notebooks/adult.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import sys\n",
    "import re\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn import metrics\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.preprocessing import MaxAbsScaler\n",
    "\n",
    "sys.path.append('../src')\n",
    "\n",
    "import columnar as col\n",
    "\n",
    "ROOT_PATH = '../'\n",
    "\n",
    "plt.style.use('fivethirtyeight')\n",
    "\n",
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "# Testing Mean Target Encoding vs categorical data embeddings\n",
    "## Prediction Task\n",
    "Predict whether an adult's income is higher or lower than $50k, using census information\n",
    "\n",
    "### Dataset\n",
    "We are leveraging the Adult dataset, which provides various census features about individuals and aims to predict whether an individual is earning over $50k or not.\n",
    "[https://archive.ics.uci.edu/ml/datasets/Adult](https://archive.ics.uci.edu/ml/datasets/Adult)\n",
    "\n",
    "### Definitions\n",
    "* **Mean Target Encoding** is a technique to transform categorical data into numerical by replacing the categorical value by the mean target value for all observations belonging to that category.\n",
    "* **Categorical data embeddings** is a potentially more expressive generalization of MTE which represents each categorical value as an embedding. embeddings sizes can be defined based on the cardinality of each feature. An embedding of size 1 should replicate closely the principle of MTE (even though values are learnt more indireclty), but weights are learnt instead of explicitly defined.\n",
    "\n",
    "## Load Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>workclass</th>\n",
       "      <th>fnlwgt</th>\n",
       "      <th>education</th>\n",
       "      <th>education_num</th>\n",
       "      <th>marital-status</th>\n",
       "      <th>occupation</th>\n",
       "      <th>relationship</th>\n",
       "      <th>race</th>\n",
       "      <th>sex</th>\n",
       "      <th>capital_gain</th>\n",
       "      <th>capital_loss</th>\n",
       "      <th>hours-per-week</th>\n",
       "      <th>native-country</th>\n",
       "      <th>income</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>19182</th>\n",
       "      <td>53</td>\n",
       "      <td>Private</td>\n",
       "      <td>290640</td>\n",
       "      <td>HS-grad</td>\n",
       "      <td>9</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Other-service</td>\n",
       "      <td>Husband</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>Germany</td>\n",
       "      <td>&gt;50K</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4616</th>\n",
       "      <td>36</td>\n",
       "      <td>Private</td>\n",
       "      <td>130926</td>\n",
       "      <td>Bachelors</td>\n",
       "      <td>13</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Adm-clerical</td>\n",
       "      <td>Not-in-family</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>3674</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12164</th>\n",
       "      <td>28</td>\n",
       "      <td>Private</td>\n",
       "      <td>70604</td>\n",
       "      <td>HS-grad</td>\n",
       "      <td>9</td>\n",
       "      <td>Married-civ-spouse</td>\n",
       "      <td>Handlers-cleaners</td>\n",
       "      <td>Husband</td>\n",
       "      <td>White</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>38</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7289</th>\n",
       "      <td>28</td>\n",
       "      <td>Local-gov</td>\n",
       "      <td>125159</td>\n",
       "      <td>Some-college</td>\n",
       "      <td>10</td>\n",
       "      <td>Never-married</td>\n",
       "      <td>Adm-clerical</td>\n",
       "      <td>Other-relative</td>\n",
       "      <td>Black</td>\n",
       "      <td>Male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>Haiti</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8103</th>\n",
       "      <td>37</td>\n",
       "      <td>Local-gov</td>\n",
       "      <td>128054</td>\n",
       "      <td>Some-college</td>\n",
       "      <td>10</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>Adm-clerical</td>\n",
       "      <td>Unmarried</td>\n",
       "      <td>White</td>\n",
       "      <td>Female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40</td>\n",
       "      <td>United-States</td>\n",
       "      <td>&lt;=50K</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       age   workclass  fnlwgt      education  education_num  \\\n",
       "19182   53     Private  290640        HS-grad              9   \n",
       "4616    36     Private  130926      Bachelors             13   \n",
       "12164   28     Private   70604        HS-grad              9   \n",
       "7289    28   Local-gov  125159   Some-college             10   \n",
       "8103    37   Local-gov  128054   Some-college             10   \n",
       "\n",
       "            marital-status          occupation     relationship    race  \\\n",
       "19182   Married-civ-spouse       Other-service          Husband   White   \n",
       "4616              Divorced        Adm-clerical    Not-in-family   White   \n",
       "12164   Married-civ-spouse   Handlers-cleaners          Husband   White   \n",
       "7289         Never-married        Adm-clerical   Other-relative   Black   \n",
       "8103              Divorced        Adm-clerical        Unmarried   White   \n",
       "\n",
       "           sex  capital_gain  capital_loss  hours-per-week  native-country  \\\n",
       "19182     Male             0             0              50         Germany   \n",
       "4616    Female          3674             0              40   United-States   \n",
       "12164     Male             0             0              38   United-States   \n",
       "7289      Male             0             0              40           Haiti   \n",
       "8103    Female             0             0              40   United-States   \n",
       "\n",
       "       income  target  \n",
       "19182    >50K    True  \n",
       "4616    <=50K   False  \n",
       "12164   <=50K   False  \n",
       "7289    <=50K   False  \n",
       "8103    <=50K   False  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "loader = col.DataLoader(task='adults', root=ROOT_PATH)\n",
    "df = loader.load_data()\n",
    "display(df.sample(5))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Define Metrics for Success"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Report(\n",
      "        scorer: Scorer(scoring_fcts=[acc,f1,auc]), \n",
      "        to_show: [transformer,acc,f1,auc]\n",
      "        )\n"
     ]
    }
   ],
   "source": [
    "# define scoring metrics of interest\n",
    "scorer = col.Scorer(\n",
    "    acc=lambda ytrue, ypreds: metrics.accuracy_score(ytrue, ypreds > .5),\n",
    "    f1=lambda ytrue, ypreds: metrics.f1_score(ytrue, ypreds > .5),\n",
    "    auc=metrics.roc_auc_score,\n",
    ")\n",
    "\n",
    "# create a reporting object\n",
    "reporter = col.Report(scorer=scorer)\n",
    "\n",
    "# define which columns should be shown when reporting experiments results so far\n",
    "reporter.set_columns_to_show(['transformer'] + list(scorer.scoring_fcts.keys()))\n",
    "\n",
    "print(reporter)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    },
    "tags": []
   },
   "source": [
    "## Define Cross Validation Strategy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# cross validation strategy\n",
    "kf = KFold(n_splits=5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Feature Selection and Processing\n",
    "* define which variables should be fed into the downstream classifiers, and which ones should be considered as categorical or numerical.\n",
    "* pass through all numerical variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FeatureSelection(categoricals=['workclass', 'education', 'marital-status', 'occupation', 'relationship', 'race', 'sex', 'native-country'], numericals=['age', 'fnlwgt', 'education_num', 'capital_gain', 'capital_loss', 'hours-per-week'], target='target')\n"
     ]
    }
   ],
   "source": [
    "feature_selection = col.FeatureSelection(**loader.get_selected_features(df))\n",
    "\n",
    "print(feature_selection)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Model Selection and Fitting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>transformer</th>\n",
       "      <th>acc</th>\n",
       "      <th>f1</th>\n",
       "      <th>auc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MeanTargetEncoder(target=target, alpha=5)</td>\n",
       "      <td>0.852646</td>\n",
       "      <td>0.628722</td>\n",
       "      <td>0.908627</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 transformer       acc        f1       auc\n",
       "0  MeanTargetEncoder(target=target, alpha=5)  0.852646  0.628722  0.908627"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# define fitting function\n",
    "mte_transformer = col.MeanTargetEncoder(feature_selection)\n",
    "\n",
    "\n",
    "pipe = col.CategoricalPipeline(features=feature_selection,\n",
    "                               transformer=col.MeanTargetEncoder(feature_selection),\n",
    "                               scaler=MaxAbsScaler(),\n",
    "                               model=RandomForestClassifier(n_estimators=100, max_depth=5))\n",
    "\n",
    "cv_score = col.cv_score(pipeline=pipe,\n",
    "                        data=df, \n",
    "                        kf=kf,\n",
    "                        scorer=reporter.scorer)\n",
    "\n",
    "reporter.add_to_report(pipe.config, cv_score, show=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Compare to Baseline\n",
    "### Baseline 1: Assign all predictions to majority class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>transformer</th>\n",
       "      <th>acc</th>\n",
       "      <th>f1</th>\n",
       "      <th>auc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MeanTargetEncoder(target=target, alpha=5)</td>\n",
       "      <td>0.852646</td>\n",
       "      <td>0.628722</td>\n",
       "      <td>0.908627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>MajorityClass</td>\n",
       "      <td>0.759190</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 transformer       acc        f1       auc\n",
       "0  MeanTargetEncoder(target=target, alpha=5)  0.852646  0.628722  0.908627\n",
       "1                              MajorityClass  0.759190  0.000000  0.500000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# baseline: assign all predictions to majority class\n",
    "maj_results = pd.DataFrame([reporter.score(df.target, np.zeros(len(df))) for _ in range(5)])\n",
    "maj_config = dict(features=str(feature_selection),\n",
    "                  transformer='MajorityClass',\n",
    "                  model='Nonę',\n",
    "                  scaler='None'\n",
    "                 )\n",
    "\n",
    "reporter.add_to_report(maj_config, maj_results, show=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Baseline 2: one hot encode all categories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>transformer</th>\n",
       "      <th>acc</th>\n",
       "      <th>f1</th>\n",
       "      <th>auc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MeanTargetEncoder(target=target, alpha=5)</td>\n",
       "      <td>0.852646</td>\n",
       "      <td>0.628722</td>\n",
       "      <td>0.908627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>MajorityClass</td>\n",
       "      <td>0.759190</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>TransformStrategy_OneHotEncoder(handle_unknown...</td>\n",
       "      <td>0.838365</td>\n",
       "      <td>0.554263</td>\n",
       "      <td>0.897662</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         transformer       acc        f1  \\\n",
       "0          MeanTargetEncoder(target=target, alpha=5)  0.852646  0.628722   \n",
       "1                                      MajorityClass  0.759190  0.000000   \n",
       "2  TransformStrategy_OneHotEncoder(handle_unknown...  0.838365  0.554263   \n",
       "\n",
       "        auc  \n",
       "0  0.908627  \n",
       "1  0.500000  \n",
       "2  0.897662  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.preprocessing import OneHotEncoder\n",
    "\n",
    "# define fitting function\n",
    "ohe_transformer = col.TransformStrategy(feature_selection, cat_encoder=OneHotEncoder(handle_unknown='ignore'))\n",
    "\n",
    "pipe = col.CategoricalPipeline(features=feature_selection,\n",
    "                               model=RandomForestClassifier(n_estimators=100, max_depth=5),\n",
    "                               transformer=ohe_transformer)\n",
    "\n",
    "cv_score = col.cv_score(\n",
    "    pipeline=pipe,\n",
    "    data=df, \n",
    "    kf=kf,\n",
    "    scorer=reporter.scorer)\n",
    "\n",
    "reporter.add_to_report(pipe.config, cv_score, show=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Baseline 3: OrdinalEncoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>transformer</th>\n",
       "      <th>acc</th>\n",
       "      <th>f1</th>\n",
       "      <th>auc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MeanTargetEncoder(target=target, alpha=5)</td>\n",
       "      <td>0.852646</td>\n",
       "      <td>0.628722</td>\n",
       "      <td>0.908627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>MajorityClass</td>\n",
       "      <td>0.759190</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>TransformStrategy_OneHotEncoder(handle_unknown...</td>\n",
       "      <td>0.838365</td>\n",
       "      <td>0.554263</td>\n",
       "      <td>0.897662</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>TransformStrategy_OrdinalEncoder(handle_unknow...</td>\n",
       "      <td>0.850312</td>\n",
       "      <td>0.614018</td>\n",
       "      <td>0.902812</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         transformer       acc        f1  \\\n",
       "0          MeanTargetEncoder(target=target, alpha=5)  0.852646  0.628722   \n",
       "1                                      MajorityClass  0.759190  0.000000   \n",
       "2  TransformStrategy_OneHotEncoder(handle_unknown...  0.838365  0.554263   \n",
       "3  TransformStrategy_OrdinalEncoder(handle_unknow...  0.850312  0.614018   \n",
       "\n",
       "        auc  \n",
       "0  0.908627  \n",
       "1  0.500000  \n",
       "2  0.897662  \n",
       "3  0.902812  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.preprocessing import OrdinalEncoder\n",
    "\n",
    "# define fitting function\n",
    "ord_transformer = col.TransformStrategy(features=feature_selection, \n",
    "                                        cat_encoder=OrdinalEncoder(handle_unknown='use_encoded_value', unknown_value=-1),\n",
    "                                       )\n",
    "\n",
    "pipe = col.CategoricalPipeline(features=feature_selection,\n",
    "                               model=RandomForestClassifier(n_estimators=100, max_depth=5),\n",
    "                               transformer=ord_transformer)\n",
    "\n",
    "cv_score = col.cv_score(\n",
    "    pipeline=pipe,\n",
    "    data=df, \n",
    "    kf=kf,\n",
    "    scorer=reporter.scorer)\n",
    "\n",
