPq7vTCwnnzy6x92vvrgJbbx4bn0MMeNQKjJx5c1tfPziDbhyfsEYfwWufK0jsT6kJ5IpuMnXO98YE65CqTrgtfXTGDr/dpjS7j/pNqMhZjxK+U5cxxa5bWr1YeR6QkkuXn88Ck37/FSqDmHf8by2ZITOjLnb00wOaMvnh4tYXa/sB9SuOH4wNJTL2lp4Z2cVT2+qoF4lBrqkjGdUUDgrtnyNQ5OWWaWwYCM/4g/4Y5Q8bUbVl0hXB02+R39bNfs6TeSqSB+ytDkvzwtC3JyC6sXjmt73LFzPb2TjD4j6REV5qtCeKCBZ1/bJJ5/w5ptvkpOTQ1xcHA8++CC33Xabuz04OJiXX36ZhQsXsmLFCsLCwnjyySeZPLnxcvc9e/Z0Vw1//fXXG+1nMBg0cxEILlR+OGClup4j8ZgzdCT+o6zfs0hz3K3dAHzMx4WWw47usDZiSEk+9cqN6qpBKdeKopM5E9dHkiRMafehC0pDKduJYs1DteahOoo1/Rz7PkIX3BXXsaXaazXBAbk5kfQWLF2fQKk+gjP3d1zHFoCrun4n9KG9MESPRB8+AEl/8kSMkiRhSr0PuXIfanV2o/2U8p04D3+Pqe31bpvqrMK2/Zl6jtcS5s6Pog+sXSF/a3AIg34uwCprA+n/OzCYK9rVbn3d2yWAQdFm/rqilN1lLuL99bw+IPi4k/xYugXrmL36QzKtDX/WncdTYOtVMxGuziB5/KskVSHbFshdo5omrs8FQty0UmbNmsUjjzzCCy+8wMiRI1m0aBEPPfQQkZGRjB8/3t3vxRdf5KmnnuKFF17g559/5u6776ZDhw706NHjD13/0KFDdOzYEaPRSO/evXnqqado27btH7wrgeD8ci4ciU91vZLKAlyyk8jgWM3bf3FFPvuOaiOM+qaNdP9bdygTSfbEqCgRMaiBIae8ply2DVTPeZJPLDqf0yuGKEl6jDHjIMbzMqhYj2HdcI9HCKhObBn/0Pr26MwYIgae1rWaC51fPOYOd2NKuh1X/pLaUHZJhyFiIIbIoae9+lQrmp6kcsMjGOTSRvs5D35Ru80V0B5VVbHvfg3Vqi2zYEq+U/NzbB9o4Mlegfxjvedn/VyfQG7toF1x7BFuYvUVkeRUy8T56TWft5C4UVyV8DVZJUUsKAmkXPZefdepBiJdnTGgFXOhBYUoV95PqKX5VuyFuGmlvPXWW0yePJm77roLgOTkZDIyMnjjjTc04mbSpElMmTIFgIcffpgVK1Ywffp03n///TO+du/evXnnnXdISUmhqKiIV155hTFjxrB27VpCQxtO+y4QtERmH7ZpHIklzo4jcV1kReZIQRa7j2xhz+HNlFTWRh91TuzDtcOmoj++JbEhU+uEG+4fS0xYW/exfl/9Lamm+tvUCwE/SZTU6aDzicaceh/2nS95jE7t/oQ+vD+S4ez+PM81kt6CMWY8xpjxp+58CnS+cRS0eZL20QZw1aAqNnBZsWe96/lZqS7su17Gp8+bOI/8jFy0RjOGoc1YDPFXeY09tZMf/kaJlXl2Lmvrw6WJDTsrS5LUoP+YJOkwxl9JsvVt2sYUsc9q5ojNxBG7kQKnAUnVEyF3xoj291doNfJ1pynM79j01AbnAiFuWimZmZncdNNNGtuAAQOYM2eOxtanTx+v4/nz5/+ha48ePVpz3Lt3b9LT0/nqq6+49957/9DYAsH54pt9Ndy7UvtGPTru7DkS2xxWlm79mS37VlJjr/Jq35m9geBN4Yzrcz1Ol4PNWSs07alttCKkvr9Nk52Ji8/c3+ZUGKKGIxdvqt3Kaaj9JAnxLhokI/rAehFhOmOdqLPjGY23P49cvEHbLSAFU4d7GvTvkSSJWzv4ea3WnA6GNqNxHJiBwVVFmq+dNN/avDg1ft1ZmtuTsuIKTf+q4Hiqu/Xn224B52x1s6kIcXMKmuoDY7PZzokD7x+hsQ/8+cbf35+0tDQOHDhw3q8tEJwuqqryvx1VPLWxwqvt3i5Nfxu1OaxU1JQQ5BeK2eh5a1ZVlV2HN/H7us+prCk76Rirds4hLqI9TpcDq8OzPeZj9qNteB3xoqpekVJNWblRrMc8Yc4AkgF9cPdTnnc6mDpMRS7f6bWVgiHgtHx7LiYMkYORo0dpHI7l4nX1OgVg7vI4kv7kuXr+CJLegjF2Is7sb902l6JjY2FfyopLNH1jY2MZO3Z0iwkeEeKmlZKamsratWu55ZZb3LY1a9aQlpam6bdx40ZNn40bN5KaevL8FqeLzWYjKyuLIUOGnNVxBYKzjaKqPLGhnHd2av1sJOClfkEMbdPEit2lR/h8wWtU1JSi1+lJiEwhOaYr8RFJrNo1l8wjGU2e008rPyTAV+s70zN5KPo6dZykkgJ0ZR4HXtVoQolPOuXY9aOkdEEdTzvXyqmQDL6YOz9am9uljm+PIXJIo0n+BGBKmYpculVbzsFNrQPx6fpGnQmGuMtw5swCuQaXomdF6aXklWmFTXh4OJdcckmLETYgxE2r5b777uP2228nPT2dkSNHsnDhQr7//ns+//xzTb9Zs2bRs2dPBg8ezC+//MKyZctYtGhRI6OCw+Fgz57a2iQ2m42CggK2bduGv78/7du3B+CJJ55g3LhxxMXFuX1uampquOGGG87dDQsEf5DDVS7+vrqMhbnalPQmHbw3NIQr2zXNN8TpcvDdsneoqKnd0pIVmYPH9nDwWMM1fQAsJl9S49JJS+iJr9mPzxa8hqzUJl5zuOwUVxzT9O+TOoKSfM/KUv3kfUq7VDCcWjh4+duco5UUfWAqxva349z/Ua1B0mGInXhOrtVakIz+mDv+vdYJux7GdjdhCDt5dfKzhc4chiX9OWqyZ7EsM4z8Mm0EVnBwMOPGjcNkOncrSGeCEDetlEsvvZSXX36Z//3vf0ybNo34+Hhee+01jTMxwGOPPcavv/7Ko48+Snh4OG+//bY71Lsh8vLyGDrUk3Dr4MGDfPLJJwwaNIjff/8dgKNHj3LnnXdSXFxMeHg4vXv3ZsGCBSQknFl6cYHgXCIrKh/sqebZTRWakG+AAKPEFyPDGBbTtBUbgEVbZlJYdvTUHQGdpGdwl/EM7365phTAxH438+uaTxs8p4MhhIiKGkpUz1zr15OSm+BvoyoycmmGxnYut4mMCdcgmYJRynagjxyKPuDUK0sXO/rQnhjiLsOV86vHFtYHY9sbT3LW2cdlSWLxwTQKygo09pCQECZMmICPz9ld7TsbCHHTSrj88sspKyvT2O644w7uuOOOk54XFRXFjz/+2OTrJCYmel2nPh9//HGTxxMImpNdpU7uX1XKxkLvqtGRPjq+Hx1G97Cmv5Fm5+9l9c6m+eklRCZz2YApRIV4l0Po3WE4OQX72LzfO3Hc4C2Z+C6YQqfgCIw+FqipQqrURiE1JXmfXLBUm6/FGITuHAoOSZIwthkNbUafurPAjSnpT6jOSuSidehDumHu+BCSdP6KC5SXl7Nw4UJKSrRbUWFhYUyYMKHF+ZqeQIgbgUBwUbK12MH42UWaSt8nGBBlYvqQENoGNP0RaXfa+HHl+6h4xgvUy1yXHEueX2/2F+ZyKD8Tk8HCsG6T6J06HF0jX1IScHVWKYVVdo4EelaNgm0uOhXXZks2lxVCWcNzOVWklOqsxJ71gcZmCOtzXr80BU1D0puxdH60Wa598OBBli1bhtOpFf8RERGMHz8es7npK5rnGyFuBALBRcnj68u9hE2gUeLp3kHcluqL7jQjC+dv/JbSykKNbUJYBeHWIsJtO+iRcBmmYa9qCkg2hnHOt5iXzWaKWc9/ekVTaa59VI85VI7eW4tpkBOSUUPCT9rHcWAGOMs8Bp0RY1vhEyeoRVEUNmzYwLZt27zaoqKiWqSPTX2EuLmIOdX2kkDQWtlU6GDlMW3tnYkJFl7pH0yM3+lHfOzL3c76ekn2evrX0NZy/BqqjOvIT7iOLcGUdAeGNpfUrpI47OhyDoJej2rxBR9f9LszMH33HgAhdpmHNh4jo100gRNvJ62tHdfG5eh3bdJkIwZQDUaUdqnYb74fALkiE7lkC/rgLuiDu7j7yRWZuHJ/15xrTJyMzjf2tO9b0PooKytjxYoVHDt2zKutXbt2DBs2DKOx5Ue5CXEjEAguOv67vVJzPLSNmS9HhZ3RWHanjZ+Wa6szhxhcDA+u9O7sLMOx53VcR+dgsYzD76MP0ZUUeverQ6DBQu/bnkeNbYsMyMMvhepKcjesIT45GdU3ANUvAExmTpRflku2YNv6JKgunIA+cgimlL8gGUNw7HkT6mydST4xGBOuO6N7F7Qe7HY7mzdvZufOne4K7ieQJIn+/fvTuXPnZsmVdiYIcSMQCC4qssqd/JZt09ge7Op/xuOt2fIlFXbPeBIqE8MqMPuEY4i7HFfOLFS7NspEqdhNTdkuaC/jVwE6V8Njq3o9tnufRo1tq23wC6Amth1KXHvvc5wV2He9AqpnULlgBdbijehDe6BU7df0N6fee04TwQlaNoqisGfPHjZu3Ijdbvdq9/PzY9SoURdcIWQhbgQCwUXF/3ZUUfe9tFuokRGnEepdF5ujhlWZ2oimPgE1tG13CabkO5EMfhjjJuE8/APO7O9AqbMVppOo6WzA1k6PKU9BNYCqB9UgITlVzDkK0qB7kTs3PZ+JqqrY9/wX1VHi3ShbkQtXa0z6yGFntdyC4MLi2LFjrFq1yisS6gSxsbGMGDGiRYZ6nwohbgQCwUXDsRqZb/bVaGwPdPU/46X2VRnfY5MV97FZUhje7y7McWPcNklvwdTuZkzOdjg3v4wjUvt2rPhK2JK8/Xwc8XqQZmDYfQhDzHh0gamnnKcrb56XgGkUvS+mlLua1lfQqrDZbKxfv57MzMwG2/39/enbty/t27e/YLah6iPEjUAguGiYvrMKh0eLkOiv5/K2Z/ZWarVXsyZzqcbWL9wH/1htHhfdgT2Yfv4Uw9a1ANjjdFT0M6L4N+FLQ7bhypuHK28eOv8kjG0no48Y3GDItlKTiyPrXe21A1Jqix/u/xRkragzJd2OznxmfkaCCxNVVcnKymLt2rUNbkEZDAbS09Pp2rUrBsOFLQ8u7NkLBAJBEyl3KHySqa0ZdW8X/zOuXrxy20zsdVZtLDqFgenXud90dYf2Ypr5sVvUnMCcoxCeZ6dycBjWtjZAG/XUGErVfuw7XkDyTcDU9gZQY9xtquLCvutlkOv4EunMtfWHfOPQRwzCse8D5PylAOhDe4vyBxcZsiyzdOnSRgsYJycn07dvX/z8zryKeEtCiBuBQHBR8MmeaiqcHm+bMJPEnQd/x7x4D67BY5G792/yWNW2StbuWaKx9Q/V4xszAgDdnq34vPx3r3DtE7j6XoLupofwUatQyrajqgqS3gw6E5LOhFy2HVfe/AaLJqo1h7Hv+jeR+jCs5aGojlJUR5nGgRjAlPIXdL612Y915jAsnR9DaXcLqqMUXWAaktRyihwKzi1Op5OFCxeSk5Pj1RYcHMygQYOIiYlp4MwLFyFuLmCCg4NP2n7DDTcwffr0s3rNGTNm8M0337B7924URaFbt248/vjjDBgwQNPvww8/5M033yQ/P5+0tDRefPFFBg4ceFbnIhA0FZtLZfquKo1thm0BAd9+AoBx/RJsf30KV7+RTRpv5bZfcNQRLr46hf5drnYLBtOcbxoUNnJyZxxX3I7cpTdIEjp80EV7X1Mf2gNj25uQSzbiOjobuWg9oA3PNcjFKJXFXucC6MMHYIgZ72XX+caCyGdzUWG325k3bx75+fkau16vp0ePHnTr1q1FVfM+WwhxcwFT1xls3rx53H///Rrbuaj5sXLlSq688kpeeuklfH19eeedd7j66qtZsWIFSUm1dWlmzpzJY489xmuvvUb//v358MMPufbaa1m7di3x8fFnfU4Cwal4f3cV+VbPFlI0VkZv/EHTx/zxq8iJKajRJ/+MVlnLWbdHm7Cvf4iMb/xxMVFZhn77ek27nNwFx5W3I3fu5c5FcyoknR5DeD8M4f1Qqo/gzP4G17ElgHLy80whmNMeuGAdQQVnj5qaGubMmeMVDRUYGMi4ceMICgpqppmde4S4OQVPfnrbeb3es7fPaHLfunkHTnxI69o++eQT3nzzTXJycoiLi+PBBx/ktts89xMcHMzLL7/MwoULWbFiBWFhYTz55JNMnjy50Wt+8IG2Hs3rr7/O77//zsKFC93i5u233+bGG290X+uVV15h0aJFfPzxx/zzn/9s8v0JBGeDYpvMq9u0CfXetS5BX6O1SbYaLG/9C+tT79QmxGuEpRk/4VQ8qzJ+Opm+HcfXbisBhvVLNas2Skwi1if+12RR0xA6v3jMnR7B2Pam4yJnEaj1Vob0FnR+bTGl3oNkCj7jawlaB+Xl5cydO5eKigqNPSwsjPHjx1+Q4d2ngxA3rZRZs2bxyCOP8MILLzBy5EgWLVrEQw89RGRkJOPHe5arX3zxRZ566ileeOEFfv75Z+6++246dOhAjx49mnQdh8OBzWZzb5E5HA4yMjK47777NP1GjhzJunXrztr9CQRN5eWMSiocni2dWJ2V8Tt+bbCv/sh+zF/8D/sdDzfYnp2/l3WZWl+bAcF2fBOvcB8bVy/UtDsHXPKHhE1ddL4xmDv+HVPSFI5kriQ2IQnJFFL7n75lVmcWnH/y8vJYsGCBV0RUdHQ0Y8eObfF1oc4GQty0Ut566y0mT57MXXfV5rFITk4mIyODN954QyNuJk2axJQpUwB4+OGHWbFiBdOnT+f9999v0nWee+45/P393WMWFxcjyzIRERGafhERERQUFDQ0hEBwzthf7uKjPdoIqRmOZeirGyiNcBzjst+QU7vhGjRGY3e6HPy06iONLUgv07vDCCRjAABSwVH0+3Zo+rgGXPJHbqFBJFMIdksa+qCUsz624MImKyuL5cuXoyja7cv4+HguueSSCz7Eu6mI+vatlMzMTPr166exDRgwgD179mhsffr08Tqu36cxpk+fzqeffsrnn39OYGCgpq3+fr+qqsIHQHDOyCncz9Ktv5JTpA1z/demcuoW/k6zOBi65RdNH8e461Dq+dmYP30dKfeQxrZoy0yKK7TFBMeHVeKTeLX72LBGu2ojp3RBjWhzurcjEJw2qqqyadMmli5d