{ "cells": [ { "cell_type": "markdown", "id": "c6202cb5", "metadata": {}, "source": [ "\n", "# Регрессия двумя нейросетями: фиксированная архитектура и эволюционный поиск архитектуры\n", "\n", "В работе автоматически определяется зависимая переменная, выполняется предобработка данных и сравниваются две нейросетевые регрессионные модели:\n", "\n", "1. **Классическая нейросеть** с заранее заданной архитектурой.\n", "2. **Нейросеть с архитектурой, найденной методом дифференциальной эволюции**.\n", "\n", "Оценка качества выполняется на валидационной и тестовой выборках. В конце ноутбука выводится сводная таблица метрик и вся информация, необходимая для краткого итогового вывода.\n" ] }, { "cell_type": "code", "execution_count": 3, "id": "274ab92f", "metadata": {}, "outputs": [], "source": [ "\n", "import os\n", "import re\n", "import glob\n", "import math\n", "import random\n", "import warnings\n", "from pathlib import Path\n", "\n", "import numpy as np\n", "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "\n", "from scipy.optimize import differential_evolution\n", "from scipy.stats import skew\n", "\n", "from sklearn.compose import ColumnTransformer\n", "from sklearn.impute import SimpleImputer\n", "from sklearn.metrics import r2_score, mean_absolute_error, mean_squared_error\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.neural_network import MLPRegressor\n", "from sklearn.pipeline import Pipeline\n", "from sklearn.preprocessing import OneHotEncoder, StandardScaler\n", "from sklearn.exceptions import ConvergenceWarning\n", "\n", "warnings.filterwarnings('ignore', category=FutureWarning)\n", "warnings.filterwarnings('ignore', category=ConvergenceWarning)\n", "\n", "RANDOM_STATE = 42\n", "np.random.seed(RANDOM_STATE)\n", "random.seed(RANDOM_STATE)\n" ] }, { "cell_type": "code", "execution_count": 4, "id": "7aeeaac9", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Используемый файл: housing_price_dataset.csv\n", "Форма датасета: (50000, 6)\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
SquareFeetBedroomsBathroomsNeighborhoodYearBuiltPrice
0212641Rural1969215355.283618
1245932Rural1980195014.221626
2186021Suburb1970306891.012076
3229421Urban1996206786.787153
4213052Suburb2001272436.239065
\n", "
" ], "text/plain": [ " SquareFeet Bedrooms Bathrooms Neighborhood YearBuilt Price\n", "0 2126 4 1 Rural 1969 215355.283618\n", "1 2459 3 2 Rural 1980 195014.221626\n", "2 1860 2 1 Suburb 1970 306891.012076\n", "3 2294 2 1 Urban 1996 206786.787153\n", "4 2130 5 2 Suburb 2001 272436.239065" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Типы данных:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
dtype
SquareFeetint64
Bedroomsint64
Bathroomsint64
Neighborhoodobject
YearBuiltint64
Pricefloat64
\n", "
" ], "text/plain": [ " dtype\n", "SquareFeet int64\n", "Bedrooms int64\n", "Bathrooms int64\n", "Neighborhood object\n", "YearBuilt int64\n", "Price float64" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "# Поиск CSV-файла\n", "candidate_names = [\n", " 'housing_price_dataset.csv',\n", " 'Housing Price Prediction Data.csv',\n", " 'Housing Price Prediction.csv',\n", " 'housing price prediction data.csv',\n", " 'housing.csv',\n", "]\n", "\n", "search_dirs = [Path('.'), Path('/mnt/data')]\n", "found_path = None\n", "\n", "for directory in search_dirs:\n", " for name in candidate_names:\n", " path = directory / name\n", " if path.exists():\n", " found_path = path\n", " break\n", " if found_path is not None:\n", " break\n", "\n", "if found_path is None:\n", " csv_files = []\n", " for directory in search_dirs:\n", " csv_files.extend(sorted(directory.glob('*.csv')))\n", " if len(csv_files) == 1:\n", " found_path = csv_files[0]\n", " elif len(csv_files) > 1:\n", " # Предпочитаем файлы, в имени которых есть слова housing/price\n", " preferred = [p for p in csv_files if ('housing' in p.name.lower() or 'price' in p.name.lower())]\n", " found_path = preferred[0] if preferred else csv_files[0]\n", "\n", "if found_path is None:\n", " raise FileNotFoundError('CSV-файл не найден. Поместите housing_price_dataset.csv в папку с ноутбуком.')\n", "\n", "print(f'Используемый файл: {found_path}')\n", "df_raw = pd.read_csv(found_path)\n", "print('Форма датасета:', df_raw.shape)\n", "display(df_raw.head())\n", "print('Типы данных:')\n", "display(df_raw.dtypes.to_frame('dtype'))\n" ] }, { "cell_type": "code", "execution_count": 5, "id": "8091fa9b", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Названия колонок после нормализации:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
original_namenormalized_name
0SquareFeetsquarefeet
1Bedroomsbedrooms
2Bathroomsbathrooms
3Neighborhoodneighborhood
4YearBuiltyearbuilt
5Priceprice
\n", "
