{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import random\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import classification_report\n",
    "import lightgbm\n",
    "from sklearn.model_selection import StratifiedKFold, RandomizedSearchCV\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.metrics import log_loss\n",
    "from sklearn.calibration import calibration_curve\n",
    "from sklearn.calibration import CalibratedClassifierCV\n",
    "from sklearn.preprocessing import KBinsDiscretizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "dir = \"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "odds_df = pd.read_csv(dir + \"Code/EDA/stacked_df.csv\")\n",
    "\n",
    "odds_df[\"B365_bookspread\"] = np.where(odds_df[\"B365_bookspread\"]==0,\n",
    "                                      -1*odds_df[\"B365_bookspread\"],\n",
    "                                      odds_df[\"B365_bookspread\"])\n",
    "\n",
    "odds_df =  odds_df[[\"match_id\", \"team_id\", \"B365_bookspread\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "match_df = pd.read_csv(dir + \"Data/match_clean.csv\")\n",
    "\n",
    "match_df = match_df.drop('Unnamed: 0',axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.5780346820809249, 0.7462686567164178, 0.29411764705882354)"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1/1.73, 1/1.34, 1/3.4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "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>BSH</th>\n",
       "      <th>BSD</th>\n",
       "      <th>BSA</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>1.73</td>\n",
       "      <td>3.40</td>\n",
       "      <td>4.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>2.50</td>\n",
       "      <td>3.20</td>\n",
       "      <td>2.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>2.00</td>\n",
       "      <td>3.25</td>\n",
       "      <td>3.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>2.40</td>\n",
       "      <td>2.88</td>\n",
       "      <td>2.88</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>6.50</td>\n",
       "      <td>3.75</td>\n",
       "      <td>1.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16301</th>\n",
       "      <td>3.20</td>\n",
       "      <td>3.40</td>\n",
       "      <td>2.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16306</th>\n",
       "      <td>2.10</td>\n",
       "      <td>3.40</td>\n",
       "      <td>3.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16307</th>\n",
       "      <td>1.45</td>\n",
       "      <td>4.40</td>\n",
       "      <td>6.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16308</th>\n",
       "      <td>2.20</td>\n",
       "      <td>3.40</td>\n",
       "      <td>3.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16310</th>\n",
       "      <td>1.45</td>\n",
       "      <td>4.20</td>\n",
       "      <td>7.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14185 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        BSH   BSD   BSA\n",
       "22     1.73  3.40  4.20\n",
       "23     2.50  3.20  2.50\n",
       "24     2.00  3.25  3.25\n",
       "25     2.40  2.88  2.88\n",
       "26     6.50  3.75  1.44\n",
       "...     ...   ...   ...\n",
       "16301  3.20  3.40  2.15\n",
       "16306  2.10  3.40  3.30\n",
       "16307  1.45  4.40  6.50\n",
       "16308  2.20  3.40  3.15\n",
       "16310  1.45  4.20  7.00\n",
       "\n",
       "[14185 rows x 3 columns]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "match_df[['BSH', 'BSD', 'BSA']].dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['B365H', 'B365D', 'B365A']\n",
      "['BWH', 'BWD', 'BWA']\n",
      "['IWH', 'IWD', 'IWA']\n",
      "['LBH', 'LBD', 'LBA']\n",
      "['PSH', 'PSD', 'PSA']\n",
      "['WHH', 'WHD', 'WHA']\n",
      "['SJH', 'SJD', 'SJA']\n",
      "['VCH', 'VCD', 'VCA']\n",
      "['GBH', 'GBD', 'GBA']\n",
      "['BSH', 'BSD', 'BSA']\n"
     ]
    }
   ],
   "source": [
    "odds_col = match_df.columns[32:63].tolist()\n",
    "\n",
    "odds_df = match_df[[\"date\", \"match_id\", \"home_team_id\", \"away_team_id\"]]\n",
    "\n",
    "for i in range(0, len(odds_col), 3):\n",
    "    print(odds_col[i:i+3])\n",
    "    \n",
    "    cur_odds_df = match_df.loc[~match_df[odds_col[i]].isnull(), \n",
    "                               [\"date\", \"match_id\", \"home_team_id\"] + odds_col[i:i+3]]\n",
    "    \n",
    "    inv_odds = (1/cur_odds_df[odds_col[i:i+3]])\n",
    "\n",
    "    booksum = inv_odds.sum(axis=1)\n",
    "\n",
    "    margin = booksum - 1\n",
    "\n",
    "    inv_odds[\"booksum\"] = booksum\n",
    "\n",
    "    bookprob = pd.concat([inv_odds[odds_col[i]]/inv_odds[\"booksum\"], \n",
    "                          inv_odds[odds_col[i+1]]/inv_odds[\"booksum\"], \n",
    "                          inv_odds[odds_col[i+2]]/inv_odds[\"booksum\"]], axis=1)\n",
    "\n",
    "    bookprob.columns = odds_col[i:i+3]\n",
    "    \n",
    "    bookprob[\"{}_bookspread\".format(odds_col[i][:-1])] = bookprob[odds_col[i]] - bookprob[odds_col[i+2]]\n",
    "    \n",
    "    bookprob[\"{}_margin\".format(odds_col[i][:-1])] = margin\n",
    "    \n",
    "    bookprob[[\"date\", \"match_id\", \"home_team_id\"]] = cur_odds_df[[\"date\", \"match_id\", \"home_team_id\"]]\n",
    "    \n",
    "    odds_df = odds_df.merge(bookprob, \n",
    "                            on=[\"date\", \"match_id\", \"home_team_id\"],\n",
    "                            how=\"left\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "wl_odds_player_df = pd.read_csv(dir + \"Data/wl_odds_player_df.csv\")\n",
    "\n",
    "wl_odds_player_df = wl_odds_player_df.drop('Unnamed: 0', axis=1)\n",
    "\n",
    "odds_df = odds_df.loc[odds_df[\"match_id\"].isin(set(wl_odds_player_df[\"match_id\"]))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-7-10b0c71b543d>:6: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  away_df[\"B365_bookspread\"] = -1*away_df[\"B365_bookspread\"]\n"
     ]
    }
   ],
   "source": [
    "home_df = odds_df[[\"match_id\", \"home_team_id\", \"B365_bookspread\"]]\n",
    "home_df.columns = [\"match_id\", \"team_id\", \"B365_bookspread\"]\n",
    "\n",
    "away_df = odds_df[[\"match_id\", \"away_team_id\", \"B365_bookspread\"]]\n",
    "away_df.columns = [\"match_id\", \"team_id\", \"B365_bookspread\"]\n",
    "away_df[\"B365_bookspread\"] = -1*away_df[\"B365_bookspread\"]\n",
    "\n",
    "odds_df = pd.concat([home_df, away_df])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(['2010/2011', '2011/2012', '2012/2013'], '2013/2014')"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "seasons = sorted(wl_odds_player_df[\"season\"].unique())\n",
    "\n",
    "i=4\n",
    "seasons[i:i+4][:-1], seasons[i:i+4][-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "i = 2\n",
    "    \n",
    "train_df = wl_odds_player_df.loc[wl_odds_player_df[\"season\"].isin(seasons[i:i+4][:-1])].dropna()\n",
    "test_df = wl_odds_player_df.loc[wl_odds_player_df[\"season\"].isin([seasons[i:i+4][-1]])].dropna()\n",
    "\n",
    "train_dates = sorted(train_df[\"date\"].unique())\n",
    "\n",
    "calib_dates = random.sample(train_dates, int(len(train_dates)*0.25))\n",
    "\n",
    "calib_df = train_df.loc[train_df[\"date\"].isin(calib_dates)]\n",
    "\n",
    "train_df = train_df.loc[~train_df[\"date\"].isin(calib_dates)]\n",
    "\n",
    "feat_col = [c for c in train_df.columns if \"avg\" in c]\n",
    "feat_col = feat_col + [\"rating_diff\"]\n",
    "\n",
    "skf = StratifiedKFold(n_splits=6, shuffle=True)\n",
    "\n",
    "lgb_param_grid = {'lgb__n_estimators': [100, 200, 300,],\n",
    "                 'lgb__num_leaves':  [2, 4, 6, 10, 15, 20],\n",
    "                 'lgb__subsample': [0.4, 0.6, 0.8, 1],\n",
    "                 'lgb__colsample_bytree': [0.4, 0.6, 0.8, 1],\n",
    "                 'lgb__is_unbalance': [False]}\n",
    "\n",
    "lgb_pipe = Pipeline([(\"lgb\", lightgbm.LGBMClassifier())])\n",
    "\n",
    "#lose\n",
    "lgb_cv_lose = RandomizedSearchCV(lgb_pipe, \n",
    "                                 param_distributions=lgb_param_grid, \n",
    "                            n_jobs=-1,\n",
    "                            cv=skf, \n",
    "                                 n_iter=50,\n",
    "                                 refit=True,\n",
    "                                )\n",
    "\n",
    "lgb_cv_lose.fit(train_df.drop([\"result\", \"win\", \"lose\", \"draw\", \"team_id\", \"m_rating\",\n",
    "                          \"cumulative_lose\", \"cumulative_win\",\n",
    "                          \"season\", \"date\", \"stage\", \"match_id\", 'match_num'], axis=1),\n",
    "                train_df[\"lose\"])\n",
    "\n",
    "\n",
    "lgb_calib_lose = CalibratedClassifierCV(lgb_cv_lose, method='sigmoid', cv=5)\n",
    "\n",
    "lgb_calib_lose.fit(calib_df.drop([\"result\", \"win\", \"lose\", \"draw\", \"team_id\", \"m_rating\",\n",
    "                             \"cumulative_lose\", \"cumulative_win\",\n",
    "                             \"season\", \"date\", \"stage\", \"match_id\", 'match_num'], axis=1),\n",
    "              calib_df[\"lose\"])\n",
    "\n",
    "lgb_calib_pred_lose = lgb_calib_lose.predict_proba(test_df.drop([\"result\", \"win\", \"lose\", \"draw\", \"team_id\", \"m_rating\",\n",
    "                                                                 \"cumulative_lose\", \"cumulative_win\",\n",
    "                                                                 \"season\", \"date\", \"stage\", \"match_id\", 'match_num'], axis=1))\n",
    "\n",
    "\n",
    "#win\n",
    "lgb_cv_win = RandomizedSearchCV(lgb_pipe, \n",
    "                                 param_distributions=lgb_param_grid, \n",
    "                            n_jobs=-1,\n",
    "                            cv=skf, \n",
    "                                 n_iter=50,\n",
    "                                 refit=True,\n",
    "                                )\n",
    "\n",
    "lgb_cv_win.fit(train_df.drop([\"result\", \"win\", \"lose\", \"draw\", \"team_id\", \"m_rating\",\n",
    "                          \"cumulative_lose\", \"cumulative_win\",\n",
    "                          \"season\", \"date\", \"stage\", \"match_id\", 'match_num'], axis=1),\n",
    "                train_df[\"win\"])\n",
    "\n",
    "\n",
    "lgb_calib_win = CalibratedClassifierCV(lgb_cv_win, method='sigmoid', cv=5)\n",
    "\n",
    "lgb_calib_win.fit(calib_df.drop([\"result\", \"win\", \"lose\", \"draw\", \"team_id\", \"m_rating\",\n",
    "                             \"cumulative_lose\", \"cumulative_win\",\n",
    "                             \"season\", \"date\", \"stage\", \"match_id\", 'match_num'], axis=1),\n",
    "              calib_df[\"win\"])\n",
    "\n",
    "lgb_calib_pred_win = lgb_calib_win.predict_proba(test_df.drop([\"result\", \"win\", \"lose\", \"draw\", \"team_id\", \"m_rating\",\n",
    "                                                                 \"cumulative_lose\", \"cumulative_win\",\n",
    "                                                                 \"season\", \"date\", \"stage\", \"match_id\", 'match_num'], axis=1))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['league_id', 'rating_diff', 'home', 'match_bin', 'avg_overall_rating',\n",
       "       'avg_volleys_imp', 'avg_long_passing', 'avg_finishing',\n",
       "       'avg_ball_control', 'avg_dribbling', 'avg_short_passing',\n",
       "       'avg_reactions', 'avg_vision_imp', 'avg_penalties', 'avg_agility_imp',\n",
       "       'cumulative_result'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(train_df.drop([\"result\", \"win\", \"lose\", \"draw\", \"team_id\", \"m_rating\",\n",
    "                \"cumulative_lose\", \"cumulative_win\",\n",
    "                \"season\", \"date\", \"stage\", \"match_id\", 'match_num'], axis=1).columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "lgb_cv_pred_lose = lgb_calib_win.predict_proba(test_df.drop([\"result\", \"win\", \"lose\", \"draw\", \"team_id\", \"m_rating\",\n",
    "                                                                 \"cumulative_lose\", \"cumulative_win\",\n",
    "                                                                 \"season\", \"date\", \"stage\", \"match_id\", 'match_num'], axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "lgb_cv_pred_win = lgb_calib_win.predict_proba(test_df.drop([\"result\", \"win\", \"lose\", \"draw\", \"team_id\", \"m_rating\",\n",
    "                                                                 \"cumulative_lose\", \"cumulative_win\",\n",
    "                                                                 \"season\", \"date\", \"stage\", \"match_id\", 'match_num'], axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_pred_df = pd.DataFrame({\"lose_pred\":lgb_calib_pred_lose[:,1], \"win_pred\":lgb_calib_pred_win[:,1]})\n",
    "\n",
    "test_pred_df.columns = [\"lose_pred\", \"win_pred\"]\n",
    "\n",
    "test_pred_df[\"sum_pred\"] = test_pred_df.sum(axis=1)\n",
    "\n",
    "test_pred_df[\"lose_pred\"] = test_pred_df[\"lose_pred\"] / test_pred_df[\"sum_pred\"]\n",
    "\n",
    "test_pred_df[\"win_pred\"] = test_pred_df[\"win_pred\"] / test_pred_df[\"sum_pred\"]\n",
    "\n",
    "test_pred_df = pd.concat([test_pred_df, test_df.reset_index(drop=True)], axis=1)\n",
    "\n",
    "test_pred_df = test_pred_df.merge(odds_df, on=[\"team_id\", \"match_id\"], how=\"left\")\n",
    "\n",
    "test_pred_df = test_pred_df.dropna()\n",
    "\n",
    "test_pred_df[\"pred_spread\"] = test_pred_df[\"win_pred\"] - test_pred_df[\"lose_pred\"]\n",
    "\n",
    "test_pred_df = test_pred_df.loc[test_pred_df[\"home\"]==1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "kbin = KBinsDiscretizer(n_bins=10, encode='ordinal', strategy=\"kmeans\")\n",
    "\n",
    "kbin.fit(test_pred_df[\"pred_spread\"].values.reshape(-1, 1))\n",
    "\n",
    "bins = kbin.transform(test_pred_df[\"pred_spread\"].values.reshape(-1, 1))\n",
    "\n",
    "test_pred_df[\"pred_spread_bins\"] = bins"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"3\" halign=\"left\">pred_spread</th>\n",
       "      <th colspan=\"3\" halign=\"left\">win</th>\n",
       "      <th colspan=\"3\" halign=\"left\">lose</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>min</th>\n",
       "      <th>max</th>\n",
       "      <th>mean</th>\n",
       "      <th>sum</th>\n",
       "      <th>size</th>\n",
       "      <th>mean</th>\n",
       "      <th>sum</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pred_spread_bins</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.0</th>\n",
       "      <td>-0.667663</td>\n",
       "      <td>-0.747637</td>\n",
       "      <td>-0.595472</td>\n",
       "      <td>0.144578</td>\n",
       "      <td>12</td>\n",
       "      <td>83</td>\n",
       "      <td>0.710843</td>\n",
       "      <td>59</td>\n",
       "      <td>83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>-0.520336</td>\n",
       "      <td>-0.592088</td>\n",
       "      <td>-0.446229</td>\n",
       "      <td>0.154639</td>\n",
       "      <td>15</td>\n",
       "      <td>97</td>\n",
       "      <td>0.649485</td>\n",
       "      <td>63</td>\n",
       "      <td>97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>-0.372072</td>\n",
       "      <td>-0.444808</td>\n",
       "      <td>-0.295559</td>\n",
       "      <td>0.235669</td>\n",
       "      <td>37</td>\n",
       "      <td>157</td>\n",
       "      <td>0.490446</td>\n",
       "      <td>77</td>\n",
       "      <td>157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>-0.218130</td>\n",
       "      <td>-0.292947</td>\n",
       "      <td>-0.145375</td>\n",
       "      <td>0.275641</td>\n",
       "      <td>43</td>\n",
       "      <td>156</td>\n",
       "      <td>0.410256</td>\n",
       "      <td>64</td>\n",
       "      <td>156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4.0</th>\n",
       "      <td>-0.068091</td>\n",
       "      <td>-0.142719</td>\n",
       "      <td>0.013661</td>\n",
       "      <td>0.321429</td>\n",
       "      <td>72</td>\n",
       "      <td>224</td>\n",
       "      <td>0.450893</td>\n",
       "      <td>101</td>\n",
       "      <td>224</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5.0</th>\n",
       "      <td>0.099476</td>\n",
       "      <td>0.015003</td>\n",
       "      <td>0.171442</td>\n",
       "      <td>0.415789</td>\n",
       "      <td>158</td>\n",
       "      <td>380</td>\n",
       "      <td>0.326316</td>\n",
       "      <td>124</td>\n",
       "      <td>380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6.0</th>\n",
       "      <td>0.245948</td>\n",
       "      <td>0.172990</td>\n",
       "      <td>0.315648</td>\n",
       "      <td>0.461224</td>\n",
       "      <td>226</td>\n",
       "      <td>490</td>\n",
       "      <td>0.267347</td>\n",
       "      <td>131</td>\n",
       "      <td>490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7.0</th>\n",
       "      <td>0.388882</td>\n",
       "      <td>0.318984</td>\n",
       "      <td>0.464139</td>\n",
       "      <td>0.559783</td>\n",
       "      <td>206</td>\n",
       "      <td>368</td>\n",
       "      <td>0.190217</td>\n",
       "      <td>70</td>\n",
       "      <td>368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8.0</th>\n",
       "      <td>0.538671</td>\n",
       "      <td>0.465553</td>\n",
       "      <td>0.609792</td>\n",
       "      <td>0.652330</td>\n",
       "      <td>182</td>\n",
       "      <td>279</td>\n",
       "      <td>0.139785</td>\n",
       "      <td>39</td>\n",
       "      <td>279</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9.0</th>\n",
       "      <td>0.679140</td>\n",
       "      <td>0.610537</td>\n",
       "      <td>0.772113</td>\n",
       "      <td>0.829060</td>\n",
       "      <td>194</td>\n",
       "      <td>234</td>\n",
       "      <td>0.029915</td>\n",
       "      <td>7</td>\n",
       "      <td>234</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 pred_spread                           win            \\\n",
       "                        mean       min       max      mean  sum size   \n",
       "pred_spread_bins                                                       \n",
       "0.0                -0.667663 -0.747637 -0.595472  0.144578   12   83   \n",
       "1.0                -0.520336 -0.592088 -0.446229  0.154639   15   97   \n",
       "2.0                -0.372072 -0.444808 -0.295559  0.235669   37  157   \n",
       "3.0                -0.218130 -0.292947 -0.145375  0.275641   43  156   \n",
       "4.0                -0.068091 -0.142719  0.013661  0.321429   72  224   \n",
       "5.0                 0.099476  0.015003  0.171442  0.415789  158  380   \n",
       "6.0                 0.245948  0.172990  0.315648  0.461224  226  490   \n",
       "7.0                 0.388882  0.318984  0.464139  0.559783  206  368   \n",
       "8.0                 0.538671  0.465553  0.609792  0.652330  182  279   \n",
       "9.0                 0.679140  0.610537  0.772113  0.829060  194  234   \n",
       "\n",
       "                      lose            \n",
       "                      mean  sum size  \n",