    "reporter.add_to_report(pipe.config, cv_score, show=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Inspect MTE output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaIAAAHdCAYAAABff317AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAACi10lEQVR4nOydeVyN2/fHP0UiYyI0KJEGVzK7psyEEsoQElJRZleGBhHKkCHJkDmizIQMzSJDSjJTJCpFRYTq/P7od57vOZ3pKekY1vv18rq35+zznP2Ma++11/osmdzcXA4IgiAIQkrISrsDBEEQxN8NGSKCIAhCqpAhIgiCIKQKGSKCIAhCqpAhIgiCIKQKGSKCIAhCqpAhIv5IGjRogGHDhkm7GwRBsIAMEfFb0qBBA7Rt21ba3fjrGDZsGBo0aICXL19KuyvEH0R1aXeAIH4GN2/eRK1ataTdDYIgWECGiPgjad26tbS7QBAES8g1R1Q6L1++ZNZocnJyMGfOHOjo6EBZWRndunXDgQMH+Np/+/YNO3fuhLm5Of755x8oKytDQ0MDpqamCA0N5WsbHR2NBg0aAADS0tLQoEED5t+MGTOYdmXXiObNm4cGDRrg9OnTQvv85MkTNGjQAMbGxnzbCwsL4ePjAyMjI6iqqkJFRQV9+vTBnj17wOGUXx3r7du3WLp0KTp37oymTZuiefPm6NmzJ9zd3VFQUMDX9sWLF5g5cyb09fXRuHFjaGtrw9raGklJSQL7XbNmDRo0aIDo6GihvytszYz7nUOHDiEqKgrDhg2Dmpoa1NXVYWFhgYcPHwrs49q1awCAdu3aMeed10XKdd2lpqbCx8cH3bp1Q5MmTWBpaYldu3ahQYMG8PT0FNrH/Px8qKiooE2bNiguLpZ8Mok/BpoRET+NvLw8DB48GDVq1ICpqSm+fv2K06dPY/bs2ZCVlcXEiRMBAB8+fMDixYvRtWtX9O3bF40aNUJGRgbOnz+PsWPHYtOmTbC2tgYANG/eHE5OTvDy8kK9evX4jI+4NSNLS0vs3bsXhw8fxogRIwQ+DwwMBACMHz+e2fbx40eYmZnhzp07MDAwgKWlJQDg6tWrmD9/Pm7dugU/Pz/W5yMhIQGjR49GTk4OunTpgiFDhuD79+949uwZtm7dCmtra9SuXRsAcPfuXYwYMQL5+fkYPHgw2rRpg5SUFJw9exYXLlxAQEAABg4cyPq3xREaGooLFy5gwIABmDJlCh4/foxLly4hPj4ecXFxaNSoEQDAyckJhw8fRlpaGuzt7VG/fn0AYP7Ly6JFixAXF4fBgwdj0KBBqFOnDsaNG4cVK1bg4MGD+O+//1CtWjW+7xw5cgSfP3/G7NmzBT4j/mzIEBE/jfv378Pa2hobNmxgXiwzZ85Ejx49sGXLFsYQNWjQAElJSVBVVeX7fm5uLgYPHozly5dj7NixqFWrFjQ0NLBkyRJ4eXmhfv36WLJkCau+dO7cGTo6Orh69SqysrKgrKzMfFZSUoKgoCAoKChg5MiRzPalS5fizp07WL58OebOncts//r1KyZNmoTAwECYmJhg6NChEn//27dvsLKyQk5ODjZv3ozJkyfzfZ6Tk8MYIQ6HA3t7e+Tn52Pbtm2MAQSAiIgIjBw5Evb29khKSoKCggKr4xdHSEgITp06hV69ejHb3N3dsXHjRgQEBDDHvmTJEsTExCAtLQ0zZsyAhoaGyH0mJSUhKipKoM3YsWPh7++PixcvCszQ9u3bh+rVq8PKyuqHj4n4vSDXHPHTUFBQgIeHB9/oVldXF926dcOTJ0/w8eNHAIC8vLyAEQJKDdTEiRORm5uL+Pj4H+7P+PHjUVRUhKCgIL7tERERSE9Px/Dhw1G3bl0ApbO0wMBAGBgY8Bkhbn9dXV0BAEePHmX12xcuXMCrV6/Qv39/ASMEAEpKSqhZsyYAIC4uDo8fP0aHDh34jBAA9OnTB8OHD0dOTg5CQkJY/bYkzM3N+YwQAGYGWtHzPmvWLKGGysbGBgCwd+9evu03btzAgwcPMGTIEKioqFToN4nfF5oRET+Nli1bok6dOgLbuUYnLy+PefE/fPgQW7ZsQWxsLDIyMvD161e+77x9+/aH+zN27FisXLkShw8fhqOjI7Od65abMGECs+3OnTsoKiqCrKws1qxZI7CvoqIiAMDTp0+ZbcLaWVpaQkNDA7dv3wYADBo0SGI/ExMTAQC9e/cW+nmfPn1w9uxZJCYmwsLCQuL+JGFoaCiwjXuNcnNzK7TPTp06Cd2uq6uLnj17IiwsDKmpqdDU1ATwP8M0bdq0Cv0e8XtDhoj4adSrV0/odu4MibsgfevWLZiamqKoqAhGRkYwNjZG3bp1ISsri6SkJJw/f17AMFWEZs2aoV+/frh8+TISEhJgaGiI/Px8nDt3Dmpqanyzgvfv3wMoXddJSEgQuc9Pnz4x/+/l5SXwec+ePaGhoYG8vDwAYDXaz8/PBwA+9yEvTZo04Wv3owi7TtWrl74aKho0IKrvADB9+nTExMRg//79cHNzw4cPH3D69GloaWmhT58+Ffo94veGDBEhddavX48vX77g7NmzAi4ib29vnD9/vtJ+y9LSEpcvX8bhw4dhaGiIU6dO4cuXLxg3bhxkZf/nqea+nG1tbbF27VpW+xY3e+Au6LOZ2XF/OysrS+jnmZmZfO0AMH0XZjgqOqv5EWRkZER+NmzYMKioqCAgIABLlizB4cOHUVhYCGtra7HfI/5caI2IkDovXryAoqKigBECwIQLl0VWVhYlJSXl/q2hQ4eiQYMGOH78OL5//8645cquxXTq1AmysrK4fv16uX9DGJ07dwYAXLp0SWLbdu3aAYDIUOzIyEgA/C41bkj769evBdrfvXu3PF0VC3c2W5Fzz6V69eqYPHky3r17h3PnzmH//v2Ql5fnc40SfxdkiAip07x5c3z48AH379/n237gwAFcvXpV6HeUlJSQnZ2NL1++lOu35OXlYW5ujpycHGzfvh3Xr1/Hv//+Cy0tLb52jRo1wtixY5GUlIQ1a9Ywa0K8pKen48mTJ6x+19jYGBoaGrhy5QoOHjwo8Pn79+9RWFgIAOjatSt0dHRw584dgWCIyMhInD17FkpKSnzRelxDFxAQgO/fvzPbc3Jy4OLiwqqPbFBSUgJQmsP1I1hbW0NOTg5Lly7FkydPMGLECGbfxN8HueYIqTNjxgxcvXoVxsbGMDMzQ7169XD37l3cuHEDI0aMEJqE2rdvXwQFBWH06NHo3r075OXl8c8//wgkpArD0tIS/v7+WLFiBQD+3CFe1q5dixcvXsDLywtHjx5F9+7d0aRJE2RmZuLZs2e4desWVq1axUrFQU5ODvv378eoUaMwa9YsHDp0CF26dEFRURGeP3+OiIgI3Lx5ExoaGpCRkYGfnx/MzMxgb2+PkydPMnlEZ86cQY0aNbB9+3a+0O0OHTrAyMgIkZGR6NOnD/r06YMPHz7g0qVLMDIyEjDyFaVv3744efIk5syZgxEjRqB27dqoX78+bG1ty7WfJk2aYPjw4Th58iQAYOrUqZXSP+L3hAwRIXUGDBiAI0eOYP369Th58iRkZWXRsWNHnD17FqmpqUINkaenJ2RlZREeHo64uDgUFxdj/PjxrAxRhw4doKenh4cPHwrkDvFSt25dnDt3DgcPHkRwcDDOnTuHwsJCNG7cGM2bN4erqyvMzMxYH6ehoSGio6OxefNmXLp0Cdu3b0etWrXQvHlzODo6onHjxnx9jIiIwLp16xAREYGrV6+ifv36GDZsGBYsWAADAwOB/R88eBDu7u44d+4cdu3ahebNm2PWrFlwdHTE8ePHWfdTHBMnTkR6ejqCgoLg6+uL79+/Q11dvdyGiLuvkydPQl9fH926dauU/hG/JzK5ubnl1ykhCIL4QTZs2ICVK1di/fr1TH4R8XdChoggiCrn06dP6NSpEz5//ozk5GQmn4z4OyHXHEEQVcaFCxdw9+5dXL58GRkZGXBzcyMjRJAhIgii6jhz5gwCAwOhrKyMuXPnYvbs2dLuEvELQK45giAIQqpQHhFBEAQhVcgQEQRBEFKFDBFBEAQhVX5JQ8QrrU9tqI002vxKfaE21OZH2vxKfRHFL2mICIIgiL8HMkQEQRCEVCFDRBAEQUgVMkQEQRCEVGFtiPz9/WFgYIAmTZrAyMgIsbGxrL73/PlzqKmpQVVVtcKdJAiCIP5cWBmiEydOYPHixViwYAGioqLQpUsXWFhYSCyO9e3bN0ydOhXdu3evlM4SBEEQfx6sDJGvry8sLS0xefJk6OjoYN26dWjSpAn27Nkj9ntubm5o06YNRowYUSmdJQiCIP48JBqib9++ISEhAf369ePb3q9fP8TFxYn8XmhoKEJDQ+Hl5fXjvSQIgiD+WCSKnr59+xZ6enoICQlBjx49mO1eXl4IDg7G7du3Bb6TkZGBPn364ODBg+jcuTMOHTqERYsWIT09XWxnGuwV/PxWz89sj4UgCIL4hdDW1mbVjnUZCBkZGb6/ORyOwDYutra2mDp1Kjp37sx29yIRdSBPnz6VeJDUhtpUtM2v1BdqQ21+pM2v1BdRSHTNKSkpoVq1asjKyuLbnp2djcaNGwv9TlRUFLy8vKCkpAQlJSXMmjULBQUFUFJSwr59+8rdSYIgCOLPReKMqEaNGjA0NER4eDjMzMyY7eHh4TA1NRX6nbKh3efPn8eGDRtw9epVqKio/FiPCYIgiD8KVq45BwcH2NnZoWPHjujatSv27NmDjIwMTJkyBQDg7u6OO3fu4MyZMwAAfX19vu/fvXsXsrKyAtsrwv/WkRSAmNL/z51COUoEQRC/K6wM0ahRo/D+/XusW7cOmZmZ0NPTQ1BQEJo3bw6gNDghJSXlp3aUIAiC+DNhHaxgY2MDGxsboZ/5+fmJ/e6ECRMwYcKE8vWMIAiC+CtgbYh+J8h9RxAE8ftAoqcEQRCEVCFDRBAEQUgVMkQEQRCEVCFDRBAEQUiVPzJYgQ0U0EAQBPFr8NcaIjaUNVZkqAiCICofcs0RBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVqFT4D8KmnDiVHCcIghANzYgIgiAIqUKGiCAIgpAqZIgIgiAIqUJrRL8ItI5EEMTfCs2ICIIgCKlChoggCIKQKmSICIIgCKlCa0S/EbSORBDEnwjNiAiCIAipQoaIIAiCkCpkiAiCIAipQoaIIAiCkCoUrPCHQQENBEH8bpAh+gshY0UQxK8EueYIgiAIqUKGiCAIgpAqZIgIgiAIqUKGiCAIgpAqrA2Rv78/DAwM0KRJExgZGSE2NlZk20ePHmH48OHQ1tZGkyZN0K5dO6xYsQLfvn2rlE4TBEEQfw6souZOnDiBxYsXY8OGDejWrRv8/f1hYWGBGzduQF1dXaB9jRo1MH78eBgYGKB+/fq4f/8+5syZg6KiIqxYsaLSD4KofCiyjiCIqoKVIfL19YWlpSUmT54MAFi3bh2uXr2KPXv2wM3NTaC9lpYWtLS0mL+bN2+OmJgYXL9+vZK6TRAEQfwpSHTNffv2DQkJCejXrx/f9n79+iEuLo7Vj7x48QJXr15Fjx49KtZLgiAI4o9FoiHKyclBcXExGjduzLe9cePGyMrKEvvdQYMGoUmTJujQoQO6desGV1fXH+stQRAE8cchk5ubyxHX4O3bt9DT08P58+fRvXt3ZrunpyeOHz+OW7duifzu69ev8enTJ9y/fx+urq6wsbHB/PnzRbb/37rE/7jV8zPf351jFKTWpuznf3sbgiAIcWhra7NqJ3GNSElJCdWqVROY/WRnZwvMksqipqYGANDV1UVxcTFmz56N2bNno3p19spCAgcSI2isqqqN0JP6N7f5f54+fSrxhvvd2vxKfaE21OZH2vxKfRGFRNdcjRo1YGhoiPDwcL7t4eHh6Nq1K+sfKikpQVFREYqLi8vdSYIgCOLPhdXUxMHBAXZ2dujYsSO6du2KPXv2ICMjA1OmTAEAuLu7486dOzhz5gwA4MiRI6hZsyb09fVRo0YN3L17FytWrMCIESMgLy//846GIAiC+O1gZYhGjRqF9+/fY926dcjMzISenh6CgoLQvHlzAEBGRgZSUlL+t9Pq1eHt7Y0XL16Aw+FAXV0dNjY2mDlz5s85CoIgCOK3hfVijY2NDWxsbIR+5ufnx/e3ubk5zM3Nf6xnBEEQxF8B1SMiKgypLxAEURmQ6ClBEAQhVcgQEQRBEFKFDBFBEAQhVcgQEQRBEFKFDBFBEAQhVcgQEQRBEFKFDBFBEAQhVcgQEQRBEFKFDBFBEAQhVUhZgfipkPoCQRCSoBkRQRAEIVXIEBEEQRBShVxzhFThLw9P7juC+BuhGRFBEAQhVWhGRPzy0KyJIP5saEZEEARBSBUyRARBEIRUIUNEEARBSBUyRARBEIRUIUNEEARBSBWKmiP+CCiyjiB+X2hGRBAEQUgVmhERfw00ayKIXxMyRATBQ1m1cABkrAjiJ0OuOYIgCEKqkCEiCIIgpAoZIoIgCEKqkCEiCIIgpAoZIoIgCEKqkCEiCIIgpAqFbxNEOaEQb4KoXGhGRBAEQUgVMkQEQRCEVCFDRBAEQUgVMkQEQRCEVCFDRBAEQUgVMkQEQRCEVCFDRBAEQUgVMkQEQRCEVCFDRBAEQUgVMkQEQRCEVCGJH4L4CZAMEEGwh2ZEBEEQhFQhQ0QQBEFIFdaGyN/fHwYGBmjSpAmMjIwQGxsrsm10dDTGjx8PHR0dNGvWDN27d8fBgwcrpcMEQRDEnwUrQ3TixAksXrwYCxYsQFRUFLp06QILCwukpaUJbX/z5k20adMG+/fvx/Xr1zFt2jTMnTsXwcHBldp5giAI4veHVbCCr68vLC0tMXnyZADAunXrcPXqVezZswdubm4C7RcsWMD397Rp0xAdHY0zZ87AwsKiErpNEL8/FNBAEKVINETfvn1DQkICZs2axbe9X79+iIuLY/1DHz9+hIqKSvl7SBB/MWSsiL8BiYYoJycHxcXFaNy4Md/2xo0bIysri9WPXLx4EZGRkQgNDS13B58+fVpmi4LU2gh+Tm1+vI30ruef0qZzjML/2sak41bPzwLfEf8b1OZPbyOtvmhra0tsD5Qjj0hGRobvbw6HI7BNGDdu3MD06dPh5eWFjh07sv05BoEDiUmXWhuhJ5Xa/FgbKV7PP7WNuIf/6dOnEl8O1ObPavMr9UUUEoMVlJSUUK1aNYHZT3Z2tsAsqSzXr1+HhYUFlixZgmnTppW7cwRBEMSfj0RDVKNGDRgaGiI8PJxve3h4OLp27Srye9euXYOFhQUWLVqEmTNn/nhPCYIgiD8SVuHbDg4OOHz4MA4cOIDHjx/DyckJGRkZmDJlCgDA3d0dpqamTPvo6GhYWFhgypQpGDNmDDIzM5GZmYns7OyfcxQEQRDEbwurNaJRo0bh/fv3WLduHTIzM6Gnp4egoCA0b94cAJCRkYGUlBSm/eHDh/H582f4+PjAx8eH2a6uro6kpKRKPgSCIAjid4Z1sIKNjQ1sbGyEfubn5yfwd9ltBEEQBCEM0pojCIIgpAoZIoIgCEKqkCEiCIIgpAoVxiOIv4CyUkEkE0T8StCMiCAIgpAqZIgIgiAIqUKGiCAIgpAqZIgIgiAIqUKGiCAIgpAqZIgIgiAIqUKGiCAIgpAqZIgIgiAIqUKGiCAIgpAqZIgIgiAIqUKGiCAIgpAqZIgIgiAIqUKGiCAIgpAqpL5NEAQAUugmpAfNiAiCIAipQoaIIAiCkCpkiAiCIAipQoaIIAiCkCpkiAiCIAipQoaIIAiCkCpkiAiCIAipQoaIIAiCkCpkiAiCIAipQsoKBEGwhtQXiJ8BzYgIgiAIqUKGiCAIgpAq5JojCKJSIfcdUV5oRkQQBEFIFZoREQRR5dCsieCFZkQEQRCEVCFDRBAEQUgVcs0RBPFLQu67vwcyRARB/LaQsfozINccQRAEIVXIEBEEQRBShQwRQRAEIVXIEBEEQRBShQwRQRAEIVXIEBEEQRBShcK3CYL4o6EQ718f1jMif39/GBgYoEmTJjAyMkJsbKzItoWFhZgxYwa6d++ORo0aYdiwYZXSWYIgCOLPg5UhOnHiBBYvXowFCxYgKioKXbp0gYWFBdLS0oS2Ly4uRs2aNWFra4tBgwZVaocJgiAqmwZ709Fgbzo6xyjwzKCIqoKVIfL19YWlpSUmT54MHR0drFu3Dk2aNMGePXuEtq9duzY2btwIa2trqKrSNJggCIIQjURD9O3bNyQkJKBfv3582/v164e4uLif1jGCIAji70CiIcrJyUFxcTEaN27Mt71x48bIysr6aR0jCIIg/g5YR83JyMjw/c3hcAS2/QyePn1aZouC1NoIfk5tfryN9K7nn9rm17/mv2sb9p//am2k1RdtbW2J7QEWhkhJSQnVqlUTmP1kZ2cLzJJ+BgIHEiO4kFhVbYSeVGrzY22keD3/1Da//DX/Xdv8P0+fPpX4gv2V2vxKfRGFRNdcjRo1YGhoiPDwcL7t4eHh6Nq1a7l/kCAIgiB4YeWac3BwgJ2dHTp27IiuXbtiz549yMjIwJQpUwAA7u7uuHPnDs6cOcN859GjR/j27RtycnJQUFCAe/fuAQAMDAx+wmEQBEEQvyusDNGoUaPw/v17rFu3DpmZmdDT00NQUBCaN28OAMjIyEBKSgrfd8rmGfXu3RsAkJubW0ldJwiCqBr4c4tIoaGyYR2sYGNjAxsbG6Gf+fn5CWxLSkqqeK8IgiCIvwbSmiMIgqgEaNZUccgQEQRBVBFkrIRDZSAIgiAIqUKGiCAIgpAq5JojCIL4hfgb3Xc0IyIIgiCkChkigiAIQqqQISIIgiCkChkigiAIQqqQISIIgiCkChkigiAIQqqQISIIgiCkChkigiAIQqqQISIIgiCkCikrEARB/Gb8aeoLZIgIgiD+QP5nrEoNFYBf1liRa44gCIKQKmSICIIgCKlChoggCIKQKmSICIIgCKlCwQoEQRB/Kb9KQAMZIoIgCEIkVWGsyDVHEARBSBUyRARBEIRUIUNEEARBSBUyRARBEIRUoWAFgiAI4of40YAGmhERBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVyBARBEEQUoUMEUEQBCFVWBsif39/GBgYoEmTJjAyMkJsbKzY9snJyRg6dCiaNm0KPT09eHl5gcPh/HCHCYIgiD8LVoboxIkTWLx4MRYsWICoqCh06dIFFhYWSEtLE9o+Pz8fI0eOhLKyMsLCwuDp6QkfHx9s3bq1UjtPEARB/P6wMkS+vr6wtLTE5MmToaOjg3Xr1qFJkybYs2eP0PbBwcH48uUL/Pz8oK+vjxEjRmDOnDnYtm0bzYoIgiAIPmRyc3PFWoZv376hWbNm2L17N8zMzJjtCxcuxIMHD3D+/HmB79jZ2eHDhw8ICgpitsXHx6Nfv35ISEiApqZmpR0AQRAE8XsjcUaUk5OD4uJiNG7cmG9748aNkZWVJfQ7WVlZQttzPyMIgiAILqyDFWRkZPj+5nA4AtsktRe2nSAIgvi7kWiIlJSUUK1aNYGZTHZ2tsCsh4uysrLQ9gBEfocgCIL4O5FoiGrUqAFDQ0OEh4fzbQ8PD0fXrl2FfqdLly64fv06CgsL+do3a9YMGhoaP9hlgiAI4k+ClWvOwcEBhw8fxoEDB/D48WM4OTkhIyMDU6ZMAQC4u7vD1NSUaW9ubo5atWph5syZePDgAc6cOYNNmzZh5syZ5JojCIIg+KjOptGoUaPw/v17rFu3DpmZmdDT00NQUBCaN28OAMjIyEBKSgrTvn79+jh58iQWLlyIvn37okGDBnBwcICjo+PPOQoecnJykJKSgrZt20JeXl5om7dv3+Ldu3coKSnh225oaPhT+pSZmYlv377xbVNXV8e2bdswfvx4KCoq8n0mKj9LGOrq6qza3bt3DwYGBmLbxMTEoGbNmujUqRMA4NChQzh48CB0dXXh4eGBOnXqsO5XYWEhQkNDkZKSAmtrazRo0AApKSlo0KCBwPH+DH70HL548QIqKiqoWbMmAKBnz56wsrLCmDFj0KBBgyrrS25urkDKg6KiIh49eoRq1apBW1sbQKnHITAwELq6upgzZw6qVasm8be+f/8OW1tbuLq6okWLFqz7yP2unJycwPbIyEg8fvwYMjIy0NHRQe/evcu134rw5csXxMXFQUtLi3kncUlNTcWjR4+Y/pQnYvdHzs+vztevXxEUFMRcK11dXZibm4t8Zwpj586dqF+/PsaOHcu3/ejRo/j48SNsbGxY70ti+HZVcOLECURGRgo1DkeOHAEAREVF4fjx43j9+rXAS/3s2bP4+PEjZs2ahdOnT0NGRgbx8fHQ1NTEvHnzoKysjCVLliAxMRF2dnZ48uSJwMMtIyOD9+/fAwBev36N69evC+0P15gWFRXhzp07Qvszfvx45OXlwcnJCadOnRL4HADev3+Pf/75B+/evYOxsTGsrKzQt29fyMjIQFFRkfXMkdvnZ8+e4fTp00L74+vrC0VFRRgYGMDKygrm5uaoX7++wL569eqFxYsXY9iwYXj69Cl69OiBSZMm4fr16+jWrRu8vb3F9q1mzZpo0aIFjI2NERQUhE+fPiEvLw937tyBpqYmnJ2dkZeXBx8fH1bnEAA+fPiAK1euCLTJycnBzZs3kZ6eDmtrayxbtozv++U5h3PnzkWrVq1gaWkJDoeDkSNHIjIyEvXq1cPx48fRqVMnrFy5EkePHkV2djaGDRsGKysrGBkZwcDAQOLvvHz5EgAgKyvZAfH+/Xu8evUK8+fPR3R0NL5//858xg0Qev/+PQYOHAh7e3uMHj0a6enp6NSpE3r27InIyEg0adKE1Qv37Nmz0NDQQGRkpNj227dvR7NmzTBixAgApc9AYGAgWrRogcDAQGhra+PNmzeYOHEiEhIS0KxZMwClA7727dsjICCA2Sbset68eRMA0LdvX4l9dnR0xIwZM9CxY0fY2Njg27dv6NOnDx4+fIgaNWogICAAAwcORH5+PmbNmoUzZ84w553D4aBOnTq4du0a1NXV0b17d7G/FRsby+r8LFq0SOx+1q5di9zcXKxZswbR0dHIzs4WeK88e/YMABAQECDyPRcZGYmzZ8/i2rVrePXqFQoLC9GoUSMYGBhg4MCBzFKJpHfBo0ePYG5ujvz8fLRp0wZAqRIO937X0dFhtZ/27dvDx8cHPXv25Pvs+vXrcHBwQHx8vNjzwgurGdHPxMXFBX5+fujVqxeaNm0q9KE+dOgQ5s+fj+HDhyMmJgZDhw7Fs2fP8PLlS8YaL1++HG/fvkVkZCSMjY2Z7w4ePBgrV67EkiVLMHfuXKiqqmLz5s0ifysoKAiOjo6oXr06lJSU+NrIyMjA0dERT548wbhx4/Dy5UtwOBxUq1YNRUVFkJOTg7y8PMaPHw8XFxfcv38fhw4dwqRJk7B161a8efMG27dvx6pVqwAASUlJCAsLw6FDh2BpaYlGjRph/PjxOHToEFRUVACU3gxubm6YMmUKOnfuDAC4desW9u3bh+XLlwMAQkNDYWVlBQMDAyQkJKBDhw5ISUnB169f8e+//wIAbt++jYCAAGzYsAHOzs4wMTHBxIkT+UasqampzI155swZ9O3bFxs2bMDt27dhZWUFb29vrFu3Dp6enhg+fDg6duwIALhz5w5CQkIwZ84cpKenw9vbG506dcKFCxf41gSNjY3h4OAAAKzO4a1btzBmzBjIy8sjOzsbzZo1Q2ZmJuTl5VFSUoIOHTrA2dkZGzduxIcPH7B+/Xrmt3jXNCWdw+XLl2Pv3r0AgMuXLyMpKQlXrlxBUFAQli9fjnPnzsHFxQXOzs64cuUKDh06hDFjxkBZWRna2tro0KED6tevj4KCAmzbtg0dOnTg+53s7GxYWFjA2tqa1fV0cHBAXl4etm7dKvI+ffz4Mdq1awcAOHXqFDp27Ijg4GBMmDABkZGRGDJkCEpKShAcHAxlZWXmWsXHxyMzMxNjxowBAJiYmODs2bOYNWuWwG9w2bFjB6OKcu3aNZw6dQr+/v44c+YMnJ2dcfToUTg5OaFatWrMAJB7P9na2sLJyQkHDhwQeT2/f/+O6tWr4+nTpyL7APzv+QsLC4OdnR0A4MKFC/j48SOePHmCgIAAeHp6YuDAgVi8eDGSk5Nx9uxZ5gV948YNTJ48GatWrcL27dthYmIicRDB5vw8ePCA7++ioiI8efIERUVFzDWys7PDo0ePMH78eCgrKwv93S1btsDb2xtTpkxBbGwspk2bhhcvXiAmJgYtW7aErq4umjZtio4dO6J9+/aoVasWPnz4gGvXrmHbtm1QV1fHkCFDsH37drHvgsWLF6Nt27bYsWMH6tWrB6BUDcfW1hZLlizBiRMnWL1T3rx5I9SboKKigjdv3og9rwLk5uZypPmvcePGnP3794tto6enx9myZQsnNzeXU6dOHU5CQgInNzeXM336dM7cuXM5ubm5HBUVFU5YWJhAm7t373Lq1KnDyc3N5SgoKHBu374t9rc0NTU58+fP5+Tk5Ihs079/f86oUaM46enpnDp16nDu3r3LiYiI4HTs2JFz8uRJpj/nz5/n5ObmcurWrcuJj4/n5Obmcvz9/Tl9+vQR2GdKSgrHy8uLY2BgwKlWrRqnT58+nN27d3P+/fdfoedn//79nH///ZeTm5vLadeuHcfV1ZXv2DMyMjgjRozgeHh48H3v/fv3nKNHj3JMTU05NWrU4GhqanKcnZ05ycnJnHr16nHu3LnDyc3N5fTu3Zuzdu1aTm5uLufevXucmjVrcnJzcznGxsbMteD9t2XLFo6xsTEnNzeXU6tWLY6WlpbAtUhMTGT2w+Yc/vvvvxxbW1vOhw8fmP08efKE06tXL468vDzn+vXrnNzcXE5ycjJHQ0OD07VrV87EiRM5SUlJnAULFnAmTJjAyc3N5XTv3l3sOZSXl+ckJycz95SNjQ0nNzeXEx8fz6lXr57QeyAlJYXj6urKkZeX51SrVo3Tr18/Tp8+fTguLi4CbV1dXTljxoxh1Zfc3FxO7dq1ObGxsWLv0zp16nASExM5ubm5nMGDB3NWrFjByc3N5SQlJTHneMaMGcz54/2uvb09x87OjpObm8tZvHgxp169ehxjY2PO0qVLOR4eHnz/cnNzOTVr1uQkJSVxcnNzObNmzWLOa1xcHKdhw4bMPR4RESHQz/DwcOYcirueO3fuZP3O4L1ekyZN4jg4ODD3F/dZV1RUZJ4/3n8hISEcRUVF1r/F5vwI+5eRkcGxsLDgeHt7M9dL2Pnh/deyZUvm3uB9bhQUFDj6+vrM/S7s39u3bzn+/v4cBQUFzoABA8S+C2rVqiV0X9euXeMoKCiwfqeoq6tzAgICBPZz4MABjqqqKutznJuby5G6+nZJSQnatm0rtk1qaiqMjIwAlEbxffr0CQAwffp0HD58GECpL71hw4YC3/348SMzNdfX10dmZqbY33r37h2srKzE+tjj4+OxcOFC1K5dG7KysigqKoKhoSHc3d0ZF1FeXh4zWqhXrx7jQuvcuTPjiuBFUVERhoaGMDAwQPXq1fHy5UssXLgQ169fR0FBgUD7Nm3aICEhAUDpiH/UqFEAgOrVq+Pz58+oWbMmFi1ahO3bt/N9T1ZWFoMHD8bOnTvh5uaGt2/fYtWqVTA0NIS8vDzc3d1x5MgRXL9+HQMHDgQAvHr1CsrKygBKXaRlp+IAGLcQAMjJySE9PV2gzevXr5kRGJtzmJycDFtbW8jIyEBWVhZfv36FsrIy3N3dUVRUhKKiIgCAqqoqLly4gHbt2iE7OxtFRUVIT09nXGLx8fHMTE/YOWzYsCGzjhMWFoZevXoBALP/sty6dQvLly/Hpk2b0LRpUzg5OaFFixaIiIhgXCy8mJmZ4cKFC6z6AgAaGhpC3bm86OnpYc+ePYiNjUVkZCT69+8PoNQdpqSkBKDUrc09f7zY2Njg6NGjAIDDhw+jQYMGSE5ORkBAAHbu3Mn827VrFwCgbt26yMnJAVA60+Q+i3Jycvj69avYfvL+trjr6enpKXY/vCgrK+Phw4coLi5GWFgY+vTpAwAoKChA9eqlTp7CwkKh7wNFRUWmzyYmJsjNzRVok5+fDxMTE9bnRxg1a9bEggULsGHDBgCApqamRHmzN2/eoEOHDsz38/PzAZS6UN+8eQM9PT2R361VqxbMzc0hIyODpUuXAhD9LpCXl0deXp7Q4+auEbF5p1hYWGDx4sUIDw/H9+/f8f37d4SFhWHp0qWwsLAQe6xlkbohsra2Zh4KUTRs2JAxPs2aNcPDhw8BlPrTuSHi7du3Fyo3tG/fPmZq7uLiAjc3N0RERCArKwsfPnzg+wcAAwcOxO3bt8X2h8PhQEFBAUBpnhV3GqqqqsoEbWhqaiI1NRUA0Lp1axw/fhwcDgdnz57lW6zPysrC5s2b0aVLF5iZmeHbt284duwY7t69i4cPH0JRUVGoD9rf358xdHXq1GHOQ9OmTfHixQsApS/Ssg/anTt3MG/ePOjo6MDPzw9z587FvXv3cOHCBWhqauLSpUtYtGgRFixYwCzQnj59mjmHioqKCAkJEehPSEgI8+B36tRJ4OWXn5+PNWvWYNCgQazPIe9iuLKyMmMsateuDQC4ePEi87mKigrWrl2LI0eOoEWLFti+fTvOnTsHAGjevDn8/f1FnkMTExPY2NjAzMwMHz58wIABAwCUuk655+Ddu3fw8fFBt27dMHz4cOTl5WH//v1ITEzEokWLsGHDBigqKuL48eMCvxMTE4NatWqx6gsArFmzBu7u7sx1FMby5cuxf/9+DB8+HKNHj2aM24ULF5iXGYfDQXJyssB3ebfdu3dP5L/ExEQApWs3s2fPhqOjI1JSUpgBysOHDxnXa+/evbF48WK8fv2a2XdaWhqWLFnCuH/FXc+MjAzmsw8fPiA4OBgbN26El5cX3z8AmDhxIqZOnYp///0XsrKyjGG8ffs2WrduDQDo2rUrVq1ahc+fPzP7LSgowJo1a9ClSxfmuvCuwXH5+vUrrl+/zvr8iCI7O5t5b3l6esLd3R1JSUkoLi4W2l5ZWZkx+Orq6rh16xZzPtiud9apU4d5rkS9C4YMGYI5c+bgxo0bKC4uRnFxMa5fv4558+Yxyxps3ilLlixB165dMWrUKDRr1gzNmjWDubk5unTpIrBmKwmprxHl5eUhODgYERERaNOmDTOi4bJ27Vr8+++/CAsLQ5s2bTBy5Eg4OTkhPDwcUVFRzGjI1dUVo0ePxqNHj1BUVMQsysXHxzMvTq5W3siRI/kuLO8icN++fbF8+XI8evQI+vr6Av0xNTWFnp4ekpKSoKmpiY4dO2Lz5s2oVq0aDhw4wLy4LC0tkZycjF69emHu3LkYN24cdu3ahZKSEmb0N3bsWISFhUFbWxtTp07FuHHj+CKyatasibVr12L69Ono3LkzE812584dvHr1CgcPHgQAdOzYETdu3ICuri4GDRoEZ2dn3L9/H+fOnWPWIbZu3YpDhw7h+fPnGDRoEHbt2oUBAwYws0V1dXXs2LEDnTt35nspAMDKlSuZGaKTkxPmzJmDqKgodOjQgQkMCQsLw+bNm5n+3Lx5E506dUJhYSGmTp2KFy9eQFlZGfv27QMAVuewXbt2iI+PR6tWrdCzZ094eHggKysLQUFB0NfXh6qqKqt7bPXq1Zg4cSKuXr0q9Bz27dsX6urqeP36Ndzd3RlDl5GRgWnTpgEonU1raWlh4sSJsLS0ZGYdvNjb28PT0xPz5s1jfuf27dsIDAzE4sWLxfblyZMnqFWrFmOMCgsL0alTJ8jLywvcg2lpaejRoweeP3+Ojx8/8t0z1tbWzIto4sSJmD17Nl68eMHXn82bN2PChAkC/f/06RNkZGSY4+eyfv16rFy5Eq9fv8b+/fuZgVRiYiJGjx4NAPDy8oKlpSUMDQ2Zda23b9+iTZs2jOEVdz25hlTcuqC6ujqcnJzg5OQEXV1dvH79GmZmZqhRowaA0pH7nDlzmPNsbm4OPT09tGnTBjIyMrh//z5q166N1atXM7PP5ORkvvNXUlKCq1evMsEVbM5P2aoCHA4HmZmZCA4OZgZeWlpaKCwsZIxmWd6/f4/evXvjwoULMDQ0xKRJk7B06VKcPHkS9+7dY95dK1euhKqqKqZOncr3/T179uDNmzes3gWenp6YMWMGjI2Nmee6pKQExsbGWLNmDQB27xQ5OTns3r0bS5cuRVJSEjgcDtq1awctLS2hxygOqUfNDR8+XORnMjIyOHv2LD58+IDCwkI0a9YMJSUl2LJlC27cuIFWrVph4cKFzI2UnJwMHx8fJCYmoqSkBO3atcOcOXOYmzwmJkZsX3r27Ck2tJhrrK5evYqCggKYmpoiNTUVY8eOxZMnT6CkpIS9e/cyrh1e0tLScPfuXbRs2ZLpj6OjIyZPnsxcWGFwOBzcunULFy9eZKL9dHV1MWXKFKipqQEodV1++vQJ//zzDz5//gxnZ2fm/KxatQrq6uro0KEDJk2ahAkTJjButrJwZ2N6enpISUnB4MGDUbt2bRQUFPC9EG/evImdO3cy/dHR0YGdnR3fcXz58gXHjh3DvXv3mGthYWHBzAzYnMO7d+/i48eP6N27N7Kzs2Fvb4+4uDi0bNkSvr6+Ql1cokhPT8fu3btFnkNxlJSU4MaNGzAwMJAYwn7y5Els374djx8/BgDo6OjA3t4eI0eOFNuXhg0bsg5rt7S0ZNWupKQEPj4+2L59OzO4aNq0Kezt7eHo6Mi8hHbt2oXNmzczs1IVFRXMnTu3XOG3XMLDw/mOiztQBCD2em7duhX//PMPjI2N0bZtW3h5eUFdXR0xMTFQUFDAtGnTMGnSJCbIgg1fvnxBUFAQX38sLCygoqLCDESFuctq1aoFLy8vTJo0idX5KZsWISsri0aNGqF3796YN28e6tatC2NjY+Tl5WHKlClCn78RI0agpKQEJSUlzHN24sQJ5jmeMmUK5OTk8M8//2Dfvn3MwIJLfHw8rKyscO7cOYnvAi4vXrzA48ePmXPDa0DYvFMqE6kboj+FDx8+oEGDBr91wm5WVhbGjx+P+Ph4vhD4uXPnQl5ennGN/CykeQ6Tk5Oxb98+pKSkMNFq586dg7q6OgwMDKCsrMzkqkiL4cOHsz43Z8+e5fubu97AXaPjsmHDBmzcuBGOjo5MNFRsbCy2bduG+fPnY968eQAqJ++EDc2bN0dYWBhatWqF5s2b4/Lly9DR0UF8fDxsbGyYkGA2of+iePXqFTgcDgwNDREWFsY3u61RowYaN27MGGm250cSzZo1w9WrV6Gvr8/6XAijSZMmiIuLEwgnT01NRdeuXSWugQOlA86SkhImR45LYWEhZGVlmRmmMBYtWgQ3NzfUrl2bVdg6W6TumhPFhw8fmBEid/1GFIqKijh16hTk5OQwbNgwvs9CQkJQVFTE5EAApQu6wm7gHj16VLi/ioqK2Lp1K2xsbFCzZk2RRQCvXbvG+re4OUufP39GUlKS0LwmXkULSUjaz9KlS6GsrIyUlBT8888/zGdmZmYCN524pGBJ7gNnZ2eh/RM1I0hJSWFmGLq6uhUqIyLu2OvUqYPx48djwIABiIqKYnzjKSkpOHz4MA4fPgxtbW1kZ2ezNkS5IhJRJfXF1NQUDRs2xOPHjwV0GbW0tHDgwAFMnz6dVWg2L2UNEJe9e/di06ZNMDc3Z7YZGRmhZcuWWLFiBebNmyc072T//v1Ys2YNk3ciapAiIyPD5Jepq6ujevXqfPcWANy/fx/Vq1eHrq6u0HUkHR0dvnUkNqH/bO5BSe8VNufH0dERbdq0wenTp8UGE+jo6ODjx49if4vrLhSFoaEh1NTUEBsbK/AMXLt2jUn5kMTkyZPRo0cPAYGBPXv2ICYmhgkAE8aDBw+YdbWyYeu8lHcw+UsYImHJqtHR0ejatStCQ0OhpaUl9MB413Y8PT2Z/BxeateujSVLlmDEiBF4+/YtbGxsEBsbCxkZGQEFcW5km7Cku+DgYJiamsLV1RXjxo0TehyRkZEICwvDiRMnsHPnTqFtuNFkwhaReeHmTERERGDatGlM33iPHWCfJClqP9zfev/+PSIjI3H69GkB5QBNTU1mEZpNUvDRo0eZtSBeDA0N4e3tDWdnZxQWFmL79u0iE5ljY2Px/v17ODo64sKFC3xJiYMHD4avr6/QqChhSDr29u3bY9WqVbCxseFz1fXq1Qu+vr4ASmWsXF1dsXbtWrRt21bo/cgmEZXNdRAVXbVkyRIEBQVh3bp1WLJkCcaNGwcvLy++vixevJj5/ocPH7By5UrmHJfdb1paGt69e8cEN/DSsWNHvHv3jtmnpLwTbuJjQUEBX0Jr7dq1oaSkxNz3y5cvFzBEjx8/xq5du3Dx4kVW60hLliyBoaEhoqOjoaOjg+joaOTl5WHBggXMIEfUPZifn4/AwEA4OzvjzJkzQs8zF1NTU4nnR05ODnJychJfvM7Ozli2bBmcnZ2hr68voEqhqKjIJLTzXqey7ydra2ssXboU379/Z4JAunXrhq9fvzLraOJIS0tDXFwcXFxcBD7z8PBgZklqamoij6lt27ZIS0tjgoEqA6kbIlHJqgoKCswNW9bNIIzU1FS0atVKYHuLFi2Y6LUlS5agWrVqiIuLQ79+/XDs2DFkZWVhzZo1WL16NQDRi6UlJSUICQmBq6uryBcg7zrAvXv3ynsqhLJ48WIMGjQIrq6ufAuop0+fZv7/3bt3WL16NYYPH86XJBkSEoIlS5aI3Q8vhYWFQqflOTk5jAuGTVLwu3fv0KhRI4HtDRs2ZF5uCxYswLlz52BmZoYuXboI3c+sWbOQkpKCCxcu8C22z58/H7Nnz0ZAQIDwk1YGSceuqqrKRILx0qBBA2bUPGXKFBQWFqJPnz6oXr26gEsqLS2NVSKquL5s3boVW7duhYyMDPbs2cO3KF5SUoLY2FhG0ufIkSO4fPmy0NDsAQMGwMvLC46Ojrh37x6sra1F9qdly5YIDg6Gk5MT3/bg4GDmeYqLi0NYWBjfrKpevXpwcXFhzpuDgwOCgoKwbds2JogkPT0djo6OGDNmDAYPHgxtbW1cvHgRM2fO5PutDh06MKNrFxcXZubg7OwMe3t7ODk5MeuCAJgAJGGh/4sWLUJsbKzIe9Dd3Z0xApMnTxb4nAt3UMDm/Nja2sLb2xvbtm0TCCzhwg1nFhcoVTYKr6ioCPfu3cP69evh5uYGoPSZeP/+PZycnJhBcrVq1TBo0CC+wqXi+PLli9B+Lly4kJnZlsetVhlI3RBt3boV69atg5WVFdTU1ODm5gZNTU38999/zIMoLG+lLA0aNMCLFy8E1L2fP3/OLDBfu3YNQUFBaN26NWRkZNCoUSN069YN8vLyWLVqFfr27QtXV1dYWFgwi6Vnz57lWywFgG3btlXyWRDNq1evEBgYKPDS4nU1jhs3Dm5ubnwP1qRJk9CxY0eEhITAxsZG5H546d69Ow4fPgxXV1dmW3FxMTZt2sRE+zx+/BhRUVFCjT4XNu6DkJAQ7N+/n28xuyxhYWE4ffo0E24LlI7+Nm3axPqhA0SfQy4NGjTA27dvBe6dxMREpr9sHsz4+HhcvnxZ7DqAuL5wZ9EcDgcHDx7km+3WqFEDzZs3x8aNG5k2ycnJAteBd6YdFRWFkydPCixs87J48WImk79r166QkZHBjRs3cO3aNWZGwSbvxMvLC4cPH+aLZFRVVYW7uzsmTJiA8ePHo2bNmkIHaLxuzPbt2zPbGzVqhGPHjgm0Fxb6r62tzRf6L+oe3Lp1K6PAwcY1x+b8xMbGIjY2Fnp6etDT02P6xuXIkSOsBtNldfKAUndsvXr14OXlxRh9Nzc3LFy4kAk00NHRKZcOZJs2bXDs2DEm34jL58+fmXtXVEAM10UvyY0I/GZrRKKSVU1MTDBlyhQmfFYc6urqGDp0KJYuXYqDBw8yD+fTp0+xbNkyZt2IN8mtQYMGePfuHVq1agUdHR3mAeZG3vEm3WlqasLd3R02NjYYM2YM/Pz8YGFhITDiKs9iPneEJUpLjduma9euePr0qVjRxejoaGZGx0uvXr2YGRGb/bi7u2PYsGGIj4/H169f4ezsjEePHiE/Px+hoaEA/pcULM4QCXMfREZGwt3dHXPnzgUAKCgoSAy/VlJSEniogdKoJrZuOUDysZubm8PV1RV79+6FjIwMioqKEBMTAxcXFybMmU2kGptEVHF94b6khw8fjoCAALHiqmxCsxs1aiQQalwWU1NTXL16Fdu2bUNoaCgTQXX16lVGnoabd7J582Zmxn3z5k2+vJN3794JTW799u0bU4usc+fOiI6ORnFxMRMMUFRUhPXr10vUfeOFTeg/m3uQDWzOj5KSEpMAKwo2g2lRaGhoICkpCUDpzNPT0xN169blcxkWFBRg0aJFzKxRHP/99x8mTJiAlJQUJsI3KioKp06dEvAycEVsgdJ1LnHrQryUd41I6lFzbdq0Yfy/PXr0wNy5c2FhYQFFRUVwOBzWayAfP36Eubk5bt++jaZNmwIozQPp2LEjjh07hnr16qFfv35YunQpBgwYAEtLS9SpUwfOzs7YuXMnzp8/j/j4eLRs2RKhoaFo1aoVOnXqBE9PTwwYMACPHz9G37598ebNG/zzzz/IzMyEkZERxo4di2HDhkFBQUHgYUpLS8Pnz5/5fOYKCgpQV1dHbGysxJyJ2NhYnDlzBqtWrYKDg4PQvCZDQ0O0bdsWU6dOFYjg2bhxI/bs2YOkpCRW+wFKlcJ3797NFwJvY2PDnNPIyEisXLlSrK8bKDVqfn5+zIu5Ro0asLe3Z/TUtm/fjkePHsHb21vkNT5w4ACCg4OxY8cOZmby5s0bzJgxA6NHj4aVlZX4G+P/kXTsbdq0wcyZM5mkY1lZWXA4HJibm8PPz09AZUOUmnpkZCQ2bdqEDRs2iAxqYHsdJMEmNPvEiRM4efIk/Pz8yjViLktubi5mzJiBixcvCuSdbNu2DfXr18e4ceOQnp6OzZs3M8eQkJDAuHIDAwPh7+8PJycnqKqqolu3bgBK9d8KCgpw/vx56OjoiBSR5Q16MDAwQJs2bSSmT0i6B4Efi76rCOICpcrO0DgcDjIyMuDp6YnU1FRERUWJDGTJyclB69at8fbtW6xfv55Zcy+bsMtdm7xy5QrWr1/PDH4MDAywYMECZtaVmpqKSZMm4cGDB3zvL319fRw8eLBCAUPikLohsrGxQbt27TBr1iysX78e27Ztw5AhQ3DlyhXo6upixYoVEoUieSNawsPD+ZKrjIyMmBs7KCgI379/x4QJE5CQkABzc3O8f/8e8vLy8PPzg5mZGUaNGoVx48ZhzJgxmDt3LhISEmBra4ugoCAUFBTg8uXL4HA4iI6OxrFjx3DmzBkUFRVh6NChGDt2LPr27QtZWVkEBATgyJEj8PPzYxYQuesIFhYWmDRpEqucCTZ5TYGBgXBwcECfPn2Y83P79m1ERETAx8cHlpaWrPbDBt79iPJ1cykoKBDpPhg7diyuX7+OevXqQVdXl3khc3O9uG4Krsow78NQs2ZNNG/eHLGxseXuc1l4+5yamsoYYAMDA7Rs2ZJpx0ZNXU1NDV+/fkVxcbHIRFQ2fSlvWKyo0Ozu3bvj1atXKC4uZiLWADAvpxMnToj9HYC/LMXz58/5cnJ4je27d+9gb2+PsLAwPmPVr18/+Pn5oXHjxoiKikJWVhYePnzI94xOmzaNub5eXl7w9fVFp06d+ER179y5g6lTp+Lp06c4f/48duzYwScjIyr0X9w9KCr6rnr16qhRowZu3LhRrvNz9+5dkfl3bAKlhCnGczgcZk22ffv2aNmyJW7evMnnjSkuLkZoaCg8PDwwZswYnDhxAvPnz8fSpUuxbNkyvHr1CidOnMCyZcuYGnKSMDExQXFxMXbs2MH3/poxYwaT38kL15NV0QGP1A0Rm2TVoUOHws7Ojm9dBChdsN++fTuj41VePn/+jCdPnkBdXZ3JJShvEuW3b99w8eJFHDt2DJcvX0b9+vXx6NEjGBgY4NChQwI6evfu3YOlpSXu37/PKmfi1atXYo+B+8K+ffs2duzYwZegZmdnx7ht2O5HUmgxm6RgLuJqQ5VdsObCdUGUDcMXBletQBKSjj0wMBCzZs0ScAN++fIFW7ZsgZOTE2bPno34+Hi4u7sLVVMfMWKE2LBXoNS9x+Y6lE3yLqvmzGa9AYBI/TZPT0/IyMiwcp+wHaBwefr0KZ4+fcq8+MW5cIUxY8YMtG7dWmB2v3nzZjx69Ahbt25F48aN0bJlS0YCp6KMHj0a9evXh4+PD1/0XZ8+fVh5YoDS88Mm/87a2hrv37/H+vXrhQZK9e3bV+DZ4ibGamlpoXHjxmKvl4yMDJYsWYIDBw7A29sbAwYMgJqaGqKjo9GiRQvs3r0bkZGROHDgAKvjatq0KS5fviz0/TVo0CBmFr5t2zZs27aNSfZt1qwZZs6cWe4iqFJdIyoqKsLx48eZl46srKxQ/y0boUig9GUsKhx47dq1yMjIYFxMQOk6RVlXCJvFUl5q1KiBLl264OXLl3j06BEjZf/u3Tu+Uulcvn79yjzcbHImhC1gCqNTp05iF6XZ7IdNaDEbX/fHjx/h6OiIM2fOiKwNVZUBH5KO3cvLC1OnThVqiLy8vODk5IQrV67A398f3bt3R7Vq1WBoaIhRo0ahadOm2Lt3L0aMGMFqHYnNdRAWFltYWIhZs2YxSZVsQrNFGeohQ4Yw/8+mLAW3nbjaNFy0tbWZyD6g1D1nYGAAWVlZVnky586dY8RzeTExMcH69evh5+cHFRUVocUGx40bh507d6JevXoiUyy4HDlyRGT03aZNm7B582bs3buX1flhk3/HJlBK3LN19uxZcDgcmJqa4sCBA3wz6xo1akBdXR3NmjXDhg0bmHpCtWvXZoJMeAOQxIVmA6X3jpqamsj3F3dt19XVFfv27cPs2bP5zs3atWuRmZmJFStWiL4AZZCqIapevTpcXV0ZPSZRcIUiy47weIUifXx84OrqCi0tLYFQVe7/6+npoWXLlujZsyfzj9cw8SIpiTI/Px9nzpxBcHAwrl27hhYtWsDc3Jypj9SnTx9mgZe7qBgfH4+5c+cykWJsciaA8vmxRa1fsNkPmxBvoFSBYdeuXXxZ9tOmTWOkS5YvX46MjAyxtaGqGnHHXtZNwuXevXvMAy9MTV1LSwudO3fG7NmzBb77I9dBGFw159GjR2Pq1KmsQrNFwTv4Wrp0KVavXs3nbTAyMoK2tja2b98Oc3NzVrVpAOHG6tChQxg9ejR2794tNE+GC3egU6tWLcTGxgqsscXGxjLyUFOmTIGnpydycnL4VBEaNmzInAc2wSyiou+MjIyYXCU254dN/h2bQClusruwc6OgoIAWLVogMTER6urqIq+3mpoaMjIyoK6uDi0tLVy9ehWGhoYYM2YMM5Mum3smDA8PDzg5OcHLy4vv/bVkyRJ4eHgAKF3D9fHxEXpu5s6d+/sYIqB0JJ+QkCB2pChJtBIoXfz28vKCra2tyP3cuXMHMTExiImJgZubG968ecMYpl69emH06NGskiitrKxw+fJl1KlTByNHjoSrqyvjz+bi4+ODGTNmYNCgQXw+8/79+2PLli0AxOdMcJUZ2GSRs1m/YLMfNiHeN27cgLm5ORo3bsyMgrj5I8ePH0eXLl1w4cIFBAQECCw8u7i4MFN4NtUx2Yzc2CDq2HmTgg0NDfl+q7i4mBFsBf6npq6urs6oqXfs2JFPTb2yroMoeNWc2YRmf/v2TeLCNRtvw+rVq+Hk5IT58+dDTU0NO3bsQNOmTfn0BUUZq7p16zJ9lqRWDZQKxy5cuBAJCQlo3749M6M+fPgw/vvvPwClYe7FxcXQ09ODiooK30x28ODBjASPJNhE37E5P2zy77S1tfH06VNoaGigbdu22Lt3L1RVVeHv7888b7wSTtx7k/dvWVlZGBsbY+7cuTh69KhQOarhw4cjMjISnTt3hr29PaZNm4b9+/fj7du3zIBJmOAt8L+ZEjeit7CwEIMHD2begyUlJahWrRpsbW2ZZ0/UuSnrkZKE1A3R5MmT4eLigtevX8PQ0FDAPWJoaIgBAwbgzp07fEKR3PBubib8x48fJc6stLS0oKWlxURbPX78GJs3b0ZAQAAOHDiA0aNHs0qirFGjBvbv34/+/fuLrFvUqFEjBAcH49mzZ3zioLw+czZuQDZZ5GyqwbLZD5sQbxcXF4wePRobN27ku0HnzZsHZ2dnXLp0SWRtqJ49ezLHyEaaqOyiPDfB78yZM1iwYIHE73MRdezW1tYwMTHB1q1b4ezszLfYz83b4eYwsVFTr6zrwEbNmU1o9qpVq/gWrlesWMG3cA2w8zaIq00zduxYODo6sjJWbNyS8+fPh4aGBnbs2MHcK61bt8bWrVuZPnDz+UQlj0ri+PHjGD16NBYuXMjU+nJ2dsbYsWNhYmLCRN+xPT9s8u/s7e0ZHbhFixbB3Nwcx44dYwKlgNIBnaurKxYsWMD37vH29saSJUsgKyuLefPmYfDgwRgyZIhIOSouI0aMgIqKCuLi4tCqVSvGJXvmzBmhz9/atWtx6dIlie9RLtxnoGzayu7duxnPEFukHqxQWdFc8+bNQ5s2bcQqBpeUlODu3buIjo5GTEwM4uLioKioiB49eqBXr16wtLREs2bNBJIogdK8CTMzs/KXwIVo+XgTExMcPHhQYEqfn5+PCRMm4OzZs2jRogVCQkKgr6+P5s2b4+rVq9DW1kZMTAyTRa6vr8+sX3DDiLW0tHDs2DEEBATg1KlTrPbDJrS4adOmiI6O5lsHAEpnHb1790ZGRgaGDRuGYcOGYebMmVBTU0NMTAw0NTUxf/58pKWlITg4uNznkJcDBw4gKipKaF0fYUg69rVr16Jr164CoejiEKamXlnXgY2aM5vQbAMDA4kL11euXMHEiROhrq4u1NswcOBA6Ojo4PTp09DV1UW3bt3g7OyM4cOHIzExEcOGDcPr16+hqqqKa9euQVNTE5qamjh//jz09fWRnJyMsWPH4v79+0y/fobWIy9FRUV4+vQp5OTk+AZ+586dw+rVq/Hs2TNkZWUJ/W7Z6Ds25+fRo0cYNmwY2rZti2vXrmHw4MF8+XfCBnbCAqWMjIywYsUKgVIRERERcHNzQ2RkJDp27Ij3798jJSWF79lKSEjAuHHj8OjRI4nnp2nTpkzSPnfgn56eDltbWzx9+hRPnjwR+V3eiM7i4mIEBQWhadOmfOcmIyMDFhYWTFFANkh9RsRmug5IjuZSVVXFmjVrEBcXJ7SukaOjIzQ0NCAvL49BgwZh1KhR8Pb2FhilsU2ilJSICkiWj2dTmItNFjmb9Qs2++EqM3BruvDCHRTUq1cPL1++FDBEL1++RP369QGwqw31I/Tu3VsgK1wcko6dd5FY3NpO2W1lt1fWdWAjD7V+/Xq8evUK2trafKHZXLgyN8IWrvv3788stLPxNrCpTSOskJq+vj5fITW2Wo8/wqNHjzB27FjGdTR06FBs3LgRU6dORWJiIqysrBAUFCTy+2UHxmzOj66uLmJjY7F7927Iy8vj69evMDMz48u/K4uwQKnHjx8LdYs3a9aMWa9+8+aN0MJ6vHJUmZmZiIuLQ3Z2tsC70sbGBleuXMH06dPRq1cv+Pv7IyUlBfPmzUPnzp0FIvc+fPjA987lFg3l3sPcpF7u+VZWVoaysrJYYyYMqRuiyormOnDgAGrXro24uDjExcUJtHF0dIS+vj4SEhIQHx+P2rVrM/94FzwXLVqEJUuWCCRROjs7M6OBW7duwcLCAjVr1hRZvEuUfLy7uzuePHnCRFhJKszFxo/NZv2CzX7YDApGjRqFWbNmwd3dndGIu3HjBtzd3ZkiaVyxWh8fH7Ro0QJRUVFo164dLl26xMwe2ER9ieL48ePlUlaQdOz5+flYtGiR2LUdNurSGhoalXIdeBE1m2bj2hS1cH3r1i2+EgCqqqp8bqWyrF69mlnnWbx4MT59+oQzZ84wtWkAdsaKjdYj24guUW0+f/6MHj16wMvLC8HBwThx4gQeP34Mc3NzHD58GHXr1oWDg4PEcwf8LxpQ3Pl59eoVUyp75MiRYgdIJ06cEBnVe+TIEejo6GD9+vXw8fFh1pa+fv0qEAknbDmAK0d19OhRzJ49GxwORyCvSkZGBjY2Nvjnn38QHh7OJLDKyMhg5cqVsLe359ufg4MDo6TAHTQIyxesDKTmmmOjVQSUuoO6deuG9u3bS4zmYsOXL18QFxfHBC0kJCRAS0sLvXr1gpeXF5MEKCqJEij1x9aqVQvPnz8XmYj6zz//CCTbAqV+YFtbW75AiLLwFuZiU0DO19cX1apVg729PSIjIzFu3Dh8//6dWb+wtbWtUDE/YXz79g0uLi7Yu3cvioqKAJSGoU+dOhXu7u5ia5nwMmHCBLFRX5aWlkIDGrgl3r29vcWKVvIi6diDg4Ml5gh1795dorp0rVq1YGdnh6VLl/7wdaiMYnXcarMLFy7E6dOnMW3aNKioqDAL17wKzBVxl339+hUpKSnQ1dVlVUhNW1sbQUFBaN++PdTV1REeHo5WrVohNDQU69atw5UrVwRyscquC9rZ2Ylt8/nzZ5w7dw6GhobIzc1FixYtsHnzZj4VjrLrF2FhYfj333+ZqDwuR48eFXt+7t+/jxUrVjDlyKtXrw4/Pz+BZx4oXT/08/NDr169hN7v27Ztw+3btzFu3DiUlJRAT08PMjIyePDgAWRlZXH06FF07NgRY8aMwcOHD3Hx4kV069YN4eHhyMjIwMyZMzFhwgQcPHgQ48ePh5OTk9g1tNu3b8PGxgY1atRAeno6hg0bhg0bNqBu3boASuXBmjVrhtmzZ0NZWVmgv2U9Ij+K1AwRN4tYlOQ98L/ZjoqKChMiXVlkZmYiOjoaly5dwokTJ1BSUsKUk5DExo0bMXnyZKxdu1ZkImqTJk1w/fp1gTDU58+f499//8WtW7dYFeYShqQCcsLWL0TtJzo6GsbGxpCTk2Mli8/l8+fPSElJAYfDgZaWloA7MysrC0ePHkVqaiqWLl0KJSUl3LhxA02bNoWmpibU1dUlRn2VvRbctZKePXuidevWYvsqCd5zyGZt59ChQxLVpa2trVG7dm0EBgYCKN914L2elVWMrSy3bt0SWLj+EXdZUlISjIyMWI+OuYM2buTYzp078e+//yI1NRX//vsv3r59K/K7bNYFDxw4gNmzZ+PJkydMKoGqqioiIyMlivRy11rKIu78lJSUYPDgwdi0aRNq1aoFd3d3XLp0SWiJF21tbaxfv14gKb8sBQUFCAoK4ksKtrCwYGbE379/FytHpaWlhcjISLESPGvXrsX69ethY2OD5cuX4+XLl7C1tUV2djZ27NiB7t27Q1VVFdHR0WLrb7HJ1WKL1FxzbNeGAHbRXIDkpLtTp04hOjoa0dHRePbsGZSVldG9e3esXbuWGYmyydbftWsXEyYuKhFVnHy8trY2NDQ0WKn/CkNSSWlh6xei9jN58mQ8efIEjRs3ZiWLz0VBQUHkyzUhIQGmpqbQ0NDAo0ePMGvWLCgpKSE8PBzPnz+Hv78/q6gvtsoJFYH3HLJZ22GjLu3i4sKX1Fqe68ALm2J1bEKzy9K5c2eBsvRs3GXloaxIJu/CO5sQZlGwWRfkipvyqiLIyspWOLoOEH9+0tLS4ObmxvTdw8MD+/fvR25urkAAUklJiYBKgTBq164tVoZHTk4Ou3btwrJly4TKUY0ZMwahoaGws7MTuY89e/bgyJEj6NevH4DS63L58mV4eHjAzMwMWVlZ6NatGx4/fizWEJV1jxcVFeH+/ft4/fq1RBHYskjNELFVDABKE9hcXFyQkZEhMpqLTdKdk5MTevToAXt7+x8aVbNJRBUnH+/m5ibWNcl94ERV1eSFO+qQpCrx9etX+Pv7Izo6mm8Rs3379hg7dizCwsLEGkYHBwdWvnVfX18mJ2rp0qV8heb69++PQ4cOASh1VaxevVpo1Fd2djb27NkDe3t7gXOQl5eHHTt2wMbGRuw6kaTRGi9s1thEqUsfPnwYb9++xdatW/HhwwemHhEvXEVjNkKRR44cYVWsTlRodmBgIMzNzSXOboHSGS6bjH82sBHJZBPCLAo264LHjx8HAL6csE+fPqFnz54C3gO2OWjizs/YsWP5xEdr164NBQUFoYbI2toaR48elZjMzTbZWVNTExkZGWjfvj2ffNaqVaswYcIEREZGChUldnJywrVr1/i8MECpW3H58uWM6KmPjw9mz56N1NRUoe/cHj16iMzVWrZsWbk156QerMAGNtFcbPIYuCM1cbAZabIp3iVOPr5Pnz5i3ZLc7WwkYwB2qhKzZ89GaGgohg4dCl1d3XLLtHOl/LmI8qsDpbNdYaXSZ86cifT0dGbtR1TUV1ZWFvr16yfUENevXx/Pnz/Htm3bRJYcB9hl1nPp1auXxBwhIyMjzJ07V0Bdevfu3ZCTk8POnTvx5csXcDgcgeq8OTk5ACDUuAiDTTG2kydPYuPGjRgwYABcXFwwbNgwtGjRArt27cLevXuFViflhfvcsMn4Z8OsWbNQt25dJCQkCIhkzpo1C2fPnuUrYW5oaIh79+4JhDBLWheU1GbChAmVFgYOiFdEAASDjTgcDh4+fMhECgKlx5qXl4fg4GBEREQIjepdu3ZtuZOdLSwsEB0dzTfA2bt3L65cuQIlJSWkpKQIPOdOTk5QUlJCYWEhQkNDkZKSAmtrazRo0AApKSlMPaLnz58jKSkJV69eFTgnkoIVpkyZgiFDhpRLQeW3MERs3Hiiku7s7e0xffp0ODo6SnSFKSoqskoCZKtHZ2hoKLRkOFu3JNtZIxtViQsXLuDw4cMSteISEhLg5+fHGO3WrVtj5syZfAu3QKlffcuWLUJH+TVr1uR7ELl06NABWVlZEiO+du/ezSQtCmPixIlwcnISa4gqqmVnZGSEmzdvCqzt+Pj4wN7eni+JWZi6dFFREePyqChsirGJCs2+e/cuevXqxXrEL85d1rBhQ7Ez92fPnjH/f+vWLVy+fJnPFamuro7Vq1eLTJAUFsJsYmLC9/IUti5Y9v4p79phQkIC33EJMx5A6fMr7vwApVGkZQeUvANI7kv70aNHjGtOVGgzm2RnXoQNZNetWwcPDw+x3osXL15gxIgRKCgoQF5eHszMzNCgQQPs3r0beXl58PHxwbx582BkZMToQ5Zn4MrV2ywPv4UhYvNCFpXHMGzYMCZUVUtLS+gJ5Q1JFDXS1NHRQXh4OGsZdaB0NBUcHMznMzc3Ny+XW5INbFQlGjduLDAdL0tQUBDs7e3Ru3dvZop++/Zt9O/fH9u2bWOdLT106FB4enpi//79zLaXL18iOTkZEyZMkLj2ww37FoWmpqZEFWteeIuJ8cItJtazZ0+MGjWKcXFw13a+ffuGwMBAjB8/Ho0bN8bx48fFqktz1yjYUFhYiBcvXkBGRgYtWrTgC6dmU4yNbWi2JMS5ywoLC9GvXz+JAUXc/kgSyQRKxTvLuoe57Nu3j9UourxrhwsWLGACZgAI1bwr633gvg/EnZ/169ezViEQJmRbFjYl0CVRXFyMoUOHim2zZMkS9OvXD97e3nxViY2NjRkD9ubNGwQHB4t9DsuWK+EqgFy5ckWkjJAofmlDVJ4Qb1F5DBoaGsyLn418vqiRpp+fH168eMGqkmRsbCyT6fzlyxdmuhsQEAAPDw8cPXqUdQE0Xv0pXnjzV4yMjHDlyhWxob2urq5YsWIF/Pz8RFb+XLlyJZYtWyYgn+Pt7Q0PDw/WhmjlypUYM2YMWrVqhc+fP8PY2BhZWVno2rUrZsyYgbVr14pd/6lWrRrS09P51pd4efPmjdiowrIEBgZi+fLlAoaosLAQR44cQWBgIAYMGCBQbOzTp09wcHDgc4mUVZfmhcPhYPfu3fD398fLly9x/fp1aGpqYuPGjdDU1MTIkSPx/ft3rFixArt27cK3b9/A4XAgLy8PW1tbuLi4MD59UbNpLqI0xdLT0zFr1iwAYBUFKc5dxpW/YQMbkcxly5Zh586d6Nq1K5SVlYVew3bt2iE8PFzAtZqbmwsjI6NyBTlxCQoKYgJmgPIFSrFxJ7JB0mDI19eXVbIzLxs3bhS4ZydMmICgoCABty4vcXFxuHLlisD55w5ugFLR5oSEBLGGqGy1Vu7MlKsNWh5+aUMkTq2XC+8akbCku7Zt2zJJd2xKGIgaaerp6eHNmzesEgkBYO7cuejWrRt8fX2Z6LCCggI4Ojpi7ty5iIiIYLUfHR0dBAcHo2nTpoxL8O7du8jMzMSwYcNw48YNJCUlITIyUqyqRN++fbFv3z60atUKTZo0EWiTmJiInJwcjBw5UqAPZmZmWLduHav+AqWu0XPnzuHatWu4d+8eU+m1T58+8PDwwKtXr8Su/zRo0ADnzp1D165dhe7/zJkzAjI4wvjw4QM4HA44HA5yc3P5jplbTExZWRkZGRlCjX1aWhpfPyUlJPr5+WHLli2YM2cO3N3dmc+aNWuGnTt3YuTIkXBzc8Px48fh7e3NF5q9YsUKlJSUMC9tLtnZ2di9ezcKCgowdOhQprKpm5sb04ZXU8zFxYUZ1ZYnCpILr7tM0ou2bGKpJJHMwMBA7Nu3T2ytKW4hv7K0aNECHA6H1dpf2eMq+/4oj0ciJCQEgwcPZu4dYe5ENkgaDPn6+pY72Zm3MCCXL1++4MCBAwgLCxO5FgVAqKLL69evmft9wIABcHZ2RnJystBgBVNTU1azPLb80oaI7ciFWziMm5OioKDALGyKQpSUiyT1WrZugUePHmH79u18Icq1a9fGokWLWEciAYC8vDwsLS0FcmqWLVsGGRkZREZGQl1dHYWFhWJVJezt7fHo0SPMmDGDybEoS69evRATEyMQshkTE4O2bduy8qu3bdsWzZs3R0xMDIyMjAR0sy5cuCA2V2vixIm4du0atm3bBhUVFdja2jIjt6KiIuzcuRM7duzA7t27Re6DC9cVKyMjI9SolZSUMP7vYcOG8Y0QS0pKkJaWxrgoJSUkAqULxZs3b8bgwYOZwQ9QOsrnaoAdO3YMW7du5XPptGjRAo0aNYK1tTXy8/MZdfaCggL07dsXGRkZqFWrFrZt24YjR45gwIABAr/NDc12dHRktrFJDyhvNVi2nwlDQUFB5BoO7+wtNDSUbwBQUlKCPn364OHDh/Dy8sK7d++wevVqDB8+nK8OTkhISKWXGLGxsUGtWrVgZmaGsWPHihwciYLtYAiAWBHWbt26SbxWQGlAFneQVnYtinvP9uvXD76+vnwBRfn5+VizZg1zX3K9IsL04soOYr58+cLM2Fq0aCE0gEkSUhc9rSyaNGmCmzdv8vk8y8JGpr8swpIAuaNh7sgvMzMToaGh0NHRYW7UXr16YeXKlUztIS4RERFYtmyZyNojZWnRogWuXLnCV7YaKF0oHjhwIFJSUvDw4UMMHjxY7LqJiooKzpw5IzaBdNeuXfD09ISpqSmf+u/Zs2eZyC9xcG/Q9u3bY//+/UJnLaqqqoiLixPpdktLS8O///4LOzs7eHt7o06dOkxARGpqKgoKCjB79my+om2iiImJEVtM7MyZM6hTpw68vLzg6OjIN2jgqm+bmpqiRo0arBISmzZtips3b6J58+Z8SZLPnj1Dr1698PbtW7GisV27dsWRI0cwePBgAKWGzd3dHVFRUVBXV4epqSlycnJYDYbYztwrqxosG/z9/ZGQkIBNmzYJjLDF5cbJycmhefPm8PDwwJAhQzBu3DgYGxsLzPj279+PkJAQsVpy5eXjx484ffo0jh07hujoaKipqcHCwgJjx45lpS4grPw3L9zKqgsXLsTnz58hLy/PNyDiJjuzycsRVsJbGG/fvmX2l5qaCgMDA7x48QLKyso4f/48XxlycXz9+hVubm7Yt28fn5t58uTJcHd3L9da5S89IyoP//zzD1JSUsQaIjYy/WURlgQ4ZswY9O/fHzNmzMCnT5/Qt29fFBQU4NOnT/Dy8sLo0aPh7OwMJycnLFq0iO+lvn79ej63iiS4M4+yhujRo0eMy0FOTk5iVIuamppE+R3uiGvfvn0C4b+87g0ZGRmxs9X//vsP7u7u2Llzp9B8BTbrPy4uLhg6dCiCgoIYBYcePXrAwsJCoPaTKLiu2MTERKipqQmUf+YqrDdv3hyjR48WKGfOC5uERE1NTSQmJgq4fi5dusSsO/7zzz/YsWMH1q9fz9dm+/btkJGR4ZsxREREYMSIEcz+uIZVkrQR74hVUl4Km2qwbPn69SuCgoL4Ciaam5sz53Xy5MkIDQ2Fnp4eWrVqxWeMevbsibNnz8LAwAARERFiXXDR0dFCk2179eolMCNKTU3Fo0ePICMjAx0dHVa5XLzUrVsXEydOxMSJE5GRkYFjx47h2LFj8Pb2ZpRRxMG2smpxcTHjSdDV1WXacNuX1w2Wk5ODlJQUtG3bVuC+btasGaKjo3Hs2DHGdW5tbQ0LC4tyzWbmz5+P8PBwbNmyhXmWbt68iRUrVuDTp098lXsl8ccYosWLF2PZsmVMCGTZrH1FRUWx5Z63bNki1ogBpS/Rpk2bIiEhgVkDOHv2LOrWrYvExEQoKytj4cKFzAudw+HAxsZGoNiVpaUla1mU8ePHY9asWXjx4gVfobBNmzYxkT5c+aMNGzaIVJVYvXo13NzcsGHDBpHZ0hVVeijL1q1b8fLlS6GFy75//856/adjx46sjY44mjdvLraqLG/EVGFhIU6ePImCggL069ePOVdsEhIdHR2xaNEiJp/o5s2bOHLkCLZs2cK4Qdzd3TFmzBiEh4ejc+fOkJGRwa1bt5CRkQEFBQW+9ZE7d+7wLTonJCSgW7duYqVweKloEb6y1WDZ8OjRI5ibmyM/P58Jed+/fz/WrFmD48ePQ0dHB/PmzcP169fRv39/oe7h79+/o3HjxsjJyRFriBo2bIjTp08LSB2dPn2aGfjk5+dj1qxZOHPmDJ+uo6mpKXx8fATWatjQtGlT2NraQl1dHevXr2cVTMU7GBJXWbVatWpMpKYwPDw8