" ], "text/plain": [ " original_name normalized_name\n", "0 SquareFeet squarefeet\n", "1 Bedrooms bedrooms\n", "2 Bathrooms bathrooms\n", "3 Neighborhood neighborhood\n", "4 YearBuilt yearbuilt\n", "5 Price price" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "# Нормализация названий столбцов\n", "\n", "def normalize_column_name(name: str) -> str:\n", " name = str(name).strip().lower()\n", " name = name.replace('%', ' pct ')\n", " name = name.replace('&', ' and ')\n", " name = re.sub(r'[^a-z0-9]+', '_', name)\n", " name = re.sub(r'_+', '_', name).strip('_')\n", " return name\n", "\n", "original_columns = df_raw.columns.tolist()\n", "normalized_columns = [normalize_column_name(col) for col in original_columns]\n", "\n", "rename_map = dict(zip(original_columns, normalized_columns))\n", "df = df_raw.rename(columns=rename_map).copy()\n", "\n", "column_mapping_df = pd.DataFrame({\n", " 'original_name': original_columns,\n", " 'normalized_name': normalized_columns,\n", "})\n", "\n", "print('Названия колонок после нормализации:')\n", "display(column_mapping_df)\n" ] }, { "cell_type": "code", "execution_count": 6, "id": "2ea7ae98", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Выбранная зависимая переменная: price\n" ] } ], "source": [ "\n", "# Автоматический выбор зависимой переменной\n", "\n", "def choose_target_column(dataframe: pd.DataFrame) -> str:\n", " numeric_cols = dataframe.select_dtypes(include=[np.number]).columns.tolist()\n", " preferred_patterns = [\n", " 'price', 'sale_price', 'selling_price', 'amount', 'cost', 'revenue',\n", " 'target', 'label', 'value', 'profit'\n", " ]\n", "\n", " for pattern in preferred_patterns:\n", " matches = [c for c in numeric_cols if pattern in c]\n", " if matches:\n", " return matches[0]\n", "\n", " # Если явный шаблон не найден, выбираем числовой столбец с наибольшей вариативностью,\n", " # исключая признаки, похожие на идентификаторы.\n", " candidate_cols = []\n", " for col in numeric_cols:\n", " nunique_ratio = dataframe[col].nunique(dropna=True) / max(len(dataframe), 1)\n", " if 'id' in col and nunique_ratio > 0.8:\n", " continue\n", " candidate_cols.append(col)\n", "\n", " if not candidate_cols:\n", " raise ValueError('Не удалось автоматически определить зависимую переменную.')\n", "\n", " variances = dataframe[candidate_cols].var(numeric_only=True).sort_values(ascending=False)\n", " return variances.index[0]\n", "\n", "TARGET_COL = choose_target_column(df)\n", "print('Выбранная зависимая переменная:', TARGET_COL)\n" ] }, { "cell_type": "code", "execution_count": 7, "id": "250e66a9", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Найденные идентификаторы / несодержательные поля: []\n", "Размер матрицы признаков: (50000, 5)\n", "Число исходных признаков: 5\n", "Пропуски в целевой переменной: 0\n", "Форма после удаления пропусков в цели:\n", "X: (50000, 5) y: (50000,)\n" ] } ], "source": [ "\n", "# Выделение признаков и удаление явных идентификаторов / несодержательных полей\n", "\n", "def detect_id_like_columns(dataframe: pd.DataFrame, target_col: str):\n", " result = []\n", " n = len(dataframe)\n", " for col in dataframe.columns:\n", " if col == target_col:\n", " continue\n", " nunique = dataframe[col].nunique(dropna=True)\n", " ratio = nunique / max(n, 1)\n", " is_object = dataframe[col].dtype == 'object'\n", " name_flag = any(token in col for token in ['id', 'name'])\n", " if name_flag and ratio > 0.5:\n", " result.append(col)\n", " elif is_object and ratio > 0.95 and any(token in col for token in ['name', 'address', 'email']):\n", " result.append(col)\n", " return sorted(set(result))\n", "\n", "id_like_columns = detect_id_like_columns(df, TARGET_COL)\n", "print('Найденные идентификаторы / несодержательные поля:', id_like_columns)\n", "\n", "data = df.drop(columns=id_like_columns).copy()\n", "\n", "X = data.drop(columns=[TARGET_COL]).copy()\n", "y = data[TARGET_COL].copy()\n", "\n", "print('Размер матрицы признаков:', X.shape)\n", "print('Число исходных признаков:', X.shape[1])\n", "print('Пропуски в целевой переменной:', int(y.isna().sum()))\n", "\n", "# Удаляем строки без целевого значения\n", "mask = y.notna()\n", "X = X.loc[mask].reset_index(drop=True)\n", "y = y.loc[mask].reset_index(drop=True)\n", "\n", "print('Форма после удаления пропусков в цели:')\n", "print('X:', X.shape, 'y:', y.shape)\n" ] }, { "cell_type": "code", "execution_count": 8, "id": "33389528", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Размеры выборок:\n", "train: (32000, 5)\n", "val : (8000, 5)\n", "test : (10000, 5)\n" ] } ], "source": [ "\n", "# Разделение на выборки\n", "X_train_full, X_test, y_train_full, y_test = train_test_split(\n", " X, y, test_size=0.20, random_state=RANDOM_STATE\n", ")\n", "\n", "X_train, X_val, y_train, y_val = train_test_split(\n", " X_train_full, y_train_full, test_size=0.20, random_state=RANDOM_STATE\n", ")\n", "\n", "print('Размеры выборок:')\n", "print('train:', X_train.shape)\n", "print('val :', X_val.shape)\n", "print('test :', X_test.shape)\n" ] }, { "cell_type": "code", "execution_count": 9, "id": "604c68d1", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Числовые