       "pred_spread_bins                      \n",
       "0.0               0.710843   59   83  \n",
       "1.0               0.649485   63   97  \n",
       "2.0               0.490446   77  157  \n",
       "3.0               0.410256   64  156  \n",
       "4.0               0.450893  101  224  \n",
       "5.0               0.326316  124  380  \n",
       "6.0               0.267347  131  490  \n",
       "7.0               0.190217   70  368  \n",
       "8.0               0.139785   39  279  \n",
       "9.0               0.029915    7  234  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_pred_df.groupby(\"pred_spread_bins\").agg({\"pred_spread\": [\"mean\", \"min\", \"max\"],\n",
    "                                              \"win\": [\"mean\", \"sum\", \"size\"],\n",
    "                                              \"lose\": [\"mean\", \"sum\", \"size\"]})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_pred_df[\"pred_spread_bins\"] = 2\n",
    "\n",
    "test_pred_df.loc[(test_pred_df[\"pred_spread\"]>= -0.079694) & \n",
    "                (test_pred_df[\"pred_spread\"]<= 0.059606), \"pred_spread_bins\"] = 4\n",
    "\n",
    "\n",
    "test_pred_df.loc[(test_pred_df[\"pred_spread\"]>= 0.060011), \"pred_spread_bins\"] = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "bet_win_bins = [9,8,7,6,5]\n",
    "bet_lose_bins = [0]\n",
    "bet_draw_bins = [4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "out_test_pred_df = test_pred_df.copy()\n",
    "\n",
    "out_test_pred_df[\"target\"] = 0\n",
    "\n",
    "out_test_pred_df.loc[out_test_pred_df[\"pred_spread_bins\"].isin(bet_win_bins), \"target\"] = 1\n",
    "\n",
    "#out_test_pred_df.loc[out_test_pred_df[\"pred_spread_bins\"].isin([bet_lose_bins]), \"target\"] = -1\n",
    "\n",
    "out_test_pred_df.loc[out_test_pred_df[\"pred_spread_bins\"].isin(bet_draw_bins), \"target\"] = -1\n",
    "\n",
    "out_test_pred_df.to_csv(dir + \"Data/bet_spread_df.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_pred_df[\"bet\"] = \"no bet\"\n",
    "\n",
    "test_pred_df.loc[test_pred_df[\"pred_spread_bins\"].isin(bet_win_bins), \"bet\"] = \"bet win\"\n",
    "\n",
    "test_pred_df.loc[test_pred_df[\"pred_spread_bins\"].isin(bet_lose_bins), \"bet\"] = \"bet lose\"\n",
    "\n",
    "test_pred_df.loc[test_pred_df[\"pred_spread_bins\"].isin(bet_draw_bins), \"bet\"] = \"bet draw\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors1 = {'bet lose':'pink', 'bet win':'green', 'bet draw': 'blue', \"no bet\":'orange'}\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(12,10))\n",
    "for k in colors1.keys():\n",
    "    #ix = np.where(test_pred_df.loc[test_pred_df[\"home\"]==0,'bet'] == k)[0]\n",
    "    #ax.scatter(test_pred_df.loc[test_pred_df[\"home\"]==0, \"pred_spread\"].values[ix],\n",
    "    #           test_pred_df.loc[test_pred_df[\"home\"]==0, \"B365_bookspread\"].values[ix],\n",
    "    #           c = colors1[k], label = k, s = 30)\n",
    "    \n",
    "    ix = np.where(test_pred_df['bet'] == k)[0]\n",
    "    ax.scatter(test_pred_df[\"pred_spread\"].values[ix],\n",
    "               test_pred_df[\"B365_bookspread\"].values[ix],\n",
    "               c = colors1[k], label = k, s = 30)\n",
    "ax.legend(prop={'size': 19})\n",
    "ax.set_xlabel('predicted spread', fontsize=19)\n",
    "ax.set_ylabel(\"bookmaker's spread\", fontsize=19)\n",
    "ax.xaxis.set_tick_params(labelsize=15)\n",
    "ax.yaxis.set_tick_params(labelsize=15)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_pred_df[\"bet_accuracy\"] = \"incorrect\"\n",
    "\n",
    "test_pred_df.loc[((test_pred_df[\"result\"]== 1) & (test_pred_df[\"bet\"]==\"bet win\") |\n",
    "                  (test_pred_df[\"result\"]== 0) & (test_pred_df[\"bet\"]==\"bet draw\")), \n",
    "                 \"bet_accuracy\"] = \"correct\"\n",
    "\n",
    "test_pred_df.loc[(test_pred_df[\"bet\"]==\"no bet\"), \"bet_accuracy\"] = \"no bet\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAu8AAAJbCAYAAABU2HoyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeXxU9b34/9eZhSEhi2AICAwiVUisECpxpogouFSRC2q1aBEBr9KAJXZ17u+28hCLtXVsrf1GkRT7FRTwSvkqaiW3dWMx6gTQBJRAtSgMi0AEshASZjLn98dkhlnObMkkmcH3sw9u5Kyfc+aE+z6feX/eH0VVVYQQQgghhBCpT9fTDRBCCCGEEELER4J3IYQQQggh0oQE70IIIYQQQqQJCd6FEEIIIYRIExK8CyGEEEIIkSYkeBdCCCGEECJNGHq6AekiLy9PHTZsWE83QwghhBBCnMW2bdtWp6pq/0jrJXiP07Bhw9i6dWtPN0MIIYQQQpzFFEXZG229pM0IIYQQQgiRJiR4F0IIIYQQIk1I8C6EEEIIIUSakOBdCCGEEEKINCHBuxBCCCGEEGlCgnchhBBCCCHShATvQgghhBBCpAkJ3oUQQgghhEgTMklTkrW0tHD06FFaWlpwu9093RxxFjAajeTn55OTk9PTTRFCCCFED5PgPYnq6+s5fPgw/fv3Z+DAgRgMBhRF6elmiTSmqiqnTp3iwIEDABLACyGEEN9wkjaTRHV1dQwZMoS+fftiNBolcBedpigKmZmZDB48mCNHjvR0c4QQQgjRwyR4T6LTp0+TkZHR080QZ6GMjAxcLldPN0MIIYQQPUyC9yST3nbRFeS5EkIIIQRI8C6EEEIIIUTakOBdCCGEEEKINCHBu0iKDRs2oCgKixYt6ummCCGEEEKctSR4F6IHDBs2jGHDhvV0M4QQQgiRZqTOu0gKi8VCbW0teXl5Pd0UIYQQQoizlgTvIikyMzMpKCjo6WYIIYQQQpzVJG1GJIVWzruiKEycOJFDhw5x5513cu6555KZmcnEiRP56KOPNI+ze/du5syZg9lsxmQyMWjQIKZMmcKbb74ZtN1XX33Ffffdh9lsplevXgwePJi5c+dy8ODBsGP6UlTq6uq49957GThwIDqdjurqapYvX46iKCxfvpy//e1vXHbZZWRmZnLzzTcHtWnmzJkMGjQIk8nEsGHDeOCBB2hsbNS8hpdeeolJkybRt29fMjMzGTFiBAsWLODw4cN8+eWXKIrC3r172bt3L4qi+P8sX7488RsvhBBCiG8U6XkXXer48eNcccUV5OXlMXv2bPbu3cvLL7/MNddcQ21tLQMHDvRv++677zJ16lRaWlqYOnUqhYWFHDlyhPfff59Vq1Zx3XXXAd7A3Wq1sm/fPm644QZmzpzJzp07efbZZ6moqODDDz9kyJAhQe1obW3lmmuuwe12c/vtt3Pq1CkyMzP961988UU2bNjATTfdxNVXX01ubi4A7733HpMnT8btdnPTTTdhNpupqanhD3/4Axs2bOC9997DZDL5j/PjH/+YJUuWMHDgQG6//XZyc3PZs2cPL7zwArfddhtjxozhoYce4sknnwTgpz/9qX/fMWPGJP8DEEIIIURcnPVO7JV2HAccWAdbsY23Yc4193SzwqmqKn/i+DN27Fg1lp07d8bc5mz17rvvqoD60EMP+ZcBKqDef//9qsfj8S9ftGiRCqiPPvqof1lzc7M6cOBA1Wg0qpWVlWHH379/v/+/77rrLhVQH3vssaBtnn76aRVQb7vttqDl559/vgqoU6dOVVtbW4PWPffccyqg6vV6dfPmzUHrWltbVbPZrJ577rnqv/71r6B1TzzxRFgbXnnlFRVQv/vd76oNDQ1B2zc2NqonTpwIatP5558fdp3RfJOfLyGEEKIr7TuxT+37+76q8TdGlUWoxt8Y1b6/76vuO7Gv29sCbFWjxKSSNpNunE4oLQWLxfvT6ezpFkXVp08fHn300aAZQufMmQPA1q1b/cteffVVvvrqK370ox9x+eWXhx1n8ODBgLcHfc2aNQwZMoSf/exnQdvMmzePCy+8kFdeeUUzpeV3v/sdvXr10mznLbfcwhVXXBG07PXXX8fpdPLggw9y0UUXBa37yU9+Qn5+Pi+99JJ/2dKlSwEoKysjOzs7aPusrCx/b74QQgghUou90k7T6SZcHhcALo+LptNN2CvtPdyycJI2k06cTigqgqYmcLmguhpWrYKaGjCn4Nc6wIgRI+jTp0/QMl8gfuLECf8yXyD/ve99L+rxdu/eTWtrK+PGjcNoNAat0+l0XHHFFXz++ed88sknjBs3zr8uIyODb3/72xGPW1xcHLasqqoKgO3bt2vWrzcYDOzevTvoGvr27at5LCGEEEKkLscBhz9w93F5XFQdqOqhFkUmwXs6sdvPBO7g/dnU5F1eVtazbYsgJycnbJnB4H3s2tra/Mvq6+sBGDRoUNTjNTQ0ADBgwADN9b7lvu18+vfvH/W4+fn5YcuOHTsGwHPPPRd1X5/6+npGjhwZ17ZCCCGESB3WwVaqv6oOCuCNOiOWwZYebJU2SZtJJw7HmcDdx+WCqtR7K0zUOeecA6BZLSaQ72Xg8OHDmut9y0NfGgLTdrRorfcdY8OGDVHHQwReQ6z2CyGEECL12MbbyOqVhVHn/VbfqDOS1SsL23hbD7csnATv6cRqhZBUEYxGb/57mrvssssA+Oc//xl1u5EjR9K7d28++OADXCEvMh6Ph8rKSvR6PZdcckmn22Rpv68ffvhhXNtfdtllHD9+PCiXPxK9Xh/0zYMQQggheoavyszQ3KEU5BVQNKCIkrEl1MyrSclqMxK8pxObDbKyzgTwRqP377bUeytM1LRp0xg0aBB/+ctf+OCDD8LW+3q0TSYTP/jBD9i/fz9lIalCy5Yt47PPPuOWW24JGzDaETfddBNDhgzh0Ucf5eOPPw5bX19fH7R8/vz5AJSWltLU1BS07cmTJ/2pQQD9+vWjrq6O1tbWTrdTCCGEEB3jrHdStLSI8m3l1ByuYVfdLvbV70vdMpFIznt6MZu9g1Ptdm+qjMXiDdxTdLBqInr37s2LL77IjTfeyIQJE5g2bRoFBQXU1dXx/vvvU1xc7J/EyG63s3HjRn7xi1/w9ttvU1RUxM6dO3nttdcYNGgQf/rTn5LWpjVr1jB58mSKi4u54YYbKCws5NSpU3zxxRds2LCBWbNm+avMTJ06lQULFvDUU08xYsQIbrrpJnJzc9m7dy8VFRWsW7eOiRMnAjBp0iS2bt3KlClTGD9+PEajkWnTpjF69OiktF0IIYQQsUWrMlN2Y2qOJ5TgPd2YzSk7OLWzrrzySrZu3cpvf/tb3nrrLf7+97/Tv39/Lr30Uu666y7/dgMHDsThcPDwww/z+uuv8+abb5KXl8c999zDokWL/NVskmHcuHFUV1fz+9//nn/84x+89dZbZGdnM3ToUEpLS/1lL33KysoYN24cS5YsYeXKlXg8HoYMGcJdd93FxRdf7N/uwQcf5Ouvv+aNN97g3Xff9W8nwbsQQoh0kexJjZJ1vHiP89eP/spTW54KW56qVWZ8lMABdyKy4uJiNVYuc21tLYWFhd3UIvFNI8+XEEKIVOFLN/H1WvsGeHY0TzxZx4v3OH/96K/c+/q9mscw6oyUjC3psZ53RVG2qaoase605LwLIYQQQoiEJHtSo2QdL97j/Hj9jyMeI1WrzPhI8C6EEEIIIRKS7EmNknW8eI/T2ha5YESqVpnxkeBdCCGEEEIkxDrY6q+J7tOZSY2Sdbx4j2PSmzT3N+lNKR24gwTvQgghhBAiQcme1ChZx4v3OE/f+LTm/pGWpxIJ3oUQQgghRELMuWZq5tVQMrYEyyBLpyc1Stbx4j3OPZfew7NTn/X3wJv0Jp6d+iz3XHpPh9rfnaTaTJyk2ozoafJ8CSGEEGc/qTYjhBBCCNENnPVOSteXYllmoXR9Kc56Z083ScSQjp+ZTNIkhBBCCNFJofXFq7+qZtWOVSlfueSbLF0/M+l5F0IIIYTopGTXPRddL10/M+l5F0IIIYTopGTXPRddwOkEux0cDrBacRRs1vzMnqt+DvBWrknFHnjpeRdCCCGE6KRk1z0XSeZ0QlERlJfDli1QXo71rV0YlfB+7JOuk5RvK6doaVFK5sBL8C6EEEII0UnJrnsuksxuh6YmcLX3tLtc2Da1kdWmD3vpgtROoZHgXQghhBCik5Jd91wkmcNxJnD3cbuZdiCLHFMOCkrYLi6PizU716Rc77vkvAshhBBCJIE510zZjWU9cm5nvRN7pR3HAQfWwdaUzdfuMVYrVFf7A3hnDhTNh8bex3Cfijzn0dGTRylaWpRSL2LS8y5EAjZs2ICiKCxatKinmyKEEEIAZ0oelm8rZ8vBLSmdr91jbDbIygKjN0XGPkGhqRe4leiTlaqoKZc+I8G7EEIIIUQai6fkYU9ORpQSEyGZzVBTAyUlYLHgGNMflz6+XVOtapCkzQghhBBCpLFYZSq7YzKiSGk7KTURktkMZd60Juv6Uqq3lYfdNy2pVjVIet5F0rz00ktMmjSJvn37kpmZyYgRI1iwYAGHDx8O2u7tt9/muuuu45xzziEjI4MxY8ZQVlaGx+MJ2i4wReWdd97hqquuIjs7mzFjxgAwceJEFEWhubmZX/7ylwwdOhS9Xs+6dev8x3jxxReZMGECOTk59OnTB6vVypo1azTbf/z4cX71q19RWFhIRkYG5557LpdffjlLliwBYNGiRUyaNAmAhx9+GEVR/H+EEEKInhKrTGVXT0YULW0nVSdC8lUHMuii92OnYtUg6XkXSfHjH/+YJUuWMHDgQG6//XZyc3PZs2cPL7zwArfddhsDBgwAYNWqVcyaNYvs7GymT59Obm4ur7/+Ovfffz9bt25lxYoVYcd+7733ePTRR7nhhhu47777cLvdQetvueUW/vWvf/Ef//Ef6PV6+vXrB8DPfvYznnzySb71rW9x5513YjAYWL9+PbfffjtOp5Nf/OIX/mMcOnSICRMm8O9//5tx48YxdepUmpub2b59O3/4wx+47777mDhxIl9++SUrVqzgqquuYuLEiV13Q4UQQog42cbbWLVjlT9IDg04u3oCqWgBeqRz9/RESL7qQPZKO3/9+K+ccp8K26aPsQ93j7k75Qb/SvAuOm3dunUsWbKE7373u/zzn/8kOzvbv66pqYm2tjYA6uvrmT9/PllZWWzdupULL7wQgEceeYRrr72W559/nunTpzNlypSg47/99tusXr2aH/7wh5rnP3bsGDU1NeTk5PiXVVRU8OSTT/KDH/yAlStX0qtXLwAee+wxrrnmGv77v/+bO+64g8GDBwMwf/58/v3vf/P444/zy1/+Muj4Bw4cAPAH6ytWrGDixIkyaFUIIURKCAxEqw5UYRlsCQo4rYOtVH9VHRREJzMVJNrLgda54cxESJ1NoelMlZ3A6kDlISk0Rp2Ru8fc3WPVg6KRtJk0kxKDPkIsXboUgLKysqDAHSArK4vc3FzAG+Q3NjYyb948f+AOYDKZ+O1vfwug2fNeXFwcMXAHbzpLYOAOsGTJEnQ6Hc8884w/cAfIzMzkwQcfxOVy8fLLLwPeXvfXXnuNUaNG8fOf/zzs+L4AXwghhEhVvkDUMddB2Y1lQQFsV08gFS1tJ/TcgTqbQpOsKjvpNsFWWva8K4pyMVAGjANOAM8CD6uq2hZjv2LgUWAsoAAfAb9WVdXRtS1OjpQa9BFg69at9O3bl+Li4qjb1dTUAHDVVVeFrRs/fjwGg8G/TaBYx9VaX1VVRW5uLmVl4W/MR48eBWD37t0AbNu2DVVVufbaa9Hp5H1WCCHE2SVWz3xnRUvbCTz3c9XPcdJ1MmjfzqTvREvXSaTHvKvvT7KlXfCuKEpf4C1gJ3AT8C3gj3i/RXgwyn7m9v0+Ama1L34A+KeiKKNVVd3ble1OhmQ9pMlWX1/PyJEjY27X0NAA4M9/D6TX6zn33HP92wTKz8+Pelyt9ceOHcPtdvPwww9H3O/kSe8/IPX19QAMGjQo6nmEEEKIZOrOiZWSOYGUVrujBb/R0lMAhuYO7VA7IqXrbN63OeFj9eQEW4lKu+AdmAdkAN9XVbUBeFNRlBxgkaIo9vZlWqYA2e37nQBQFOV9oA64EXim65veOV094KSjzjnnHA4ePBhzO19qS2j1GYC2tja+/vprhg8fHrYuVjUXrfU5OTnk5OTwxRdfxGzXOeecAxDXNQghhBDJkKrfpscSrd2xgl/beBsralbgOh0cy6ytXYtjvwPrEGtCbbEOtvLxoY9xq8GFLHbV7cJZ70zp+9gZ6ZgjMBn4R0iQ/j94A/rwfIwzjIAbaApY1tS+LC1q/cUqBdVTLrvsMo4fP87WrVujbldUVATApk2bwtZ98MEHuN1ufxnIzrJYLOzdu5dDhw7F3Hbs2LEoisLbb78dVq4ylF7vndHBNwhXCCGE6IhULaEYS2fabc41Myhb+1vuua/Pjbl/6Li/maNnoteFz7TkUT2a7UnFcYMdkY7BewGwK3CBqqr7gOb2dZH8v/Zt/qgoSr6iKPnAn4DjwN+6qK1JlaoDKubPnw9AaWkpTU1NQetOnjzpT0u5+eabyc7OZunSpezZs8e/zenTp/n1r38NwKxZs0iGBQsWoKoq9957L42NjWHrd+7cyZEjRwAYOHAgt9xyC9u3b+eJJ54I29ZXbQbwl6EMXCaEEEIkqiu/TdcKUpMVuHa23fsb9msu33N8j+ZyH63BqZNXTeZbfb8Vtq1We5I1uDUVpGPaTF+8g1RDHW9fp0lV1YOKokwC/g7c3774EHC9qqpHk97KLpCqAyqmTp3KggULeOqppxgxYgQ33XQTubm57N27l4qKCtatW8fEiRPJzc1lyZIlzJo1i7Fjx3L77beTk5PD66+/zq5du7jrrrvCykR21JQpU3jggQd4/PHHueiii/je977HoEGD+Oqrr9ixYwcfffQRH3zwgT9ffsmSJWzfvp0HHniAV155hSuuuIJTp07xySef8OWXX/pfNkaOHMl5553Hiy++iMlk8leiefDBiMMthBBCiDBdVb5RK63lhe0vANDsau50ik5n2z2873B2HNmhuTyaSD3+ep0eo84Ysz2pOm6wI9IxeAdQNZYpEZZ7VyrKecBaYBtwb/viHwNvKIpyeXvvfeg+PwJ+BDB0aMcGUyRbqg6oKCsrY9y4cSxZsoSVK1fi8XgYMmQId911FxdffLF/u5kzZ3Leeefxu9/9jhdffJHW1lZGjBjBn//8ZxYsWJDUNtntdiZMmMDTTz/NG2+8QVNTEwMGDKCgoIAlS5YwatQo/7YDBgygqqqK3//+97z88ss8+eSTZGdnM3LkSGy2M99sGAwG1q5dy3/913/xwgsv+Ae9SvAuhBAiEbEmVuoorSC1obUBBQUPHv+yjgaunW33sqnLGPfXcagBIZuCwrKpy6LuF6nHX6foyOqVFbM9qTpusCMUVY0Y76YkRVGOAE+rqvpwyPImvOUiH4+w3xPA94GLVFV1tS/rBXwGvKqq6v1a+/kUFxersXK6a2trKSwsjPtahEiEPF9CCHF2cex3MPf1uew5vofhfYezbOqyhAdthrIss7Dl4Jb4th1kwTE3uFp2PBVwfNt0NAugI9ddur5UcyKlkrEl2MbbYrYn2v6p1imqKMo2VVUj1slOx+B9E3BAVdUfBiwzA/uAaaqqvh5hv/WAqqrqlHiWh5LgXfQ0eb6EEOLsEZre4usxDkxl6UgpSa0gVWn/n6/nHbQD13ja1FM627ZUvrZQsYL3dBywWgFcryhK4FSetwOngI1R9tsLXNLe2w6Aoigm4BLgyy5opxBCCCGEplhVWzo6wFKruEWOKYdsU3bQskxjJo2tjUEDWFO5Ao5v3F/J2BIseUWUnCyg5g0z5gft4Ix8T3wDdW9dcyvXXHANw/sOp4+xDwV5BVTcWZFygXs80rHnvS/eCZo+AR4DhgNPAE+qqvpgwHafAxtVVb2n/e9jgQ+BfwJL8ObI/xi4FihWVTV8as8A0vMuepo8X0IIcfaIlN7iS2XpTJqHVloL4F9W2L+QdbvW+Qew+nqhh+YOpeZweDiklV7TY5xOKCqCpiZwucBohKwsqKkBc3h6T2BveyDpee9GqqoeB64B9MDrwMN4Sz4+FLKpoX0b337bgBvwTtT0AvA8kAlcFytwF0IIIYRIpmhztzjrnaz5dI32AMt/PofzJ3MoXTMnYtlHX3ELx1wHZTeWYc41Y841YxtvwzLYQsXnFTS0NoT1sHtUT4fmk+nW+ul2+5nAHbw/m5q8y0M3DfkmIZDL4+JEywkWvrOw69raRdKy2oyqqjuBq2NsM0xj2dvA213ULCGEEEKIiAJz2C/Ou5hMY2Zw77chk5lvfkVR5QXUG9vCppA0uqHwy5MUFayg6RNw6Ym77GO0XmhIrHJLtON2+UyxDseZwN3feBdUVYWNEdi8b7PmtfqoqKzcsZLFVy9Oud73aNKu510IIYQQIlm6q9c4NId99SerAZgxagaWQRZKRs6g5hlY+dn/o0nfhickcFc8kNUehzb18gbuEH9eerReaPD2sE8YOuFMXvkgCyVjS2IG4d2eJ2+1elNlghpvxPndwrAxArvqdmHQRe+nVj1tKZHTn4i07HkXQgghhOis7uw11gpym13NZPfK9uaTl5bCV804rlf9gXmg/s2w9S9w63TC1sdTr1yrzrlPYA97ovPJdHv9dJsNVq0Ky3m3j4em3cH316AY0Ov0uHFHPJxHIe1qvUvPuxBCCCG+kbqz1zhmkNueDmI9AMa24H2Nqo7ptQrmBrTXx5GXrpVjr0NHfmZ+XD3siR63sH8XFVgwm72DU0tKwGLx/qypwVG/M+z+ulU3BXkF9Db0jng4o6rr9Ky23U163oUQQgjxjdSdvcbWwVaqv6oOqx7jDxytVqiuxlbpYtXoM6kxRlUhy2PA9r4bFLBVqsHr45zhNNLMqFt/tDWhoN2XV77583fwHPkK96mTKNnB+fkePKzbtQ5nvTNp32CE1bx/JLjmvdb9VVA41HSI83PPZ/fXu8MPqkJmrz6dntW2u0nwLoQQQohvpJgBdRJFCp79gWN7Ooi5qYmaZ1zYJyhUDVGwHDZg2+TBfNwDOh3mkzpqjt6C/fo+VNXXRp3hNDTgrbizgpXbV3Z4ZlRfmlFjayNu1e2NInPAP/dTQADf7GrGXmlPyuyl8aQ3hd5f8A5IPXLyCMeaj/mbp9L+fxRQFECXfkkoErwLIYQQ4hspZkDdUU6nt3Shw+HtUbfZMJu9kwyF1l/3z6aaA/al03Bsr8B6AGw5kzHXA2+sBld7zrbHA0Yj5syBlE0Pnhm1dM2coH2ZN5+i1yb7r+3jQx+z7KNlFOQVMGHohLDAPdpsrr51az5dQ31L/ZmZWn3Bukb8m8xvMKKlN/leDnyTOPnaWddc52+nW3VjUAzkmHI41nLM326V5L5kdJe0m6Spp8gkTaKnyfMlhBDJpzWhUadSPWJMIhRULjLzfNixg49b97E7t5U2nQ636j4zgdAbZsybt4efw2Lxvhi0t79oySiaTtV702jaIOs0TNvTi9WXeHCp4YM1QycoCu3ZDlwPRC0xGUm8E0pFvY/tL0CW7+1li/FI+G2IMHlUpAmw+hj7cNJ1Mu7j9JRYkzRJz7tIW4sWLeLhhx/m3XffZeLEiT3dHCGEEGko0eoqMUWZRMj5iC0oEN6iboHeQAbebmDV21Ps71m+XKXsQ2NwXXOj0Ru8+05XaaeptfFM6Ug9NBmh4vzTuCL0z4b2XMcauBtX4K62p6Uo4Xn40Xr1NYW8AFnzFaovDa6yEy29KVI61PC+w9lVt6tb0qS6Uvol+gjRTSZOnIiiKLE3FEIIkbK6dfZPiDqJUFitdYUzqSch/+/G5XFRZdZ5e+19dc19vfi2M2k9jgMOXIoneN/2rllj8OLw47entUQbuButxCTtLwcGN+SehlkX3RZWHz60vn35tnKKlhZF/xxCXoBsm1WyTnsrw0DsQbq28TayemX5q+D4tl82dZnmchmwKoQQQgjRSQn31kY4RrfO/gn+qjFaveVRA+EQRp0RywUToMbmDWarqrw97jabt1yi73SDrVQf2BYUwBvdMPnfCq+N6kWTzqN5zsAe51gDd7WquJxrOofzmhR0zaeY0OtCbLOXYS60hp0nnnz1MCEvQOYGqHkG7N/Po+ryYTHTmwLz30PToaKNO0gXErwLIYQQIqUkK+juUODYWREmEcJmw7rDHhYIB2mvghLUI5xrhrLIbbWNt7Gq5oUzOe9u70ysi7fmsPi3/8D+xUo279vMrrpdtKltuD3usB7nWAN3tdZ9FOdn0aFynBovQOZTRsqypsPc+D63SOlQvuW+l8Nb19za4ZfDniJpM6LTNmzYgKIoLFq0iA8//JBJkyaRlZVFv379uPPOOzl69GjYPqqqsnTpUi699FIyMzPJzc3l6quv5h//+EeH2vDUU08xcuRIevfuzYUXXsjjjz+OxxP+fWFbWxtLliyhuLiYPn36kJ2dzaRJk3j77beDtlMUhY0bN/r/2/dn0aJFHWqfEEKI+CVr8qRun/0TIk4ihNkcls7hzXP3/qfBDaY2KOp3cUKTJplzzdTct4OSS2ZjceVTsj+fmlOzMVfuwFxopezGMqrnVfNZ6WfMGzsvLK3Ff4x5NZSMLQlbH21dPLQmcYqZZ26zxUwX6owOpfKkEOl5F0lTVVXFY489xnXXXce8efOorKxk9erV7Nmzh/fffz8of3zevHn85S9/Yfjw4cyfP59Tp07x0ksvMXnyZJ599ln+8z//M+7z2u123n//fe644w6ysrJ45ZVXsNls7N+/nz//+c/+7VRVZfr06bz88suMHj2ae+65h9bWVl599VW+973v8eKLLzJ9+nQAHnroIZYvX87evXt56KGH/MeQgbFCCNH1khV0J7OOe0JpPGbt3vLQtI3CzKGwYwe1rfuxmIZHTD2JxZxrpmz6cpgeY5v2nuhI1xJtfVgvtkY5zMB0ntd2vcbdr97N8ZbjAOgVPW1qm7/nfubomUZPzgAAACAASURBVJSuL2XTvk2oqkqbpw29To+iKFw59EpslRWYl6yMmC4USTyfk9bL4YmWE4x4agQ5vXKYfOFkFl+9OGV74qVUZJykVGRkGzZsYNKkSQCsXbuWW2+9FQCPx8N1113HO++8w/vvv8+4ceMAeOedd7jmmmsoLi5m48aNZGZmArBv3z4uvfRSmpub2bt3L/379496Xl+1mYyMDD7++GNGjhwJwMmTJ/nud7/Lp59+ykcffcSYMWMAWLp0KfPnz+enP/0pf/zjH9G1T8zw9ddfc9lll9HQ0IDT6SQjIwPwBuobN24kVX5HvqnPlxDim6d0fSnl28rDgu5ESw9GK4HYkcmJOnucVBDrWmKVjbRX2nF8sRnrW7uwbWrDfMwdVg7ztV2vcdNLN4Wd+6J+F3H9t65n5uiZTF41+cxkTyEMioFsU3aXfU6RSkkGyjXlsmP+jh75fGOVipS0GZE0EydO9AfuADqdjlmzZgEQ+OLzwgsvAPCb3/zGH7gDDB06lPvvv59Tp07xt7/9Le7zzpo1yx+4A/Tp04df//rXqKrKqlWr/Muffvpp8vLyePzxx/2BO8C5557LL37xC77++mveeuutBK5YCCFEV4hULSTRqiCdTfnwSVYaTyqIdS2R1i98d+GZVJO6GspHtVI0140zh6BymAB3v3q35rm/bv6ashvLWLl9JU2nmzQDd/BOqtSR+xvv56SVyhOqobUhZT9fSZsRSfOd73wnbNngwYMBOHHihH9ZTY337f3KK68M2/6qq64K2iYeV1xxRdiy8ePHA7B9u3dyi+bmZj799FOGDRvGI488Erb9Z599BsDu3buZOnVq3OcWQohvqmRUg4l2rA5VBYkws2lnB6f2SO58DB25/856J2s+XaN5LZv3bQYiX2vF5xXBgXF7PXn7eCirwF8OE+BEywm0+JZv3rc5ZtWdwDbFK1Lb/+eT/2HNp2tAgckXTub7Bd9n2UfLoh5LRe3RzzcaCd5F0uTk5IQtMxi8j1hbW5t/WUNDA1lZWfTp0yds+wEDBvi3iZdWek1+fn7QcY4fP46qqnzxxRc8/PDDEY918mT4zGtCCCGCJbMEY7RjJRR0h85sWl3trfrSnsrRGcnMnU+GuO5/yIuM876ZFL02mfqWes1j7qrbhbPeGfFaUQkPjA1QNdi30ZnJo3JMOZxoDQ/gc0w5OOud7KrbFdd1+tqUyOBYrWo+dafq/P+9omYFK2pWoFf0obsHUVBSdvImSZsR3S4nJ4empiaam5vD1h0+fNi/Tby0qtkcOXIk6Di+nxMnTkRV1Yh/AgenCiGE0JbMNJKkHSvKzKadlaw0nmSJec98LzLl5bBlC5SXY//VVTS1NuJBe+Ymj+rBXmmPeK2TL5wcXjXGDZYDhFWDudx8ueY5Ljdfjr3STpunTXN9qDZPW0LPQWjbdVHC3DY1ehtyTDkpO3mTBO+i2xUVFQGwadOmsHWbN3u/IvMNMo3He++9F7assrISgNGjRwOQnZ1NQUEB27dv13xp0KLXe9/KA781EEIIkdw0kqQdK8rMph0RODOrvdJOxZ0VlIwtoWhAEQV5BQzNHYq90h6zvGBXzPDq+CI87STontntOJVGSq91YbkXSq91sWlgK64IOeaB+0caJ7D46sXBQb1iJEtnwnayyF8O05njHWy8ce9GzXPUNdfhOODQzHVXQqeYxZv7nshzENr2vMy8uPfNy8gjv08++Zn5zB49u8cGq8ZDgnfR7e666y7AW46xpaXFv9xX2jEjI4Pbbrst7uM9//zz7N692//3kydP8uijj6IoCjNmzPAvX7BgAceOHeMnP/kJp0+fDjtOVVVVUGDfr18/AA4cOBD/xQkhxFkqMAhtcbdgUIIzbzuaRtKhOuCaB7KeqQvuP9CZVI5EaNUBn7xqMjNHz2Rf/T5q62qpOVzDU1ue4vwnz+cHa34QFpQ7653MeWUOF/z5ApZsWaJZTzxWYK+53unk4o21/vrwgQrzvBXJnDWbKJrrprwYtgyB8mLYnQd67U53763SuOdq+0kONh7EXmlnYNZAsntlk2HIoKB/ARVzN2LeWA1lZTiUg1xUdhFPbXmKk67wFFTf8SN93pfkX5Lwc6B1f3wlLx1zHUz/9nTNlwKttt1xyR0c/uVhDj9wmOW3LE/ZwB0k5130gKuvvpq5c+eybNkyRo0axU033eSv837s2DGWLVsWs0xkoIkTJ2K1WoPqvO/Zs4fS0tKgQbT33Xcfmzdv5tlnn+Xtt99m0qRJ9O/fn/3797Nt2zZ27drFoUOH/BVwJk2axNq1a7ntttu4/vrrMZlMXHnllZoDbYUQ4mwWmmNt0BloU9swKAbcaviMnYmINbtn/AeKPLNpoiKlpcx9fS6Npxtxe870HKuorK1dy5t73vT31vru14mWE/4AOPA4vvSUaHnrEfPaD04Dt8Zgz/bZWQHsl6s09fIOKgXvT2MbqBHiWB06/z0PPe/Wg1t5astTYfvU1tUyedVkfwnJq5ZfRWtbq+bx45mxddnUZUxeNTnu5yCevH/beBsvbH+B+tbwPP9kPLs9RXreRY8oLy9nyZIlZGVl8fTTT/PCCy8watQo1q9fzz333JPQsWw2G4888gjvvvsuZWVlqKqK3W7nySefDNpOURRefPFFli9fzpAhQ1i7di1PPvkklZWVjBgxghUrVpCXd+Yrtrlz5/Lzn/+cQ4cO8bvf/Y6FCxfyzjvvJOX6hRAinYQGs26PG4POQGH/wk6VYITklXOMNrNpNFq9t5FSefZ8VRsUuAdqbG0MK7eoanSP+9JTOlqy0d5Qwc5zVcI6lBWoPVoLgMOs8wfu/vPqwWTordn2vMw8/z0PPa/WNYD3GfC1115pjxi49zH2iWvGVusQa0LPQTxjJcy5ZnbM38Hs0bPJz8wnv08+s4tm8+E9HzKvWHu22XQgkzTFSSZpEj1Nni8hRE+JNKmNZZAFx1xHl503meUoIx1fa1KfaSOnsXrH6uCKK24o+Bp25BMeOLezDLKwdvpaiv9SzJHmI5rb+CabchxwRL2nEe+5Kx9L9VHKL1WDAnSjqlDScgll7/Sm9OoWyjNrg3LcjYqRgv4F7KrbFXXyq3gmMAptr4oacZ8Fly3odKnOIO1VdCz659jSNzw9J9Iz2dXPUjLJJE1CCCGE6JSk5aUnQCvvPDBfPBki9d6i4h2cqXrDJKMbslyw7FUwtaGZb65DR2FeIUVLi6hrrgvfgOD0lFj3NOL60ZOxbc8h67Q3FQa8P7NaVWzLanHu3kLjvz7B0+b253sbdUayTN7UlFhVc+KZwCi0vdbBVgy68Exsk96U3HSUgCo61n+d9F9/aHvCduuGZ6k7SfAuhBBCiKh6olRid8xqGik9prau1pvCsTcPy34o2QY1z4D1IGz8v9BLY+BntikbFGg63RSxHOP3C7/vT9GIdU8jrr9+MebKHdScmk2JMx+LK5+SppHU/MXAwd5uLrofVoxWadMBqope0TPjkhlxp6b4zhtxoGf7i0tge23jbWT3yg4K4E16ExvnbExu73ZAOVBbJcEvMFGeybNphlyQAatCCCGEiMGXp5zwjKed0B2zmkabfMmca6Ysa7q3VnpACUrrUSOf189g4USo+LzCP2vn4kmLuXXNrRFnDjXqjAzMGui/Z7HuadT1uWD+83L8ySgWC063m6vuhlY9/rQeVQGdoiPblB103GhpLL7zLnxnISurV+ABVB1BA2J1HphRNIPFkxb7j9stz0dAOVBzg/eFyj4eqkb2wXLd3RHPmYoz5HaGBO9CCCGEiClW0Jds3TGracxKNxEq2Jh/uZjlGgNhI83wCdrBYjyBdFz33GrFnr+VVkP4QNaOBKnmXDPLb1nO4g0wxbUiOM9fAT06sntlBwXK3fJ8WK3emXMDAviyt4zwrbshyrlTbYbczpK0GSGEEEKkHF/6RmA9eZ2iY+bomXHtH8/kSDEr3SRYwSZayomCQn1rfVIna/Jf630zWXOxRgWadqfcpzp0PvMvF9Nb1Ye/ECiexHutnU4oLfXex9JS798TZbN5y3/66vnHWQ401WbI7SypNhMnqTYjepo8X0KIbxrHfkdQ/XCDzkB2r+ywPO3QSiIzR8/UrBkeVhLQ6cT5h4XYGypwDAbr6MnenPJOpHs4653elJMdK2lTA0ZUBqSdGBQDep2egrwCJgyd4E/3iKsiSnu1FRwOsFpx3jeTotcmU99yAk+Eso4GxUC2Kfy+RWp/YBu+OrKHtV+uDwvgZxfNZvnNy+O8Ke0DTUNr8Ed4EYp6H3zXX1XlfRGw2WKWAw08ZnelfXVGrGozErzHSYJ30dPk+RJCfNOUri+lfFt51NKGWuUedYqONk8b7sBSiSH74XTiHD+Kohn1/gmNjG2QlZFLzX07Oh3YzVn1A57/11rUGDkOvheLijsrYr9waATBpVN0lH+nLagsZKTzBF2/Bq17qaBw2hM+K/ns0bNZfsvyWLfBq7Q0bOwARqP3m4yy4PZEKt+ZbrXYO0NKRXYzeRkSXUGeKyHEN1E8Aw21Kom0trUGBe5a+2G3s7C4nhOm4JlImwImW4on9SaSnbWbYgbuvnYdbznO+P87nhMtJ6JWRHH+YSGll5/AMttF6WRwZrhw9G+NGbhrXr8GrXupFbiDd4bVeDjrnZQ2rTnT5hxfg1ze3vM42pDOlWG6ggxYTSK9Xo/L5aJXr1493RRxlnG73RgM8usqhPhmiWegoVaAr8WoM1LYv5DS9aU4Dji4+NQeVo4iLMD25XOH9gBvO7iNZ7Y+w8xRM1l8tXZqjS9lpuLzChp1R1E84cePJCjFxtcWj4vnqp8D8KYCZa2kqX1ypurz4IUiGNhAUEoOeOvJq+3/C7z+WAM0472XCgpf1n9J6frSqOkn/nt4fj0uxdvmVaO9VWLMp4zetJc42pDOlWG6gvS8J1F2djYNDQ093QxxFmpsbKR3b+1prYUQ4mwVz0BDrUmFfANGAycpyjRmsm7XOv9EPc8P+Zo2jQGeigqWwZawHmAPHtrUNp7f/rzmBD/OeiejnhnFiu0rONJ8hFN6FVXhzIROKpqTO8Vy0nWS8m3lXLX8KhqMbUHfEtSbYHd/ggJ3o85ItimbHFNOwgM0te6lQTFg0puClquoHDl5hCVbljDqmVERv5Hw30PF429zkxHsE5SIA027Y0KwznyjkgokeE+ifv36cfz4cerq6jh9+rSkOohOU1WV5uZm6urq6N+/f083RwghulXMajCEB/iAv8dZQfFPUnTzyJtpdjX7g3FVQbM6i06nwzbeFrEXWkXVTOOwV9ppaA3pwFP8O4HifTHQocQ9g6mPLxUo7GVD4xoK8grYMX8HO+bviHrftGi9LGWbstk4ZyMlY0s4N+PcoO09eKhvrWfhuws1j6fZi26AqjH9Iw5WjfnC1smqNWfDbKvyPXwSmUwmhg4dyrFjx/jyyy9pawv/CkyIRJlMJgYMGCA970KIb6R4JxWyV9pZs3MNR08e9QfvHjwYFW8AGk9KiA6FmaPvwpxrTrhmu+OAIyhNxS8guFZ13oo5w/sOZ/fXuyO2Q0HRPlYcMgwZ/iA90brr0SaGsg6xsmbnGs39Kj6v0FweMe3p8ukRK8REnZwqdMBudbW3Dn+U8p2houXUd+c8Bp0hwXuSmUwmzjvvPM4777yebooQQohvgLjKC57lzLlmbONtrNm5Jizo9QXa1sFWPv7qY9ye4MGdOnTeIL+9h3fxpMWAN8d82UfLNM9nVHVYcoOrf1kHW9l6cGvMoNvlcbG/YX/Ubfpl9KOhtSGu/PNQnU0vifqyFOnSIiyPOQlWom2w288E7uD92dTkXV4WX+B9NuTUS9qMEEIIkabOhhSARETKVfbdh6Mnj4bt48uXnjl6Jm2e8G/Ev1/4/bDUEme9k8mrJgdv3x6gGt2Q1eLB9st1/pQNZ72TxtONmpMzabVneN/hEVNn9OjJy8zDo3rQtYdpRp0Rk96EHn3UY5v0pqDA2FnvZM66OQz4wwAGPD6AOa/M6dCz4bvvWoNqASZfOFlzeTxpT/Ge27LMQmnTGpwZLpw5UDoZLPdC6bUunNs3x3287sip72pS5z1O8dR5F0IIIbpTPHXQzxbR6n/bK+1h9wG8veq5vXP92yzdujSohKRBMTCveF7YvdK6r3igtxtyTsPkz2DxZgPmGfNwPmILapcv5SUvI4+Jwyby5p43/bn2gTXdr195PfWt9ZrXatAZcHvcKCj+WWXnF88PqgMfyqQ3sXHORqxDrP77NeqZUWHnyDXlsmN+/HXsQ+97IAWFHFNOQseLdA6tb4/CPnNVR2ard/BrszGgNr9ioubnn8XVhnSoIx+rzrukzQghhBBp6mxIAYiXVq5yfYt3sOTOozs1A9q8zDy2/mgr5lwzjgOOsNrvbtUddK98QeRz1c+FH08HLUZo6QWrR8NrBW5qNr6NvZKgdqmoGHVG7rjkDspuLIs4s+fNI2/m+e3Ph6XZKCj+1B4VFZ2iI7tXNtYh1qBc8MK8QlCg9mit5oyh9ko7ja2NYfekobUhofzu0PvuvRU68jLzmP7t6Z1O0woNpqu/qmbVjlX+aw36zBUPjb3aC/e054649NCkeOK+pqg59WlCgnchhBAiTcVTB/1sofWi4sHDyu0rmTlqpuZ9mP7t6f6gLNa9itbDDATVUveXPMzbjeOLXlFfoCLlb++s26mZHx8pZz/asbQ4Djjw4NE8fiIvd5Hu+7BzhnX4253AnvYWdwuNpxv9LyyBA0g1z62R8O1SE3thTeQ+piLJeRdCCCHSVDx10M8W1sFWfw54II/qAYWY9yHWvVr4zsKgGU7DhKSzuwxQdZ4Hq9PToRzqSPXpQ6+xoy9jke6XgpLQ8ZKdIx46TmPHkR1hg4gDBxmHnluHLmxswdn6whqJBO9CCCFEmkrGgMB0YRtvQ1HCB4SqqNQerY15H6LdK2e9k5U7Vmr2hPcx9mFU/igMIZ3YRjdY9oPtA12HXqC0XiZyTDlkm7KT8jJmG28j25QdtjzHlJPQ8ZL9gqiVhhPKF4xHqjvfkQmoziYyYDVOMmBVCCHEWcfp9JbZczjAavXOeBlnveyeMOeVOWF54jEH6MZxjaXrS1myZYlmmsns0bNZfPViip64iCa11TtI0g2Zbrh5F+w8z8DFOd+CUaOobd6XUA61Vj48kLR8bGe9k4XvLvTWYVe9VWEWX7044eP52/nFZiz/bsH2ah3mJj1MngyLFyf0zFiWWdhycEvE9aEDSJN2j9LoWY81YFWC9zhJ8C6EEOKsEjrhjdHonbI+gQlvulvClULivMZoAaWvOgsHD2L/1VVU9W+l8AisKwypeOJSqJn1AeZCa5IvWjvo7Pb6/k4njBoF9SEVcnJzYceOuJ8ZrUo+BsVAYf9CMgwZMYPxDl13mj3rsYJ3SZsRQgghvomiTXiTogJTX4oGFFGQV8DQ3KHYK+3a9cvjvEat3GqfZlcz9ko75kIrZf/nMxymBWTre/sDd2gfwGpQsa+Ym8zLPRN0lpfDli3en0VFOGsd3V/f326HxvDqNTQ0JPTMREqFeWPGGzjmOii7sSxq4N6h607DZz0aCd6FEEKILhJpUqHObpsUDseZYMbH5YKq1C0z6et13bxvM7vqdnlz3Q/XRA7i4rxGX0CpJaj0ptkMZWU4zHp/4O7fzgBVrXvivo64PusIQad9xdywspm+Ci1JPX8ghwM84WlFqGpCz0xnxmlolQuN67rT8FmPRkpFCiGEEF0gWv3q0EAlkW2TxmqF6urgoMZoBEtqVu2IVsoxMIjz5b47653Yr27BMQas+8FWCeYGgq4xMAVj2ohpfHjgQ3Z/vTvo2FqVTKy9h1PdtiMogDe6odA0hNL1pVFTOhL6rCMEnY7WPbgyOlbfv8PPmtUK27aFB/CKkvAz09FSjR2e1yDNnvVYpOddCCGE6AKJ9BJ2uEexM2w2b96vsT1dxJcHbEvNqh2xqpQEBnH+9IrMWrYMhvJiKJoPzn4G/zWGpmCs/mQ1XzV9Ra4pN2YlE9vsZWS5FIxt3r8b3ZDZButyDsZM6Ujos7Zaz3w+PkYjVtNwjEpw/6tRia9cYoefNZsNssOr15CT023PTIfLVqbZsx6LBO9CCCFEF0ikl7BHZko1m70D9kpKvD2QJSUJD+DrzlQfrXsUKDCI8weo7TOquvTQZAL73EL/NWoFsc2uZm4eeXNQSkfFnRXYK+1B12gutFIz6wNKTo/CcqIPJa5R3Pzt22hua4kZFCf0WUcIOm03PELWqbagl4esU25sF8wM2l3r8+nws2Y2ewemzp4N+fneP7NnJzRYtbM6XLYyCc96KpG0GSGEEKILJDL7aY/NlNqew90R3Z3qo3WPfEKDOM0AVQdV38rwB2yRgtjauloccx2xr7HQStnvt/v3tSyzxBUUJ/RZ+4JOu92bn22xgM2G2W6nZrUeu9VN1WCwHACbQ4+5aSWUWaO2/Zph12hW1inMKww/v1Z7li+PvV0X8eXLd6iUZiee9VQjPe9CCCFEkgT2dDaebiTTmBlXL2E6zpTa3ak+WvfIpDdRNKAobMKlFndL2P6hAXI8KRiJXGO8KR0Jf9a+oNPh8P40m8HhwHzMTVkFOJ6FsgowH3MHDcC0V9ppam0MbntrA2998Zb2ecLnv0pJvnz5WJVpzmYSvAshhBBJEJZDvWM1ADMumRGzqkYiFTi6vSpNBF2V6hPp+nz3aMaoGeT3yadv777c8e07eP2Hr2Mbb8NeaadoaREXlV1E7dHaoGMaFENYgBwWRCtGstw6bL/fBKWl4EwsxcQ23kamvjdK+/Q5igqZ+t5hQXnYZz1yBjUHp2G+7lb/eWOKkAsfOADTsfsdf9qQv+1qGydaT2geMvSeidQlaTNCCCFEEkTKoc42ZfvTMKKJpwJHj1SliaArUn3iub7Xdr/mX7/6k9Ws270O8NZjj5QTX9i/kDdmvBF0j4JSML7YjOWtXdg2tWE+th0+rIVVq7AunRb/NR48CE2NKEZQFVCBptZGFv795yz+jyfCzl12Y1n45EHV1bBqVex8bJvNu13opEMBAzCtH31F9QUEl7RU0exh75YULZE00vMuhBApJlV6VkViumPQaY9UpYmgK1J9Fr67kPqW+ojXp3X9Da0N1LfWRwzcATIMGVFfbtRDh8B1GtztPdUuF06lkcaPPsSjelDaI95o12hfMZdmA3h8kZUCbTp4/rO1YVVn/L/j5cWUXn4Cp6/sY7yTB8UxANP21imyTuMf1BopcAfQKTo27dsk/96kCel5F0KIFJJKPasiMd0x6LRHqtJE0KnBgxqc9U5Wbl+Jh+A64oHXp3X9KmrU40b6DIJ+14wuqi+FVZdAzTPe9UVz3TSZdtOmgg4dOkXHjEtmsPjqxZrX6GjZgysj/PyqQlAN+mjnNTcQ/+RBMQZgmvtfSM0zO7CPh6rBcMoAtf3BHTK5lA4dbZ42th/eTu3RWvn3Jg1Iz7sQQqSQVOpZFYkJ7YnWocOjemhsbUxab6bWoEiAU+5TXdJjGutboGQOHrRX2vGoGjN44k17gcjXH4lWT7nvmor/Uhzcy6+HJiPYx3v/NPXyVqgB8OBBp+jINmVHvEZr7+FnerlD+F5AnPVOpqyewvGW45rn9TY6SZMHLVuGuVHxD2p9YzVkn8ZfH9434Fen0+H2ldSUf2/SggTvQgiRQlKpZ1UkJnBApV7Ro6LSprax+pPVmpP1dITvBcEQMkFPbV1t0s7hEzoAN9KkQ8niOOCI2Iu+btc6nPXOsBekaEKr0EDwNR1pPhLey2/w9lI7hoTkihP799A3cZNO4/3DqDNSmFdI0dIidhzZEbbed96kTh5ktcIHH8CoUdCnD+bzR1Fzw6uUFM/zD4ouyCvA7QkZ1Cr/3qQ8Cd6FECKFdHgGQZESzLlmsntlo1N0/kA0mb2ZvhcEX0+0j9vjpr6lnoXvLuz0OXw6+i1QR8dsROtVb3Y1Y6+0Y841U3FnBQV5BfQx9qFf737oleAoW0Fh9ujZVM+rDvs2INYsrUZVh8Wdz8X9RqLTSBD3fcOhdY2+iZvuah2J3oO/6gx4792ru1+lobUh6nk7PHmQ0wmlpTivLKL0/xuN5ekx3nYVDILt27159Nu3Y544LeibkglDJyT13xsZr9M9FFWNnismvIqLi9WtW7f2dDOEEGe50Jx339f+koOaPizLLJqT4FgGWeKqOtOZc+gVPV/85IukPCsduY7OPL++fY+3HNduzyALa6evDTq+QWegzdOGXtHjVt0xzxfpmuBMik3FnRVcv/J66lvrw7YxeKAPvcBk8s+mGnhO8L4gbN63mcbWRvac2BP1mn369u7b8d/x9oo1TqXRm6ffy/utgVExkGXKjnrcZP57I/92JY+iKNtUVS2OtD4te94VRblYUZS3FUVpVhTloKIov1GUkFfvyPt+X1GULYqinFIU5WtFUf5XUZQ+Xd1mIYSIRyL1vkVq6o5vT6yDreg0/l+4R/XE38Pf3luLxaJZX7wj19GZMRu+Z39U/qiwdb7zhh7f7XFj0Bko7F8Y1+9LpN59g85AQV4BFXdWsHL7SppdzZr7u3XQwGkaQyc/Ot3EwncX+lNyag7XxB24j8of1bnfcbsdmpqwW88E7gAu1R3z3ifz3xsZr9N90q7nXVGUvsCnwE7gMeBbwB+BP6mq+mCMfe8FngLswDtAX+Bq4EFVVcNfsQNIz7sQQoh4dEcPpLPeyQV/voA2NXyEZNGAIqrnVcc4QEh9cV+udUDKRkeuo6O99fZKO44DDqyDrcwcPZPJqyYHnTfTmMnNBTezdudaTrpOJnT80HMFXlMg3/WZc81sP7w95rFC5ffJ5/ip4xFTckIZdUZ0io6ReSO5cuiVwZV6nE5vUO5wwMUXe5ft3OnNY7fZgtNqLBbYsgXLvbBlSPh5kvmNTzSxPvvQEXYkTwAAIABJREFUz7kzlYnOdrF63tOxVOQ8IAP4vqqqDcCbiqLkAIsURbG3LwujKEoe8CegVFXVZQGrXunyFgshhEhpyQwsOlVCMTBo0wrUAtp6jukcvm75OuwQu+p2eXOwo52vvbcWl0Z98fbygx25Dq1ymTp0fHniS0rXlzJz9ExWbl8ZMVD3lUb19YBXHaiiMK+QdbvXsXrHas3AOJFvNQKvac2na6hrrvMPWvX1FKuqilFnjByEe7xpC56ALz6MOiOoxBW46xU9fXv3paG1QbtEYwPBL1ZbAgJirUmcrFaorsZ6wEX1ecEDbZMxaVa8vxfRSqVKCdzkSsee903AQVVV7whYNhTYC0xTVfX1CPvdh7en/lxVVU8nel7peRdCiLNTyuTqdqA3XItBMTCveF702Vrbe2s1lzs63ksbrX0GxUCb2oZep8ftcft7nts8bf5SheAN+ErGlvjbX7q+lPJt5RED945+Vpanx7ClriZsedGAIr488aVmzjsq9GqDDDc0t5eS9LVh2ohprP5E+wUjlA5dWKUb/3VXAOXlZ16sQhmN3oGtvhrvEXPejWSZOv4cJ/p7EW17e6U97DMM/ZzFGWdjznsBsCtwgaqq+4Dm9nWRWIHdwD2KouxXFMWlKIpDUZTLu66pQgghUl3K5OpG6w2P0FYtbtUdu9Sf1eoNAgN1sL54YIURe6WdijsrKBlbQn6ffP/MpL52qaj+0oQuj4vWttagwN23PLD9WuVTAfoY+3Q8R9vpxPrWrrC67EbFyIShE7i54Gbt/RQoPAo7lkDJF+cG5Ykvvnpx2Iyz2YY+jHSdg8EDgVUwQwP3oOt2OCIH7hA+iVP7bKvmGfOo+aCIktOjsOQVUVLcufEyif5eRMuflxK4yZWOaTN9gRMay4+3r4tkIDASeBCwAV+3//xfRVEuUlX1cLIbKoQQIvWlTGChFbSFBGqRAtlAcaVK2Gze9IvQXv4E6os7650sfHehd1ZU1YOKGpQO4Tjg4MjJI3EfL1L7I6Vj3D3m7o732trt2Da1sWokZ3qq3ZCl02Ebb+PWNbdq76fChH1gPmWkLPeHMDf4/EFpRrmF2H65Dk42UfyfcCQrzuu24k2PidbzHvqS1T7bqhlIVj92R34vfJN2heqO2Ye/SdKx5x3QnMVBibDcRwdkAfeoqrpKVdX/BW4G2oAFWjsoivIjRVG2Koqy9ejRo51tsxBCiBSUMrX14+gNj1QtxVd5RmtGUU3tvbWUlHiPn0h9cacT50/mUGS/gBeqV9CmtmnWtI9nNlTfLJ+BvdWh7Q+dmCnWNcZTa9xZswm71Y35BBQchaJDULINajYUYM41Yx1sDZsICxV0Ksz8VB/xRSdoxtnN2XDyJEVz3RzN1L5+3zcTgdfkvG8mpVN0WOZC6WRw5gTesCRO4hRDPL8X8dZ11/oMM42ZNJ5ujP45Sd14TemY834EeFpV1YdDljcBD6uq+niE/V4CpgMZqqq2BCx/C6hXVTXCa7aX5LwLIcTZKZ1z3gMrsdQerU1scGwn2lh6+QnKL1XDZiH10azJrpHz7qur7hucGqn9voGTsQbOxvNZOuudFD1xEU1qq7fHvQ2yTkPNMgPmGfOgrMx/nIbWhqCKPnoP5GCi5o6NmAut0e+VxUJp3hbKi8Nna9WhI9uUzc0jb6a27sznBnjb39qIS3Vj9ECW2ouaA9Mwf7LP+6KlMYi5K8S6lx3Jifd9hr5ByM2u5uifUyr8XvaAs7HazC5CctsVRTEDfQjJhQ9Ri7dnPnTKNAU0ks+EEEJ8I3SqOkxSG9LeG263e1NlNAK1Hm9re16+47zIgbuvd1arrb5qM6Fttw6JHghHSseA4IooLe4WGk83BuXV+74J8O1vr7TTpG/D1d536dJDkxHsV+opa+/R9rV9yuop7Diyw3+uNh3U46L479OY/sX06PfeasXRukXzPuVl5rH1R1vD9i1dX+oNVtvHAbh00KRTsd86kLK//i3qPUom3z0dmjsUj+pBp+iYMHRC0PVGy4nX+qwCP8PS9aX+wD3Svoke/5skHYP3CuABRVGyVVVtbF92O3AK2Bhlv78DDwGTgPUAiqLkAmOBP3Rdc4UQQiSqu2tCRwsOu1V77nKQkPKRZput59ranpdvPUBYWULwpoEEprRo3dfQQD3SZx3PMxBP9Z2gPG2nE8f7a3AZQwbJGqDq2oKwF6Xeht5hx/Pg4UjzEcq3lUcvd2izYb3/L1QPPB1cvrENpp8/WXOfTfs29fj4i0g93qH3vzNjReLZN2XGoqSgdMx5Xwq0Ai8rinKtoig/AhYBTwTWeFcU5XNFUf7q+7uqqluBV4G/KooyW1GUKcBrgAt4ujsvQAghRGS+4KF8WzlbDm6hvGoJRfYLcP5kTtgsoCkpxsylCR+rqMhbOnDLFu/PoqKeuw/tefm2Sm+qia9ai0711i+fNXpWQmkNYZ/1tnKKlhbh2O/QXB6a87zwnYWcaDkRdRCvDh3/PvZvRv+fAsY8NoyWY0cwhFaZ0RmxXDAh/HKj5O1rVV8JytHeYWdm/2uD7pPR7b1vtkrfDmeeFedP5rD7aHgCgUExRB1/key88HirzHRmrEg8+6bMWJQUlHY57wCKolyMd6bUcXgrzzwLLFLVM4lpiqJ8CWxQVXVOwLIs4HG8ue+ZQCXwM1VVz3wnFoHkvAshRPfQquttdEPJxwpl758T/8DKnhBH3npCSkvDa36H1vlOtH0xJoGKuX/79TkzXNgnKFSZdVgmzcR2/eKEvx3R/Kx1RgryCthVtytqXfBos8xqak+cNbR501/0HnDrwajqyMrI1XzpiKdnP3AG0bAe6xYPFcvbWFkEVYPBcsAbuJsLLLB2bdCzUjpFYemlKu6QbzNMehOflX7W4Rz/RMU7S25nzh332ATJedeUjj3vqKq6U1XVq1VVzVBV9TxVVRcGBu7t2wwLDNzblzWpqjpfVdVz2/e9Np7AXQghRPfR/LrcAFXnqWF1z5Pay50McdRqT0iE8pHO7ZsT721NRi9+QJUac4GFsgt+jONXX1A2fXmHAqpIqRF7ju+JmTJhr7Sj1QGpoDAqf1RYnXnff7r1YPB467Vb9kPJ3ryIAWFg7fKw4xHcE6zZY21UWfkdhbIKcDwLZRXeMpNYLGHPiuO88MAdoCCvIOK97Yo5CuLt8Y5W1z2WePbtzPHPdumY8y6EEOIsplkT2u3ttQyqex7ay601dXx3i6NWe0Ks1rCa385+Boom7qJp287EppqP9mKRSC++Vl5+B0Wq/z2873DNnvfAANJxwKE52ZFO0fHGjDe4dc2tEevMu/TeWVIdK4xQMh2i3Ddf3r5tvC1iLrivPWEvHIqHKrMejLrwevq33hr0uWqNIzDqvJNGRdIVeeG28TZW7VgV8ToDdWasSDz7psxYlBSTlj3vQgghzl7+mtBqe+1yN2S52vOEA+ueJ7uXOwmc4y6mdIqC5d6AGt0dnLkU8AZ5WVln6r8bjdiv1HurpSTa25rsF4skiFTDfdnUZTFru2v1EOvQMXP0TH+t9kj56kY3WA4pCdVMj9UTrHU+g2LgVOGFWGx9KZ2Tj3P+jDMvlyF1/f3jCNT4a/YnOy/cN0jYnGumIK+AogFF0uOdgtIy570nSM67EJ3X3RVERPpy1jux/2MhVe+uxOL0YNv8/7P37vFxlde993fPxbJkyQLH2CbOEGygtgzyONSZCTiGAyUvR9AaBwhNHQHiJI6ARqUnTaeXFzekzntOOqHkkyjEdtwGTGTSGto4SUEhTUOw6jhjc9GFWCaAcdgYjDEGXSxZnsvz/rHnui8ze89FGtnP9/PhIzzal+e2R+tZe63fElq4QXb8eCCghX/oCQQ0Q3UK2uz/TjOj40O5+uGPNuLbPVD824BUnHpSPjKwtId9x/oMh+ljkg2UO36+TFhpuBfSdrerQz4yMUJMZNRlvEKhPuqib7QV35c2lu0tjb49Vrr2aUPYJD9CXVBH+P617B2yp9lfzrjwMznGvNooFPMujXebSONdIikN+YdBUhQ6wzUnwbLKjFHT5Evhon3ZrXTe8nBl76NL5gSTzfKiVnyrWsqXTFsF2DHww7vD9LzeQ+LUSVxH32X1YReh2S1lNdzN2jMeG2fw2GBacx5M5inf+i7inqXo/ttdV5LKI433MiGNd4mkNOQfBnPk24gSKJOyS3oOXushqCYI/UrB57/CsSFlV6WjVIpR6lCEJufYeuICNr73EXjlZcKXJYj4XAQXra7MunOgbJP9HCw7ZxkI2H9sf/meCYdrpRzPZaXWQ6W+MyZr/UoKczpWWJVIJNMQWXDDiN7Asp14KNGwUZG0EOk5SJaj750B2y+Hvq378TlMfrVKviy3LrWdKqt6FRKhQFyB7896lZ0zXoULGhiLnyR6LErv8f3lX3cOkon1z0G2AVm2Z8JBsm65nsvgwiAvHHnB4HkvZT1U8jtjstavpHRkwqpEIpkUZMENI5WQeTvjSCmfRCLaT4chB+k5SJWjd8OoF8LBmOPkV6vky3wJhymcFtpJqXBE1kfovK7TYLiZbZYBEi4Y8cLwqZHKrjsHycT65yCbsrXNQbJuuZ7L1uWtxBO5GvSxRIzW5a3O2l6BtplRyvqVTC7SeJdIJJOC/MNg5HR+G1Huqo+VaoulpvxCHCuxFKNLrQ6ptO1sY9E3F/HgvgfzVhN10s+TsZN4FPOX6wmX5onPpph1l3eOk8ayOltT3Ql8Djqu0fTp9VhtNHLa9prxPEca/zplF8BSBcjquex5vYe2H7Yx/+vzmX//fNp2tqX7bDYWXf1duHRmlkCw6dlN1u0sQM/rPaZt2/GbHQS2Bmjb2UbbD9usn7s8Y+Z0/VbTM36mIWPebSJj3iWS0ilXYtXpwumaB1BNycmGtggX9VElrTTSMRA2r+b6HHT+vLLJr6m2DZ0cMuiVW60Dq3hng9KJy0M8EdeKGOkMdSBdbbTQ/Qq13XKOOzpQH92Mf32M0Rlk1HeUGvq+mFst1Ow5yBmLGLS/WEPnt17OvFlxmu/g4Hiz9ngUDy7FxanEqZxjG2saear1KVq2txjGYkH9AgaPDRqa4sLFoT8/VJQazEWdFzERnzC9ptkaKqRuU2zCcjU946cjp2WFVYlEMj0p9Kr/TKPcbyOqxRNWTeFAhrYoCUZdccKvPgJ+P6FFrdocJL3UaU35iMeRBngpbTMrNGTmCU8ZTFue22Lw0Ov7GUvEcLvcKIqiGep6dIa703VXcI5DIU2Pfkam6FDUDaPuhGEd6J8DBOk2p+djVzw35Mapxn9WZVgCAe2nhdFq9ly6XW7TzcXIxAjrf7LedCzeHn3btCkJjGNgh/DusCEMJ/uaegxzUsa6CNX0jJ+JyIRViUQimSLsJB7apZqSX6spHMgyLOZcAaOj+L7TRd9Xk3PwWo+mKb/HhW/d6qJk+0ptWwqzfJB8BpPZtdKJkmae9ySzvLO4Y8UdjtddwTn2+Yhcs5SoTo8+KozrQP8cNEVehWPvMniOVlU3tBt8w7HcEKZiCk7ZrAxr9lz2vN5D39tGbf0ECQ6+d9B0LMZj45b3KOZZiByO5OjV2yFnTspYpKuanvEzEWm8SySSSUVKI+ZSrvLf+Qy7yQ7BqSbVCtO2xDSjMGW4TFYJdv3aX3bOMkPbABQUU094PoPJrJ+F8CietOHu9Jm0M8fBRavpPb7f1jrImYP+DnjSRL8/Oz49GNQUbPIdk4dC30P6NdHxZAf9b/cjdK8xXLhYfPZiDhw7YOjnhXMu5MWjLxrOUVAyY+BATrOYOc4Zb92YqbMhvFohsuIQwSc7HH0XV9MzfiYiY95tImPeJZLSkXGSlaOaNJqraZ7TbRkfIqok0mEYfZvQKrZOUkEnszGp89YBMBYdI5qI4sKFoii0Nrey8eqNhrHKlyMRWhUyXN+luIgn4pbe2hp3Dc+0PWMar20n0dap1rztdWAnNruE+O1i2qUOqTRvamZoYijn83wx792f6ebarmtNzxm4awDfMI61580qxlqRL+ZdrY3iv4tMPoLDZ7SanvHTkUIx7+777rtvEpszffnud7973+c///mpboZEMq259xf3sueNPWnjIyESxBNxJmITXHfRdVPcuulN/9v9vHDkBRIiE/vqdXm5YckNkz62jTMbWde8jomYllh3w5Ib6Lqxa0r+qKfbMvwuvPgiN7wEXf+WNNzr66GrCxobK94Os7WfEAk+texTXO67HIC1S9fy2Kce4/YVt9M409im5nnNfO+F7xFPxEmIRNpgSo2tfsy/ce03eGz/Y5yMnTRt0yXzLmHo5FBRz6SdOS56HTQ2wrp1MJFMzLzhBm2esg1aO8dYUMz3UOPMRj7T/BmOjx3nrZG3mDVjFjcvu5nHPvUYy+YtM+3nsnnLtHPGj/PW6FvM8s7i5qabeeyWx7QxuPde2LMn8/YgkYB4XOvTdcZ2pMbzZwd/xtETR3N+p6Awb9Y8bl52Myvmr6DGXWMc76wxu/eCQ+yZO07UJWyPgVlbquEZPx35yle+8tZ99933XavfS8+7TaTnXSIpnWryDp9uSE+YDcpQir5YyrX2nSo2qYMRrt9+HQOe46bqMpHDken7TNoIOTELj7lpx00F+zwp4X2BAOwztoNAQOuT1WllWEvFXEOGPE4essKqRCKpGmScZOUoZ/LraYvNhMVKUK61bzc+Xx1SCT+1gcjTXVz6dpzXl8CYNxkioSTVZRa1Et71DL01EM3SnitHFdBKGHk5121cRuhLO/EdGbOs4GqVxL3m99bkrXw6acnfRcbtl2MtOb1GucdEbgRKQ3rebSI97xJJ6UjvsKRUpusf/clc+4Y4/zjURWHtIAzOg8CcZkK3b8W3qgVVGcnVYle81NcU365K9dNUr/9kQstdGE4e5M3NYbDKEbhhyQ382+C/5SSSKijs+ewegh8KTl79hSLj9ssxxk6vUc4xkX8HCiN13iUSSdVQTAXKcjAZ+udlu4eTqpFlpFo04vO1w0znvHlTc/6KkpPUxlSl1Pn3z2f+1+fT9sO2nLZM5tpPKw8pWv5D1A1jHmg4BZGt0PlTF74/Xg/vvYfveIy+TdD+rKbA0/7+YvreXIPvEzcVtf42/GID7598P1f1aHyI8FMbgOLXmalevxfCq7Tfq7O1Cq4Bz0Pp61qp8+x6fRdunfnjxs2mfZvoeLKDh3ofyiuD6LQPlsc70J7Pphxryek1yikNKTXiS0d63m0iPe8SyfRkMrw8ZbtHGSsgTkn7K9wOq0qcCgoCMSnttlKNSYgEI6dGco5Nq4qUcZ3ZeetgGc/8BkS2ecHlyiR66nG7td8Xsf7UIZVF31xEXBgLCQWOuHn8z3bj/7FzZZtCfXp8B6bKKWuWrOHRgUcN3uKzZ8zm6Ml3jV1X3LgUl6kUYz5Vn3x9qJZnq1TK6XmXuU+FkZ53iURyRjMZXp6y3aOMFRAd3bZKPGGF2mFV1CgV/jAZ7TZr4/DEsMFwBxieGC5bW/JVV9UTXBjMVCtN4o1B4C1FM8zj5lU6cbk0xZNoNOPJvvk9Oh683paHPLw7jJlDUElAQE0Q3mZeidTOGFn26bDmfc+p5Jq8LgLTCsYt78/FqxsCJQEiEbc03FO6+06flWp5tkqlnNWgTedS5j45QhrvEonktGYyKgGW7R5lrIDo6LZVUi2xUDvM/ujrqXS7zdqoL8KT/Xm52uLECDQYWsJFfcJN6ILbYMkSiFlohCsKCIE6W/Nkb1kJ+xbClpoBy41CNpHDERIkDJ+7BIR6BJEJ80qkdsbI0KeEptcf2g2RhRnDPfu6g8cGTUNDNu6uof4UaQPeG9OMoYRJJdpZ3lk5ISVOn5VqebZKpZxhX+XcCJypSONdIpGc1kyGl6ds9wgGtVCFnAvZrxpZLMW2v+g4eYu4/kLt0P/Rd5n8CUsfb5U7UEROQaqf/s1+Xj3+quH3ikX0aU4lTd21bI1ZVlsjv9ph2whMGVrrLlnHvLp5nF0/lzUfbYW/3whXXGFcYwDNzdDaCl6v0ZPtwpa32Gz+XAlo7dd09YM1iwuvs6w+q/e00bFDy2cI7w7T/ZnujPH43gX0bdJOOekB/f5JQeHQ0CHCu8OEVoWIrI/QeV0nvkYfPv8V9G31aHH+b0D7c9A6oOAVLsM1ZrlnMrLrP9nwFx8hcO98To4P41FyhfryPSu2n60Sc13srqtCOSX5rpFSOsoey2KYqtyn0wkZ824TGfMukUxPZMx7ZdpfieqZ6mxsVe1MyWE2ndPEzgM70xVK08ev6ca3qsV4j+5uaDH5PM/4Wla1FIACnhjMikHMBSdmGM//9Wd/TfBDQedjphunjusVtlwqcjzM+WKOLe9lNTZ9fdqJfj+Bm99j30JjXwrFJKfvOTFCVMQy1Wy3evCJBtTd3flj3u1WAD3wJnzsY+k3BCMzIObOzEk2pmNssgbVBXXa/WJjxtCZlJmkgCcOcTe4FQ8xEStPzHuJz73ddZXvOCj87EkmD1lhtUzICqsSyfRkMioBlu0eJVSNLIVi2l90tdw8VSUbP/lpW1U7r7voOtb//nrWLl3LZ5o/Yzz+HzaZ32PPHnjzTdsVLbP7aShHrwACLn4H/vt78HYD9M8nx3j0KB5muGekx8PRmOnGqflt+N6lEPe4SCiZ5NyuG7tMq7Fa3qt2Btd9uct8jSXXX/+LP+cF99GcMJKC1XpVlcaNYdb1vM/Egrkwdy43DC+ga8+5+K65Cbq6aLzAvBJpen6z+nzv1bDHl/H+54zVPZ1w9Gj6mFhqQ2MS9mI6xibPWeP3trPu43czEZvg0PuHGI+OZ8KhlMy1Ey7wCoWmeRdzfuP5BZ8VW8+Ww0qreuyuq3zH7VZ3y+rXVUShCquySJNEIjntsVvYpiruMUWFhJy2v+hY3gJx/U7bYXq81T0OHnScU2CVJAuAArUxTWd8/wdA6IzHmIjljEfP6z32x0zXB98w9G2C8I1z2Xv5+QWLcOWdn+vyrDGfj9DfPMF2Ey+sZUxylufYF43S+essz/HXctuXd36z+mwVx7738F5tHi2OMcN0jE2eMx9oYSGHIxwdO2p9PUVQ66m1rYxScE2XmOti91nMd5xAnBax+WcKMuZdIpFIJI4pOs6/yLh+dTBCx18vJ/Dn9XT89XLUwTyGk9U9Fi92fO98SbIptROA4FvGmGl91c4Dxw4Yr2E1ZiZ98I176ay/xVbMcXBhEK8+Nluxl4dRKCbZEBt9/wZTlST1/g3OciKy+hw8jEERxuvyEmhsgpoay2NcuFB0LnhH+SeqSvDVk3iNebeZewg49P6h8tUVKDHXxe6zmO84qQAzvZAx7zaRMe8SiUSSoRIx75Zx54MR/I9cxqhXi/n2xqE+qtB32x58TUH79yhjzLs3DvWnNG+4b9wYM22mUb/52c2G8Jsadw0vd7xcMObdcRy0fsxiUB/LM2Y2MZ33kwn6vh3PVDpFK5rk/4Kb0Zku++ujUMy7p04b7zdHIR5Px7ynj8FNXU09gDEPwk7sdvL++qqz2THv2XH1ZYsLlzHvEh0y5r1MyJh3iUQyXVCHVO79xb3c98v76H+7n+Z5zaZx0aVQdJx/vrh+VdXif++7D/r7NQWUxkbu/cfr2DPj7Uz8swviwMTzEa675i7791i2zHFOQbqf8Qkm4hPMmzWPc+vP5cYP/0+6Dq7Ad7LGEDNtNh73/fI+3hh5w3D9S+Zdwl99/K+cjZMZurFr/MVu1v3gRSZcmgv5hpeg68cefCdn2IqjtsI0btqlMOGF617OOu4aRYtZJ545rlAMdVafGydg3YJrmFi5AmbUaOPZtxjfL59PS102TsC6AZjwAArc4L2E7f/7v7n7o9bzkL9zWux544lYznWvOX4WK0Zn8VY9jLkyrv6yxYWXmOti91nMd9xk5AZJ7FMo5l163m0iPe8SiWQ6YPCuKR7q4276nl6Cz38FhEIVT4Atijzex8A/NrHv7BOGUwLvzyLyjdEpaKxzLCtULllHZ0+DFvccDDqeH3VIJfzUBiJPdxFUE4R6BL5xbybpUU8goN2rSCyrYx5xE/nnTGXWwGcT7FtgUmm1lCqagQDsM9475/cl9M3y+snrysqgkslCVliVSCSSMwhDMR8RYzQxQbi+H7Zs0QxkhxrSk0Ke6rLBmYuN8c8xCNQsLu5edjS1dceog5GSdLT1GvUexYMLhZ5d36fjtQdRX9rneH7SVVf3f599C+JsuVTgvwvU2qhmvLt0f+Kd1AzQj1EkAh0dBH/1O/PY/qtaob1dO769neBVreWPoTaLDU9f3Gbf8s19gdhzJ3Hh2eugbWcbbT9sM6wJ/VqJvGFvjUkk0vNuE+l5l0gk0wFL7+AbEPknNGOkvX1KFG3yksfrqT78rfLFb9uJL9Ydo87x4P98nNFat6ZfXkRMsa/Rl9ao73m9hwPHDhCPnSKmCEP8fPb8pM6JHI4QXBjMUZgx9ebHtKJDnd2A260Z8E7jqPVj5PFoXny3G7UuZq2/7mAsiiLVrpGR3CqxdvtWaO4L/L7Y2PJsUud0f6ablu0ZzXuPy0M8EcetuG3px0tOb6TnXSKRSM4gTL2DWaooTiToJpU8Xk9fU5C+2/bQfqqZwPuzaI82F594mcfDb3VMOBjTNg7JZNNoImpacdTw1kN3XEoycPV5q0mIBLFkadaoG0a9EF5FzvykPevPbWHfm/vY8twW/Jv9aY+sqfSfB/YuTI5da6433HaxL/0YxWIgBMRiacnK9ucVAtF5ltUxK1JF0+fT+nDnnZqR3dys/bTbt0Jzn7q+xZjZ7ZN+HWSTWhPrf7I+55hYIoZApBOardaYRAJS510ikUgqRj6vaaUIrQqxfWB7xjuYrHIZ2p08wEnoxGQSCsH27UavZ0jTFvc1Ben8Wn/p97Gjqa07pue8PJrj2ZcuRW872+hOzk++zUDndZ0EFwbpPdJr8LwH3lK0sdu4sbgCzUDsAAAgAElEQVT8BrMxysI3DJ1PCHjnfPiq9RucitRXKKUOgp25L3B9O33KWxsAbR4Pvncw7zGp46TOusQM6XmXSCSSClDIa1opcryDc/20v1ijlacfxmAQVxUFvJ5lIxhEneOhowUCn4OOFlDneHI3NFlvAdTZcGAuGbnAJHZ1tBUUDg3laoJbvh1JGd3J+Sm0GdDH0XuFi/qEm9AFt5U2dvliy9MNrtJNYD5K1FO3fZs8tQFAWzuLz16c95jUcVJnXWKGjHm3iYx5l0gkTrBUF/n99opXe81BVbWwgL17NSOlWtVmJglbmvFZsc8d10TZvBJiOs+7mT57oVjnOm8da5es5YW3X+ClYy8RF3FiiZhmdEcV+kZb8X0p4y23WkPrLllHQ00DPa/3cDJ2kmMnjuF2u2m5sIWNV20s/e2OWex3LKbF0MdijnXIK4mjt1smuQzhK9xErllKcNHqsr0ZKybm3evyEkvEZMy7BCgc8y6Nd5tI410ikThByspVJx1PdrDl2c3p+HXQKo+2r9RtqpKbHr/3n+lvHDdcxz/fT++dvYbPU8bkjt/s4NjYMRLklupUUBAIPIoHt8vN0rlLWX2eueFoliBZ560D4ET0BLFEpg8exUNDTUP5jD39pq+1VdMer6JNYFFJscl+qf09+P/HAUbd8bxJyKW0Lbw7zN7De2k6pwkEDB4bJLAwkJ7r7GMCCwO0Lm+lq78r/e/JCLOTVCfSeC8T0niXSCROqBrPe7WRMgqL1DUvFSebKnVI5aLOi5iIT+R87lE83LnyzrzzaHWfbOysB72BN3JqhEcHHjWNl/YqXtrHltL5i5lTMraTTSnPmHw+JdWMVJuRSCSScmNDJ9wQj5z07IVWhezpjDtpjoW2+JSQr2+psIUtWzRZSBu65qX0zexcJ1rd4d1h4sJYaMjtcmvzmIdCcc9QOCExOyQk5Ynd/85+y0THqIjykHuAjrn7UB/dXL2a/mXCMifgtZ6iz32o96Gpf4YkkgJIz7tNpOddIjl9KSVuNl/8r95rGloV0hJHbZ5vt+1l19MulkJj09GhGezZih95dOdL6ZvVuWaxxlbXtPKeW4XMGO7/nWZGx4eIusGVgIQCKFldz+PptWr/mt9bw6N93yeqJAznILTr5+jGf7gZnnjC8dqqtFJSOa5vGgIVg/YXa+j81st5+2zmeU9fQ8abS6aYoj3viqIkFEWJF/FfzOqaEonEGVXlUT1NcawKY0cnPElKVi6yPkLndZ2aIeDgfDsU0hafVAr1zY5UX/blSuib1bld/V229cetvPSrz1td8P6+Rh99R9ZqeuhvwK190DhBulJsKn595NSI6fNt1X4ULcHWk3ohILJ+JjcGObrxAwOOPfBlV0pSVdR72ui4Yz6Be+fTtv1TNG9qLvn6oVUh6uPuzJimZFF3xQs+T/o3Y9lIjXVJtZMvbGaXyX8von09JIDXgEjyZzz5+YtA4fdVEomkIFMlNXim4dhAdGiAlv18/eVsaotPCoX65lCqr5S+5TvXdFNlQt7QpyR5N9gvvAAJgQAaTsFTj0D7sxB4fxbrLlkHwKMDj7LvzX18Z993WPTNRbT9sA11SLVs/+A7g/SNtnLn8wpNb8OcMVAS5Hj0IUs3HhxvDsu6IVRV1FXN+Gu3scV3lH3eo3z/t48zNDFU8vV9jT76nl6ijekbWmXZvk3gOx4r+DxlS6rO8s4y/F5qrEuqGUvjXQjxP4QQV6X+A+4A5gDfA84XQlwohLhcCHEhsAh4CDgbaJuEdkskpz1V5VE9jXFsIJaqFV1mrWknMdwVp1DfQiEtjCZ1TAHd+VL6Vo5xKVRRM98GWx1S8V/1EltWwr4PwZaV0HIrhCIeIrE7aKhpYCw6ll57CRLERZxH+h/Bv9nPsnOWWbbf96WNhPpnc2Q2jMwE4cKoQ19CVd2ybgjDYcLLhxmdkSl0lTCxPIq9vs9/BZ0/9xL5J+jsJlPPwMbzlNrE3bHijup5hiQSGzhJWH0