MGnSJInvqI8fP8LR0RFnzpxh3hWamppMGDb3/q1Vq5bYFAlheYC8ODo64vTp0zh48CDfUoOmpiYaN24MKyurv9MQcaNbJk2axHexeUOzxUm52NjYSAxVBcAET2RlZaFNmzYIDw/H8OHDIScnh927d2PGjBlMhndl4OHhAWVlZWzbto0JI23SpAlmz57NREdVr14dDx48gJycnEhVCWtra3z9+hWdOnWCvLy8gGuEbd4JG8S5hR4/flwp6z/lQVRV2Y0bN8LU1JTJCSkqKsLgwYOZMgy1a9fGyZMn0blzZ1YJiRMnTkRxcTEjhGlnZwcVFRV4enoyOW49evTA7du34e/vz5QVMDMzw7Rp0zB58mScO3cOc+fOxf379/HmzRs+Mdq0tDQ0btwY379/h62tLVxdXcVGNZU3L4UX3mqwbFi8eDHatm2LHTt2MOs7+fn5sLW1xZIlS3DixAmcOnVK4MVVlpcvX0qc3S9duhQODg6Ijo5mruft27cREREBHx8fpj/Jyck4e/YsM+i5ceMG5s+fjyVLlkh80ZYlKioKwcHBzFrW8OHDRXpRhNG8eXMkJydj3759QiurtmvXTqwnISAgABs3bkTPnj1hZWUFExMToR6O5cuXIyMjA5GRkTA2Nma2Dx48GEuXLuXbJgphEZtFRUXMWmWjRo3g6OgIBQUFkWUryuOWA/6gNaKydTR4ycrKgpmZGXr16gVPT0/06tULI0eOhK6uLlavXg1fX1/4+Pjg8uXLYn+jpKQEmZmZGDVqFFRVVREeHg4DAwPs27cPvXr1wr179zBy5Eg8f/68sg8PQOmDDQhK/xgZGWHEiBGMqkRMTAyfqoSjo6OAYnFZuLOCc+fOwdfXly/3aebMmeXWjhLHypUrf3j9pzwMHDgQ+vr6AlVlW7ZsCSUlJabGypEjRzBv3jycOnUKrVu3hr29PWRkZHDkyBGBtRReZGRkcPLkSSYirFmzZsjJyUFJSQlfeC1vyYAePXpAT0+Pbz/nzp3DlClT0KlTJzx9+hQdO3bkSyR2c3PDq1evsHfvXmhoaEgUt2RThE9cNdjevXtj165drM5xs2bNEBYWJnBMycnJGDhwIN68eQMDAwMEBQXxuZ7KwlUEX7lypdjfu337Nnbs2IHHjx8zeVZ2dnaMG7xFixY4dOiQQLrFtWvXMHHiRKHh0KL6c+LECbx79w79+vXD2LFjMXToULFuXGGEhYVh/PjxGDBgAC5fvoybN29CU1MTPj4+uH79Og4fPozu3bvj5cuX+P79u4AngcPhwM3NDQEBAQgNDUXdunUxZswYTJo0iU/zUV9fHwEBAejQoQPfOmVKSgrat28vVvUfEK+akJWVBQcHB8YQrl+/HsnJydi2bRvj0vvy5QscHR2hq6vLlHZnw28/I5o4cSK2bdvGTIFv3LghoL/UsGFDpgKrOCkXNmGdmpqamDNnDlavXg19fX2oq6szkiLcaqlcvn37hgcPHghN3mObCMeLKO05SaWc7ezscPnyZYkjaB8fH6xcuRLjxo1jDNOtW7cwffp0LFu2jJmBsSU9PZ3vuKtVqwYVFZVKWf8pD0lJSdi2bZuAGOb379/5AjzCwsJgYmLCjKAXLlzIlA9g46N3dXVlrmvZEe21a9cwZswYsSUDhg8fjmPHjuHixYsYPHiwgFJGrVq1GFeZiYkJzp49K/aasMlLKTvy5eaCTJgwoVwq3/Ly8kzOHS/5+fnMs7hkyRKsXr0a27ZtE6lF9vnzZwQHByM8PByGhoYCqu7cSL1OnTqJDbzhlebhRVFRUahmoCji4uIwb948jB49WqLYsDhWrVqFVatWwcbGhm99lFvGBZAcYDJkyBAMGTIE7969Q2BgIA4fPoydO3fC0NAQVlZWGD16NHJzc4UeN1eSTFVVFRMmTMDIkSPLHd2mrKwMZ2dnTJkyBSYmJrh16xZiY2Ohp6fHGMMHDx6guLgYBQUFfHqPkpS4f/sZUcOGDfH48WNm5Kmuri6gv6SoqMgoTPPCVqa/LN+/f0dAQAAaNWqEvn37Mg9VaGgo6tevz9QJsbOzY0QqeSlPYSk2BeTYlHJmM4LW0dHB0qVLhUYjrV69WqJW3927d+Hq6spE7qipqTEvXq6L9Pz580wuTFXRunVrbNu2TSDsuWnTpqhduzYzg+3YsSNmzpyJadOmASg9t507d2aS/CRhamoKGxsboS+UoUOHom7duqxKBrDB09MT27ZtQ48ePdC+fXuBF7ajoyOMjY2Z2ayNjQ3ev3+P+fPn48CBA7h//z6rTH1JXLt2DV26dMGsWbNw9+5dbN68mXGX3bx5E/PmzUOHDh2wbds2ptZXSUkJ1NTUhFaVlTTz5I0Ke/v2rdCcLkNDQ5iZmaFu3brYsWMHc24KCgpgb2+Pjx8/4tSpUz987OVBVVUVsbGx0NDQ4JuppKamomvXrozbvbzExcUxpUo4HA7atWuHYcOGYebMmXy/M3/+fLx48QLDhg3DwYMHkZqaipEjR8LKyqpcg7+EhASYmJggLS0NM2fOZP09SZJbv/2MqOyLWdIaDy9sZfrLIicnJ1Tqhxt2C5SOpgcPHoz//vuv3FpNvDg6OootIAewq47JZgRdUFAgtEBer169WGXZ79y5U+Blf/DgQaiqqoLD4WDPnj3w9/evckMkqqpscXExEwmXmpqKlJQUPsHMN2/eMDMbNrVXJk+eDBcXF7x+/VpgNJ+UlITQ0FBWJQPYcPjwYTRo0ADJyckCxoybOyYuL2Xv3r0ASmftJSUlAj79wsJCyMrKSoy0NDExwePHj+Hp6YkZM2bA2NiYT4vP2NgYa9asAcAupJzNzDMxMRF2dnbMGlvZY3///j1WrVoFCwsLZrQuIyOD+/fvo3bt2hLXcNkol3NhGybfoEEDvH37ViDYgFtZtSJwOBx8/PgReXl5+Pr1K+rVqwdXV1eMHj0ajx49QlFREXx9ffHo0SNGPsjQ0BDTp09HYmIiAgICYG5ujiZNmmDSpEmYMWMG4zUoew64Llt/f39m3bmieo7C+O1nRGVnO8KKXCkqKmLWrFkS69+Ik8UoS2hoKDZt2sREYeno6GDu3LmMa0Zcsa3ywKaAHJvqmGxG0La2ttDT0xNwyWzatAnJyckS1ws6deoEHx8f5kYtew7i4+NhZWWF+/fvV+BMVBxRVWW7deuGW7duwcTEBPHx8WjYsCFCQ0OZ761btw537tzBkSNHBEZ/ZWuv+Pr6inXdcDgcPH36lG9WXln3SHkoW4Rv/Pjx6NGjB18iLFD6komJiZG4tlj2+Xv+/DljIHR1dcXWsxGHOC2+vn37omHDhli0aJHQwRnXxf7lyxcEBQXx9YdNiDJbF1x5PBtubm64fv069u7dK7SyqpOTE+saU6mpqQgICEBgYCAyMzNhZGSESZMmYdiwYZCTk0NycjJ8fHyYekXt2rXDnDlzhHp9srKyMG3aNFy7dg3Pnz9njr3sOeCWwOjduzc8PDzQtGlT5rO7d+8iJSUFgwcPRu3atVFQUCA0IErsufwTDNHJkyeZUeWwYcMECm317dsX2traEk8MW1fFgQMHsGDBAlhYWDCj++vXr+P48ePYsGEDJk2ahGnTpmHQoEGs9dlE0b59exw+fFhgEbi8iJPE4eYFcY1Vp06d+KKRbt26BQcHBz5DXvbFBZQuWMfFxTEvgt27d2PMmDFMqGxaWho6duyIrKysHzqWiiKsquzBgwdx8eJFNGnSBE5OTmjSpAnTfsGCBejbt69YdxG39sqSJUvEJhS3a9eO7z4FhN+rFZGPycrKQqNGjQTypNigpaWFc+fO8a1tAsDDhw9hYmLCp7AtDEVFRTx9+pR1EqQk2GjxqaioICoqSmzV1WvXrqFr164Cz3xRURHi4uIqtVQEGyRVVq1WrRrc3Nz4lP+XLVvGKP8vWrQI9erVQ0BAAGJjY6GqqgpLS0tMnDixQl6dqKgoBAQEICQkBAYGBpg4cSIsLS3L5bnJysrC+PHjER8fzxcqPnfuXMjLy8PLy4v1vv4IQyRJj07YaPRH6NChA+zt7QUWk3fs2IGdO3fizp07yMvLg62tLbS0tKCnpydQoEpUDkdZTpw4gZMnT/60AnK8sNFvA0QntLZo0QKHDx8WmQh5/fp1WFpaso5YqiwyMzNRVFTEpwINlAZTyMnJiZQ8YsOzZ88wZMgQVi9strqJbPj+/TtWrlyJPXv24MuXL7hz5w40NTXh5uYGdXV12NjYoLCwENu3bxepjxcbG4tmzZohMjJSQHrn8ePHMDIykrg+pqioiFGjRkkM1/X19WU14l+6dCmOHz8ONzc3AS0+CwsLeHh4YMCAAXB3dxdrTMquHfP+RqtWrVif58omJSWFSSLlrawKlD5/3t7eGDBgANTU1BAdHY0WLVpg9+7dcHJyQvXq1TFkyBBMmjQJ/fr1E2o0xB13y5YtsWTJEhw+fBhfv37F2LFjMXHiRLEGXRw2NjYoKCiAn58f/vnnH2Z2HxERgUWLFuHmzZus9/XbrxGx0aOryChTHK9fvxaq9zVw4EAm/DQsLAyRkZG4dOkSFBQUBIrVsTVE69evr/QCcqJG0Nz8mYrSrl07hISEiDREp0+fZm3sKhM7OzuMHDlSIAjj6tWrOHnyJE6ePMlsGzNmDLZs2cLnehDH06dPUVRUJDG5MTAwUGDW8SN4eXnh4sWL2LFjB6ZPn85s79ChAzZv3gwbGxssWLAA586dg5mZGbM2VpY2bdrg2LFjAmW4g4ODWc/C379/L3EtCRBdVZa31pc4Lb7Zs2fDw8MDLi4ucHNzg7Ozs9AqpIqKikxwjLC+SnLRl4Vt1VQ2tGjRQmTkqijlf24NrOTkZIlCtKIGOl+/fgWHw8HBgwdhaWmJQYMGoVq1avj06RNz73KlkdisWTk6OiIyMhKnT58WWOPU1NTE69evJe6Dl9/eELEJuS5PAAMb1NTUEB4eLuD/DgsLY6bJLi4umD59OhYvXlzuG58XcYuh5Skgx2YE/aPY2NjA2toaGhoamDZtGmPoiouLsWPHDvj7+zOL5FVJfHy8UPXw7t27w9XVlW9bbGys0Lo64mqvfP78mVXdnsochXNf2D179uQbUOjr6zOzs5CQEOzfv19A75CX//77DxMmTEBKSgoTqBIVFYVTp04xJc4lsXPnTlbeBja1vvLz84W+qFu0aMG8mM3MzAAAI0eOFEhe53A4GDJkCGRkZGBra8tnIEtKSvDgwQNGjoYNFVWnAErvGTc3N0bsWBxr164VW2OqXr16AkaIVwXkzZs3aNiwIWRkZLBnzx6+d05JSQmz7PD69WusXbuWeR5471nu/0t6p3KDYQoLC4UOQHJycsqdZ/XbGyI2ODk5/ZAxKMusWbOwaNEiJCYm8kVhHT16lMlzyMvLw9SpU3/4d8UJXJangJyoEfSDBw8QEREBGxubH1JiBkrzYObMmYNFixZh5cqVTIQQb7JqZSbGsqW4uFho7khhYaFISZWyiKu9YmRkVK76SJUB94VVlqKiIkYmSEFBQcAdWZbBgwfjyJEjWL9+PXOvGRgYIDAwkFEeF0d51hTEjfi5SczitPi4EY7ihD03bdqEhg0bgsPhoEGDBnwuwxo1aqBbt24Stfp4+RF1igcPHjDux7L3Dy/ccyhO+b9jx45wdnZmyoSUVQEBShVXuLMe3sEJV8D30KFDEvUSAfZlMrp3747Dhw/zDeaKi4uxadMmAdV9SfwVhqi8FR0lMWXKFDRq1Ai+vr7MQ6Gjo4O9e/cytVZMTEwQEREh1lCIIywsDL1792ZccR8/fuTTxyosLERJSQnrAnKiRtDv37/HixcvALB7WCTh4uICY2NjBAcHM/vt0aMHzM3NxUb+/Uw6duyIPXv2CJQG8ff35yv7DkDA/cmlMmuvVAa6urpMXgovJ0+eZKq4zp49G76+vvD29hYbyDBgwAChrmY2lMfbwKaqrDgtvuDgYADi64r17NkTJSUlKCgogI+Pzw8PBH+kairvPcPm/ilbY0pVVZWJfl25ciWf+PKxY8fw5MkTXLx4kU8F5NOnTwgICKhQSkB5cXd3x7BhwxAfH4+vX7/C2dkZjx49Qn5+Pl/0KRv+CkP0MzAxMRE7utfU1MTKlSsRGxsrslidOMzNzfkWHfX19fkSdfPz8/HlyxfWAqKiRtBbt25lXkKV9bKVlPVe1bi4uMDU1BT3799nynlHR0fj3r17AomN169fl0IPy4+TkxPs7OyQnp6O4uJinDp1Ck+ePEFgYCA6derE5D3FxsbiypUr0NXV5bsHs7OzcfHiRbEDneDgYLGuX6A0CEGU4geXoqIiVK9eXWKtL0C8Fh9vdKE4EVtu4nRmZmaFw8e5lLdqalmMjY0xadKkCikZ8D5H3NQKLqJUQB4+fFiu3xCHpBQV7mBo9+7dkJeXx9evX2FmZgYbGxvWa6xcfvuouV8VNuHS4pCUH5WVlQUdHR3IysrCw8NDqICoq6srdu/ejREjRqBPnz6ws7PD+PHj+fa1evVqxMTE4Pz585Vz4L8oSUlJ8PHxwb1798DhcKCiooJRo0ZJfFFt2rSJVVRdeZSGK4urV69iw4YNfPkitWrVEipEWZbAwEC++6usIklWVhZ0dXXLva5la2uLHj16wMzMDPXr10deXh6srKxw+vRpgbZla31J0uLjIkzE9tatW8jOzsbx48fRpUsXdO/eHZs2bSrXepAwflSdwtraGhcvXkSNGjUwevRoWFlZCczCuaxcuRKqqqoCiud79uyBk5MTbt26xVwfcSogz549w+nTp0Uq8bOBTYqKKL58+YLt27eXSyKKDFEFEFfsqmbNmmjRogUmTZoEe3v7H/oNSYZIV1cX8+bNYyUgeuHCBdjZ2WH27NnYsGEDnJyc8OTJExw5cgTGxsYS1xKA8lfk/JUpG05dtlQH9++SkhI+xYywsDD8+++/AqNbXnFSaZOWlgZVVVWx7ji291dFDNHDhw+RmpoKBwcHnD59GkVFRbh165bY77HR4uMiSsR23rx5ePjwIS5duoTLly9jw4YNWLt2Ldq2bVthZZOrV6+ioKAApqamSE1NxdixY/HkyRNGnUKYEklZPnz4gCNHjuDQoUN48OAB9PT0YGVlhbFjx/K50P755x/s27dPwJsQHx+PwYMHY8OGDbCyskJqaio6dOiA2NhYRkA2Li4OU6dOhbe3N6ysrGBgYCBUiV+S5hsXSSkqly5dwu3btyEnJ8eskX7//h27du3Cxo0bUVxczLjm2UCGqAL4+/vD09MTw4cPZ3Sa7ty5g5CQEMyZMwfp6enYt28fli9fDjs7uwr9RnleFHfu3OETEG3ZsqVQAVFhI+hPnz6x8iezrf74q5OdnY3du3cjOzsbAwcORKdOnXD79m24uLhgwYIFfAW+vL294e7uzmeIpKGGIInc3Fy+tZqWLVsKzSXh5WcZIi5HjhzBjBkzUKdOHRw6dEii+27u3LlQVFRklJzFafGJq3Lbu3dvZGRkQE1NjVlHrV69ukAU14+UPSmrTlEe4uPjERAQgOPHj+Pr168YNmwYrKysYGRkhCZNmiAuLk7g3kpNTUWnTp1Qo0YNiSogb9++lajEzwZlZWXcuHFDwGPw4sULdOnSBbVr10Z+fj5kZGQYLcGJEyfi+/fvmDFjBiZOnCig3iIOWiOqAFevXoWrqyujzAyU1kHq0KEDLly4gMDAQGhra2Pr1q349OlThafIycnJjJHgVmrNzc0FwF/WnG0Buf79+6N///4sjvDPYM6cOeBwONiyZQuAUi29vn37MnVV9u7diyNHjmDVqlXw9PQUWuDL1dWVzxD9Krx69Qrz589HdHQ0X2IoN4S5qvDw8ICWlhZfccHMzEysWbMGw4cPx7NnzzBixAhWSecyMjKwsLDAzp07xWrx1atXDy9fvhQwRC9fvkT9+vUB/LzZe1paGgoKCiocDNChQwd06NABq1evxrFjx+Di4oKTJ0/i/fv3UFNTQ2xsrIAhunbtGtTV1TF//nxcvHgRvXr1EpAjy8jIwMSJE2FnZydWiZ+tIRKXoiInJ4f+/ftj4cKFCAgIgJ+fH8aOHYtFixZh3LhxFTLQZIgqQFRUlNCiWD179mQi9KpXr44XL17g4sWLIovVSWLUqFF8Dy/vww6UL3SWi6jSDH8i169f56trExQUhI8fP+LOnTtQV1eHg4MDtmzZgsePHws9B82aNcPTp0+rssuscXBwQF5eHlNgjfdeYBsiz3agI46goCAcOHCA+TsvLw+jRo1Ct27d4Ofnh9DQUCxbtkyi2nW7du1w5MgRHDhwAAsXLsT58+ehoKAg1BCJErF1d3fH6NGjAQg+K+XlxIkT+PDhA7MGA5SWxt63bx+AUkX3EydOVOjZefbsGQICAnDkyBF8/PiRGehYW1tj6dKl+P79OxNUExkZCXd3d8ydOxeTJk0SuTazYcMGAECdOnWYPLimTZvixYsX0NfXR1FREXNt2SAuRUVOTg7//fcf9PT04OzsjO3bt8PNzY3J76oIZIgqgKKiIkJCQgSUrENCQhg5na1bt6JOnTq4fPky1NTUsGPHDr4psiTYKEawoWxphq5du/KVZuBwODh16hT69Onzw3lEvxpv3rzhk66JiIjAiBEjmDwJe3t7mJubQ1dXF15eXvD19eUr8LV27VqxRdykSXx8PC5fvixSrUFS6DKHw6mUgU5WVhZzzxcWFmLcuHHo2rUrNmzYABkZGejp6eHNmzesclO4CaI2NjZISEgQMI5AqUrKihUrwOFw4OjoyCdiO3XqVL6iillZWTh69ChSUlKwbNkyKCkp4caNG2jatKlE1+rOnTv5dCIjIiKwd+9eLFu2DK1bt4aHhwfWrVuHjRs3SjwuoFTn8MSJEwgICMDNmzfRokUL2NnZwdLSktE3nDVrFt6/f88IoAKlOUD29vYwMjISqt5Rv359aGhoMGtlbJT42SAuRWXixImMtqCCggIUFBR+WDGFDFEFcHJywpw5cxAVFYUOHTowgn9hYWHYvHkzgFK/LtfVU5EpMtukMklIKs0wbtw47NmzB3369KmUPKJfiWrVqjHJnUDpOh6vS6N+/fr4+PEjvL29MXbsWIECX9WqVYOHhwffC0DYyxGofBkpSWhoaIhNxuUNzRZGq1atcOLEiUrpx549e2BpaYlFixahXbt28PT0ZD5PSkpiPWvgNYzcZ4fXOHLVKWrUqAEvLy+4ubkJiNhySUhIgKmpKTQ0NPDo0SPMnj0bSkpKCA8Px/Pnz5ny8KLgVsjlEhISgn79+mHhwoUASoOS2FQgjYuLw8GDB5mgDRMTEyxbtkxkkIObmxsWLlzIVJ7V0dFBnTp1RGoVysjIoEaNGpg5cyaWLVuG1atXMyXeFy9ejE+fPuHMmTOMEn95EJeikpubi+rVqzMu1fz8fHz48IGvTXkKCZIhqgCTJk2Cjo4Odu7ciQsXLjA3zIULF5hRh6KiItzd3QH82BT5R7l16xbfWhZQqi/GHREGBgaWqwrp74SOjg7OnTuHuXPn4v79+3jz5g3fCyAtLQ2NGzdGhw4dkJiYyFcyYMyYMTA3N4eamprAC0DYzKGqRTTXrFkDd3d3bNiwQWgIekhISKWJ/Ipj4cKFsLe3x9atW9G+fXvcuHEDDx8+hJ6eHm7evImlS5cy7jJxVMQDoKCgILKgpbOzM+zt7bF06VK+hO/+/fvj0KFDEvddUFDA9yK9efMms/YClObQsFGRHzJkCAwMDODm5gYLCwtmDUsctWvXRocOHfi2iTo/eXl5SEhIwOrVq5lyN1wUFBQEkrglwUZIGQBf7iKHw0G/fv34/i7vM0GGqIJ06dJFbI5CZU2Rf5T09HS+0Gx3d3c+zarGjRtLrSzDz2b27NmYMmUKQkND8fTpUwwcOJDPJXP58mVm1KugoABra2uBfVSWi7Qy4BpFLoWFhejUqZNA7ZeqDFawsLBA165dISMjAxUVFcyePRvdu3dH9erVUVxcjK5du0p0+QLsPAAODg6s+uTr64vExERs3bpV4LMmTZoIrZpcFhUVFTx69Ajq6urIz8/HgwcP+MoasBVP5ZY8F8W4ceOwc+dO1KtXj1XxRWG0bdsW9erVg6enp9jCl2zYvn07Xr16JVZIecyYMRITncsLGaIKUHYKWhZFRcVKnSL/CDVr1kR6ejrzoPMuvgKlIoi8D9SJEydElg1gm4PwqzB8+HAcO3YMFy9exODBgwVyImrVqsUkD16+fBn+/v5ITU3F8ePHoaamhgMHDkBDQ6Pculk/C7ZrdGxf2JUFrxHx9fXFrFmz8ODBAzRr1ox1YA4bsrOz+f4WldMFlN73wjwPbMvBmJmZwcnJCRkZGbh69SqaNGnCN4BMSEgQiNoThiSXLVeolPv/FaVdu3Z49eqVwGClLJLC1i9cuMDnWi0LV0hZnMxSRSBDVAG0tLTEXuz379/zjbwrMkWuLMpTmsHFxQV+fn7o1auXyLLkvxtGRkYiDQk3wjEoKAjz58/HpEmTEBkZySyAFxcXY/Pmzb+MIWIbCfbq1atKFfktL7q6uj8lyKNs0rCamhq2bNkiNPBg6NCh8PT0xP79+5ltL1++hJubG6uowv/++w/p6elwdnZGkyZNsHPnTj5h22PHjlVKWD9vue0fKb2dmZmJRo0aCYR1FxUV4d69ezhz5gwWLFggcT+pqamshZR5KW/5lLJQQmsFiImJ4fube7F3796NOXPmIDs7u9KK1f0o586dg7W1NdasWSO0NIOrqyv27t0LExMTaGtrY/369RgxYkSV9O1XoUePHpg/fz5Gjx7Nl9iZlJSEUaNG/ZIh3L9C4bfyVOAs+4KsDMQlF+fn52PMmDFITk5GQUEBmjRpgqysLHTt2hXBwcFSNdSVDbfcd/PmzUXmJx44cABRUVESgzQ0NDQQFBQkUr8yLi4OY8aMwcuXL/m2/2iiN82IKoCwaWmfPn2goaEBNzc3dOnSpdzF6n4W5SnNUFJSwkom/k/jxYsXQtft6tSpg48fP0qhR5IRVwCNTZG6yqCsflxaWho+f/7MaN29ffsWCgoKUFdX/ymGSBz16tXDxYsXERkZyVREbdeundjaTJLYuHEjpkyZUunK1mzW0I4dOyZ0tpGfn4/MzExoa2uLNTK9e/cWKH4ojHbt2rEWUq5MyBBVIgYGBnj58iV8fHxEtuEtVldVsC3NYG1tjaNHj2LJkiVV1rdfgaZNm+L58+cCC+bXrl2rcBmPnwV3AV5cATQ2axeVAa/gJzdB08/Pj1F5T0tLg4ODAywsLKqkP8IQ55otL97e3hg5cmSlGyJxaRNAadpBYWGhwBonANStWxfa2tpMFVdRHD9+nJUXxsbGBlOmTIGKiopQIeUdO3Zg9+7dAt8TVT6FLWSIKolPnz4xPt6K+Fh/NqJKM/COxjgcDoKCghARESG0dMXvltDKFmtrazg5OTFSQK9fv0ZsbCzc3NwqvZbVj7Jz504AEFsAjW2SZWWydu1aHDp0iK/UiLq6Ojw8PGBpaVkpUVZlEzrF5XQJi5jjha3UTdnf+xmISpu4fv06U6Nozpw5rO7F7t27C2zLysrChw8fWK1Tm5qaYu7cuVi6dClWr14tVEhZmOv+R8unkCGqAGUjUzgcDj5//ozatWtDQUGBdbG6X4GyozGua+7Jkyd82/+EwAVRzJkzB/n5+Rg5ciQKCwthYmICeXl5ODo68lWz/RXgVuQcPnx4lRVAY8O7d++Ellj/+vVrpa1X9e3bl3VOF9dgcykqKmI0Bhs1alQhQ1RVPH78mBF9HTduHPbu3StUHX/BggVYunQpXzqGqakpXxtuJeGePXvyqYyIw8XFBUOHDuUTUu7Ro4dQIeXPnz8jKSlJaJRt2b6Ig4IVysGYMWOwdetWXLlyhW8792J36tQJVlZWaNeuHZ/GGS/Lli3DvXv3/ggl6z+Nz58/4/HjxygpKWEy2n8HsrKy0KhRI7FlH34248ePx8uXL7F582YmGTM+Ph5z586FhoYGDh8+/MO/wdaTIConKSsrCw4ODrCysqpQyfrXr1+jWbNmP20gmZGRgVWrViEwMJApny6qJhMgWEOqqomIiMC0adOEDjQoofUn8unTJ/z777/w9vYWGVlWUR+rtOEWJSsqKkKPHj1+WY21n4GDgwM8PT1Rt25dvqJlBQUFWLRokVSK3kmiqKgIK1aswJ49e/DlyxfcuXMHmpqacHNzg7q6OmxsbKq0Pz4+PpgxYwYGDRrE3PMlJSXo378/4/L8UX5U9kpZWRnOzs6YMmVKhQyRKC/Hj/Lx40ds2rQJ27dvh76+Pk6fPo0ePXpI/J44V2FViBsvXrwYgwYNgqurK6tijOKgGVE52bZtGzw8PGBiYoJ169YJjY5buXIlq2J1vwrlKUr2JyIqFDonJwetW7dGTk6OlHomGg8PD5w5cwZubm6YPn06Uz7g9OnT2Lx5M8LCwqTSr2fPnjEySTo6OmjVqpVU+iGKhIQEmJiYiE3slJQUysuP1DXi0rJlS3z58gV2dnZiUyfKJsfyhkyXFTdWU1PjEzfmlk/nVlutDFRUVCotoIcMUQV49uwZZs6ciTdv3mD69OkCU3VHR0fWxep+BYYOHYq6deti06ZNEouS/Ul8+PCBuTY3b95kFIWB0jyr0NBQeHh44OHDh1LspXC4C/I9e/bkeyE9ffoU/fv3l0pAzK/EmTNn+P7mcDjIzMyEv78/NDU1ERQUJPK75XEj/mi5CYBfHFRU3SZJrq4ZM2ZAV1cXc+bMAQBG8Z8rbsydOUvKIyoPI0eOZGbBPwq55ipAq1atMHHiRMybN08g41pGRgaOjo6si9X9Cjx48ADnz59nptfiipL9SXAVMmRkZITmTcjIyPyyoewZGRl8EWpcioqK+BTHq5Jnz57h9OnTFS4EWZlMnjyZ728ZGRk0atQIvXv3hoeHh9jvVoZxKQ8V0TP88OEDX4DAtWvXBMLUecWNp0yZIiB+/KNMmTIFLi4uyMjIgL6+vkCUbXkU6ckQlZN3795h9uzZuHbtGrZs2YIJEyZIu0s/TF5eHp9bihv996cborNnz4LD4cDU1BQHDhzgG5nWqFED6urqP+z7/lno6uoiNjaWSVDmcvLkSbRr167K+xMaGgorKysYGBhUuBBkZSJJD1ISYWFh6N27N/Ny/fjxI+rWrct8XlhYiODg4EoJSy/P2ldiYiIcHBwEol1LSkowY8YMRji1KsSNucaeOwvjhYIVfiKnTp3CggULoKenh5iYmEqrGfQrwFutExCeo1HVNXd+NlyFjMTERKirq/9WIepOTk6ws7NDeno6iouLcerUKTx58gTHjh0T63b6WaxevRpOTk6YP39+hQpBViaVEXhjbm7Ot26or6/PF6GWn5+POXPmVLoKtSQcHR2hoqICT09PKCsrM/fsgAEDsH79eqadJHHjyqAylenJEJUDe3t7uLi4YObMmb/VS4sNZat1AsKLkv0pJCQkwMDAALKysnj//r3YY/sVDbCxsTH27t2LDRs2QFZWFl5eXky57R+Rsakoz549Y+r1VKQQZGVRWYE3ZZ+FqiytIY4XL15g//79AjWo2rdvj8TERJEqFrzixpVFZQ7EyRCVg4iIiD8yrPlXqrlTVfTt2xdPnjxB48aNhSZKcvmVDXD//v3Rv39/aXcDQKkuHzehVZqFIFetWoWePXvyBd64ubn9MRGg3bp1w+PHjwUMkY2NDaytraGhoSFU3Njf3x979+6t9P4UFRXhzp07QtcFx48fz3o/ZIjKwZ9ohIDKHdn8LiQmJjJRcn+jIa5sfpVCkH964I2Pjw9mz56N1NRUvgABRUVFjBo1ipW4cWXx5MkTjBs3Di9fvgSHw0G1atVQVFQEOTk5yMvLkyEiCEnwGt/fyRAL0xIri4yMDK5du1YFvfkfv0ohyMoMvOFdNy27ZiqtWfLz58+RlJSEq1evCnwmIyODy5cvSxQ3riyWLFkCQ0NDREdHQ0dHB9HR0cjLy8OCBQvKLepMeUQEgcrTzPrZiKuemZOTg4CAgErVd/vdUFRUxMmTJ/mMzrBhw+Dv788XASlp3U9RUVGku5aLNNy2nTp1QocOHTBv3jy+YAUuVVXnDCgVdw4JCYG+vj6aN2+Oq1evQltbGzExMVi0aBGfOrskaEZE/PVUpmbWz0aYAvOXL1+wdetWHD16FFpaWnB3d5dCz34dKiPw5ld117558wbBwcG/RHkSDocDBQUFAICSkhLevHkDbW1tqKqqIiUlpVz7IkNE/PVUpmZWVVJSUoIDBw7Ay8sLcnJy8PLywrhx46osopM7a2BDVRnzyjIgv6q7tk+fPkhISPglDJGenh6SkpKgqamJjh07YvPmzahWrRoOHDhQ7v6RISL+el69eoXAwMDfygidO3cO7u7uyM7Oxvz582Frawt5efkq7cO+ffuY/3/37h1Wr16N4cOHM8EJt27dQkhISJWqU/yqBqSyGDBgAJydnZGcnCxUzaAq3cgLFy5EQUEBAMDZ2Rljx46FiYkJlJSUyh2hR2tExF9PZWpm/Wxu3LiB5cuX4969e7C1tcXcuXN/iWiwcePGwdjYWEBaZ//+/QgJCZFKku2fCK/6R1l+BTfyhw8f0KBBg3LPyskQEX89Z86cwapVq+Dg4PDDmlk/G0VFRdSqVQuTJ08WWiyNS1UXflNVVUV0dLRAfsuLFy/Qs2dPvHnzpkr7Q/x8MjMzUVRUJHAfpqenQ05ODsrKyqz3RYaI+Ov51UeZvLRt21biaFNGRqbKF9vbtm2LqVOnYt68eXzbN27ciD179iApKalK+0P8fMzMzDBy5EiBWfCBAwdw8uRJnDx5kvW+yBARfz2SSib86esOlUFgYCAcHBzQp08fZo3o9u3biIiIgI+PT5UrWv/JfPjwAVeuXBGqZuDk5FRl/eAN2ebl2bNnGDBgAFJTU1nvi4IViL+e393QpKeno1mzZlIvFa6trY0dO3bgwoUL4HA40NXVRWho6E9JpvxbuXXrFsaMGQN5eXlkZ2ejWbNmyMzMhLy8PNTV1avUEBUXF+Pr168C2wsLCwUMpCRoRkT8tWzdulXo9nr16kFbW1sq5Qsqgrq6Op8yNPHnYmxsjLZt28LLywvq6uqIiYmBgoICpk2bhkmTJmHMmDFV1hdTU1O0atUK3t7efNvnzp2Lp0+fIiQkhPW+aEZE/LXs3LlT6Pa8vDzk5+ejS5cuOHLkiNg1pF+BX0UZmktmZqbAiFhYET+i/CQnJ8PHxwcyMjKQlZXF169foampCXd3d9jY2FSpIXJxcYGpqSnu37+P3r17AwCio6Nx7949nDp1qlz7IkNE/LXcu3dP5GepqamwtbXFypUrBUZ8hCB5eXlwcnLCqVOnhLplfqWAj98ZOTk55v+VlZWRlpYGHR0d1K5dGxkZGVXal86dO+PSpUvw8fHBuXPnwOFw0K5dO6xfvx5t27Yt177IEBGEEDQ1NeHm5gYHBwdpd0Ui8+fPl3oukYuLC+7fv49Dhw5h0qRJ2Lp1K968eYPt27dXqejpn067du0QHx+PVq1aoWfPnvDw8EBWVhaCgoLQpk2bKu9P27ZtRXoWygOtERGECF6+fInu3bsjPT1d2l355dHX14e/vz+6d+8OdXV1REZGQktLC8eOHUNAQEC5XTWEcO7evYuPHz+id+/eyM7Ohr29PeLi4tCyZUv4+vr+dGP04cMHxlUtqRx7eVzaNCMiCBEkJyf/Umsb5Zmd+fr6/sSeCJKXl8ecq3r16uH9+/fQ0tJC586dMXv27Crty59M+/btmf9v1KgRjh07VqW/37JlS6aEupaWltCcNg6HU+78OzJExF+LqBFdXl4eEhMT4eLigokTJ1Zxr0STnZ3N9/f169chIyMDfX19AMDDhw9RUlLCqmZRZaOpqYnU1FSoq6ujdevWOH78ODp27IizZ8/+8sEeBHvOnDnDXM+zZ89W2n7JNUf8tYhTj5aRkYGVlRW8vLxQo0aNKu6ZZLy9vXHv3j34+vqidu3aAICCggLMmjUL+vr6WLhwYZX2x9fXF9WqVYO9vT0iIyMxbtw4fP/+HSUlJfD09IStrW2V9udPRdIgozw1gH4lyBARfy0xMTFCt9etWxctW7ZEnTp1qrhH7NHR0cHp06cFytc/fPgQI0aMwJMnT6TUs1LS0tJw9+5dtGzZUiqL6H8qZQsjFhUVISkpCTdu3MD06dPLXRm1Mnj79q3QgpLl0Wgk1xzx19KzZ09pd6HCFBQUICMjQ8AQZWZm4suXL1Lq1f9QV1