признаки: ['squarefeet', 'bedrooms', 'bathrooms', 'yearbuilt']\n", "Категориальные признаки: ['neighborhood']\n", "Число признаков после кодирования: 7\n", "Первые признаки после кодирования:\n", "['num__squarefeet', 'num__bedrooms', 'num__bathrooms', 'num__yearbuilt', 'cat__neighborhood_Rural', 'cat__neighborhood_Suburb', 'cat__neighborhood_Urban']\n" ] } ], "source": [ "\n", "# Предобработка признаков\n", "numeric_features = X_train.select_dtypes(include=[np.number]).columns.tolist()\n", "categorical_features = [c for c in X_train.columns if c not in numeric_features]\n", "\n", "print('Числовые признаки:', numeric_features)\n", "print('Категориальные признаки:', categorical_features)\n", "\n", "try:\n", " ohe = OneHotEncoder(handle_unknown='ignore', sparse_output=False)\n", "except TypeError:\n", " ohe = OneHotEncoder(handle_unknown='ignore', sparse=False)\n", "\n", "numeric_transformer = Pipeline([\n", " ('imputer', SimpleImputer(strategy='median')),\n", " ('scaler', StandardScaler()),\n", "])\n", "\n", "categorical_transformer = Pipeline([\n", " ('imputer', SimpleImputer(strategy='most_frequent')),\n", " ('onehot', ohe),\n", "])\n", "\n", "preprocessor = ColumnTransformer([\n", " ('num', numeric_transformer, numeric_features),\n", " ('cat', categorical_transformer, categorical_features),\n", "], remainder='drop')\n", "\n", "X_train_proc = preprocessor.fit_transform(X_train)\n", "X_val_proc = preprocessor.transform(X_val)\n", "X_test_proc = preprocessor.transform(X_test)\n", "X_train_full_proc = preprocessor.transform(X_train_full)\n", "\n", "feature_names_processed = preprocessor.get_feature_names_out().tolist()\n", "\n", "print('Число признаков после кодирования:', len(feature_names_processed))\n", "print('Первые признаки после кодирования:')\n", "print(feature_names_processed[:20])\n" ] }, { "cell_type": "code", "execution_count": 10, "id": "356c3585", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Использовано log1p-преобразование цели: False\n" ] } ], "source": [ "\n", "# Преобразование целевой переменной для более устойчивого обучения\n", "use_log1p_target = bool((y_train > 0).all() and abs(float(skew(y_train))) > 0.75)\n", "\n", "if use_log1p_target:\n", " y_train_base = np.log1p(y_train.values.astype(float))\n", " y_val_base = np.log1p(y_val.values.astype(float))\n", " y_test_base = np.log1p(y_test.values.astype(float))\n", " y_train_full_base = np.log1p(y_train_full.values.astype(float))\n", "else:\n", " y_train_base = y_train.values.astype(float)\n", " y_val_base = y_val.values.astype(float)\n", " y_test_base = y_test.values.astype(float)\n", " y_train_full_base = y_train_full.values.astype(float)\n", "\n", "y_scaler = StandardScaler()\n", "y_train_scaled = y_scaler.fit_transform(y_train_base.reshape(-1, 1)).ravel()\n", "y_val_scaled = y_scaler.transform(y_val_base.reshape(-1, 1)).ravel()\n", "y_test_scaled = y_scaler.transform(y_test_base.reshape(-1, 1)).ravel()\n", "y_train_full_scaled = y_scaler.transform(y_train_full_base.reshape(-1, 1)).ravel()\n", "\n", "print('Использовано log1p-преобразование цели:', use_log1p_target)\n" ] }, { "cell_type": "code", "execution_count": 11, "id": "be58189a", "metadata": {}, "outputs": [], "source": [ "\n", "# Вспомогательные функции\n", "\n", "def inverse_target_transform(y_scaled_pred):\n", " y_base_pred = y_scaler.inverse_transform(np.asarray(y_scaled_pred).reshape(-1, 1)).ravel()\n", " if use_log1p_target:\n", " y_pred = np.expm1(y_base_pred)\n", " else:\n", " y_pred = y_base_pred\n", " return y_pred\n", "\n", "\n", "def regression_metrics(y_true, y_pred):\n", " return {\n", " 'r2': float(r2_score(y_true, y_pred)),\n", " 'rmse': float(np.sqrt(mean_squared_error(y_true, y_pred))),\n", " 'mae': float(mean_absolute_error(y_true, y_pred)),\n", " }\n", "\n", "\n", "def evaluate_regressor(model, X_tr, y_tr_scaled, X_va, y_va_true, X_te, y_te_true):\n", " model.fit(X_tr, y_tr_scaled)\n", " pred_val = inverse_target_transform(model.predict(X_va))\n", " pred_test = inverse_target_transform(model.predict(X_te))\n", " metrics_val = regression_metrics(y_va_true, pred_val)\n", " metrics_test = regression_metrics(y_te_true, pred_test)\n", " return model, pred_val, pred_test, metrics_val, metrics_test\n" ] }, { "cell_type": "markdown", "id": "82dd00ca", "metadata": {}, "source": [ "\n", "## 1. Базовая нейросеть с заранее заданной архитектурой\n" ] }, { "cell_type": "code", "execution_count": 12, "id": "b2fdda72", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Метрики базовой нейросети:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
modelr2_testrmse_testmae_testr2_valrmse_valmae_valn_original_featuresn_processed_featureshidden_layers
0Классическая нейросеть0.57400249449.05878539483.8251530.57263849821.04593539964.57365457(64, 32)
\n", "
" ], "text/plain": [ " model r2_test rmse_test mae_test r2_val \\\n", "0 Классическая нейросеть 0.574002 49449.058785 39483.825153 0.572638 \n", "\n", " rmse_val mae_val n_original_features n_processed_features \\\n", "0 49821.045935 39964.573654 5 7 \n", "\n", " hidden_layers \n", "0 (64, 32) " ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "# Фиксированная архитектура\n", "fixed_hidden_layers = (64, 32)\n", "\n", "baseline_model = MLPRegressor(\n", " hidden_layer_sizes=fixed_hidden_layers,\n", " activation='relu',\n", " solver='adam',\n", " alpha=1e-4,\n", " batch_size='auto',\n", " learning_rate_init=1e-3,\n", " max_iter=500,\n", " early_stopping=True,\n", " validation_fraction=0.15,\n", " n_iter_no_change=20,\n", " random_state=RANDOM_STATE,\n", ")\n", "\n", "baseline_model, baseline_pred_val, baseline_pred_test, baseline_metrics_val, baseline_metrics_test = evaluate_regressor(\n", " baseline_model,\n", " X_train_proc, y_train_scaled,\n", " X_val_proc, y_val.values.astype(float),\n", " X_test_proc, y_test.values.astype(float)\n", ")\n", "\n", "baseline_results = {\n", " 'model': 'Классическая нейросеть',\n", " 'r2_test': baseline_metrics_test['r2'],\n", " 'rmse_test': baseline_metrics_test['rmse'],\n", " 'mae_test': baseline_metrics_test['mae'],\n", " 'r2_val': baseline_metrics_val['r2'],\n", " 'rmse_val': baseline_metrics_val['rmse'],\n", " 'mae_val': baseline_metrics_val['mae'],\n", " 'n_original_features': X.shape[1],\n", " 'n_processed_features': len(feature_names_processed),\n", " 'hidden_layers': str(fixed_hidden_layers),\n", "}\n", "\n", "baseline_history = pd.DataFrame({\n", " 'epoch': np.arange(1, len(baseline_model.loss_curve_) + 1),\n", " 'train_loss': baseline_model.loss_curve_,\n", "})\n", "\n", "print('Метрики базовой нейросети:')\n", "display(pd.DataFrame([baseline_results]))\n" ] }, { "cell_type": "markdown", "id": "21bf5dfb", "metadata": {}, "source": [ "\n", "## 2. Поиск архитектуры нейросети методом дифференциальной эволюции\n", "\n", "Оптимизируются:\n", "- число скрытых слоёв;\n", "- размеры скрытых слоёв;\n", "- коэффициент регуляризации `alpha`;\n", "- шаг обучения `learning_rate_init`.\n" ] }, { "cell_type": "code", "execution_count": 13, "id": "3db75211", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Размер подвыборки для эволюционного поиска:\n", "train_search: (4000, 7)\n", "val_search : (1500, 7)\n" ] } ], "source": [ "\n", "# Подвыборки для ускорения эволюционного поиска\n", "search_train_size = min(len(X_train_proc), 4000)\n", "search_val_size = min(len(X_val_proc), 1500)\n", "\n", "rng = np.random.default_rng(RANDOM_STATE)\n", "search_train_idx = rng.choice(len(X_train_proc), size=search_train_size, replace=False)\n", "search_val_idx = rng.choice(len(X_val_proc), size=search_val_size, replace=False)\n", "\n", "X_search_train = X_train_proc[search_train_idx]\n", "y_search_train = y_train_scaled[search_train_idx]\n", "X_search_val = X_val_proc[search_val_idx]\n", "y_search_val_true = y_val.values.astype(float)[search_val_idx]\n", "\n", "print('Размер подвыборки для эволюционного поиска:')\n", "print('train_search:', X_search_train.shape)\n", "print('val_search :', X_search_val.shape)\n" ] }, { "cell_type": "code", "execution_count": 14, "id": "267ad3d0", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Лучшая найденная архитектура:\n", "{'hidden_layers': (80, 96), 'alpha': 1.8274889467365341e-06, 'learning_rate_init': 0.00042874921144929587}\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
hidden_layersalphalearning_rate_initrmse_valmae_valr2_val
0(80, 96)0.0000020.00042949520.45809139914.3894770.570645
1(128,)0.0000930.00019249584.79976840046.1962760.569528
2(120,)0.0000460.00111249650.09369539974.6919250.568394
3(64,)0.0002860.00428149655.42698539866.6043770.568301
4(88,)0.0001630.00010149671.96210739922.9692430.568013
5(112, 16)0.0084730.00023749693.86517440039.1385920.567632
6(64,)0.0000040.00029549699.37875439982.0231500.567536
7(64,)0.0000130.00153849703.25761939942.6915780.567469
8(120, 96, 8, 56)0.0013320.00071149712.48979540074.5857690.567308
9(112,)0.0007750.00020649712.82364140067.8064040.567302
10(120, 32)0.0000040.00010649724.37702039948.5485670.567101
11(64, 72, 56)0.0063610.00365449725.06206640046.0394630.567089
12(80,)0.0004630.00041349746.87447940093.4085460.566709
13(72, 32)0.0000040.00028449756.83219540060.5631690.566536
14(112, 32, 72)0.0000070.00022449766.75818140050.7114950.566363
\n", "