AOCiE+JwQ4nD2L4QQh4UQnwUOJY+TSCQlUlUe1dMYx0aeQwO07OfrL2fDOzxpFOqbw0JMpfStXOOSz0ufb4Md3h3WZAiTBms6lOUKN4RCllU4BUILjxFYt9/nI3z/WkZnujKVX7Nj6UusqlvWDWEkQuRcYahQq8fp9dNvPJb20HG9Syu0BUU9T1X1DEkkNnBivF8F/HeBY3qAq4tvjkQiSVFVHtUqxE4+gJ1jHP/hLrUSaJkridrxDk9a3oSdvqXKz0ci2s88/db3bV3zOtb83hpu2nGTeV+ylG5894bpW9NtK7a9EFZj2PN6j+UGO3I4kpNICclQlvNcqPdv4OT4sOX9ookog8cG885rZGi/adLqLE+teVXd1lbbCkdOn4m8aywYJPiWko5LTyOS4T5oEpf1M+ppXd5K2w/bmP/1+cy/fz5tO9ssn+v0G49jfWz5SBz/F9yoV/iLep6y15l/zjKWTjTgO3KS8P+9HnVQ1mSQVB+21WYURRkGfiqEuCXPMY8DnxBCNJapfVWDVJuRTDZVpSJSZdgZGyfjZ6oKcxqMcdnW0BRrs4ONvjhQATJeXEW9fwPh4W4iCyG4vIXQtRvThXSsVGuufPhKgwZ8SkEGMOqIx2DdAPx4KYzMMFZt1V8jn954x442try4Lcej7Y1D+yW303nZRmOBpWuvhZERSCTA5YKGBi2Z1ecznV91NraeCTvzoq5qxr9uKNPnBMxIwJoD8PrZEJjTTOsXtnJt17UMTQzlXL+xppGBuwZy7m2mFOPCxa3+W3l47cOWY1YIW9V3JZJJoJw6708DNyqK8nlFUXJSYxSNduCTwC+Ka6pEIsmmkEf1TMZOPoCTnAG7CYxFU2Zdd7uUJW+iCG32SlCwL8Wq+CSNy+a6bTx4npZQ+eBvttH84MXpTZ3Zfdf/ZD3xhFED3qW4CK0KGb3XyVAWgNEChrudkI3Qbk0OMkdp5VQyVEb/dmPTJhga0gx30H4ODcGGDZbz6xvG1jNRcF58Pny7B+ieuBm3QIvNd2mymf91ATz+r9D5dC1d/V0MTxjfRoxMjBjWq1nIUYIEXf1dJb1ZCm9bnzbcIRnq5BGEt60v+ppnJFP0fXcm4XFw7BeBjwGbgL9WFOXXwDvAOcnPP5z895fK3UiJ5EwlZVRKcrGTD1A1OQN6j3BvL2zfXlKojF3KMgb5jGITbfZKUbAvxar4hMNsWDnEUA3puHHhgqFTI2z4xQb2HzMWRYomohx87yAxYVRGXjp3adrQ7buzT/Ne/+whmg6dAODxizGN/543ax7nN55v+82Pb89++l7S5CD3LtSSU0O7wbd00Hhwd7f5Rbq7NQ98CfNra435fHR9YgGJva50rEwslQOwWqFzUYDI4QhCn3WLZpTr12twYZDn3nyOBLlhQwmRILw7XPR3ZuTkQaK1uZ9FPbB39GBR1zsjmcLvuzMJ2553IcSrwO8D/wKcC3wa6Ej+PBf4AbAyeZxEIpFUDDv5AFWTM1BmXXcnOBkDy7jlMktbFkvBvhSr4hOJ0L0Yg9QiCnS/0m1538VnLy6oAZ/afD8ububHS+HR5XBiBkZlGJeXW5bd4uzNTzCIb9xLZzcZpZVxh6pFx4/Djh0lza/dNRY5HDHE6Ec9sNfnglCI4MIgimEStHAY/bVCq0LoAgAALdm3lM15cOZiQ2y+NwaBmsVFX/OMYwq/784knITNIIR4QwjxGaAR8AOrkz8bhRCtQog3KtBGiURSQaZjISg7CXWTrSBhNo7qkErH6A4Ct0fpaAF1dvLgfMZRCa+c9W1oXd5qawwMkod7v4M/vAj1njZYtqys0pZO2p+9FvXzqaCQEAlGTo1ox5kp3bhcsGtX/nEMBo2Ge4pEgtBTI9SfTOAV2p9Lr8tLvaeOrb9dYvzcYn2FV5EjlYhC2oAvqB2/o02bB/16MOtvXZ0W164/tqXFvH+xGBw7Zvzc47E9v63LW3EpGVPCo3hMx8HUyI9DIPBJ8PkIrQoxu2Y2ehpqGgitCmljsqONwL3zCf/5Sj45/mGDsV/q5jx0+1bqo0puKFJMIXT71pzjstvSccd8bX6KCA2ptu/esrSnSjb7pzu2E1bPdGTCquR0ZDonxdpJMp2sRFSzcazz1gEwdnKEqJLILVk/7tVUMfRhCSUkXeZLrOzq78o7BmYJgN4YtL+g0PnfSYNqbMxWm4ote283CXnDLzbQNdCFEIIEidzjhtE8fD09cOAAxOOagZqvzapK219cyLZlpwxG/O2/mcHDOwVqbZTwaoW9PheBwCcJffk/8R0Zy/38qtZ0kquewNYA+97cZ/h81im4w397znmGcdCvm+x+pBJN9+6FpibYudN8ngCam2F4GOz8zVcU2LNH29jYmLORUyPEEloIUY27hmfaniH4oaDx2O80Mzo+pCWDJnMA+h5txLdbS5xNzW/3K92gQMuFLWy8aiNA7rlxqEsu1bFaD1ERK9t3lzoYIbxtPXsnDhKoWUzo9q05yaqGfqTmJ6sftu5TZd+9ZWtPR4eWN5FtwHstvu8klpQzYVUikZxmTOdCUHaSTCueiJrEbBxHJkYYnhhOhwpEs2J8LXWoS3jlbDWXXf1dBcfANG7ZA3vPFZoxuHat9sfX74elS+G887Q26byNpVQFtrMWfY0+GmoacCmudLxzznGpRM3Vq7WkzFis8Dj6fGz8yi4axQxcSbvWhUJjYgYb/ysB0aiWvPmEIPLPLjq/8RK+I2PGz3saLNdXcGEw7aFP4Y3BHX2K4TzDOKTWzSqTfmQnpjY0ZAx3fZ99Pk1Z5k//FGYZq4ka8Higq6vgYam2pgx30OLOu/qN5/oaffQdWUv780puNdQjY+n++Bp9PPzJh3n7L9/m7S+9zcNrH8bX6NPuMzGSk0g65oG1Lym0jzWVNaHfV/9BOk9cSWT3MjpPXImv/oPGPuvaMuqF8PLhqatiWwbK1p4y17GQmOMkYRVFURYBfwv8AVqc+wyTw4QQwtF1JRLJ1FA1SZ3THCv1Cz1RD+xdcQ5setbcQ1fCK+dS5jK4MEjvkV6D5z1wOHn/wUF4/HEt8Sy1udi/35CIls8AKJREaLf9to5zOI6+piADX3wl9y3N13bhO95vvMbBg47nKLQqxPbdmxh1k+N1DvUIeMdG/zxaUmreexXqc8rQB6NnVE+F1pxvz3469+k9/4XvZRUvP/gBQeTpWvhambTYbSRbWsbunyumroptGShbe1K1HrKlSqdAWvZ0x7bnXVGUJcBzQBswDNQArwMvAVG0F479aIWaJBLJNKBqkjqnOWbj6MJlHpN7+S3Wf8iKTbq0aIPduUzHk6fit7MrdKbub+OtgJUBsGP/joJxtHbbb+u4IsbR16jFXQcWason4ctFpmpn9jUWL865tjobOq5XCHziUKZ/urwF3zD0jbYavc4mCaam/UttpPL1w26f9Z5Rl4kZUKk1V+T6tnpzEXhLKW/uhY01Xq62VNt3b1nb46AQm6Q4nBRp+hc0HferhRC7FUVJAPcJIf5eUZR5wDeBjwKrhBBvV6zFU4SMeZecjlRj3GUx8dJTTd6Y9+iY9djqi+O0tmrJhWWMeTedS6uiPE9tYO/TXQTUBKEekRtjfdNNmg64nkBAuw7msfOgJZcKRMFCWXbab+u4ArkDZusMyL2u4qF+PE7fd934jmfFzXd3p+dIrY3ivyuTjJpKZk2Hg0SjqHM8hK9wE7niAoK7XiW0K557Pd3cGvqXig03i3nXz6ndfAm7sfLlXHNO26i/T4F4+exji/4OCQQKrnEnbclHNX73VlN7znQKxbw7Md6PAP8thLg5+e8E8BUhxFeS//YAzwPPCyHaSm14tSGNd8npSrVUF53ufzzMxhHyVKm0MmS6u7V44yJeOduay0IGVLZhl31/G4lo+jl04TKED+WrHmp3LeY7Lm28vdZDUE0Q2uPCt3x1uh9W62zNkjU8OvBobuiQ4qV9bCmdT9fmjkVyjDpGd7Dlw8dywii8wkX7s4LOJwTqbGi+W6ummkg6a91CofWNOWyc9Yf4vrTROulXv5EaBmpq4JlnrBNJreYuD6l7Rfq7CR6G0OwWy3ZZnu/k+6OINma3c29/t6Zpb9JOQwKtgJqEwjMjNxH84gOF72Mz2dJOW+xgNXZT5cSolr8FkvIa7yeBB4QQf5v89wTwLSHEX2Yd8w3gT4QQC0prdvUhjXeJpLKYKp7YKBM/bZkqVYZi72vTa5ptABx6/xBHx44aLhX4YIDI+jLFKWc30cYG0GqdnV17NkdPOGurlYpM4A1Ne73tBti2AoOKjYLCWTPPyr8xnYT1Md03zHo6nuxg83ObcxJoEVATg5cfseEZL0HtqVycbnMiKY5yqs28DczJ+vcRoEl3TD1Q5+CaEolEAlRfAlfFmSo95GLvm0pEa2/XPKbt7dpbgnA4R1c8W+HnlotvmdS4XjuKGVbrDIHjtprGCQuXFv8MdF+EqYa8QBRW8ijD+iik211tiielEjkcyTXcARSYcNtUgzFb45NcGfR0mxNJZXBivPcBF2f9exfwCUVRPqUoSp2iKNegVVvdX84GSiSSM4NqS+CqOCUkp07ZfbMT0UIhLfZ7yxYtTnjLFs1rmSUfOdmFsuxsAK3WWcuFLY7batq/mgZC/bNR53gYNtNjs2iXgRLXhx3ZztNtwxxcaBFO5HKgBjPFyZan25xIKoMT430ncKmiKOcl//1VYAL4F2AEeAqoBf6urC2USCRnBJNt6E05U6WHXK772lDm8DX66Luzj/bfby9Zi9tO9Uc7G0Crdbbx6o2O22rav7sH4Kmn8H/BzYTX8tTCG9MS58mOB/d02zCHVoWocdekq9em8FRCmaZCnG5zIqkMJVVYTcpH3gMsAlRgqxDCJFV7+iNj3iVVj4mCyHST6CpnwtS0UK4pJXmvlL4Ved8cbChzlIuCccDJ/qh9u/Bf9RKj7njeqpu211mRz5SV6k4KffVby3lM9au/h/BlCSI+F8FFq7XjUtVkLdpmGY+fFcN/OsZXR96IcOVDVzAR1yrmemLQUIQaTJpJ/l49HedE4pyyJaxWE4qiLAM6gcuA94F/QlO+ids83wXsAy4F/kgI8R+FzpHGu6SqMUu0qqvTKmPu3z9tjflimew/gJO5UaiaP+5ZCZXqbK0CaORDEJzTTOhvnjA1Qos1gPImMzeHctZ+Sppx7zVLCaQM3WLGpYTkRSvDeaZnJovPWow7Hmfs6Fu85h4moaAZmS4PDTMa7EljZslRqrVRwqsVzbC/qpXQtRvxNfpsJ4CXvGG2kh0t8nkox7NULjUYwxpQFE0Xv7UVNhZxPSftl6ovZzQVM94VRZkDzBJCFK57XUYURTkb+A1abP0/ABcA/wh8Qwhxr81rfB74e2A+0niXnA6YKVOA9ocmkZgS1YSpxNJwWbKOzp6GsnrRJtuYNuubCxe3+m/l4bUPl/VeeQ2ppGGjKiP418cyWueKh/qahlyvuIlxHblmacaLXGCc8nqR+wOVUWVJPlNqbVTbmCyE4FsKrRfdRNcnFuSOic4L3rF6hC0vPWpYf+suWcePD+xkZHyImBstvCMrodXMuDZdy0k5ylCPyNWZFy7qaxvpu7MPoPLr0mSDoy6o09oUG8vo5cfd9D29BJ//irzPXNVsTFPk+15tbDxjvk8lk0851WZQFGWOoijfVhTlKPAO8FrW7wKKojypKMrvF99cW9yJFlt/oxDiP4UQm4GvAF9UFGV2oZOTxv//B/y/lW2mRDKJmClTgGa4g2k88umMZdLX0115EyyLYbLVIcz6liBBV3+XZfXSYiiY8JhU5givb2K0RjMeAaIiltv/rNh4dTb418fY0jzBvmN9pkmUZuSNAzZZ+2ptlI6hHxSs6pqXSCRdhGnLStj3Idh8qeCyxsfZ/OzmzJh8pxl1VXPOugp9aSf1njpDXD0KjE6MaIY7GJRozBITTdeykmDvuYLwqozhnvo8NfZF5RvoKsMWfDZM8h7Cy4cZnRjJPA8ixmhignB9f8FnruqUVvJ9r55B36eS6sO28a4oyny0UJO7gUPAILlfPf1oYSy3lrF9ZrQATwkhhrM++xc0g/5KG+dvBHYD/1WBtkmqEDuJbtMeM2UKPZMhQ1glWEr4qYm8CZbFUEgdotzrL7gwiMvkqzshEmU1cmwZUj4fkQtmEtU1J8cIzTKADMZmIsrQySE2PL0hb1vyJjMHg+DxpI9VZ6MZ3Oe/a6myYotgkPBqJae9MbfmLI+JWGZMJkY0GcKsdeU7MkbfkbUGw3n/O/tzCjqZoU9MzCdHGVmYaVuK7LHPlu3svK6zsOHu9zvb3JoYt5FzhaGPUQ/sXUjBZ67qlFbyfa+eQd+nkurDief974HzgU8KIQLAY9m/FEKcBJ4Bri5b68xZChzQ3ft1YCz5O0sURVkO3AF8qWKtk1QVduTSyntDh56rcqFXplBMxKUnQ4awSjA19qIKoR5dmGAZ/gDn8wpXYv2FVoVQTOZXIMpq5Ng1pAqqY2QZQGbGpp23Bnm9yKEQuDMXNdsgFOW9DYWI+FyG9hq85UkveO6HUXy/HjQYzsGFQbzC4s+ugBp3jUFdKZ8cZfAtBa8u06toZRIb6kEGTIzb4FuKoY/eGAQOk7muxTNXdUorqe/VM/z7VFJ9ODHe/xDYKYT4UZ5jDgEfKqlFhTkbLUlVz3vJ3+WjE3hQCPFK2VslqUom9TVsMZ6rcqEvLnLbbVpM5mTLEFYJpsbeaCu+8fLrqufzCldi/fkafbQ2t6LorEhbRo6DzaVdQ6qgxGfWxjJ4GFwmjmc7bw0svcg+HyxZkj6ukDfagNWY+HwEr2q1NraTZBdlynxovq5Cq0LU1zTgSRncIvOzxj2DZ9qeMXjHreQofbsHCF1wG/Vxd7qNJcmrFlMUykTOMtQ/m/qahsx6iEF9FEK7SR9j9cxVnVxs6nv1ttu0DaIruRbOsO9TSfXhxHj/AFDI6E0AM4tvjm3MsmwVi8+1XyrKp4ElaPr0tlAU5fOKojyrKMqz77zzjvNWSqacSX0NW4znqpxkFxd5+GEYGJjSSoFTjcHY+9JG+7rZDozcfF5hp+vPbojNxqs3ctbMs5wZOQ43l3YNqYKx1Vkby9AJv+23Bo7Cja64Ij2vwcPY90YXGJPQtRupr23MGQMFBY+ihel4FS91CTcjXoH/Tlh+J6y4Ezqud6He3Wq4na/RR9/dA9x5ye34ox+geWQm/uhcvrDsdl7+s1cIfsi8yJDpxsXng7/fyJqPtnJ2/Vzm1c1j3SXrik/uLKYolElFUt/uAfruHtDWw1w/7S/W0LfVoyX1FjB6fcPQ9+Ya1v1uNvPiMzl7xmzWLFnjvC/lxOfTvk9few3uvvuM/T6VVBe21WYURTkEPC+EuDH57y8DfyeEcGcd81/AB4UQTRVoa+oeR9G851/RfT6KJhf5dZNzvMBB4AHgoeTH56FVjf008KQQYiTffaXazPTErlxaWZhE3WtJkdjRNy9BIlCPk/XnVGnDsZycmXJGAVWWSkjWtf2wjUf6H0Fk+Vr0SkDqZcvwL9iZUSwppDqSNWepJNO0Aku+c22MiX4MWpe30tXfRc+rv+Dky4O8crYgAQgXafUYj+KhocYo+1hOyq7MUsZ1b7iunZoCdtWLJJIzgLJJRSqKsgn4LPBxIcRevfGuKMr/A/wU+LoQ4q9Kb7plO3YBh4UQf5L1mQ94HVgjhPiJyTlnoYXVWPGqEOLCfPeVxvv0ZFKlx4owjiRVSBnn0cn6q/hG0+7mssJFaQrplhON0nG9wpZLRU74S8GxyDIS1Y81EV4Fe4cG8286itxwq0Mq/gcu4n0msIqqqZiTIElF1ks5incVS/K567gmypaVOJt7ieQ0o5Dx7rH6hQkbgU8Cv1QUZTPw4eQNWtFUZj4HvAFUOkagG/hLRVEasrzlfwyMoyXMmjEKXKX7bAHwA+BvgV9UoqGSqSf1Sn9SCl6EQrB9u9FzJeMipxfFxP5a4GT9VTzEKxiE3l7jpiQ7LELvfe3t1dZ0GUMETMfkqRF8b22HmKbiEjlXmMet/2oH3Je7qcjRom8JEvrq41qYSYF2qEMq4atPElkBwTe0mGzfMFpcc1OTtca9qhJ+8HpGZ0wg3NbXjyai7Ni/o2LfNxVZL6nQu6kg+dw5zlmQSM5AbBvvQog3FUW5EtgG/HnWr7ahxZs/C6wTQrxb3iYa2Az8GfDviqL8A7AYuA94IFs+UlGUV4BnhBCfFULEgF9mX0RRlPOT/zsghJAxDacxqXjRyt8oGf85VZ4rSXmwY+Q6wO76Cy4M0nuk1+BJLZvShp3NZb68jTIadYYx+Rt/2nAHLW6991yd9zUOgd53YN/R9KZC3d2N/8ctaS9+75Fetg9sL/hmLe39rxshWgu9C2D7cjTv/3AC9ef/jv87mbCd9HXXdONb1ULk5veILizcz3dOvIN/s78ib/oqvl4mm+RzFzwcNc79dO6XRFIBHBVpEkK8JIT4GLASTe/9XuAe4GNCiMBkqLgIId4D/gBwAz9BK9D0DeDLukM9yWMkkskjO2m0s1Ma7tMREwWNyXiDUnGlDZPkQoNHvYxvHRyhC98M7Yb6U+BNKtN4hYv6U2SkPpObivC29UWp+aRVgFJ67W4Y9WoykwDhFaOMTAwbr7ttPYyOEnzDmBSr9UP/T1ExdauqU2YpleRzF4p4tLlPjq9Xmeb9kkgqgJMiTc8rivJ1ACHE80KIzUKI/yOE+LYQYlLfZwkh9gshrhZC1AohzhVCbBBCxHXHnC+EaMtzjUNCCEUI8R8Vb7BEItGYKh18J/h8qLu76fjiUvz/u5bloQZW3HcuHQPhstUHMFNSSYWTrGtex7xZ8zh75tms+T1rpQ2nxZ/UIZWOgTCB5RE6vhxA/arJWyETxRF1joeOq8aLLjKVbueDK+j46+WoV/iNc+9yoc6GjhZY0VU/9/IAACAASURBVA7Xr4P5w7B4tIa5tXM5e8LFmgO6C0ejRCYOFhU60vN6j/E8D+y4GAKfgx8sE8SUXEs8mojyA9d+1NpoZnOR/KujJMAlYIZrhuFe0USUHb/ZYTpupRTwKqqCqhlT9Uzq7wvQ14dv3Z307fHTfqpZU6tZWWS/JJLTGCcJq8NoKi9/U9kmVScyYVVypmMZA2z7AhVSsygzqZCKkYmRdCVNAI/LQ8OM0lUv8iWxArYSXItRprF1vG6O1Dke/J+PM1rrJipi5uflSXBN33diRDs/rhm9fVs9+ERDeu7Ve9rw125jZIZWxTSl2pL+Celzu78PXX6IfAiG55/Fbz1DRuWaPMmN6pDKRZ0XMRGfyP2FAEVkqcaAoSATAs4+qYXXgOap3/shCMxppvULW7ny4SuN1wVcuGic2ZgzbpOaTG/FVD2T1fpdUOFkbYnELoUSVp2EzTwPVEwCUiKRVC9lqRQ61Tr4NkmFVGQb7gCxRKwsIRD5CjfZLerktPiT7eN1oTXh9U1pw930vAI66ZbhKcFYztynqqLGUoGOiu5n6twZcOUdsGUl7FsIL3nezzHcAeq8dXlDLMK7w8QTupiX5CXSyjEKRsM9+fnoDAivVvANQ+fPvUQeP5vOP32Crv4u4sIslkarIqsf70ktIGfFVD2T1fhdMJVF9iQShzgx3r8CtCiKckulGiORSKqTshgaNuOpSwklKAdmKh4pyqF6kU8lxK6CiFOlEUfHZ+VtRC6YmTa8Tc8rYISZ3tcDexeSM/eRof0GhREzom6Y8BjVSFIoKKxdshZfow91MELHXy8ncE8dHbd+APVjTdDRQeS1HsPGzNJYt2jD3hXnQCCAetc6OjavIfDTm9ixfwexRMz6PN1455uTSXsGpirHYarum49q3FBIJBY4kYpcDewCfqAoyl+gqcscxVjVVAghNpapfRKJpAooiyydDRUXfSiBXfWQcmKm4pFubhlULwqphBRUEFFVgq+epLcGolnuF/1x2WFOJ2Mn8bg8OcalR/EwHtNi2a3CoAoqmhQwwkzPj0HgMDlzn2/MDeQxsgWCwWODqIMR/I9cxugMoanJNI6zfeFx+ra+TPAKD70f8eRsShQUFCFI2DDgvS4vgctvQf3LkLZWX9LWqlLA+tfPj9XYNp3TNHnPQJmVlcpy36kKXanGDYVEYoETz/t9aCovCvBR4C40hZf7TP6TSCSnEcGFwbSqRQrHhqwNFZdqCCVIqXh4lFzfRrnUPPKphBRUEEm+2g9tHaR+wlqRQx/mNPjOIPFEHI9L65NH8RAXcQaPDeYNgyrYnmXLzDvZ1JR7fnIsvTGoj0Io4smZ+/SYu6z9SV6Xlxp3jWFe9McEFgY0FRpvRis+E64TJ7QrTn3cnemT4mX2SUFD9njGodHbwM1Lb8atuHEl/1Rm91+/VlPhOy6TP6tma8dqbBFM3jMwRcpKlvdtbZ260BWTZO1J2chIJEXgxPOuL3IkkUjOEEKrQmwf2G5IrnNkyNrQwa94oSI7zcwqItTzeg8JkcCluFh93uqyFNwpVLgpb1Gn5Kt9XzRG36bshMmlhO55In2c3rCMiRgexUPT3CZqPbWMx8YZPDaY9sRnG4jZiZ6lFjnLOf+1HgJqgtAeF751q3Pm3mrMVyxYAQIGj2mVUluXt9KyvSWnbymy1+RNP3uIaG1uW1LhOr7jMfp+eTHhv1qt9enVcUJbByEW08ZzIQTeUghdcCO+v304/QZD33+r8Kq5dXM5/6zzaZrbBAoMvmNe5dVqbG/acdPkPQNTVZvC6r6TVGfAFFlkTzKNcFKkyap6qURSeU4XFYDp1o9ke32RCH2XrbEsOW9biaZABcdqKTwzWYW99MmWoFX57OwGIgKCQDPQiDYXO3akDZv0cQCzDsKJcHo9mRmWMRGj1lNLZH2EwNaAIT5bbyDq5/TxWx4HSH92Tt057P7gTxn5OzhrHB7aCWteTp48OJjpj82xtHtctsFrZSAHZy6mNz6QW+gnK1zHt3x15l6BABzXxiI9nggIDOZtl+laVbzc8u58Oh8TEGyAUAh1tjZmN+24yfBsmF17Up4BG99DOfPfuEyrQLtnf/m+t8y+C6YydEUW2ZNMI2xLRZ7pSKnIKaRaZcWcMt36YbO95ZS8qwr5vAqTt4/DmI95dze0tMD77xsKGqXJmp+OgTBbnttiMABTEoodT3bk/b1ZG+u8dQCMRccy52VLOgI/ehTWHPJqajWV9pTmIR3zngydSYXr9G114xOzc9dwR4cWnqGPvy7QB8MYKV7qx2P0fdeN73gMvF7UBXX47yJdqdXOeq74M2DjuTa0ISXxuQl84xX83ipyLiSS041ySkWmLvhHiqL8q6Iog4qiHE7+/FdFUf6otKZKJBacLioA060fNttbzjj1shWeyabKCkNteHoDQyeHcsbr/fH32PC1T8CGDeZjvl6r7GlpuKeOfe89uP56Qota88aqF4plN2vj8MQwIxMjuR59naTjHZ+kMqEGDufQ1xSk77Y9WqGf47W0H5pD3zNN+NbdZTQ6i4z7NqzVsaX0fdcNsRgdLRC4Pcr1fzhkXqk1z7NRkWcgGxvPteGZzq5A6+R7y+mzV2IMvl6lJ/JGZEqVqySSSuGkSNMM4F+BNWhf1VHgXeADgBfN9/Ij4NNCiFMVae0UIj3vU0ggoCUvmX0eiUx+e4pluvXDZnsDWwPse9N4XOCDASLrp7hfVfa2Qx1SWfTNRaZ64O4EvNbpwvdewnjirFlw4oT9GzU2ou55ivBrXZax6lax3PnaWAiXgPhnXy/v2E7GHKbCSEoJlwgEUF/ap3naZyTlLLOKTOUcOpXPho3n2vKZfgMi/2Q83pRi563IudC/LfC4PMQTcdyKm5hVgTGJpEopp+f9y8ANwM+By4CZQogPAjOBy4H/Sv7+74pvrkRiwumiAjDd+mGzvWZKNC5cHHr/0NR7u6rsbUd4d5iEMDHOgQQQ/lgCXNrXsjobzYO7Hjo+WYM6R5eipCioZynaMZ/TjlVnJ383NITv7+6nsxsiWwWd3VqMfDapeOvI+gid13XmJLtatdFMSSWbs2rn2Dd67XplN2zQwoUqOYdZ2vZ0dha3KQgGCa9WMoY75IQUpciW6Gzb2UbbD9sm1zNs47k2VZdK5QyYHG+Kw2cv7TX/6U3aWv7Z447mwpCknYghEGlN/ykpgiWRVAgnnncVOAZcKkxOUhTFBbwAzBFCnHbbWul5n0KqzHtaNNOtH3Zj3gcj+P/lSkaZyNEdBybH25Uv+a7K3nZYeTRT+N+C3n9yozYq+NfH0oagV/FQPx7PjaeeNxP/rSOMepPHZMclDwOKAh6PNnceD7jdsHQprF6d15uZr40NMxo4GTtpqcf+oz/+EWuWrik8EHafBVWFRYsgbvIWoMAc2k6iLtN5qCqB/7OIfQus31ikJDrdLrchYXjSPMOTFfPu4NmzyrFYu3Qt+9/Zb2seCj1b6eOy3noUPdcSSYUpp+f9A0C3meEOIIRIAE8mj5NIyoeuZDvt7dVr8OZjuvXDTntVFd+qFvq+Had9H8wbBSXLaVtxb1ehkuZWXsampvLGwdv0Ipt5NNMISLiA1lbC65sYrcl4cKMixmith/D6pvRchB+4kdFaj1HLfFXqeiLj9YzFYGJCm78C2tnBhUHTgkMKCjcuvTEdi33dhdfRWNOICxdzZs7hR3/8Iz5y7kfsxRjb9cqGw+Zx/oqSdw71OvdWOvZ6LM8bjOSf3+QGMnjiLLy6lxZel5fmec0EPhig6ZwmU8Md8j8rVhVXi6rEauO5NsTdX3I7feO341vq4HvLwZtGs7yZoYkhvt/3fdvzl/fZSt0+S7Wn2DUikVQDTjzvLwAvCCH+V55jHgL8QohLy9S+qkF63iUSE3TqEIHPwb4PGQ+rWIxvIXUKMy9jnaaawthY4TcgdqQ9HbxRSRkM7518z7Q7/qMuev/6EIGf3mTqRfTP99N7Zy9gMy7ZijwKHvli3vPNYz6VFCDXw/m1Xfh6+o0X0Xtlrby3Lhc0NFjOYSE1HStMz1O8tL/govOJRO5bjCVL4IortMJCLS0wOopaG82Jedd70+14hwNz/UQOrE6vOfXuVvw/bjGMa/dnunM0760891PmXXbwXNgZl0LzZ7b+YomYZcx7sWtEIpkMyul5/xrwaUVRPm5xo9XAp4H/66yJEolkUqiE6opOlzl4OFOlMkVFddotdKHV/p5M/OzmNah3rdP6vW4dLFgAQ0OFvb6FvPopzLzIQ0NarLaOlEezeV6z4XdeobD6ylvB5yO4MGhaSfTAsQNpz2DeuOQZM8DjycTN62Pi82hn+xp9tC5vNcS358yjyVqyUh3a8PSGXA/ns5vxr34x05b0DUy8smbeW5cLLrooY7gDam2UjsvfJ7BlJR1PdrDr9V3mhY5e6zHtcwrTImEiyt5zJoxvMfr7tTVx5ZUwMgLRKL5hLbSk/XmFQHReRilmGOjoIPir3+EV1n92vYqXwM8P5Ky58N9eyWiWyk9qXNf/ZL21ylNyftQr/PgfuIgtz26efO+ygzeNdrzmhQpVman07PnsHu5ceaepak81FISTSIrFief9NuCPgf8J/BTYDbwDnAN8HLgW6AZ26M8VQjxSpvZOGdLzLpnWqCrqqmbCy4eJnCsIvqUQ6p+Nb/dAaWE7Os+3OhvN8zjTRVRJVD6O18Tzrs7x4P+Cm1FPItcjuaYb36oWTU7RDL3X18Krr961jvC1DYW9yG43vPaa6fgW0vJWh1Qu6ryIifhEznl5tdiFi/qoQt9oK751d6F+6lr864YyXmB93LLO857toV12zjJ2HtiZ1nS3o0Uf2Hge+471Gfo6b9Y83ht/L9fDGYP257KLIiW5+WZ44IHMmFl5b887TzMEyVpzWd5ul+IiHo8RI557zxdr6PzWy5Zr3tQbG4f2ZzNtVWdroUmRhdpmNbTbmAwMZNZTVh/0nnn93NbHXPR9O67lNaQusx72LTRefpZ3FieiRgWiwFw/kQ2vw+goHddE2bIy916T4V126u3Xr2UFxVDArNztNptrFy5u9d/Kw2sfLss9JJJiKeR5d2K8J7AUvsoh+4IKIIQQbquDpwvSeJdMZ9R72vDXbjMacuO34/vmwyVc2GhcqQvqCN+/1rQSa9kxuX/H9S62fCROVGQMIBcu5sZncMvzE4R6hNHYMgsjMQnZUGejbQxmujJGbcxF37dO4Xtf912qKPCnf2pZXMZKqjHFis0r6HvbaAybJdyZXaNjRxtb9n+faHYSgoDmo/DEfzTmbNwsEwaXrGXwmG4eLTY1bV9czLbalwzt/UDtB3h3/F1jP8zCe1wuaGzM9dCaSQeGw+k2dLRgaqC64nESiYSuSJMH37o7886JofDSWDSdBGzYKOiThLPGI72eTDa44dUKe1ecQ9PKFhBkxthkI9jRAls+Sk4yuNflZencpRw4dsAY9nFiKZ0PHIBodPLD2Ci+yFT2Wm6a28TOlyw2j2X6LlGHVJo3NTM0MZTzeWNNIwN3DcjEVcmUUsh4N76XteaOMrRHIpFMAeHhbkYbyUlufL8GNhz7Dx4u5cImJcV9oRCdk5WEa3L/yNIeojoPcIIER90n2XIpbL/ExNgyKwQTDEJvb46RGl6taFU7s0MV3F7Clws6n9S1TYi8Zd1TUo1WrK65iP3xvlyjNA6BxiZb14gM7c813AEUGJinGaB9syE1S2YhL2PRMRpqGoxGnlUJ+2PHMhfMorGm0Wi8C2h6x6TRiUQmhCllYCdlHNPe3J/eRHD1MkI/qsN3ZIzIwqjBix1NRPEP17L6wDh7F2qhRJqHPGaYE72XuPsz3XT1J/XxXx0ntHU/EKejBXZcrD03qciXdJLwFS46n3Lnvh1IrSfdePmGofMJAe+cD199OLfh3R3w68Gc40MRD9s/6mbUlci8YTmZYOtvl9Ay73VD9dbQfyrp84OHofdc3cYmAYFXx7VNUQWe03xF2/Ktd/1aLrS5LRVfo4+1S9bySP8jOV7+sehYwbZKJFONbeNdCLGtkg2RSCSVI7LQ+JpeuKDrQ8fZOKSW9kcxpZE9VejuH3yyg97j+03lDLMVWdIhG83N8MQTRkMmFILt23O8+hFfgqiSG9QfFVH2Lp8L3e/mKqOUouGvqoRCP2b7Z7O8vEnvcWg3cEvhSwQXBnnhyAtGZRMFRmO5Boqj+F+TTQ1eL/vPMW/He+MWYUpWmMTj6725vUd62X5XHX2v3EDwzceNBqriZfWMC+n8+QFjMnPWnJhed2B7xsMbCKDG4pahLgBRD+z9g6Vw/tXmhYUsxst0bZisOZ+op+/T3YQHNrH36S4CaoJQTwLf+I/oW1BH+P51uW+53ginNwCh3bB9eabtSkKrJzDy2xdRVzWXHjZnQrliyQttbsvB/mP7DeE5Mu5dMh1wkrAqkUimKcHlLeidsABCYdoWLbGSyQutClE/o94yAS7qgb2pGOKaGti6VUvs1F/LJOEueFWrMUnU5SXwkevhrLOKK+tulkgcDuM7dkpLfnxWCzFpfy75xuCZXlvj07q8lXjCXHNcb6BYJQyOx8aNyY1mJezr6ggeGMFjcrv3JkyMdwUGLYx9M8M2vDvMyCld0mZsjPA5LxH6tYf6U5lEaW8M6uMuQrdvNbZTNyeWXuKnNmhz8bvfEV5lbbhDMsn09662LvJkNl5Wa8MiydPXFKTzSYh8V9D5RDLsKxrFd2SMzp6G3EJbWffzDWuhQuteVHAntDjWuAsevUTgXzeEer8xqbpUTBOpK5m0XgLTqa0SSTZOYt4/glZZdbsQYij5WT3wbeCPgAngASHE/RVq65QiY94l0xlN/u984iaVMwPReUR+9mFrKcQqxE7CZ3h3mB37d/DOiXdyvGs5yZIeD+q5szTPaix/fG3kjQi377ydl97NxHUbEjmdlnW3SsicPx8OHDA/p7lZUzopQMeTHWx+dnO6wmQ2+uS/1HiOnBrJ8dR7FA8NNQ3GWGN9HPrICOpPtnPR3TEm3ORkRqU043PmQLhoP3g2nXvnwiuvaG8sEglLOUGr+H//UC293xhPJ5GmQ2RO+LVNjlm8fNZ1LeU2j7iJ/LMrb9x4ahxtxWIXaEdBnBar0t2vY+gHbDn/3dy3EzFof2MenQ+9bb8ddppaZMx7Oe7rVBJzqtoqkRSinAmr/wZ8VAhxXtZnm4B24DBQB5wFfFII8eOSWl2FSONdMt1p29nG9/u+T4KMAe+Na7J2nU+I8lR8taOL7uQ4i3M6Pj7Mltm/JapkGYQmahSmiiwnEzkx7x3XK2xZqeTEh6ev1awlR0YO/JzLVh1AZKfrC7j5rbN4YOYN+L60sbgxs9Kpb2iA48fNz/H7UZ/5ibWhkhyngPsh9p1tVCMB86Q8dUjl+kevZ+DoQM6xViofOcbSr35H6N+P8od/Av3nGu/nVty4FJe5gWRl2EYisH49HDzI8vUxBs6aMFx3pnDzuWcTuUnIeTTss8eHSISOq0+ypW4wJ7nZIxSajgpmRrWY8ZEZ8OjyXM+7Cxdz6+Zyy8W3lByLrQ5GCG9bT+TkQYIzFxO6fSu+pmDuMfe0EX55m1HhxuWCu+8uGLYWuHc++7xHjZ9H5xH5ah7jPTlWat8uwpcLIj4XwUWrbanHVDJe3ex+xRrhk91WicQO5TTeXweeFkLcnvx3DfAu8Gs0mcgGoA94RQjxB6U2vNqQxrtkumPHkC1o+OS9gYOS9zaLt1id40RFI+eP868OEfr3oznJqpbXSknujYyw/HMxBuaTq7UloPlt6P/nEjY9VkWIamthfNz0FPXPbsf/wR+bGypZMo4d10T59kcxDY68uelmHrvlMWNzrDzRunG1WktrDhgNXa/Ly7rmdTTMaLBvIEUicNll6RwCf7v5pgBMJDBTcwHGDSLkrCN1jgf/5+OM1rqJihgexUM8EcOdgFhSTaYuua8am6EpvpTTO6sORvA/cpmWBJ1Sr4kq9N22J23Aq0Mq/vAiRt1xo8LNCWs50mw6drSx5cVthuTn9ktup/OWhy0apz1zqjKCf30sk3uheKg3exszhciCS5LTjXIWaZqL5mFP8TE0b/tWIURcCPE+8GNgWVEtlUgkFcVQxOR3c42qK3mK9xTEScl7O8flOce0GJRwmcaqphLfIusjdNbfohl4WQTfUgyFc7wuLwE1oRXficU4OAejSK6C9rmd9lthVUL+wgu1Sp56amq0GGyr4jxZ4xTabdLmJLt+t8u8OTZjgA2x4kqC0RmAS8mNP1c0Q3fjVRszc5CKzc7H+vU5yb9XvI5pPD0kk5BnugjfOC9TCAjMC2xt2JCzjnzHY/R91037WBOBDwZoOqcJNwqxLFWmMQ+sfUmhfaLZtNhPKYS3rU8b7um+eAThbeszx+wOG49JJl3T2mprwxi6diP1tY3pde4VLuprGwlduzFP47S1FA7GcmL+oyKWWW9Vgiy4JDnTcGK8nwBmZf37KjRN96d1xzSWoV0SiaTMZIc5BBYGCM1uMRiyxSikpBNH3Q/RcU00t3Km2WbASmow36YheU6qYmjPeeBKZAw6TYlFIbQqlJMAqt7TRseOtkwi6t2thuTBUP9s6msa0kZryrPa+lyUjk/ECHwOaqKA/iWlgMWpyBanm55UG3ft0kIfUoZ66i3E1q1a6Ey2AV9TA888o0lAWhkqWWPrG4YPjFncP2GSvUwm2dfjytzXpbhonf+JnKTayGs9xja4YXDZOfTt8dN+qpnAXD/tK4s0dA8ezG3Xbmg4Zdywpe+tJNi7IAE7dsDKlfCJT2hx+LWaDnzgdq0Cq9rzH4a15zseI/TQb2n61UsMHhkgpujURzwwOE+h88ilRLYKOrvBd+BNW9WKrZKqU0ROHjTKXHpg70Sm/5HDEYPkZ9QDez/sho15jO/sPjb66Lt7gPbA3doGJHA3fXdnwqZM25lcS2ZKVdVmGMvEU8mZhhOd998CLYqihNDUpj4N9AkhsgPpzgOMgXUSiWRKMZXEW1BH3wJNK9tUn9rpdc+O0rtSk6ZLe/StSt7blc7LOkd99YWc1/eeOLgFXPwWrH5DIXRBa07YiFobxX/1PkZf1I5PywDu7sb3na4cXfq+2eTEvbYub6Vl7EpGhXauO2U0psrUCe3H1lR2j0n7LRPo9GFDHo9WjfXii2H16kzMt06/PvX5slc+zD6xzxDC01R3HgQX5IztH/4Wtq3AcGzLc8OmOt++Rh/dn+nmyoevJIYWBx6Px2h5fC19j7q1yp+9vQSvd9H7EU9OrLjX5SVw+S34vtpJyYEKixfDQCb23jesranwmjnsWO7h2NgxQ+5GoO9Y5q/P0aOGgkq95wq2Nx+n77uenAqm6mxo/l8TDNVMmL6p8AqFwNseeOJRbVxfeAEefFCbs5g2HmzfbgibKihDCQRnLqY3PmBIJA3ULE7/O7gwSO+R3tyQEOEicJU9r3t6DC2kFy3bedkafL29BA9HjVKcVWYYh1aF2D6w3RBKFlpl/7tMIplOOPG8dwIXAq8Ag8BFwKbULxVFcQEfR4t7l0jOTMxk/6bgmnpP2oZfbDCGWsTGCN+/1iBLZyqbaIEhfCL7lX5qM9Damtv+VqP3u+CmIRQifIU75/V9zK3FIK84qtD5q7O0pNGssBG9xF86vOS1LtSvhuj4coDA8ggdA9rr/+ywjq7+rnSMMUDcDZ4EzBmDWae0KqV7HnITfNO8/SmDaMtzW9j35j62PLsZf/h81HPrYMUKGB7ObF5iMc0Tvnq11q6BsDb2A2HUr4aM8oMDuQmlaQYGDLKEG3s8NE6QkQkVmqF7QjlFYMtK0/nt6u8ikaVKFCOuhXIEkwZvNEpoV5z/n713j47qOu/+P3suCAkJGQzyBY8vOA4SiRg3ITNxqHDt2nUUEkyLQxuiBGhKhBNP219XOr2s8L5pSdt03rxpU9kGQrOMbaApoQ4lwSq/JE0MP+KMgBihGEFt8OUAtsE2CAkJMZf9+2PP5VznphFIeL5reRnNnDln38+zn/19vk9twm05rQjPCxvG3vLty1n+g+UFjSUDNmxQGWp18PULOh55lv1f3E/9xPrss6WL2ksp/XsdLP3vhr4qydwvxAktEJkTosg8OD8BW8PdhaBWTiC8O5nNkFoTJ/RxSWB5nFAraNX2tKlcyYrSCC/bQG1MGGUu40LJXKavMUmfel3evJSXfB7/gso5D6itJRz12FKhHA3j0VgD88BCCSwjtamCCsYiCg5YBRBChIDlqT//XUoZ0X13P7AV+Esp5fpyFnIsoBKwWkaUojQyHlBKIOYo3NNOeSEpkySklXOQNxAxT3CeY4DjuUlE4yuUod7aai1/Zydsynq/aWtTf+cYE4HH7mTf21bfgFsKXvn9VICfLgDUKRDVf52f1/tez1lHx3qdhOi2Kdbym8prF0BHUhn9O7eY4gzSbT/fj/8TucsFEPjTWlsVmcC5SUT/acCi3hLt6eTu3zzGsIfMqQGofxdV9xMQ/VdjeSN/0WIIQgUM4wfInFZ4pUsFOn6pwNTzOrUZZs5UBn0wG8SZKwgZnPsfUmUZTtL9OCxe4nxdwwXYv6cJ3y971XNN3nxD8GijUbKx4ODflNpM1/BxAlUzCX/86/j+48eGuaCZToZyBfyWbQ7HGoh+agds2oR2aA+Ru5J0+VwEcqnNjMYaWCiu1vdKBe9J5AtYLYY2g5SyA+xPRKWUPwamFFe8Ct5zMC/uDkfO4xK5AjFLzUBawj3tPGkuXAiEUWu7kEDEPKnNbY/0XV4C96+AT3Qoz5td+Tdtypa/wDERvK2FA2/3GOgSAEkBkVc20dEUNFBybFPDpzYy+epoW68kBKY2Q3cqG2vQKOenh10AHS7ouU4Zf5ZAYa+XyF35ywUFUC1MGWc3/eUcki6ynmWdh7ngusfVxkVfXt+cFsuYCD0bMhruuufFRJKBoT4i2ko22AAAIABJREFUu1Y7K5zoEQw66tkbKCCHQtD/GOaghOBJOHijOp0xIyaSDFS5iLRIgicl+29UGYf1cCVhyWGB74JHGaF2pznpk6YWQcdtxrnkODdMc87XFKTjG6l6OswFX3d3waopZZnDcQgcPAOPt6okUb4CqVCjsQYWgqv5vVJBBTaoZFit4PKiFKWR8YJSAjFH4Z52hmOSJC6pO/qWglpPjeXou1jVBrsj/RpvDf3D/erIfmCrohXkKr9uTGiTIXRfjMBDZwk9tkAd96eO4cPf2I2wOSmUyGz5dLSR8F6Mx/1JqI27ELF43jraUhVqphD+q50FGQNOGUsROlqRHi4X0esSBbV9IVQLPaI+l2N2ULtnWOouvOr+UVNQbThsoWfsed0ayGp4lhu6DnU6FwaKp12EwzB5ciaYOfBH6v9txyZRq6PXWMoilCc5HHUz+RJG21+qANnwHgmnTkFNDdpUD1tn2wRveqDL57LQvmzHkBPdJF3nuXOhr29E62NJc9hTgye9J06CS0LbCxLOnVMKPQU/3GG92rp1dGk0V/N7pYIKbFAx3iu4vBgNA3eswEn2r0j1lpHe01Z5IQFtByXt+xX9ob1L0v1Y0kI1KEW1YeH7FzJl4hQaJjXw4KwHAdjy6y2K633L2/gfxqhAYy6/TknG/zCsnwv7ZsD6qh78jzejzWuG9evx7TlE2yGJSXjDWD5denlfY4DuMw/R3jOBwElo3wfdjyaY/7PjeIXH+R4Ux6G14xenDTc7xDwqE6gBiQTB3cfylguUp7b7888rRZdzk2iPNRt0wc0I3tZiuW9RdZ/bru6/dJUlNsLA6z+wniNneh0lHcHGg29G2oNqlnjMZfD5fGjP78L/JxPU2LkJ1gcErcs9dH5uF+0fbqdhUkMm06uh3ve04Vv6MD0/b2JZ7wQa+qFhAJYdhJ7HU6cj776LVpvE/2UXb9t0qVPwaMFjSF/n06etSkBFro/FzmHfeeh8Kok7idrAuCDhgtbPgVYn1SlZoQa33XoF8PbbhfdnKbia3ysVVGCDojjv72VUOO9lglNGyVITA40ljFXOu10yJlABgV/+sqHdi+HL2l3rEi4SyQRxvQpJKotreI8k0iJUlsZ72gg/sEbdc/lyePppQg8kWT/XRHORLtr3S5UBFh3neKKLmEjmT5hjM960qR78j7gZ8CRLSomuV5GZPW02249uZzA2aLkXoDKWvtVjoKp449B+ADo6VX0i8yA6A2a/A9ubJzDokWVN1W7uJz1G8gw7Xr9HCtwJSVKk+lHPr49DbQy6h5bh+/ZGh5uWtj7kS9Kj9Ubxf+9uBhhWiZaEl9oqU73Tc62vz2JA22XhVdUSXDPxmpH1kV2d9ShyfSw622goROiVx1j/IWmhYrUfgI5dKour9vWwc1bfzMNN65UQBr3+UuqTrlPOZ1/N75UK3pMoW4bV9zoqxnsRyBU4dCUDmi4HnFK9X+Z7FhLQB6j76YLsLL/NERxnG5TpAH9sGq/LsyrZjN7oXtiJ764HoK/POdOpOVByMkR+r4Guj92aP1unQwZTu2DLQg13vWFkjiMARc9pH26m48s71WZDnx0zbcCmdLrMwY81CReLPvI5es/0lidVuym9fZfPRdONd4Ig7zPyGUxOgY7+N6DldXW60HRGfdY7HQJvCMKHJuPb2+M8fp0yztqMU8PXuYJDP74tkyk0EozTdRMEzlQR/vvn8NXeaFyr2tpg4ULlAdffx2FsNkxqYP/K/SPrI6c6Q/710WGtLXQOp58f8O/LOfcKDaY2lKmrC1591dKWmTrn6E/D7QrZjFzt75UK3nMoa8BqBRXkRb7AoRz61VcFTMGCV+qeloC+gcfBFOiJELb0Gyc9aDNsgzJt4JUukkMDDNTJjOcyE0T35Eo6BlUmIdsAU+ki8IZET0j2DXnpqF0CKwtoEwdNebtgy0JgDgY0G+6gAiS73u0Bvx9fdzfdA21Ejj1F1w2SwEklaejrF4RapSX4cdAFdRPqDGok+VCInrwvFqPjl2mD5oc5T20ieyPsfn03R98+SkImiCfj9hrlDvrjLSckHZ26dnG5YNo0WLIE1uaZ66XkAHAqS5oqkuJD+2JxOtJ0e28S6tfCjh3WtWrhQtiyxVCG4BuCgzOMnnevy8uS2UtGLkdoV2chYPp01WZO62OOtdbnK2wOp58ffGU/B2+wet4DJykqmBowrldOHvEiqISRvRH6L/UTT8adn321v1cqqMCEiue9QFQ87wWicnw59qBp0Nys6AB61NcrbfASX3BOVAWXcJGUKUpKSkrPdw4O3WC9R+DsJKLfVrKHFhk+l5daT42i+5gTSRXqUSuzR87Jw6tHhm7wk9S4D4etZaipIfCZfvZdb810apYSzFm9XF7Jr0aKmou56DVgpKE4PvsK9VfOdrh/sb1nu6EBzp61ts/SpUaj3utFu75Gjc24lR41YuO91DFarrVW09DmNeNf2pede+kTog0efLKOwJqbbWVa847VMsy/O9fdSfdb1mf7r/NzcNXBgqtZQQXjCfk872UJWBVCTBZC1JTjXhWMc1QCh8YefD5lpC9bpgyWhgb171yGewGKH05qGs8tf04F6cUaaP+VoHstzH/dmtreIzwM1VYRWKnUQUDRSdp/JdRvP9yuNMH39lgSSRVseOsCWEv6vQlOKjKulA8kbfSE95Id93Zl6OkheM/n8Jr0CQWCV989Rugv56DN9+dV58iZCMg0FzNKPp4nbBP3mO9lhlmxxDYgs5z95fdDY6P6LBJxbIf0aYGv3kfjtEb81/mNwaFOQd9gv1b19lr6y7e3h+4v9YxOEqBSx2i51lqfT9VvaBntWgOBi9NovzCL7uf9Kki5u5vgbS14XPmDqctWNx30CcMK+byCCt4LKNjzLoT4beAB4O+llOdSn10PfA9oAeLABinlI6NU1iuKiue9QFQ87+MfRXjLDNza+iZFCXn+sDKYdu/O6HSbveoe4SEhE7iFm7iMZ5PdpDx9mWddwcQrdnQUwOLhrfHWsOjc9fS+e5TAiRQt5jx5x31OT7eEqgQ895SbYP/kjLqLuTyLty525nofCmTmou2pRoHJmdLwCi/tg410/PfE0e+LAsfgiPjQNvSY8bJWaX0akX9YQPTdHoJFjLlSET0R5a7v3mWgiQkEz3/heYI3Oec7KAf86/wcesuq91/xvFdwNaOcnvevAAvThnsK/wzMB7qA48DDQohlJZW0glFDMamyRwxTena71PEVjHEUoZnsq/cRnhcmUN9E9GebiBx7Cu1oShLu6FHwKG+d77zRq940vQm3y51RpYm5YaAKIiubjIZ7sbKBhcLpZCH1uTbfj/9bd7B+/7qMDKJ/nR/A4m3uebiHjV/+MdFtU+j4iTdrRJnGvXke6u/VUGOSMhQw7Ia7P5dAE/1Ev/Vn3NFxB4/ue5R9p/ax7sA6/Ov8zJ4221kWUDcXLcmF0h76Xasz7RA8dtFRUtIrvNQOxQlv6LXvi2K12fOhwDHoePKgqxeRiMqIa/b+rlkzOmtVudvCfPvUhmV9Ta+SVJ2rNmbaNUKp5PT3l/2Zmw5twi2MAvdu4WbToU1lfY4d5t88H49pXHqEh5abW0b92RVUMFZRjOf9JLBLSvmHqb9rgbeBH0opPy2EqAJeAM5KKc0pSMY9xqvnvWjZsLI8dBQUV0YTlbTaRhSh+JEZX0N9SkVGny5+yKuCFZNJi/c08F85PMZpDu1oneI4eWI7O6G1FQYGCN0Xs8pWmjjftvd1GPdWqUIPtVV1mXno6PVOwrJu+N4cZczr4XV5Wdq8lB1HdzjP71SZAp4n2HfNBcvtA2+6iX7XpbzzUz34v5hgoNpNTMbxCA9ul5vGaY20vJokvKEX37txXQFycPpHqvRR4Bh0VJnR1Stnecq9Vl0G1RPbWBOd/OhoPDOnmk8RwdWl4Iq8wyqo4AqjnJ73qcAbur/nAROApwGklMPALuCOEspZwSghJyd2tJBWG4hG1f/HsiE8mt7d8YpCE0Np6uh+YPBsVkXGDWcnwoKlqMyqjY22fNeCEsmUi9Nr9oSuXm3v1V25MvN5dIZNJs0cWSoB53GvaUT++m4GpDLcAWIyzsBwdh4GZzhQD1zQeQcM26zUsWSM3jO9uRMBpcoUvH+Ftb2li4CWzLSD79043d9x0z7YRODGAKvmruKl0EscXHWQjv+eaDTc0+3W1ZXXS17oyZ/hunsvok01nQLYjEHbcWSqV85sm+Veq0aS6bNAj71tBlV94q9RyC5aSvK2cqGYhGkVVPBeQTFSkX3AFN3f9wAJYLfusxhQCVwdQyg2VfZ7DrletmOc91ouWLjdX2rDt3mz1XuopxOkNj3Rh84SM2cLFdBznTrK7x66E9+3re0Ynhdmc89mizfNkDq+RNlAY+Vs5PSSSUiYomdjMTh+PPMsW9nKUo2VSITo9GHrZkBm52F4XpgNv9rAcGLYcI0njpINFPYnpIEZgYKkPW3b+2KS8B5j0J/v3TgdP6uGb5i8qbn6Iscmy+w1TUtOdn62k02HNmXGXNucNlo3t2avq/Gw+YsJur/jUZsGB0pLofW6bEHzpW44c8g+mmMdZk+fbZXFNGeuLXN9C5qvo4hC5WsrqOC9gmI8778GFgohrhVCTAaWAlETB/424M1yFrCCkeFKekzGBcqojnNZYwuKKdPW5QS+eh2hFdeh/clyg0cvw5/Vpbj372hF22vDEdZ7JVObnuAJq4oMAEJxrCMOBDpHb9p5st7H/n6oqRkZJzlVTq06RqgVAstihH4nofjBeni9MHNmhqMf3qvoP+m6jchYiUZt28mbJDMPffU+nlv+HFXuqsz3niTUiQm0JmfisRHWmCBdhL+xW2WoXb4867GNRi0eXNv2HmhT1CY9PB4YGrJ6f3PFsuQ4qbE7+esf7ufujXcbxtzdG++m/1J/9joZZ6Dao2IgcqiUFFyvYjd9Nihofhd6agWqbZcvh+uug/e/3yhbmXIiaN9cbZmf249sp8Zbk1V5kq6swlGZ6mteN/hfq+le2FnxfldQwRhBMZz33wH+CxhGedgnAQ9JKX+Q+n4CcAL4mZTy90enuFcOFc47448bXkh5y8SrHg1eZt6U4IWU6fFmxUdPZfCsvQTdW+ozWS7zpZV3RCCAdnQfq++BTX5ICEDYXFYMJ9aOL1xTA4sWKfm+FCfZTnUlV/ZI7eg+SybTtLIN8TiRFkHU5yJ4+3zCa36WyUSrTVabj6577iDwgQdyPyfXWAuF0Lasw78ybtTRjgu6Bz+P7ytrstz4Po3IrtV0Heok8EqM8LPnIClp/hL0VenaWELdMLz4OMbMual09Fo9RD4G0ZsgqMsmGv3Wn7HS9SOO18aZOfkWNvzraYLHLqr29nggoTY2kY8mic6A4CkIv3WHyoLb1gabNqmNbVOTKu/xF4jMHSY69BJBTRLeI5XhnOJc+3/wAIfO9hbW/3bdVwqnehR45wXPb9Oztame1PhyE3w9TvjN2/H9y5Nw4432uRdMCC1vYP3Ms5b5ufSDS6mrqsuqPH1lu6O2frHriOO68R0XvoWfU4G+Y3ndr6CCqwD5OO9FJWkSQiwE0moy/y6l3Kr77h7gW8A/6D+/WjBejXcoPN197puMs/TThZa3TPUq2Qh2Kn4ZNgOhZ0Os73rcmBUyDu0vCDpu+zJ0dJQciKb9yXL81U9mFUzSy4jOgC+6/gVspIpul1CI0CuPsf5DpuyRUrD04vvZ4X6ZAa/K/OpNQu1wKthWbxA3N2ckL+0bI88YSn2viX4iwThdM8hmWh2yv5b+fohnOebLH4Sn/KCXhTcEKeqLY5aFTEAtE9i408uiT11QXSUAqWz95889RPAXr8PQENqJF/F/MWmz0XHjk0qyEsjUx7AhkYLamEt5vr+yBm3gFHds+agKtLXZ2OXDSOZPuQNR7ea3QDB90nSWzF5iXFNTz9YO7cE//zADImZsy3Xgu/8heOYZReFygtdLIDyFfd7Tlq8s89OhvqWsI47rxgHo2OVSyd3G6rpfQQVXCcqapElKuUNKuTj131bTdz+TUv7G1Wi4j3ek+YLRlVE6PtFRmid4JIFYVwKFlrdMSXzKHVtQjkDj6Mmo4QUMqcC2G2SGFlQqrcosPZg2zlzJ7D2KppkUQGEqul3CYaI+l5VvLiSd084yMNGVDbZ1wYDXhupz/Hjucucba6kx5lu6io4DDUS/K+joTG0QzNem7xU3Bocenm403MEUpKgvjlkW0g0DXGLFfTrDPfV/KWHl+4+qtp84kchHk9bfeiESTGTLmSpjJBg3XiskAxNdRB6oA5+PyJMrrScyUiW0Mif88QgPVe4qS8KvkjnVZQ5EtZvfEsnpC6czMqIZGk3q2ZG/aGHAFbO25ceAnTvzGu7U1hKc01rY/HSobynriOO6MQNV5rG87ldQwXsEZcmwWsF7AOMtc2ox5S3Di77csQXl2AwEZwQtGTy9cQi8ITJ8WKcsqfmMpmjfYYtBjIBpg2SzoxZLGXLgC2sfbcpwjbce3mrfLq/sAWx4yZMhGPhdK988DlyK5VTt0CarzK/+lXHmrJ3DnevuJPRsiOiJqPEZ3bvzj7X0GLvlFmUxO11rN25RAbR2dTAEKaZgq5TjhnMTsXrABRw/m9qcBIP2v023SbqcqTI6KfLseX0PoWdDPOF9kbjNGGkcmEjdhDrDmKurqstm5h2DnGqnzLrgbBBHT0Yz6kKZa/Vt6XJ4/TY0ZJwI4QfWlDQ/DWUoch1xXDfSY20sr/sVVPAeQU61GSHEfPNnUsrddtdWMEZhx8WF4rnr5VD+uJzQlTfNXY7eBMGpQ4T7tLIbBRk1huF+YjKuKBhJF+Hb2kor/oygVVGiyM1AeF6Yzd1PZ7mrcVRg26HJsFaNg3TAXyG0Kj139mL8Ih7cxMlalN44LHm5io5/2e+sp51rzIXDSmFDRz/Rrq/Bf/12Bg4MEkvGjImMdM8N/OQI2r1R/Dtarcomb/42m6sw8s1j0PriOTb7jXXwpIyUNPWkfwLE3cNwugeAQ28d4tF9jyIQSKR6xj0uul/0GKUUXS549VUVkAjwwgvKaD91KsNLz1ZAN4/s5hmKYrN5jpEK4wJ236w2GJkMmzgo5SSg7iK8W4PFEz5zysxM+we/9DgHb0gaf5s23PTlPHiQ4MmYrSLPkbePcPjMYWKepKJTCeO97vXcQXjVTtsxN9rZOkuFWW3FDDuDODgjyMET+wwGfKYtb78d3nzTynmvr4f92fnjg4Lnpx1KWUcc1410QOxYXvcrqOA9gpycdyFEEiObVUopzb6U9wTGJefdKQAQYNA+uKmoe40DzruFl2tKjlPWR/ZGifz13XRNHyZwAsJRDz5ZV1Ib5eKqQuEBm4YAyJMQntxqCJAstTwel4dEMoEbF3ESarNCFd1/8By+JpMBlmfsGALq6mcrQ/SXKkA11NLP+qNbjAZTinqRdGUNi+4NSplk/aQj1riD41MIP3NaBZ/q+OanauGulSCNiU15/r9nsuma11l/Z9ziObWDBw+rDrrp2JlEq46pjeIMZUTrjWpbOPDjzZx3QAWhfv529tzq4kjfMRIySdwN3qSgNumh+7EEvj7QapNGznuqjTY+A4uWYuW8f+GXGaNZ+/kO/P/1oOW33Rt0Yxkc5pYXl8tFIpnIZM5N9xdCd6+fN+L7yH1jP+DdhOiJKCt/uJLeM73G+mHPz9f6NPyPfkBt6PVtuQ58u36pglZXr1bJwUAlCCtzMGipsTOZdeOFnQQOvU34FwLfOTn21/0KKrhKMKKAVSHEMvNnUsony1S2cYVxabzbBQAKof7T8y0LVVcZh5lTQ48tYH1Vj9H7NZJAuFwoc0ZQu0Bj4IpkG3QKyG2c1ki1pzq3RzBHu2hfD+esj1NAbUM/3NqnC/w8D4E/nWSfRTTWQPTvrEF/oVZYNxcDtSM9NqIno/YZTx1Q7a5iyYVb2S7+h0GPNAYomgNgQXnmp02DJUus8yg9z/bsgYsX4aWXlKdeSvB4CN0ft2Z/FV7aG5fSsadO6at/tImI+5d0DRxVG8lUG0VnwMrfdXN8qmDmtbez4feetHi7td4okSdX0jX0MoF3qwm/fB2+ub9tLKcuKDNyV5Iun4vAbS3sfn03h96yBvdOiglWdAtlBL6bGHdGoNkI1iPXHNT6NCLP/BldB3cSeDWWVZsJXr4ThhELFoy3db+CCq4C5DPec9Jm3quG+lUDO/5s2gjQo1AOY5q3O17g8xG9fSKxU8aPSwkkLUhurcxxAXaJSULPhhwD0Ow2IyOVm0y/uKPuJ4hNsXJnqz3V+aX8nNpl61YiLf0562N77C9dLDki6dhppJ4Eq2Zy0GX1vAfmtIJ4yjLuozOwcLLTY8PuubkwlBjmqYn/g5QyQxPJBCjOsyrCkEzCrbcqQ8iOTpSeZ6GQIXkU8bg911zG6OrrhQ7VFz6gIxAA0/4jeBIO7fmgouY4wNcUpOMbedR1UmX2BVvo+HLWmFu+fTm/fuvXJNEplbi8rIg30rHrSNHJ0HKN3xGP7SJgDvyElNpMYiJLXq1TJ1rngXrj73z1PjpWfH9UylQoRpzgaLyt+xVU8B5AwRlWhRDPAF1Sym+MYnkqKCfs+LNOnverlMNYDu64U5ZIi6ftMsQFFBOAVnC5naCjuwTvi3HQ7O0ttB0deNycOUP0p08Ru9FoVOvrY5vZ0VND+BDg1VG/XC7CP4+x+QEXA26PijtIB/c9sAYWX4Bt24zFOgkHZwhiuuyl6Tqln9s/3K8oEibuth0k0nKNkyIMXq/SSrfLqtnZqfTUo1F47TVLu9ly2nUJnzJG7e+8RrBBKO11vef/yBHVt6V4T/NkAt1+ZLvBcAeo8dYQ/oUoemNrHr8HTh1g7f61tDW38fBHHjZmZC12bBcJJ7WZW98comPjEHi3wNM7xs1JQgUVVDC+UYzazANY/AoVjGnYZUWcPBnq6hyzVo7FLKEjQalqKnoULLeWKwtlmVCwqo2mEfmHBQwMns1fbhMyY2D9XEIfO4dWHRtZxtF0uwiTZSslQU1aVVR09clk0Jy1VKnYHJ9C95uL8H1/l6Ij+f2KgpJI4IseofvRBO0vuAlM8xsVS771LRUMmFb4cLkIH6qjtmqy7dhIP3dV42fxn3Yx6+2UDGbhaTHUPe0UYdLjAqwSk/39cPfdima0bx+cOZNRvQn8kfp/W7epL+Iq3iA8L2zMmOs9zfoPSfwPqwDcDJJJxbVOZ/e87jr1b62Aue4gi6l9czULtiygb9gYgCkQLJq1CJ9/fsGZR9Pjb+6GuZy7eC4zfpMkScgETx16ypqRtQQp1WJgO+/iEDiR+mOsS+dWUEEFVxWKybDaBRyXUv7B6BZpbGJcct7Bnq8IZUvoMR4wUs5nUYmMRpkfWlAfpbyjgYfOss/G65srAZPl/jreNqQyjs6aROD+FcW1o6bB3Llw2sg9tyQUki5qq+tt62Mb8BqJFB5noGlo31xN5HynCiid00rbxx5m06FNzmNDx9fXJpM3o6weniTUyQl0n1yI78BLymh2uaClRY2LxYuVgZ6r2ewSLl2Czn9zsekDSbpugkA6g2pT0D42wZTMSZsMkd+E6I2moNr6eujpyT1eAwFLmbXJ4H/EzdkJCfuf3Bgg+vFtBQW85+KWF4KSMrIWAMu8cEroFQioE5MKKqigghFgRJx3E/4P8JQQ4i4p5fMjL1oFlwVOfEWbz3J5mMse3HkZMVLOZ1HUm1HmhxYk7ZjyjgZPwMHri6O6WMaAibfd8RMv3L4Cim1Pn08FZ5oMbd95ZQBllGDi0wj/837b+tjypYuIM9Amg//GHQxcq+Q8D774JJtf/J5SyHGqj+7+vvOw8T9hzc9g7pc8nJ4Yt/9NCk3XN7Nz6U7nDY4TnUgH24RLVYJNi++gY2893B6Af8luEG1pVSbtev1m4OANSoKyey34+vvzctDtyhxpEQx47Z1AmfGWToaWZ2Nrxy0vFCPJq5APlnl3bIjwhl5853Vj4DLRDy8n17+CCioYmyjGeK8GngOeS/HfDwCnsTlIllI+VZ7iVVAsRrKwlztLaFlRiE74KMGWd22ijNi2+3lGpcyZzUi6Tb622Hj/lMFp0QZPQu3E3FSXnMafjgZkrm/bnDY2HdqUe9yZddzT9Tmf8gp7vdC+BBwCgTN6/TMgeDJG+NAefMGW/HEGqXaKDGxl4JZzGY57zAUD8WEif303Hf/ykn3f2BirviEvS2Qj603BsWZUe6pVGziNXRtd+zQFKC0TaRuc6pJ0Db0M215RbbI3QvS/VLvPnjbbutFMqMRc2mTJgs/CWV2yppgb+qpg7kpYcjhJ+NAecFhDtD6NSEs/0XiSoJbi0g95ifqSxIS9190wTwrY2EZf2VOQ4e4RHtwut0GW0iVctM0pLa9CITA4ATQNvuMHr+kkoYwUOTuMOI6lggoquCpQDG0mzfg0Hxbrb3DVasGPB9rMSGkvTnKAoyKrWAzGgMa8hXpzWxu+xzehde9m9ceG2VTzEkkpkchMUGX3WvC9mVtPX+vTWP3fq+l8uZOETDCtZhoTPRNpubkl98YrR5to31xN5NhTRG+QzD6jLu9tgMDUZsLLNuB7fJP9hsJJWlO6aH9tGh21StZQm4y95rtwE9cFitqOO70M4pEjWUM1VX7tmY2s3hWmk2Mwwcv82+5h0tFjvHDuKEenKbpKPE0fESld+XmtzmND106BZTH23WRtysBJiFY9Ym9YOuRK0Bbfj//6HzDglZZU8qCbN81h237S9nYSeWUT0Vf2EDx2kfB/vo1vwA3z58OPf5zJwxBaIFj/IWlJmrT0RQGz3s+m6peRUpIkidflpcar8jgMxgYNAb6dr8yj9dpOzlZZg2oz901AjZhAcmIVFy5dIEkSFy7qqurY1bbLECDqlS5qY4LugTYi87Bq8QMTPRP5/Vs/yZrtAS86AAAgAElEQVSuSfieP5x/A6tphP74DtY3Dxvqm06Ilf5/up7zfPPofLlTBQoDHpeHugmjk8PBqbyOJwnFOBuKuHa01uj0+rb79d1IKXEJV/41yPTbyklABRWUDyPSeTfdyKL57oSrUWJyPBjvI13Yxyznvcz66SOGKQFUX5VKGKSHV7po32+VM9SXWevTaF7bbAnygwIMEYc20R5eqjKSprMjpjnrW+pVkGerg6ELjol3aquMY8BunJlR0LgzGT/a4vtp3vUgfVVkDUzp8G9S2uZzUwaykxGla6dQK1Z99DQf/JTfWT5RX86mJti+HQYHVUKmFsGeWwRHrveQIEk8adq8fNXKydemevA/4mbAk7TEFfiGUonUFi2CgwfRTvbi/8IlQ9KkmhRT43yVMcFUut2XNi+lbkKdYaMZ+eu7LUaxHdIGshmzrp3F8bPHbdeW8Lywswa6uW65Nt2hENqWdcbxF4daVxWdK5/LxCY0TWti+9Ht9A/3W5Rtxp2zoUjHRFHxN4UWN7XuZ5SVUihkMzRm3xkVVDDOkc94L1htRkr5ZKH/lafoFRSLkdJeMsoeH24ncGPAqNZxJaBpyvh64omy6qePGCkOdiSojAyz4Q4QE0m6bsitpx/ZG+H8sH36zXgynls9w4HrHTnfyUB80MiRnugi8s1FSn7QiTueqpPv3Tjda6F9v/JItw82WsaA3Tiz1L+QcZemUUSj0NHB6l1ho+FOjn+T0jY/2WW5j8Ho0bVTW7fKzJq2Tb0JXdr3pNV7blvOurqMV9x3Hjp2Sg6ug5c21rHqQpNV5camnyLBOAMM28YVEIup+9fVQUsLvj6Z7Y8TaqOxqBcueK2GO6h27z3TS8cnOoiujNLxiQ58j28iOj2/4Q7YGu4AL7/7suPaol83GiY1IHQdZalbLkWWaNQ4/lL17f55I8Gbgpk61VXVMRgbtBju+jJdUeSK0RjJtRShNlVMcVNxBuassXnXIKwxCqOt+lNBBRUoFMN5r2CMoxya5iNO6FEumD1SZlxJbfqUQWbHR07DK10E3tBZimApc/Rk1NFYgjyGiIOmfHQGViNLJFUCn6h03gTJ7HcZDjpAoBq+Ydy8FZLAyOvyEqhvUpuvAugAWp/GpglH8yq4GJ6RQD3D9oYpb/lrr4EQaHWS1s/pVGKkkn7sfDqlFuIq0I9ht2lKJvEdf4eOb51PeU3DWd6+TT9Fb8JASwKTHrypT3wxY5KnO9utyaUybWI336NRgtOswcuFaNenkZAJPC4P8WTWwDNLenZ8ooPoySinLxgVhWzrloKBcnHvRcLHPPjejWfr6/VCe4uxOjk2j6MZtFowiknWVmRit0Lib4oubo72TK9BTtSYMR0nVUEFVzGK0XkHQAjxkBDi+0KIQ0KIl3WfNwohwkIIG1ZpeSGEmC2E+KkQYlAIcUoI8bdCiJx+JSHER4QQTwghXk797qgQ4n8LISaOdnkvF8qhaT5mYPZI6VFAcNio6tUHgyqj50ksGuWgqAe1VXWED03OqfkenBE0eCnNyGmIOGjKB+e0Wj1z0qWM3FS5jV+mNhS5vkO156f//dNU/101a/evVcac8GTKKRCGv2s9NYS/sj2rV75+vdqMOWiJR/ZGbD3JFqS95nFFxwjvtbkmvfFbvx7t4mlCH5fM/SKcq9IZvQKSQsk+4vEo+cb0zx3GjtanEbr3IoGVSm/doJ0OaqyePQsLFmTradNPs89NsPa7hCFP6p7pBE4XL9o2QbK6yrF5MvM9fWoVCMDFi4SjboM2vJ3h7kLgFV7zLTPXuxNJvPo+tllbHPXQ01r3pjHlX+dn3YF17Du1j0ere7jjS3GiN7uz15rnuaYRPHYRr81BiUCMjfUuz1wq+VpG53TUrs8yRXF5aZrelM0dcGof6w+sx7/Oj9anjcpJQAUVVJAfxXDevcD3gU+lPhoCqtPBqUKI6cBJ4G+klH83CmVNl2MK8CJwGPhH4Hbg/wL/JKX8ao7ffRMIAE8DLwFzgDXAT6SUi/M9dzxw3mHkmuZjBjZ60gBMmgQrVuT14o4qD9PEeU/zc10IhHDR1tzGmnvXZNVmHKTxcnHeCyqzTcCcNhn8jzdnOe9xRQ0plPNuGwA7GT7w+Afov9RvKcLsabO597Z7M2ozmXG3qx/f2i0Fxyk4cXn1RqYnDm4JjW9Dy+spffLbbLjqKZ67Vh0zyCLaIXACot8V8PzzEAw6jp3Oz3aqgM1hJTNp4HLbMZ+mTDEGzab6SftoE83XPUNfTNeWqTp6ElB3Cbo31eEbcMGFCxnVmUz71dbi/9r1HDrba3nktdXX8kL7C6o8+r70eCCRQJviIhJM8MRvwIUJ1iJfOwiX6ifRH7tg21b+N6HlzSq67mskcJt9MGOuPAFmznvo2RDrDqwzePMBqqSLl37WjG9OizUItJB5d6XXu1HkvI9KcR047+mxv/D9C9ny6y0FxTtUOO8VVFAelFPnPQwsBP4F+CrwFWB1+ksp5RkhxG7gk8CoGe/AKpRs5e9JKc8DPxZCTAa+JoSIpD6zwz9KKc/o/v65EOIisF4IcYuU8rVRLPNlw5ihvYwUDrQQVqzIG6Rakl59AYoPhqPjdQsJ74Xu5w8SuStJl89lNWjqyVlWX72Pnod7SlObAVvpPR/Q/eYiIseeousGSSCdhGdoUHHebbS2M3KDa24mqCUJP+8yGE6RZ0O2hjuA2+XOtGnwpmD2i68F7OkATzyh/t3WpsqTau9gi1XiEAk3186gemItJ978H2a+I9mwA4KndPfUc9XTfZiKkYjcl9twz3iEPR5VlmDQceys/OFK9XnKuIm51b3T+vcWpHnLaQ5+qp8iz4YYPGDyqKc2J/F0fMKSG+l49DhaTVwnjQnhd2bi2/Zj5vdEOLz/JWNwofDwmQ9+Ro2Xr4aMp1apDYBv0EvHzycAQ7aBuzf2w5HqIVsqjTcOLa9Bx0+SaDffSeQ2WLx1sUVdxKKHXt+kxl9jr2UDG31lj8VwBxgWSSJ/0WKdq+m4jFg8mxvgJghM/SDhv8qhp18EyqKcUqCmfTHXjqaii77P9ry+h6RMGtRmFm9dnDfe4apwGFVQwThCMZ73I8DbUsrfTP39v4H/pZeFFEKsBRZJKW8YjcKmnrEbOKXP9CqEuBl4DVgopfxhEff6CNAFBKSUOVMdjhfP+1WDPB6pXC+zohUZCvB+jStVBadTC5vsj4XUy9ErDkzyTmLgrwesX9ip4aSR8gTjdmdkIrXra5SXPJ6VOLTIHtp5u/0pz7tNjETgj7CVhgTdiUT6Xqm2carrJO8kLth4pAMnIPqv9s+wa+9cbZm55uwktj1xwZpZNeGmO/yKqnauPnPq/xQsWVtTbXHzOei2W7mTMGU4m2XX/4ibgYmukc2DlCzko/5h282C7VwtYlyXgrE6x690ucashHAFFVzFKJvaDHAr8Is815wHphRxz1LQCBzRfyClfB0YTH1XDD4GJIGj5SlaBWVD2iO1dCk0NCgawsKFQPZlZsfBhMIUGfS85uWP38/y3zpHYJmSE9SqrYoP40lVQbtrNqEFgsAf6bjZDjzaQuoVnBG0/C6NmVNm2n9h5nrrEY+j1UlC98dVGe+LwYULdL+5iPZZSwnEGmg/PoVFfdczeOmCvSoLqE1AMqnqtWAB9PcbNgt2MQmuJDT0p1RMUsZoqBUCH+0htHU5s6fNto4d4WXmYJWFZ+2VgsAbTqLp9u2di1+cxq+uGeTOVXDenFnVK4nsjVh5z7OW0n1qIb77F6tN0+zZmXbXJqfqpxsL6ay2BkWXtdByQuCVpleChObT2U1OOpvqSOeB9s3V9IthZbeb/Ece4bHnTBfJDy+sINnYgMg/LFC0KKe66eMIQiHH+A2n+xf8GxOu9NpzVcVSVVDBVYJiPO+ngZ1SyhWpv+0879uAoJRy1NwBQogY8OdSyn82fX4CeEpK+dcF3ud64BDwrJRyeb7rK573KwAHj3ho3UJLUhi9Jyifp8r8fcZ4ECZd6sasR2809JVt6zvCjKxan2bkvOt13vf2WO5XSL20Po0PPNqkuNB6zXXgl5/6T4IfXpi7Pk88ofjb6Y/Nnt90GZ9rwnf0zUx/B1bCvhnW2wZOQPRJr/Lap733do938DB3rwVfv1KgMZejZmIduFxZb7/wUjsUp3OLoPUzSft72RH1RJZHb+kfB010AxyUYCzjzSGJFIDmvmDUTM/F0/d40G6YZDz9SNW9+ztufO+qE5LAF5Lsu94apV3MPND6NPyR2xhwJ9TmRDf/PEmoq5lC52c7rRl7zVz+kfLDTW3nON5uDBD9+Lbin10mPvtlWXvy4KqJpaqggnGCcnre9wIPCiGudXjQrcAC4L+LKWCJsNtxCIfPrRcKMQHYCgwA/0+O674ohNgvhNh/5swZp8sqGC04aCBHD3XmlCfLp8hg9mSRlg9E591tEQaP3qirKugUUgpRZ3FCZG/EWefdxmAopF6+eh8vnn6Ih3ph4iUVWDnrDPzyCTfBjT92rk96IzJzpvKSp8s4z8hFz7T5+94y9HfwhNVz7pUuAvEGaGzMabiDCpDsPvMQ7T1VSrP+AHRv8OAT9fD+99uWY/DSBRY1LsqOncFGur/jJvh60uqtXocKLLWD26149OYymcbmMv8yls1ZxkSPSfTKZjWzHW92c2RwEBYtIrLsdvt2nme8BdXVsGoVvr09dH+pJ1v3ue10f/55fEtXqbnQ3k7wnrYRz4PI3ojy3uuUf4SEhgFYNdycCQ62nKxNRhm+7e2Z8owosNPUdrbjLV23IvXY7e5f0G9sMBYUXTJSoOncARXDvYIKriiKCVj9e5Rx/v8JIf4CmAyQkoa8C6X8IoHRPss7C1xj83k9cC7fj4UQAngK+AAwT0p51ulaKeV3gO+A8ryXVNoKSoeDBnLwJByc6c2pZ58rcHfP63tyej1jHujyuQwSdaOhr2xArhd9EVlkbXWX0zrvNii0Xr7nD/N9i/MvYa9HbfY4pjnuHg/E40rn3BREGvNA19QhQ3+H98LmOTBQpXTRM2X75/1w/2Jbw12brIzT6E0QPCMIA+E3Z7L6mjfY6u9n61wPrbPuZ836/7HV6Y+JJNsOb2PFnSvYtmSboqK8mwr4PG8TnOp2oM3E445a3XZjc1vvNuuF+myyTuPNSSf84EH2/OZL9u2s9y57vfCFL2TGmA+s86Yje3oQ7tPY/MqO0ueBphH9xVZiJg6SdMGtfdBx6UOEfrE2d8B5ubIqm9rOcbzNC8PXFhelx253/4J+Y4O8c7QMJ3YVVFDB+EIxGVb3AUuBGcAPgD9FvV5eA74HTAeWSilfHIVy6nEEE7ddCOEDJmHiwjvgn4AHgQellIVcP/ZRLK+yDDzMywIHjmt4cmvJHEytT+PI27m73StdBO5pM7wARz37bLEveoc+LNZLV3C9dH2R4VGvhNA9QxYNfe2bqwl9TBdDUBNXhntTEwQCBKc2Z/TCM2UUXgLV7zP0t++88pS3Dzdby2YzNrSpbvyPuFj/EUV/WN98ieaGbXzgnl6evOUcpycmOO2+yJMvb+N99/Zwy3kbnX4JF2IXWLd/nfL23jXbOgbTbbBAEPjzawi1TbVqvuu52LnmW+q7me8kHM8NJ3kn5ewXbaonw2tf/iAs/13BnfccpWdq0uq9T5Dl6ReQL8FYaQ3fVyN07/TRfqHRmk22gN/j9xM8eMbS7p44DLkhMOFJtu5/sryJf5za3zSGih1vefn2dr9xueDVV4tad3PO0VJP7MbLO6CCCiqwRcGc98wPhJgGLAM+AkxFBal2ARullKdz/bYcEEL8FfDnwC1Syv7UZ18B/ha4PodUZPq3XweWSCn/o5jnjlnOe7G8yjGgK1wwcpQ1LW9YLAcz9GyIdfvXWVKBp3HFFCbs1FmcdNHztIuTTrmFQ1ykOoidxrZXeKitqjPGE+j4zHYxBI4xCQs78c1z0KG30cfW5jUTmXOe6A2S4BuC/gmCLc1JQ/ZSkUQlf7JxkE9IQHUMLnhTyZtMPHOP8LCq8bN0rNqhgmFTnn4Ll154qB1KGLjh1NTAokXwwgtw9Kg6eUgp69hp60enx/joynShs2Vw4+bhjzzseIqk9UbxP3VXloai449bePMSqpKCl04uxvej3eqz1lZYsyb/3C/HuuGgv++JQ8IN7qTqB1dSJdAynzyUpG6Sq9wwunrsubJEl2vdLWbdGEldKqiggsuKsnHehRC3AEgp35ZS/l8p5R9IKX9HSvmQlDKSNtyFEA+MvNg5sQ4YBp4RQtwnhPgi8DXgW3rDPZVJ9bu6v5eiqD9PASeFEB/V/Td9lMs8eiiWV1kmHuZlQVpxxobjWioHM3oyamu4X1t97eh41G1gm8HTIWOqrVc0Rx/aeekcOcTFZJ1N9UVkZZOiFaR51DJuUL4w85nNMQTpwDdfvY/GaY34r9N5b5uCBXOa0wb0+rmCfTfB+g9JNpkMd1B0DKcktpfcsEjOoqm/yjZANC7jPPHyNkLrFqJ96bPK4GluJvLgtUoqUd8G1R4iK9XJAkuXqi+2bIFDh2B4OEvx0c+3SCSjkBM8Bf+5RRmuaQPcHYfJF5OEb2tz7JbIK5sYqHYb+OOZepjrLaAxdg2+H/xUZYI9fVqVsRBPrd2Y6+9He+h+QiuuI/DV6whtXZ57TKVOl8xqN01vZw13gKQrXdxsBVzCRdsc53YA7L3Juda7HOuLBcVca/ebhgYVxJxGudbdUqg54+kdUEEFFdiiGM77j4UQv5nLuy6E+CQqC2v1iEvmACnlWSHEbwOPAj9E8dz/CWXA6+EB9IzP30n9f3nqPz1WABvLW9LLhGIX7zLxMC3QNFi9GjpTpOBCPXr5YJOIKGcx8iQzCc4IWpIBeV1ePvPBz4y6ZnH0RJRl25dx9J2sMunBNw+yuWezMl4LTexSYB/KlBW4dl8eDnGh8PmI3j6R2Cnjx7FkjH/79b8RPRnltXOvERNGPnPMA103CbTBN5VX3iuJiaTyWCfchLdIfH6y9bU7aTBxeiM9qcDc1LNibuVldyWzxh/k9rwD9HrOMvF8wv77NIXm6GY2THcza9Us5t98N7tf303srXeMdZQx9nhOgZwJv/qVNTuq4eJUX128aLhm4Uvw6rdTyYdmqARS4agb38AmA+880yx9Gltf3JpJHJUPIgl3XpoCA1rxsRU2Yy7aEOfu3znKsBsQ8MKvn2Tzse10f6nHfvOrS7yWiR9wuQh8IZkx3LOFNf6ZkAke2PQAixoXcfjMYevcNnuTDx6EzZvh5ptzz5Vi1pci1yLDb6JRtVlyKkepcEpml4vOU+g7oMKlr6CCMYtijPfrgF1CiN+SUlryuQshHkQpuLxaprI5Qkp5GLg3zzW3mv5ejtVoH/8odvEuZbHPB02D5mbo0w2LJ5+E7duhxypPaPv7MrwkzHQMg2GcesmPeuCpA6Inotz13bsyBnUaxQbjaX0akXsvEr1TqWOE96Zk/1J9aNcGSZkkIY0k41I5xMEZQV544wXL6cU7Q+/wztA7uGwO87xSEHjLQyT2HwzMkMTSyj4yzkAyTqT2EB3re5WxZfZmOhhl0TU3W3jR0qWMd28iK+dYE4e4gAsTsBiE3gQEDp6BpOSFaRgNSJ0nPp6MEyfOobcO0XumF5dw4cZNgoTh+kOed1h+4zus+Rn4ctnT6fn23HOWr6xBsfaBr+l+7rtoWYqN0NVDumB79Wto1Ql8+qYzG25289G0bmiT4e4VZAx3SGWIHT7vvCkMh1Ufm2Qtg2eGOHj9JUNwrQthmCnxZJy+4T6e7n6aJEnr3HbyJieT6jnlXO9KwWisu2DfpvniGAopi9NmqEKtqaCCMYFipCIXArOAnUIIg2ddCPEQyuP+CvBbZStdBflRDN2ilOsLQfr434zz55U3PldgVJkkEqGwZCa+eh/dCztVwN3ZSbRfaFRc6/OUNYDLTI1Ztn2ZxXBPo1BDOpOcqqZXBWTOVdQRbaon04d2bSClNFAQoHSpufC8MG6X2U2aRRLlCU8/z+vyUisnEN6dIHqDtFU/2XOzStQUeOgsoX+6H+0Ln4brrlP/3X+/omdUq