f/pdbX/hSEFUYEgC1btiAtLa1K+5KYmAg7Ozs8efJEIJetvGtENCMiiN+QGTNmICoqCitWrGAkdG7fvg03Nzf06tULfn5+Vd6n27dvIzIyUujoeO3atVXen7+JlJQU9OnTBy9fvqyy3+zbty8aNmyIRYsWoWnTpgJu7vJIZ9GMiCB+Q7y9veHs7IyZM2fi+/fvAIDq1atj0qRJWLlyZZX3x8fHB66urtDS0hJ4KVVVxdi/mWvXrjFlu6uKx48fIyoqCq1atfrhfdGMiCB+YwoKCpCSkgIOhwMtLS0mcKGqadOmDebMmUNBCT+ZcePG8f3N4XCQmZmJe/fuwcnJqUpFTwcMGAB3d3f06NHjh/dFhoggiB+mefPmiIqKIuHVn8zMmTP5/paVlUWjRo3Qu3dv9OvXr0r7EhkZiZUrV8LZ2Rn6+vp88kNA+RJayRARxG9IYWEhtm/fLnJNpqrDeOfNm4c2bdrAxsamSn+XkB68hobX/UoJrQTxl7BgwQKcO3cOZmZm6NKli9TXYVRVVbFmzRrExcWhTZs2AuXWq7p0OfHzoYRWgvjL0dTUxL59+9CnTx9pdwUAYGBgIPIzaZQu/9N4+vQpPD09sXv3bgCAmpoaPn/+zHwuKyuLyMjI3zZni2ZEBPEboqCgAFVVVWl3g0FcSQ3ix9m5cye0tLT4tnl5eUFZWRkAEBQUhB07dmDLli0/tR8JCQkwMDCArKwsEhISxLalhFaC+MOZPXs2fH194e3tLdUS4UTVEB0djXXr1vFtGzhwIBMcoqysjBkzZvz0fvTt2xdPnjxB48aN0bdvX8jIyAgtzEhrRATxFxAeHo7r16/jypUr0NXVFViTOXLkyE/vw6JFi+Dm5obatWtj0aJFYttSQuuPkZaWxpcgamNjg/r16zN/q6mp4e3btz+9H4mJiWjUqBHz/5UFGSKC+A1RUlLC8OHDpdqHBw8eMMm0ycnJYgVkiR+jWrVqePfuHTQ0NAAAy5cv5/s8MzMT8vLyP70fvMawPMoJkiBDRBC/Idu2bZN2F3Du3Dnm/0NCQqTYkz8ffX19REREMHJOZbly5Qr09PSquFfAt2/f8ODBA2RnZwukEAwaNIj1fsgQEQTxQ3z//h1t2rTB6dOnpfIy/BuYMGECnJyc0LZtWwwePJjvs3PnzmHLli0Cytw/m/DwcNjZ2eHdu3cCn9EaEUH8JQQEBOD48eN4/fo1vn37xvdZVYZLy8nJQU5OjlxwP5FJkyYhOjoa48aNQ+vWrRl9t6dPn+LZs2cYOXIkrKysqrRPCxcuxODBg/Hff/9BWVn5h64/hdsQxG/Ili1b4OzsDENDQ7x69QrDhg2Dnp4ePnz4IJWqsra2tvD29kZRUVGV//bfws6dO7F79260bNkSz549w7Nnz9CqVSv4+/sz+UVVSWZmJhYsWIDmzZujZs2akJeX5/tXHiihlSB+Qzp27AhXV1eMGDECampqiImJgaamJtauXYvXr1//9HwSLmlpaVBVVcX48eMRGxuLmjVrQk9PT0AJuiqi+IiqZdq0aRg0aBDGjh37w/si1xxB/Ia8efMGHTp0AADUrFkT+fn5AABzc3P069evygxRu3bt8PjxYygpKcHExKRKfvNv4+PHj6hbt+5Pa19RvL29YWtri4SEBOjp6QmIno4fP571vsgQEcRviLKyMnJycphKqLdu3YKBgQFevHhRpWs13GTGXyGK70+lY8eOmD59OiwtLUWqaZSUlODq1avw9fWFkZER5s2b99P7FRYWhsjISFy6dAkKCgoCNajIEBHEH07v3r1x4cIFGBoaYtKkSVi6dClOnjyJe/fuwczMTNrdIyqRCxf+r727j6mq/uMA/j6oPAwELw9RyE2eYWAlaWL2h7mg/sgYKW5BD4aoKy9YzD/UGiO4TnHlrkC2GXOU0+WAdVspZmVKIZLTEEyBTWVAZUTyFJgpcH5//ObNK2LAvfd879fer7+857Bz3n/c3Y/f50MwGo2YM2cO4uLiEB8fj8DAQLi7u6O3txfNzc04deoU3N3dsX79eqxYsUKTXLm5uVi9ejU2btxo8zlYHCMiktDIyAhGRkYsOyp8+umnqKurQ0REBDIyMkZ1kziKTqdDdnb2v/4QaXlg273q559/htlsxokTJ9DR0YFr167Bz88PDz30EJ566ik8/fTTmm73pNfr8d133yE0NNTmZ7EQEdGk6XQ6REZGjtpi6HZan49Ejrd27Vo89thjyMjIsPlZ7JojksS/7XZ8q4nsfGyrgwcPIiAgQLP30T+GhoZw7do1eHl5af7ukJAQGI1G1NbW2nwGFVtERJLQ6XRj7nZ8q4muareFr68vWlpaWIgcrLq6Gt3d3Xj++ect10wmEwoLCzE0NIQnn3wSu3fvxowZMzTLZM8zqNgiIpKEMx4u929FkezDZDIhMTHR8vn06dMoKCjAyy+/jKioKJSUlGD79u0wGo2aZbLnGVQsRESSsOdux/ayYcMGm2dM0b87f/681Y7bn332GRISEizrxYKDg7F582ZNC9GtBgYGoCjKpL8LLEREkvr9999RWlqKlpYWKIqC6OhorFq1ynJqpxY2btyo2bv+y/r6+iznAAHADz/8gKSkJMvn+Ph4Tc4jul1paSmKiorw66+/AgCCgoLw5ptvYtWqVRN6DveaI5JQXV0dHn30UVRWVlr2+aqoqMDcuXNx8uRJ0fHIzgIDA9Ha2goA+Pvvv9HY2Ij58+db7g8MDMDV1VXTTNu3b0d+fj5eeuklmM1mmM1mvPjii8jPz4fJZJrQszhZgUhCSUlJiI2NhclksqwdGRkZQU5ODpqamvDVV18JTkj2tH79etTX1yMvLw+HDh1CeXk5mpubLcWnvLwcu3btwpEjRzTLNHv2bLzzzjtITU21ul5eXo6CggL89NNP434WW0REEjp79iyysrKsFjC6uLjAYDDYdRCZnMNbb70FNzc3pKSkYN++fSgqKrJqAe3duxeLFy/WNFNXV5dlv8NbzZ07945nFN0Nx4iIJOTt7Y22tjZERkZaXW9ra4OPj4+gVOQofn5+OHToEPr6+uDl5YUpU6ZY3f/44481nzQSHh6OioqKUbtmVFRUWM5LGi8WIiIJLV26FNnZ2cjPz8f8+fOhKArq6uqQn5+PZcuWiY5HDjLWfzJ0Op3GSf4/USUjIwO1tbVISEiwfAePHz+Ojz76aELP4hgRkYSuX7+O3NxclJWVWQ6jmzZtGlauXIn8/HzNB67JsRYuXHjH6z4+PoiIiMDatWuFHNN+5swZfPDBB2hpaYGqqoiJiYHBYMAjjzwyoeewEBFJ7OrVq2htbYWqqggLCxt1IB3dGwoLC+94va+vDw0NDTh16hQ+//xzLFiwQONk9sFCREQkOaPRiJMnT+KLL77Q7J39/f04duwY2traAAChoaFYtGjRpA7lYyEikojBYBjX3+3cudPBSciZNDc349lnn8XFixc1eZ/ZbEZOTg76+vqsrvv4+GDHjh0TPhOLkxWIJPLHH39Yff7222/x+OOPw8PDQ1Aicgbj2QzXXs6dO4c1a9YgJSUFWVlZiImJgaqqaG5uRklJCdasWYPIyEjExcWN+5lsERFJLDg4GDU1NQgJCREdhQR69913cfToUVRVVTn8XevWrUNXVxc++eSTO95PS0tDQECAZR+88WCLiIjIyb3//vt3vN7f34+GhgYcOXIElZWVmmQ5ceIEtm7dOub9zMxMbNq0aULPZCEiInJyH3744R2vT58+HZGRkaiqqrLae86RLl++jKioqDHvR0ZGWjZBHS8WIiIiJ+dM2zYNDg7edUzSw8MDf/3114SeyUJEJJHbjwtXVRVNTU3o7e21uq7lUeEkRl1dHeLj4+Hm5qb5u8+dOzfmabCTOR2YkxWIJDKe48K1PCqcxNHr9fj+++81n6jiiO8gW0REEnHG48JJDFHHtDviO8hCRCQRZzwunP5bHPEd5HlEREQSMplMCAgIEB3DLjhGREREQrFrjohIEkuWLIGiKKOuK4oCd3d3hIaGIi0tTbpZk+yaIyKSRHR0NBoaGtDZ2YmgoCAEBQWhs7MTDQ0N8Pf3R11dHRITE1FdXS066oSwRUREJAk3Nzekp6ePOp/o7bffhqIoqK6uxoYNG7B582YsWrRIUMqJ4xgREZEkQkND8c033yA8PNzq+oULF5CUlITW1lY0NTXhmWeeQXt7u6CUE8euOSIiSdzcSeN2zc3NlnVF06ZNu+M4kjNj1xwRkSTS0tKQnZ2NS5cuIT4+Hoqi4Mcff8SOHTuQnp4OADh+/DhiY2MFJ50Yds0REUlieHgYxcXF2LVrFzo7OwEAgYGBeO2115CdnY0pU6ago6MDLi4umDlzpuC048dCREQkof7+fgCAt7e34CS2Y9ccEZGE7oUCdBMLERGRJHp6emA0GlFdXY2urq5RG592dHQISmYbFiIiIklkZWWhsbERr776Ku6//37pZseNhWNERESS0Ov1MJvNmDdvnugodsV1REREkvD394enp6foGHbHQkREJInc3Fxs2bIFAwMDoqPYFbvmiIgksXDhQrS3t2N4eBh6vR5Tp1oP89fW1gpKZhtOViAikkRycrLoCA7BFhEREQnFMSIiIhKKXXNERE5Mr9fjzJkz8PPzQ3Bw8F3XDnFBKxER2d22bdvg5eVl+fe9soj1VhwjIiIioThGREQkifT0dHz55ZcYGRkRHcWuWIiIiCTh6emJlStXIjY2FgUFBbh48aLoSHbBrjkiIon09/ejoqIC+/btQ319PRYsWIBXXnkFKSkp8PDwEB1vUliIiIgk1dTUhD179qCsrAyurq5YunQpXn/9dURHR4uONiHsmiMiktDly5dRVVWFw4cPY+rUqUhOTsYvv/yCJ554AiUlJaLjTQhbREREkrhx4waqqqqwd+9eHD16FA8//DBWrFiBZcuWWaZ4m81mvPHGG2hvbxecdvy4joiISBLR0dFQVRWpqak4duwYZs+ePepvFi9ejBkzZmgfzgZsERERSWL//v1ISUmBu7u76Ch2xUJERERCsWuOiMiJvfDCC+P+2/379zswieOwEBEROTFfX1/RERyOXXNERCQU1xEREUnCYDDgzz//HHV9cHAQBoNBQCL7YIuIiEgSvr6+aGlpQUBAgNX1K1euICoqCleuXBGUzDYcIyIicnI9PT1QVRWqqqK3txdTp/7z0z08PIzDhw/jvvvuE5jQNixEREROLiwsDIqiQFEUJCQkjLqvKAo2bdokIJl9sGuOiMjJ1dTUQFVVJCcnY8+ePdDpdJZ7rq6u0Ov1eOCBBwQmtA0LERGRJNrb26HX6++548LZNUdEJInu7m50d3ePeX/OnDnahbEjtoiIiCSh0+mgKApU9Z+f7VtbR3crUs6MLSIiIkk0NDRYfR4aGkJjYyPee+895OXlCUplOxYiIiJJPPjgg6OuhYWFwdvbG9u2bUNSUpKAVLbjzgpERJKbNWsWzp49KzrGpLFFREQkiZ6eHqvPqqrit99+Q2FhISIiIgSlsh0LERGRJG4ubL2VqqqYOXMmysrKBKWyHWfNERFJoqamxuqzi4sL/P39ERYWZrXtj2xYiIiISCh5SygR0X9MfX09KisrceHCBQBAREQEli9fLu1C1pvYIiIikkBBQQFMJhM8PT0REhICVVXR1taGq1evIicnB7m5uaIjThpbRERETq68vBw7d+7Eli1bkJmZCVdXVwDA9evXUVpaioKCAsTExGD58uWCk04OW0RERE4uMTERycnJWLdu3R3vFxUV4cCBA/j66681TmYfXNBKROTkmpqasGTJkjHvP/fcczh//ryGieyLhYiIyMndvtHp7e52TwYsRERETi42NhYHDx4c8/6BAwcQFxenYSL74mQFIiInt3r1amRnZ8PNzQ2ZmZmWxas3btzA7t27sXXrVhQXFwtOOXmcrEBEJIG8vDwUFxdj+vTpmDVrFhRFQWtrKwYHB2EwGGA0GkVHnDQWIiIiSZw+fRrl5eW4dOkSACA8PBypqamYN2+e4GS2YSEiIiKhOFmBiIiEYiEiIiKhWIiIiEgoFiIiIhKKhYiIiIRiISIiIqFYiIiISCgWIiIiEoqFiIiIhGIhIiIiof4HAAB1IbXtw/gAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# define fitting function\n",
    "mte_transformer = col.MeanTargetEncoder(feature_selection)\n",
    "\n",
    "pipe = col.CategoricalPipeline(features=feature_selection,\n",
    "                               model=RandomForestClassifier(n_estimators=100, max_depth=5),\n",
    "                               transformer=mte_transformer)\n",
    "\n",
    "# train model on entire dataset (no CV)\n",
    "pipe.fit(df)\n",
    "\n",
    "\n",
    "\n",
    "# get Mean Target Encoder learned weights for each categorical variable:\n",
    "mapper = pipe.pipe[0].mapper\n",
    "for k, v in mapper.items():\n",
    "    pd.Series(v).sort_values(ascending=False).plot.bar()\n",
    "    plt.title(k)\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot Feature Importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_feature_importance(pipe: col.CategoricalPipeline, *args, **kwargs):\n",
    "    fi = (pd.Series(pipe.model.feature_importances_, \n",
    "                   index=pipe.features.categoricals + pipe.features.numericals)\n",
    "          .sort_values())\n",
    "    fi.plot.barh(*args, **kwargs)\n",
    "    plt.show()\n",
    "\n",
    "\n",
    "col.plot_feature_importance(pipe)\n",
    "plt.savefig('../figures/feature_importance.png', facecolor='white');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'workclass': 9,\n",
       " 'education': 16,\n",
       " 'marital-status': 7,\n",
       " 'occupation': 15,\n",
       " 'relationship': 6,\n",
       " 'race': 5,\n",
       " 'sex': 2,\n",
       " 'native-country': 42}"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# number of unique values for each categorical column\n",
    "{col: df[col].nunique() for col in feature_selection.categoricals}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Compare relative performance of MTE for various types of downstream classifiers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\tevaluating encoders when fitting a RandomForestClassifier(max_depth=5)\n",
      "\tevaluating encoders when fitting a LogisticRegression(max_iter=500)\n",
      "\tevaluating encoders when fitting a KNeighborsClassifier(n_neighbors=10)\n",
      "\tevaluating encoders when fitting a LGBMClassifier()\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>model</th>\n",
       "      <th>transformer</th>\n",
       "      <th>acc</th>\n",
       "      <th>f1</th>\n",
       "      <th>auc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>RandomForestClassifier(max_depth=5)</td>\n",
       "      <td>MeanTargetEncoder(target=target, alpha=5)</td>\n",
       "      <td>0.851970</td>\n",
       "      <td>0.626713</td>\n",
       "      <td>0.908543</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>RandomForestClassifier(max_depth=5)</td>\n",
       "      <td>TransformStrategy_OneHotEncoder(handle_unknown...</td>\n",
       "      <td>0.836952</td>\n",
       "      <td>0.550605</td>\n",
       "      <td>0.897631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>RandomForestClassifier(max_depth=5)</td>\n",
       "      <td>TransformStrategy_OrdinalEncoder(handle_unknow...</td>\n",
       "      <td>0.848991</td>\n",
       "      <td>0.612731</td>\n",
       "      <td>0.902580</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>LogisticRegression(max_iter=500)</td>\n",
       "      <td>MeanTargetEncoder(target=target, alpha=5)</td>\n",
       "      <td>0.845429</td>\n",
       "      <td>0.639919</td>\n",
       "      <td>0.899764</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LogisticRegression(max_iter=500)</td>\n",
       "      <td>TransformStrategy_OneHotEncoder(handle_unknown...</td>\n",
       "      <td>0.850220</td>\n",
       "      <td>0.656654</td>\n",
       "      <td>0.904498</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>LogisticRegression(max_iter=500)</td>\n",
       "      <td>TransformStrategy_OrdinalEncoder(handle_unknow...</td>\n",
       "      <td>0.822978</td>\n",
       "      <td>0.545481</td>\n",
       "      <td>0.851828</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>KNeighborsClassifier(n_neighbors=10)</td>\n",
       "      <td>MeanTargetEncoder(target=target, alpha=5)</td>\n",
       "      <td>0.835877</td>\n",
       "      <td>0.608622</td>\n",
       "      <td>0.874924</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>KNeighborsClassifier(n_neighbors=10)</td>\n",
       "      <td>TransformStrategy_OneHotEncoder(handle_unknown...</td>\n",
       "      <td>0.830012</td>\n",
       "      <td>0.598894</td>\n",
       "      <td>0.866346</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>KNeighborsClassifier(n_neighbors=10)</td>\n",
       "      <td>TransformStrategy_OrdinalEncoder(handle_unknow...</td>\n",
       "      <td>0.830012</td>\n",
       "      <td>0.594483</td>\n",
       "      <td>0.866827</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>LGBMClassifier()</td>\n",
       "      <td>MeanTargetEncoder(target=target, alpha=5)</td>\n",
       "      <td>0.872056</td>\n",
       "      <td>0.710777</td>\n",
       "      <td>0.927866</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>LGBMClassifier()</td>\n",
       "      <td>TransformStrategy_OneHotEncoder(handle_unknown...</td>\n",
       "      <td>0.873683</td>\n",
       "      <td>0.715355</td>\n",
       "      <td>0.928402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>LGBMClassifier()</td>\n",
       "      <td>TransformStrategy_OrdinalEncoder(handle_unknow...</td>\n",
       "      <td>0.873161</td>\n",
       "      <td>0.714183</td>\n",
       "      <td>0.927786</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                   model  \\\n",
       "0    RandomForestClassifier(max_depth=5)   \n",
       "1    RandomForestClassifier(max_depth=5)   \n",
       "2    RandomForestClassifier(max_depth=5)   \n",
       "3       LogisticRegression(max_iter=500)   \n",
       "4       LogisticRegression(max_iter=500)   \n",
       "5       LogisticRegression(max_iter=500)   \n",
       "6   KNeighborsClassifier(n_neighbors=10)   \n",
       "7   KNeighborsClassifier(n_neighbors=10)   \n",
       "8   KNeighborsClassifier(n_neighbors=10)   \n",
       "9                       LGBMClassifier()   \n",
       "10                      LGBMClassifier()   \n",
       "11                      LGBMClassifier()   \n",
       "\n",
       "                                          transformer       acc        f1  \\\n",
       "0           MeanTargetEncoder(target=target, alpha=5)  0.851970  0.626713   \n",
       "1   TransformStrategy_OneHotEncoder(handle_unknown...  0.836952  0.550605   \n",
       "2   TransformStrategy_OrdinalEncoder(handle_unknow...  0.848991  0.612731   \n",
       "3           MeanTargetEncoder(target=target, alpha=5)  0.845429  0.639919   \n",
       "4   TransformStrategy_OneHotEncoder(handle_unknown...  0.850220  0.656654   \n",
       "5   TransformStrategy_OrdinalEncoder(handle_unknow...  0.822978  0.545481   \n",
       "6           MeanTargetEncoder(target=target, alpha=5)  0.835877  0.608622   \n",
       "7   TransformStrategy_OneHotEncoder(handle_unknown...  0.830012  0.598894   \n",
       "8   TransformStrategy_OrdinalEncoder(handle_unknow...  0.830012  0.594483   \n",
       "9           MeanTargetEncoder(target=target, alpha=5)  0.872056  0.710777   \n",
       "10  TransformStrategy_OneHotEncoder(handle_unknown...  0.873683  0.715355   \n",
       "11  TransformStrategy_OrdinalEncoder(handle_unknow...  0.873161  0.714183   \n",
       "\n",
       "         auc  \n",
       "0   0.908543  \n",
       "1   0.897631  \n",
       "2   0.902580  \n",
       "3   0.899764  \n",
       "4   0.904498  \n",
       "5   0.851828  \n",
       "6   0.874924  \n",
       "7   0.866346  \n",
       "8   0.866827  \n",
       "9   0.927866  \n",
       "10  0.928402  \n",
       "11  0.927786  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.preprocessing import MaxAbsScaler\n",
    "from lightgbm import LGBMClassifier\n",
    "from sklearn.base import clone\n",
    "\n",
    "reporter = col.Report(scorer=scorer)\n",
    "reporter.set_columns_to_show(['model', 'transformer'] + list(scorer.scoring_fcts.keys()))\n",
    "\n",
    "models = [\n",
    "    RandomForestClassifier(n_estimators=100, max_depth=5),\n",
    "    LogisticRegression(max_iter=500),\n",
    "    KNeighborsClassifier(n_neighbors=5),\n",
    "    LGBMClassifier()\n",
    "]\n",
    "\n",
    "models = [\n",
    "        RandomForestClassifier(n_estimators=100, max_depth=5),\n",
    "        LogisticRegression(max_iter=500),\n",
    "        KNeighborsClassifier(n_neighbors=10),\n",
    "        LGBMClassifier(),\n",
    "    ]\n",
    "\n",
    "transformers = [\n",
    "    col.MeanTargetEncoder(feature_selection),\n",
    "    col.TransformStrategy(feature_selection, OneHotEncoder(handle_unknown='ignore')),\n",
    "    col.TransformStrategy(feature_selection, OrdinalEncoder(handle_unknown='use_encoded_value', unknown_value=-1)),\n",
    "]\n",
    "\n",
    "\n",
    "for model in models:\n",
    "    print(f\"\\tevaluating encoders when fitting a {model}\")\n",
    "    for transformer in transformers:\n",
    "        pipe = col.CategoricalPipeline(features=feature_selection,\n",
    "                                       transformer=clone(transformer),\n",
    "                                       scaler=MaxAbsScaler(),\n",
    "                                       model=clone(model),\n",
    "                                      )\n",
    "\n",
    "        cv_score = col.cv_score(pipeline=pipe,\n",
    "                                data=df, \n",
    "                                kf=kf,\n",
    "                                scorer=reporter.scorer)\n",
    "\n",
    "        reporter.add_to_report(pipe.config, cv_score, show=False)\n",
    "        \n",