6iVskpOTGTNmzEUjbECs3JySpvrA2Gy2c+LA+0doSEycLYExffp0nn/+eb7//nt69erltoeFhaHX671WaYqKirxWcwSCs0Fu0UHen/0sqqqyJONnbhx5Pzr8WJNvZ1a9GlIzHEvRVXt8EFRfPxyX3YJr0Fh8npmK5KwtjyA5bPi88QQ1T70D/oEcKdjH6l1zNWOl+9eQbI5E53Pcz01VMa6pvyWlTegnEJwLZFlm+fLl7Nu3z6utV69e9OjR46J7uRQrN62U1NRU1q7VJg9bs2YNaWlpGtvGjRu9jlNTU0869ltvvcVzzz3Ht99+6xUCbjKZSE9PZ8kSrV/CkiVLvFaSBIKzwfo9i9xVjBVV5ocV71FhLeWJ9eWafoODXPRY95PG5hhzLfgFoCQkYb/lAU2bLj8Hy1v/xGmvYeaqDzWVkgP1MiOCq/Bbtt+9WqPLzkKXd9jdR9XrcfUddlbvVSCoj9PpZP78+V7CRqfTMWLECHr27HnRCRtoQeLmtddeIzg4mEceecRtU1WVF198kbS0NKKjo5k4cSK7d+/WnGe323nkkUdo3749MTExXH/99eTm5p7v6bc47rvvPr799ls++OAD9u/fz3vvvcf333/P/fffr+k3a9YsZsyYwf79+3n99ddZtmwZU6dObXTcN998k6effpq33nqL5ORk8vPzyc/Pp7zc80Vyzz338NVXX/HZZ5+RmZnJo48+yrFjx9y+PQLB2UJRFDJzMjQ2m6OG33b+wNYira/N+3bvVRvnGM+WkmvoBJxDtLlhDLu3sPzLJykqz9PYx4VWYHEomI8omD9+BV12lveWVNe+EBD8B+5OIDg5NpuN2bNneyXns1gsTJw4keTk5GaaWfPTIralNmzYwIwZM+jcubPG/sYbb/D222/z9ttvk5KSwssvv8yVV17Jhg0bCAiodeCbNm0as2fP5qOPPiIkJITHH3+cyZMns2zZslaZmKipXHrppbz88sv873//Y9q0acTHx/Paa69pnIkBHnvsMX799VceffRRwsPDefvtt+nZs/EqwR988AFOp9NLqNRNGHjVVVdRUlLCK6+8Qn5+Ph07duS7774jISHh7N+o4KLmSOE+qm3ezsEuez7D+ZWFXA1IXN1GJeXXHzV9TqzauJEk7Lf9DV1+Lvq92yiyGJjbLojNUiHgefPt5melvY8Dyy4ZSQEcdixvPglOh2Z810CxJSU4d1RVVTFnzhzKyso09sDAQMaPH+/lB3mx0ezipry8nD//+c/873//4+WXX3bbVVVl+vTpPPjgg1x++eVArZ9HSkoKP/zwA1OmTKG8vJzPP/+ct99+mxEjatOev/fee3Tt2pWlS5cyatSoZrmn5uDyyy/3+pDfcccd3HHHHSc9Lyoqih9//PGkfeqyffv2JvW78847ufPOO5s8rkBwJuw5sqXRto7SZvLUePZI/XjJvBvpJKs2bowmCv78ECs/f5y1ITqUenWnAvQyI0NqxZTPPk+eGV2R1tFYtfjgShcZuQVnF1VVKS4uJjs7mz179lBToy2GGhYWxrhx4/D19W2mGbYcml3cnBAvw4YN04ib7Oxs8vPzGTnSk5rcx8eHgQMHsm7dOqZMmUJGRgZOp1PTJy4ujtTUVNatW3dRiRuB4KKhuhLjkl/BZGZ35SZNk05vQpE9KyjD+I3R7ZOIy9T6ljmHTtSu2gCyIrNq51yWZvyMM8x71VdCZXxoBRadil4OQe3ZDxb93OAUXb2GgrllBRgILlzKy8vZuXMn2dnZVFVVNdgnJiaG0aNHXxQ5bJpCs4qbGTNmcODAAd577z2vthN1MBrKl5KXV7v/XVBQgF6vJywszKvPyXKqZGVledksFgtmc+NZSZuCzWY7dacWxokkfE2lJd1jRUXFOcmd09Dno7Vxru9RUqz41GxClczYfLqh6v7Y39YJdHYbHT59EXPhUQp8DBT399RJ0kk6luimMMD1CSapVuDoJRn14BvMLqxmrF7CItc6BWeHx1NV52dQXlPEqqxZFFU17K8XoXMyKryKtpbacUvCRpITP5jkzB3453hHqGQndKTyPHyOxGe1dXCyeywtLWX79u1e4d11iYiIIDk5mezs7HMxvbPC2f49pqScPMdTs4mbrKwsnnnmGebMmXNSpXkm+VJO1aehH0p5efkfCuVuiaHgp6L+NtapaGn3GBgYeNZrVWVlZZ3yj+ZC51zfo6qq2Lc+gVxau6oiWaIxp92PPrRxX64TKPYSXMcWofNpgz68P5KuziNKVTFPfwZj4VEAtkdol94V3xS2VSVSw1WM5xuPXZVZGmNhY3gsEw6U0bdUps2IcWAwoigKa3bPZ+G2H3DJTq/5hBh8GXPoMIkDZNw7VJKRNl0nIxkDkR75N8o//4KuzFO9WwkKIXr0pUTrz+3jVXxWWwcnu8fS0lJWr17dqLCRJImuXbvSp08fdLoWEx/kRXP8HptN3Kxfv57i4mJNKLEsy6xevZqPP/7YHcZcUFBAXFycu0/dfCmRkZHIskxxcTHh4eGaPqICtUDQPKjVB5FLPNtFqu0Ytox/YIi+BFPKXUjGhh0dlapDWDc/DK7aZXfJJxZT+9vQRw5GknQYF/6EcZ0nxcCOcG1tnGXVHQDYRzc2qUfpJS3XtFeZ9HyXFsaPKkhfT0VCQkVtUNT4mv0Z2eMqeqUMQ1nzBE5HhrtNH97PfQ9qcBi2+57G58UHkVy14zhHXgHnWNgIWj81NTXMnTsXh0PrqK7X64mLi6Nt27YkJCS0qBfOlkSz/QVOnDjRq37RPffcQ1JSEn//+99JTk4mKiqKJUuWuKN3bDYba9as4ZlnngEgPT0do9HIkiVLuPbaawHIzc1tMDuvQCA4P7gKVjVsP7YQV/EG9El3UWJO4ljJEQrKcjAazPRI7I5l99NuYQOgWnOx73wB3eEUzJYRmL5+x91WadSRHajd6jqgdnT/e79pNE/06sDKbT9QWVOm6SdLQAOC5gQdQgIZF23Cv3oOzozfUVzabSdDG21ZBiW5M9Z/vIlx6SyU6HicY69tdGyBoCk4nU7mzZvn5V/To0cPunfvjtFobKaZXTg0m7gJDg52F1s8ga+vLyEhIXTq1AmAqVOn8tprr5GSkkJycjKvvvoqfn5+XHPNNUBtJexbbrmFp556ioiICHcoeOfOnRk+fPh5viOBQAAgF632sqkqrKnwZWe1gZL9M1DRbhtv3T2LKZEFmBpYWVcqs7BWZqEMkghaCZICu8J8UOtsPTudgVQZgt3H05KdDEwbQu82HVn3nztZnBiI6xTb2Ra9nkuCS+jsm49UDQ1tBEimUPShvbzsSlJH7EkdGzhDIDg9FEVh8eLFFBUVaexpaWn06tXrokzIdya06LXTBx54AKvVyiOPPEJZWRm9evVi5syZ7hw3AC+88AJ6vZ4pU6Zgs9kYOnQo77777kWd40YgkCpK0W9ehRLbFiWly3m7rlJzFKXqYB2LDskvnnV5hSwvD2j0vFKHwo5qH3oGWBvtY2+npxwIWuFke2IkdeVHt8IaOF7C6br2PgwKrQ2R9dm7gzGF5ST3s7OsMoC9VgtOVfvloJd0pPg6GRVUSIChcadNAEP0JUg68WwRnBvKyspYs2aNV1K+uLg4Bg0aJITNadCixM3vv/+uOZYkiWnTpjFt2rRGz7FYLLzyyiu88sor53p6AsEFgT5jNZb3nkeqqc3Qa/vTo7iGjj/FWWcHV6F2S0oX3JWKtlNZvudfgNrgOSdYX+lLur8VozkFv6Nx2OQ1OKK00Xn2dnoKI9uyt1SFOk6WE3IP0zngCDv94/lbtwAoKgbAsG09lX0MBJpVJpkrUNQKjgdMHZ+NhF5S0TfhO0PyjceYeN2pOwoEp4nD4WDLli3s2LHDy3k4NDSUUaNGtWiH4ZZIixI3AoHgD6DImH6egemXzzRm8xdvIHdMPy/VqeVC7ZaUFNafmas+Rla1wiZI7yLC5OKA1YxyfIuqzGVgf7k/A2dvR+fcjg/gDJMoH2pEDvQ82PdKBTiVYPdxiM1FbJWTu44u4r+9/kRasIF9RYCqIuetxj7Qs9Kik0Dvn4BafaIGlLfgOuHILJlCQJJA0oHOgs6/LZIkvmAEZ5djx46xbt06rFbvVUs/Pz/GjRsnctecAULcCM6YyZMnExoa6i67IGhGqiqwvPschu3rvZokuw3zJ69he+SV2i/rc4RiL0ap0NZ+W1VYw9Fibe6NK1Pak2pfA8DvxYFsr/ZEPW3M1TO4jq+vsVglZL6DkvFmFL/auWdZtY7EyRYbRVebuca4gl6WEpTyKYAZ6XAmVZ1rqFtCTxeQiqX367hyfsWx7yNQXZ6BJCPGtpMxJlyHpBdfJoJzT2ZmJnv27GmwLSoqimHDhuHn53eeZ9U6EK8hFzhTp051O2eHhYXRpUsX/v73v592DhvBhYuUdxjff/2lQWFzAsPOjRhWzD2n86i/apNvTGL5zkUaW5e2fek56J+YuzwBBn/6BmiLW2YHWDhYJwpKlXTQpgt+6iQwBnHAamJ3jTb0NSnSieIvgRnaqTuxbX6Y0MJ3cGRORw6q84hTwZR6D5Kkxxh/JZbe/0EXmAaSEX1YP3z6vYup3c1C2AjOC2VlZaxe7e187+fnx8iRI5k0aRJBQUHNMLPWgVi5aQUMHz6c9957D5fLRWZmJvfeey/l5eV89NFHzT01wR/FbsOQsRrVaEbuMdB75cVag89//oGuUFu1WgkMQQ2JQJ+9120zf/0Wctc+52yqrjrixqXCb3m1CfRO4G8J4tL+twJgiByMLjCVNoe+JqlyO/urPbk8liQEklhown7tn5E79XKXSdi+NZLZ+3/TRFqZJYV4szYPCIDFthtHvaebkY7oAzu4j/UBKfj0/i+qIgsnYcF5RZZllixZgsvlWTnU6/V0796d7t27YzCIr+Y/ivgJnoIPPvjgvF7vz3/+82mfYzabiYqKAiA2NpYrr7ySr776Cqj9I3rggQdYvnw5BQUFxMTEcNttt3Hfffe5HdSmTp1KSUkJw4cP580336SmpoaJEyfy6quvuguw1dTU8OCDD/L777/j6+vL3Xff7TWPsrIyHnvsMebMmYPdbqdfv3689NJLdOxYGyL75Zdf8n//9398+umn/OMf/yAnJ4dhw4bx3nvvsXTpUp5++mmKiooYN24cb7zxBj4+Pl7XuKhQVSxv/wvD1tqElq6eg7Dd8zScePCpKuYZr6PL10ZWyEmdsN37L3A68X3iDiSHHQCpphrzjP/AhNvO/lSdlShlW93HK8v9KazR5ui4bODt+Fk8EVN7bcE8tucKnti0jv1pnkfRjnAfci67j9DUWiGmqArzN37Lqp1zoV4I+ZDgqqY5A9tVTOnen1lACBvBeWfz5s1eod6DBg0iNTW1mWbU+hDbUq2MQ4cOsWjRIneSJ0VRaNOmDZ9++inr1q3jySef5LXXXuOLL77QnLdmzRp2797Nzz//zCeffMJvv/3Gu+++625/8sknWb58OZ999hm//PIL27Zt81pSnTp1Kps2beKrr75i0aJF+Pj4cM0112gc5ex2O2+99RYffPABv/zyCxkZGdx222189dVXfPbZZ3zxxRfMmzePDz/88Bz+lC4M9Ls2uYUNgGHzKswf/dsdJWRYMQfjmoVAreOtPVaHY8BwrNP+ixoaiRoVi+MabWV2w5ZVRKxfCJVlZ3WurqK1oNbOq1rWsb5C6yfQI2kwHRM85Rfya2SunFvIHaveZVheLrGVntUXVZJYVbKLKms5Ow6t58tF/z0ubLSkBvqSoQ7iUEY0obPshMx1YDzWcCi334EAaCO+OATNT15eHhkZGRpbu3bt6NChQ8MnCM4IsXLTCli4cCGxsbHIsuwubPn8888DYDQaefzxx919ExMT2bp1Kz/++CO33nqr2x4QEMDrr7+OwWAgNTWVK664gmXLlvH3v/+dqqoqPv/8c/7zn/+4K62//fbb7mSLAPv372fOnDn8/vvvDBo0CID33nuPrl278v3337uv5XK5ePXVV911Rq655hreeecdsrKy3AVQJ0yYwMqVK7nvvvvO1Y/sgsA493tv2+oFqL7+uEZchvnzNwCo7qSnqs/xjKX6rRj2v4shZjz6wBSco6/CsH4p+n073WPELfgOFnyHEhmDnNQJuXt/XP1Gwh8INa3rb7OtyqJJgBfoG8L4vje6jx2yyp0L8/n79s+5saD2vBFHKviik6eEyobMJazPXNzgtYwGE4N63cXVa2MAWCTtZnHJc4BKyDwHjhgdVT0MuMJr78d4TMEQPJTGcxILBOcHu93OkiVLNDaTycTgwYNFDpuzjBA3rYCBAwfyxhtvYLVamTFjBocOHdJsG3388cd89tlnHDlyBJvNhtPp9Co4mZqaqtnnjY6OZuPGjQAcPHgQh8NB79693e3+/v507tzZfZyZmYlOp6Nv375uW1BQEJ06ddJEA5jNZk0BtcjISKKiojSV3SMjI8nMzPwjP5ILHin3EIZt6xpsMy38CePq+UgOO7IvVPWo82csW3EdnY3r6Gx0/kkYEydj+9P/4fvkne7aRyfQFRxFV3AU45qFuNYuxnbfM54tr9NAdVndtaRUFbZWa7cTB3Yeh4/Zs5Lz9u9beHPu63St9mynpRdU81NSGNXm2ge82khOnACfYG4a9SA/5IUDFQAsD+7IP4Y/zrM5v6DftwPzUQXTUQfOCAnFImE+qmD724AGxxMIzheKorBixQqqq7VO9B07dhT1oc4BQtycgqb6wDRnxWxfX1/at28PwMsvv8yll17Kyy+/zLRp05g5cybTpk3j2WefpW/fvgQGBvLBBx/w22+/acaoX6tEkiTU47lJVPXkyddO1afuG0l9RzlJkhq0NVYF92LBNO+Hk7afSNBX1dMIhobf+JSq/dh3voCacjeGa/+MuU5tpvoYMlZj/vQ17H/6v9MOF5dLNoJSu610yGaizOX5fRp0RnokDT7e0cXOTz/lsRVfY6zjaAxglcxsZjCpNFyXCiA6NIGbRz1IkF8Yc9YVatoi+g/A2nE0ugN7MM7/AcO6JZgKZUBFiY5H7ph+WvckEJxNTjgQHzx4UGPv1q0bISEhzTSr1o0QN62QRx99lGuvvZbbb7+dNWvW0KtXL+666y53e/0/sFPRvn17jEYjmzZtcju8VVdXs2vXLtq2bQvU1j1RFIX169e7t6UqKirYtWsXN954Y2NDCxqiogzD6nkak+OK2zAumIlUXem2OcMkbEmndoZ17P8E/bD3UaLjMWxdi3PXFnwKcpDqCUjjijmowWFefjonQ5XtOLO/cx9vqdKu2nRq2xtfkw/69UtxzZxBvzzvz16+TxjXpd7DelNbotQ9BEu12YV1ko42YYm0jUqjfZuOJMd0RafTUWSTWVegjZAaF1/7YqG0T8N+9xPs6TuWDvkHkKorcY6YBAZRaFDQPDidThYsWEBubq7GHhYWRu/evTlw4EAzzax1I8RNK2TIkCGkpaXx6quv0qFDB77++msWLFhA+/bt+fHHH1m9evVp5U/w9/fnlltu4bnnnqNNmzZER0fz8ssva1ZXkpKSmDBhAn/729/473//S1BQEM8++ywBAQHuiu2CpmFc/AuS07OFpIRH4bjsFlzd+uHz778j2W2oQGXveqttfm0xtbsZV9485OKNuLPvKnYc+95Hl/4kcvoAsrKySEmIR9n9K86DM5BsTvy3ODHlq5hmfYEaFIpz9FWnnKeqqtj3vIFSmQVAlazzSrDXv8aM+dFbMRbkNjQEhb1HU3n9vexeUIPTrvAt95CoZuJv9uOLS3sRG+jvdc7cIzbNplXXUCPx/tpHmSsgGGfPyae8B4HgXGKz2Zg3bx4FBQUau5+fH6NGjRI1EM8hQty0Uu655x7uueceNm7cyPbt27nzzjtRVZXLLruMe+65xyta6lQ8++yzVFRUcPPNN+Pj48Ndd91FTU2Nps8777zDY489xg033OAOBf/hhx9ESPfp4LBjXPSzxuQcfQ3oDShJnbA98DyWN5/AHmHHGa11ADYl/xlDWC8MkYNx5s7BkfmGu00uXIWreCOGsFq/KVflVuylMyBIhiAdpWNM+Ge48N0uY/ryf0iVZShBYaDXg06PGhaBnJqu8clxHZmJnO9x+t1W5aPJQRNplUn7egYNbXLlGwPZfdUD9J4wCh/g3SFmrltYjAMLWXQHO0yaX8V3o80kB3lEnKKq/HxQm6Z+fILwVxC0PKqrq5k9e7ZXQtWgoCAmTJiAv7+3cBecPaSysrJTO1RcBJSXl/+hbJDN6XNzvmhp9/hHf2cNkZWVpXF4Pt8Ylv2O5WNPEVjV4kv1f78Hnzqh1SX5WLf9DUUpcZv0ob2xpD/nOU9VsG36G0qFxzFb8onBp9+7HN45n/Di90Dxjh8y5coErXSis3k1obRJwPbnaShJHXEVb8S+9SlOVOZWVHg3L4qKOtUMrtxbwtDcSu0YSHwX2Z/My6by0JAETdtTG8p5c4c2N06wSeKLUWEMjjazr9zJ/avKWJ2v3ZJaOimC9HBtVuHm/j2eD8Q9tlxKSkqYO3eul/NweHg448aN07zwXaj3eDo0xz2KlRuBoKWgql7h387hl2qFDeCsXqMRNqDDlKz1k5EkHaYO92LbeD8ntqdU61Hsu/9DaNEqUBsOjHbE6imepCNgrRNzroJUxy1Hl3cYn2fvwXbZJKrDV0KdgO+DzkCNsDHKCr3zPQ92h6Tny6jBvNXuUsb2S+XR7p5kfid4slcg20qcLD1qd9vKHCpXzivi+iRfvjtQg13rh0ycn57uYcKfRtByyMvLY/78+TgcWhHepk0bxowZI4pgnieEuBEIWgj67evRHz3kPlYlnZfvi1y2A8fBLzU2Q8xYdP5tvccLTMEQOxFXricyTs5f4p25U+8LsmeLUfGVKB9pQnKqmI4qmI8o6KtVHNE6HG10OAPngqvuZpOObWonwOMsnF5Qg6+rVvx82GY4H6Rdw6Seifyc6keIueF8OkadxHeXhPHA6jK+3ueZj1OBz7NqvPr7GyReHxAs8oMIWgwHDhxg6dKlyLJWhScmJjJy5EhRVuE8In7SAkELQCorxvzp6xqbq88w1PBoAJSqgzj2f4pcXC/3jd4HU/tbaQxT+9twFawAZ3nD7cl/Rh85FPvOl1DKd2raVKOEPVFPXowJq6Ij0ujE2IAuscbdwN412gKZA4/WbketC05Fd+fDzG3vh1F3ahFi0ku8MziY5EADz26uaLTf6Fgzrw8M9nIkFgiaix07drBmzRove8eOHRk4cKC73I3g/CCeDAJBc2O3YvnPNHTF+Rqzc9y1qM5KHFnv4jq2GBpIbGdMvA7J1HieDMkYgCnpDhx7/uN9brubMSZcDYClx8s4D36GM/tbTZ9NlT4sLA1ARcIkKaT42OnkZ6OtxUGh08COXAs7Di/RJN1rU+UgscKBQ9Ljd/fDXJd8eo6TkiTxUPcA2gXombqyVLMVFWrW8WK/IK5r7yNWbAQtAlVVWb9+Pdu2bfNq6927N+np6eKz2gwIcSMQNCeKjGX6c+gP7dWYHaOuQG6Xim3Tg+5Q6/oYokdhTDh1mL2hzWhcR+egVHgyRRvir8bY9ib3saTTY0qagj58QG0oedE6sisq3cIGwKHq2Fnjw84aHwyouJBAD6Bdgh94tBIJKB09mbYdz9yJ8Kr2vsT563l4TTkHKlxcmmjhub5BhFtE+KygZaAoCsuWLWPfvn0auyRJDB06VNSLakaEuKmDqqpCYV8gNCVr8oWA6et3MGzRZuV1de2L46Z7cR2b36Cw0QV2xJR0B/qQrk26hiTpMHd5vHbrqSaXCt/BRCff2eBnXR+Uhj4ojcqaUn799QlUqhoYkVph0wDBNhe9j1Vjj4zF79o/Xn28b6SZZZdFoKigb8K2lkBwvnA6nSxcuJCcnByN3WAwcMkll3iVuBGcX4S4OY6fnx9lZWUEBwsHxZaOqqqUlZUREOAdcXMhYZz/I6b5P2psckIStnv+hYqMs57jsOQbXytqwvuf9mdUZ4nAp9drABzNyjrp+Yqi8MOK96myNSxsGsLfIdP7WDXDj1RgkVWsUx4Ck/nUJzYBSZLQiz9JQTPjcrmoqqqisrKSyspKMjMzKSoq0vSxWCyMGzeOiIiIZpql4ARC3BzHYDAQEBBARUXjTowno6KigsDAwLM8q5ZFS7rHgICACzryQJdzANPXb2tsSkg4tr+9CD6+uA7/iGqv8+DUGbGkv4DOcu4fmku3/cKBvF0a25AuE0mO7cL2g2vZfmgjdkc1iioRWw7jjuTTqdiK4fhimnPwWOROPc/5PAWCc01paSkHDx7k0KFDFBcXn7RvYGAg48aNO+u5twRnxoX77XAOMBgMZ/zBLCgoaPXLkBfDPZ4NVNmGXJqBas1DH9oTnV+iVx/jT59Q1VNHTQcjhjKVwA06nPe/hBoaieqqxlHPsdcQO+m8CJv9R3eyNOMXja1tVCq9Ol/JvBwHv1mjWOochb96jEqCUfxNxOg/pJtau7WmhEdhv37qOZ+nQHCuqKqqYvfu3Rw6dMgru3BjhIeHM3bsWHx9fc/t5ARNRogbgeAsoFjzkYvXIRdvQC7N8GT/lXSYUh/AGDPW3Vc6lEmNZS22pNo/P1eEROkEH8zBKnrAefhHcNZZQdT7Yko893WSCsuO8u2ydzWRT7LOn+8d1/J/3xYgu80Gaog7Pjd4b9gDjIu9Av+So8g9B0NA8Dmfq0BwtlFVld27d7N+/XqczoaTXDZEXFwcl1xyCUajSCbZkhDiRiD4gzgOfY3zwGc0FKqNquDY8x9UWwHGdjcDKq6Nz2CvV81bxYpty6OYO/4N55GZmjZjwtVIpnO31K2oKj/v2MqmLe+iUzx1m1RVYpZ8HUdKGn8bHdbGzFejQrEYI3E12ksgaNlUVFSwYsUKjh49etJ+/v7+BAQEuP8fFRVFbGys8NNsgQhxIxD8AZTqw40Lmzo4D32Jai9EqqzAGdrI3r1cg33H81qbMQhj/JVnZ7L1sMsq7+2qYuGOJXSx/YS+bq0FYAMjOEJyg+fG+em5OcWXB7sGYDGIB7vgwuTEas26detwubzluSRJxMbG0rZtWxITE8W20wWEEDcCwR/AcegrGhI2kiUK1VEKiqe+jCtvvnc/l4RqaFwYmdregGQ4+w9Um0vl8rkF6At/o5e0nPqR3TvVXqxnpMbWMdjAxEQfLk2w0D3MKN5WBRc8mzZtYsuWLV52o9FI7969SUlJwWw+O1F/gvOLEDcCwRmiVB9Bzl+msRkTrsbQZgySbwJKxW5sW/8JrsoGz9fVqPjGPYg9uBzngU+82iVzJIbYCWd93qoKD60pJaDwGzpKm73aM83jSUwYwxP+RmL99MT66WkXoCdOlDoQtCJycnIaFDYxMTEMHTr0gk81cbEjnlYCwRlSf9VGX6HDf3kpcs9c5E7R6IM64dPrdWxbn0S1HdOcq6tRCdqTiGPiWEyShGT0x5H5lmY8Y/ubkXRNqyBsc1gpqcwn0DcUP0vASVdVfjxmYM/+uQysJ2xUyUiP7nfwTPcBYlVG0KqxWq0sXbpUYzMajfTr14+0tDTx+W8FCHEjEJwBDa3a+G+2YcqeDctmo5osyJ17oVp8sMjRVMSXIfvbANBZVULmOXD+dSocf4gaYyci6X2x730LXNXoIwZjiB7VpLnszdnKF4v+487abDZaCA2IIjQwkuSYLvRIHoxeV/unvibfzvcH9jEB7RaZryWQW0Y9SFxE0h/6uQgELR1VVVm2bBlWq1VjHzt2LG3atGmmWQnONkLcCARnQO2qjccB11CqYM72HEsOm6asQtgmsKboUSwSPpku1MR05I49NGMaokegD++L6qxEskQiSaeuIqyoCr+u+VRTjsLutJFXkk1eSTY7D21gQ+Zirhz8ZxRTDPct2s1o9TskydPfZPTlrgmPExYYfSY/CoHggmLHjh0cOXJEY+vRo4cQNq0MIW4EgtOkoVUbv62NVVuqRZLBd4+nwGTNlVMa7mfwQzL4NXkuhwuyKK8uOWmfo8XZTJ/1Tw6bRzDAvhmT5KjTquPGEfcKYSO4KCgqKmL9+vUaW1RUFD17iozarQ0hbgSC08Rx6Gvqrtroj6/aOIeMRzWZMWxeia60qPHzR1+Fktb9rMxl+4G1TeqnKDJx1oVeUVET+91IUkznszIXgaAl43Q6WbRoEYri+ds1mUyMGDECne7Uq6SCCwshbgSC00CpyUHOX6Kx+W91oaR2xz7lIdAbcNzyALpDe9Ed2Q86ParZAiYLqtmCGhaJGnF2lr9lxcWOQxs0thtG3E98RBKLDx5h3tYFhDm2Nnp+z5Rh9Eu75KzMRSBo6axevdqrduCQIUNEVFQrRYgbgaCJqLIN+/Z/o4mQKlMwlgZie+ZJ0B//c5IklHapKO1Sz+l8DhzdRY3dE2ZuMfnSIa4bOTUSd20KxS5PJolODOdXfKVqzbmxESlM6n+riAoRXBTs37+fvXv3amxpaWm0b9++mWYkONeItTiBoAmoihP79udQqrM0dr9tMo67n0QNCT/vc9p+cJ3muFNibwx6Ix/srsZ+3L1nP135kgfIUru6+/lbQrh55P0Y9OLdRtD6OVFaoS7BwcEMGDCgmWYkOB+Ip5tAcApUVcG++zXkko0auylPRkq/GVfnXud9Tk6Xg12HN2ls3dr1p8al8EWWdpWmd3Qot3W+h07mXMqqCzHY/fH3CTyf0xUImgVFUViyZImmEKZer2fkyJEYDOLrrzUjfrsCwUlQVRVH1rvI+Us1dkOxgl9uGo6Hb22Wee3N3Ybd6cnT4W8Jol10R77YZ6Xc4dk2CzXr+GF0+PH6T8kkkkxWVlYDIwoErY9NmzZRUFCgsfXt25ewsLBmmpHgfCG2pQSCk+A89BWunF81Nn2FQvAKHc4p/wCdvpEzzy31o6S6tOuLJEm8v1u7anNrB19R2FJwUXL06FEyMjI0toSEBDp3FtGBFwNC3AgEjaBUZ+M8+IXGpqtRCZnvxHnVvWct6ul0sTmsZOZkaGxd2/VnXYGDHSWe5XedBFNSm54zRyBoLeTn57NgwQKNzdfXl6FDhwon+osEIW4EgkZw5S+jbmSUZFcJWeBAbd8H17CJzTav3Yc34ZI9IibYP5z4iCQ+qLdqMy7eQmKA2HkWXFzk5OQwe/ZsHA6Hxj58+HB8fHyaaVaC840QNwJBI8iFazTHARtd6B1+2O94xF0T6nygqiqKWpt4rNAq895abeRH13b9yLcq/HJIWyvnz2li1UZwcXHw4EHmzZuHy+XS2Hv27ElsbGwzzUrQHIjXOoGgAZTqXJTqg3UMKuYjMvbb7kcNjTjl+XtztrI/bxfhgdF0SuyFn+XMopM2Zy1n7oZvsDqqMRnM2FQTYa4qTabhjY6ubNhTjcuzyERKkIFhMeYzuqZAcKFhs9nYt28fa9eu1dRZA+jVqxc9evRo5ExBa0WIG4GgPoqC9PtLUCd1jbFQRek8GNfA0ac8fc2u+cxe/6X7eNbaGbSL7kiXtn1PS+jsOLSen1Z95D52uOzosGuETbEayVd7AoBKzbl3pvmhE74FglaKoijk5uaSm5vL0aNHKS4ubrDfgAED6NKly3menaAlIMSNQFAXVcX0zXSqlT2AJxLKVB6I7U+n3o7afXgzc9Z/VW9IlQN5uziQt4tZa2fQNiqVTom96ZjQiyC/0AbHOVywjx+Xv3/K6e6lO/ULRvkZJK5P9j3luQLBhUheXh4rV66krKys0T6SJDF06FA6dOhw/iYmaFEIcSMQ1CFq5e8YVv+C43rtlo56+VNI/kEnPTencD/fL5uOitpoH1VVOXhsDweP7eH3dV8QF55E13b96JE8GB9zrY9MSUU+Xy76Ly7F2eg4AIXEksEgL/v1yb4EmYQ7naB1YbfbWb9+PXv27DlpP51Ox8iRI2nXrt15mpmgJSLEjUBwHMOKOcQs+wVrO11tHPVxJFM0UptuJz23tLKQLxb9F6fsidDQSTrCg2IoKMtp9Lycov3kFO1n0ZYf6ZE8mO5Jg/hxxfuamlEAxWGT+b4oFRMOjNhJCjDxxfgOFK8qY0GuXdP3TuFILGhFqKrKwYMHWb16NVartdF+oaGhxMTE0LFjR4KDg8/fBAUtEiFuBAKAqgrMX70NgD1em5jPED3kpKda7dV8tvA1qm3aisOX9r+VPqkjKCo/xo5D69lxaD35pUcaHMPhsrNuzyLW7Vnk1dYx5VL+urc7AE4sADzQO5Q2fga+Gx3Ge7ureWlLBVVOlUfTA+gYYmzaPQsELZyamhrmzJlDbm5ug+1JSUm0bduWmJgYLBbLeZ6doCUjxI1AAJh++xKppgpVB45Y7ZaOPuLkBfZ+WPEeReV5GtuQrhPpkzoCgPCgaIZ3v4zh3S+juCKf3Yc3sSt7I0cK959yXt3bD+THmhGAZ3WmW6iRSYm1D3JJkri7kz93pPphk1UCxXaUoBXgdDrZsmUL27Zt84p+AggKCmLIkCG0adM8iTQFLR8hbgQXPVJhHsYFMwFwROlQTXUcdI1B6AJTGz03r+Qwe3O2amxd2/Xjkp7XNNg/LDCKwV0mMLjLBMqrS9iybyXrMxdRWVPm1bdtVCqJHW5mzhxt2+M9A72yrJr0Eia9iI4SXPgcOnSI1atXU11d7dWm0+lIT0+ne/fuovCl4KSIT4fgosf040dIrlrnXXu8duXDEN4PSWq8flTG/lWa49iwdlw56E500qlXUIL8Qhne/TIGd5nAruwNrNm1gJyi2tWcmLBEJo+4j5uWan0MekcYGRMn8tcIWid79+5l2bJlDba1adOGQYMGERIScp5nJbgQEeJGcFGjO7QX45qFQG2hhfr+NvrwxrekZEVm235tFuOBncdiNJhOaw4GvYFu7QfQrf0A8ktzqLKW0zY6jee3VLEsT+ss/HgP71UbgaA1kJuby/Lly73svr6+9O/fn/bt24vPvqDJCHEjuChRVRln9g+w5TucQ4wYShUkByj+dR6eOhP60MYzm+7L3U6Vrdx9bDb60DGh1x+aV1RIHFEhcXy9r4bXt1Vp2gZHmxgusg4LWiElJSUsWLBA418jSRJxcXGMHDkSk+n0XhgEAiFuBBcdqqriyHwL19E5EATOID11E/adQB/aA0nfeATGlv0rNcdd2vY97VWbhliTb+eBVaUaW6SPjulDQsSbq6DVUVVVxdy5c3E6tXmdRowYgaIoQtgIzggRWiG46HAe+LRW2JwCfXj/Rtus9mr2HN6isfVIHvyH53ao0sXNi0pwKB6bWQ9fjgwj3l+8iwhaFw6Hg3nz5nk5D/fr14+kpKRmmpWgNSCeloKLCufhmTizvz11R8mI4STiZseh9ciKp/JwSEAECZEpZz4vRWVlnp1p68sptiuatrcHh9AnUry9CloXFRUVLFq0iJKSEo29U6dOdO3atZlmJWgtCHEjuGhw5i3EsU9br0myqfhtd+HsEI8zPhhX5UF0egOmpD8hmRqPytiyT7sllZ406LS3jGRFZWGunV8OWZlzxEqp3Tufx/+lB3BNe1EnStC6yM7OZunSpTgcDo09MTGRAQMGiO1XwR9GiBvBRYGraB2OPa9rbJJTJWSRA0ONLzV3/htDWBRZe/eScopie0XlxzhSuE9j65F0eltSJTaZyQuL2VDYeP2oK9v68Fh6wGmNKxC0ZBRFYcOGDWzbts2rLTIykpEjR6LTCW8JwR9HiBtBq0cu2YJ9x3Og1tnukVWCljgxFqlY7/0/1LCoWnsT3hgz6jkSt41KJSQgosnzKbHJXD6vmO0ljQubETFm3hkSgk68wQouULZv38727dux2z3pDFRVRZZlr76JiYkMHz5cJOYTnDXEJ0nQqpHLdmLb9i+oW2FbVQla4cScp+AYdQVyn2FNHk9RFTL2r9bY0pO8K3M3Rqld4YpGhE2oWceEBAuXt/VhVKxZCBvBBcv+/ftZu3btKftJkkTfvn3p2rWr2IoSnFWEuBG0WuSKvdi2PgmKNhFewFoXlmwFOSEZx/VTT2vMQ8cyKa8udh8b9SbiYnrzv+2V+Bt13NzBF6Ou4Yd0mV3hynlFbKsnbAZEmXgsPYBB0WYMjZwrEFwoVFRUsGLFilP28/PzY+TIkURHR5+HWQkuNoS4EbRKlKqD2DIeB7lGY/ff4MR3r4xq8cF2z7/A1PSkeC7ZxeKMmRpbYmxPxs6t5FBl7VL7r9lWvhsd5iVwyuwKV84vIqNYK2yGRJv4dnQYvgbhZyC48JFlmcWLF3vlrKmLXq+nbdu2DBgwAB8fn/M4O8HFhBA3glaHYs3DumUauCo1dr8tTvx21YoQ++0Po0bHNXlMVVX5dfUnZOfv1di/KuzMoRqPD8GSo3b+trqM/w0Kdi+zF1plrp5f7LViMyjaxDeXCGEjaD1s3LiRwsJCjW3AgAGkpnqKz+p0OvT6xuu1CQRnAyFuBK0Ox953wVmmsflud+G3rVaEuHoPxTVg1GmNuXz7b14ZiYv1SayraefV94usGtoFGHioewA5VS6umFfMvgqXps/AKBPfXRKGn1EIG0Hr4MiRI15RUImJiXTu3Fn40wjOO0LcCFoVijUPuXi9xuaTqeK/2cWJx6tz5OWnNeb2g+tYuPkHja1GF8ZM1w00luT72c0V6CX4cE81OdXa6JABUSa+Gy2EjaD1UFNTw9KlSzU2Pz8/hg4dKoSNoFkQT1dBq8KV+xu19b1r0RFBwFq7W9go4VHIHRsvhlmfwwVZ/LBCm/jPgQ8/yrdhw5Ncb3iMmWCT9iH+r00VXsJmVKyZH0aH4S+EjaCVUFJSwq+//orNZnPbJElixIgRWCyN12YTCM4l4gkraDWosh3n0fkam+9elbqSwzV4HDQxSVhZdTnT576BUqfMgqzq+U29iTLC3bZLYs18e0kYX4wK42Sa5fK2Fr4eJVZsBK2H7Oxsfv31Vyortf5tPXv2pE2bNs00K4FAiBtBK8JVsEzrRKzzxXd9jqaPc/C4Jo2lqirvLfoEg6J9aC/hCnJp7z4eHmPm85FhmPUSg6PNvDW44ZINt6T48vGwUEx6sUQvuPBRVZWtW7cyf/58r8io+Ph40tPTm2diAsFxhM+NoNXgyvlNc2yuikGSPUX5XB17oEY07W1y24F1VJVoq35vUoeym17u40mJFt4bGoKPwSNYJif5crDCxUsZHlF0T2d/nusTKHwPBK0CRVFYvnw5WVlZXm1paWkMHDhQlFAQNDtC3AhaBXJFJkpl3TBtCb+VRzR9XEPGN2msKms5P6/5TGPLJ46OqVcwLsBEYoCBjsEGOgQbGzz/0fQA4v31LD1qZ1y8hava+QhhI2gVyLLMokWLyM7O1tglSWLAgAF06tRJfNYFLQIhbgStgvqrNgZDEsa8He5j1eKLq/fQU46jqiqz1szA5ap222RVjzH+Jl4bGNakuUiSxE0pftyU4tfE2QsELR+Xy8WCBQvIydFu9ZpMJkaNGkVcXNPzRgkE5xohbgQXPKqzotbfpg4++7Vvj65+I8B86siNQ0W72HV4k8a2jkt4q1fKH5+oQHCB4nQ6mTdvHnl5eRp7YGAgY8eOJTg4uHkmJhA0ghA3ggse59F5oDjcx5IpAsvKXdo+QyeccpwqaznrD8zV2I6pcUTGXULHkIa3oASC1o7D4WDu3Lnk5+dr7MHBwUyYMAE/P7FCKWh5CHEjuKBRVQVX7u8am9naFp3L42+jtIlHSep0yrHmbPgau8vqPpZVPQu5mm+7B529CQsEFxCVlZXMnz+fkpISjT00NJQJEyaI2lCCFosQN4ILGqU0A9V2zGOQjPgtrLdqM2Q8nMLJsayqmB0H12lsa7mEjlHx9I1senFNgaC1kJeXx8KFCzXJ+QDCw8MZP368SNAnaNEIcSO4oHHla31tjM54DMc84kY1GHENGtvguTtLnCzOtWFXwJo7G0VV3G0lagRbGMx33QLOzcQFghbMrl27WL16NaqqauyRkZGMGzcOs1kIfkHLRogbwQWLqjhxFa7S2HxXHdIcOy+5EjXYO8ppbb6dy+cVYZfBgIMprMJSZ3Eng0F0CjFzSax4iAsuHhRFYfXq1ezevdurrV27dgwbNgyjUfifCVo+QtwILljkki3gqnIfS4oRc7YneZ7q649j0s1e57kUlYfWlGE/XvYplS1YJI+vjVX1IZN03u0WIHJ2CC4aVFVtNDlfr1696NGjh/h7EFwwCHEjuGCRC5Zrjs377Uh1VtEdk24G/0Cv8z7bW8PO0hP1olS6s0bTvpO+pEf4cUVb4SwpuDhQVZU1a9Z4CRuDwcDw4cNp165dM81MIDgzmi1H9gcffMDAgQOJj48nPj6e0aNHM2/ePHe7qqq8+OKLpKWlER0dzcSJE72WSu12O4888gjt27cnJiaG66+/ntzc3PN9K4JmQJUduApXa2w+BzwFLpXwKJyXXOl1Xpld4fnNFe7jePYRJhV4xkXHVT1H88PocAw68ZYquDjYtGkTO3fu1Nj8/Py47LLLhLARXJA0m7iJiYnh6aefZtmyZSxZsoShQ4dy0003sWNHbVbZN954g7fffpt///vfLF68mIiICK688kpN9dlp06Yxa9YsPvroI2bPnk1lZSWTJ09GluXmui3BeUIu2QRyjftYZ1Ux5nscgh1X3wkmb3+Zl7dWUGz39OspaQVS27A0/tw9nmCzqI0juDjYtm0bW7Zo66j5+PgwceJEwsKalpVbIGhpNNsTfOLEiYwePZr27duTnJzMk08+ib+/Pxs2bEBVVaZPn86DDz7I5ZdfTqdOnZg+fTpVVVX88MMPAJSXl/P555/zzDPPMGLECNLT03nvvffYuXMnS5cuba7bEpxrXE6M839EXTNdYzZny+4tKTmxA67+o7xOzSp38v4uT1mFIIpIIFPTp2NMn7M/Z4GgBaKqKjt37mTdOm0KBLPZzIQJEwgKEvmdBBcuLeL1VJZlfvzxR6qrq+nbty/Z2dnk5+czcuRIdx8fHx8GDhzo/kPMyMjA6XRq+sTFxZGamur1xypoJcguLG8+iemb/+E012ZLVVXYUOHLm/pIPuoSgdUg4bj+bmigKvET68tx1fHJGWRcq2mPDWtHRICojyNo/djtdpYsWcLq1dqVS6PRyLhx4wgNDW2mmQkEZ4dmdSjeuXMnY8aMwWaz4efnxxdffEHnzp3d4iQiIkLTPyIiwl3bpKCgAL1e77VsGhERQUFBASejoWiAs8G5Grcl0Wz3qKrEz/4C/61rsSXqUI21/jC7aiwsKgsAMxRHGPnMrz19jQFQb55rSnXMy/EkHQvnKMnyRupm8WgX2g1JksTvsZUg7rFhysrK2L17N3a7XWOXJIlOnTpRXl5OeXn52ZriH0b8HlsHZ/seU1JOXu+vWcVNSkoKK1asoLy8nF9//ZWpU6fy22+e6s71ww5VVT1lKGJT+pzqh3ImZGVlnZNxWxLNeY/G377EvKU2OsrWVg+AVZZYVKpNsrfH18VAfxft2nR02xRV5dafC5Bw0JY9pLOKOOkgdfOTBfgEM6r/ZRw8cFD8HlsB4h69cblcbN68ma1bt3q16fV6LrnkEhISEs7mFP8w4vfYOmiOe2xWcWMymWjfvj0APXr0YPPmzbzzzjs8/PDDQO3qTFycZ5ugqKjIvZoTGRmJLMsUFxcTHh6u6TNw4MDzeBeCc41hzSLM338AgGIAe1ztltOisgBqFO/tpzkbvubuS/+F7vjW1KJcO5Vle7iZnwmWSrz6A/RJG4lBLzIjCFofqqqSlZXFxo0bqa6u9moPDQ1l5MiRhISENMPsBIJzQ4vwuTmBoig4HA4SExOJiopiyZIl7jabzcaaNWvo168fAOnp6RiNRk2f3NxcMjMz3X0EFz7S7gz037yEI0rC2k5HVW8DGCQO2UzsqG44D01eSTYZ+z2Zi7/YvInLmNGosIkNa8fATmPOyfwFguYkJyeHmTNnsmzZsgaFTbdu3bjiiiuEsBG0OprtVfVf//oXY8aMITY21h0FtXLlSr777jskSWLq1Km89tprpKSkkJyczKuvvoqfnx/XXHMNAEFBQdxyyy089dRTREREEBISwuOPP07nzp0ZPnx4c92W4CyiHt2FffdjVF2uB/Ruu1OBuSUnr/m0cPMPdG7bh01HDtKm+BP0knd6gHbRaQzoNJbUuHT3Ko9A0BooKChg48aNjeb98vPzY9iwYcTGxp7nmQkE54dmEzf5+fncddddFBQUEBgYSOfOnfnhhx8YNao2hPeBBx7AarXyyCOPUFZWRq9evZg5cyYBAZ4vtRdeeAG9Xs+UKVOw2WwMHTqUd999F71e39hlBRcIquzAvuUp5GDvtlUV/pS5PB9dCYlrht7NTys/xKU4Aai0lvHL6k/Ylp2BUXJqzu+RNJgBncbQJizxXN6CQHDeKSoqYtOmTRw+fLjBdp1OR6dOnejZs6cofilo1TSbuJk+ffpJ2yVJYtq0aUybNq3RPhaLhVdeeYVXXnnlbE9P0Iyoqopj8/PIPlVebfkOA+sqfDW2vmkj6da+P8dKD7Ni++9u+/aDa6nvWh7d/jquGjLxXExbIGgWFEUhLy+PXbt2cejQoUb7JSUl0bt3bwIDvUuSCAStDeFBKWhxuHJn4arU5irS1RiQEvqydH8hKp7yCYG+IVzS81oAhnadxOasFVTbKmiIbfrRPDFowrmbuEBwnlBVlcrKStauXcv+/fupqalptG+bNm3o16+fV2oNgaA1I8SNoEUhl27FsfddjU1fruCb+BDF7bpwcPPDmrZL+9+KxVTrWGwx+TCqx1X8uuZTr3E3q0Po32kSZr2oFyW4cFFVlZycHDZs2EBxcfFJ+0ZGRtK7d29iYmJENW/BRYcQN4IWg2IrwLbjecBT+0lyqATsb4vziuFkbPtV0z8uPImOCT01tl4pw1i3ZyH5pTlu2w61D+ukcXyQ5n9O5y8QnEsKCgpYv369O5FpY4SHh9O7d2/i4uKEqBFctAhxI2gRqKqKfedL4NRuKQWudCLf/FdUIGPfKk1bj+RBXuPodDquG3YPr//2KjpnGdsYwCrGcUWiLzF+wtFccGFht9vJz89n7969HDx4sNF+RqORtm3bkpKSIlZqBAKEuBG0EJSq/SjluzQ2vy1ODEG9sKWlczh/LyWV+e42vc5Al3YN5zOy6iN53/UQKjLq8RDyuzr5nbvJCwRnCVVVOXz4MEeOHOHYsWOUlpaetH98fDwpKSkkJiZiMIjHuUBwAvHXIGgRyAUrNcfGPBm/bTLWf/4JQJOUDyA1Ph1fc8PbTO/trkJR4URunK6hRvpHms76nAWCs83GjRvJyMg4Zb/Y2Fiio6Pp2bPnKfsKBBcjQtwIWgSuQq148dknI/cagtI+DafLwY6D6zXtPZIHNzjOgQoXH+7WZmK9q6OfWKYXtHhyc3NPKWzCw8Pp27cvsbGxF0WxRYHgTBHiRtDsKNWHUWuO1DGomHJV7H+6A4A9R7Zgc3