+DXwLIYoY+d45YT/dY6xaGt2yjnuOsp8EglRQhkKCkeqbj44T2S8F6oS0lgZq5x8NTHkjGSMmntS6E8/E/5U31iDmDNFFI331wFLL95kmul2zsDXbEEwlKPC54EC5YakzYBKpgYnOdjW5th3YjMg2EPlvvHhDSO5RSVRZvvJ/TYAgKrbyD0Z41o8/2KTtLTQ/jvd1MrskmwvMKLEC7b+ZKur2VuO3mTXa68650tja3cyLPullyGUug8hbwDKtSaCioY0ygqYDVFi3kGpeX+SSllXAjxB8DTwP8A90op3xqVkl5hjNmAVch6yvLRLczX79mjPFNCwPz5pR+L5krH7narF6hTYFQpAVdOxSgkmUmOpDYMDpYlgMsuIDOXNGWhwXihZ0OsO7COeDLr1vUmoX24mY4v7wSfz7EN3MKNS4wwpX0Kd667k+63unNe0zCpgVvrb1XBxGt+iu+XvYRa1YZDb8B7E1mqi1MiIXOAqCcBcTOXXUL9MPQ8rvTdSSYhkSDUCuvmYqFlNJ+fyM5/vWh4RpquMuSFXrMn3oRqTzVD8SHb77xxtXHIeNA9HjUPGhuhpSU7z0IhWLfOSq8RAqTMOQ6d+vnaC/DJVyfQe3cTrw6+welBG5ZjanNiaOsbZsGPf6zWBaf5GA6rzfimTQRWJNh3k03lJTwSeESN5QKDnMGaBKh/uJ8tv96Sc96Abm7nWkfCYcf1MV9Ct7LCYZ2+rGXIU5YMnNb0hga45ZbST0grVJwKKigI5UzShJTyRyh++P3AFiHECmAT0Av81tVquI956HmVHR35F0OfTy2ar78OR46ooLoReLwJBp29iFLm9t6UkYNfkEyiQ+Ad58+Xzctk5/3OhUJpO3te32Mw3EHpUe+51ZXpc6c2aJvTVjaZy5abWyzPMD9vyewl2WDiC4qdF96rjMW0TKA3Di4JCRc5EwmZkynZGtUC7memUrRpb1eeYq+XPTfbX++qrlVGfgq+89Cxy0V0RwM7vcuoq67HK1Nj2iy3KF28b+r7HNtAcfxRVLJAAFatgpdeUtQD/fwMh6GuzugBra+Hz38+rxfVtp/j8Jlfw8btkuiRFpZ8YIm1jLpTBUNbHz2q5v/u3c7z0edT5XW5CJ7UnVTo7l3lnpAdy6m5FgnGjcmiTEHOYE0CtObeNQY5WPMpC5jmdi5vco71sZDTurLBoRyXtQx5ypKBncwlwJkzpZ+QlvGUtYIK3usoyngHkFJuBkLAQ8C/Ar9GGe6VFKSXGyPR6i3nsWjaCDHD5VIeUD3Mhnkp+slOxZgXzq//brdZSCbVJiNXOYuAXaIkJzzU9FDBhnRSJvN+7tQGa+5ZU7YMieZn6GHb5imVDbPKSPsBlanVbFybEwnZJVOyo7XsnjqQNUbWrIHaWhW8anO4mGy41mrs1dfD/v34vr1RZRp9bRr+N6AqnjVUvXGojQk2fGoDtRNq7Tn+cQicqYKdO3NvqO0oDz09sHFj3o14pg/SVJM41MbUBik9dm37yUxz0bf1wEDW42+okG4+puaPHdWoyj2B51bszo6t1LW2ybDyUMXMqkmfn/N56qvqnee2z4e2t5PQnzUS+NNJipqztzOvI8M2qdlINOVLwFgogwXmzVDaOZNeJ0t5X1SoOBVUUDY4BqwKIebn+N2vgR8BHwP+Bvig0MlgSSl3l6uAFThgpAFF5VSd8fmU0aFXm5k/X3129KjxWrNhXkrAlVMxUi98R/13TVMKH2a4XOqlZDbgh4bUb4o81rVVtEnCg5du5+gNHo4PnGDmlJls+NQGgjdZVUScoJ9jhuKL1ItV0/BFInR3+4h8TNLlcxG4raU4TXebY21t4BSRJ1cSvXic4MSZhJdtMLRz07QmENB7ptdec3/+fDh8GOJxS0BmqBV6G0xUmrhSWkkjeBIO3mBjwFsaSPfvlGEs1v8GiHcsl7omTFRzxYE64Kv30VG7JKNLnlGAeUMQvr0N6m5k4fsX8qOXfsS7Q+8CSt3Hm4RaVxXhv3/OMm5slZBKUTBBN9b/YQFd7/YQsAleNs+HofgQvWcOE9cFL3sTurbWc8Sd5mMq2NF3Pkb32hTV6CYITG0m/Fc7jf2eujZ4Mmbpv4zXPAeNwpyJ1kyt0Y8zrU/Dv6OVgUkDxKpjHHQdYfOOVrpvzL0xdlKfyhkPUmbqR0llGE2k63fzzcqx4XLBG2+MXC1ntJTOKqjgPQhHzrsQQqc47Px7u2uklPles+MOY47zPlKuuN3vhVBH9hs3qr9LfUmlNxa6xDaA4v7W1dkrihTD2S8FTmXSc97N8n4eD0yahLb4fiLJ3URnQHBOK+EH1uQ0CDIc1uF+YjKe8Yp2b/Dgk3XQ2Ym2ZS2Rsz8iOmWQ4NkawlM+ie8ruaU17RJMefCwaqiJjv9yKQqUXTIg3T1zSmnaxANoDRPxf66fAa+Okx4TdH/+eXy1NxY2PnIkq7EkPEq3lZ7zfq0X/x/FMtc4yT8u8y9j46KNhnpuPbyV0xeMRocnCat0cQLOHWmfzEbb26kMxRTVwSM8uF1uGqc10nKz/WbJwmtOyWR2/2wWPn/+eBOtT2P1f6+m8+VOEND6vlbW3LNGtVGBCXeiB3Zw1w8fVAt2auUWwPMbIHiKgjjiRSX4ceK8S0Ft1WS6F+0qPClXHoSeDbH+wHqLAZwvlqRovrndOlJVpZSDgoVvxEdUhtGEU/8uXKhyAowkNqmM8U0VVHC1Ix/nPZfx/jXyG++2kFL+TSm/G8sYc8a7U0BRQGWxzAs7eUdQ1IGeHvXvUrPw2S3SoJ63M4/BNFrIVyaABQuydU/BYlwmoLa63lnLOv27Ps3eK+rxoF3jwv+FS8Z7XoLuLfWwaxeRV+wzoVoNQC+1Q/FsZljOccYAACAASURBVE8zTC/GvEaCTRuFPgHrP2z1jLdfmEXHxtPFZfaNRGDrVsWb1a07+mDRwEldWwE89BB85Ston36AyJzzdN0gaXoHnmmCCxMESSQuXNRV1dHzcE82y6uunoYmMW+k8o1nm41lqCdStKFoa1ymA1t/krvttD6N5rXN9A0b52p9Vb2q83nyb341jdAf38G65mEDTckTh1UFlCFfm+TMMrp6NdqOp4l8NJk9vTg0Gd+9i0ZuEKbgFMA7yTuJFXeuyHn6lMujb4FTkHFVlYprKHFtK6oMowknA3vpUtixY2QbrUpm1woqKBj5jHdH2oyU8mujUqIKyoOR6gb7fEqGb9s24+eDg1kOohM/Md+L1e54FKC6+sot0oWUaeJEy9fmYMmYGwaG+3MmOMp4ty8eN2qxA8TjRD5ic08vrJ57nh3fu5sBT5JYMsb+U/tZu38tbXPalJfVTAs6NkR4Q6+94Q6WI+nIrtUMDPVlExuZNebtEvHcaMNX9kDXxWMwIAofHzmS1Vi1zVPweuH66yEYxPf9XXSsXAk/Pw4zZ7Lmj79OZPDHtsaOOQAQVGDstAuw5MV0f8TBW8B4tqG1RP+reI6yLa85zTfP03aRvRHOD5+3fN4/3E9k12o69tTlP/2IRIhOH7bEF8Q90DVrEty+wuphdzpVyUX1sftdXR2+ATcdnenYDAneQUWxK2Owupl6AirJ1voD6y35HvQw03NyIhq1T741PFzY2uiAosowmnCitvT25qSZFYR0jMdon7JWUMF7AMUkaapgLKFYrrj5pdrWBj/4gfW69MtTrxJj/i4fRishiRMKofcUUiaba3bfbGO8iqSjoWbw+k6JcXAubJ5jpIHYBvB54N+bJBfFMGnpbokkIRM83f00O47uyBgfmZd8IABOhru5fppG9GebiF1vyn6ajPFEdB18fytt1bVsWiCI3iAJpjzgdnxzbxwCb3kgZoofKGR82PWDE9L30zSVsTc91o8cwfd7y+no7oZPWF/8doZyUsCt50ybhDIaivk4yra/iZv45lu3WsewphH9xVak117zvOtnm+C7rvxxL9EowWnwwvXGAGEhoenuh2CRznC0i6d5+mlYtEjFLjjNMac4HJ/Pfi2BsiVQMidfMzwqGWNgqE9tdJZsLPreBgSDzrK4VwN3O9c6WWJ8hgHluIcJ+bJqV1DB1Yii1WYqGCMoJjmHnUTX3XdblWBA8d4DgZGpwIxGIignFCo/VkiZwmHFf08FhmqT4eh0LOQxjxSOhppF9k0vx+f1QlUVwTdERi4xAwkXvdiqqCRJ2kvHOcm5petXU6O4uYEALFhAUEvaPveCK84632nuuuc46z4k2XdTNgFUW7dV3rE2Lghf88nSxoddPwiRaXPb+xWpUmEroyhdBN4wp1gdRVWjfL/Rq8Ok8fbbxjEcjYLfT/DgGYTNVHVJCGjJwtolGKTtRbfSx9eNZwk80/uMMSmQXXv39cFTT+WeY0795KRg09patnVCr04zyTvJ8n1MpDY6I5UlDIcVRcYMj+fyZ20dDVzOtbsMyCSuO7Cefaf2sf7Aevzr/KOTaKuCCsYQijLehRC3CSE2CCGOCyGGhBAJm/9yJQavoJwoVN/d7qU6PGxVVwGlLBAOj2wRLyXrXxEwZCN8bAGa6M9vwJRQpsg8SJgDIyW43V5HQ82RHjFrknrmc88RPjTZYBBnDDOHrJ7gQMuw66OqKmVYPfigUst58kllcPX0EN4jDc/Va37H3erPuInKs8lvknc8PFEFq37tW6WND1M/aA8vJfTdxQT+YiqhT4BWj/V+RapU2BrXVXWED00ur6E4aymBWAPtx6fQfWphlhqV6zcfbicwzU/7r6sU7/68qnOoFQJ/mFRZT6tTY3jlShgYILxHMvkSlk1kXdxNeE+BEqfhMJs+5FGZZvXjTGDdGDpRzPLJBBab5XTNmrKuE+lTqRV3rshq9Kfgjac2OiOVJfT5VHCq3oBPB+KPUQO3KIzy2l1uXBGN/AoqGAMomDYjhJgFPA/UAS8CtwKvAUPATGAC0A2cK3spKxgZnF7GZggB73sfLF6svLqdnbBpU2n8xFE4HgVr0OXBKti80khLcTRg8pUpElGKMykjJTrDJsGPgMbpTY7Hso6UivtXQIru4tvbQ/c3VxM59SO6pgxybNIl3qkyu8SNsKVlOHFIAe64Ay5dMl6e0lmPzIMnfgMuTLDWTY80J9vASV/2+9CUUtUolb+a6odMX54YIDYxxsGgh81z3XT/vBHfnJbs/YqkYfnOQ/ephUTOd6oAybRCUBtl49v6zkPHKn0A3xZ4eoezklI0ii8YpCMcVuNgbhSOrUSb8DL+ZUOZGIj9N8Laj0Bbd4w10ZfxxWL4Yipz7Op7oPN9wAQPrR9YxJp/6cF3Po8Uq67No/c1It+2ZsaVSOPGsBBqk90cmz0bDhwwnuh5vdmssk5tX+Z1IjwvzOa9axlwGxWMwnsknCkDtSUYVMGpVxF320I9+fr4oJ6MSY38Ciq4DCiG8/43wCRUQqa9KSnJJ6SUfyuEaAC+DXwE+PgolLOCkcDuZZxO2Z5MHbt7PEpm8NgxFZBVrG78ZYLF0+LK0lIyBqaNAVMQL9IUjGbL93Z5abm5xbF8Zu6tLaXC58P37Y10pMp1R8cdqu0dkJOWYbchCYXUyYoN0oZ4/0R4cg6WUwX932a9dQC2b89q349wg2bpSxlnwCOI/EWLMXjPLr5DTwkyccTx+/ENDNCRiQXZgbb3YUKvbCI6J0qwtQy82FxUnnSbOHHAOzuhtRVN9LNgSZyzE8m0u3RBAnjaDzs+MEz3Og++d5U+/sb/JKv8sW6Hqr8eHk/O04TgbS0ceLuHJEYOjsBEAzO3ty7ZWloZKHoTBKcOEe7TVDtqmhobZipeMpktZzFjZQRa6r56H90DbUSOPUXXDTKrYDRUxribUXJOXAlYHCJvHswZ4DuWMOY08iuo4DKhGNrMbwE/lFLqWZoCQEp5GvgcMAj8Y9lKV0F5YEevqKtTx7/p49GmJmXMp43XMZr9zpGWclPqDxs6RMG8SJNOc3ivie9dALfZnBmy/cPtOV+Ckb0REkmr4T7BNYFl/mUF3cOCXFKhzc2qv2febvt1msLjFV5qEy4jJxuMakQjRMFeM/NR/tKl6vMtWwwcbK03SuixBQQeOkvovhjaZCAWUzrj/zaf9V2Pq/7vehz/481F82INdK2BrYreYii8yRvtZOCvXJnRPu+5Dvs4BxcMVLmIzHdb6Sag7mNWPWlqsvf8p7Iwh3f1U2fDB59cNdmyuTS09+c+B/X1aFM9+B9W8RD7ZsD6mt7sPIpE1NgwI5FQ/WTHkXdCoXEsOeD7yho6fnEN0Se9dHSmDPcxzN2+khjP1JNS4k8qqOBqQDHG+zXA/+j+jgG16T+klHHgp1Q872MPTjzGYDDLmZ840WoMjMHsd7bBiMJLYGqzI0ez4JeTKRgtTTNp/5VQ3OYCjeg09za6MkrHJzpyXh89GTUkXUqjaXoTGxdtLOgeFgSDygtrRlWV0rTfto3DF161Go0CptdcqzYMc9tVAiEzj7vYMaEzHgmFQNMyRvBrfa8hTIVw9Jrp4zvq6pShqDOKNdFP8/fm89jEHvbNgMc+As1fSnmKg3EGuJSVyBTJjPpIQVXo01i+fTm3ffs2Htv3mNoA3PI2/ofV/bXJKc76SgjdM5TdFOSQ3YsEVdKinHEOMk7XfY3WeXv4cGFSrCYj2Ld2Cz3rXDx0yyeo9lTjER5mXTuLXW27rONL394bN0JPD5GVTQxU6SROZTw7j3JR84p1BBQZoGyLy8jdNmzqng1d2WBJm/mWD+OZelKss6SCCq4WFEObeQuYqvv7TaDJdE0tUDPSQlUwCsh3zHu55R1LhCMt5U92wjfsF+yiPLzPPaeUeFK0E9+gh45f1MHa/aPy4nc69s1FzcmLNO3BLgukTyVjCmpJDk63SkAuue2TSk5P0+CFDdZ7FzMmbGgj2n8+rZJexQetCZQK9ZrZGIqrW+L06YKLpQv6qhRP/HCDjTSnG7oOdcKSPFVIndr0Xewz0E1iIsnABFj927Dj/Vnd/oOil83r/MqAcOKOx+O2cqFmeF1eAre1wJdN87bQuWpnBF+4wE9f+SlxT5K4jHP87HFaN7fmN3h8PqK3TyR2yvhxZh7l48kXs+krMkA5V5lHm9oypignTjStPJuW8U49GTMa+RVUcBlRjOe9G/iA7u/dwP1CiE8LIWqEEPcBfwAcLmcBK7hMGCcSYaV4Wmy99U4vp3Qw2iOPKGNo1arR8dilPGThb+ymNu7CKzyZchVz7Gvr9Ut7HVetUnV45BFVpzQtKBq1KM9YpAsjEXsefiIBu3cX5tWzMR4jc84zMNxvTKCEi4Ya3cnGeXJ7D20kMjvvwPYkofMOFbtglsj0xiHQez6vhzJ9amPmiYMyvjs/ONHZG22eUzo5TLsy6eHC5TwOHOaq9qU241jo3m0xgtUpxHBJFIngjGBmnKbhFal5ZC6TGanNRUFe6pHI1JaAkXjOxxTlpMQTiwr1pIIKxh+EtJMLtLtQiD8EOoAmKeXrKfWZfagg1jQk8Akp5f9b9pJeYcydO1fu37//ShdjdFFM2vMxDn2A6uzps9l+ZDuDsUGDt76s3rFiAuxMHjJtqofIfDdd9zUSuK2l4GBKs9fP6/JS66mh+81F+J7PkUwnFIJHH80EH3bNIBvU1xhQdQgEnJPRQGGpzW3uEfgj2HeTzaU3BoiujOZMoa5NVllioy/8iNlH3gEJB69X/PDD0yBh48luSFSz/+QC/NO3/f/svXuYXNV5p/uu6ipd+4IwksFiYyNjIymWyja4CqwRmDiORyajKLFHkyjCiBzLDYSO53hyyjOZaOIMmcRT49ymCdCWHXORlEQhOYRYdIiJMeqRcbWQ6VbbkvAFbG0EGGShlhrdqrrW/LHqumtfq3ZVV0nrfZ5+pK7al7XW3tX1rbV/3+8rrY4XJyrjQwLjuHTtS2Jrgr0v249DLBJjwdwFvPbmazXvVfUnnYaREXX8Amav8tGvTFatave8RTz76Wed7wPLZ9W8cyPxx9ZU3wu5COP3TFdV4E1sVnp1x/a6YB7MsOKh6zgxSz3ZEHnoPQcTn/w2xrJkdV8PHVITvVyuPLnYM1zbRrvPocs9EPbfI9vPUIC/DU73h5/xDB2nz2wi4Z4HQ/nvpV3FYo1G03qEEPuklNc6ve9bNiOl/Cvgryp+f14I8QHgM8CVgAlslVK6fONrfNOA20LdnCcOCnaPsufF5rHhPRs4ePRg+F9OQR9XW1bIjGM5BncJuNxGIuGC7arf6UnSP3qIwb3SuR2pFGzdinHibHXF0crVTT8yCKu7ihWbYyRfEYwtFiX9OViegjisHppf3EL80keZOj1Jdg7sjRd2FpRdcqwe5hLWXH0zxuY/ZXzV10mvPFHtPnLCxrfc0hc7SYE6raB7VjdrrlrDjokdzpKD4mdqYAAmJkpuLMV8ips3UJO0GpuG9W9f4yljqWxr+vGB2nuhK0b6hi51bxWC4OXHBfsWZ8lXmMb7lkjcfx/Msxnq+++Dv0hWt8lmISA94bxKXSV7cLJAbcLfP7eVcz9SjLaSnDQgfdTSE42ms2iowqqU8nkp5Z1SyjVSyk/rwD0kQnBbuJCx+0I+lT1Fz+ye+hJAPU8Y8HF1SJpeWy1/F4xe5lJMpxhULVmiLACLWGVSxWqzbni12UbekdrfS/fsHudH9A5jkz4xrOQ2xdX1yuJZlf8Wuh7JQ99ZuHt0vrLm3DPB4JW/RWZ/gsF9i3wn4lolBREJXXn45Jl3M752mLtvurtaciAjdJ/Jk3riZPXnNZOpsVE0TsCuv4YFZyzypXPUuvxUYCfzsL0XZLYq4dW8YwOPvnd2VeAOMC82z5dEIn1imFMx9aQD1L+nour1GmwKyAVKjPRbgK5BGk3WbCvJSYdIHzUaTeM0FLxrmkQYbgsXMC13TwgajIek6bXV8lu92SvbUTkpPHhQBe/FiqxWNw7DgHXrqnTaNXi12cbxw9gzwfidE845Cw5jk1lM1Wq9IwLmn4M796rCRsa3D5bbUgwG16/3Pf5V1VRf7eLOZwUv/jk88CcvYKxao1bQK97vf1Yqqcp9FntEu35FIhjRt1RXsN1XKDhWbLcFJ9vT5Zcst8/ruHJ1qd/pj/ZwavqMZbgE665e52sya5dkWyzk5YdAuSctotE2tZXbSYdVR9VoNPUTxG0GACHE+4EVwNsAu+wkKaW8u9GGXdCE5bYQFkElPF7b+31/925V7TQSKVdp9PFFVM+jbF9FnJzav3x5sMfVqRQ8/DCcOKH6J4Ra5Q6yQmaapJ44yfbuPFOxCFmRJyYF3VlZvWorBObRH5H+zyvJiJdJfvA4qRGJkUXpkYVQY1spFyn275FHStVma/Ba1bOO0SOPlK6dAc6P6O0KMnV3k1y5hrEDD3sG8LEc3PZcoWCX0zXYuBG2VjjpePTF6DMYHOmBr0QgW8wyLU+ojcFB5/e3bFHWlrt3q/s4Gq3SgfNLv4SxYweDw/7uHbunSsfPHOdrP/gaEREhKqLkZM52BdhuUiuRHDxaMVGovG7Ll6vXDqj8ieTSaxn76ePVDkXTqoKtH3wVMHMg0OczAMU2nTx7smTZGhERNq7c6PsYbSU5OU+kjxqNxp0gCauXATuAG4ovOWwqpZQeJmidR0sTVgcG1OqoNRjs72/9H+agyWNe2/t9v9LmEFTQ09PjayUpaBKa4/ZrhzHu3VY9yYDa9hflJUXvcT9jtGKF6mM+r4K6nh6YmFBJmV5BSsUYmnOzpFcLRi8XJF7tIvVUtkoSUkyMLCVrTitZxvh9lLerTGizXh8rQsDChWr12mkyFfCeqQnMrtyoxr1C72z2QvzeFZw8PUmui1p9OxCdhp5i3047nNPu/iraaFqKdFW18YY46e79ZBYrpxilma8YO6dkwa4udX2LVYy7umDp0vJkFAKNlVsCbTQSpUt0sfSSpay+ojbxeeDxAYb2DdVMavuv6VfBp9u1j0Yx5+eI30514u+cHsbv+p7vQLqexMhGk0q9yLyU4cYHbuTstLKHjUai9Mzq0X7hGo1mxggtYRW4F7gR+CdgJ8rnvba6jKZxHFYfZ0S76KcMfJDt/b5vLRiVy3knSBYoPsr2GyTYJq2dnSL9uzcyuCtfnYS6dm1t+0+dUlU/e3r8JdgVq1EWNdD5PJw6pZIy3/aYt2d0xRgaWRjcJSEiQOawyJlJryoHW6D+nYqp121Xp63Xp5Liffish+d9gHvG0Sf7D6v7bADjd05w84MfYeLY8zWB+1vmvoVff8cvFRxzDlZfg8rV5DNnaieG+Txs2+YYvJuTJvGbnmdKFrzcL4PtK2F8axTDLcFXCPXkovia05OOAMmZTgm0ALl8DhERrL5ite1KsOfKt9u1z+UwJtXEqORQ9Iog9c5fDRTg1rNK3WhSqRfb9m8jL8tPdHL5XKjH12g0mrAJErx/BHhSSvnLzWqMpoCX20IhGDHHd5P+oCRjREgGsBgMRFAJj9f29bzv57wWggQJjsl+C4Hiy8UAdHjYvv3Dw95BLahrt3Onc1LmJTZByhNblCyj+ARgZKR2/3y+ZP1YuTrsqlO2mxQ6jf/8+XDbbZh3biQ9kSbzzxmWX7IcBBx4/UD1U4IA94yjY85/vJbB7vLqfnHF9oWpl2yf+b1zwTtVcSlr0SWvJwkubatqY9c02WIecHEClMwxuLEgr7CZcJvd06Svy1tW623O5UfqUPjMp8Z3s/2mCFNdUbI2lXndcjtqJrV9y1SbPvJxdV/trvWFrznGCSociiQk7LX5dm33I7uzk8f4ymFpwJ2rkyuMajSaC5MgwftpVKEmTStw+kIvBCOmOEl8syqxnj0KYz/7XnMq+wW1H/Pavp73/Zy3AWw18nmVQFhFsU2xWG37XntNSWEmJrz93Y8fr365F9KrBV+94hjZfHVAls1nGX3yQcydkP4gZM7uJWlESJldGMfKVX7MiwTxflmu9FlYHV57SP2/Rqd89i3Q/+u1QY7N+Jt9kP7wNCMX/TWHHvkS01JV56yUbzz36nPl+8/pGp4+rcag4ny2gZPIMxp9TUnHtm+v8Qe34pbPYH5xC+kPHidzmayWu1QdwP2+yhzJ1ATK2SiMXk55xd4y4TavW0a876+Z4pzzar1fKiYgRjbL+PdUXYCd1/Vx9MwbVQWkvHI7SpNaO3vTSETJeuyKc9nh5/MYwEbV6SnM2qvXuuew1FlZtEhb2T1qNBqND4K4zTwJ6L9mM4lpws03wxtvqEqJlXKIysqOYRLUfsxre7/vRy3zyiZKh2zt3phNKmPThjVrVDvsXFgmJ1WCohNFWUJFnklRkz70fsmbkdqV1FgOlr0G8dth6FpVYGfofXnim6cxL46W2pX+UMxWHgPUVlI9B6lv5uxXJy3Xx+wtnPvnzjAe/Rlnp8+VEvsqqZQaOF7DgwdrLE9dHXMKTzvSD252DNzB2erQnDSJd29j6P2SvZer8YvfofpUPpn3fZVcnCRmyZON5QqTu8pV9ApHm/RHe5iK5mvlSjd0lc9VqLLrVeXVvi5Anmdf/iX65vTVZ1NoJ23K550TlIt0FTrk9/MYwDnLSR6DxN2OsUF3rraye9RoNBofBAne/xPwdiHEHwsh5jSrQRoHiqtLExOAgxyiGY96g9qPeW3v9/3bb1f9XbHC3srQhaDlzm3t3n7taQzZUzvJuPtu1Y45Dh+BYRvP6yI2chKrJr2SYiVQsNGtz4L05mWlMcz8wjJbeczBhdhbEb56yjm4WbsWFiyAOXNq2+fiHFm6/4rXcNmy6g0q8xYK1AROhT6XHHOyWTJnX3AM3CNEWLfU3uowvSfNVEzWBtCrhbqvfNrppVal6GZ29QQoi5rcOaw8O67W/8LS6sRsP7UcHGRIxrcPMn77OBtWbGDR/EUsmLOAte9e69gPP8dk9mznfbq61IQsyOcxgITKSb5y8OhBdzvGBt252sruUaPRaHwQpMLqy0KINcAe4E4hxA8A6wPowqbyw2E1UFOguLpUIHnERg7R4KNeRzu2oPZjXts3+r4LjgmQHl/Gthp5t7yDnh4lAwlCMgnPPVeVLJm53D5wn58V3PYdZfv48fU2E7UuGH3nXPiCcolJ7tzE2OsT1ZVLCyvY1Trl0hHgq19V/61M7Kx0sbkJvvo++/Y5UVlZ1LxIkF6Dq+a7Sof9rZ0kxl5XVpbFvyyxGMnZSxiLHLIN4PPkOfi6ve46cyRTYy2ZjcKoEYH7dtUGng66aaPPYPzXnib9uzcyuvAsiZdU4G7IHrWNzX6OUiwzX97ebyK4g9TMvG4ZW57aUkq4lEh2fHcHj33/Me/g00m+dtVVcOiQvWxterps2+r38xlAducmX3HNYWmgsmiRtrJ71Gg0Gg+CWEV+AHgCuMhjU20V2QwsVnQ1FoAiRvfs+u3Tmm3H1io87fDCYtMmePDB2tff8hb4dRstOajg7vrrq6QJA2tgKNlFlrLOOJaD/rEIg/8cgVxObXOtZaImYvRfW7b4M1etIL5hstrGL1ttCWmX0FplqZhOw9AQ5txs1b1VY8toY9MIMLtrNj8Y+AFGn0owjf/pu5iSZ6stKrdGMf7db5S9zyt9/DduVLIki8tSUfN+/MxxpMVOx+3a2t4L09B/5ScYvO3vqjf2Y29ZDLorJ3NQvV8korz1b/0V4ld9nansm2Rlrnw9thaCfsOA/furm9AL6V9dROaDb6+ePNu0zbx0HvE7YPLcySrNu9eYePZ3eLj6GthRaS3qRQDb0Lr/BgW1s3U6Rp0JrxqNRhM2XlaRQYL3/wMkgc8Bfwu8IqX0UfLw/GDGg3cb73ezF9JrL2b0OoNEg24zLQt6w8Lhy9bJBzvxtgSZzT4DDr/nX7GiXGipEqfgYWAA7r+/auXdvLiL+O2Sqa58ddC9NYrxZpdykbEE0zUTtcK9Yc7Nlm38jkDqGYFxXKpgb162yqM7Og1deVh6FFa/JEi9/m6M770Eb75pO1koBewSIrLgSBkpO1PO7prN05ueJnm5slsceHyAoWfvr5KOxHLQ/91ZDD41F958097Hf3hYJYFannaYkyZbvrGFbRPbkFKSJ+8c3FW4McU/dKiUNFoa2x19GHsmaq9NPbUV7PYDFcBf3kv6E5cyOvW8Wq0vJsvGYqqvx46Vm1ycjM8pFNyy9s0ycRhYfZKh53c4yokSl8TJHFrtHozaTUYqrTV37oTXX6++v+upN+F0HrtN6/CBD3oO230bDf41Go0mRMIM3t8E/l5K+cmwGtdJzHjw3uQvmJYFvWHgMhYDE+nWTUIqg5yjR8u+7WAf5DgU8jEvnUv6fafLQXcxyIvH1Yp0wb0kvQpGJw/WBjVOBYIWLYJ3vEMFe+9/laGf/EO1jKQQkFsLNyU+BXsvt+lvYfvoNMzPwjqu5uCSPtsgy/F+OnMJmT+ZdHYTqhgzOxkX4B7cWe4Ns0+59FSN7Wn/18Zzldlpv2J/FixQTkQeDNwsGHq/rJHBOd23bsWaYiJG/3ORco2Cev9WXChBbTsVxdNoNBrCLdL0M+CY51aa5uDl/d4gbWeX5vYY20UvnPrD+kuwe57XSoW7SE2AZpcw56DNNRZexeCTh0or5x9fD8mXIXXpuzAKwYMBOIYRTprf9etLwUdma6JG/12UvlgLN9nlU1RKZXJdcFLA8Nkfsn7xHbaro47306sRX37rbrkLjpOwCjemIsakg97f57Xx1E27WZu6WYtayLxzDtmu6hwKtwR0p2JNAkH3dITU7mnI5srtmJpSTkg9Pf6lIU3+m9M2NJjwqtFoNK0mSPD+18AvCyFmSynPNqtBGhcaSOT0wrP6Yivx8m12+bL1W121tKr74ghJM0/qWwJjyfvg0UdV9dMgftF+A7+KQj7m3CxbPiwYUeRvJAAAIABJREFUvioHF73EDd+VfP3tcCpW4dOefYTxgxmMZfaVP+2O61SRN7k4ydgrz9kW9gGVyLlzudLDL38d5mXhFA6adyAfgdfmTjP07P01CcHmpMnJcyfJyzwCgUQSIUJe5jl56QLMi39W5VFvN2a+q2oWJ1u7d8Pzz8NZ7z9N5sVR0qsmyfzeW5X2v3cNxsY71BgWqq+avZC+YZrMkm+w/NFNvHn2TXYf3g0C1ly1hjuuvYNt+7eRWTpC8uYIqW8WJErW/qxZA4895q4hB5Inehi7JOd78mz9vEaIIIRg44qN3P1nz2Ecq9bTk80qOVIkEuzeruNvjmPie7sSQsJrK+m48dVoNKETRDYzG6V17wV+D9gvpZxy3+v8YcZlMwWq/nD3LSf1xJsYX9ut3lyzRlkZ1rkyVrfeNEyKq6cFS8wqVqyAXbtKiZU1pegXLlSrzR6rg6VV3bMnVTKhRTZShcfj88xLGTb/w6288NPnWfIGbP1HSL7uIi8wTcwvbmFFzzYmo9PloLj4MawIkmM56M+uYPALlkDMtlPuml9rAqldEmpEqqA8Nq2C93UHldXk6YLlZM4hDb1S3mFOmsTvXaHGVuQREmRln0SU7lO52rEuat4LY+ZLxuWnemqRSETlD1wcJf7pHFOFSVLp2u/ow7jnIVi/HnP22eqEXQeiIkpO5lSfchHG78mpAF7KaokJlK/N6dPK775S7x+LYd6xgfjbHguUrOn4ebWTgUQiql2N6tc96MjE9w6SB3Xk+Go0msCEqXkvLpUV0tYckVLKICv6HUE7BO81f7jtgs6+Pvcqn+1M8Uu0QvZQw4IF7o4YPr54bZNzc8oDvVZigaPuOfNShuu/cn3ZAUWqD8czk58g+dk/dT3/vXvvrXEJsSNxfD6ZPwtnjmzeECfdvZ+RK+DQQpgWKiAX+UKAbZ04FMbD6mxk285CUD2wcxND333QUXJjPXaJ4sSsMGa+EqidkkWt9PXBunVw8CADN51maM53yYryn7BYDvqfEwwefg8cOsTAL2RrE3Y9iEVi9F+9gdQeSJ8YVo4+K9eQ+ujdrpr8yvvV7PXQ8/vFzp2me5r0dXmLbSfBnGN80HGJ70UaSXhtIR07vhqNJhBhat5HcA/aNU2mRkpg0SoDyv3Ezis6DJptp2bxsrdlako9/i9qca3Jom5+2QVsi8FEVVJjDS6Pzzf/0+Zq60KhPiCb3/08+13GJXMk4ytwj+UgMXuJ53Z+MeI3MDh0UCVyFmwjRxfDjy+C17qrt60cD+OEmiCmV8HOn4PX54GsKO9WKe/I7B8mW10wtUZyYzvWc+dW3Uu+ZFx28ikrlklBZmuC7MvVf8ayURi9TMI3X1BFoWwKoHmRzWcZOTbG9rcdZuqSgk7/+R1sf9Hiue6iIzcgnADMcg7zumXE+/665LozdhlsX1lwNApZGuJUaCn04nFh00RJYph07PhqNJpQ8V1hVUr5ISnlTX5+mtngCxlfQaeUqviOW7n1eghSEbJe/ARj1kSyN9+sdnkpbjMyosbgve+FlStVWwtjklycLFX0LFIsaFSDSwn4F4790P71N15w7UJycZKI3UdPQiRfbk93TpC6davrsQB1DQYGVCDodt1TKdWfWKxUuCnzZVj/PUrVQ4tYx8N4s4vBb8zm2b+KctHZ8vYxUR1UJ4/UHss65a8Z60LBocqquIB31ctkslwB10rR6WVXdTGm5OIkMVk99rEcJF4RsGQJxGIsf718HfwSFVHyMu+o06+iMtF5cLA5K7wV50h/tIepaL620uwNXY73dr3YfrZmMvH9PEOPr0ajgQDBu1+EEF5FnDR14jvofPPN8INrt4qQYeEWjBWJxWByEq68Ev7yL1Vf7bY5dEh5qo+PKxnR/v3q93ic1JUb6Z7VTUyoB0+lcvd7LMe5+mpYuxY+/vHaoNg0WfLKWdtnUUsWuK+Wp1al6JndU/N6T2w+t5y9msTx+fRnVzD+yWe8k1WDTKqKK7L9/cpGMqI+/qk9Sn5VDsij1eMRjUJvLzz9NMaG2xl/Jk7/uRUkLonTf211UJ3qXcO8LLXjUvi9ZqyLBYcufZShfUPsfXkvQ/uGiN8fB9RKdGZzhsGPDdZKSComI6V2zp6t+t/fbyudSq1K0T27p9zXYnv298LWrZiXzuPRZZC3KULlxrScJpvPNndV1O8kzULmSKYmUTkbhdFfWBr6xCG1KqU+W4W/UzOa+H4eosdXo9FAMM37kJSy32ObtwBfl1K+P4zGtRMdo3mvJMyEtHp9sINg1epGo6oke1eXSvCLxcqJfk73bSymgtLp6eqkwEpWrMD8262kJ+5j9LldJA6dIPWNc+UxjEZh/nz1/6LzjFVLPzBA5rH7uP62aRWXFjJBBPDMp75dKlbk2NVJky1PbWH4h8MglYPJ3T9/d2CNs/mZTaR/9BCZy2R11VSv6271Q784SvqGLkZ/Yakq+HXlRox7a4sleTfIZNN/uoqHlp2rktZE8nDJm7D+QCFwjwjSPz+bzLJezixawMHJH5HLVxR0khH6l9/C4PoHyu21k2xVapWXLVPbHjjgKusyJ03ST2xhdP+w8n6P3IAxPR8OHGDg35xgqPf7VZp4geDKi65k6twUCFgwZwE/PPZDpmVFVdxIjKWXLOXQ0UPVemQRo//UUga/MacxqVkDSZWt1km3ReL7eYweX43m/CfMhNU88D+llP/F4f1LgG8AS6WUs+ppbDvTDsE7VPzhfnGExJMHST11DmPSZYewgutWFTKxJo5t3Kg07iMj6r1jLqUG5s+H225TtoH7PRxa+vrUv9ZKn6BWb//dv4N//Efn/hYmM5m3wea18MLFsOQYbP3RMpL/cqC+vgfEnDSJp69kqmu62jnlPjCWJuCRR9xzFDIZ2LwZXnhBSUa2blXbNUjiL9/L3qPjta+/pGQ6pQTY2ZB1efaXeLWLzO++qH7xClzrDW4t+yU2w16b3IdKlxsnJ5z4W+McnjxcnlyLGN2nc4x/qQvjWK4xF5N6P38Fd6N49zamYtK+eivaflCj0WjaCa/gPYhs5n8DKSHE52xOsgj4JrAMuCVoIzX+MfoMJSU4tJrBXdI9cA/Tq9gqUbDxEg8Fqx64GHQePuweuMdiKnAfHIQbbvCW35w8qZJ77Vbn83k1AXAr3FKQ+CRfhv33w9Qfwf6vxEhe/eFg/W2A9J60CsisWubVQq1Cu8lpTFM59hw6pCYwhw6p30OQWSWvXG2vKy/Iu9KrCs41lZvY6eLNvJp8bNkCx4+7S7bqlXVZ9ku+ZKP/t2iKnXTHq69YXa3TP7W0HLgHaZMd9RQSKkxMjPt2MH7PNP3PShKvdtF/9YaawD1+f7xGtmROhpjPotFoNJrQCJKw+h+Bh4E/EkJsLr4uhLgMeBp4N/DrUsq/Db2Vmlq8kjuDBtdeetpKvXQi4agpbgpeLjRCVPe1ONGIupgp5fPO0pvKypiVVE6GUinMS+cxcLMg8SlV3t68dJ7v8TYnzaoEzZpAyYe+eeTwSE3V1GwURo3Cx9otmG1iDoOjrrygc89cbuPmUmFAW9p+RKonLtu21V4ra+Bab5VMy36pPcrjvpiwGiHCvNi8sqbYNEk9cZLuM/nSBKVSd1yaXG/OMPiNOeXA3dImz+tvxS4fxGtyXnGNjRMwuEuS+UqEwZGeqlV1p4JYW57aEqyNGo1Go2kJQf3YfxPoA+4VQkwC3wKeAq4A1kspHw25fRonnKoCLl2qbPcC6pRdK5oWmSk7NbeJSleXktZUFqeqtMpzktvYFa0pYlcZ0zIZKkk/zgqyQjK2WLB9FYz3gteIW3MXxl4dq65S6uN6mJMmh44eqm26FCRu2ghfPOAezDaxJLzRZzB+54TSlT+3i8T3p0jtm4Nx2Vvhyjksv/gMe3m+eqdCkagVr8LqwxXafadJlhDVgWu9VTLt9qP8IKDKCrS4kj01xfjcadKrBaNGF4mbNtT6ubu0ybxumfv1t8NHFd0afF5jJ/vBbfu3ERER/23UaDQaTUsI5DYjpcwD/wG10v4QKng3gE/owL05OK7QOclYdu1yt6CrXNHdtEn9XHutcnAJcxXWa+U4iHOGkwvNihXw4ovwwAO1fS1ONMbG1M+CBdVj1dOjHFQKq/NmLwysgcRmGLg5gvm5O1yfNKT3pJnKnSqtfGdFnuPnTrDla5/17Fd6T1pVIK1c6TxbYSnoY1U8vSfNdN7qyQiR6CxSH71bjZn1yUM0Wg5m7cZUCPjxjz2vh59VY6PPYHD9A2T++HUG//40xo/fUNKcsTFIXld7UAF5YPVLQtUsiEYZuDlC4qYfMvDRPGavtaORcuBqmkoGlc+rPoD/J0+pFMybV/o1vQpOxco+9hLJqVOTpAfepyr/njxJZmGWmzfAV+OS02Kaj3z7ddJ70t6f0WgUIhHSb3yNqdOT3paSVQNqqOJkS5eq3I6lS2F4WN23TtfC52q9nQwoQoS8zPtqY+CnCF7U6arTMefTaDSaBvGdsFq1kxDzgSeBOPBxKaVdXcrziplIWPUshR20KmCQcvJQf7KrV/Jg0ORC01SB+okTahVWCBV4B6kkazdWAOk05rP/Svznn1de2JGCVeLsnupVRovbSWLpiG1SZlceXrwn6pqg6JjQeUmczG+NeTv7mCaJoWvZG3utZpP4W+OM3T6mtrv++upVayHgmWdUUOd2L7hcjzDKszslfAIksot4ZOQy4h86pBJxZY5IXqlqNo7D3U+BMRWBW25RkzZrPyIR1U/r0xg3Nm2CBx9U5/8U7L3cpl2FZNvM2+D6zVQ5DAF0RbqYltPOn9GRETV5mZ4msSlnf46KpNga7KqmXjpPPf3JnXL+++Djc2Z3TfMyX+Wm49TGMO4Hr37WneTbjufTaDQaH9SdsCqEeMHpB5gA3oFaLPtLy/s/Cr0XFyhOWtTS6lfQYi/WFV03Gkl29Vo5rldvXVxVFQFNuMF+rAqvpbd8mKk5XaUEyqzMVY+zjZd68slDCGvpUFQsl066Jygmzbx9USSzILR2WzEttCU59rptYuXqK1arX7ZtU5KiSqJR9XpxPGw8393aDT7uSR84FakSCBIfXE/6c6vVRKrgTZ6PwHQEHoorqZK5uEcF5lB7L+Xzqi89Pf6DrwNldyC7IlOVybab11YE7hX/FgNdx8/o6tWqbbmc/Tm8Cu3YfGbSK0/UPsGxnttHnorRZ9QUxNq4YqOvYkBh3A9e/Qy9nsRMnk+j0WhCwE02E0F9NTn9nAFet3k99MJPFyqhl8L2U8EUGneS8dLaBtVbp9PKb71YSTWfV7+H9AXrOc42X/Cp3dNEbB5a5SOWirc2/Up9S1QXRSomaD5T+Oi4OfsU2pIakdXHkCqx8uTZk0q+MLUTc559smSJYmD59rfbV6m1uR6B7kkHOYJTkare2b2kVqVszwFKyjI1J0L6i+vKAWgY2v0KiVFNwSpLsu0LF0PNnM3yu+14VLSz5hx+Cu3Y9DNzmaxNWLae2+cEvyrR9mOD3P3zd/sqBuR1PwSW1DQxF6MtzqfRaDQh4BhoSynfIaW8sp6fVnbgfCb0UtheFUyFUKuwjTrJeGltgzpnOH3B3n8/XHKJ0v/G4ypAzGQC61dtx1nESPzotDrOzp015