" ], "text/plain": [ " hidden_layers alpha learning_rate_init rmse_val \\\n", "0 (80, 96) 0.000002 0.000429 49520.458091 \n", "1 (128,) 0.000093 0.000192 49584.799768 \n", "2 (120,) 0.000046 0.001112 49650.093695 \n", "3 (64,) 0.000286 0.004281 49655.426985 \n", "4 (88,) 0.000163 0.000101 49671.962107 \n", "5 (112, 16) 0.008473 0.000237 49693.865174 \n", "6 (64,) 0.000004 0.000295 49699.378754 \n", "7 (64,) 0.000013 0.001538 49703.257619 \n", "8 (120, 96, 8, 56) 0.001332 0.000711 49712.489795 \n", "9 (112,) 0.000775 0.000206 49712.823641 \n", "10 (120, 32) 0.000004 0.000106 49724.377020 \n", "11 (64, 72, 56) 0.006361 0.003654 49725.062066 \n", "12 (80,) 0.000463 0.000413 49746.874479 \n", "13 (72, 32) 0.000004 0.000284 49756.832195 \n", "14 (112, 32, 72) 0.000007 0.000224 49766.758181 \n", "\n", " mae_val r2_val \n", "0 39914.389477 0.570645 \n", "1 40046.196276 0.569528 \n", "2 39974.691925 0.568394 \n", "3 39866.604377 0.568301 \n", "4 39922.969243 0.568013 \n", "5 40039.138592 0.567632 \n", "6 39982.023150 0.567536 \n", "7 39942.691578 0.567469 \n", "8 40074.585769 0.567308 \n", "9 40067.806404 0.567302 \n", "10 39948.548567 0.567101 \n", "11 40046.039463 0.567089 \n", "12 40093.408546 0.566709 \n", "13 40060.563169 0.566536 \n", "14 40050.711495 0.566363 " ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "# Дифференциальная эволюция для поиска архитектуры\n", "search_history = []\n", "\n", "\n", "def vector_to_config(x):\n", " n_layers = int(np.clip(np.round(x[0]), 1, 4))\n", " units_raw = []\n", " for val in x[1:5]:\n", " units_raw.append(int(np.clip(np.round(val / 8) * 8, 8, 256)))\n", " hidden_layers = tuple(units_raw[:n_layers])\n", " alpha = 10 ** float(x[5])\n", " lr = 10 ** float(x[6])\n", " return {\n", " 'hidden_layers': hidden_layers,\n", " 'alpha': alpha,\n", " 'learning_rate_init': lr,\n", " }\n", "\n", "\n", "def de_objective(x):\n", " cfg = vector_to_config(x)\n", " model = MLPRegressor(\n", " hidden_layer_sizes=cfg['hidden_layers'],\n", " activation='relu',\n", " solver='adam',\n", " alpha=cfg['alpha'],\n", " learning_rate_init=cfg['learning_rate_init'],\n", " batch_size='auto',\n", " max_iter=220,\n", " early_stopping=True,\n", " validation_fraction=0.15,\n", " n_iter_no_change=15,\n", " random_state=RANDOM_STATE,\n", " )\n", " try:\n", " model.fit(X_search_train, y_search_train)\n", " pred_val = inverse_target_transform(model.predict(X_search_val))\n", " rmse_val = float(np.sqrt(mean_squared_error(y_search_val_true, pred_val)))\n", " mae_val = float(mean_absolute_error(y_search_val_true, pred_val))\n", " r2_val = float(r2_score(y_search_val_true, pred_val))\n", " except Exception:\n", " rmse_val = 1e9\n", " mae_val = 1e9\n", " r2_val = -1e9\n", "\n", " search_history.append({\n", " 'hidden_layers': str(cfg['hidden_layers']),\n", " 'alpha': cfg['alpha'],\n", " 'learning_rate_init': cfg['learning_rate_init'],\n", " 'rmse_val': rmse_val,\n", " 'mae_val': mae_val,\n", " 'r2_val': r2_val,\n", " })\n", " return rmse_val\n", "\n", "bounds = [\n", " (1, 4), # число слоев\n", " (16, 128), # units 1\n", " (8, 128), # units 2\n", " (8, 128), # units 3\n", " (8, 128), # units 4\n", " (-6, -2), # log10(alpha)\n", " (-4, -2), # log10(lr)\n", "]\n", "\n", "maxiter_de = 7 if len(X_train_proc) <= 10000 else 5\n", "popsize_de = 6 if len(X_train_proc) <= 10000 else 5\n", "\n", "result_de = differential_evolution(\n", " de_objective,\n", " bounds=bounds,\n", " strategy='best1bin',\n", " maxiter=maxiter_de,\n", " popsize=popsize_de,\n", " tol=0.01,\n", " mutation=(0.5, 1.0),\n", " recombination=0.7,\n", " seed=RANDOM_STATE,\n", " polish=False,\n", " updating='deferred',\n", " workers=1,\n", ")\n", "\n", "best_config = vector_to_config(result_de.x)\n", "print('Лучшая найденная архитектура:')\n", "print(best_config)\n", "\n", "search_history_df = pd.DataFrame(search_history)\n", "search_history_df = search_history_df.sort_values(['rmse_val', 'mae_val'], ascending=[True, True]).reset_index(drop=True)\n", "\n", "display(search_history_df.head(15))\n" ] }, { "cell_type": "code", "execution_count": 15, "id": "652e54a8", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Метрики нейросети с найденной архитектурой:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
modelr2_testrmse_testmae_testr2_valrmse_valmae_valn_original_featuresn_processed_featureshidden_layersalphalearning_rate_init
0Нейросеть с архитектурой, найденной DE0.57320149495.53441839527.8663140.57299549800.2600139955.01779157(80, 96)0.0000020.000429
\n", "
" ], "text/plain": [ " model r2_test rmse_test \\\n", "0 Нейросеть с архитектурой, найденной DE 0.573201 49495.534418 \n", "\n", " mae_test r2_val rmse_val mae_val n_original_features \\\n", "0 39527.866314 0.572995 49800.26001 39955.017791 5 \n", "\n", " n_processed_features hidden_layers alpha learning_rate_init \n", "0 7 (80, 96) 0.000002 0.000429 " ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "# Финальная модель с найденной архитектурой\n", "searched_model = MLPRegressor(\n", " hidden_layer_sizes=best_config['hidden_layers'],\n", " activation='relu',\n", " solver='adam',\n", " alpha=best_config['alpha'],\n", " learning_rate_init=best_config['learning_rate_init'],\n", " batch_size='auto',\n", " max_iter=700,\n", " early_stopping=True,\n", " validation_fraction=0.15,\n", " n_iter_no_change=25,\n", " random_state=RANDOM_STATE,\n", ")\n", "\n", "searched_model, searched_pred_val, searched_pred_test, searched_metrics_val, searched_metrics_test = evaluate_regressor(\n", " searched_model,\n", " X_train_proc, y_train_scaled,\n", " X_val_proc, y_val.values.astype(float),\n", " X_test_proc, y_test.values.astype(float)\n", ")\n", "\n", "searched_results = {\n", " 'model': 'Нейросеть с архитектурой, найденной DE',\n", " 'r2_test': searched_metrics_test['r2'],\n", " 'rmse_test': searched_metrics_test['rmse'],\n", " 'mae_test': searched_metrics_test['mae'],\n", " 'r2_val': searched_metrics_val['r2'],\n", " 'rmse_val': searched_metrics_val['rmse'],\n", " 'mae_val': searched_metrics_val['mae'],\n", " 'n_original_features': X.shape[1],\n", " 'n_processed_features': len(feature_names_processed),\n", " 'hidden_layers': str(best_config['hidden_layers']),\n", " 'alpha': best_config['alpha'],\n", " 'learning_rate_init': best_config['learning_rate_init'],\n", "}\n", "\n", "searched_history = pd.DataFrame({\n", " 'epoch': np.arange(1, len(searched_model.loss_curve_) + 1),\n", " 'train_loss': searched_model.loss_curve_,\n", "})\n", "\n", "print('Метрики нейросети с найденной архитектурой:')\n", "display(pd.DataFrame([searched_results]))\n" ] }, { "cell_type": "code", "execution_count": 16, "id": "73713364", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Сводные метрики моделей:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
modelr2_testrmse_testmae_testr2_valrmse_valmae_valn_original_featuresn_processed_featureshidden_layersalphalearning_rate_init
0Классическая нейросеть0.57400249449.05878539483.8251530.57263849821.04593539964.57365457(64, 32)NaNNaN
1Нейросеть с архитектурой, найденной DE0.57320149495.53441839527.8663140.57299549800.26001039955.01779157(80, 96)0.0000020.000429
\n", "
" ], "text/plain": [ " model r2_test rmse_test \\\n", "0 Классическая нейросеть 0.574002 49449.058785 \n", "1 Нейросеть с архитектурой, найденной DE 0.573201 49495.534418 \n", "\n", " mae_test r2_val rmse_val mae_val n_original_features \\\n", "0 39483.825153 0.572638 49821.045935 39964.573654 5 \n", "1 39527.866314 0.572995 49800.260010 39955.017791 5 \n", "\n", " n_processed_features hidden_layers alpha learning_rate_init \n", "0 7 (64, 32) NaN NaN \n", "1 7 (80, 96) 0.000002 0.000429 " ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Лучшая модель:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
modelr2_testrmse_testmae_testr2_valrmse_valmae_valn_original_featuresn_processed_featureshidden_layersalphalearning_rate_init
0Классическая нейросеть0.57400249449.05878539483.8251530.57263849821.04593539964.57365457(64, 32)NaNNaN
\n", "
" ], "text/plain": [ " model r2_test rmse_test mae_test r2_val \\\n", "0 Классическая нейросеть 0.574002 49449.058785 39483.825153 0.572638 \n", "\n", " rmse_val mae_val n_original_features n_processed_features \\\n", "0 49821.045935 39964.573654 5 7 \n", "\n", " hidden_layers alpha learning_rate_init \n", "0 (64, 32) NaN NaN " ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "# Сводное сравнение моделей\n", "results_df = pd.DataFrame([\n", " baseline_results,\n", " searched_results,\n", "]).sort_values(['r2_test', 'rmse_test', 'mae_test'], ascending=[False, True, True]).reset_index(drop=True)\n", "\n", "best_model_row = results_df.iloc[[0]].copy()\n", "\n", "print('Сводные метрики моделей:')\n", "display(results_df)\n", "\n", "print('Лучшая модель:')\n", "display(best_model_row)\n" ] }, { "cell_type": "code", "execution_count": 17, "id": "da001667", "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABW0AAAHqCAYAAAB/bWzAAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAPYQAAD2EBqD+naQABAABJREFUeJzs3XecVOX1+PHPvXfa9l3K0osURbFjL8EKKia2aGLFXmJijRj9xhpr7D1iQWw/lZhorFiw9wII0vuysL1Nv+35/XF3hh22sCAyC5z364XuzNy58+zM7My5557nPJpSSiGEEEIIIYQQQgghhBCiS9CzPQAhhBBCCCGEEEIIIYQQa0jSVgghhBBCCCGEEEIIIboQSdoKIYQQQgghhBBCCCFEFyJJWyGEEEIIIYQQQgghhOhCJGkrhBBCCCGEEEIIIYQQXYgkbYUQQgghhBBCCCGEEKILkaStEEIIIYQQQgghhBBCdCGStBVCCCGEEEIIIYQQQoguRJK2QgghhBBCCCGEEEII0YVI0lYIIcRm5+GHH6ahoSF9+f777ycajWZvQEIIIYTIMHv2bF577bX05RkzZvDWW29lb0BCdEG33XYbrusC4Lout99+e5ZH1LY333yTGTNmpC+/9tpr/Pzzz9kb0Ca2cuVKnnnmmfTlZcuW8cILL2RvQGKrIUlbIcSv7plnnkHTNL7//vtWtz3xxBNomsaxxx6L4zhZGJ3YHL3xxhvceOONlJWV8cILL3DdddeRk5OT7WEJIYQQm42O4jOAgw46iB133HGD9x8Oh7ngggv4+uuvWbhwIZdeeimzZs3a4P0JsSWaPHkyd999NytXruSee+5h8uTJ2R5Sm2bNmsWll17KwoUL+frrr7nwwgsJh8PZHtYmo2kaF198MVOnTmXZsmVMmDCBzz77rFP3vfHGG9E0Lf0vNzeXgQMH8tvf/pZJkyaRTCZb3efMM8/MuE/Lf6FQaGP/eqIL82V7AEKIrdd///tfLrroIg488EBeeuklDMPI9pDEZuLaa6/ld7/7HQ888AC6rnPPPfeg63IeUgghhOgq9t133/Q/gG233Zbzzjsvy6MSomu5+eabOeOMM7j66qsJBoM8//zz2R5Sm84991yefvpptt12WwCOP/549tlnnyyPatPp168f5513HkcccQQAffr04eOPP16vfTz22GPk5+eTTCYpLy9n6tSpnH322dx///28+eabDBgwIGP7YDDIk08+2Wo/csy8dZGkrRAiKz7++GNOPvlkdthhB9544w05YyjWy+jRo1m+fDlz585lwIAB9O/fP9tDEkIIIcRaXnvtNebMmUM8HmennXYiEAhke0hCdCl/+MMfOPjgg1m0aBHDhw+nZ8+e2R5Sm3r27Mns2bOZPXs2ubm5bL/99tke0iZ3//3385e//IWamhp23HFH8vLy1uv+v//97+nRo0f68vXXX88LL7zAGWecwYknnsjXX3+dsb3P5+O0007bKGMXmy8pSxJCbHIzZszgmGOOoU+fPkydOpWioqJW2yxbtqzdKSEt3X333ey33350796dnJwcRo0axb///