pCXf0sAaTEdm1wrCc2lOPw+CMT46vjmva+DfYVCFoKNpuNZcuWNdgWGhpKVFQU8fHxJCQkCKEuEDQBIW4EzY6rcLXm2HRMQe06BCWuHeC9JdWt3QD0Ou+P7pJcG7MP2zS2f/UOwscgvgwELRdVVVm5cqVX7adhw4aRmJgoMgkLBGeAEDeCZkfOX645NmcrOC+tzSJcWVPGvtztmvb0BrakXIrKP9aXa2x9I0xc277h4poCQUuhoUio7t2706FDh2aakUBw4SOqBQqaFcV6DKX6gMegqpjKg5C79gZg28G1KKpnnykqJI42oQle43ySWc3uMpfG9lK/ILGEL2jRlJeXs3q1duUyPDycXr16NdOMBILWgRA3gmZFrrclZSxQUXqPBV1tpNOWfdooqvSkQV6CpcQm8/xmbX6cG5N96RkhIqQELRdFUVi6dCkul0eU6/V6RowYIYr/CgR/ECFuBM2K6+gSzbH5sIxz8DgA8koOk1/qcTSWJInu7Qd6jfFiRiVlDtV97G+QeKqXKA4oaNmsXbuWgoICjW3AgAEEBwc3z4QEglaEEDeCZkOxl6BUa8NZjbRHjUkEIKPeqk1yTBcCfIM1tk8zq/loj9YR86HuAUT7ijdfQctl79697Ny5U2NLTEwkLS2tmWYkELQuhEOxoNmQC1ZBnR0mQ5GC0vey2jZFZtuBtZr+6UmecguqqvLK1kpe2FKp6dM2QM/UTqKGlKDlUlBQwMqVWuHu5+fHkCFDhI+YQHCWEOJG0GzI2fM0x+YccF06AoADR3dSZfNEP5mNFjom1DpZyorKY+vK+aDeio1OgtcHBGMRod+CFkpNTQ0LFixAlmW3Ta/XM2bMGHx8RGSfQHC2EOJG0Cyozkpk+37tyk1Qb1y+tasuGQe0jsadE/tgNJiwyyp/WV7Kz4esmnaLHj4eHsrIWMs5n7tAcCbIssyCBQuoqanR2IcOHUp4eHgzzUogaJ0IcSNoFlx5y0HyOAHryxTU/lcBYHda2Z29SdO/e9IgVFXl7gaETZBJ4ptLwhgQJZKdCVoua9as8XIg7tatG8nJyc00I4Gg9SIcigXNgrJ/lubYXGBB7tgDgF3Zm3DKDndbkF8obaNTeWN7FT/VEzZtfHXMmRAhhI2gRXPw4EF2796tscXGxtKnT59mmpFA0LoRKzeC845izcelHNJsSemjh6Icz22zdb92S6pb+wEsOerg6U3aXDbJgQZmjg0jwV98jAUtl+rqalasWKGxBQQEMHLkSHQ68X4pEJwLxLeC4Lzz/+3deVxU970//tfsDAgM67AJyCaLKIqK4r7hQoxNbJrtpq2pbW4e6c3SxJuY5KZbGmNM+m3an0l6b5a2SdomMSYxiYhaNaKAWySIC6IIorLILjAzMDPn94d14DDgOsyZGV7Px8PHw/M+nxneH44ybz7ncz4fy4l/iOfaNFlhnXE3AKC9sxmVtcdE7YP1WbhjVzOEPjGdWoYNOSxsyLUJgoBvvvkGJpPJFpPL5Zg/fz68vDg/jGio8NcGcirB2oOexh2imOZSJBA2EsDl7RaEPmWMPiAaPz/ghbY+i/TJZZcnD8f6srAh11ZWVobz58+LYpmZmZxATDTEWNyQU1kufANB0TufRtYtQD7mXttx/x3AjwnjcazfnlG/zPTjU1Hk8pqamrB//35RLDw8HGPHjpUoI6Lhg8UNOZX55D9Ex5pzKlgnzAVwZbuFc7ZzVsjweUuqqP2do7R4dAwX6SPX1t7ejp07d8Jq7d30Va1WY/bs2ZxnQ+QEHNcnp7F2nIUF4iF6ddAcWJSX/xluKxNPuqwREtEFX9txWoASf5qm4yqu5LKam5tRUlKCyspKCIIgOjd9+nSMGMHCnMgZWNyQ05hPfCg6VjVYYZ1/P3qsAl4paUVTZRF8+pw/gQzb30eOUODDeUHwUfG3XnI97e3tKC4uRnV19YDnExISEB8f7+SsiIYvFjfkFILFBHPLXqDPfpYaQyyOK4Pwn19dhKFpP3JkvftEdQtqVCIVUT4K/DTFBz9K8oFOw8KGXE9XVxe+/PJLu5WHr9Dr9Zg2bdqA54hoaLC4Iacw1+RDUPRODJYZBWwJXI4fbmqAySLgXuwWtW/1zsQ7WWHIjfaCUs7bUOS6SktLByxsdDodMjIyEB8fz3k2RE7G4oacwnJ6g2htG3m1GvdqUmGVAbEoR7Csvk9rGdYsvgOBftxIkFybwWDAsWP91mUKDsb48eMRExPD+WFEEmFxQ0PO2n4KFpl4T52/tk6HNezyb7MT+o3ajImdhEA/vdPyI7pZpaWloh2+vb29sXTpUiiV/NFKJCWOldLQ6mgHvvyNKKS6YMUfAnIBAGE4i0hZlej8jPRcZ2VHdNOMRqPdqM24ceNY2BC5ABY3NGTk1RXw/tVD6NbWieLHGkaiVhMAAMiUfSM6Fx+ehoigWGelSHTTjhw5ArO5dx6ZVqtFcnKyhBkR0RU39StGeXk5qqqq0NLSYreWAwDce++9A7yKhhPlnnxo/vIaLNoeWHS9O3YLVuDn3j8BAASgAXEQ75TMURtyB0ajEUePHhXFxo4dy1EbIhdxQ/8Tq6ur8dBDD2H//v0DFjUAIJPJWNwMc8p/fQ6vv/0BAGAarRCd22dIRJkmBgCw1KcQ6POQSURQDOLCxSsSE7misrIy9PT02I69vLyQkpIiYUZE1NcNFTdPPPEESktL8bvf/Q7Tpk2DTqcborTIbXW0Q/Px/9oOjTHi4uYLYxYAIEHbgQDjIVj7nJs+JpdPl5DLM5lMKCsrE8XGjh0LlUolUUZE1N8NFTdFRUV49NFH8fDDDw9VPuTmVNs2Qma8PBxj8QHMwb3TuqyCDFsM46GSA/8VcQinKnufMgn0DUVazCSn50t0vdrb21FRUYGKigrRqI1Go0FqKkcciVzJDRU3/v7+CAoKGqpcyN0ZOqHe9qnt0DRSPGpzyBSPBosOT47R4HzFHtG5aWmLudAZuaSamhqUlJSgrq5uwPMctSFyPTf0aXLffffh888/H6JUyN2pdnwBWWfvFgrGOLXofF7XBADAOFUZDN2dtrhW44PxCdOdkyTRDTh+/Di2bNkyaGGj1Wo5akPkgm5o5GbBggXYuXMnli5dihUrViAqKgoKhcKuXWZmpsMSJDdhMkCV97Ht0KoBeoLFTfIMExDuLceFml2ieGbiLKiU4kKISGpVVVXYu3fvgOfkcjliY2ORmZkJtZr/dolczQ0VN0uWLLH9faD/9IIgQCaTobm5+dYzI7ei2vUV5JdabcfGOC0g632i7mj3SJw1h2BFRAPO15yxxWWQYVLSbCdmSnRttbW12LFjh91ToaGhoUhMTER8fDw0Gs0gryYiqd1QcbN+/fqhyoPcWbcJqs0fiUKGDD2A3qH8zf++JRVrLkZjn3YJkWO41QK5lObmZmzdulW0rYJMJsP8+fMRGxsrXWJEdN1uqLi57777hioPcmPygs9h1jRDppNB2S7AqtXArGkE+vzSm9c1AT4yA1obDoheOzl5npOzJRrcpUuXsGXLFnR3d4vi06dPZ2FD5Ea4nCbdEuulanSZ/gJh0b/nHVgFyAQtIBhtbSp79CjvicSdvvtg7uh9hNbfJwhJkeOcnTKRHUEQcOrUKRQVFcFkMonOTZw4kdsqELmZqxY3a9euhUwmw1NPPQW5XI61a9de8w1lMhn++7//22EJkmvrObAGglefIRq5DAKMojZ5XeMBCIjuKYKlT3zS6Dl8/Jsk19HRgT179qCmpsbuXGpqKjIyMpyfFBHdkqsWNy+//DJkMhkef/xxqNVqvPzyy9d8QxY3w4eloQRmVF2zXV5XJkaiEhbTRVtMIVcgM3HWEGZHdHWCIODEiRPYt2+faFG+K+Li4jB16lSumk3khq5a3LS0tFz1mIY3y8HfA159AwKgEH8Q7DcmoKQ7Fncq/4G+wzapMZMwQuvnnESJ+hEEAUVFRXabXwKAQqFAZmYm0tPTObJI5KY454ZuinBqF3q8GkQxH9NcCAsfgbWzGp8cOY5ddWZ80TkZvmhFuOWYqG0WJxKTRKxWK7755hucOnXK7lxYWBhmzJjBffOI3ByLG7pxgoCe7/4/IKA3pGxRQLjtMchUXpD7p+KluiCc7bAAEJCDDyHr8+iUPiAK0aGJzs+bhj2z2YwdO3agurpaFFcqlcjKykJKSgpvQxF5gBsubk6cOIG33noLJSUlaGtrg9VqFZ2XyWQoKSlxVH7kivZ/jJ6ADlFIHfkDyDSX71FVtJn/XdgAo1GCaJn4N+RpaYv5AUJO193dja1bt6K2tlYU12q1WLJkCQIDAyXKjIgc7YZuKO/btw9z5szB119/Db1ej6qqKsTGxiI8PBw1NTXw8fFBdnb2UOVKrqDbhJ7K90UhZYcWmPyA7Tj/3OWnpbTowAx8JWo7KiwFGfHThj5Poj4sFgvy8vLsChtfX1/cfvvtLGyIPMwNjdy8+OKLiIiIwL/+9S9YLBYkJCTgF7/4BWbNmoV9+/bhBz/4AV588cWhypVcgGzLG+jWm0UxZfJPIZPLUdNhxqvfXcIHFV0AgBn4GlqZobedQoVl2T/mqA053cGDB9HQIJ4jptPpsGTJEvj4+EiUFRENlRsqbg4fPoxVq1ZBp9PZnpy6clsqKysLP/rRj/C73/0Oc+fOdXymJAnBYoS17TgsbccQVFuETk0FgN7iRNEdgMbIHPy+uBXvlXei+993KWNQjtGy70TvNWfc9xDkF+bE7Iku7xNVWloqioWEhGDRokXw8vIa5FVE5M5uqLiRyWTw9/cHAHh7ewOAaJPMhIQEvP322w5Mj6RkaTkC45FfA+bL82s0AASVeNTleOTP8LONxQgxl2MiADOUsECFMdgnahcWEI1pYxY5KXOiy7q7u7Fr1y5RzMfHB4sXL+bGl0Qe7IaKm+joaFRWVgIANBoNYmJisHPnTixfvhwAUFhYyHvXHsJqvAhj2Yu2wmYgXd16PHjIiiXC25DLhEHbyWQyfG/ag1DI+XAeOVdRURE6OsT/hmfNmsXChsjD3dCE4jlz5uCLL76AIFz+IPvRj36EDz/8ELfffjuWLl2Kjz76CHfdddeQJErOI1h7YCr7HdDTNuB5mVGArEaFOXWPYprw9VULGwDITl2IyOBRQ5Eq0aCqqqpw8uRJUWzMmDGIjIyUKCMicpYb+lX6qaeewve//32YzWaoVCo8/vjjEAQBn332GRQKBZ555hn84he/GKpcyUm6T70Da/sJUUxz1gJNjRWqBisU7QLuSX0IAaFVCJA1XfW9gv3CMTfjzqFMl8hOd3c3iouLRTGdTodJkyZJlBEROdMNFTc6nU60iZxMJsMvfvELFjQexFy/G+Zzn4tiqhZv+O9stk0jPuAbh00h4/BDvCZqFx+RhpEhCTBbemC29MDHyxcTk2ZDreItAHIeq9WK48ePw2js3cBVJpNhzpw5UCp5a5RoOOD/dLKxdtbAdOL/iWIyuT90+fXoO414ddw9mCArgLes0xZTKdW4c/pP4ecdACIpffvtt3b74GVmZiI4OFiijIjI2a6ruHn33Xeh1+uRm5sLAGhvb8f9999v1y46Ohrr1693bIbkFILVDGPZ7wBL77o0kCnhV6yE3NQb2hI4FgcCRuIBfCx6/bS0xSxsSHLV1dU4fPiwKKbX6zFu3DiJMiIiKVxzQvFXX32Fp556yvYIOHB5f5Y9e/bgzJkzqK+vR319Perq6vCPf/wD+fn5Q5owDQ1L0z4InVWimJdsBryOn7cdWyHDs3H3IAvboZL12OI+Xn6YPmaxs1IlGlBbW5vdY99arRbz5s3j7t5Ew8w1/8dv2LABmZmZmD59ut259evXY//+/di/fz8OHDiAzMxMfPTRR0OSKA0tS7P4t12F/yT4frRbFPu7PhvnRnghBYdE8bkZ34NGpR3yHIkG09PTg+3bt6O7u9sWk8lkmDdvHlcgJhqGrlncHDp0CAsWLLiuN1u4cCEOHDhwy0mR81laxCu4eh+og7yrd05Nj0KFX8UuxwyIH/0O8gtDZtIsp+VJ1J8gCNizZ49oQVHg8qrp4eHhEmVFRFK65pybhoYGu3UhvLy8sHLlSkRFRYniYWFhuHjxomMzpCFnNTVB6DrbGxBk8DpYIWrzavLdSNb+y26H75zMu7g4H0nGbDZj9+7dOH36tCgeGhqKMWPGSJQVEUntmp9KGo0GBoNBFPP29sa6devs2hoMBqhUKsdlR05h7Tdqo2yyQt47pQbNCeOwN6QFibIyUbsYfRJSojOdkSKRna6uLmzbts1uQ8yAgAAkJSVxg1aiYeyat6ViY2Ov+1bTgQMHEBsbe6s5kZNZWkpEx+oLFtvfzd4+eGV0hF1hEzAiBHfN/E9+gJAkmpqa8Pnnn9sVNl5eXpg/fz7XsyEa5q5Z3CxcuBBffPEFysvLr9ruxIkT+OKLL7B4MZ+acTeWFvHu3eq6y1t7W2TA+/MyYeo6Jjqv0gThwUXPwN8nyGk5El1x9uxZbNq0CZ2dnaK4TqfDsmXLoNPppEmMiFzGNYubRx55BP7+/li2bBk2btwIs9ksOm82m7FhwwYsW7YMgYGBePjhh4csWXI8q6EOgrGuN2ARoG6wQgDwx/EpKO2oFrVvRwB+vOgZ6EZwQTRyvvr6emzfvt3u51BUVBSWLVsGPz8/iTIjIldyzbFbnU6Hjz/+GPfddx9WrlwJrVaL+Ph4jBgxAh0dHTh16hSMRiMiIiLw4YcfIiCAC7m5k/5PSakuCpBZgC9j9Djr3yU61yYEoCX8EUQHhDozRSIAlxcP3bp1KywWiyg+ZswYZGVlcS0bIrK5rhvT48aNQ1FREd577z1s2bIFJ06cwKVLl+Dr64tx48Zh8eLF+PGPf8zfmtyQ6dwe0dYK6loLygK98a9RWsjQ+8h3u6DDZ1iJ1xK5ozI5n8lkQn5+vmi/KADIzs5GWlqaRFkRkau67ll3fn5+eOyxx/DYY48NZT7kTM0XIbt4AOiz/l5LsxzvpIdDJusd9u8WNPgSP4JJpsOikV4SJErDmdVqxfbt29Ha2iqKZ2RksLAhogFxHHe46uqA4s0nAW3v6IyhG3grfhQg7y1sBEGGfNyNZuiRHWCBn5r/ZMh5rizQd+HCBVE8Pj4eEydOlCgrInJ1/KQapjT/eAOCrM++UQLwSUMIuvrukglgLxahCskAgPkh4rkOREPtyJEjdk9q6vV6zJw5k8sQENGgWNwMR8YuKAq3oTus9/Kf6NLgglX8z8Hqn4XDuLynWKyvArMCWdyQ89TU1GDfvn2imK+vLxYsWMB1bIjoqvgTYhhSlB2EzNyD7jANAEAQgOJ28eaCMaFJ+FHOSkw+Z0H1JTPuivdGx/lLUqRLw1Brayt27NghiqnVaixatAhaLTdpJaKrY3EzDDUUfIOIABkEr8vD+meMajT09G6bIYMM35v2E6iUaiyL7X1dBYiGnslkwtatWwfc4ZsL9BHR9eBtqWHmYocJfkeLRbek+o/apMZMRLB/mLNTI4LVasWOHTvQ1tYmimdlZdlt1EtENBgWN8OIIAj4v8+KoDN3whSjAACcN6lw1qQWtZuRnitFejTMCYKAffv24dy5c6J4YmIid/gmohvC21LDyF/KuxB+fC/as5Xo0V+ua4vbvUVt4sJTERk8Sor0aBgzm83Ys2cPKirENz9DQ0Mxffp0PhlFRDdEspGb3//+95gzZw5GjhyJ+Ph43H333Th2TLxBoyAIWLNmDZKTkxEWFobc3FwcP35c1MZkMmHVqlWIi4tDREQE7rnnHpw/fx4kVtHWg2f3teDOkUUwJlyuaRt7FKgwiBflm5l+mxTp0TDW1dWFr7/+2q6w8fb25g7fRHRTJCtu9uzZg5/85CfIz8/Hpk2boFQq8b3vfQ8tLS22Nq+//jrWr1+PtWvXYseOHQgJCcEdd9yBS5d6n9pZvXo1vvzyS7zzzjvYvHkzLl26hLvvvttu/5nhrMcq4Ge7W/Br9d+gHN27QN++fnNtIoJiEBee6uz0aBhramrCF198gYaGBlFcrVYjJycHPj4+g7ySiGhwkv1KtHHjRtHxn//8Z0RHR6O4uBiLFy+GIAh488038fjjj2PZsmUAgDfffBOJiYnYsGEDVqxYgba2Nrz//vtYv3495syZY3uf9PR07Nq1C/PmzXN6v1zR+ye7MMG0Gffrd9ti7WY5jnaKR21mpN/G4X9ymtOnT2P37t12O3z7+/sjJyeHT0YR0U1zmQnFHR0dsFqtth9o1dXVqK+vx9y5c21ttFotsrOzbQt7lZSUoKenR9QmKioKo0ePtlv8azj77sx3+E3gP0WxwhYfWPtsmRnoq0dqNJezp6FnsViwd+9e7Nixw66wiYyMxLJly1jYENEtcZmb2c888wzS09MxefJkAEB9fT0AICQkRNQuJCQEtbW1AICGhgYoFAoEBQXZtek/zN1X/3v7jjJU73srLppkGNezHdD0xva1aFFiEE8kTgrNxOnTp6/5fq7YR0djH4eOwWDA0aNH0dHRYXcuMjIS8fHxOHv2rEO+Fq+jZ2AfPYOj+5iYmHjV8y5R3Dz77LMoLi7Gli1boFAoROf63yYRBOGat06u1eZa35SbUVFRMSTve6t2Hr2EWV4nbMffdXhh5yU/URt/nyDkTL0DKqW6/8tFXLWPjsQ+Dg2TyYQzZ87g22+/RU9Pj+icXC5HdnY2UlJSHPb1eB09A/voGaToo+TFzerVq7Fx40Z8+eWXiI2NtcX1ej2Ay6MzfRfvamxstI3mhIaGwmKxoKmpCcHBwaI22dnZzumAiys6exb3a5oBACe7NNjSLC5sNCov3Dvnv65Z2BDdqI6ODlRVVaG6uhq1tbUQBMGuja+vL+bNm2c3QktEdCsknXPz9NNPY8OGDdi0aROSkpJE52JiYqDX67Fz505bzGg0oqioCFlZWQCAjIwMqFQqUZvz58+jvLzc1mY4azBYoO0oBQBUG1X4otEfQp95Ngq5EvfNfZzr2pBDCYKA/fv345///CeKiopw4cKFAQub2NhY3HHHHSxsiMjhJBu5eeqpp/DRRx/hgw8+gE6ns82x8fHxwYgRIyCTyfDwww/jtddeQ2JiIhISEvDqq6/Cx8cH3//+9wFcfqrigQcewAsvvICQkBAEBATgueeeQ1paGmbPni1V11zGV9VGTPU6gVazHJ9e1MHSp7CRyWT4wayHERfuuFsBRFarFQUFBTh58uSgbeRyObKyspCWlsan84hoSEhW3Lz99tsAYHvM+4qnn34aq1evBgA89thjMBgMWLVqFVpbW5GZmYmNGzfC19fX1v6ll16CQqHAihUrYDQaMXPmTLz11lt2c3eGo8+ruvD/NCdwoN0H3YJ4kG5Z9oNIjeHTUeQ4FosFu3btQmVl5YDndTodoqOjMXr0aD4NRURDSrLiprW19ZptZDIZVq9ebSt2BuLl5YV169Zh3bp1DszO/TUaLahtrIY+vA2fG8VPk80fNRuZiTOlSYw8ktlsxvbt21FTUyOKa7VajB07FtHR0SxoiMhpJJ9QTEPj62ojpmpOoNmsQKu59zIrrAKmTL1HwszI03R3d2Pr1q22JRqu8PHxwZIlS1jUEJHTsbjxUF9UGfADrxM4bdCI4nFyP2jUWomyIk/T0dGB/Px8NDc3i+J+fn7Izc3FiBEjJMqMiIYzl1mhmByn2WjB7loDsr1OoNIofsQ7aWSGNEmRx7myL1T/wiYwMBBLly5lYUNEkuHIjQf6+qwRCcoL8JV14qxR/Jht4vhFEmVFnuTs2bPYsWOH3YJ8oaGhWLhwIby8vAZ5JRHR0GNx44E2VV0etakyqkX7RwVZVQgKjLrKK4mu7dixYygsLLRbuyY2NhZz5syBUskfK0QkLf4U8jDt3VbsqjXhraDjOG0Q35IaHRQvUVbkCSwWCwoLC3HixAm7c2PGjEFWVhbkct7pJiLpsbjxMMX13bBYrchWn8D7LeI5Dwnp8yTKitxdV1cXtm/fblts8wqZTIapU6ciLS1NosyIiOyxuPEwBXUmpKnPwmg145KldyFDlQDERmdIlxi5rYsXL2Lbtm3o7OwUxZVKJebOnYuYmBiJMiMiGhiLGw+zp86EbK8TdrekErRh3ByTbojRaERpaSnKyspgsVhE53x9fZGTk4PAwECJsiMiGhyLGw/S1m3Fd009eDqoDKc7xOvbJCZOlSgrcjcGgwGlpaU4duwYzGaz3fmIiAjMmzePT0QRkcticeNBiupNkAtmjFWfQqEpQHQucfR0ibIid2G1WnH48GGUlpYOWNQAnDhMRO6BxY0H2VPbjfGaM7jQLYfQ5xFwvdwbuhHBEmZGrs5qtWL37t2oqKgY8LxWq8WUKVOQkJDg5MyIiG4cixsPUlBnQq6mBKeN4ltSSZHjJMqI3MHVChutVotx48YhJSWF69cQkdvgTysP0WqyorSpB38KPoyvW/rNt0nlDuA0sMEKGy8vL2RkZLCoISK3xJ9aHqKw3oRAeTu8lU3osgbZ4hpBjmh9koSZkauyWq0oKCiwK2x8fHxw2223wc/PT6LMiIhuDYsbD7GnzoRZ2qM402+jzASfCCjkvMwk1tHRgT179qCmpkYUZ2FDRJ6An3oeYk9tNx7yKsWZTvEtqfi4iRJlRK5IEAScP38ee/futdv00sfHB7m5uSxsiMjtsbjxAC0mK8qaTZgefgTvNvuKziUkTZMoK3I1zc3NKCgoQENDg925K4WNv7+/BJkRETkWixsPUPjvLRdazBbRLuDBVhUC/EIlzIxcQVNTE7777jtUVlba7eQNAKGhoZgzZw5HbIjIY7C48QAFdSbM8SrDmX6PgCf6c8+f4UoQBNTW1uK7777DuXPnBmyjUqkwadIkpKamQiaTDdiGiMgdsbjxAHvquvFb7REcbhZPJo5PnCJRRiQFQRDQ1NSEM2fO4MyZM2hraxu0bUxMDLKzszFixIhB2xARuSsWN26uxWRFTUsb4sPPYIe5dxViuSAgNpnzbYaDnp4elJaWoqKiApcuXbpqW51Oh8jISEydOpWjNUTksVjcuLk9dSZM0x7D2X6PgMdatdCovSXKipzFYDBgy5YtaGxsvGq70NBQZGRkIDo6GqdOnWJhQ0QejcWNm9tTa8JsrzL79W0C4yTKiJzl0qVL2Lx5M9rb2wc8L5fLMXLkSKSnpyMsLIwFDRENGyxu3JggCNh+3oAHR5Thk9Z+xQ13Afdozc3NyMvLQ1dXlyiuUCgQFRWFUaNGITo6GhqNZpB3ICLyXCxu3NjxVjMUhhrAuwMmIdAW9+mxQp/MycSeqq6uDvn5+eju7hbFIyMjMW/ePBY0RDTssbhxY19WGy4/Am4Qj9okYgTkcoVEWdFQqq2tRV5eHiwWiygeFxeH2bNnQ6HgdScikkudAN28TVUGzNWW2s+3CeFGmZ6ooaEB+fn5doVNamoq5s6dy8KGiOjfWNy4qcp2M+ramjFefRK13SrRubjUmRJlRUOlubkZW7ZssdsPKjMzE9nZ2ZwsTETUB29Luamvqg1Y7P0takwqCH22XAjrMsM3IUO6xMjh2trasHnzZphMJlF80qRJyMjIkCYpIiIXxpEbN7Wp2oCl3gfs5tskyXUA59t4jI6ODmzevBkGg0EUz8jIYGFDRDQIFjdu6HynBWeampClKccpg/jJmAR9skRZkaOdP38en3/+OTo6OkTxtLQ0TJw4UaKsiIhcH29LuaGv/31Lqq5biU5r7yiNxmxFdPocCTMjR7BarTh06BBKSkrsziUlJXHrBCKia2Bx44Y2VRvwX94HcKLLSxRPa+2BIi5FoqzIETo6OrBjxw7U19fbnRs1ahRmzJjBwoaI6BpY3LiZRqMF5RcbMSWiHH9uDhKdG+Mdyfk2buzixYvIy8uzmzgMAOPHj8eECRMgl/NOMhHRtbC4cTObzxqxSPst6nuUuGTpLWTUZisSErMlzIxuhclkwvbt2+0KG29vb8yZMwcRERESZUZE5H5Y3LiZL6sMWOl9ACe6xBOJU5sMkN82GYJEedHNEwQBe/bssZs4PHLkSMyaNQtarVaizIiI3BPHuN1IW7cVR+ovYoqmHOX95tuMNSgh6KMkyoxuRUVFBSorK0Wx1NRULFy4kIUNEdFNYHHjRvJrjFjg9S0u9ijQ3ueWlMpiRWLkWIATTd1OW1sb9u7dK4oFBgZiypQpnDhMRHSTWNy4kU1VBtw2wFNSKU0GKNMmSZQV3SyLxYIdO3bAbDbbYgqFgvtEERHdIhY3bqKjx4rDtQ2Xb0n1W7hv3MUuWFInSJQZ3ayDBw+isbFRFJs6dSoCAgIkyoiIyDNwQrGb2HbOiGnqUjSaFWg19142pUXAaE0IBF3QVV5NrkQQBBw8eBClpaWieGxsLJKTucI0EdGtYnHjJjZVGTFPewzl/Z6SSm42QJUyE90S5UU3xmKxYPfu3Th16pQo7uPjwwX6iIgchLel3ECX2Ypt57owXXPMbr7N2ItdsKRlSpQZ3QiTyYS8vDy7wkahUGDOnDnw8vIa5JVERHQjOHLjBv513oQYWQ0EqwEtZm9bXGEVkNZshGX0OAmzo+vR1taGrVu3orW1VRT38vJCTk4O9Hq9NIkREXkgFjdu4MsqA2Zqj6KsU/ybfVKLEZqYFBi0PhJlRtciCAJOnDiB4uJi0VNRAODn54fFixfDz89PouyIiDwTixsXZ7IIyKsx4s+6oyhtFhc3mfWdsEzhLSlX1dXVhd27d6OmpsbunF6vR05ODm9FERENARY3Lm7nBSN6zCaEogqd1t7f8DUWK9IvdsHM+TYuqaqqCgUFBTAajXbn4uLiMGvWLCiV/O9HRDQU+NPVxW2qMmKy5iROdqlF8XH1XVApNeiOT5EoMxpMZWUl/vWvf9nFlUolpk6ditGjR/OpKCKiIcTixoV1WwR8fdaAn2tLcbJd/Aj45LqOyxOJlSqJsqOBNDc345tvvrGL6/V6zJ49m/NriIicgMWNCyuoM6GtW0C811EcRu9v+gE9ZoxqM8GUs1zC7Ki/7u5ubN++XTRxWCaTYeLEiRg7dizkcq68QETkDCxuXNimKgNSzWfRYDIB6L0tNamuAz3f/yksY7OkS45EBEHArl270NbWJopPmzYNKSm8dUhE5Ez8VdJFCYKAwsom/F/LH3C+WzzfJiNuJnpuu0+izGggJSUlqK6uFsWSkpK4nQIRkQRY3LioE61mPHPkr7gQ1COKR8tl8L33CYATUl3GuXPncPDgQVEsODgY06ZN48RhIiIJsLhxUUVVLbiroQjfqbxF8XFj5gKcu+Ey6uvrsX37dlFMo9Fg/vz5fNSbiEgi/JR0UYZv96EmUok2q8IWU8oEpKfeLmFW1FdDQwPy8vLQ0yMeXZs7dy58fX0lyoqIiFjcuCCzVcCoimLsixwhio/W+UDrpZMmKRJpaGjA5s2b7QqbyZMnIyoqSqKsiIgIYHHjko40GDC38TCOeWtF8XExGdIkRCIXL14ccMRmwoQJGDeOm5gSEUmNkwJcUNWBbxGks8LUp/b0lluRmJAjYVYEAI2Njdi8eTO6u7tF8fHjx2PChAkSZUVERH1x5MYFjSjdi++ixTt9J46QQ+ETK01CBABoaWlBXl6eXWGTkZGBzMxMPhlFROQiWNy4GJPZiszq/TjmJ74llRI1hh+eEmpvb0deXp7dRpjjxo3DxIkTeW2IiFwIixsXc6LkOCw+l9CB3qek1DIr4pNukzCr4a2zsxObN29GZ2enKJ6eno5JkyaxsCEicjEsblyMYX8BSmLFt6TifeTQ+HOlWykYjUZs3rwZly5dEsWTk5ORlZXFwoaIyAWxuHEx0eVFKPMXL9yXHJnCD1EJ9PT0IC8vD62traJ4fHw8Vx8mInJhLG5cSFftBYywXkBzn4fY5BAwejRvSTmbIAj45ptv0NjYKIrHxMRg9uzZ3OGbiMiF8VFwF1JbsBsX4sSjNjHeMngHpEmU0fB16NAhnDlzRhSLiIjA3LlzWdgQEbk4/pR2IdqSvSgLEBc3KRFJvP3hZKdPn8bhw4dFscDAQCxYsID7RRERuQH+pHYVHe3QtZ3ABUT0CQpISVoiWUrDUXt7O7777jtRzMvLCzk5OVCr1RJlRUREN4LFjYswHixEWbx4bZsILxn8QzKkSWiYEAQBXV1daG5uRlNTE8rKymCxWGzn5XI5FixYwI0wiYjcCIsbF9H63WGUBfa7JRUWx1tSQ8RsNuPgwYOoqKiwW5ivrxkzZiAsLMyJmRER0a1iceMizHXlqPIX3/ZITVosUTaezWQyIT8/H/X19Vdtl56ejqSkJCdlRUREjsLixgUIPT24GNQCATpbLFAlICR8knRJeaiuri7k5eWhubn5qu1GjRqFyZMnOykrIiJyJBY3LqDy5BmcCVCJYsn6GN6ScrC2tjbk5eXZrTasVCoRGBiIwMBAmM1mpKamIjQ0lN9/IiI3xeLGBVSWleO8Sg1Ye2PxIydKl5AHampqQl5eHgwGgyiu1+uxcOFCaDQaAEBFRQX0er0UKRIRkYOwuHEBxjNluBgmvhRRUdkSZeN5uru7kZ+fb1fYREdHY968eVy7hojIw/CnusRaTVb4yk5CQO8tEJ1SgLdPiIRZeZbDhw/b7eidmJiImTNncrVhIiIPxJ/sEtt53gD5iDZRLMJnhETZeJ7W1laUlZWJYsnJyZg1axYLGyIiD8Wf7hI7cLIWDd4KUSwiJFaaZDyMIAgoKiqC1do7mcnHxwdTpkzhZGEiIg8maXGzd+9e3HPPPUhJSYFOp8OHH34oOi8IAtasWYPk5GSEhYUhNzcXx48fF7UxmUxYtWoV4uLiEBERgXvuuQfnz593ZjdumlUQ0FxejvNK8ZNSkWHjJMrIs5w9exbnzp0TxbKysqBSqQZ5BREReQJJi5vOzk6kpqbi5ZdfhlartTv/+uuvY/369Vi7di127NiBkJAQ3HHHHaJHeVevXo0vv/wS77zzDjZv3oxLly7h7rvvFi2h76q+a+pBSucxNFrEU58iI6dIlJHnMJvNKCoqEsXCw8MRFxcnUUZEROQskhY3OTk5eOGFF7Bs2TK7+Q+CIODNN9/E448/jmXLliE1NRVvvvkmOjo6sGHDBgCX1y15//338Zvf/AZz5sxBRkYG/vznP+Po0aPYtWuXBD26Mfk1RiR6nQT6TCYOUgrQav2lS8pDHDlyRFQEy2QyTJ06lbejiIiGAZedc1NdXY36+nrMnTvXFtNqtcjOzsa+ffsAACUlJejp6RG1iYqKwujRo21tXNm2c0aovcWTicN9uEHjzRAEAe3t7Thz5gwOHTqEkpIS0fmUlBQEBQVJkxwRETmVyz4KfmXfn5AQ8SPRISEhqK2tBQA0NDRAoVDYfWiFhISgoaFh0PeuqKhwcLY3/r7N3cDJOqDJSxDF/bT6IcvPEVwtN0EQUFNTg7Nnz8JsNg/YRqVSISAg4Lpzd7U+DgX20TOwj56BfbxxiYmJVz3vssXNFf1vIwiCcM1bC9dqc61vys2oqKi4off956kupHcW4UKQGugzPSg5eRZGxTo+P0e40T4ONUEQUFxcjMrKyqu2mzJlCpKTk6/rPV2tj0OBffQM7KNnYB+HhsvelrqyBH7/EZjGxkbbaE5oaCgsFguampoGbeOqtp0zYob5CJr7TCaWQUB4BLdduF6HDx+2W8Omv5iYGO7sTUQ0zLhscRMTEwO9Xo+dO3faYkajEUVFRcjKygIAZGRkQKVSidqcP38e5eXltjauyGIV8K/zRiRrykXxYCXgpfaRKCv3UlZWhkOHDoliSqUS4eHhSEtLw4wZM3DHHXdgwYIFXKyPiGiYkfS2VEdHh+2WgtVqxblz51BaWoqAgACMHDkSDz/8MF577TUkJiYiISEBr776Knx8fPD9738fAODv748HHngAL7zwAkJCQhAQEIDnnnsOaWlpmD17toQ9u7pT7Wa0dgvQeLUA0Nji4d6cTHw9ysvL7R7zVqvVuO222zhpmIiIpC1uDh8+jKVLl9qO16xZgzVr1uDee+/Fm2++icceewwGgwGrVq1Ca2srMjMzsXHjRvj69hYBL730EhQKBVasWAGj0YiZM2firbfegkKhGOhLuoSTbWYorBa0aayincAjQj37vqsjVFVVoaCgQBRTKpVYtGgRCxsiIgIgcXEzY8YMtLa2DnpeJpNh9erVWL169aBtvLy8sG7dOqxbt24IMhwaFW1mpBjO44JavFJuVMxUiTJyD01NTdi5cycEofcJM7lcjgULFtjmaBEREXEyggROtvZgnuVbtPabTBzGbRcGZTQasXXrVtHj3jKZDHPnzkVUVJSEmRERkathcSOBU+1mpKnFe2SFKGXQqLwkysi1Wa1WbN++HR0dHaL4tGnTMGrUKImyIiIiV8XixskEQcDJNjO8NM2ieISWk4kHU1RUZFu48YqUlBSkpKRIlBEREbkyFjdO1mCwot1kRbtKvLFnZAgnEw/kxIkTOHbsmCgWHh6O7OxsiTIiIiJXx+LGyU62mRHb3YBa9JtMHD9Dooxc18WLF7F3715RbMSIEZg3bx7XriEiokHxE8LJKtrMWGw+gHZL76PqcggIC0uXMCvXIwgC9u7dC6u191l5pVKJnJwcaLVaCTMjIiJXx+LGyU629SBNeVQUC1ECSoVqkFcMT+Xl5bh48aIoNmvWLK5lQ0RE18Tixskq2szwUYsnE4drRkiUjWsymUw4cOCAKDZq1CjExcVJlBEREbkTFjdOVtFmRpeyRxSLCOLjzH0dOnQIRqPRdqxQKDBlyhQJMyIiInfC4saJusxWdDW3oF4mXhg6fBRXJr6iqanJ7umojIwMjBjB0S0iIro+LG6c6HS7BfNNh9Fi6VvcCAiLnCBZTq5EEAQUFhaKtlfw9fXF2LFjJcyKiIjcDYsbJ6po7cEk5XeiWJBCgJfaW6KMXEtFRQXq6upEsalTp0KplHQLNCIicjP81HCik21mJKnqUNUnFq7hY82NjY04ePAgampqRPGRI0ciOjpaoqyIiMhdsbhxooo2M2JVJgAaWyzMP0K6hCTW3NyMQ4cOoaqqyu6cXC7H1KlTIZPJnJ8YERG5NRY3TnS+sR0XFXKgz84L4dGZ0iUkge7ubpw5cwbl5eWor68ftF1WVhb8/f2dmBkREXkKFjdOYhUEBNWWodFX/C2PjBkeeyQ1NzfjyJEjqKyshNlsHrSdXq/HxIkTERExfEe0iIjo1rC4cZLznRZMwUE0oPc2i5/cCh9vz15xVxAEHD16FPv27RNtpdBfcHAwJk6ciKioKN6KIiKiW8Lixkkq2swIV9WgoU8sTO3ZWy6YTCZ88803qK6uHrRNWFgY0tPTERMTw6KGiIgcgsWNk5xsMyNQ3Ym+k4kjfEOkS2iI1dfXY8eOHejo6LA75+Pjg8TERCQlJXFeDRERORyLGyc509QFKGVAnzszYZGeuRP46dOnsXPnTtFifACgUqkwffp0xMXFQS7nEktERDQ0WNw4iVBVjgZB/O2O8MDJxB0dHSgoKLArbIKDgzFv3jz4+flJlBkREQ0XLG6cJLKtGI0jekcrvOVW6HSetWGmIAjYs2cPenrEG4Omp6dj0qRJUCgUEmVGRETDCYsbJ2jrtkIvO43GPrEwldzjJtCePn3abpXhKVOmID3dM2+/ERGRa+LEByc41WaGSt0mioV566RJZogYDAYUFhaKYnq9HmPGjJEoIyIiGq44cuMEp5qN6FCJ56BEhI2WKJuhUVRUBJPJZDtWKBSYOXOmx41OERGR62Nx4wRtZypxUaYC+tQ3EdFZ0iXkYNXV1Th9+rQoNmHCBOh0OmkSIiKiYY23pZxAc34fjELvt1otsyIwdKyEGTmOyWTC3r17RbGgoCCMHesZ/SMiIvfD4sYJtMajouNQJaBQuP/qxD09PcjPz0dnZ6ctJpPJMHPmTK5jQ0REkuEn0BC7aLBAqWwSxcK1IyTKxnHMZjO2bt1qt7P3uHHjEBwcLFFWRERELG6G3KE6AzrU4l2ww4NjpUnGQaxWK3bs2IELFy6I4nq9HuPHj5coKyIioss4oXiInTtRgQa5ut9k4onSJXSLrFYrdu3aZbcZZnBwMBYtWgSlkv+kiIhIWhy5GWLW6j0w9JlMrJFZERYxRcKMbp7VasXu3bvtnozS6XRYvHgx1Gq1RJkRERH14q/ZQ8hiFeBjEU8mHqkGFGofiTK6eSaTCaWlpWhtbRXFfX19sWTJEnh5eUmTGBERUT8sboZQeZsZPZp2AL1PRsX6BkmX0E1qb29Hfn6+XWHj4+OD3Nxc+Pi4X7FGRESei8XNECqpaUGtUnznb1SMe823qa+vx9atW2E0GkVxX19fLFq0CL6+vhJlRkRENDAWN0Po4rG96ELvTthqmRXhcXMkzOjGnD59Gt988w0sFosortfrsWDBAmi1WokyIyIiGhyLmyHk1VoM9JmKEqU2Q+kdIV1C10kQBJSUlODgwYN25+Lj4zFz5kw+FUVERC6Ln1BDpL3bCpmiThSL1vi4/EaSVqsVe/bsQXl5ud25mJgYzJkzx+X7QEREwxuLmyHy7UUTmpUWoM9tqTh9knQJXQeTyYTt27fbLc4nl8sxY8YMyGQyFjZEROTyuM7NEDlafhSdst7CRiUTEJmUI2FGV2cwGPDVV1/ZFTZqtRqLFy9GUpJrF2ZERERXcORmiHSf3SU6jlB3QxWYKk0y12AymZCXl4fm5mZR/MoTUTqdTprEiIiIbgKLmyEgCAKUPRV9l7dBtFIBmcL1VvDt7u5GXl4emprEm3uGhoYiJyeHT0QREZHbYXEzBKoumdGp7ELfu35xfq73lFRPTw+2bNmCixcviuKRkZHIycnhE1FEROSW+Ok1BAqrz6FT1lvYKGUCRsZmS5iRve7ubmzbtg319fWieFhYGBYsWMDChoiI3BY/wYbA+ZMFouMIdQ9UMTMkykaso6MDR48exfHjx9HT0yM6FxoaioULF0KlUg3yaiIiItfH4mYIKNoOi55Di1JYIPcKli4hAC0tLfjuu+9w6tQpCIJgdz44OBiLFi3izt5EROT2WNw4mMEsoBvip45iNH4SZXPZyZMnUVBQAKvVOuD5wMBALF68GBqNxsmZEREROR6LGwfbd64OXX1GbeQQMEo/RpJcBEHAt99+i2+//XbA815eXkhLS0N6ejpvRRERkcdgceNgJ6qPio4jND1QS7BZpsViwe7du3Hq1Cm7c35+fhg7diwSExM5cZiIiDwOP9kczFpXIjrWK82QBzt35MZoNGLbtm2oqxPvbaVQKDB9+nQkJCRALufi1ERE5JlY3DiYzHhWdOxvVUImVwzS2vE6OjqQl5eH1tZWUdzLyws5OTnQ6/VOy4WIiEgKLG4crMd6Ceizt6Sfwt9pX7u1tRV5eXno6OgQxf39/bFo0SL4+Uk7sZmIiMgZWNw4WJfMgr7VTZBflFO+bmNjI/Ly8mA0GkXxK4vyeXl5OSUPIiIiqbG4caDGLhM6RREBYeFDv1lmbW0t8vPz7Rbli42Nxdy5c6FQOO+2GBERkdRY3DjQsboL6Dtq46uwQh05bsi+ntVqxdGjR3HgwAFYLBbRuaSkJMyYMYMTh4mIaNhhceNAddWlomOdwgJZQOyQfK3a2lrs3bsXLS0tdufS09ORlZUFmUw2wCuJiIg8G4sbBzI0Hhcd6yBAJnPsLaGuri4UFxfj9OnTA56fNGkSxo0bx8KGiIiGLRY3DmQ21oqO/eC4wkYQBJSXl2Pfvn3o7u62O69SqZCdnY2kpCSHfU0iIiJ3xOLGgbot7aJjX6WvQ963ra0NBQUFqK2tHfB8YmIiJk+eDG9vb4d8PSIiInfG4saBDDCj73bgAb4Rt/R+VqsVZWVlOHjwoN2EYQAICgpCdnY2wsLCbunrEBEReRIWNw7SbRHQDvE8l4jwtJt+vwsXLqCwsHDACcNKpRITJ05EWloan4YiIiLqh8WNg1S1tcHSp7jxklvhF515w+9z6dIl7Nu3D2fOnBnwfGRkJKZPn87VhomIiAbB4sZBWhsrRMeXHwOPvu7XC4KAkpISHD58eMBbUGq1GlOnTkViYiKfhCIiIroKFjcOYmkXP5qtgxUy2fXdMhIEAXv27MGJEycGPM8Jw0RERNePxY2DWLsbRMe+1/mtFQQBe/fuHbCwCQ4ORnZ2NnfyJiIiugEsbhykRxDvxD1COeKarxEEAUVFRTh+XLz4n0ajweTJkzF69GjegiIiIrpBLG4cxCDrAYTeRfv8vcOv2l4QBBQXF+Po0aOiuFarxW233QadTjcUaRIREXk8PkfsAIIg4JIgHmEJDUu+6msOHjyIsrIyUUyr1SI3N5eFDRER0S1gceMA51paYOzzrVRAQHhc1qDtT58+jZKSElHMy8sLS5YsQUBAwFClSURENCywuHGAqlOHRMc6hQWKwJgB27a0tKCgoEAU02g0WLJkCQIDA4csRyIiouGCxY0DtNYeER37y6wDTgTu6enB9u3b0dPTY4vJ5XIsWrQIQUFBQ54nERHRcMDixgEMnTWiY98BdgMXBAEFBQVobW0VxadOnYrQ0NChTI+IiGhYYXHjAEZzm+jYR2H/GPixY8dw+rR4ob+EhASkpKQMaW5ERETDDYsbB+gSukXHvlrxontnzpxBcXGxKBYQEIAZM2ZwHRsiIiIH4zo3DtDe7zg4dDQAwGAwoLCwEJWVlaLzKpUK8+fPh1LJbz8REZGj8dP1FrV3deKS0HcATMDI+Ck4deoUCgsLYTKZ7F4za9YsrmVDREQ0RFjc3KLKk/sB9N5a8pUBRUcqcebMGbu2MpkMWVlZGDVqlBMzJCIiGl5Y3NyixvPf2f4uExTQmVMGLGyCg4Mxc+ZMPvJNREQ0xDxmQvHbb7+NsWPHQq/XY9asWSgsLHTK1+1oPwsAUAhq6M1jIRN0ovMKhQKTJk3CsmXLWNgQERE5gUcUNxs3bsQzzzyDJ598Ert378bkyZNx1113oaam5tovvkUGcyuUghZ681io4SM6FxgYiDvvvBMZGRmQyz3iW01EROTyPOITd/369bjvvvvwox/9CKNHj8a6deug1+vx7rvvDvnX7rRaoDenQwkvUTwsLIy7exMREUnA7Yub7u5ulJSUYO7cuaL43LlzsW/fviH/+laZGe1y8QhRTEwMFi9eDI1GM+Rfn4iIiMRkra2tgtRJ3Ira2lqkpKTg66+/xrRp02zxtWvX4pNPPsHBgwftXlNRUeHQHEzGTpw4dhgt7WaEBAUgJS2dt6GIiIiGSGJi4lXPe8zTUv1X+hUEYdDVf6/1TbkZao03VCoVYmNjPXbV4YqKiiH53rkS9tEzsI+egX30DFL00e2Lm6CgICgUCjQ0NIjijY2NCAkJcVoeMpmM69cQERG5ALe/d6JWq5GRkYGdO3eK4jt37kRWVpZEWREREZFU3H7kBgAeeeQRPPTQQ8jMzERWVhbeffdd1NXVYcWKFVKnRkRERE7mEcXNnXfeiebmZqxbtw719fVISUnBxx9/jOjoaKlTIyIiIifziOIGAFauXImVK1dKnQYRERFJzO3n3BARERH1xeKGiIiIPAqLGyIiIvIoLG6IiIjIo7C4ISIiIo/C4oaIiIg8CosbIiIi8igsboiIiMijsLghIiIij8LihoiIiDwKixsiIiLyKCxuiIiIyKOwuCEiIiKPImttbRWkToKIiIjIUThyQ0RERB6FxQ0RERF5FBY3RERE5FFY3BAREZFHYXFDREREHoXFDREREXkUFjcO8Pbbb2Ps2LHQ6/WYNWsWCgsLpU7puqxZswY6nU70JykpyXZeEASsWbMGycnJCAsLQ25uLo4fPy56D5PJhFWrViEuLg4RERG45557cP78eWd3xWbv3r245557kJKSAp1Ohw8//FB03lF9am1txc9+9jNER0cjOjoaP/vZz9Da2jrU3QNw7T4+/PDDdtd1/vz5ojau3Mff//73mDNnDkaOHIn4+HjcfffdOHbsmKiNu1/H6+mju1/H//u//0N2djZGjhyJkSNHYsGCBcjPz7edd/drCFy7j+5+DQfy2muvQafTYdWqVbaYK15LFje3aOPGjXjmmWfw5JNPYvfu3Zg8eTLuuusu1NTUSJ3adUlMTER5ebntT9/C7PXXX8f69euxdu1a7NixAyEhIbjjjjtw6dIlW5vVq1fjyy+/xDvvvIPNmzfj0qVLuPvuu2GxWKToDjo7O5GamoqXX34ZWq3W7ryj+rRy5UqUlpbik08+wYYNG1BaWoqHHnrIJfoIALNnzxZd108++UR03pX7uGfPHvzkJz9Bfn4+Nm3aBKVSie9973toaWmxtXH363g9fQTc+zpGRETg17/+Nb755hvs3LkTM2fOxP3334+ysjIA7n8Nr6ePgHtfw/4OHDiAv/71r0hLSxPFXfFachG/WzRv3jykpaXhj3/8oy02YcIELFu2DL/85S8lzOza1qxZg02bNqGoqMjunCAISE5Oxk9/+lM89dRTAACDwYDExET89re/xYoVK9DW1oaEhASsX78eP/jBDwAA586dQ3p6OjZs2IB58+Y5tT/9RUZG4pVXXsH9998PwHF9Ki8vR1ZWFrZs2YIpU6YAAIqKirB48WIcOHAAiYmJkvURuPzbYnNzMz766KMBX+Nufezo6EB0dDQ+/PBDLF682COvY/8+Ap53HQEgNjYWv/zlL/HjH//Y465h/z6uWLHCo65hW1sbZs2ahddffx2vvPIKUlNTsW7dOpf9/8iRm1vQ3d2NkpISzJ07VxSfO3cu9u3bJ1FWN6aqqgopKSkYO3YsHnzwQVRVVQEAqqurUV9fL+qbVqtFdna2rW8lJSXo6ekRtYmKisLo0aNdsv+O6tP+/fsxYsQIZGVl2dpMmTIFPj4+LtPvoqIiJCQkIDMzE48++iguXrxoO+dufezo6IDVaoVOpwPgmdexfx+v8JTraLFY8Omnn6KzsxOTJ0/2yGvYv49XeMo1fPzxx7Fs2TLMmjVLFHfVa6m84VeQTVNTEywWC0JCQkTxkJAQNDQ0SJTV9Zs4cSLeeOMNJCYmorGxEevWrUNOTg6Ki4tRX18PAAP2rba2FgDQ0NAAhUKBoKAguzau2H9H9amhoQFBQUGQyWS28zKZDMHBwS7R7/nz52Pp0qWIiYnB2bNn8eKLL+L222/Hrl27oNFo3K6PzzzzDNLT020fGJ54Hfv3EfCM63j06FHk5OTAaDTCx8cHH3zwAdLS0mwfVp5wDQfrI+AZ1xAA/vrXv6KyshJ//vOf7c656v9HFjcO0PdiAJdvf/SPuaIFCxaIjidOnIiMjAz8/e9/x6RJkwDcXN9cvf+O6NNA7V2l38uXL7f9PS0tDRkZGUhPT0d+fj5uv/32QV/nin189tlnUVxcjC1btkChUIjOecp1HKyPnnAdExMTUVBQgLa2NmzatAkPP/wwvvrqq0Fzc8drOFgfU1NTPeIaVlRU4De/+Q3y8vKgVqsHbedq15K3pW5BUFAQFAqFXVXZ2NhoV8W6gxEjRiA5ORmVlZXQ6/UAcNW+hYaGwmKxoKmpadA2rsRRfQoNDUVjYyMEoXe6miAIaGpqcsl+h4eHIyIiApWVlQDcp4+rV6/Gp59+ik2bNiE2NtYW96TrOFgfB+KO11GtViMuLg7jx4/HL3/5S6Snp+ONN97wqGs4WB8H4o7XcP/+/WhqasLUqVMRFBSEoKAg7N27F2+//TaCgoIQGBgIwPWuJYubW6BWq5GRkYGdO3eK4jt37hTdN3QXRqMRFRUV0Ov1iImJgV6vF/XNaDSiqKjI1reMjAyoVCpRm/Pnz9smhrkaR/Vp8uTJ6OjowP79+21t9u/fj87OTpfsd1NTE2pra20fKO7Qx6effhobNmzApk2bRMsTAJ5zHa/Wx4G443Xsz2q1oru722Ou4UCu9HEg7ngNc3NzUVhYiIKCAtuf8ePHY/ny5SgoKEBCQoJLXkvelrpFjzzyCB566CFkZmYiKysL7777Lurq6rBixQqpU7um559/HosWLUJUVJRtzk1XVxfuvfdeyGQyPPzww3jttdeQmJiIhIQEvPrqq/Dx8cH3v/99AIC/vz8eeOABvPDCCwgJCUFAQACee+45pKWlYfbs2ZL0qaOjw/ZbkdVqxblz51BaWoqAgACMHDnSIX0aPXo05s+fjyeeeAKvv/46BEHAE088gYULFzrlyYWr9TEgIAAvv/wybr/9duj1epw9exa/+c1vEBISgttuu80t+vjUU0/ho48+wgcffACdTme7p+/j44MRI0Y47N+mK/exo6PD7a/jr371K+Tk5CAyMhIdHR3YsGED9uzZg48//tgjruG1+ugJ1xCAbX2evry9vREQEIDU1FQAcMlryUfBHeDtt9/G66+/jvr6eqSkpOCll17CtGnTpE7rmh588EEUFhaiqakJwcHBmDhxIp577jkkJycDuDwk+PLLL+Mvf/kLWltbkZmZiVdffdX2Dxq4XKH/z//8DzZs2ACj0YiZM2fitddeQ1RUlCR9KigowNKlS+3i9957L958802H9amlpQVPP/008vLyAACLFy/GK6+8YvdDwNl9/P3vf4/7778fpaWlaGtrg16vx4wZM/Dcc8+J8nflPg72/k8//TRWr14NwHH/Nl21jwaDwe2v48MPP4yCggI0NDTAz88PaWlpePTRR21LRLj7NbxWHz3hGg4mNzfX9ig44JrXksUNEREReRTOuSEiIiKPwuKGiIiIPAqLGyIiIvIoLG6IiIjIo7C4ISIiIo/C4oaIiIg8CosbIvIIH374IXQ6Haqrqx36vrm5ucjNzXXoexLR0GJxQ0TX5UrxcOVPUFAQUlNT8fOf/xx1dXVSp3dLioqKsGbNGrS2tkqdChE5ALdfIKIb8swzz2DUqFEwmUwoLi7G3//+d+zduxeFhYXQarVSp3dTiouLsXbtWtx33312q6F+9tln0iRFRDeNxQ0R3ZB58+Zh0qRJAIAf/vCHCAgIwPr167F582YsX75c4uwcT61WS50CEd0g3pYiolsyc+ZMAEBVVRWsViv+8Ic/IDMzE6GhoUhJScGqVavQ1tYmek1ubi4mTZqEI0eOYPHixQgPD0daWhr+8Ic/iNoVFBRAp9OhoKDA7uvqdDqsWbPmqrkVFhbixz/+McaMGYPQ0FAkJyfj8ccfF91+WrNmDX79618DAMaNG2e77Xblaw4058ZgMOBXv/oV0tPTERoairFjx+LFF1+EyWQStUtPT8fy5ctx6NAhLFq0CGFhYUhLS8Mbb7xx1byJ6NZw5IaIbsmZM2cAAIGBgXjyySfx3nvvYfHixfjP//xPHD9+HO+88w4OHTqE/Px8qFQq2+va29uxfPly3HbbbbjjjjuwefNm/OpXv4LFYsGTTz7pkNw+++wztLS04Ic//CH0ej3Kysrwt7/9DcePH0d+fj4AYOnSpaioqMDGjRvx0ksvISgoCMDlXYoHIggCHnjgAWzfvh333HMPJk6ciOLiYrz66qs4fvw4PvzwQ1H76upq3HPPPbjvvvtw1113YePGjXj22WeRnJyMuXPnOqSfRCTG4oaIbkh7ezuamppgNBqxb98+vPLKK9BqtUhMTMQTTzyBH/zgB/jf//1fW/vExESsXr0a//jHP/DDH/7QFq+vr8cLL7yAX/ziFwCAlStX4vbbb8err76KlStXwt/f/5Zz/fWvfw1vb29RbOLEiXjooYdQXFyMKVOmYMyYMUhPT8fGjRuRm5uLmJiYq75nfn4+tm/fjqeeegrPP/+8LfeQkBC8+eab2LVrF2bPnm1rf+rUKXz++ee22H/8x39gzJgx+Otf/8rihmiI8LYUEd2Q5cuXIz4+HmlpaXjwwQeh1+vx0Ucf4cCBAwCARx99VNT+wQcfhJ+fn22k5Aq5XI6VK1eKjn/605/CYDAMeBvqZlwpbARBsBVlWVlZAICSkpKbes/8/HzIZDL8/Oc/F8Ufe+wx2/m+4uPjRcWORqPBxIkTUVVVdVNfn4iujSM3RHRD1q5di9GjR0Oj0SAqKgpRUVGQyWT47LPPIJPJkJiYKGqv0WgQExODs2fPiuKhoaHw8/MTxeLj4wEANTU1Dsn13LlzeOGFF7Bt2zZcunRJdK7/PKDrdfbsWej1erunqsLCwuDv72/Xz5EjR9q9h06nw9GjR2/q6xPRtbG4IaIbMmHCBNvTUtdLEATIZDJRrP/xlXbXagMAFovlml/TarXizjvvRGNjI5544gkkJSXBx8cHVqsVy5cvh9VqvYEeXJ/++QOAQqG47rZE5BgsbojIIaKjoyEIAioqKjBmzBhbvLu7G2fPnsWMGTNE7evr69He3i4avamsrATQO9pxZXSk/yhL/9GRgZSVleHkyZN44403cN9999nip0+ftms7WBE1kOjoaOzYsQOtra2i0Zsr/YmOjr7u9yKiocE5N0TkEDk5OQCA9evXi+Lvvfce2tvbsXDhQlHcarXi7bfftjv28vLC9OnTAVwuJBQKhd0cnL4TlgdzZcSk/wjJn/70J7u2V+bmXM8KxQsXLoQgCHaPc//xj3+0nSciaXHkhogcIi0tDStWrLAVM3PmzMHx48fx3nvvYcKECbj33ntF7fV6Pd566y2cO3cOKSkp+Prrr1FQUIBnn33WNiLi5+eH5cuX4+2337bN5ykoKLiuybhJSUmIj4/H888/jwsXLiAgIADbtm3DhQsX7NqOHz8eAPDb3/4Wy5cvh1qtxsyZMxESEmLXduHChZg/fz5eeeUVnDt3DhMmTMD+/fvx8ccfY8mSJaLJw0QkDRY3ROQwr732GmJiYvC3v/0NW7duRVBQEH7yk5/g+eefF61xA1wuXN59913893//N/7+978jMDAQL7zwAp544glRu7Vr18JsNuODDz6AXC5HTk4ONmzYgISEhKvmolKp8M9//hPPPPMM/vSnP0Eul2P+/Pn49NNPkZSUJGo7adIkPP/88/jLX/6CRx55BFarFV9++eWAxY1MJsP777+Pl19+GZ9++ik++eQThIWF4amnnsKqVatu8jtHRI4ka21t5aw2InKq3NxcNDQ02B4fJyJyJM65ISIiIo/C4oaIiIg8CosbIiIi8iicc0NEREQehSM3RERE5FFY3BAREZFHYXFDREREHoXFDREREXkUFjdERETkUf5/bGAViqmQQ9IAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_preds = pd.DataFrame([y_preds.ravel(), \n",
    "                         y_preds_t10.ravel(),\n",
    "                         y_preds_t15.ravel(),\n",
    "                         y_preds_t20.ravel(),\n",
    "                         treatments,\n",
    "                         df_test[y_name].ravel()],\n",
    "                        index=['All', 'Top 10', 'Top 15', 'Top 20', 'is_treated', y_name]).T\n",
    "\n",
    "plot_gain(df_preds, outcome_col=y_name, treatment_col='is_treated')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-11-29T23:31:15.523386Z",
     "start_time": "2021-11-29T23:31:15.346076Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "All       0.824483\n",
       "Top 10    0.832872\n",
       "Top 15    0.817835\n",
       "Top 20    0.816149\n",
       "Random    0.506801\n",
       "dtype: float64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# print out AUUC score\n",
    "auuc_score(df_preds, outcome_col=y_name, treatment_col='is_treated')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this notebook, we demonstrated how our Filter method functions are able to select important features and enhance the AUUC performance (while the results might vary among different datasets, models and hyper-parameters)."
   ]
  },
  {
   "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": "3.8.8"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}