zeO5dj40sU1K8giDz++SOnnzV77fhnxGxjfGqX/WSXH6N8H41ujGCsLq+ZuK6aFsTBOKD/34jE2vLQAgB3f3aHs/t5+VK1SV2rFncY4wPXwfU8Wn1bcf796WnHPPfCud0Emg9FnMHHHBLfGb2XR/EUsmreIW1feysQdExgnIPmjM8Qsc4kiWZFndPKgd9uXLfN/H6RSaqU+Gq0Z1/591UXQlryBY/VY1/GoaGfpHN8RJLKL6L+m31tmYtPP5Cui1pKzkb8PFditxtu10e1+qMuCsh5XnUZw+pt4+rTWvms0mralLs37hUhbat4DH9BD5xyJwNVXK5/0eitB2p2nUc27XYEaO7q61ApyZUVWH/pV82CG+N/cyBRny5r309Nlj+5IpHZlOhbDvGMD8bc9ViUbQAKiomDSjj6MPRZtfiM6W4diPQOfXcrQfEt1z2kVIA7ukuoc8+bBunW1FUgDtMf3PTkwoAJ3uwJYP/iBc25DPI4pThLfnGNytnqSUTXs1sqgdm0vJqA6Vce1w1qp9dFHawt4RaNkrprL9b92shyvFzXveZguFsqa28f4nRPV49GotroezXsLcLsf0nvSwSu7zpTm/eTJmmtNT4/Wvms0mhkhtAqrloPOB94D9AIngO9KKd+su5UdwExVWA29FLZdOfmxsVIyXZCg1/d57JJpgyTbBk20raRonznHoTx9RcCYTuYYvRwSr0RIfUtgHKtN2DN7VRGkjBEhedNGNn7wDrbt38bO7+3k6Kmj5CkH+TEZoX/5LQyufyD4+Fi3K1ZJ3bhR2VhagpvE3VfYJ8FmF5H5+jsw33MF6WOPkVl4juRLkMpEMWRP9YTKqz2Fbczx3aQ/KBk1IiSuXG1/Tzol3QLmb99K+qM9tdU8KyYmZi9suQm2rVQBvBRqPLtn99gHxpVtP3kSduwIXo3UbtxHRso6+tWrIZUiI15m8z9t5oVXD7Lk9Rx/+CR8/Soll0q8Iki985Nw+x2kH9xM5swLJOcsIXXrVozut6mCU8OF/P41a9yTau2u/bZtVdfI7CXcvw914PQ3yikx2TUxF4In4jfcAVM5CU1MVL/ejArSGo1G44NQg3chxNuAPwN+BajMhpsG/gH4rJTy5Trb2tbMVPDeEuotvd5Kil/oX/2qWhGtB7tJid9VfSq83ecob/fKVcaP7/x4fYGK9RyVZer7lpP6nUcxXrWsIA8P1wRxA89sYei7D1YVP4pNw4arPwHz5rNt/CGklOQjFU8FtkYxNtzu7xrX87Tknnvsx/CuLqbmRGpXiz/y8ZqA3+xV9o2lwHh/b+2TDCtebj1h4XAe87plyi2nUP02Ng3dWcH4Rx/F+NVN/sbQz9OrysC+2QFuHQw8PhB85X2maNU9o9FoND6o223G5kCXAd8G/j3wY+BBIF3490VgPfDtwnaaTqJJSVuh+j8XE+9uu60+txmwd5Lwm8SLCiKnZkFWqAlvpbNGGPkJNRrhAw8T3zCJOdfihLFtW00Sol2i5bwsPPr9x3ho/0NMC1mSoGS7YCpWcMVxu8aVCacFj3NXV47K7U+exHxLTHnnf6qcA5BeLVRQa+dOYqM/Nk7A4LCyaRzcJdVExitRuVW6aYfzpK/6aSlwh8J4RyXpr9zm39nEzQXFxv2IeLx9NNqF+yD1hd105yLEhEoG9pWYO1O0UmuvfeU1Gk2DBHGG+QPgcuBO4Gop5W9KKf+LlPI3gaXAHcBi4PfDb6bGL3UFzE344qorWa3qAA5fcKmU8nivF+ukxK7v0ajSZlvILKZqZRsKzhrf2knqyo2+3DncqLHdE3kVZK9yaX8B45kDNYmW6w7CKc5VVwktHiYKo5fjfI2tAeLERI1+3ZybZWBqp7rXdm7CXLWitL35T9uJb84ydK3yTR+6Vj21GHmHcHZIsbrs2OFnUunm1lPZv0YDKIfzZC4+XXufRGF07jH/k2S3CXUAe8PQCyh5UXHfGCP7Gb9nmv7nukhcEveXmDtT+LlnwqDdJ14ajaYjCBK8fwz4Rynl/dKitZGKIeCfgF8Ks4Eae+y+lOsOmJvwxdWQ/7PbF5xhqEDy1luVM86cObX7RyLqvRUraquMWicldn3v6YGnn1b7V+DoAT72OsaqNYyvHfZ053DD1nYvarGedHGMMU7HyqvUw3BgESXveiuxHCRen+18jT1qAhQlRENvP2r7lCCdzDEVo3r1eU6E/Lvf5fyEwuqy4+f61TSsICcxjLILkdU9KawAysEVKDn3Kvv75CWbY9TjAOTzSVnDE+h6sNw3xrEcg7vyZA6tZvBjg+0ZuINvT/yG0b7yGo0mBKLem5RYCBzw2OZ7wL+tvzkaP1gdHsZeHWP7xHbWXr3WMWB21ZgWv7hCTBJryKPe6QtuyxYVWBd1vsUcBDttsNt7lQGrW9937araP5WJsj0+zVRESWdKHuAjEk5PYdy7jcEGcgSSi5OMvTpW4xiTeKVQytNtUpVKwfbtVX1Nvh5h7PLpUrGjIhEJ3ZHZpP7oaedr7CEnSq8WTM0q+4xXPiUYHHZ4SiHyRGbNoXtWd9mdRMTozkVIfWE37NykNiy64dgl586bp+Q7iUSt1ttJJ269l+3ur+PH4bOfhUsvDaYjL8q5KkjdupXtD13PFAXNu8UrvoQQga5nldf/2FhtjoplEuA2gW6a5ryTfdNtrmXodPL4aDSatiHIyvtPgfd6bBMHamu2a0LF6Ut5+IfD9QfMQau1etCQBtzpC27bttrVUnBeMfO7mubUd8v+xobbGf/kM/T/ZGGtB3gIX8CpVala6c3cPuVe4rUaaNPX1B89TffsntLxBIIu0cUt772V8c/+AGNZ0rkxTnKiFSsgkSDz3oX20pDCUwKnKqKrr1hd9g+/JE7/cxHG75nGGNkPDz6oforXd80alZxb7NOGDepAO3bYrpibX9zCwAePk7g1qzT2cx1WNe3uLynhkUfKvvQNyBmMZUl1n5xbQeL4fPoPzKnyii+xcGGg61na1ueTstCLvPmh1T7tnYYeH41GEwK+3WaEEPcC/UAK+HMp5XTFe13AAPAnwP1Syt9qQltnlJa7zbi4SThZsC2at4g3zrzRFu4ODXnU2znACKF+Kv3WZ8oRJ6A7T5WDTKU1og3mpEn6iS2M7h8mcQRSvWswfufu6tXlivvCvHMj6Yn7yOwfJmmzfd1Wox5uJ7ZOIhW+8ubFUeKfnmZqbhdZmbO//l5OP9YxdRl38w9TxNNXMtU1XXZ4OVeYXF0ZVzaPxc/SyZPw0EMqYHcbgoujpDcvI/POOZ7XzRWv+6Ue5xgfdooz4vbSap/2TkOPj0aj8UFoVpFCiLcCo6ik1ZeAZ1Cr8W8Fri+8bgIJKeV5t/re0uDd4j2euRySryuZg7Es6filvOE9G3js+4+FV9Sp0W6EGTjm88qH3kqllVur7PPCKmp0gtr2gvOxLe+pALmsLXctDFVvPx0CRNt+Recx/uo6jG8fhERCTSxe3OZ8/V284Ku2KV5fFzu/gd9PMDR6b1UybCwH/c8JBp+cpe4fq/TGWnirsutFW9DZKm/A72epNFF7cYSkmVf1Apa8TxV9sisaBU0L5kIv8ub7xC32ae809PhoNBoPwvZ5fyvwv1B2kZV2HGeAvwNSUsqf1tnWIO1YDgyiJg3HgS8Df1D5NMBhvz7gz4F1KMnQ14DfllL+zOucLQ3eBwYwd9xPfHNOWRMWAzMxm/HP/gDA8UsZZr5oSygELbwTZEUrjCDf5xew4+rn1RsYvP2x2vauXevcT6hawR1Yo1xcqrzdiwHrlb/V9CcSDRcQC3HlPbEyY++z/7Ig81dd1U45sRgsWaIqvToE8LZj67FqXQqWz55UTxsq/fSn56vqtgcPVt8vTa6xEHqRN41Go9E0nWZVWJ0FXE25wurzUspzdbcy2LkXoBJjDwD/E3gnSq7zZ1LK3/PY959R7f4dIF/Y/6dSytVe521p8J5IMHDJ3trgIQ/9ybsY/Njghfel7BWc+w2CWvzY2rHKZHYRmfQbte1dsABes3lwlUgomUfFynPiU8qGsWbTlyCzvwOKy7hVzrW7Li7XbmAiXTtJkoL+/bMY/P/P1p47HofDh8vHikbVk50uFegnNsPexbW7FYtu2Umh0nts2pBTuRGDTzoE5Lo4kEaj0WgseAXvQdxmShQC9QnPDZvD7cBc4FellCeArwsheoHPCyHShddqEEJcD3wUuFFKubvw2hEgI4T4BSnlky1qvzfJJJmze2sTAiOUks2MPqP9qhQ2Ey9HHL8uDm5WbU1YqbZ1kInESBzBvr2gglInJ5EKl5HkERi7rHblPfGK6IwEOOs1XbZMvW5dnXbavmKbVG+K7RPbq11szuTYOHqWgTXK/SZ5RDm+GKdjSgNfdG4pHmvjxlLl2uTFpxkTB6uceooJ105uT1f0XeFs9emU0JxM+nKO0Wg0Go2mSODgXQjxFpRsJk555X0MeMSP/CQE1gBPWIL0v0Gtot+I8pp32u+nxcAdQEo5KoR4sfBe+wTvqRTJ397K2KVnqwMzEaxiZ4lWl1Jv1vncrNz8BkFBrNqK/di9W616RyKloM/sxVcSamqVJagsFm/qXQMxG3nMmjXw2GPVK8KRCIyMwHvfq/TaBe20sq7MMQXVloT7e+G+xorLBEmybWj/oPZ8DtsbfQbjt4+Xn0b96DQb//oAazZOl6RnY5fB9pUwvmMeRvGetB4rqRx4UpMmD9+3ghNnTyCRCATzYvPUCvsTW5g6PVm2ySy4PeVlnlgkVrPynjhC1b1YNTarl5P6x3mqcqyTnalGo9FoNBUE1bxvAu5BrXxba9SfAu6SUj4QVuMc2vAacK+U8vOW198EPi+l/F8O++0EFkkpP2R5fReAlPJmt/O22m3GPJgh/jc3MsVZlTAnYnTPriPZrNXuBjPlpuD3vEHlNSdPVuulo1HMy+arZMbcKV+JgLYSpxO4J6am0ypgP3RIyTlyuXKyZYV2uug24+hOU89QNpjoOGOJkpU4Sc+mof89tzK4/gHX3c1JkxX3reDk2ZPkyRMhQs/sHiZ++Qk+/r9XsffS2vSa+FvjHJ48XNa8FyZT41ujGLIHxsdVIqxHoq9OYNRoNJoLGy/ZjG+fdyHELwJfAU4Dnwc+BCwr/PvfCq9/WQjxkfqb64sFqCRVK28U3gt7vxnBWJZk/LM/oD95l6rYeW2dpcVbXdFvpioI+vV091tNttiPXHWBI3I50itPqADNZ/XYosQpszlTrjLp1t7iivDq1SqhstiGbFatuhcLVQ0OYixLMrj+ATJ/+FMGv/pTjL94oOHAr6HquE4kT8uSAAAgAElEQVT7n54k/cSWhtoViGSSzOU2haK6YHTyoOfu6T1pTmVPkUetrufJcyp7ivSDm0ma+VoPexkpe9hfe7vysM+uYPyZOMaG20vX1nZscqdIf7QntBoLnYZdtWjvnUw1EU8k1L91+PFrNBpNpxJENvM54BjwPillZaHv54HdQogHgOcK2309tBbaY/e4QDi8Xvd+QohPA58GuOKKK4K0LxRC0bW3uqLfTFYQ9CPB8FtN1qXCaOYySVZU3zJ1Fb/xau8MjWWjxX1s9xd5Rp/aBterpwKNynI8aVB65jgGZ17gkRHJ9vdQdoLKQXdelPrg9pmdkcJJbYxT/oDrQoX1KdvYmKpGq73SNRrNBUKQCqvXAH9rCdxLFF7/W8BxmT8k3gAusnm9D/uVda/9LnLaT0r5JSnltVLKaxcuXBi4oW1Bqyv6dUIFQT/VZO36UXzrFUFMVn90fFePDYKfsWzCCmRD1XGL+1vHJwcJMw/pdClgG9o3xN6X9zK0b4j4/XF/K65+MQxVYVbMJlZwgyxKz1KrvPXkjmMwZwnG6Rjj90H/s8rZp/85wfjURl+Tj0bH9nyjrqc8M/V0T6PRaNqEIMF7DCWNceN0YbtmcghYWvmCEMIA5hfe871fgaUe+3U2fmUinXo+P9QT4Bb7EbU8nIrFSO3vpXt2TykIi8kI3WfypJ44Gd7je9MsFxISonTuqrEsrkAODSm7waEhzFUrGNi5KZgEwUJqVYruWd3l/hWTbH0EvaX9s6IkLSkl0o5IGB1tWJbjl0akZ45jcOtW6O7GOB1jcBgyD8YY/NZFKs/AB42O7flGXU8iZvLpngt1yX80Go2mDoIE74eAdUKIOXZvCiFmA79M8wPhYeCjQoieitf+A2ri8LTHfpcKIf5N8QUhxLXAksJ7HYnnF0ZRJrJhAyxapHzE164NsQGWwBj8ac+bSNWY7NyEuWpFVYBLPO4dZBfGzbzzNxi45S0kfnsuA7e8BfOODRh7Jhi/c4L+qzeQeLWL/mcl4/dMY9y3w9+xPTtQCMp37FDJqkIo//ENG6rH0rICac7NEt8wydCBhxta0S46t/Rf06+C3muC5VsYfQbjUxvp/45QK9P7YPy+gkVjItFS6YhtvoHfPtiNwbKk9/3tMllsdGzPN+p6EtGGT/da8jRJo9FoCvh2mxFC3A7cC4yiElRHpJSnC8H8DcAfAAngTinlUJPaWyzSdAD4Lsoecgnwp8CfVxZpEkL8EHhaSvn/VLz2z8C7qS7S9FrbFWnyiTlpEr93hUqeFHmiUtDVFePqhUu54YobyjriZjnAzJSzjFuTrE4nhVXx8ftQDi/gu4Klp2uKH+caD9tMW+3376VrjxuJwC23wAMPlPf5l6+S/P6byrv8RPCqoOakyZantjD8w2GYnmbN8Uu4e89sjPgNysXmxW3+NOl2fYRgBZU8qpd2DK2u9Nvu+Lj/AzsTteHfHcdKyufDPa3RaFpOqBVWhRBDwGbKCZ5TQHfxbWCrlLK/zrYGacdylGXl9Si9+pdRNpHTFdv8GPimlHJTxWsXAX8G/ArqqcPXgN+WUh71Omc7Bu8DOzcx9N0Hqx01JCCo/hK0CwbDKMHe5NLudTXJ7ku0WOWy8vmKjwqWnl/IXtUxPYIMx8Bll4Exsr/2uF1dmBN7iD+2pmxHOA3d59TK9sfXO1RcLVQFraRohTh5drL8ooS+s/DEji7WbMgzNbdLncMtoHLrI9gmBnsGbJ0c1LZppd8ZwWcf66oWXbxH3JLOW4hjJWWbz55Go9F4EWqFVSllvxDib4FbgJWoIk0/BMaBh6WUTzXS2ADtOAD8vMc277B57ThwW+Gn48nsHyZrzTAoyKMrdcSDzdKIOhzX3D9C+vGB5jmJuDXJTpJRrHJZxOcjdk95h1dhKI9qro7a7w9KBvdElN69knye9IObmZp3slT5M9sFUzFIr1KJtGOXyZqVdzsJQnpPmpNnT1a/KODELNj8sWmmYpTPUXkvWVcRvSrW+imoVBmwdbqTSJtW+p0RfPaxLletoMW9moxjJeULNBFZo9E0lyCadwCklN+QUt4mpbxGSvmuwr+/2arAXVMmeYQav+lKSoFmAxrRGk39wUxZz3vmTE1Cp9kL8Q/uZyhzj9J+Pnt/S7WfthraaUi8UpH0OW8enDyJeUOcgf+8ksRfvrc6X6CgWU5+6yfurjJ2ybmRiKrIOjCgiiw5TG4GHh/gq2NftZ8cGJFykmolUpI5+0IpqC7tU5icpF5/N91z+3wlQ2aOZEoe5lWniMALF9v4oztp0uucGDpq0cNwEplJD3C/n7WZSrps5di0aWJpM9CJyBqNppUEDt4rEUI021lG40Kqdw3d5yoCeIsCqhRo1ukAU5OE9ez9xB+6HnPH/UoucvCgSqisCODTq2AqJskW7qyszDF1NnwnESdsv0Tn9pF65ydVwLJhg+rbP21Xk4xZE+w9Ol6eZBzMlBxcUv/wGt1nygV5ar6QKwstxeMqcJ+ehv37lXTi0KHayc3FUeIfOsTQviHezL5Z0/5YJEbiytWwcWNtAB+LkZy9pGR9WHo5B4kjYNCnEml9JEMmFyeJ2Hz8RR6WHKudFDquIoadPNhowGfjwBNKErFf/H7WZiLpstVj04aJpc1CJyJrNJpWEih4F0JEhRADQoiMEOI0cEYIcbrw+4AQIpAMR9MYxu/czfiOPvq/I4i/ArNzELULNP1WH7VQI+uQOaaiknSysPKbyykXlN7e0j6ZxTartrJ1RWhsv0TvnFCVRzMZVZ301CnSyVypyE6pb+emSD+4ubTya5xQWvL+7wgS2UX2X8hu1VCnp9X4VARy6Ru6mOqarllxB8s1u/tuuOiimiAwdetWuplda8OYiUIi4dtdJbUqRc/snuoXJfSeg62Pdymrx8LH2XUVMWxr0EYDvpn2AA+70m+YtHps2tE2tonU62yk0Wg0QQniNtMH/AuqCFMe+AnwGrAIuALoAvYCvyilPOF0nE5lRhNW3RL4KhK3zOuWkV6lyr+7Jn75TAh0TMJ6CTJfrnhh/nx4U60i2zqe5KE/eZfStc50MmIhyTTxKYfkzjfmk/mL2hVxzwRXp+TVeFwF9oXEusTSEfYeHa/ZbH5sPre997bqa+aQlGcezJD+3RsZXXiWxEsqcDdkT2BduK3bzLfmYKxcXXKb8ZVEGGbyYKOJnF5JxO1Eq5MuZ2Js2iyxVKPRaDqBMBNW/xj4APBXwO9LKY9UnGQx8N9RiaB/DPxWfc3V1OCVwFeRuGUAnilcARICbZOwChKN8gsxWLJESUSyWVJ7YPtKS+n4yGy1atsOyYiFJNPkkSxjl9XaKibmLIHYIeckVI/j1uy3enVVYl3y8QHGjh2oSWy77b231SbtOSTlGcuSDP7vH6ig6GejsKG+oMjoM3hg3QP27wGDy5I+D+SePGhrh+k0ESiuXNcb8HklEbcTrU66nImxabPEUo1GozkfCLLy/irwfSnlDS7b/B/gXVLKt4bUvrZhxlbew7ZjDHC8Gks/EaP7dI7xL3VhHMuVV0WHh2HNGlURNJfD7FXa99HFkDg6i9Qf7VbFbRrwRa8JAK/ciHHvNv8r+MXjjozAoUOY83PEN0+XJxkiRvfsbsbXDmOsWlO78js8DNtczhfAFi+wr/UMEijwdti/pf0Ny4Jxpp8QhUVlP5Yvh0cfhVOnzl97So1GozkPCM3nXQjxJqoQ0n912eZ/AJ+RUnY7bdOpzFjwHvaj7oDHq/FgLgbN1lVR04QtW+Dhh8sWh0IoPfzEhNqmTl90c8+w8jYvTSKidJ+erp1EOAUh1uNGo8oz/X1LSN8YZdSIkLhydbVdYeXK78aNanLiFRD6lAj49rW2Bl4ABw60LJgMI/CekeI1jUo1zhcPdrt+zJsH69apZHMtY9FoNJq2JEzZzBjwTo9trkJ5vmvCIuxH3QGPZ+vBPGgjpzAMlQza1VUO3qVUq3xFX+c6fdHTD25mar41cRbSyZwqvOTlkW09bi4HQmBc+2EGv2CzvfVR/8CAP09unxIBX77W1sCrctLTIrmRow+9nd+7A55e+c2gUanG+eLBbtePU6fU57Td9P8ajUaj8U0Qt5n/BqwTQmywe1MIsRFYC2wJo2GaAmE7Nrgdz8MDusbz3erd7mXz59UXh/0zZ1/wLrzkZifo137Qqf8z4VdtDbys526Bg0oYgbet7367F685X/zJz5d+aDQajaYKx+BdCPHfKn+AVcDTwMNCiO8JIb4khPgfhX+/CzxYeH9Va5p+gVCnzWPg44GrB3SN5/vovcTTV2J+ZlM5yPWy+fPqi8P+ydlLSraFpZftEmednkb4sR9088CeCb9qu8CrkjqDMM8JWAW+A2+XSV+N776I0Z2LkPrC7tYXUPLZ5vPGn/x86YdGo9FoqnDUvAshassv+kNKKbu8N+ssZtQqshV4JJPaapdz0P+cYPBbF1VPAOrVCjtp3v/hAeJPrFPFn4oONlkY39qFcWzal+bdXLWC9MoTZC6TJF8RpPb3YuyZKG/v1v9UqvUaaLv2VGKTZOyVXBpUw+5rex/68JLG/8UREk8eIrV72l+uQrPwavP5rHnvxH5oNBrNBYaX5t1NNnNTnT8/H0rLNa3F4xG7rYQiCqOXybKEwzCUK8vSpcr7felS9bvfQMFhZd74+68z/qUu+p9VHvP9+wqB++Llvp5GmL0QvwOGrhXsvVz9G79Dve6r/2E//fCDVWJkJRJRibQFap6M7BtSFWMrVtbdNOx2+Koa6aPwT6l4zaHVDO7Kq8DdYduW4NXmmbjezeB86YdGo9FoqnBMWJVSPt3KhmhmGI9kUlfP92KQa5rVriyHDqnfgwQMdsmGmQzGsUJyaolpuGqur8S79J40U7lTZAsPk7Iiz1TuVHXipVcybav9qq1+51dcAY89BufOqfenp6vG1k9y6e7DuwNr2D2Ta4PoqttFg+2nHeeLP3m79ON8sd7UaDSaNiBIwqrmfMKq+d240TWZtKRdluqWKUpXUnsoB7nNKr/eoHbXV+JlO5ZyLwZemQxceqly7ymSy1WNrVcfzUmT548+X3OKqIg2ljwa5NrYbQtw+nRrte9aC95a3PJJNBqNRhMYHbxfiFi/TO+9F1atgg9/GDZssH3EXpJQLL+FxKtd9D8nGL8PjNMVQa7TiubOnY4ONr5oMLD2lXjZ7hIDj9Virz6m96SZltM1h+2KdKnqt/US5NoUt41aHvgdPNhYMOfhkmTbjnnzVB0CUP/OmzezE7WZIOi41UuzJvUajUZzgeK7SNOFznmVsOqUDCkEXHSRd9DqVATH6biRiPJ+95Mw5/R4vYHCO51W1RSoHYeTJ2HHDseEYttquNMRxp+6GiN+A4mlI+w9WluCIf7WOGO3j4XTVj/XxjTh5ptV4a5K6q0aXE9SpmnCihVqTPN5dX/29JSLiV0ItDKZNexCcxqNRnOe45WwipRS//j4ueaaa2RHc/iwlHfdJeUHPiDlokVSKhFG7U80qrar9xwLFkgZi6ljCVF7/FjM+fiHD0vZ11feTwj1++HD9fe7eOjjh+Vdu+6SiS8l5F277pKHjzd+zKZhHcdYTI1DX1/1awsWVI1NqY/3xOVd62bLwxdHS9vetW62jP1BVPJ5Sj+x/x6Td+2q81o3wgc+YH/vJRLBj3XXXeUx8XOP1bvP+UYrx0CPt0aj0QQCeFa6xKRBKqxqOhXrKlvERS2Vy8HISH3nqUyyHBmB55+HM2eqt3FLUNyyBSYny79LqX7fsgUeeKC+NhWb5qeqabvgVBlzwwa1Quywwl3q48AA7DoA2bKrS2p3lO0ru5iKiqqnDw1JZuolzKrB9STBtkvi7EzSyjFIpVRFYOsq/4UmU9JoNJqQ0Jr3CwFrMJj3sPD3et8Nw1BfyocPl51RKnEL0oaHg71+vuIUWB08WE5gHRx0ljfY7G8cyzH+zaXuto+tIszk4HqST3XCamvHoN3zSTQajabD0MH7hYBTtc4uh1pabivzfihOFqyTACH0ipsfGg2sHPY3Vq5WfuubMwx+bHDm9P5hBnP1TARm0lmoVUmiXrR6DCqdk9wmnhqNRqPxRAfvnYRpYn5mEwO3vZXE772VgZ2bXMvbl1i+vDYgj8XU61bnj1gMVq9urJ1Ok4WFC92DtDVr3F8PI/Cp5xghnNecNBl4fIDE1gQDjw+4X7dGA6t2tL20ElYwV89EYKZWgtvJMlGvhms0Gk3Hot1mfDLjbjOmiblqBfENk0zNgmwXxKahm1mM//pujGVJx/1YsaJaSw7Q1wdPPFFdVKnScQLqL6pi5zrjx03EzQUEGnfHqNeZpMHz1uV204C7Tij7a8Kn3s+FRqPRaC4ovNxm9Mp7p5BOk155ohS4g/p3Sp4j/bs3Oq/epdMq2bESIWDdOhWU262+QWMrhPWu/BqGCtTvvFO15847y/Z9YXhF13OMEM5rV/30jTNvcPOOm51X4BtdmdYyhfZDJ8pqNBqNJgR08N4pZDJkLpOlwL1INgqjC886B5N2AYOUKvkR7IO8RgPWRh7JOwWdYQQ+M+RMYlf9FGDitQni98f9SZ80nY9OlNVoNBpNCOjgvVNIJkm+IohZimTGcpB4Cedgsp6AIYxAOeyV3zACnxlyJrGrflpk6twU6T2FSVG7JDNqmkMn5CJoNBqNpu3RwXunkEqR2t9L9zlKAbzIQz4CJ+cIzOuWOe5XGTCYF0cZuDlCYumIc+Kkz4A1UBJmo4QR+MyQM0lqVYruWd2272XzWUaPjLZXMuP5SMCJUVPubZ0kqtFoNJoQ0AmrPpnxhFVQSauf/yxbjj3CtpUgUcF7bBq65/YxfueEfQJkIXnR3D9C/EOHmOqaJitzzomTPpI060rCDKH/DSdh1nOMEM5rTprcvONmJl6bqHo9FonRf00/g8PoZMZmETDpeEbubY1Go9FoCuiE1fMJw8D4yt/R8+u3EhER8oWrl+2CqdypsvzCZj8GB0l/bjVT0TxZqSpvZvPZatlG5fYeK4R2SZi2xwqTMKQ49RwjhPMafQa7NuxiwZwFJQlNVZXTEKRKVavFOzdhfmaTluBA4ByOGbm3NRqNRqPxSdR7E027kZk8QFZUF0AqyS/c9rNJnHTcrxiwhnGsmaK4Yp7JKE97gAMH/FlfVu5r3d7uPfC01jT6DMZvHye9J83okVESixOkVqXUam4yCWNjtSvvPrT15qTJlqe2sG3/NvIyj0QyZu5l+1wYfx6MsTFVnv5ClWgEnBh1xL2t0Wg0mgsWHbx3IMnFScZeHasKMGIiRuJHp1Ww5xA82u4XiZFY7BEg2gSrdR/LC7egOehxKqUSe/eW3/MKZq37Vm4Pte89/LB6/dSp2u1tAvjBj9lMilIptY9V2uGhrS9KPCbPTJKnPKHLdsFUDNKrYHC4YqXZpwTHnDRJ70mTOZIhuThZnmR0IgEnRk27tzUajUajCQGtefdJW2jeC9RockWM7tM5xr/UhXEs56jprbtYkI1e2NwzTPyxNUydPan083noZjbjv/a0c8Eoz47Vnsu8dB7pL64jM3kgWBBpVxCnEjc9uVsxHah9Twj1k69+GsKKFbBrV7AiUgG19QOPDzC0b8jWihKUE1Hmy8VfEmpS5NWM803zrTXvGo1Go+kgtOb9PKQov+i/pp/E2xL0n1paDtzBUdNbs981/d4BiYNe2Lh3G+Nrh+l/rovEEejfC+P3TGOsWuNPX23n/mE5lzk3S3zDJEMHHmbvy3sZ2jfk3xfdTipRiZue3E1m4eSbbw3cQRWYCuIYU4e23slDHgo2okeKv/i3tzzvNN8BXV7q+pxoNBqNRtMitGymEzFNjHSawaK0ZGQEioF7EUtwak6apJ/YQmb/MMkj8EjvGox/mwKvgMQlkDWAwV15KL2dA96Am292X3G2roTu2wf33QcXXVR1rvQqVEXZgr6/Moi0lZ5UYieVqCQWg2XL1MTBKtFxklksWwbf+U7tsZxW3iGwXCUodhIPUDai3VlI7SGwvaWj5vvFkbCa3Xo8cjhqNneSN2k0Go1GM8PolfdOw84P/NAhiFrmYRUrreakSfzeFQx990H2xl5jyHiN+NwHMVet8F4VdvN8d1rd9lpxtq7m5/MwPQ0/+1nVZpnF1FaU9Zs4aPVnt7Z/3jx49FF7X3U7b/fi9sXKtJXH6u2Fnh77dgQtbhWQood80cEmQoQu0cUn3/0Jxk/firE0uJ94cnGSmKi+n2I5SDx56MJ2rdFoNBqNpg3QwXunYSdjmZ6Gri7HQkLpPWmlTS8EwqVkxpUnHO3ySrgVKbIL7Iu4WPF5SlqEALCvKOs3cdAqlbj1VvVTlE2sW1dOMIVqqZGdzKK4fc7yhGPpUjVZmZhQGncrQavABsQq8bjzA3fy4mde5IHf+DuMv3igLnvL1KoU3dNdpbGP5Qqr+Lunve8XjUaj0Wg0TUXLZjoNu8A3l4Of+zlYvdo22TFzJFNrLRmF0cuk96pwMZC1S6QsOqS88UbtftmskvN4yFLMXiWPySyG5BEl8zDmLIR3vIPUO5exfe6jTOVOVSUOplb5kH94udYkEu72gVaZhd32AHPnlo+7a5d9YmSQKrB10LDEwzJWRirF+FNXk+7ez+hipZtP7QHjRK6pTxEaIgyXorCcjjQajUajaSI6eO80nPTYq1c7anqTi5OMHdlXFcDHcpB4RfhbFXbSCxcD+5tvVivPlcRiSs5z4ECtfWIh6DflceL9Uunau2DsMti+EsZPr8H4iwcwgPGCZWGNL7obblaPxWAsqK+6n+3dJjrtisNYGWvXMrjjYF2+8y3Hz/VuxTE0Go1Go2kBWjbTabjJWJx2WZWie3ZPrQxif2/jq8KGoVacFyyoblMkouQ8LrKU9G3vLgXuUJDzzFIr8aXDF1aVM5szDH5s0J/jh5+KmkHH0e/2YVSBbSVOYwXO/bVzCppJAlZQbdoxNBqNRqNpATp47zQC2t5BQRd95wT977mVRHYR/S8tUsmMeybCCS7t2nT11bX6cIssJbOstzYhtQtGJy1JoUHxU1Ez6DjWMe4dgdNYHTxo31+oTZgOYofZDAJWUG3aMTQajUajaQFaNtOJBLS9g8IK9voHYH1zmlTTpoEBFQC6yC6aVsnSryTGbRyt+ueNG2HbtvNPD+02VnbjMzDgvELdJDtMT4JKoJp1DI1Go9FoWoCusOqTdqqw2hH4qGrZtEqWAStqeu4fjZYdfXLOFWw7kqBjlUioFXe7131Ub20KjV7vsI6h0Wg0Gk0I6AqrmpnBh8ykaZUsG5W4WPXPuZyqoppzr2DbkQQdKzff/5kiDEnT+SqL0mg0Gs15h5bNnE+0m9WdD3mPcQIGh4GMhCSwAuir41x2MhdQQXfQ4+zc6e5DD8H10E26NmbBjSdzJENycdKfG4+VIDKsoj1oi+0wPalDStaUY2g0Go1G02S0bMYnbS+b6cTH/mG1OSyZS/E4x497B/2xmFqd9RPsNenaNE125Hlis7PsMDUajUaj6SC0bOZCoROt7sJqc1gyl+JxrIG7EOonWnhQFXS1uUnXJr0nXQrcAbL5LFPnpkjvafI17zQ7TI1Go9FoziN08H6+0M5Wd06+4E5t3rkzmPWg3XGsFMfCzaPc6TgLF8Izz8Dtt9enhw7j2ti0O3MkU+XUAyqAHz1SfVxz0mTg8QESWxMMPD6AOTnDvuwajUaj0WjqRmve25hAeuZ2tbpzq1xp12aAo0fVPn4DZKfjVBKLwbJl7lU0ncZw/Xr1XjLp3Vc7XXuj18ZhDJP3r/W02rRKa8ZeHWP7xPbmS2s0Go1Go9E0Bb3y3qYUg66hfUPsfXkvQ/uGiN8fd141raPyauON9FFp000yUmyzENX75PPBZCV2fbeTuYC7fKWRMSwG2HbFixq9Ng5jmNoD3bO6iUXUcYua99Sq8nEdpTVPbGmvKqkajUaj0Wh8oYP3NiWwnrnVVnduwWolbpKRYpsXLqw9fhBZiV3f7WQuBw64y1caGUO3SUqj18ZhDI1vH/S02nSU1jy1rb2qpGo0Go1Go/GFls20KX71zFW00urOLVitbIOXZMQwlCxlaKgxyY9d360yF7u2RCLw4x+r1eeizKWeMfTStTdybVzG0OgzGPyY83Ftq9jKCAkzD9npcjtnukqqRqPRaDQaX+iV9zYluThZkkMUseqZZxS/SZh+JCOtkvxYzwNKovPaa42vPodRvMhJhtTA+KRWpWqlNVlBasTiqNMuyc0ajUaj0Whc0cF7m2IbdFn0zDOK32DVj2SkVZKfyvMsWlSttW/UvrHRCYibDKmB8bGtYju1EeN0m1VJ1bQXfvJZNBqNRjMj6CJNPpmJIk1Ft5nRI6MkFifqq57ZLDqxKFQliYQKku1ez2RKvwZy/GmkeNHAgL10yG8hqCB0+rXTNBd9f2g0Gs2M4lWkSWve2xgvPbMtTnaFoTeusBrcqZU2fdg3BrZZbETX3kqf/k6/dprm4jefRaPRaDQzgg7eOxGnAN3NU71ZAXynfpmnUmpsrKuLFTIXN8efwJMqL1rt09/J107TXNq54JtGo9FotOa943DTRrutmHUyzdDf+tCR1+X444RXH2bCp1+jsSOM5GuNRqPRNA298t5puAXo5+OKmWlirlpBeuUJMnFJ8sVnSa16GGPPhOvTBF9adY/VZ1ubxXocf/w8EdFSFk274OOplEaj0WhmDp2w6pOZSFi1xS3RMpFoXdJjizA/s4n43AeZmgXZLohNQ/c5GD99K8ZfPGC/j0WrXnTqcdSqO507pOO0NBlVowmDRpKvNRqNRtMQXgmrWjbTabg90m6G9GKGLePSJ4ZLgTuof6di6nXHfYJWp3XA1mYxaOAO5+cTkUbQNoTtT/GpVCaj/tWBu0aj0bQNWjbTabg90g5betHqBFgbMovLgXuRbBRGF7vsE6JWvS7HHyutTkZtZ0oNSqEAAB4QSURBVNrgntJoNBqNppPpyJV3IcRmIcQPhBBnhBD7hBAf9rFPvxDi60KInwohJoUQe4QQv9iK9oaKV6Kl24pZ0BXPNkiAXb7sBrAquyQsW3aD4z5tV51WJ6OWaYN7SqPRaDSaTqbjgnchxK8B9wMPAWuA7wFfE0K8x2PX/wq8CPQDnwB+CPyzEGJtE5sbCuakycDjAyS2Jhh4fACzl+CPtN1capxoB7nH/PkgLK+JwusOtF112lZVkO0E2uGe0mg0Go2mg+lE2cwfAA9KKe8GEEI8DbwP+M/ARpf93i+lPFrx+9eFEO8C/l/gsWY1tlECFwpyop7CK20g9zjw+gHb1w++ftBxn6JWva2q02pfdUUb3FMajUaj0XQyHbXyLoRYArwb2Fl8TUqZB/4OtQrviCVwL/IcsCjMNoZNWMmXda14toHco14JTFGrntmcYfBjgzMbuDeLTkz8bIN7SqPRaDSaTqajgndgaeHfQ5bXDwIXCyEWBjze9YD90m6bEFryZT2FV9pA7tF2Eph2oR4ZVDvQBveURqPRaDSdTKcF7wsK/x63vP6G5X1PhBC/iZLb3BtCu5pGaMmX9a54zrBlXGh2jWEz06venZz4qW0IW8NM36MajUajaQozXqRJCNEHXOa1nZTykBDiN4BtwEVSysmKY3wE+Bfg3VLKH/g45zXAbmCrlPI/umz3aeDTAFdcccU1P/nJT7wOHTqhFQoC58IrxdczGbVCrwuyuGO1OyxOhFq5guxWrCuTaU0bNO1LO9yjmrbAV7VpjUbTVngVaWqHhNV/D2z1sZ2gvMJ+ETBZ8d5FhX+tK/K1B1G6+V3AvwL/yW1bKeWXgC+BqrDqo42hE2rypV3SpPbdDk49yb9hoxM/NW60wz2qmXFCMzzQaDRtxYzLZqSUX5ZSCq+fwuZFrftSy2GWAseklK+7nUsIsQh4AvgJ8GtSyulQO9Mkmpp82cnyi5miHewOdeKnxo12uEc1M05ohgcajaatmPHgPQhSyheA76NW6wEQQkQKvw+77SuE6AYeL/z6S1LKU81qZ0ehv+SDU0/yb9joxE+NG+1wj2pmnDCrTWs0mvaho4L3Ap8HbhNC/J4Q4ibgr4B3AV8obiCEuFEIkRNC3Fix3z8AK4HfB94phLiu+NPCtjedmoJOkx5Jaq34km/jxLnA4wXts+qtEz81TrTLPVovbfw3o5Nou2rTGo0mFGY8YbUehBCbgc8BBqrC6v8npfzXivc/BDwF3CSl/GbhNceOVshyHLn22mvls88+21jDm0xdya3NTmyzO/68ebBuHRw4MKMJsg0lAzsl/3YqOmn5/KNT71GdbBsaoRoeaDSaluGVsNqRwftM0AnB+8DjAwztG6p6TBqLxOi/pp/Bj7kkqTXzS35gQHmQW6U5kQjk8zP6xVz3eJ1v6GBJ007Y/c2IxZQ0TCfbBqboNtM21aY1Go0nneA2owmJuvWNdi40oTXKRlMPKnCHGXXB0HrQAtqZRNNO6DycUCkaHmj+b3t3HyVZVR5q/HkHGgM0zKBAiJNGQkQGNcz1MukWBhQURVQ0ahQ0RFGRxqUd47rY8VsI5q7QuUluMmr48APFWSJEg1cEJDGBxFF7MsTpEAQEErCBIKDDSPM1LbPvH/vUTE1R1VNdXdVVp/v5rXVWTe/a55y3qveqeWv3e/aRFo4y1ryrgZ6sb6xXU1+rnf8xz6JWtiffr24wWVIv8WJbSZqRyfsCMrp6lP7d+rclpH1L+ujfdQ9Gv/1w9y78qr1wLupcXtCu/5gr5R8XXJBvYHTBBfnnBq+57vu1Wz+jq0tyUV+7mCypl5T9YltJ6jBr3ptUhpp3qKlvXHoYo2ddwcB9j3a3lrm6pv6ww+CKK+DRDsTUQq2s9aBY867eU9aLbSWpDbxgtU3KkrzvoFcv/OrUf8yDg3nGvV77+Pjcj7+QmSxJktQTvGB1MevVWuZOXSA7NAQbNz71y4rlHzvXyYuWJUlS21jzvpD1Qi3zfN5sxVpZSZK0wFk206RSls10u5a5G+e3/EOSJJWYZTOL2cBATpS7lcx2Y/1wyz8kSdICZvK+0HUzme3VmntJkqSSsuZdndOlmvvJzZOMXDXC4EWDjFw1wuTmeV7bXpIkqUOseW9SKWveu60LNe+TmydZef5KprZMMb11etuNlybOnFh867dLkqTS2VnNuzPv6pxKzf3wcJ5tHx7u+MWyY+vGtiXuANNbp5naMsXYurGOnVOSJGm+WPOuzprnmvvxe8a3Je4V01unWX+PdfaSJKn8nHlfjOZz7fV5NrR8iL4lO9bZ9y3pY3C5N2qSJEnlZ817kxZMzXu3137vMGveJUlSmVnzrh3NtPb6bPXgDP7A0gEmXnM1w4+sYHDTngw/soKJ11xt4i5JkhYEa94Xm3atvV47g79xI6xd2/0Z/MlJBlafyJptf1m4BS48sftxSZIktYEz74tNu9Zeb+cMfjv1alySJEltYPK+2IyO5hr3SgJfqXkfHZ3dcXr17qm9GpckSVIbmLwvNu1ae71Ld0/dqV6NS5IkqQ1cbaZJC2a1mXbp1VVrejUuSZKkJrjajDqjC3dPLXVckiRJbeDMe5OceZckSVKnOfMuSZIkLRAm75IkSVJJmLxLkiRJJWHyLkmSJJWEybskSZJUEibvkiRJUkmYvGthmZyEkZG8xvvISP5ZkiRpgdi12wFIbVN7d9WNG2HtWm/SJEmSFgxn3rVwjI1tT9whP05N5fYWTG6eZOSqEQYvGmTkqhEmNzuLL0mSusuZdy0c4+PbE/eK6WlYv37Wh5rcPMnK81cytWWK6a3TbLxvI2tvXMvEmRMMLHUWX5IkdYcz71o4hoagr2/Htr6+XP8+S2PrxrYl7gDTW6eZ2jLF2LrWZvElSZLaweRdC8foKPT3b0/g+/ryz6Ojsz7U+D3j2xL3iumt06y/Z/az+JIkSe1i8q6FY2AgX5w6PJxn24eHW75YdWj5EH1LdpzF71vSx+Dy2c/iS5IktUuklLodQymsWrUqbdiwodthaJ7U1rz3Lemjf7d+a94lSVJHRcQNKaVVjZ535l2qY2DpABNnTjB8xDCDzxxk+IhhE3dJktR1rjYjNTCwdIA1r1zT7TAkSZK2ceZd6jGuLy9Jkhpx5l3qIa4vL0mSZuLMu9RDXF9ekiTNxORd6iGuLy9JkmZi8i71ENeXlyRJMzF5l3rI6OpR+nfr35bAV9aXH109+7vESpKkhcfkXZqNyUkYGcl3cB0ZyT+3kevLS5KkmXiH1SZ5h1UxOQkrV8LUFExPQ18f9PfDxAQMmFxLkqS58w6rUruMjW1P3CE/Tk3ldkmSpHlg8i41a3x8e+JeMT0N610JRpIkzQ+Td6lZQ0O5VKZaX1+uf5ckSZoHJu9Ss0ZHc417JYGv1LyPuhKMJEmaHybvUrMGBvLFqcPDebZ9eNiLVSVJ0rzatdsBSKUyMABr1nQ7CkmStEg58y5JkiSVhMm7JEmSVBIm75IkSVJJmLxLkiRJJWHyLkmSJJVEKZP3iHhXRNwWEY9HxA0R8dJZ7v+CiHgyIh7sVIySJElSu5UueY+IU4DzgS8BJwI3AVdGxPOb3D+ATwEPdCxISZIkqQNKl7wD5wBfTCmdm1L6J+A04Hbgg03ufyrwq8DnOxOeJEmS1BmlSt4j4mDgOcBllbaU0lbgcvIs/M723ws4DzgL2NKhMCVJkqSOKFXyDqwoHm+pab8ZeHpE7LeT/T8O3JxSuqLtkfWKyUkYGYHBwfw4OdntiCRJktQmu3Y7gFnap3h8qKZ9U9XzdWvZI+JQ4D3AUGdC6wGTk7ByJUxNwfQ0bNwIa9fCxAQMDHQ7OkmSJM1R15P3iFgK/NrO+qWUqmfbU+1hGrRX+yvg4pTSjbOI7QzgDIADDzyw2d26Z2xse+IO+XFqKrevWdPd2CRJkjRnXU/egTcCFzXRL9g+w74M2Fz13LLisXZGPu8YcSKwGnhvRFT6/kp+KpYBj6WUnqjdL6V0IXAhwKpVq2b6YtAbxse3J+4V09Owfn134lkgJjdPMrZujPF7xhlaPsTo6lEGlvqXDEmSNP+6nrynlD4LfLbJ7pXZ9xXAXVXtK4Cfp5QaLf94KNAP3FbnuU3Ax4BPNhlD7xoayqUy1Ql8X1+uf1dLJjdPsvL8lUxtmWJ66zQb79vI2hvXMnHmhAm8JEmad6W6YDWl9J/Aj8mz9QBExJLi56tn2PVvgeNqti8Cvyj+fUmHQp5fo6PQ358TdsiP/f25XS0ZWze2LXEHmN46zdSWKcbWjXU5MkmStBh1fea9BWcDX46IO4F1wNuAQ4C3VDpExIuB7wAvTSldn1K6G7i7+iARcSwwnVK6bl6ing8DA/ni1LGxXCozOJgTdy9Wbdn4PePbEveK6a3TrL/HUiRJkjT/Spe8p5S+EhH9wB+Ry11uAl6dUvqPqm4B7ML2C1kXj4EBL05to6HlQ2y8b+MOCXzfkj4Gl1uKJEmS5l+k1PvXYfaCVatWpQ0bNnQ7DM2z2pr3viV99O/Wb827JEnqiIi4IaW0qtHzpap5l+bbwNIBJs6cYPiIYQafOcjwEcMm7pIkqWtKVzYjzbeBpQOseaWlSJIkqfuceZckSZJKwuRdkiRJKgmTd0mSJKkkTN4lSZKkkjB5lyRJkkrC5F2SJEkqCZN3SZIkqSRM3iVJkqSSMHmXJEmSSsLkXZIkSSoJk3dJkiSpJEzeJUmSpJIweZckSZJKwuS9LCYnYWQEBgfz4+RktyOSJEnSPNu12wGoCZOTsHIlTE3B9DRs3Ahr18LEBAwMdDs6SZIkzRNn3stgbGx74g75cWoqt0uSJGnRMHkvg/Hx7Yl7xfQ0rF/fnXgkSZLUFSbvZTA0BH19O7b19eX6d0mSJC0aJu9lMDoK/f3bE/i+vvzz6Gh345IkSdK8Mnkvg4GBfHHq8HCebR8e9mJVSZKkRcjVZspiYADWrOl2FJIkSeoiZ94lSZKkkjB5lyRJkkrC5F2SJEkqCZN3SZIkqSRM3iVJkqSSMHmXJEmSSsLkXZIkSSoJk3dJkiSpJEzeJUmSpJIweZckSZJKwuRdkiRJKgmTd0mSJKkkTN4lSZKkkjB5lyRJkkrC5F2SJEkqCZN3SZIkqSRM3iVJkqSSMHmXJEmSSsLkXZIkSSoJk3dJkiSpJCKl1O0YSiEiHgDu6nYcC9C+wIPdDkKl5NhRqxw7mgvHj1rV7Nh5Vkppv0ZPmryrqyJiQ0ppVbfjUPk4dtQqx47mwvGjVrVr7Fg2I0mSJJWEybskSZJUEibv6rYLux2ASsuxo1Y5djQXjh+1qi1jx5p3SZIkqSSceZckSZJKwuRdHRERz42I70TEoxFxb0T8cUTs0sR+SyPiCxGxKSI2R8TaiHjGfMSs3tDK2ImI3y7Gze3FfrdGxCci4lfmK271hlY/e6r2XxIRN0REiohXdzJW9Za5jJ2IeH1E/GtEPBYRP4uIayJiz07HrN4wh5xnVURcW4yZn0fEP0TE0M7227U9YUvbRcQ+wD8APwJeC/wm8OfkL4sf3cnuXwUOBU4HtgLnAVcAx3QqXvWOOYydk4u+5wG3AYcD5xaPb+hgyOohc/zsqTgdWN6RANWz5jJ2IuJ04FPAGPABYB/gJZhjLQqtjp2IGCj2+zfgrUXzB4BrI+LwlFLjewullNzc2roBHwI2AXtXtY0Cj1a31dnvSCABL6pqGyzaju/263Lr6bGzX522M4qx86xuvy63+dlaHT9VffcBHgDeWYydV3f7NbnNzzaHz559gYeBd3X7Nbh1Z5vD2DkTeBJYVtW2T9H27pnOadmMOuFE4NsppV9UtV0K7A68eCf7/TSl9M+VhpTSeuC/iue08LU0dlJKD9Rp/mHxuH/7wlOPa/Wzp+JcYB3wnQ7Ept7W6th5U/H4xU4Fpp7X6tjpA34JTFW1TRVtMdMJTd7VCSuAW6obUko/IX8LXTGb/Qo372Q/LRytjp16jiKXXt3antBUAi2Pn4g4HHg7cFbHolMva3XsDJE/Y94ZEXdHxHREjEfEUZ0LVT2m1bHztaLPn0fE/hGxP/CX5Fn8y2c6ocm7OmEf4KE67ZuK59q9nxaOtoyBiDgA+AhwSc1siBa2uYyfNcCnU0q3tz0qlUGrY+cA8nVaHwX+CDgJeAS4JiJ+td1Bqie1NHZSSvcCx5Gvy/ppsb0eOKHBX5O3MXlXp9S7gUA0aG/Hflo45jQGImI34DLynx/f38a4VA6zHj8RcQo5Aftkp4JSKbTy2bME6AfemVJam1K6Bvgdct3ye9sfonpUK587vwb8LXADufTmxOLf34qIA2c6mVdCqxM2AcvqtC+l/rfT6v32q9O+bCf7aeFodewAEBEBfAl4HrA6pbSpveGpx816/EREH/Bn5JWKlkTEMmDv4uk9I2KvlNLDnQhWPaXVz56fF4/XVRpSSr+IiBuA57YtOvWyVsfOB8h5+O+mlKYBIuIfySumnQX8QaMdnXlXJ9xCTZ1XsSTSntSvaW+4X6FRLbwWnlbHTsVfkpfqem1KyTGz+LQyfvYEfh34C/J/wpuAieK5S9l+4bMWtlY/e24mz67WXmAY5GtutPC1OnZWADdVEneAlNIW4CbycpMNmbyrE64GToiIvaraTgYeA67fyX4HRMTRlYaIWAUcXDynha/VsUNEfAgYAU5NKX23cyGqh7UyfqbIdafV25uL5z4M/F5nQlWPafWz50pyon5cpSEilgJHsP1LoBa2VsfOXcDzi1JPACLiacDzgTtnOmEU60pKbVPcsOBHwH+Q/xR9MHlW6/+mlD5a1e924PqU0jur2q4BnkP+k1HlJk33p5S8SdMi0OrYiYi3AGuBi4ELag57x84u/tHCMJfPnprjHEReovaklNKVHQ5bPWCO/29dQV515oPAg+Q1vp8LPMfSvYVvDv9vHQH8ALgW+Az5S+B7gOOBVSmlhl/+nHlX2xUfVi8FdgG+CZxDLmf4RE3XXYs+1U4hf1P9PLl2+QbgdZ2MV71jDmPn5cXjacD3a7ZXdS5i9ZI5fvZoEZvj2DmVfCfwvyBfgDgNvMTEfXFodeyklG4AXgHsBVxCznn2AF42U+IOzrxLkiRJpeHMuyRJklQSJu+SJElSSZi8S5IkSSVh8i5JkiSVhMm7JEmSVBIm75IkSVJJmLxLUglFxLERkSLi7Jr2FBHXdSeq5kTE2UWcx3Y7ll4TEddFhGs4S2rI5F2SBEBEnFYk1ad1OxZJUn27djsASVJbHQY82u0gJEmdYfIuSQtISumWbscgSeocy2YkaQbVteURcXxEfC8iHo2In0bE30TEspr+BxX9L46IlRFxdUQ8FBGbqvosi4jzIuLHEfF4RDwQEZdGxLPrnD8i4n0RcVvR99aIeO8M8dateY+I5RGxJiLuiIgnIuL+iPjHiDi5eP5i4AtF9y8Ux0kRcWfNcQ6NiC9HxL3Fce6MiD+LiL3qnHOviPjriLiveM++HxEvnfENr/+aTine9wcj4rGImIyIr0fEEVV9tpX8RMTJEbGx6Ht3RPzviHhazTGrf68viYjrI+LhiNhY8579TUTcVbzWeyPiwog4oE6M74iIb0bET4q+P42Ir0bEigav6dkR8Y3inA9FxBUR8RuzfW8kLT7OvEtSc44CPgR8A7geOAY4E1gVEatTSltq+h8CfBdYD1wI7AsQEfsV7c8B/r443nLgDcDLIuLIlNKPq47zJ8V57wI+DewBfBL4XrOBR8TzgH8C9ivOeRmwD7CqeA1fBa4AlgGvLWKqJLEPVR3naOBq8v8d3wAmgZXAWcCxEXF0SumJou8uwLeK92m8OP/BwFXF+9ds7CPAXwN3AF8BHiG/X8cBRwI31OzyJuAlwKXANcCJ5PfvecVrq3U08OGi72eK10ZEHFrEuS9wJXAb8GzgdPLv6bdTSg9WHedTwA+L4/yc/Pt9PXBCRByRUrqj6jUNAOvIv4+/A35cxPEvxb6S1FhKyc3Nzc2twQYcC6Ri+/2a5z5XtL+vqu2gqv4fqnO8S4GtwOtr2geBaeCqqrZDgSfJyd3eNe2PFuc4u+Y4Cbiupu2HRfsb68SzvOrfpxX9TqvTbzfgJ8CDwCE1z72/2G+0qu1dRdulQFS1v7Xq/Tm2iff/34B7gD1q2pcA+9SJPQEvrmrfBbi2aH9tg9/rm+uc9wfA48CRNe2vK/b5TE37QXWOcQzwS+BzNe1ri2O8u8F4St0e925ubr27WTYjSc25BfhyTdsnyMn1qXX6/zfwf6obiln3NwL/L6X09ernUkrrybPfJ0TE0qL5zeQkdSyl9IuqvrcCX2om6Ih4IfA/gG+llC6vfT6ldE8zxwFOAgaAT6aUbqt57q+A+4GTq9p+j5yIfiylVL304SXk93I2niAnwduklLamlDbV6fvtlNL1Vf2eBD5e/Fjv97QhpfSV6oaiHGcIOD+l9P2a8/4dsIE8w1/dfmftgVNK/wLcBGwrFSrKd95A/kvKhTW7VMaTJDVk2YwkNWddTRJKSunuiLgLOLxO/4mU0nRN2ypyMr531KzPXnhm8fwh5ASxctzv1osHGG4i7lXF47VN9J3JYPF4eIPYf0n+i0DF4cD9tYl+SilFxPeAurXgdXwV+FPgxoi4lFx+84OU0uMN+td7r9aT/6pR7/e0oU5b5bUe2OC17gE8IyL2TUXpTFHb/hHgxcABQF9V/+qSqkOBpxWvYYdEvWo8HVznnJIEmLxLUrMeaNB+P3BwRDwtFfXeVe21nl48HldsjexZPO49w7nrHb+eyiz+vU32b6QS+9ub7L83cHOD55qNHWCMXHf/bvIM+seBRyPiEuADKaWHa/o/5b1KKW2NiJ+x/f3cWSyV1/q6YmtkT+DBInFfD+xOvqbgcmCKogQJeFbVPjP9TivxmLxLasjkXZKas1+D9v2BLTWJO+TErVal9OXslNI5TZyz0n8/4Gd1ztuMygWnz2yy/85iOba6LGUn/Wd6z5pS/LXjAuCCYpWX48j19MPkGfC31uzylHNGxBLgGcB/1TtFnbbKaz0tpfTFJsL8A2Avcu38pTXnPrmmb/XvtJ6m3xtJi5M175LUnKMiIqobIuLXgQOBf2/yGBvIyeILm+xfOe7RdZ5b3eQx/rV4fHkTfStlHLvUeW598Tib2PePiEOqG4v38Mgmj7GDlNJ9RX36y8kXsZ5Up1u992qQXMbS7O9ptq+1MlN+ZXVjROzPU2fRbyXX8L+wWJGnun9lPElSQybvktScw3jqBY/nkP+CubaZA6SU/hv4GvCKiHhH7fMRsWtEVCfllZVpRiNi76p+h/LUGedG51xPXrHlVRHxu3XOubzqx8oyhctr+5GXhrwb+HBEvKDOcZbWtK8FAji35kvP75Pfy6ZExMtrk1ygnzzrXvvXDsgX/L6oav9dgD+uimmnUkrj5C9ap0fECXVi2j0ihqqaJovHo6r69JEv5N2t5thPkMfAs4Azag5dGU+S1JAfEpLUnL8HPhsRJwG3Ay8iz35vIK8P3qx3A88FPhcRw+SZ8SfIydwxwCaKizlTSrdExHnkdcr/PSK+Rk5a3wxcR17DvBmnFv0vj4hrycn8UuB/Ao+xvf6+sjzi+yPi6eRlIR9KKX0qpfR4RLyJvM77hoi4hlzTvjvwG+SlF79EXjce4PPkLxgnAwdFRGWd998hv5cvazL2y4CHI+K75KUq9wReQ16n/oN1+l8NXFtc3Hof+T06nLzCzxVNnhPgLeSLY68pYp8gT3gdRL4o9QfAK4q+FwHvAL5enHeKvMLMnsV+K2uO/UHgeODTEXE8eTb+GPL7eCPwW7OIU9Ii48y7JDXne8CryMsl/iF51ZALgJelp96gqaFidZIh8oWXu5OTvjOA55NvavSeml0+Upzvl8B7yTcg+ij5Qs5mz3kzcARwPnnW+3+Rlyt8hKovHimln5GT7f8k15SfS74BU+X575OXnbyI/AVkpOh/ALCGPNNc6fsk8EryzYsOBt5HLgl5JbO4wRT5i8tG8hel95GX2ryDvE7+eXX6Xwa8rYjzD8k3WfpTdlzGcqeKVXJeQF7uczn5S9fbgN8kL3f5saq+68lj40fAKeTEvxLzQ9RIKU2Sy3u+SS4Beg/5moZj8CZNknYialY+kyRViYhjyTOw56SUzu5uNGokIk4DvgC8PaV0cXejkaTOceZdkiRJKgmTd0mSJKkkTN4lSZKkkrDmXZIkSSoJZ94lSZKkkjB5lyRJkkrC5F2SJEkqCZN3SZIkqSRM3iVJkqSSMHmXJEmSSuL/A4IWC0Ba2AiEAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 864x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors2 = {'incorrect':'red', 'correct':'green', 'no bet': 'white'}\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(12,10))\n",
    "for k in colors2.keys():\n",
    "    ix = np.where(test_pred_df.loc[test_pred_df[\"bet_accuracy\"]!=\"no bet\",'bet_accuracy'] == k)[0]\n",
    "    ax.scatter(test_pred_df.loc[test_pred_df[\"bet_accuracy\"]!=\"no bet\",\"pred_spread\"].values[ix],\n",
    "               test_pred_df.loc[test_pred_df[\"bet_accuracy\"]!=\"no bet\",\"B365_bookspread\"].values[ix],\n",
    "               c = colors2[k], label = k, s = 30)\n",
    "ax.legend(prop={'size': 19})\n",
    "ax.set_xlabel('predicted spread', fontsize=19)\n",
    "ax.set_ylabel(\"bookmaker's spread\", fontsize=19)\n",
    "ax.xaxis.set_tick_params(labelsize=15)\n",
    "ax.yaxis.set_tick_params(labelsize=15)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "match_df = pd.read_csv(dir + \"Data/match_clean.csv\")\n",
    "\n",
    "match_df = match_df.drop('Unnamed: 0',axis=1)\n",
    "\n",
    "match_df = match_df.loc[match_df[\"match_id\"].isin(set(test_pred_df[\"match_id\"]))]\n",
    "\n",
    "match_df[\"result\"] = \"BWD\"\n",
    "match_df.loc[match_df[\"home_team_goal\"]> match_df[\"away_team_goal\"], \"result\"] = \"BWH\"\n",
    "match_df.loc[match_df[\"home_team_goal\"]< match_df[\"away_team_goal\"], \"result\"] = \"BWA\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Series([], Name: result, dtype: int64)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bet_lose = test_pred_df.loc[(test_pred_df[\"pred_spread_bins\"].isin(bet_lose_bins)),\n",
    "                             [\"team_id\", \"match_id\", \"home\", \"result\"]]\n",
    "\n",
    "bet_lose[\"result\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Series([], Name: result, dtype: float64)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bet_lose[\"result\"].value_counts()/len(bet_lose)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "nan"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "book_l_df = match_df.loc[match_df[\"match_id\"].isin(set(bet_lose[\"match_id\"])), \n",
    "                         [\"result\", \"BWH\", \"BWD\", \"BWA\", \"match_id\"]]\n",
    "\n",
    "book_l_df[\"pred\"] = book_l_df[['BWH','BWD', 'BWA']].idxmin(axis=1)\n",
    "\n",
    "np.mean(book_l_df[\"pred\"]==book_l_df[\"result\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bet win     1655\n",
       "no bet       591\n",
       "bet draw     222\n",
       "Name: bet, dtype: int64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_pred_df[\"bet\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "correct      990\n",
       "incorrect    887\n",
       "no bet       591\n",
       "Name: bet_accuracy, dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_pred_df[\"bet_accuracy\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6531034482758621"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "947/(947+503)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "compare_matches = set(test_pred_df.loc[test_pred_df[\"bet\"].isin([\"bet lose\",\n",
    "                                                                 \"bet win\",\n",
    "                                                                 \"bet draw\"]), \"match_id\"])\n",
    "\n",
    "follow_book_bet = match_df.loc[match_df[\"match_id\"].isin(compare_matches)]\n",
    "\n",
    "follow_book_bet = follow_book_bet[[\"match_id\", \"result\", 'BWH','BWD', 'BWA']]\n",
    "\n",
    "follow_book_bet[\"pred\"] = follow_book_bet[['BWH','BWD', 'BWA']].idxmin(axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5466169419286094"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(follow_book_bet[\"pred\"]==follow_book_bet[\"result\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "943.48"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "follow_book_bet[\"dollar\"] = follow_book_bet[\"BWD\"]\n",
    "\n",
    "follow_book_bet[\"dollar\"] = np.where(follow_book_bet[\"result\"]==\"BWH\", \n",
    "                                     follow_book_bet[\"BWH\"],\n",
    "                                     follow_book_bet[\"dollar\"])\n",
    "\n",
    "\n",
    "follow_book_bet[\"dollar\"] = np.where(follow_book_bet[\"result\"]==\"BWA\", \n",
    "                                     follow_book_bet[\"BWA\"],\n",
    "                                     follow_book_bet[\"dollar\"])\n",
    "\n",
    "follow_book_bet[\"dollar\"] = np.where(follow_book_bet[\"pred\"]!=follow_book_bet[\"result\"],\n",
    "                                    -1, follow_book_bet[\"dollar\"])\n",
    "\n",
    "\n",
    "\n",
    "follow_book_bet[\"dollar\"].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1877"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(follow_book_bet)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1075.4300000000005"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "894.9600000000006+180.46999999999997"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5466169419286094"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(follow_book_bet[\"result\"]==follow_book_bet[\"pred\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "correct      990\n",
       "incorrect    887\n",
       "no bet       591\n",
       "Name: bet_accuracy, dtype: int64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_pred_df[\"bet_accuracy\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['lose_pred', 'win_pred', 'sum_pred', 'league_id', 'season', 'date',\n",
       "       'stage', 'match_id', 'team_id', 'm_rating', 'win', 'lose',\n",
       "       'rating_diff', 'home', 'cumulative_win', 'cumulative_lose', 'match_num',\n",
       "       'match_bin', 'draw', 'avg_overall_rating', 'avg_volleys_imp',\n",
       "       'avg_long_passing', 'avg_finishing', 'avg_ball_control',\n",
       "       'avg_dribbling', 'avg_short_passing', 'avg_reactions', 'avg_vision_imp',\n",
       "       'avg_penalties', 'avg_agility_imp', 'result', 'cumulative_result',\n",
       "       'B365_bookspread', 'pred_spread', 'pred_spread_bins', 'bet',\n",
       "       'bet_accuracy'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_pred_df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "odds_col = match_df.columns[32:63].tolist()\n",
    "\n",
    "dollar = 0\n",
    "for t, m, h, r in zip(bet_lose[\"team_id\"], bet_lose[\"match_id\"], bet_lose[\"home\"], bet_lose[\"result\"]):\n",
    "\n",
    "    if h == 0:\n",
    "        bwh = match_df.loc[(match_df[\"away_team_id\"]==t) & \n",
    "                           (match_df[\"match_id\"]==m), \"BWH\"].values[0]\n",
    "        if np.isnan(bwh):\n",
    "            continue\n",
    "            \n",
    "        if r == -1:\n",
    "            dollar += bwh\n",
    "        elif r == 0:\n",
    "            dollar -= 1\n",
    "        elif r == 1:\n",
    "            dollar -= 1\n",
    "            \n",
    "    elif h == 1:\n",
    "        bwa = match_df.loc[(match_df[\"home_team_id\"]==t) & \n",
    "                               (match_df[\"match_id\"]==m),  \"BWA\"].values[0]  \n",
    "        \n",
    "        if np.isnan(bwa):\n",
    "            continue\n",
    "        \n",
    "        if r == -1:\n",
    "            dollar -= 1\n",
    "        elif r == 0:\n",
    "            dollar -= 1\n",
    "        elif r == 1:\n",
    "            dollar += bwa        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dollar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       " 1    936\n",
       " 0    386\n",
       "-1    333\n",
       "Name: result, dtype: int64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bet_win = test_pred_df.loc[(test_pred_df[\"pred_spread_bins\"].isin(bet_win_bins)), \n",
    "                           [\"team_id\", \"match_id\", \"home\", \"result\"]]\n",
    "\n",
    "bet_win[\"result\"].value_counts()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5673716012084592"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "book_w_df = match_df.loc[match_df[\"match_id\"].isin(set(bet_win[\"match_id\"])), \n",
    "                         [\"result\", \"BWH\", \"BWD\", \"BWA\", \"match_id\"]]\n",
    "\n",
    "book_w_df[\"pred\"] = book_w_df[['BWH','BWD', 'BWA']].idxmin(axis=1)\n",
    "\n",
    "np.mean(book_w_df[\"pred\"]==book_w_df[\"result\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "odds_col = match_df.columns[32:63].tolist()\n",
    "\n",
    "dollar = 0\n",
    "\n",
    "for t, m, h, r in zip(bet_win[\"team_id\"], bet_win[\"match_id\"], bet_win[\"home\"], bet_win[\"result\"]):\n",
    "    \n",
    "    if h == 0:\n",
    "        bwa = match_df.loc[(match_df[\"away_team_id\"]==t) & \n",
    "                           (match_df[\"match_id\"]==m), \"BWA\"].values[0]\n",
    "        \n",
    "        if r == -1:\n",
    "            dollar -= 1\n",
    "        elif r == 0:  \n",
    "            dollar -= 1\n",
    "        elif r == 1:\n",
    "            dollar += bwa\n",
    "        \n",
    "    elif h == 1:\n",
    "        bwh = match_df.loc[(match_df[\"home_team_id\"]==t) & \n",
    "                           (match_df[\"match_id\"]==m),  \"BWH\"].values[0]\n",
    "        \n",
    "        if r == -1:\n",
    "            dollar -= 1\n",
    "        elif r == 0:\n",
    "            dollar -= 1\n",
    "        elif r == 1:\n",
    "            dollar += bwh        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "894.9600000000006"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dollar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-1    96\n",
       " 1    72\n",
       " 0    54\n",
       "Name: result, dtype: int64"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bet_draw = test_pred_df.loc[(test_pred_df[\"pred_spread_bins\"].isin(bet_draw_bins)), \n",
    "                           [\"team_id\", \"match_id\", \"home\", \"result\"]]\n",
    "\n",
    "bet_draw[\"result\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.3918918918918919"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "book_d_df = match_df.loc[match_df[\"match_id\"].isin(set(bet_draw[\"match_id\"])), \n",
    "                         [\"result\", \"BWH\", \"BWD\", \"BWA\", \"match_id\"]]\n",
    "\n",
    "book_d_df[\"pred\"] = book_d_df[['BWH','BWD', 'BWA']].idxmin(axis=1)\n",
    "\n",
    "np.mean(book_d_df[\"pred\"]==book_d_df[\"result\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "odds_col = match_df.columns[32:63].tolist()\n",
    "\n",
    "dollar = 0\n",
    "\n",
    "for t, m, h, r in zip(bet_win[\"team_id\"], bet_win[\"match_id\"], bet_win[\"home\"], bet_win[\"result\"]):\n",
    "    \n",
    "    if h == 0:\n",
    "        bwd = match_df.loc[(match_df[\"away_team_id\"]==t) & \n",
    "                           (match_df[\"match_id\"]==m), \"BWD\"].values[0]\n",
    "        \n",
    "        if r == -1:\n",
    "            dollar -= 1\n",
    "        elif r == 0:  \n",
    "            dollar += bwd\n",
    "        elif r == 1:\n",
    "            dollar -= 1\n",
    "        \n",
    "    elif h == 1:\n",
    "        bwd = match_df.loc[(match_df[\"home_team_id\"]==t) & \n",
    "                           (match_df[\"match_id\"]==m),  \"BWD\"].values[0]\n",
    "        \n",
    "        if r == -1:\n",
    "            dollar -= 1\n",
    "        elif r == 0:\n",
    "            dollar += bwd\n",
    "        elif r == 1:\n",
    "            dollar -= 1        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "180.46999999999997"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dollar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.09048767967145789"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(132+185+211.81)/ (1877+1956+2011)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.07032498668087374"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "132/1877"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.09458077709611452"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "185/1956"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.10532570860268524"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "211.81/2011"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}