    "reporter.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### plot the 3 metrics for each model / encoder pair"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAACBkAAAGSCAYAAACsKouJAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAACK+klEQVR4nOzdeXyM9/r/8XeIrZRpnexEbI1YYisRpbYetautlipVsUsdtEQ5tRaRVmsJ1Sq1RGuL0lJLia2WWmo5lthSxZEoTigtkeX3h1/m28kMRsxkJvF6Ph55PMw99+dzX/fcV+LKzJXP7ZKYmJgmAAAAAAAAAAAAAACAR8jl6AAAAAAAAAAAAAAAAED2QJMBAAAAAAAAAAAAAACwCk0GAAAAAAAAAAAAAADAKjQZAAAAAAAAAAAAAAAAq9BkAAAAAAAAAAAAAAAArOL6qB2Sk5N1+/btrIgFAKxWsGBBubo+8kcYAAAAAAAAAAAAABt66Cd0ycnJ+uOPP2QwGOTi4pJVMQHAQ6WlpSkxMVHPPvssjQYAAADAU+j8+fOqXLmyIiMj9cYbbzz2+EqVKqlOnTqaPXu2HaIDAAAAACBne+jtEm7fvk2DAQCn4+LiIoPBwCorAAAAQA6wZMkSGQwGVa1a1WExXLp0SZMmTdKRI0ccFgMAAAAAANnFQ5sMJNFgAMAp8bMJAAAAyBmWLVsmX19fxcXF6eeff3ZIDP/9738VHh6uo0ePOuT4AAAAAABkJ49sMgAAAAAAALCH+Ph4bd++XaNGjVLx4sW1bNkyR4cEAAAAAAAegSYDAAAAAADgEMuXL1f+/PnVrFkztWvXTtHR0bp3757JPvHx8erevbuKFSsmPz8/DRw4UH/88YfZXM2bN1fz5s3Ntk+aNEkGg+GBMezYsUP//Oc/JUkDBgyQwWCQwWDQpEmTnuzkAAAAAADIoWgyAAAAAAAADrF06VK9+uqrKlSokNq3b6/r16/rxx9/ND5/584dtW7dWuvWrVO3bt00YsQI/frrr+rbt6/NYvD391dYWJgk6a233tKcOXM0Z84ctWzZ0mbHAAAAAAAgJ3HNzCDD/Eu2juOhEnv4ZOnx4LxWr16t7t27KzEx0dGhAAAAAACewIkTJ/Sf//xHw4cPlyRVrFhRAQEBWrZsmZo2bSpJWrBggWJjY/XZZ5+pU6dOkqSQkBC1aNHCZnG4u7urUaNGmjx5smrUqKGOHTvabG4AAAAAAHKiHLmSQb9+/WQwGBQaGmr23AcffCCDwZClbxo0b97cuNyipa9KlSplWSwZ9evXz+Jr8aBY582b54AoAQAAAAA5zdKlS1W4cGE1btzYuK1du3b64YcfdPPmTUnShg0b5Obmpg4dOhj3yZ07t/r06ZPl8QIAAAAAgPsytZJBdlCsWDGtWrVKkydPVsGCBSVJycnJWrp0qYoVK5alsSxevFhJSUmSpP/973+qVauWFi5cqKCgIEn33yB5HElJScqbN6/N48xo+vTpevXVV022FS5c2O7Htbesev0AAAAAAJalpaVpxYoVeumllxQfH2/cXqNGDd25c0dr1qxR165ddeHCBZUsWdLs9+YyZcpkdcgAAAAAAOD/y5ErGUhShQoVVKpUKa1atcq4bcOGDcqXL5/q1Kljsu/ixYsVFBQkDw8PVa9eXZGRkUpNTTU+P3PmTNWuXVve3t4KCAhQaGioyXL9UVFR8vHx0bZt2xQcHCxvb2+1aNFCv/76qyTpueeek4eHhzw8POTm5may7R//+IfGjBmjwMBAeXp6qlq1apo2bZrJ8dNXG/j0009Vvnx5lS9fXpK0f/9+vfzyy/Lw8FDdunW1ceNGGQwG7dixwzj25MmTev3111WsWDGVKVNGPXv2VEJCgiRp0qRJ+vrrr7VhwwbjSgV/H1ukSBFj3OlfBQoUsOqc//6aN2rUSJ6enipZsqQ6duyoO3fuSJISExPVt29flShRQp6enmrdurVOnDhhMv7rr79WxYoV5eXlpY4dO+rKlStm1/qHH35QvXr15OHhocDAQI0fP97Y1CFJlSpV0qRJkzRgwAD5+vqqV69eZnMAAAAAALLOjh07dPHiRf3www+qXLmy8at169aSpGXLlkm634zg4uJiNj4tLc1sm6X9JCklJcWGkQMAAAAAgBy7koEkvfnmm4qKilLXrl0l3W8meOONN0w+CF+wYIEmTpyoKVOmqHLlyjpx4oQGDRqkPHnyqHfv3pKkXLlyadKkSfLz89OFCxc0bNgwDRs2TJ9//rlxnrt372rq1KmaOXOm8uXLp379+mnIkCGKjo5+aIypqany8vLSV199paJFi+rgwYMaNGiQnnvuOXXr1s24308//aTChQtrxYoVSktL061bt9SxY0c1aNBAc+bMUXx8vEaMGGEyd3x8vJo1a6Y333xT48eP17179zR+/Hh17txZP/74o0JDQ3Xq1Cn973//05w5cyTdb36w1qPO+ccff1SXLl00ePBgRUZGKjk5WTExMcYGin79+unMmTNasmSJDAaDxo8fr/bt22v//v0qUKCA9u/fr/79+2vkyJF67bXXtGPHDo0bN84khs2bN6t3796aNGmSXnrpJV24cEFDhgzR3bt3NWHCBON+s2bN0rvvvqutW7dafDMKAAAAAJB1li1bpueee04zZswwe27btm368ssv9d///le+vr46evSoUlJSTFYzOHv2rNk4g8Fg1vguSb/99tsj43lQgwIAAAAAADCXo5sMOnTooH//+986e/asChUqpM2bN2vKlCmaOHGicZ+IiAiNHTvW+NcSfn5+iouL05dffmlsMujfv79x/xIlSmjcuHHq0qWLPvvsM+XKdX8xiOTkZH300UcqW7asJCk0NFQDBgxQamqqcR9L8uTJo5EjR5rMf/jwYa1cudKkySBfvnzGD/Mlaf78+UpJSdGMGTNUoEABBQQEaOjQoSZ/pf/ll1+qYsWKGjt2rHHbnDlz5Ofnp19++UXVq1dX/vz5lS9fPnl4eJjF1qdPH5Nzl6SNGzeqQoUKVp1zRESEWrdurVGjRhnHV6xYUdL9N4R++OEHrV27Vi+99JIxtkqVKmn58uXq1q2bPvvsM9WrV0/vvvuupPvLYR48eFCLFi0yzvfRRx8pNDTU2EhSsmRJjRkzRn369NH48eONbxTVrl1bgwYNeuB1AAAAAABkjfTbITRp0kQtWrQwe75ixYr64osvtGLFCjVu3FhbtmzR8uXL1alTJ0n3VyZIb5T/u1KlSmnjxo26cuWK3N3dJUn//e9/tW7dukfG9Mwzz0iSyaqFAAAAAADAshzdZGAwGNSiRQstXrxYRYoUUZ06dVS8eHHj81evXtXFixc1ePBgDR061Lg9OTnZ5K/dt23bpk8++USnTp3SzZs3lZKSoqSkJCUkJMjLy0vS/SaA9A/bJcnT01P37t3TjRs3Hrk6wLx587Rw4UJduHBBd+7c0b1790zilKSAgABjg4EknTp1SgEBAcbbF0jSiy++aDLm8OHD2rVrl3x8fMyOGRcXp+rVqz80rnHjxumVV14x2VasWDHjvx91zkeOHFGXLl0szh0bG6tcuXKpZs2axm1FihRR+fLldfLkSeM+TZo0MRlXo0YNkyaDw4cP6+DBg5o2bZpxW2pqqv766y8lJCTI09NTklS1atWHnisAAAAAIGv88MMPunnzppo1a2bxeT8/PwUEBGjp0qX68ccf9eWXXyo0NFRHjhyRn5+f1qxZoz/++MNsXLdu3TRz5ky1adNG3bt3140bNzRv3jyVLl1ahw8ffmhMpUuXVuHChTVv3jwVKlRIhQoVUkBAgPF2hQAAAAAA4P/k6CYDSeratav69eunggUL6v333zd5Ln3Z/qlTpyooKMji+N9++00dO3ZUt27d9P777+v555/X4cOH1bNnTyUlJRn3c3U1fSnT/4I+/RgPEh0drREjRmj8+PGqWbOmChcurC+++ELff/+9yX4FCxY0efyg+1JmPL/GjRub3DYgnZub20PHSpKHh4dKlSr1wOcze86S5ftnZpzHmtsapKamavjw4XrttdfMnvvHP/5h/HfG1w8AAAAA4BhLly5V3rx51ahRowfu06RJE33yySc6e/asVq9erbCwMC1YsEB58uRRixYt1LdvX9WpU8dkTJkyZfTll19qwoQJGjlypEqWLKnx48fr9OnTj2wyyJcvn+bMmaPx48fr3Xff1b179zR8+HCaDAAAAAAAsCDHNxnUq1dPefLk0bVr19S8eXOT59zd3eXt7a24uDh17tzZ4vhffvlFSUlJmjRpkvH+j+vXr7dZfLt371b16tWNt2aQ7q8y8Cj+/v765ptv9NdffxlXMzhw4IDJPpUrV9aqVatUvHhx5cmTx+I8efPmVUpKyhOcwYMFBgZq27Zt6t69u9lz5cqVU2pqqn7++Wfj7RJu3ryp48ePG1c/KFeunPbv328yLuPjypUr69SpUw9thgAAAAAAOI9vvvnmkfuMHj1ao0ePNj5euHCh2T6Wbm3Qpk0btWnTxmz7iBEjTB4fPXrUbJ+mTZuqadOmj4wNAAAAAICnXS5HB2BvLi4u+umnn3T48GGT2w2kCwsL0/Tp0xUZGanTp0/r+PHj+vrrrzV16lRJ95dMTE1N1axZs/Trr79qxYoV+uyzz2wWX5kyZXTkyBFt2rRJZ8+e1ZQpU7Rr165HjuvQoYNy586tQYMG6eTJk9q6dasx5vSVAEJCQnTz5k316NFD+/fv16+//qqtW7dq0KBBxqUlfX19deLECZ0+fVrXrl3TvXv3jMe4ceOGEhISTL5u3bpl9bkNHTpU3377rSZMmKCTJ0/qxIkTioyM1J9//qnSpUurWbNmGjx4sHbt2qVjx46pd+/eevbZZ9WhQwdJUp8+fYzndfbsWS1YsMBshYdhw4ZpxYoV+vDDD3X8+HGdOnVKq1ev1gcffGB1nAAAAAAAAAAAAAAA62RqJYPEHj62jsOunn322Qc+161bNz3zzDOaPn26xo0bp/z58ysgIEC9evWSJFWsWFGTJ0/WtGnT9OGHH6pmzZoaP368evToYZPYevTooaNHjyokJERpaWlq1aqVBgwYoMWLFz90XKFChfTNN99oyJAhevnll+Xv76/hw4ere/fuyp8/vyTJy8tLGzZs0NixY9WuXTvdvXtXxYoVU4MGDYwNF927d9fOnTvVoEED3bp1S999953q1q0rSXrnnXfMjvvuu+9q1KhRVp1b48aNtXjxYoWHh2v69OkqVKiQatasqZ49e0qSZs2apbCwMHXu3Fl3795VUFCQVqxYYVyZoUaNGpoxY4YmT56sKVOmqE6dOgoLC9OwYcOMx2jUqJGWLVumiIgIzZw5U66uripdurRxNQQAAAAAAAAAAAAAgO24JCYmPvDG9zdu3FCRIkWyMh48gbVr16pr1646c+aMihYt6uhwALvjZxQAAAAAAAAAAACQtTK1kgGcw5IlS+Tn5ycfHx+dOHFCI0aMUJMmTWgwAAAAAAAAAAAAAADYBU0G2djvv/+uSZMmKSEhQe7u7nr11Vc1ZswYR4cFAAAAAAAAAAAAAMihuF0CgGyLn1EAAAAAAAAAAABA1srl6AAAAAAAAAAAAAAAAED2QJMBAAAAAAAAAAAAAACwCk0GAAAAAAAAAAAAAADAKjQZAAAAAAAAAAAAAAAAq9BkAAAAAAAAAAAAAAAArEKTAQAAAAAAAAAAAAAAsIprZgYV6l7fxmE83K0FW7P0ePby1Vdf6aOPPtKlS5c0bNgwjRgxwtEhAXZ37do1lS5dWt99953q1q3r6HAAAAAAAAAAAAAAPIEct5KBwWB46Fe/fv0cEldiYqLeffddhYaG6sSJEwoNDc3yGH799Vf17t1b5cuXl7u7u8qVK6fXX39dhw8fNu5jMBi0evVqmxzv/PnzMhgM+uWXX2wyn62dOnVKPXv2VNmyZeXu7q7AwECNHDlSiYmJNj/Wjh07ZDAYdO3aNbPngoODNWnSJKvnioqKko+Pj9n25s2bW8z5t99++4liBwAAAAAAAAAAAIB0mVrJwJnFxsYa/71hwwa98847Jtvy589vsv+9e/eUJ08eu8f122+/KTk5Wa+++qo8PT0zPU9SUpLy5s372OPu3bunNm3aqGTJkpo/f768vLx0+fJlxcTEPPaH6snJycqdO7dcXFweOw5nceDAAb322muqXbu2oqKi5OXlpWPHjmn06NHatGmTNm7cKIPB4OgwH9sbb7yhDz74wGRbxpzPjjKb9wAAAAAAAAAAAABsK8etZODh4WH8KlKkiMm2O3fuqESJElqxYoVatmwpT09PzZ8/X9evX1fPnj1Vvnx5eXp6qlatWlq8eLHJvM2bN9fQoUM1btw4lSpVSmXKlNGoUaOUmppq3GfNmjWqXbu2PD095efnp2bNmunKlSuKiorSyy+/LEmqUqWKDAaDzp8/L0maP3++qlatKjc3N1WtWlULFiwwOa7BYNAXX3yhrl27ytvbW+PGjdOkSZMUHBysJUuWqFKlSvLx8VH//v2VlJSkuXPnqkKFCipZsqTef/99Y3wnTpxQXFycPvroIwUFBcnX11dBQUEKCwtTvXr1JEmVKlWSJHXv3l0Gg8H4OP14UVFRqlKlitzd3XX79m39+OOPatq0qUqUKCE/Pz+1bdvWpKGjcuXKkqQGDRrIYDCoefPmxucWL16soKAgeXh4qHr16oqMjDR5Lc+cOaNmzZrJw8NDL774ojZu3CgfHx9FRUVJklq2bKn33nvP5LW6efOmvLy8tGbNmofmSFpamgYOHKhSpUrp66+/Vs2aNVW8eHE1adJE3377rS5evKgJEyYY969UqZIiIiL0r3/9S8WLF1f58uU1ffp0kzlv3LihQYMGqUyZMipWrJiaNWuW6RUcEhMT1bdvX5UoUUKenp5q3bq1Tpw4Ien+iggDBgzQ7du3jSsV/H0VhGeeecbke+Dv3wfpK0usXr1ar732mry8vBQUFKSYmBiT4586dUqdOnWSr6+vfHx89M9//lPHjh2TJKWmpmrKlCmqUKGC3N3dVbt2ba1du9Zk/MGDB1WvXj15eHiobt262r9/v9k5njx5Uq+//rqKFSumMmXKqGfPnkpISDA+369fP3Xs2FGffvqpypcvr/Lly2fqtQQAAAAAAAAAAABgWzmuycAaY8eOVUhIiPbs2aPmzZvrzp07qly5sr755hvt2bNHffv21eDBg7Vt2zaTccuXL1fu3Lm1ceNGRUREaPbs2YqOjpYkJSQkqGfPnurcubP27t2rdevWqVOnTpKktm3bauXKlZKkLVu2KDY2VsWKFdN3332n9957T/369dPu3bvVt29fDR06VD/88IPJccPDw9W4cWPt2rVLISEhku6vjLBu3TotXbpUCxcu1OrVq9WlSxcdPHhQ0dHRmj59uj7//HN99913kqR//OMfypUrl9asWaPk5GSLr0v6h83Tp09XbGysyYfP58+f14oVK/TVV19p586dyp8/v27fvq2+fftqy5Yt+v7771W4cGF16tRJSUlJxnOVpJUrVyo2NtbYuLFgwQKNHz9e77//vvbu3asJEyZo2rRpmjt3rqT7H2R37dpVrq6u2rRpk2bNmqXw8HDdvXvXGE/37t21fPlyk20rV65UwYIF1bRp04de/yNHjujEiRMaOHCgcuUy/Rbw8vJS+/bttWLFCqWlpRm3z5o1S+XLl9e2bds0aNAgffDBB/r5558l3W9a6Nixoy5fvqylS5dq+/btql27tlq1aqX4+PiHxmJJv379dODAAS1ZskSbN29WgQIF1L59e/31118KCgrSpEmT9Mwzzyg2NlaxsbGPfeuNCRMmqE+fPtq5c6eqVq2qt99+W7du3ZIkXb58WU2aNJGLi4tWrVqlbdu2KSQkRCkpKZKk2bNna8aMGRozZox27dql5s2b680339SRI0ckSbdv39brr78uPz8/xcTEaMyYMfr3v/9tcvz4+Hg1a9ZMAQEB2rx5s7799lvdunVLnTt3Nmk0+emnn3Ts2DGtWLHCZrfwAAAAAAAAAAAAAPBkctztEqzRu3dvtW7d2mTbO++8Y/z3W2+9pe3bt2vFihXGv/KXJH9/f40cOVKSVKZMGS1YsEDbtm1T+/btdfnyZd27d0+tW7eWr6+vJJn89fXzzz8vSSpatKg8PDwkSTNnzlTHjh3Vu3dv45yHDh3StGnTTD4ob9Omjbp162YSb0pKiiIjI1WkSBGVL19ejRo10k8//aQTJ04ob9688vf3V1BQkHbu3KnWrVvL29tb4eHhGj16tCIiIlS5cmXVrl1b7dq1U0BAgKT7jQiSVKRIEWOM6ZKSkjRnzhy5u7sbt2V8DSMjI1W8eHEdOHBAwcHBKlq0qPHc/z5fRESExo4daxzv5+enuLg4ffnll+rdu7diYmJ0+vRpRUdHy9vbW5I0ceJEvfrqq8Y5WrZsqWHDhun7779Xu3btJN1fHaFTp06PvP3F2bNnJUkvvPCCxef9/f2VmJioq1evys3NTZLUsGFD43Xq06eP5syZo23btqlmzZravn27jh49qjNnzqhAgQKSpFGjRmn9+vVaunSpBg0aZJw7MDDQ7Hh//fWXSWw//PCD1q5dq5deekmSNGfOHFWqVEnLly9Xt27dVLhwYbm4uJhdI0n66quvtGTJEpNt6U016fr372/Mrw8++EDffPONjh49quDgYM2dO1fPPPOMFixYYLw9QZkyZYxjZ86cqYEDB6pDhw6SpJEjR2rXrl2aOXOmPv/8cy1fvlxJSUmKjIxUoUKFVL58eQ0dOlR9+vQxzvHll1+qYsWKGjt2rHHbnDlz5Ofnp19++UXVq1eXJOXLl08zZ85Uvnz5LF4nAAAAAAAAAAAAAFnvqWwyqFq1qsnjlJQUffLJJ4qOjtbly5eVlJSkpKQk1alTx2S/ChUqmDz29PTU77//Lun+kvr169dX7dq11aBBA9WvX1+tW7c2fnBvSWxsrN544w2TbcHBwWYrGWSMV5KKFStmXAZfktzd3VWmTBmT+9a7u7sb45OkXr16qVOnTtqxY4cOHDigdevW6dNPP9XMmTONqy48iLe3t0mDgSTFxcXpww8/1P79+3Xt2jWlpqYqNTVVFy9efOA8V69e1cWLFzV48GANHTrUuD05Odm4csCpU6fk5eVlbDCQpGrVqpmsOpAvXz517NhRixcvVrt27XTy5EkdOHBAkZGRDz2Pv3NxcbG4PT2Ovz//sGt/+PBh/fnnnyYfxkvSnTt3FBcXZ7Ltu+++k8FgMNmW/oG9dD8ncuXKpZo1axq3pTeSnDx58pHn1KZNG4WFhZlsS2/2sHQuXl5ekmQ8lyNHjig4ONgkj9LdvHlTly9fVq1atUy2BwcHa+PGjcb4K1SooEKFChmf//u5SPdfr127dsnHx8fsGHFxccYmg4CAABoMAAAAAAAAAAAAACfzVDYZFCxY0OTxjBkzNHPmTE2ePFnly5dXoUKFNG7cOJMP6CWZ/YW8i4uL8QPp3Llza9WqVdq3b5+2bNmiRYsWaezYsVq7dq0qVar0wFgsfdCdcVvGeB8Ui6urq9m29GXu0z377LNq1qyZmjVrplGjRqlt27b68MMPH9lkYCmGTp06ycvLS59++qm8vLzk6uqqoKAg4+0SLElfDn/q1KkKCgqyuM/fb1PwMN26ddNLL72kCxcuaNGiRapZs6bKlSv3yHGlS5eWJJ08edLiygKnTp2SwWAw+XD+Ydc+NTVV7u7uZs0h0v3X++9KlChh9qH/3z/Qf9i5P6gp4u+KFCmiUqVKPXSfv59L+pzpx7X2tX9QbNaMT01NVePGjTVhwgSz59JXjpAs5xwAAAAAAAAAAAAAx8r16F1yvt27d6tJkybq1KmTAgMDVbJkSZ05c+ax53FxcVHNmjUVFhammJgYeXl5adWqVQ/c39/fX3v27DGLxZoPym3BxcVFZcuW1e3bt43b8uTJY9aYYMn169cVGxurIUOGqH79+vL399cff/yh5ORk4z7pH57/fT53d3d5e3srLi5OpUqVMvuS7r8uly9f1uXLl43jfvnlF2ODQrqAgAC9+OKLWrBggZYtW6auXbtadd6BgYHy9/dXZGSk2ZyXL1/W8uXL1b59e6s+1JekypUr68qVK8qVK5fZ+fz9Q3NrlCtXTqmpqfr555+N227evKnjx4/L399f0v3X1ZprlBmVK1fW7t27LTaKFC5cWF5eXhZzNj22cuXK6fjx4yY5tW/fPrNjnDx5UsWLFzd7vTI2ZQAAAAAAAAAAAABwLjQZ6P4957dv367du3fr1KlTeu+99/Tbb7891hz79u1TRESEDh48qAsXLmjdunW6dOmS8cNXS0JDQ7V06VJ98cUXOnv2rObMmaPly5frnXfeedJTMnPkyBF17txZq1ev1smTJ3Xu3DktXLhQUVFRatGihXE/X19fbdu2TQkJCUpMTHzgfOl/6b9w4UKdO3dOO3fu1JAhQ0xWU3Bzc1OBAgW0efNmXblyRTdu3JAkhYWFafr06YqMjNTp06d1/Phxff3115o6daokqUGDBipbtqz69euno0ePat++fRo5cqRcXV3NPvjv1q2bpk+frj///FNt2rSx6rVwcXHRzJkzdebMGXXu3Fk///yzLl68qA0bNui1115T8eLFNWrUKGtfWtWvX1+1atVSly5dtGnTJv3666/6+eefNXHiRO3atcvqeaT7qyw0a9ZMgwcP1q5du3Ts2DH17t1bzz77rPG2Cr6+vrpz545iYmJ07do1/fnnn8bxf/75pxISEky+/ve//1l9/J49e+r27dt66623dPDgQZ07d04rVqzQkSNHJN3P2ZkzZ2rFihU6c+aMPvzwQ+3evVsDBw6UJLVv316urq4aOHCgTpw4oZiYGH388ccmxwgJCdHNmzfVo0cP7d+/X7/++qu2bt2qQYMG6Y8//nis1wsAAAAAAAAAAABA1srU7RJuLdhq4zAc67333tP58+fVoUMH5c+fX126dFGHDh108uRJq+coXLiw9u7dq88//1w3btyQj4+P3nvvPXXs2PGBY1q0aKEpU6ZoxowZGjFihIoXL66PP/5YTZs2tcVpmfDx8ZGfn5/Cw8N14cIFpaamqlixYho4cKAGDx5s3G/ChAkaOXKkKlSoIC8vLx09etTifLly5dK8efMUFham4OBglSpVShMmTFC3bt2M+7i6uio8PFxTpkxReHi4goODtXbtWnXr1k3PPPOMpk+frnHjxil//vwKCAhQr169jHMvXrxYoaGhatSokXx9fTVhwgS9+eabyp8/v0kcbdu2VVhYmFq3bv1YfwVfo0YNbd68WVOmTFGXLl1048YNeXp6qmXLlho2bJgMBoPVc7m4uGjZsmWaMGGCBg0apN9//13u7u4KCgpS586drZ4n3axZsxQWFqbOnTvr7t27CgoK0ooVK1SgQAFJUlBQkN5++2317NlT169f1/DhwzVixAhJUlRUlKKiokzmq1WrltavX2/Vsb29vbVu3Tp98MEHatmypVxcXFS+fHl9+umnkqS+ffvq1q1bGj16tK5cuaKyZctq4cKFxttOFCpUSEuXLtWQIUNUr149lS1bVmPGjDF5Hby8vLRhwwaNHTtW7dq10927d1WsWDE1aNBA+fLle+zXCwAAAAAAAAAAAEDWcUlMTHzgTdRv3LihIkWKZGU8gEVHjx5V3bp1tXXrVlWpUsW4/fLly6pYsaLWrl2rWrVqOS5AOAQ/owAAAAAAAAAAAICslamVDAB7++6771SwYEGVKlVKv/32m0aOHKmKFSuqcuXKkqR79+4pPj5e48ePV2BgIA0GAAAAAAAAAAAAAJAFaDKAU7p165bGjBmjS5cuyWAwqE6dOpo4caJcXFwkSXv27FHLli1VunRpzZ8/32Tsrl271KFDhwfOfenSJbvGDgAAAAAAAAAAAAA5FbdLQI7z119/6fLlyw98vlSpUlkYDeyJn1EAAAAAAAAAAABA1mIlA+Q4BQoUoJEAAAAAAAAAAAAAAOwgl6MDAAAAAAAAAAAAAAAA2QNNBgAAAAAAAAAAAAAAwCo0GQAAAAAAAAAAAAAAAKvQZAAAAAAAAAAAAAAAAKxCkwEAAAAAAAAAAAAAALCKa2YG3d7SxNZxPFTBhuuz9Hj28tVXX+mjjz7SpUuXNGzYMI0YMcLRIcGGVq9ere7duysxMVGSFBUVpWHDhunSpUs2PU6lSpXUu3dvhYaG2nReW/vll1/UoEEDHT58WCVKlHB0OAAAAAAAAAAAAABsIMetZGAwGB761a9fP4fElZiYqHfffVehoaE6ceKEQz4g/vXXX9W7d2+VL19e7u7uKleunF5//XUdPnzYuI/BYNDq1attcrzz58/LYDDol19+scl8tnbq1Cn17NlTZcuWlbu7uwIDAzVy5Ehjk8CTatu2rQ4dOmSTuR6mUqVKFnN9zJgxdj82AAAAAAAAAAAAgKdLplYycGaxsbHGf2/YsEHvvPOOybb8+fOb7H/v3j3lyZPH7nH99ttvSk5O1quvvipPT89Mz5OUlKS8efM+9rh79+6pTZs2KlmypObPny8vLy9dvnxZMTExj/2henJysnLnzi0XF5fHjsNZHDhwQK+99ppq166tqKgoeXl56dixYxo9erQ2bdqkjRs3ymAwWBxr7TUoUKCAChQoYOPILRs2bJh69uxpsq1gwYJZcmx7ymy+AwAAAAAAAAAAALCPHLeSgYeHh/GrSJEiJtvu3LmjEiVKaMWKFWrZsqU8PT01f/58Xb9+XT179lT58uXl6empWrVqafHixSbzNm/eXEOHDtW4ceNUqlQplSlTRqNGjVJqaqpxnzVr1qh27dry9PSUn5+fmjVrpitXrigqKkovv/yyJKlKlSoyGAw6f/68JGn+/PmqWrWq3NzcVLVqVS1YsMDkuAaDQV988YW6du0qb29vjRs3TpMmTVJwcLCWLFmiSpUqycfHR/3791dSUpLmzp2rChUqqGTJknr//feN8Z04cUJxcXH66KOPFBQUJF9fXwUFBSksLEz16tWTdP8v4iWpe/fuMhgMxsfpx4uKilKVKlXk7u6u27dv68cff1TTpk1VokQJ+fn5qW3btiYNHZUrV5YkNWjQQAaDQc2bNzc+t3jxYgUFBcnDw0PVq1dXZGSkyWt55swZNWvWTB4eHnrxxRe1ceNG+fj4KCoqSpLUsmVLvffeeyav1c2bN+Xl5aU1a9Y8NEfS0tI0cOBAlSpVSl9//bVq1qyp4sWLq0mTJvr222918eJFTZgwwbh/pUqVNGnSJA0YMEC+vr7q1auXJOnrr79WxYoV5eXlpY4dO+rKlSsmx4mKipKPj4/xcfrruHLlSlWpUkXFihVTly5ddO3aNeM+Bw8eVJs2bVSqVCljTD///PNDz0eSnn32WZPc9/DwUKFChSRJO3bskMFg0LZt29SoUSN5eXmpfv36Zqss7Nu3Ty1btpS3t7d8fX3VqlUrXb58WZJ09+5dhYWFqWzZsvLw8NArr7yi3bt3m4z/8ccfVaNGDXl4eKhp06Y6c+aMWZx79+5Vs2bN5OXlpYCAAA0ZMkQ3b940Pt+8eXMNGTJEo0aNUunSpfXqq68+8twBAAAAAAAAAAAAZJ0c12RgjbFjxyokJER79uxR8+bNdefOHVWuXFnffPON9uzZo759+2rw4MHatm2bybjly5crd+7c2rhxoyIiIjR79mxFR0dLkhISEtSzZ0917txZe/fu1bp169SpUydJ95fNX7lypSRpy5Ytio2NVbFixfTdd9/pvffeU79+/bR792717dtXQ4cO1Q8//GBy3PDwcDVu3Fi7du1SSEiIpPsrI6xbt05Lly7VwoULtXr1anXp0kUHDx5UdHS0pk+frs8//1zfffedJOkf//iHcuXKpTVr1ig5Odni6xITEyNJmj59umJjY42Ppfu3PlixYoW++uor7dy5U/nz59ft27fVt29fbdmyRd9//70KFy6sTp06KSkpyXiukrRy5UrFxsYaGzcWLFig8ePH6/3339fevXs1YcIETZs2TXPnzpUkpaamqmvXrnJ1ddWmTZs0a9YshYeH6+7du8Z4unfvruXLl5tsW7lypQoWLKimTZs+9PofOXJEJ06c0MCBA5Url+m3gJeXl9q3b68VK1YoLS3NuH3WrFl64YUXtHXrVn3wwQfav3+/+vfvr7feeks7duxQkyZNNHHixIceV7p/3aKjo7V48WJFR0fryJEjGj9+vPH5P/74Qx07dtQPP/ygzZs3q1KlSurQoYNJI0JmjR07VqNHj9a2bdv0/PPPq3fv3sZzPHr0qFq2bKlSpUpp/fr12rRpk9q0aWPMlQ8++ECrVq3SzJkztX37dpUvX17t27dXfHy8JOnixYt64403VL9+fe3YsUO9e/fW6NGjTY5/7NgxtW3bVk2bNtXOnTu1aNEiHT16VAMHDjTZb9myZUpLS9MPP/ygzz777InPGwAAAAAAAAAAAIDt5LjbJVijd+/eat26tcm2d955x/jvt956S9u3b9eKFSuMf+UvSf7+/ho5cqQkqUyZMlqwYIG2bdum9u3b6/Lly7p3755at24tX19fSVL58uWNY59//nlJUtGiReXh4SFJmjlzpjp27KjevXsb5zx06JCmTZtm8kF5mzZt1K1bN5N4U1JSFBkZqSJFiqh8+fJq1KiRfvrpJ504cUJ58+aVv7+/goKCtHPnTrVu3Vre3t4KDw/X6NGjFRERocqVK6t27dpq166dAgICJN1vRJCkIkWKGGNMl5SUpDlz5sjd3d24LeNrGBkZqeLFi+vAgQMKDg5W0aJFjef+9/kiIiI0duxY43g/Pz/FxcXpyy+/VO/evRUTE6PTp08rOjpa3t7ekqSJEyea/FV7y5YtNWzYMH3//fdq166dpPurI3Tq1OmRt784e/asJOmFF16w+Ly/v78SExN19epVubm5SZJq166tQYMGGfcJCQlRvXr19O6770q6f+0OHjyoRYsWPfTYycnJmjVrlnGVjbfeesu4OoMkk3yTpClTpmjNmjX68ccf1bFjxwfOO378eE2ePNlk25dffqkmTZoYH48cOdK4osawYcPUpEkT/fe//5WPj4+mT5+uihUratq0aSavgyTdvn1b8+bN0/Tp043X4JNPPtH27ds1d+5cjRo1SvPmzVOxYsU0ZcoUubi46IUXXtCZM2f04YcfGuebPn262rRpo9DQUOO2jz/+WC+//LJ+//1342vt6+trMg4AAAAAAAAAAACA83gqmwyqVq1q8jglJUWffPKJoqOjdfnyZSUlJSkpKUl16tQx2a9ChQomjz09PfX7779Lur+kfv369VW7dm01aNBA9evXV+vWrY0f3FsSGxurN954w2RbcHCw2UoGGeOVpGLFihk/qJYkd3d3lSlTxuT+9e7u7sb4JKlXr17q1KmTduzYoQMHDmjdunX69NNPNXPmTOOqCw/i7e1t0mAgSXFxcfrwww+1f/9+Xbt2TampqUpNTdXFixcfOM/Vq1d18eJFDR48WEOHDjVuT05ONv5V/alTp+Tl5WVsMJCkatWqmaw6kC9fPnXs2FGLFy9Wu3btdPLkSR04cECRkZEPPY+/c3Fxsbg9PY6/P5/xGsTGxpp8gC9JNWrUeGSTQfHixU2um6enp65evWp8/Pvvv+vDDz/Ujh079PvvvyslJUV//fXXQ19TSRowYIDefPNNk20ZG0X+nr+enp7G4/n4+OjIkSNq0aKFxbnj4uJ079491apVy7gtd+7cqlmzpk6ePCnp/uvx4osvmrxmNWvWNJnn8OHDOnfunFatWmXclv5ax8XFGZsMqlSp8tBzBQAAAAAAAAAAAOA4T2WTQcGCBU0ez5gxQzNnztTkyZNVvnx5FSpUSOPGjTP5gF6S2V/Iu7i4GD8kzZ07t1atWqV9+/Zpy5YtWrRokcaOHau1a9eqUqVKD4zF0gfdGbdljPdBsbi6upptS0lJMdn27LPPqlmzZmrWrJlGjRqltm3b6sMPP3xkk4GlGDp16iQvLy99+umn8vLykqurq4KCgoy3S7AkNTVVkjR16lQFBQVZ3Ofvtyl4mG7duumll17ShQsXtGjRItWsWVPlypV75LjSpUtLkk6ePKnAwECz50+dOiWDwWBciUEyP39rY8zI0nVLf00kqV+/frpy5YomTpwoX19f5cuXT61atXroayrdXy2iVKlSVh87PcfSz+Nh52Op6eJB8zxMamqqunXrpv79+5s95+XlZfy3pVwDAAAAAAAAAAAA4BxyPXqXnG/37t1q0qSJOnXqpMDAQJUsWVJnzpx57HlcXFxUs2ZNhYWFKSYmRl5eXiZ/tZ2Rv7+/9uzZYxaLNR+U24KLi4vKli2r27dvG7flyZPHrDHBkuvXrys2NlZDhgxR/fr15e/vrz/++EPJycnGfdJXVfj7fO7u7vL29lZcXJxKlSpl9iXdf10uX76sy5cvG8f98ssvJh/GS1JAQIBefPFFLViwQMuWLVPXrl2tOu/AwED5+/srMjLSbM7Lly9r+fLlat++/QNXOpCkcuXKaf/+/SbbMj7OjD179qh379569dVXFRAQoEKFCikhIeGJ532UypUra/v27RafK1WqlPLmzavdu3cbt6WkpOjnn3823lKhXLlyOnDggEmzwb59+8yOceLECYvXvUCBAnY4KwAAAAAAAAAAAAC2RpOBpDJlymj79u3avXu3Tp06pffee0+//fbbY82xb98+RURE6ODBg7pw4YLWrVunS5cuGT+EtSQ0NFRLly7VF198obNnz2rOnDlavny53nnnnSc9JTNHjhxR586dtXr1ap08eVLnzp3TwoULFRUVZbJMvq+vr7Zt26aEhAQlJiY+cL70v/RfuHChzp07p507d2rIkCEmqym4ubmpQIEC2rx5s65cuaIbN25IksLCwjR9+nRFRkbq9OnTOn78uL7++mtNnTpVktSgQQOVLVtW/fr109GjR7Vv3z6NHDlSrq6uZh/8d+vWTdOnT9eff/6pNm3aWPVauLi4aObMmTpz5ow6d+6sn3/+WRcvXtSGDRv02muvqXjx4ho1atRD5+jTp4+2bt2qqVOn6uzZs1qwYIG+//57q47/MKVLl9ayZct08uRJHTx4UG+//bbJLTAe5I8//lBCQoLJV/rrbY3Q0FAdOXJEgwYN0tGjR3X69GktXLhQFy5cUMGCBfX2229r7Nix2rhxo7G55Pfff1dISIgkqUePHvrtt98UFham06dPa/Xq1Zo/f77JMQYNGqSDBw9q8ODBxlsnrF+/Xv/6178e6zUCAAAAAAAAAAAA4DiZul1CwYbrbR2HQ7333ns6f/68OnTooPz586tLly7q0KGD8X7z1ihcuLD27t2rzz//XDdu3JCPj4/ee+89dezY8YFjWrRooSlTpmjGjBkaMWKEihcvro8//lhNmza1xWmZ8PHxkZ+fn8LDw3XhwgWlpqaqWLFiGjhwoAYPHmzcb8KECRo5cqQqVKggLy8vHT161OJ8uXLl0rx58xQWFqbg4GCVKlVKEyZMULdu3Yz7uLq6Kjw8XFOmTFF4eLiCg4O1du1adevWTc8884ymT5+ucePGKX/+/AoICFCvXr2Mcy9evFihoaFq1KiRfH19NWHCBL355pvKnz+/SRxt27ZVWFiYWrdurWeffdbq16NGjRravHmzpkyZoi5duujGjRvy9PRUy5YtNWzYMBkMhkeOnzFjhiZPnqwpU6aoTp06CgsL07Bhw6yOwZKZM2fqX//6l+rXry9PT0+FhYXp2rVrjxw3ZcoUTZkyxWTb66+/rs8//9yq4wYGBurbb7/VuHHj9M9//lN58+ZV1apV1bhxY0nS2LFjJUkDBgzQjRs3FBgYqBUrVsjT01OSVLx4cS1atEgjR47UV199pSpVqmj06NHq3bu38RgVK1bUunXrNGHCBLVo0UIpKSny8/NT8+bNrYoRAAAAAAAAAAAAgOO5JCYmPvBm6jdu3FCRIkWyMh7AoqNHj6pu3braunWrqlSpYtx++fJlVaxYUWvXrlWtWrUcFyAcgp9RkKRZs2bp/ffflyRt3rxZ1atXf+C+f/75pxYvXqx169bp+PHj+t///qf8+fPLz89PdevWVdeuXVW+fHmzcffu3dPy5cu1evVqHTlyRNeuXVOePHlUvHhxBQcHq3PnzqpZs6bdzhEAAACO86gm9IkTJ6p///6SpMTERC1cuFBHjx7VkSNHdObMGaWkpGjFihV65ZVXsiBaAAAAOAtr3recNGmSwsPDNXz4cI0YMcLiPDt27FDLli310ksvae3atWbP894lAEfI1EoGgL199913KliwoEqVKqXffvtNI0eOVMWKFVW5cmVJ9//TjI+P1/jx4xUYGEiDAfAUW7hwoVxcXJSWlqb58+c/sMng4MGD6tatmy5evChPT081aNBAPj4++vPPP3Xy5EnNnTtXs2fPVlRUlJo1a2Ycd+7cOXXt2lXHjx/Xc889p/r166tEiRK6d++ezpw5o2XLlmn+/PmaMmWKyeodAAAAyFmGDx9ucXuNGjWM//7tt9/0wQcfSLq/omDRokV15cqVLIkPAAAAzsXa9y2fBO9dAnAUmgzglG7duqUxY8bo0qVLMhgMqlOnjiZOnCgXFxdJ0p49e9SyZUuVLl1a8+fPNxm7a9cudejQ4YFzX7p0ya6xA8g6u3bt0smTJ9W+fXvt379f0dHRmjhxogoXLmyy3+nTp/Xaa6/pjz/+0KhRozRo0CDlyZPHZJ+EhASFh4crMTHRuO33339Xq1atdPHiRfXq1UtjxoxRwYIFTcYlJiZq5syZunnzpt3OEwAAAI73oL8s+ztfX1+tXr1agYGBeu6559SvXz99/fXXWRAdAAAAnIm171s+Cd67BOBINBnAKXXu3FmdO3d+4PN169Y1+SDw76pWraodO3bYKTIAzuSrr76SJHXt2lUvvPCCJk6cqGXLlikkJMRkv2HDhunmzZsaNGiQ3n33XYtzeXh4aOrUqbp7965x24QJE3Tx4kW99tprioiIsDjOYDBo1KhRJuMAAADwdDIYDKpXr56jwwAAAICDWfu+5ZPgvUsAjpTL0QEAtlagQAGVKlXqgV8AcobExEStWbNGxYoV08svv6zOnTsrV65cZqubnD9/XjExMcqfP78GDx78yHnz5csnSbpz546WLl0qybq/WksfBwAAAAAAAODpZe37lk+C9y4BOBorGQAAsqUlS5bozp07xiK9ePHievnll7V161bt379fL774oiRp9+7dkqQqVarIYDBYPf/Bgwd1584deXt7y9/f3x6nAAAAgGxk0qRJZts8PDz09ttvOyAaAAAAOCtr37d8Erx3CcDRHtlkkJaWJhcXl6yIBQCslpaW5ugQ4GALFiyQi4uLunTpYtz2xhtvaOvWrZo/f76xWE9ISJAkeXt7P9b8mR0HAACAnCk8PNxsW8WKFWkyAAAAgAlr37d8Erx3CcDRHnq7hIIFCyoxMZEP8wA4lbS0NCUmJqpgwYKODgUO8tNPPyk2Nla1a9dWyZIljdtbtGihwoULa9WqVbpx44ak/2tIedyGucyOAwAAQM6UmJho9rVz505HhwUAAAAn8jjvWz4J3rsE4GgPXcnA1dVVzz77rG7evJlV8QCAVZ599lm5unLHl6fVggULJMmkG1iSChQooHbt2mn+/PlatmyZevXqJU9PT0nSpUuXHusYmR0HAAAAAAAA4On0OO9bSlKuXPf/Fjg1NfWBc6Y/l76vxHuXABzvkZ/Qubq6qkiRIlkRCwAAj/S///1Pq1evliQNGDBAAwYMsLjf/Pnz1atXLwUHB0uSDh06pMTERBkMBquOU61aNeXPn1///e9/dfr0aZUtW9Ym8QMAAAAAAADIeR73fUtJKly4sHHsg1y/fl2STD6r471LAI7GnwEDALKVJUuW6O7du6pUqZKqVKlicZ+YmBgdP35c+/btU40aNdSgQQPFxMRo2rRpGj169EPnv3v3rvLly6f8+fOrY8eOWrBggSZPnqwvv/zSqnEAAAAAAAAAnj6Zed+yYsWKkqQ9e/Y8cN7059L3lcR7lwAcjiYDAEC2kr7kWHh4uGrXrm1xn88++0xhYWGaP3++atSooSlTpqhRo0aaNm2aihQpooEDB5rdbuP3339XeHi4qlWrZlzObNSoUdq8ebNWrlwpd3d3/fvf/9YzzzxjMu6PP/7QjBkzlDdvXr377rt2OGMAAAAAAAAAzi4z71vWrl1bJUuW1H/+8x8tXLhQ3bp1M9n/yJEjWrx4sVxdXdWxY0eT53jvEoAjuSQmJqY5OggAAKzx008/qXnz5nrhhRf0888/P3C/xMRElStXTrly5dKJEydUpEgRHTx4UN26ddPFixfl5eWlevXqydvbW3fu3NHJkye1a9cuJSUlKSoqSk2bNjXOde7cOXXt2lXHjx/X888/rwYNGqh48eJKTk7W2bNntXPnTv3xxx+KiIgwLnMGAACAnCP9dluJiYlW7T9q1Chdu3ZN0v2/OouLi1P9+vXl5eUlSQoODjZ78xgAAADZ25O8b7lv3z61a9dON2/eVHBwsF588UXlyZNHp06d0oYNG5SSkqIpU6YoJCTEbD7euwTgKDQZAACyjV69emn58uWaMGGCBg4c+NB9e/furWXLlmnKlCnq3bu3JOnPP//UokWLtG7dOh0/flyJiYnKly+fSpQoobp166p79+4KCAgwm+vevXtavny5Vq9ercOHD+v69evKnTu3ihcvruDgYHXt2lU1atSwyzkDAADAsR63yaBSpUq6cOHCA5/v3LmzZs+ebYPIAAAA4Cye9H3L3377TTNnzlRMTIwuXryolJQUubm5qVatWurTp49q1qz5wPl47xKAI9BkAAAAAAAAAAAAAAAArJLL0QEAAAAAAAAAAAAAAIDsgSYD2NxPP/2kTp06KSAgQAaDQVFRUY8cc+zYMTVr1kyenp4KCAhQeHi40tJYZAMAAOBpR20JAAAAW6G2BAAAsA2aDGBzt2/fVvny5TV58mQVKFDgkfvfvHlTbdq0kbu7u7Zs2aLJkydrxowZmjlzZhZECwAAAGdGbQkAAABbobYEAACwDVdHB4Ccp3HjxmrcuLEkqX///o/cf/ny5frrr780e/ZsFShQQOXLl9epU6c0a9YsDRw4UC4uLvYOGQAAAE6K2hIAAAC2Qm0JAABgG6xkAIf7+eefFRwcbNI93KhRI12+fFnnz593YGQAAADIbqgtAQAAYCvUlgAAAJbRZACHu3Llitzc3Ey2pT++cuWKI0ICAABANkVtCQAAAFuhtgQAALCMJgM4hYxLi6WlpVncDgAAADwKtSUAAABshdoSAADAHE0GcDh3d3ezzt+rV69KklmnMCw7ffq0o0NANkCewBrkCaxFrsBZUVs+Ob6/YS1yBdYgT2AN8gTOitrSNvgehzXIE1iDPIE1yJOsQZMBHK5mzZravXu37ty5Y9wWExMjLy8vlShRwoGRAQAAILuhtgQAAICtUFsCAABYRpMBbO7WrVs6cuSIjhw5otTUVF28eFFHjhzRhQsXJEljx45Vq1atjPu3b99eBQoUUP/+/XX8+HGtWbNGn376qfr378+yYwAAAE85aksAAADYCrUlAACAbdBkAJv75Zdf9PLLL+vll1/WX3/9pUmTJunll1/WxIkTJUnx8fGKi4sz7l+kSBGtWrVKly9fVoMGDfTee+9pwIABGjhwoKNOAQAAAE6C2hIAAAC2Qm0JAABgG66ODgA5T926dZWYmPjA52fPnm22rUKFCvrhhx/sGBUAAACyI2pLAAAA2Aq1JQAAgG2wkgEAAAAAAAAAAAAAALAKTQYAAAAAAAAAAAAAAMAqNBkAAAAAAAAAAAAAAACr0GQAAAAAAAAAAAAAAACsQpMBAAAAAAAAAAAAAACwCk0GAAAAAAAAAAAAAADAKjQZAAAAAAAAAAAAAAAAq9BkAAAAAAAAAAAAAAAArEKTAQAAAAAAAAAAAAAAsApNBgAAAAAAAAAAAAAAwCo0GQAAAAAAAAAAAAAAAKvQZAAAAAAAAAAAAAAAAKxCkwEAAAAAAAAAAAAAALAKTQYAAAAAAAAAAAAAAMAqNBkAAAAAAAAAAAAAAACr0GQAAAAAAAAAAAAAAACsQpMBAAAAAAAAAAAAAACwCk0GAAAAAAAAAAAAAADAKjQZAAAAAAAAAAAAAAAAq9BkAAAAAAAAAAAAAAAArEKTAQAAAAAAAAAAAAAAsApNBgAAAAAAAAAAAAAAwCqujg4AAAAAAAAAAAAAzslgMJg8TkxMfOrnBoCnHU0GAAAAAAAAAAAATxnD/Et2G5e89Q2r5kpuXcV0Q/f6jxyT0D2/VXNndHtLk0fuU7Dh+kzNDQBPG5oMAAAAAAAAAAAAYNknxxwdQaZciq7l6BAAIMfK5egAAAAAAAAAAAAAAABA9kCTAQAAAAAAAAAAAAAAsApNBgAAAAAAAAAAAAAAwCo0GQAAAAAAAAAAAAAAAKu4OjoAAAAAAAAAAAAAAEDOZjAYTB4nJiY6JA48OVYyAAAAAAAAAAAAAAAAVqHJAHYxd+5cBQYGysPDQ/Xq1dOuXbseuv+qVatUp04deXl5qWLFipo+fXoWRQoAAABnR20JAAAAW6K+BIDsxWAwyGAwqEaNGmZ/CQ/nYZh/6ZFfmRljaRwcj9slwOaio6MVFhamjz/+WLVq1dLcuXPVoUMH7dmzR8WLFzfbf9OmTQoJCVF4eLheeeUVxcbGatCgQcqfP7969+7tgDMAAACAs6C2BAAAgC1RXwKA8yjUvb7dxt1asDVTc8POPjnm6Agyhds8mKPJADYXGRmpLl26qHv37pKkiIgIbd68WfPmzdPo0aPN9l+6dKmaNGmikJAQSZKfn58GDx6sadOmqVevXnJxccnS+AEAAOA8qC0BAABgS9SXAGAffAiL7Or2liZ2G1Ow4frHnju7oMkANpWUlKRDhw4pNDTUZHvDhg21d+9ei2Pu3r2r/Pnzm2wrUKCALl26pN9++00lSpSwOO706dO2CTqH4PWANcgTWIM8gbXIlf9TtmxZR4eQI1FbOg6vB6xFrsAa5AmsQZ78H2pL+8mq+pJ8Nsdr4qyecXQATodcNVdjZ+byxJol7pOtnCu5dZXHPj7X0hHs9zPFmutZo0YNk8f79u2zam7vTEVkHWfIQ3vVljQZwKauXbumlJQUubm5mWx3c3PTlStXLI5p1KiRwsLCtGXLFtWvX1/nzp3TzJkzJUkJCQkPfCOYX7j+z+nTp3k98EjkCaxBnsBa5AqyArWlY/D9DWuRK7AGeQJrkCfIKllVX5LPpvged2I7uc95RuSqBdk0T7iWDmDHXKk6oZfdxiR0z//IfS5F13rs40s5Ow9pMoBdZFwmLC0t7YFLh3Xv3l1xcXHq0qWL7t27p2effVZ9+/bV5MmTlTt37qwIFwAAAE6M2hIAAAC2RH0JAHbwyTFHRwAgC+VydADIWYoWLarcuXObdf5evXrVrEM4nYuLi8aOHatLly7p6NGjOnXqlKpXry5J8vX1tXvMAAAAcE7UlgAAALAl6ksAAHKu5NZVTL5gXzQZwKby5s2rKlWqKCYmxmR7TEyMgoKCHjo2d+7c8vb2Vt68ebVixQrVrFnzgcU9AAAAcj5qSwAAANgS9SUAAIBtcLsE2NyAAQPUp08fVa9eXUFBQZo3b57i4+PVo0cPSdLYsWN14MABrVmzRtL9e6F9++23qlOnju7evauoqCitXr1aa9eudeRpAAAAwAlQWwIAAMCWqC8BAACeHE0GsLm2bdvq+vXrioiIUEJCggICArRs2TLj8mHx8fGKi4szGfPNN9/ogw8+UFpammrUqKHvv//euOwYAAAAnl7UlgAAALAl6ksAAIAnR5MB7CIkJEQhISEWn5s9e7bJ46JFi2rTpk1ZERYAAACyIWpLAAAA2BL1JQAAwJPJ5egAAAAAAAAAAAAAAABA9kCTAQAAAAAAAAAAAAAAsAq3SwCAHM5gMJg8TkxMzBZzAwAAAAAAwL54bwcAAGQGTQYA4AQy+wtdoe71H/tY1o65tWDrY88NAAAAAAAAxzPMv2S3cYk9fDI1NwAAyDloMgAAO8vML3XWjkm2Zp/WVR77+JJ0e0sTu40p2HD9Y88NAAAAAAAAG/vkmN2mZpUEAAByLpoMAMAZ2PEXOnu6FF3L0SEAAAAAAADgKUMDAwAAjkWTAQAAAAAAAAAAsEpmbt9p7biE7vkzNbc1q2uysiYAALZDkwEAAAAAAAAAAMg2WF0TAADHoskAAAAAAAAAAADYVHLrKo4OAQAA2AlNBgAAAAAAAAAAAACeOgaDweRxYmKiQ+IAsptcjg4AAAAAAAAAAAAAAABkD6xkAAAA6NjNYbieAAAAAAAAAAB7ockAAAAgmzDMv2S3cYk9fDI1ty3RHAEAAAAAAABbuL2lid3GFWy4PlNzAzkJTQYAAORwhbrXt9uYWwu2PvbcAAAAAAAAAAAg+6LJAACAbIK/8obVPjnm6AhM0DkOAAAAAAAAADkHTQYAADiBzCyDb+2YZGv2aV3lsY8P2Nql6FqODgEAAAAAAAAA8Ag0GQAAkF042V+nAwAAAAAAAEB2xh+9AJlDkwEAAACsxm07AAAAAAAAAODpRpMBYGd8GAMAAAAAAAAAAAAgp6DJAHgC9ryHemIPn8ee25ZojgCAp0uh7vXtNu529/yZmhsAAAAAAAAA4HxoMgCyMRoBAABZLbl1FUeHAAAAAAAAAABwIJoMAHv75JijIwAAAAAAAAAAAAAAm6DJAHBSmVm22toxCZlYtvr2liZW7Vew4frHnhsAAAAAAAAAAABA9kCTAZCNsWQ1AAAAAAAAAAAAgKxEkwEAiy5F13J0CACyAWtWOfFpu8fksbU/X1gZBQAAAAAAAAAA50OTAQAAsCualgAAAAAAAAAAyDlyOToAAAAAAAAAAAAAAACQPdBkAAAAAAAAAAAAAAAArEKTAQAAAAAAAAAAAAAAsApNBgAAAAAAAAAAAAAAwCo0GcAu5s6dq8DAQHl4eKhevXratWvXQ/ffvHmz/vnPf6pYsWIqVaqUOnfurDNnzmRRtAAAAHBm1JYAAACwJepLAACAJ0OTAWwuOjpaYWFhGjp0qLZv366aNWuqQ4cOunDhgsX9f/31V3Xp0kXBwcHavn27vv32W925c0cdOnTI4sgBAADgbKgtAQAAYEvUlwAAAE+OJgPYXGRkpLp06aLu3bvL399fERER8vDw0Lx58yzuf/jwYd27d0+jR49WqVKlFBgYqMGDBysuLk7Xrl3L4ugBAADgTKgtAQAAYEvUlwAAAE+OJgPYVFJSkg4dOqSGDRuabG/YsKH27t1rcUyVKlWUJ08eLVy4UCkpKfrjjz/09ddfq1q1aipatGhWhA0AAAAnRG0JAAAAW6K+BAAAsA1XRweAnOXatWtKSUmRm5ubyXY3NzdduXLF4pgSJUpo1apVeuutt/Tuu+8qNTVVgYGBWrFixUOPdfr0aZvFnXnPODoAp+Mc18XZkCcZkSeWkCcZkSeWkCcZOUuelC1b1tEh5EhPX23pPHg9YC1yBdYgT2AN8uT/UFvaT1bVl86Rz/z+mJFzXBdnQ55kRJ5YQp5kRJ48CLmSkTPkir1qS5oMYBcuLi4mj9PS0sy2pUtISFBoaKg6deqkdu3a6datW5o4caLeeustfffdd8qVy/KCG07xC9fOS46OwOk4xXVxNuSJGfLEAvLEDHliAXlihjx5Ojw1taWTOH36NK8HrEKuwBrkCaxBniCr2bu+dIp85vdHM05xXZwNeWKGPLGAPDFDnjwAuWImJ+cKTQawqaJFiyp37txmnb9Xr1416xBO98UXX+iZZ57RuHHjjNs+//xzVahQQXv37lVwcLBdYwYAAIBzorYEAACALVFfAgAA2IblP+MBMilv3ryqUqWKYmJiTLbHxMQoKCjI4pi//vpLuXPnNtmW/jg1NdU+gQIAAMDpUVsCAADAlqgvAQAAbIMmA9jcgAEDtGTJEi1cuFCxsbEaPny44uPj1aNHD0nS2LFj1apVK+P+jRs31uHDhzV58mSdPXtWhw4d0oABA1SsWDFVqVLFQWcBAAAAZ0BtCQAAAFuivgQAAHhy3C4BNte2bVtdv35dERERSkhIUEBAgJYtWyZfX19JUnx8vOLi4oz716tXT3PnztW0adM0Y8YM5c+fXy+++KJWrFihggULOuo0AAAA4ASoLQEAAGBL1JcAAABPjiYD2EVISIhCQkIsPjd79myzbe3atVO7du3sHRYAAACyIWpLAAAA2BL1JQAAwJPhdgkAAAAAAAAAAAAAAMAqNBkAAAAAAAAAAAAAAACr0GQAAAAAAAAAAAAAAACsQpMBAAAAAAAAAAAAAACwCk0GAAAAAAAAAAAAAADAKjQZAAAAAAAAAAAAAAAAq9BkAAAAAAAAAAAAAAAArEKTAQAAAAAAAAAAAAAAsApNBgAAAAAAAAAAAAAAwCo0GQAAAAAAAAAAAAAAAKvQZAAAAAAAAAAAAAAAAKxCkwEAAAAAAAAAAAAAALAKTQYAAAAAAAAAAAAAAMAqNBkAAAAAAAAAAAAAAACr0GQAAAAAAAAAAAAAAACsQpMBAAAAAAAAAAAAAACwCk0GAAAAAAAAAAAAAADAKjQZAAAAAAAAAAAAAAAAq9BkAAAAAAAAAAAAAAAArEKTAQAAAAAAAAAAAAAAsApNBgAAAAAAAAAAAAAAwCo0GQAAAAAAAAAAAAAAAKvQZAAAAAAAAAAAAAAAAKxCkwEAAAAAAAAAAAAAALAKTQYAAAAAAAAAAAAAAMAqro4OAAAAAAAAZB8Gg8HkcWJiokPiAAAAAAAAjkGTAQAAAAAAcAo0MAAAAAAA4PxoMgAAAAAAIAfhg3oAAAAAAGBPNBkAAAAAAAAV6l7fbuNuLdiaqbkBAAAAAIDzockAAAAAAADY1e0tTew2rmDD9ZmaGwAAAAAAZA5NBgAAAAAAZBOG+ZfsNib5sWcGAAAAAABPo1yODgA509y5cxUYGCgPDw/Vq1dPu3bteuC+kyZNksFgsPj1+++/Z2HUAAAAcEbUloBzSW5dxeTLli5F1zL5AgDAHqgvAQAAngxNBrC56OhohYWFaejQodq+fbtq1qypDh066MKFCxb3Dw0NVWxsrMnXSy+9pDp16sjNzS2LowcAAIAzobYEAACALVFfAgAAPDmaDGBzkZGR6tKli7p37y5/f39FRETIw8ND8+bNs7h/oUKF5OHhYfy6d++edu/ere7du2dx5AAAAHA21JZAJnxyzPQLAAAYUV8CAAA8OZoMYFNJSUk6dOiQGjZsaLK9YcOG2rt3r1VzLFq0SEWKFFGrVq3sESIAAACyCWpLAAAA2BL1JQAAgG24OjoA5CzXrl1TSkqK2VJhbm5uunLlyiPHp6amKioqSp06dVK+fPkeuu/p06efKFbbeMbRATgd57guzoY8yYg8sYQ8yYg8sYQ8ychZ8qRs2bKODiFHevpqS+fB6+HM+L8gI/LV+XGNYA3y5P9QW9pPVtWXzpHP1AwZOcd1cTbkSUbkiSXkSUbkyYOQKxk5Q67Yq7akyQB24eLiYvI4LS3NbJslmzZt0sWLF9WtW7dH7usUv3DtvOToCJyOU1wXZ0OemCFPLCBPzJAnFpAnZsiTp8NTU1s6idOnT/N6ODP+LzBDvjo3fqbAGuQJspq960unyGdqBjNOcV2cDXlihjyxgDwxQ548ALliJifnCrdLgE0VLVpUuXPnNuv8vXr1qlmHsCVfffWVgoKCFBAQYK8QAQAAkE1QWwIAAMCWqC8BAABsgyYD2FTevHlVpUoVxcTEmGyPiYlRUFDQQ8devnxZGzdutOovzQAAAJDzUVsCAADAlqgvAQAAbIMmA9jcgAEDtGTJEi1cuFCxsbEaPny44uPj1aNHD0nS2LFj1apVK7NxixcvVsGCBdWmTZusDhkAAABOitoSAAAAtkR9CQAA8ORcHR0Acp62bdvq+vXrioiIUEJCggICArRs2TL5+vpKkuLj4xUXF2cyJi0tTYsWLVKHDh30zDPPOCJsAAAAOCFqSwAAANgS9SUAAMCTo8kAdhESEqKQkBCLz82ePdtsm4uLi44cOWLvsAAAAJANUVsCAADAlqgvAQAAngy3SwAAAAAAAAAAAAAAAFahyQAAAAAAAAAAAAAAAFiFJgMAAAAAAAAAAAAAAGAVmgwAAAAAAAAAAAAAAIBVaDIAAAAAAAAAAAAAAABWockAAAAAAAAAAAAAAABYxdXRAQAAAAAAACBnMRgMJo8TExMdEgcAAAAAwPZYyQAAAAAAAAAAAAAAAFiFJgMAAAAAAAAAAAAAAGAVmgwAAAAAAAAAAAAAAIBVaDIAAAAAAAAAAAAAAABWcXV0AAAAAAAAAMgeCnWvb7dxtxZszdTcAAAAAICsxUoGAAAAAAAAAAAAAADAKjQZAAAAAAAAAAAAAAAAq3C7BAAAAAAAgKeQwWAweZyYmOiQOAAAAAAA2QtNBgAAAAAAALCp5NZVHB0CAAAAAMBOuF0CAAAAAAAAAAAAAACwCk0GAAAAAAAAAAAAAADAKtwuAQAAAAAAIAcxzL9kt3HJmZoZAAAAAJCT0GQAAAAAAAAAwCKDwWDyODEx0SFxAAAAAHAe3C4BAAAAAAAAAAAAAABYhZUMAAAAAAAAnkafHHN0BAAAAACAbIiVDAAAAAAAAAAAAAAAgFVYyQAAAAAAAAB4yhjmX7LbuMQePpmaGwAAAED2wEoGAAAAAAAAAAAAAADAKqxkAAAAAAAAACBHMRgMJo8TExMdEgcAAACQE7GSAQAAAAAAAAAAAAAAsAorGQAAAABANsFfZQIAstwnxxwdAQAAAAAnw0oGAAAAAAAAAAAAAADAKqxkAAAAAABglQQAQLZwe0sTu40r2HB9puYGAAAAnjY0GQAAAAAA7IoGBgAAAAAAgJyD2yXALubOnavAwEB5eHioXr162rVr10P3T0tL06xZs1SjRg25u7vL399fY8aMyZpgAQAA4NSeltrSMP/SI78yM8bSOAAAgKfZ01JfAgAA2AsrGcDmoqOjFRYWpo8//li1atXS3Llz1aFDB+3Zs0fFixe3OGbkyJHasGGDxo0bpwoVKujGjRtKSEjI4sgBAADgbKgtAQAAYEvUlwAAAE+OJgPYXGRkpLp06aLu3btLkiIiIrR582bNmzdPo0ePNtv/9OnT+vzzz/XTTz/J398/q8MFAACAE6O2BAAAmXEpupajQ4CTor4EAAB4cjQZwKaSkpJ06NAhhYaGmmxv2LCh9u7da3HMunXr5Ofnpx9//FGvv/66UlNT9dJLL2n8+PFyc3N74LFOnz5t09gz5xlHB+B0nOO6OBvyJCPyxBLyJCPyxBLyJCNnyZOyZcs6OoQcidrSdjJzfvZ8TZzj9c6u+L8gI/LJkuyZJ1zLrJY9/9/xfuyZbR+DvVFb2k9W1ZfOkUvZ8/8Ce3KO6+JsyJOMyBNLyJOMyJMHIVcycoZcsVdtSZMBbOratWtKSUkxK7Dd3Nx05coVi2N+/fVXXbhwQdHR0Zo1a5ZcXFz073//W506ddKmTZuUK1cui+Oc4heundzfNiOnuC7OhjwxQ55YQJ6YIU8sIE/MkCc5G7WlBZ8cy9TUVSf0stuYWwu2PvbcTvF6Z1f8X2CGfLIgm+YJ1zKL2TFPHvdanj592uoxty9kJiLrkIM5X1bVl06RS9n0/wJ7corr4mzIEzPkiQXkiRny5AHIFTM5OVdoMoBduLi4mDxOS0sz25YuNTVVd+/e1Zw5c1SmTBlJ0pw5c/Tiiy/q4MGDevHFF+0eLwAAAJwXtSUAAABsifoSAADgydBkAJsqWrSocufObdb5e/Xq1QcuH+bh4SFXV1djkS5JpUuXlqurqy5evEihDgAA8JSitswebm9pYrcxBRuuf+y5AQAAHoT6EgAAwDYsrxUKZFLevHlVpUoVxcTEmGyPiYlRUFCQxTG1atVScnKy4uLijNt+/fVXJScnq3jx4naNFwAAAM6L2hIAAAC2RH0JAABgGzQZwOYGDBigJUuWaOHChYqNjdXw4cMVHx+vHj16SJLGjh2rVq1aGfevX7++KleurAEDBujw4cM6fPiwBgwYoBdffFFVq1Z11GkAAADACVBbAgAAwJaoLwEAAJ4ct0uAzbVt21bXr19XRESEEhISFBAQoGXLlsnX11eSFB8fb9L5mytXLi1dulTDhw9X8+bNlT9/fjVo0EAffvihcuWiDwYAAOBpRm2ZdZJbV3F0CAAAAHZHfQkAAPDkaDKAXYSEhCgkJMTic7Nnzzbb5unpqQULFtg7LAAAAGRD1JYAAACwJepLAACAJ0OrJQAAAAAAAAAAAAAAsAorGQAAAAAA7OpSdC1HhwAAAAAAAAAbYSUDAAAAAAAAAAAAAABgFZoMAAAAAAAAAAAAAACAVWgyAAAAAAAAAAAAAAAAVqHJAAAAAAAAAAAAAAAAWIUmAwAAAAAAAAAAAAAAYBWaDAAAAAAAAAAAAAAAgFVoMgAAAAAAAAAAAAAAAFahyQAAAAAAAAAAAAAAAFiFJgMAAAAAAAAAAAAAAGAVV0cHAAAAAAAAAODpYzAYTB4nJiY6JA4AAAAAj4eVDAAAAAAAAAAAAAAAgFVYyQAAAAAAAACAzRTqXt9u4253z5+puQEAAADYDk0GAAAAAAAAyDZYYh8AAAAAHIsmAwAAAAAAADjc7S1N7DauYMP1mZob9pXcuoqjQwAAAACQCTQZAAAAAAAAINu4FF3L0SEAAAAAwFMtl6MDAAAAAAAAAAAAAAAA2QNNBgAAAAAAAAAAAAAAwCo0GQAAAAAAAAAAAAAAAKvQZAAAAAAAAAAAAAAAAKxCkwEAAAAAAAAAAAAAALAKTQYAAAAAAAAAAAAAAMAqNBkAAAAAAAAAAAAAAACr0GQAAAAAAAAAAAAAAACsQpMBAAAAAAAAAAAAAACwCk0GAAAAAAAAAAAAAADAKjQZAAAAAAAAAAAAAAAAq9BkAAAAAAAAAAAAAAAArEKTAQAAAAAAAAAAAAAAsApNBrCLuXPnKjAwUB4eHqpXr5527dr1wH3Pnz8vg8Fg9vXjjz9mYcQAAABwVtSWAAAAsCXqSwAAgCfj6ugAkPNER0crLCxMH3/8sWrVqqW5c+eqQ4cO2rNnj4oXL/7AcStXrlTFihWNj5977rmsCBcAAABOjNoSAAAAtkR9CQAA8ORYyQA2FxkZqS5duqh79+7y9/dXRESEPDw8NG/evIeOe/755+Xh4WH8yps3bxZFDAAAAGdFbQkAAABbor4EAAB4cjQZwKaSkpJ06NAhNWzY0GR7w4YNtXfv3oeOffPNN1WmTBm9+uqrWr16tT3DBAAAQDZAbQkAAABbor4EAACwDW6XAJu6du2aUlJS5ObmZrLdzc1NV65csTimUKFCGj9+vGrVqiVXV1etW7dOPXr00OzZs9WxY8cHHuv06dM2jT1znnF0AE7HOa6LsyFPMiJPLCFPMiJPLCFPMnKWPClbtqyjQ8iRqC3hHNfF2ZAnGZEnlpAnGZEnlpAnGTlLnlBb2k9W1ZfOkUt8j2fkHNfF2ZAnGZEnlpAnGZEnD0KuZOQMuWKv2pImA9iFi4uLyeO0tDSzbemKFi2q0NBQ4+OqVavq+vXrmjZt2kPfCHaKX7h2XnJ0BE7HKa6LsyFPzJAnFpAnZsgTC8gTM+TJ04Ha8unlFNfF2ZAnZsgTC8gTM+SJBeSJGfLk6WHv+tIpconvcTNOcV2cDXlihjyxgDwxQ548ALliJifnCrdLgE0VLVpUuXPnNuv8vXr1qlmH8MNUr15d586ds3V4AAAAyEaoLQEAAGBL1JcAAAC2QZMBbCpv3ryqUqWKYmJiTLbHxMQoKCjI6nmOHj0qDw8PW4cHAACAbITaEgAAALZEfQkAAGAb3C4BNjdgwAD16dNH1atXV1BQkObNm6f4+Hj16NFDkjR27FgdOHBAa9askSQtWbJEefLkUWBgoHLlyqX169dr7ty5GjNmjAPPAgAAAM6A2hIAAAC2RH0JAADw5GgygM21bdtW169fV0REhBISEhQQEKBly5bJ19dXkhQfH6+4uDiTMR999JEuXLig3Llzq3Tp0po5c+ZD75kLAACApwO1JQAAAGyJ+hIAAODJ0WQAuwgJCVFISIjF52bPnm3yuEuXLurSpUtWhAUAAIBsiNoSAAAAtkR9CQAA8GRyOToAAAAAAAAAAAAAAACQPdBkAAAAAAAAAAAAAAAArEKTAQAAAAAAAAAAAAAAsApNBgAAAAAAAAAAAAAAwCo0GQAAAAAAAAAAAAAAAKvQZAAAAAAAAAAAAAAAAKxCkwEAAAAAAAAAAAAAALAKTQYAAAAAAAAAAAAAAMAqNBkAAAAAAAAAAAAAAACr0GQAAAAAAAAAAAAAAACsQpMBAAAAAAAAAAAAAACwCk0GAAAAAAAAAAAAAADAKjQZAAAAAAAAAAAAAAAAq9BkAAAAAAAAAAAAAAAArEKTAQAAAAAAAAAAAAAAsApNBgAAAAAAAAAAAAAAwCo0GQAAAAAAAAAAAAAAAKvQZAAAAAAAAAAAAAAAAKxCkwEAAAAAAAAAAAAAALAKTQYAAAAAAAAAAAAAAMAqNBkAAAAAAAAAAAAAAACr0GQAAAAAAAAAAAAAAACsQpMBAAAAAAAAAAAAAACwCk0GAAAAAAAAAAAAAADAKjQZAAAAAAAAAAAAAAAAq9BkAAAAAAAAAAAAAAAArEKTAQAAAAAAAAAAAAAAsApNBgAAAAAAAAAAAAAAwCo0GQAAAAAAAAAAAAAAAKvQZAAAAAAAAAAAAAAAAKxCkwEAAAAAAAAAAAAAALAKTQawi7lz5yowMFAeHh6qV6+edu3aZdW4s2fPqlixYvLx8bFzhAAAAMguqC0BAABgS9SXAAAAT4YmA9hcdHS0wsLCNHToUG3fvl01a9ZUhw4ddOHChYeOS0pK0ttvv63atWtnUaQAAABwdtSWAAAAsCXqSwAAgCdHkwFsLjIyUl26dFH37t3l7++viIgIeXh4aN68eQ8dN3r0aFWoUEGtW7fOokgBAADg7KgtAQAAYEvUlwAAAE+OJgPYVFJSkg4dOqSGDRuabG/YsKH27t37wHEbNmzQhg0bFB4ebu8QAQAAkE1QWwIAAMCWqC8BAABsw9XRASBnuXbtmlJSUuTm5may3c3NTVeuXLE4Jj4+XoMGDdKiRYv07LPPWn2s06dPP1GstvGMowNwOs5xXZwNeZIReWIJeZIReWIJeZKRs+RJ2bJlHR1CjkRtCee4Ls6GPMmIPLGEPMmIPLGEPMnIWfKE2tJ+sqq+dI5c4ns8I+e4Ls6GPMmIPLGEPMmIPHkQciUjZ8gVe9WWNBnALlxcXEwep6WlmW1L17t3b7399tuqUaPGYx3DKX7h2nnJ0RE4Hae4Ls6GPDFDnlhAnpghTywgT8yQJ08Hasunl1NcF2dDnpghTywgT8yQJxaQJ2bIk6eHvetLp8glvsfNOMV1cTbkiRnyxALyxAx58gDkipmcnCvcLgE2VbRoUeXOndus8/fq1atmHcLptm/frvDwcBUtWlRFixZVaGiobt++raJFi+qrr77KgqgBAADgjKgtAQAAYEvUlwAAALbBSgawqbx586pKlSqKiYnRa6+9ZtweExOjVq1aWRyza9cuk8fr1q3Txx9/rM2bN8vb29ue4QIAAMCJUVsCAADAlqgvAQAAbIMmA9jcgAED1KdPH1WvXl1BQUGaN2+e4uPj1aNHD0nS2LFjdeDAAa1Zs0aSVL58eZPxv/zyi3LlymW2HQAAAE8faksAAADYEvUlAADAk6PJADbXtm1bXb9+XREREUpISFBAQICWLVsmX19fSVJ8fLzi4uIcHCUAAACyA2pLAAAA2BL1JQAAwJOjyQB2ERISopCQEIvPzZ49+6Fj33jjDb3xxhv2CAsAAADZELUlAAAAbIn6EgAA4MnkcnQAAAAAAAAAAAAAAAAge6DJAAAAAAAAAAAAAAAAWIUmAwAAAAAAAAAAAAAAYBWaDAAAAAAAAAAAAAAAgFVoMgAAAAAAAAAAAAAAAFahyQAAAAAAAAAAAAAAAFiFJgMAAAAAAAAAAAAAAGAVmgwAAAAAAAAAAAAAAIBVaDIAAAAAAAAAAAAAAABWockAAAAAAAAAAAAAAABYhSYDAAAAAAAAAAAAAABgFZoMAAAAAAAAAAAAAACAVWgyAAAAAAAAAAAAAAAAVqHJAAAAAAAAAAAAAAAAWIUmAwAAAAAAAAAAAAAAYBWaDAAAAAAAAAAAAAAAgFVoMgAAAAAAAAAAAAAAAFahyQAAAAAAAAAAAAAAAFiFJgMAAAAAAAAAAAAAAGAVmgwAAAAAAAAAAAAAAIBVaDIAAAAAAAAAAAAAAABWockAAAAAAAAAAAAAAABYhSYDAAAAAAAAAAAAAABgFZoMAAAAAAAAAAAAAACAVWgyAAAAAAAAAAAAAAAAVqHJAAAAAAAAAAAAAAAAWIUmAwAAAAAAAAAAAAAAYBWaDAAAAAAAAAAAAAAAgFVoMgAAAAAAAAAAAAAAAFahyQAAAAAAAAAAAAAAAFiFJgMAAAAAAAAAAAAAAGAVmgxgF3PnzlVgYKA8PDxUr1497dq164H7njx5Ui1atFDZsmXl4eGhypUra9y4cUpKSsrCiAEAAOCsqC0BAABgS9SXAAAAT8bV0QEg54mOjlZYWJg+/vhj1apVS3PnzlWHDh20Z88eFS9e3Gz/vHnzqnPnzgoMDFSRIkX0n//8R4MGDVJycrLGjRvngDMAAACAs6C2BAAAgC1RXwIAADw5l8TExDRHB4GcpVGjRqpQoYKmT59u3FatWjW1bt1ao0ePtmqO999/X/v27dOmTZvsFSYAAACyAWpLAAAA2BL1JQAAwJPjdgmwqaSkJB06dEgNGzY02d6wYUPt3bvXqjnOnTunzZs366WXXrJHiAAAAMgmqC0BAABgS9SXAAAAtkGTAWzq2rVrSklJkZubm8l2Nzc3Xbly5aFjGzduLA8PD1WrVk21atXSBx98YM9QAQAA4OSoLQEAAGBL1JcAAAC2QZMB7MLFxcXkcVpamtm2jObNm6dt27Zp7ty52rRpkz799FM7RggAAIDsgtoSAAAAtkR9CQAA8GRcHR0AcpaiRYsqd+7cZp2/V69eNesQzqhYsWKSpHLlyiklJUXvvPOO3nnnHbm6kqYAAABPI2pLAAAA2BL1JQAAgG2wkgFsKm/evKpSpYpiYmJMtsfExCgoKMjqeVJTU5WcnKyUlBRbhwgAAIBsgtoSAAAAtkR9CQAAYBs0GcDmBgwYoCVLlmjhwoWKjY3V8OHDFR8frx49ekiSxo4dq1atWhn3/+abb/Ttt9/q1KlT+vXXX7Vq1SqNGzdOrVu3Vr58+RxyDlFRUfLx8XmsMf369VPHjh0fuk/z5s313nvvPUlodpWZ834SBoNBq1evNj4+deqU/vnPf8rDw0OVKlWyuA8enzW5aSs7duyQwWDQtWvXjNvWrl2ratWqqWjRourXr5/FfZ42tvhZUKlSJc2YMcNGEWVvzv6z9e86duyofv36OTqMbCkrc37SpEkKDg4221a2bFkZDAZFRUVZ3Af2QW35YM7+84/aMueivnQu1Ja25ew/W/+O2jLzqC2fbtm9vqS2zBrUllmH2tK5UFvalrP/bM2I+jLznsb6kiYD2Fzbtm01adIkRUREqG7dutqzZ4+WLVsmX19fSVJ8fLzi4uKM+7u6umrq1Klq2LChXnrpJYWHhyskJESRkZEm81oqNtavXy8vLy+NHz9ekyZNksFgUGhoqMk+58+fl8Fg0C+//PJY53Do0KHHPHPnt2bNGrVs2VK+vr7y9vZW7dq1NX78eP3+++8OiSc2NlZNmjQxPp4wYYIKFCign3/+2dhRnnGfrPaoIvfIkSPq2bOnypUrJ3d3d1WsWFEdOnTQd999p9TUVEn/l4PpX+7u7qpevbrZfzjpOdy6dWuz48ydO1cGg8HsP4WkpCRNnz5ddevWlbe3t/z8/PTKK69o/vz5unv3rg1egccTFBSk2NhYPf/888Zt77zzjlq1aqWjR49q8uTJFvdxNvb+5Wbx4sX64IMPrNr3QcVATEyMevbsadUc6b8cpX+VLFlSLVu21J49ex4rbmf1OK+ndP/6pr8WRYsWVcWKFTVkyBAlJibaL8gs9vdz/PvXkSNHHBbTw35Jv3LlioYPH64qVarI3d1dAQEBat++vTZu3OiASKXQ0FCtXbvW+Pj48eMKDw/X1KlTFRsbq7Zt25rtA/uhtnRe1JaZQ335eHJCfUltmb1QW5qjtnwy1JbOxx71JbXlk6O2zBxqy8dDbflo1Ja29bi1pUR96SjUl4+PG0bBLkJCQhQSEmLxudmzZ5s8bt++vdq3b//Yx/jmm2/0zjvvaOzYserXr58mTZqk/Pnza8mSJerfv78CAgIyFbskFShQQAUKFMj0+KyUmpqqtLQ05c6d+6H7jR8/Xp988on69u2r999/Xz4+PoqLi9PixYv15ZdfKiwsLIsi/j8eHh4mj8+dO6dmzZqpRIkSD9zncSUnJyt37txycXF5onksWb9+vbp166Z69eopMjJSpUqV0v/+9z8dO3ZMH3/8sapVq2bSYb1y5UpVrFhRd+/e1fbt2/Wvf/1LPj4+atu2rXEfDw8P7d69W+fPnzd5HRYvXmy891+6pKQktW3bVkeOHNH777+v4OBgFSlSRL/88osiIyNVpkwZ1a1b1+bn/TB58+Y1uWaJiYm6du2aGjZsKG9vb+P2J72uSUlJyps37xPN4UjPPffcE8/xj3/847HH7NmzR88995yuXr2qjz76SK+//roOHDjwyPtOPomsuFaZeT3r16+vOXPmKDk5WbGxsRo4cKBu3LihL7/80g4ROkb6Of5d0aJFMzWXPa/j+fPn1aRJExUqVEijR49WxYoVlZqaqm3btmnIkCH6z3/+Y5fjPkyhQoVMHp87d06S1KJFC5v+f5Ldf5ZlJWrLrENt+XD2rC0l6ktLqC8fjdrStqgtLaO2zDxqS+dk7/qS2vI+asuHo7aktnRG1Ja2ldnXk/ry8VBfOqa+ZCUDZEuzZ8/WO++8o+nTp5ss3VKyZEk1atRIY8eOfej4//73v3r77bdVokQJlShRQq+//rrOnj1rfN7S8ltTp05V2bJl5ePjoz59+mjy5MnGpbEyxhYQEKASJUqof//++vPPP02eT05O1vDhw43H/ve//23sGpXuFzZ9+/ZViRIl5OnpqdatW+vEiRNmsW3cuFHBwcFyc3NTbGysjh07platWql48eIqVqyYXnrpJW3fvl2SdODAAX388ccaN26cJk6cqODgYPn6+qpevXr64osv1LdvX4uvU1xcnDp37qwXXnhB3t7eevnll7V+/XqTfdasWaPatWvL09NTfn5+atasma5cuSJJunjxojp37iw/Pz95eXmpRo0aWrlypXHs35cUMxgM+s9//qMpU6bIYDBo0qRJZvtYc+3SuyejoqKMXWW3b9+2eH5P4vbt2xowYIAaN26s5cuXq1GjRipZsqSqVaumN998U1u3bjUpTCXp+eefl4eHh3x9fdW1a1dVrFhRhw8fNtuncePGioqKMm77z3/+o9OnT5t1Cc+ePVs//fSTVq9erb59+6py5cry8/NTmzZttHHjRlWuXNli7D/++KOaNm2qEiVKyM/PT23btlVsbKzJPuHh4apYsaLc3d31wgsvqE+fPsbnfvrpJ73yyivy8fGRr6+vGjVqpOPHj0sy7fbbsWOH/Pz8JEmtWrWSwWDQjh07LHYE7t27V82aNZOXl5cCAgI0ZMgQ3bx50/h88+bNNWTIEI0aNUqlS5fWq6+++qhLZDc//fSTGjVqJA8PD5UtW1YjRoxQUlKS8fnbt2+rT58+8vHxUdmyZTV16lSzZaYyLpP1oO+jqKgohYeH68SJE8ZuzvTcyLj80s2bNzVkyBD5+/vLw8NDNWvWVHR0tEnsbm5u8vDwUIUKFfTuu+/q5s2b2r9/v/H5kydP6vXXX1exYsVUpkwZ9ezZUwkJCcbnk5OTNWLECOP334gRIzRkyBA1b97c5NwsXatHzf2wn2H37t3TsGHDjF33FSpU0JgxYx74ej7q5+i5c+e0Y8cOnTx5Uu3bt1fXrl0l3f/ekKSUlBQNHDhQgYGB8vT0VLVq1TRt2jSTn9XpXeMP+5n/559/ql+/fsZc+Pjjj5WRtT/zN23apBo1asjLy0udOnXSjRs3tHr1alWrVk2+vr7q3bu3/vrrL5O58+XLJw8PD5MvV9f7/aWPymNbX8fz58+rZcuWkqTSpUvLYDAYvyfeffddpaWlKSYmRm3atFHZsmXl7++v3r17a+fOnWavWbqZM2eqdu3a8vb2VkBAgEJDQ006um/cuKHevXurTJky8vDwUOXKlTVr1izj8/Pnz1f16tXl4eGh0qVLq23btkpOTpZk2ok/adIkY44899xzMhgMZvukW7x4sYKCguTh4aHq1asrMjLSJG8MBoO++OILde3aVd7e3ho3btwDzw9Zi9qS2tLRtaVEffm01pfUltSWErUltSW1ZU5DbUltSW15H7Vl1qO2zP61ZVRUlJYvX65bt27ptddeU40aNTR16lQ1atRIW7ZskUR9aY9rSX153+PWlzQZINuZMGGCxo0bp0WLFqlTp05mz48ePVobN27Url27LI7/888/1bJlS+XLl09r167Vpk2b5OHhodatW5sV1ulWrlyp8PBw/fvf/9a2bdvk7+9v8s2ebvfu3Tpx4oS+/fZbzZ8/X99//70+++wzk32WL1+u1NRUbdq0SZ9++qkWLFhgMle/fv104MABLVmyRJs3/7/27j0qqqr/4/h7ECVF8NJ6EBWSxgulohEmD1YiKmaaiBopKGaPyZK8pVbi/QbLMm8gKOZjmVCm4B1RHkFRUURRA0PFC6LkTy2pRPAGzvz+YM2JcRgY8Jb1fa3lWjJzZp99zj5nz2dm7dk7idq1a/Puu+/qdcB37txhwYIFLF68mLS0NOzt7RkxYgS2trYkJSWxb98+goKCeO655wBYv349lpaWemGrLF2n86DCwkI8PT3ZtGkTKSkpeHl54e/vz5kzZwC4du0aw4cPx9fXl7S0NOLj4/XaZOLEidy+fZtt27aRmprKvHnzqFevXrn7ys7OpmXLlowePZrs7GyD6ePA9La7ePEisbGxrF69mpSUFOU8PEq7d+8mPz+fcePGGd3G2IgxrVbLoUOHOHPmDC4uLgbP+/v7s3btWqVzj4qKwtvb22B02vr16+nSpQvOzs4GZZiZmWFtbV3u/ouKihg5ciS7d+8mLi4Oa2trBg0apLxJb9myhfDwcBYuXMjRo0dZt26dUs+SkhL8/Pz497//TUpKComJiYwcObLcEemurq7KtFa6dQ5dXV0NtsvKyqJ///68/fbbpKSkEBUVxYkTJxg9erTB8Wq1Wnbs2GFwXz0p//d//4ePjw/t2rVj3759LF26lA0bNuh9QTBt2jQOHDhAdHQ0W7du5aeffiI1NdVomRXdR/3792f06NG0bNmS7OxsZaqjB2m1Wnx8fDhw4AARERGkpaUREhJCzZo1y93nrVu3lNCv2+bq1av06tWLl19+maSkJDZv3kxhYSG+vr7Ktbh06VK+//57wsLCSExMRKPREBsba1D+g21lStkV9WGRkZFs376dVatWcfToUb7++mtatGhh9Jya0o9qNBoWLVpEeHg433zzDfn5+crzGo2Gxo0bs3r1atLS0pg+fToLFy4kOjpabz+V9fnTp08nOTmZNWvWsGXLFjIzMw3em0yp6927dwkPD2flypVs2bKFH3/8kffff19ZQzQ6OpqEhAT++9//Gj0nZZlyHT/qdrSzs2PNmjVA6cj07OxsPv/8c37//XcSExMZMWKEQR8Hxt+foLSfmzdvHqmpqaxcuZKjR4/y2WefKc8HBwdz8uRJ1q1bx+HDhwkPD1e+QDl+/DiffPIJkyZN4siRI2zevJlu3bqVu58xY8YQFhYGoNyH5fn222+ZO3cuU6ZMIS0tjeDgYEJDQw3a5YsvvqBHjx4cPHjQ6C+nxJMl2VKy5V8hW4Lky39ivpRsKdmyLMmWki3Lkmz57JJsKdlSsuWfJFs+WZIt/z7Z8v79++Tk5BAeHs7//vc/fv31VzZu3KicD8mXj74tJV/+qSr5UpZLEM+UPXv2kJCQwLp164yOBmzTpg2DBg1i5syZ7Nq1y+D5DRs2oNVqWbZsmRKklixZQosWLUhISKBfv34Gr4mMjMTPz4+hQ4cCMGHCBPbv38+5c+f0trOysmLRokWYm5vj6OiIt7e3MmWKTqNGjZg/fz4qlYpWrVpx7tw5li1bxujRozl//jw7duxg+/btvP766wCsWLECJycnYmJilP3fv3+f+fPn88orryjl5uXlMXr0aFq1agWAWq1WnsvJycHBwcHoG7cxTk5OeqOeP/nkE3bu3MmWLVv49NNPuXLlCsXFxfTt21dZt65169Z6dfLy8lLK0I0MLY9ulJqlpaXR6ahMbbt79+6xYsUKbGxsqnS8VaEbhdyyZUvlsaysLHr06KH8vXjxYt577z3l7169emFmZsa9e/coLi4mMDAQLy8vg7K7d+9OSUkJe/fupVOnTsTExPDdd9+RnJyst11OTg5vvPFGlev+4KjiiIgI7O3tOXr0KG5ubuTl5dGoUSO6du1KzZo1sbe3Vz4M3Lx5kxs3btCzZ09efPFFAOWae1CtWrWU6awaNGhgtF3DwsLo16+f3ge0hQsX0rlzZ3799VeljBdeeIGQkJAqH++jtGrVKho1asTChQsxMzPD0dGRmTNnMn78eKZOnYpGoyE6OprIyEg8PDyA0oBb9r54UGX3kaWlJebm5hVO05acnMzhw4c5dOgQjo6OQPn3W7t27YDSsK7VanF2dsbd3V05trZt2+qFtRUrVuDg4MDx48dxcXEhMjKSjz/+WLmGPv/8c2UEbVkPtlVISEilZVfUh+Xl5dG8eXM6deqESqXC3t6+3A99gMn9qFar5fDhw/Tu3Zs7d+4ApcFPo9FQs2ZNpk6dqpTZrFkzMjIy2LBhg/J6qLjPLywsJCoqivDwcCUARkRE6LWtqXUtKSlhwYIFSn/z7rvvsmzZMs6ePatMIdarVy9SUlL07qPExES9X7e4ubkRGxtb6XVcp06dx9KOuunh/vWvfyn1Pnr0KFqt1mg/UpGPPvpI+X+zZs2YM2cOfn5+REZGYmZmRl5eHu3atVO+aCg7jWNeXh6Wlpa8/fbbWFlZAZT7Kx8onX5M90VTRffhl19+yezZs5X7w8HBgQsXLrBq1SoCAgKU7fr166d3HYmnS7KlZMu/SrYEyZf/xHwp2VKypWRLyZbGSLZ8Nkm2lGwp2TJZbzvJlk+WZMu/V7a8ceMGXl5e3L9/X8mXkydPBpB8KflSz9PMlzLIQDxTXn75ZQoKCvjiiy9wdXU1OkpoypQpdOjQga1btxpMu5SRkcHFixcN1oi6desWFy5cKLe8M2fOGNxULi4uBmHd0dFRmc4FwNbWVm9KH4AOHTrojRLt2LEjISEhFBQUkJ2djZmZGR07dlSer1evHq1bt+b06dPKY+bm5gYdykcffcTYsWNZu3Yt7u7ueHl5KR2fVqst97gqU1RUxBdffEFCQgJXr16lpKSEO3fu0KZNG6C0U+vSpQudOnXCw8ODLl260LdvX2XNpZEjRzJhwgSSkpJwd3fnnXfe0fuAUVWmtl2TJk0ee1AvT8uWLdm/fz8Ab7zxBsXFxXrPr1y5ktatW1NcXMzJkyeZNGkSlpaWTJs2TW+7GjVq4OvrS3R0NL///jsNGzbEzc3NIKhXt10vXLhASEgI6enp5Ofno9Fo0Gg0/PzzzwB4e3sTGRlJ+/bt6dq1K927d+ftt9/GwsKCBg0a4Ofnx4ABA3B3d6dz5854e3sbtElVZGRkkJOTw6ZNmwyO7cKFC0pQf5hr51HJzs7mtddew8zsz4mA3NzcuHfvHjk5OWi1WoqLi/VGeVtaWlYY1iu7j0yRmZmJra2tEtSN2bZtG9bW1mRmZjJ79myWL1+ufIjPyMjg4MGDBlMuQmk7tGjRgmvXrvHqq68qj6tUKpydnbl8+bLe9g+2VWVlu7i4VNiH+fn50a9fP1xcXOjatSuenp54enrqtYOOqf2oSqXiwIED3L59m2+//ZZjx46Rnp7OjRs3aNCgAV9//TVr1qwhLy+PO3fuUFxcjL29vd6+KurzL1y4wL179/TqUbduXaX/rEpdLSws9L4UsLGxoVGjRnprlNnY2BiMUu3UqROhoaHK37oR1pVdx23btgUefTuWp7r9GMDevXtZvHgxZ86coaCggPv373Pv3j2uXbtG48aNGT58OO+//z4ZGRl4eHjQs2dP5csNDw8P7OzsaN++Pd26dcPDw4M+ffooob2qrl+/zs8//8z48eOZOHGi8nhJSYnBMZb3Cw7x9Ei2LCXZ8q+ZLUHyZXU9K/lSsqVky7IkW/5JsqVky2eVZMtSki0lW+pItnyyJFv+fbKlmZmZkr1u377NvHnziIuL05uNRfKl5MuqeFz5UpZLEM8UW1tbtm/fTkFBAd7e3nrrl5RlZ2dHQEAAc+bM4f79+3rPaTQanJyclPWVdP+OHj3KBx98YHTfxqaPKuvBEbcqlapKnVBF25bdv4WFhcEUT5MnTyYtLY3evXtz+PBhXn/9daKiooDSNWRyc3P11qwxxfTp09m8eTNTpkxh+/bt7N+/HxcXF6WcGjVqsGnTJjZu3EibNm2Iiori1Vdf5cSJEwAMHTqUjIwMBg8ezLlz5+jRo4eyZll1mNp2lpaW1d6HqZo3bw6gTMEGpaNf1Wo1arW63OuladOmqNVqHB0d6devH4GBgSxZskQZiVjWkCFDiI+PZ/ny5cp6OuXVoez+TTVo0CCuX7/OkiVLSExMZN++fZibmyvtamdnR3p6OosXL8bKyopp06bRpUsXZY24ZcuWkZiYSKdOndixYwcdOnQgKSmpyvXQ0Wg0DB06VK9NU1JSOHbsmN6H0ifRrpXRarVG+4Kq3u86ld1HptbLFM2aNaNFixb079+fSZMm4e/vz927d4HSdujRo4fB/XXs2DG9X2CY0hc+2FamlF1RH/bKK6+QmZnJjBkz0Gg0BAYG4u3trbdelCnnomzdzczMUKvVtGnThvnz5+udh40bNzJ58mT8/PzYsGED+/fvZ/jw4QZ9aEV9viltYmpdy34Y0D1X3mMPno86deoofZJarVam26rsOtZ51O1YnubNm6NSqarcl126dImBAwfSqlUrVq9eTXJyMuHh4QBKO3l6enLixAnGjBlDfn4+AwcOVEYQW1lZsW/fPr755hvs7OxYvHgxHTt25MqVK1Wqh47u3C9atEjv3KSmpipTL+r8Ffoy8SfJlqUkWz79bAmSL/+J+VKypWTLsiRbSrbUkWz57JJsWUqypWTLsnWQbPnkSLb8e2VLXfZq06aNspyMbkp8yZelJF+a7nHlSxlkIJ45TZo0IS4ujlu3buHl5cVvv/1W7nYTJkzg+vXryjoqOu3btycnJ4eGDRvqdWBqtVqZDuVBrVq14tixY3qPPfi3qXTTq+gcOXKExo0bY21tzUsvvYRGo+Hw4cPK8wUFBZw8ebLSkX5Q2uGNHDmS9evX4+/vr3SOPj4+FBUV8dVXX5X7OmMfeg4dOsSgQYPo27cvbdu2pUmTJgajplUqFR07diQoKIg9e/bQuHFjvVGdTZs2ZdiwYaxevZopU6bw7bffVnocxlSn7R6Xrl270rBhQxYtWlTtMmrUqEFJSUm5H6LUajXOzs4cP3683DX8oLRdk5OTOX78uMFzGo2GgoICg8d/++03srOzmTBhAl26dMHR0ZGbN29SUlKit91zzz3HW2+9xbx589i9ezenTp0iLS1Ned7JyYmPP/6Y7du388Ybb7B27dqqHr6iffv2nDp1yqBN1Wo1tWvXrna5j8NLL73EkSNH9AJRamoqtWrV4sUXX0StVlOzZk29/uHWrVucPHmywnIruo9q1apl8KXDg9q3b8/Vq1eNrrdUnkGDBlFcXMzKlSuVMk6fPo29vb1BO1hZWVGvXj0aNWqkd2xarbbc66+8+lVUto6xPgxKg5W3tzeLFi1i/fr17Nu3j5ycHIN9Vbcf9fX1BUrXmktNTcXFxYWAgABeeeUV1Gq10V+MGKO7Fo4cOaI8VlRUpHctPGyfX12VXcfGPGw71qpVC0Dvem7QoAHdunVj5cqVFBYWGuzT2PvT8ePHuXfvHvPmzaNjx460aNGi3JD9/PPPM2jQIJYvX87SpUtZu3at8gHV3Nwcd3d3Zs6cyYEDBygqKiIhIaGCM2ecjY2N8h5ZXl8m/tokWxon2fLJknz5z8uXki0lW5pKsqVkS8mWzw7JlsZJtnyyJFtKtgTJln+XbKmzYsUKrly5IvnSCMmXxj2ufCmDDMQzydbWlri4OIqLi+nTpw/5+fkG29SvX5+JEycSGRmp97iPjw82Njb4+fmRkpJCbm4uBw4cYOrUqcpaVQ8aOXIk33//PVFRUZw/f57Q0FDS09NNGhn3oKtXrxIUFMTZs2fZsmULYWFhyuik5s2b06tXL8aPH8/BgwfJysoiICAAKysrfHx8jJZ5+/ZtPvnkE/bv38/FixdJT0/XW+OoQ4cOjBs3jhkzZjBlyhQOHTrEpUuX2L9/PwEBAQbnSKd58+bExcXx448/KnXRdXBQ+kHjyy+/5NixY+Tl5REfH8/ly5eV/U6aNInExERyc3PJzMwkMTHxod58qtN2j0JBQQGZmZl6/65fv87SpUtJSkpiwIABJCYmcuHCBbKysggPD+fOnTsGo7Z/++03rl27xuXLl9m1axeRkZG8+eabWFtbl7vf2NhYzp07h62tbbnPBwYG4ubmpkwRlpmZSW5uLlu3bqVnz55kZGQYvKZ+/fo8//zzrFmzhpycHFJSUpgwYYLeyMLvvvuONWvWkJWVRW5uLt999x01a9ZErVaTm5vLrFmzSEtL49KlS+zbt4+srKyHatdx48Zx7Ngxxo8fr0w/tnPnTj7++ONql/kolNfunp6eXL16lYkTJ5KdnU1CQgKzZ89mxIgR1KlTh7p16zJkyBBmzpzJ3r17OX36NGPHjq1w9GVl99ELL7xAXl4eP/74I/n5+Xr3oI67uzsdOnRg6NChJCUlkZuby549e4iLizN6fGZmZsqI9KKiIj788EMKCgr44IMPSE9PJzc3l+TkZMaNG8fNmzeB0r4wNDSUbdu2cfbsWaZOncq1a9cq7QsrK7uyPiw8PJzY2Fiys7PJyckhJiYGa2trZXRrWdXtR3UjzyMiImjRogWZmZns2rWL8+fPM3/+fA4ePFjhMT6obt26+Pv7M2vWLPbs2cOpU6cYPXq0Xjiubl0f1vDhwyu8jo152Ha0t7dHpVKRkJDA9evXlWC+YMECtFotHh4ebN68mbNnz3LmzBlWrVpldO3G5s2bo9FoWLZsGbm5ucTGxhq8l4WEhBAXF8f58+fJzs5m27ZtODg4YGFhwc6dO1m+fDkZGRlcunSJmJgYCgsLq7W+mk5QUBBhYWFERERw9uxZTp48ydq1ax/qCx3x5Ei21CfZ8vFmS5B8+U/Ml5ItJVtKttQn2bJiki2fbZIt9Um2lGwp2fLRk2z5z8qWOi1btmTBggWSL42QfFmxx5EvzSvfRIi/JhsbG7Zt20bfvn3p06cPbm5uBtsEBASwYsUKZb0mKJ2CJT4+nlmzZjFs2DAKCgqwtbXlzTffNLpW2oABA8jNzWX27Nncvn2bd955h//85z/Ex8dXud4+Pj5oNBq6deuGSqXC399fCetQOp1TUFAQvr6+3L17F1dXV2JjYyscFVmjRg3++OMPAgMD+eWXX2jYsCFvvfUWc+fOVbaZPXs2zs7OrFy5kujoaEpKSmjWrBm9evXiww8/LLfckJAQxowZQ69evahfvz6BgYF6QcHa2pq0tDS++uorbty4QdOmTfn0008ZOHAgUDoi9bPPPuPy5cvUrVsXd3d3goODq3zOdKrTdo9CamoqnTt31nvMy8uLNWvWsGvXLkJDQxk1ahT5+flYWVnRvn17IiIieO+99/ReM2DAAKC0vWxtbfH09GT69OlG91unTp0K3zQtLCzYtGkTy5cvJyoqilmzZinrHw0ePBhXV1eD15iZmfH1118TFBSEm5sbarWa4OBgvbX76tWrR2hoKNOmTaOkpARHR0eioqJwcHDgl19+4dy5cwwbNoz8/HxsbGzw8fF5qFDdtm1b4uPjCQ4O5p133uH+/fs4ODjQu3fvapf5KBhr95iYGGbMmMGbb75JvXr1ePfdd5kxY4ayzdy5cykqKsLX1xdLS0s++ugjfvnlF2VNqQdVdh95eXkpfd2NGzeIiIhg8ODBemWYmZkp9QoICKCwsBAHBweCgoIqPMYhQ4Ywb948IiMjmThxohLYBgwYwN27d7Gzs8PDwwMLCwsAxowZw7Vr1xg1ahQqlYrBgwfTu3dvfv311wr307hx40rLrqgPs7KyIiwsjJycHFQqFU5OTsTExBi9P6rTj+rExMQwadIkTpw4wYcffohWq8XLy4tRo0YRHR1d6evL0l0LQ4YMoXbt2gQEBHDr1q1HVtfqatKkSaXXcXketh2bNGnC5MmTCQ4OZuzYscooXQcHB/bu3cvChQuZOXMmV65coWHDhrRt25bFixeXW5e2bdvy+eefExoaSkhICB07dmTu3Ll6U1BaWFgQHBzMxYsXsbCw4LXXXuOHH34ASvu57du3M3/+fG7fvs2LL75IWFgYnTp1qvZ5HTp0KHXq1CEsLIw5c+bw3HPP8fLLLzNixIhqlymeLMmWf5Js+XizJUi+/CfmS8mWki0lW+qTbFkxyZbPPsmWf5JsKdlSsuWjJ9nyn5kthw8fzqRJk0hPT5d8WQ7JlxV7HPlS9ccff1R9IRYhBIMHD6akpIR169Y97aoIIf7C7t69i5OTE2PGjGHMmDFPuzqPXOfOnXF1deXLL7982lURQohnmmRLIYQpJFsKIYQwhWRLIYQpJFsKIR6GzGQghAlu3brFqlWr6N69O+bm5mzdupX4+HiDddOEECIjI4MzZ87g4uLCzZs3CQ0NpbCwkP79+z/tqj20S5cusXv3bl5//XVKSkpYvXo1P/30E6GhoU+7akII8UyRbCmEMJVkSyGEEJWRbCmEMJVkSyHEoySDDIQwgUqlIjExkUWLFnHnzh3UajUrVqygT58+T7tqQoi/oIiICM6dO0eNGjVwcnIiPj6epk2bPu1qPTQzMzN++OEHZsyYgUajwdHRkdjYWJydnZ921YQQ4pki2VIIURWSLYUQQlREsqUQoiokWwohHhVZLkEIIYQQQgghhBBCCCGEEEIIIYQQJjF72hUQQgghhBBCCCGEEEIIIYQQQgghxLNBBhkIIYQQQgghhBBCCCGEEEIIIYQQwiQyyEAIIYQQQgghhBBCCCGEEEIIIYQQJpFBBkIIIYQQQgghhBBCCCGEEEIIIYQwiQwyEEIIIYQQQgghhBBCCCGEEEIIIYRJ/h/Pm4x1V+yDrQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 2160x360 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2160x360 with 3 Axes>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plt.style.use('fivethirtyeight')\n",
    "col.plot_model_encoder_pairs(reporter, title='adult'.capitalize())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# testing embeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# define fitting function\n",
    "mte_transformer = col.MeanTargetEncoder(feature_selection)\n",
    "\n",
    "\n",
    "pipe = col.CategoricalPipeline(features=feature_selection,\n",
    "                               transformer=col.MeanTargetEncoder(feature_selection),\n",
    "                               scaler=MaxAbsScaler(),\n",
    "                               model=RandomForestClassifier(n_estimators=100, max_depth=5))\n",
    "\n",
    "cv_score = col.cv_score(pipeline=pipe,\n",
    "                        data=df, \n",
    "                        kf=kf,\n",
    "                        scorer=reporter.scorer)\n",
    "\n",
    "reporter.add_to_report(pipe.config, cv_score, show=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}