e82H/f5559nr732Ijc3l5KSEn7zm9/w3nvvZWzzzjvvMHr0aAoKCigsLGTPPffkxRdfTN9+0EEHcdBBB2Xc57vvvms1to8//hhN01qdgT3zzDMZPHhwxnWu63L//fczcuRIQqEQvXr14oILLqC+vr7V77Ah47v11lvRdT1ju88++4wTTzyRgQMHEgwGGTBgAJdffjnxeLzVY/773/9mjz32oKCgION1uPvuu1tt21J70y5ramrQNI0bb7wx4/ry8nLOPvtsevXqRTAYZOTIkTz99NMZ27R8XouLi9l3333p378/48aNa7XP1FSkefPmcdJJJ1FYWEj37t259NJLSSQSGfu1bZt//OMfDB06lGAwyODBg7n22mvbnK60rtcA4JtvvuGII46gqKiI3NxcRo8ezRdffNFqbB39S713fun0VCGEEGJjev755xk1ahQ5OTl069aNP/7xj5SVlWVs0zIe2WGHHRg1ahQzZ85sM5bTNI0///nPvPDCC2y33XaEQiFGjRrFp59+2uqxp0+fzpFHHklhYSH5+fkceuihrRIdAA0NDVx++eUMHjyYYDBI//79OeOMM6ipqUlvk0wmueGGGxg2bFg6FpowYULGd/+6vqtTv2MqPmkvBl2XRCLBjTfeyLbbbksoFKJPnz4cf/zxLF68uMP7vf7664wbN46+ffsSDAYZOnQo//jHP1q1HEvFEj/88AP77bcfOTk5bLPNNvzrX//K2G78+PGEQiHmzp2bcf3YsWMpKSlh1apVwJoYb9myZeltXNdl5513RtO0dO/PjqZ4p/613Mc777zDgQceSF5eHgUFBYwbN65V39S2Yunnn38eXde544471utxx48fT48ePbAsq9VzO2bMGLbbbrv05Y39Pm35HJaWlqaPZ9Z+DtdXapxrO/roo1s9b509jkrF2MFgkFGjRrH99ttz1113ZfwNtNRenNvWtuuKmVvur+XfL8D333/f6rk688wzyc/Pb/U4//73v9s8NpsyZUr686xHjx6cdtpplJeXZ2zT8j03dOhQ9t57b+rq6sjJyWn1Hl5fp556Kueeey7ffPMN77///gbvR2y5pNJWCLFJLV68mCOOOIJgMMjUqVPp06dPh9uff/75HHjggQD85z//4b///W/G7Q888AC/+93vOPXUUzFNk5deeokTTzyRN998k3HjxqW3u+mmm7jxxhvZb7/9uPnmmwkEAnzzzTdMmzaNMWPGAF7wdPbZZzNy5EiuueYaiouLmT59Ou+++y6nnHJKu2O8+uqrN/TpAOCCCy7gmWee4ayzzuKSSy5h6dKlPPzww0yfPp0vvvgCv9+/weObNGkSf//737nnnnsytpkyZQqxWIyLLrqI7t278+233/LQQw+xcuVKpkyZkt7uq6++4qSTTmKXXXbhjjvuoKioiJqaGi6//PJf9DuvrbKykn322ScdaPbs2ZN33nmHc845h6amJi677LJ27/vpp5/y9ttvt3v7SSedxODBg7n99tv5+uuvefDBB6mvr+fZZ59Nb3PuuecyefJkfv/733PllVfyzTffcPvttzN37tyM91xnXoNp06Zx5JFHMmrUKG644QZ0XWfSpEkccsghfPbZZ+y1114cf/zxDBs2LL3fyy+/nO23357zzz8/fd3WWMEghBBi02tsbGyVDAHaTGTdeuutXHfddZx00kmce+65VFdX89BDD/Gb3/yG6dOnU1xc3O7jdBQvffLJJ7z88stccsklBINBHn30UY444gi+/fbb9InLn3/+mQMPPJDCwkImTJiA3+/n8ccf56CDDuKTTz5h7733BiASiXDggQcyd+5czj77bHbffXdqamr43//+x8qVK+nRoweu6/K73/2Ozz//nPPPP5/tt9+eWbNmcd9997FgwYL0AmrPPfdceoyfffYZEydO5L777ktXy/Xq1Wudz++6OI7D0UcfzYcffsgf//hHLr30UsLhMO+//z6zZ89m6NCh7d73mWeeIT8/nyuuuIL8/HymTZvG9ddfT1NTE3fddVfGtvX19Rx11FGcdNJJnHzyybzyyitcdNFFBAIBzj77bMCLq6dNm8b48eP56quvMAyDxx9/nPfee4/nnnuOvn37tjuW5557rlXP4gsuuIDDDjssffn000/nuOOO4/jjj09fl6osfe655xg/fjxjx47lzjvvJBaL8dhjj3HAAQcwffr0VgnHlPfee4+zzz6bP//5z/ztb39br8c9/fTTefbZZ5k6dSpHH310+raKigqmTZvGDTfckPFYG/N92tnn8NfU2eOotTU0NHRqwbRUOwCAa665ptXtnYmZf02p468999yT22+/ncrKSh544AG++OKLdX6eXX/99a2KQDbU6aefzsSJE3nvvfc4/PDDM25r67M5EAhQWFi4UR5bbAaUEEL8yiZNmqQA9eabb6qhQ4cqQI0ZM6bD+yxcuFABavLkyenrbrjhBrX2x1YsFsu4bJqm2nHHHdUhhxySsS9d19Vxxx2nHMfJ2N51XaWUUg0NDaqgoEDtvffeKh6Pt7mNUkqNHj1ajR49On357bffVoA64ogjMsb2ySefKEBNmzYtY1/jx49XgwYNSl/+7LPPFKBeeOGFjO3efffdjOs3ZHxvvfWW8vl86sorr1RrW/t5U0qp22+/XWmappYvX56+7pprrlGAWr16dfq6pUuXKkDdddddrfbRUup1/+677zKur66uVoC64YYb0tedc845qk+fPqqmpiZj2z/+8Y+qqKgoPd6PPvpIAeqjjz5Kb7P33nurI488stU+U++X3/3udxn7/NOf/qQANXPmTKWUUjNmzFCAOvfcczO2++tf/5rxGnbmNXBdVw0fPlyNHTs243WJxWJqm222UYcffnibz9WgQYPU+PHj27xt9OjRauTIkW3eJoQQQmyo1Pd0R/9afv8sW7ZMGYahbr311oz9zJo1S/l8vozrOxsvKaXSj/X999+nr1u+fLkKhULquOOOS1937LHHqkAgoBYvXpy+btWqVaqgoED95je/SV93/fXXK0D95z//afU7p76bn3vuOaXruvrss88ybv/Xv/6lAPXFF1+0+3wtXbq01W2p+GTKlCmtbluXp59+WgHq3nvvbXe87WkrnrvgggtUbm6uSiQS6etGjx6tAHXPPfekr0smk2rXXXdVpaWlyjTN9PVTp05VgLrlllvUkiVLVH5+vjr22GMzHmPt5yKRSKiBAwem47FJkya1Od61Y7WUcDisiouL1XnnnZdxfUVFhSoqKsq4vmUs/f3336v8/Hx14okntorxO/O4juOo/v37qz/84Q8Z1997771K0zS1ZMmSjH1szPfphj6H6wKoiy++uNX148aNyzgGUapzx1GpfbZ8/iZMmKBKS0vVqFGjMv7OU6699loFZMT1I0eOzNh2fWLmVExfXV2d8Tjfffddq+dq/PjxKi8vr9WYpkyZknEMYZqmKi0tVTvuuGNGXP/mm28qQF1//fUZ+2z53M2ePVvpup5+rdr6TGipvfGn1NfXKyDjfTR+/Ph2P5fHjh3b4eOJLYu0RxBCbDJnnnkmZWVlnHLKKbz33nsZFZ1rM00T8BqwdyQnJyf9c319PY2NjRx44IH8+OOP6etfe+01XNfl+uuvb7VYVWqK3vvvv084HOZvf/tbq/66a0/jS1FKcc0113DCCSe0OmteWloKwMqVKzsc/5QpUygqKuLwww+npqYm/W/UqFHk5+fz0UcfbdD4vv32W0466SROOOGEVpUWkPm8RaNRampq2G+//VBKMX369PRt4XAYXdc7PNO8LqkKntS/urq6jNuVUrz66qv89re/RSmVse3YsWNpbGzMeD1b+s9//sN3332Xng7Xlosvvjjj8l/+8heAdHVu6v9XXHFFxnZXXnklAG+99RbQuddgxowZLFy4kFNOOYXa2tr07xGNRjn00EP59NNPcV23/SerHY7jpPeV+tsQQgghNoZHHnmE999/v9W/nXfeOWO7//znP7iuy0knnZTxXd27d2+GDx+ejlnW1lG8lLLvvvsyatSo9OWBAwdyzDHHMHXqVBzHwXEc3nvvPY499liGDBmS3q5Pnz6ccsopfP755zQ1NQHw6quvsssuu3Dccce1epzU9/WUKVPYfvvtGTFiRMbvcsghhwC0+7usSzgcpqamhoaGhk7f59VXX6VHjx7p+KSt8banZTyXeuwDDzyQWCzGvHnzMrb1+XxccMEF6cuBQIALLriAqqoqfvjhh/T1Y8aM4YILLuDmm2/m+OOPJxQK8fjjj3c4jkceeYTa2tpWlamd9f7779PQ0MDJJ5+c8XoYhsHee+/d5uuxZMkSxo0bx6677spzzz23QQvS6rrOqaeeyv/+9z/C4XD6+hdeeIH99tuPbbbZJmP7jfk+XdsvfQ5bSiQSGc9jTU1Nm5XznTmOWlt5eTkPPfQQ1113XZttCFKPD3S4ZsmGxMx1dXUZv1NjY2O7+1/792/5+oLXWqGqqoo//elPGeMcN24cI0aMSMf/bbnmmmvYfffdOfHEE9vdZn2knse1xxgKhdr8bO7ouEdseaQ9ghBik6mrq+Oll17iuOOOY86cOVx66aWMGTOmzZ62qWC3vWAg5c033+SWW25hxowZrXqQpSxevBhd19lhhx3a3U+qZ9j69A594YUX+Pnnn3nllVda9TQdMmQIvXv35u6772aXXXZJTydbu0fqwoULaWxsTCd511ZVVbXe4ysvL2fcuHFEo1Fqa2vbDPhXrFjB9ddfz//+979WvXNbBkD77rsvDz/8MJdeeikTJkygqKiozV67HWk5Pa0t1dXVNDQ0MHHiRCZOnNjmNqnnoSXHcbj22ms59dRTWx1YtjR8+PCMy0OHDkXX9XT/qeXLl6Preka7AoDevXtTXFzM8uXLgc69BgsXLgS8nnDtaWxspKSkpN3b2zJv3rz09MHUWG+44YYO23YIIYQQnbHXXnuxxx57tLq+pKQkY2ruwoULUUq1+l5NSbVzWltH8VJKW/vcdttticViVFdXAxCLxTJ6jKZsv/32uK5LWVkZI0eOZPHixZxwwgltPk7L32Xu3LntLvrUVtzRGak2A+DFsL/97W+57777OmyjsHjxYrbbbjt8vvU/NP/555/5+9//zrRp01olA9dOaPXt27fVwknbbrst4K0lsc8++6Svv/vuu3n99deZMWMGL774YrtxaupxbrvtNq644ooNbheRip9SSfO1rT0VPBqNMnbsWCorK+nevfs6k9sdOeOMM7jzzjv573//yxlnnMH8+fP54YcfWvX7hY37Pm1pYzyHLT311FM89dRTra4fNGhQxuXOHEet7YYbbqBv375ccMEF7fZwrqmpwe/3k5ub2+5+NiRmbut5bUs0Gl3ngm6p+L6tfY4YMYLPP/+8zft9/vnnvPHGG3z44YesWLGiU+NZl0gkAkBBQUHG9YZhrPM4Smz5JGkrhNhk7rrrrvQZyYkTJ7LPPvtwzTXX8Oijj7batqKiAvASZ+357LPP+N3vfsdvfvMbHn30Ufr06YPf72fSpEntHhRsLKZpct1113HOOeekA96WAoEATzzxBKeccgq77LJLxm0tAybXdSktLeWFF15o83E2ZAXZRYsWsfvuu3Pfffdx+umnM3ny5IyAyHEcDj/8cOrq6rj66qsZMWIEeXl5lJeXc+aZZ2ac1f7jH//Ijz/+yEMPPdRuQnVdHnnkkYznqKmpKeNgKvV4p512WruBW1tJ2aeeeoply5YxderU9RpPe4HoLwn4U1K/y1133cWuu+7a5jbrOhHRlsGDB/PEE08AUFtby4MPPsjpp5/OkCFDMg6yhBBCiF+L67pomsY777yDYRitbm/r+21d8VK2uK7LTjvtxL333tvm7QMGDNig/V5//fUceOCBWJbFDz/8wM0330xDQ0OHvfc3VENDA6NHj6awsJCbb76ZoUOHEgqF+PHHH7n66qs3aGZPyvTp09OJ61mzZnHyySe3u+2dd96JrutcddVV1NbWbtDjpcb63HPPtRn7r53QrqmpIS8vjzfeeINjjz2W22+/fYMrVFML5T3//POcccYZPP/88wQCAU466aQN2t+G2BjPYUvHHHNMq8XI/v73v6ePr2DDjqPmzp3LM888w/PPP9/uSRrwTgIMHDiww9h6Q2LmV199NSOBv2DBglYz6sCrUH3jjTcyrvvss8+4+eab2x1PZ1199dWMHTuWQw45ZIMXi1vb7NmzAVoVkAgBkrQVQmxCv/nNb9I/77nnnlx88cU88sgjnHHGGa0ST3PmzEHTtA7PqL766quEQiGmTp2a0UZh0qRJGdsNHToU13WZM2dOu0FBapGH2bNnd+oL89FHH6Wqqoobb7yx3W2OPvpoysvL+emnn4jH44AXmMyfPz/jcT/44AP233//jClKv2R8ffr04e2336ZXr168/vrrXHnllRx11FHpBPCsWbNYsGABkydP5owzzkjfr60VS3Vd5+6772bWrFksXbqURx99lMrKSk477bQOx9DS2hU8azfU79mzJwUFBTiO0+mzybFYjJtuuok//elPraoG1rZw4cKM6W2LFi3Cdd30ghaDBg3CdV0WLlyYsfhXZWUlDQ0N6f135jVIbVNYWLhRz4zn5eVl7O/AAw+kX79+vPfee5K0FUIIsUkMHToUpRTbbLNNpxOwnYmXYE3VXUsLFiwgNzc3Hb/k5uZmxFAp8+bNQ9f1dKJ16NCh6SRIR7/LzJkzOfTQQzfKSduUnXbaKf19feSRR7JixQomT56MbdvtVtIOHTqUb775BsuyOkyEre3jjz+mtraW//znPxkx9tKlS9vcftWqVUSj0Yxq2wULFgBkLPIVjUY566yz2GGHHdhvv/345z//yXHHHceee+7Z5j4feOABbr/9dgoKCjY44ZiKn0pLSzsVP+Xm5vLuu+8yYsQILr/8cm677TZOOumkDV7E9YwzzuCKK65g9erVvPjii4wbN67NWVEb832asrGew5b69+/f6nm8//77M5K2nT2Oaumaa65h11135Q9/+EO729i2zcyZMzniiCM6HOOGxMy/+c1v0osAAu22b2urQnXtliWp+H7+/PmtKrznz5/f5vHFa6+9xldffdVh+4gNkVrwcOzYsRt1v2LLID1thRBZc+utt9KnTx/OP/98bNtOX2/bNq+++ip77bVXh1WJhmGgaRqO46SvW7ZsWXrF35Rjjz0WXde5+eabW1UdKKUAr39XQUEBt99+e6uVQFPbpITDYW699VYuv/zyDiuBwZvmsv/++3PYYYdx2GGH0adPn4zbTzrpJBzH4R//+Eer+9q2nQ4w1md82267bXpq1UMPPYTrulx66aXp21PVMS3vp5TigQceaPN3eOihh5g2bRovvPAChx12GPvvv3+Hv/P6MgyDE044gVdffbXNg6zUdLOWHnjgAaLRKP/3f/+3zv0/8sgjGZcfeughwDuYAjjqqKMAL5htKVV9k1o9tzOvwahRoxg6dCh33313eqrTun6XDZF6H7dV6SSEEEL8Go4//ngMw+Cmm25qFXsopVolm9YnXlo7EVJWVsbrr7/OmDFjMAwDwzAYM2YMr7/+erq9EXgnWF988UUOOOCAdAXeCSecwMyZM/nvf//b6nFS4z7ppJMoLy9Pz2JpKR6PE41GO34yOsl1XXRd7zAxfMIJJ1BTU8PDDz/c7njb0lY8Z5pmmzPYwIsrW/amNU2Txx9/nJ49e2b0ab366qvTyeZ7772XwYMHM378+FYtvgBuuukmevXqxYUXXtjuODtj7NixFBYWctttt7XZe3Xt+Klnz56MGDECgJtvvpn+/ftz3nnndfh8deTkk09G0zQuvfRSlixZ0m5xwsZ8n6ZsrOdwfXX2OCrlq6++4vXXX+eOO+7o8P383nvv0djYyDHHHNPh42+qmLk9e+yxB6WlpfzrX//KeG+/8847zJ07Nx3/p6Tasp1yyintFgFtiBdffJEnn3ySfffdl0MPPXSj7VdsOaTSVgiRNQUFBTz00EMcf/zx3HPPPVx99dV88MEHXHfddfz000+tprWsbdy4cdx7770cccQRnHLKKVRVVfHII48wbNgwfvrpp/R2w4YN4//+7//4xz/+wYEHHsjxxx9PMBjku+++o2/fvtx+++0UFhZy3333ce6557LnnntyyimnUFJSwsyZM4nFYkyePDm9vx9//JEePXowYcKEX/wcjB49mgsuuIDbb7+dGTNmMGbMGPx+PwsXLmTKlCk88MAD/P73v1+v8bXUu3dv7rrrLs4991xOO+00jjrqKEaMGMHQoUP561//Snl5OYWFhbz66qtt9qr9+eefmTBhAjfeeGObFRYbyx133MFHH33E3nvvzXnnnccOO+xAXV0dP/74Ix988EGrxcvee+89br31Vrp3777OfS9dupTf/e53HHHEEXz11Vc8//zzGW0rdtllF8aPH8/EiRPTUw2//fZbJk+ezLHHHsvBBx8M0KnXQNd1nnzySY488khGjhzJWWedRb9+/SgvL+ejjz6isLBwne/rtkQiEd59913A6w394IMP4vf7WwWUQgghxK9l6NCh3HLLLVxzzTUsW7aMY489loKCApYuXcp///tfzj//fP7617+mt1+feGnHHXdk7NixXHLJJQSDwXTi8aabbkpvc8stt/D+++9zwAEH8Kc//Qmfz8fjjz9OMpnkn//8Z3q7q666in//+9+ceOKJnH322YwaNYq6ujr+97//8a9//YtddtmF008/nVdeeYULL7yQjz76iP333x/HcZg3bx6vvPIKU6dObbPP77rMmDGD/Px8bNvmhx9+4Nlnn+WYY47p8CTrGWecwbPPPssVV1zBt99+y4EHHkg0GuWDDz7gT3/6U7vJr/3224+SkhLGjx/PJZdcgqZpPPfcc+0mLvv27cudd97JsmXL2HbbbXn55ZeZMWMGEydOTFf4Tps2jUcffZQbbriB3XffHfAqLw866CCuu+66jOcZvHjshRdeIBAIrPdz1VJhYSGPPfYYp59+Orvvvjt//OMf6dmzJytWrOCtt95i//33bzOpDd5iWhMnTuSwww7jscce409/+tN6P37Pnj054ogjmDJlCsXFxe3GVxvzfZrSmedw2bJlbLPNNowfP36jTcnv7HFUy3EefvjhHVbFvvzyy/z1r38lGAwSj8d5/vnn07c1NjbiOA6vvfZauqDm14iZO8vv93PnnXdy1llnMXr0aE4++WQqKyt54IEHGDx4MJdffnnG9itXriQQCPyiVif//ve/yc/PxzRNysvLmTp1Kl988QW77LJLmwt027ad8Ry2dNxxx7XqUS22UEoIIX5lkyZNUoD67rvv2rz9mGOOUbm5uWrJkiXqL3/5i/rNb36j3n333Vbb3XDDDWrtj62nnnpKDR8+XAWDQTVixAg1adKkNrdTSqmnn35a7bbbbioYDKqSkhI1evRo9f7772ds87///U/tt99+KicnRxUWFqq99tpL/b//9//St48ePVoB6r777lvn2Noyfvx4NWjQoFbXT5w4UY0aNUrl5OSogoICtdNOO6kJEyaoVatWrff4Ro8e3Wr/hxxyiBo4cKAKh8NKKaXmzJmjDjvsMJWfn6969OihzjvvPDVz5kwFqEmTJimllEokEmrnnXdWBxxwgLJtO72vpUuXKkDdddddHf6u7b3u1dXVClA33HBDxvWVlZXq4osvVgMGDFB+v1/17t1bHXrooWrixInpbT766CMFqD59+qhoNJpx/7X3mXpN5syZo37/+9+rgoICVVJSov785z+reDyecV/LstRNN92kttlmG+X3+9WAAQPUNddcoxKJRKvfa12vgVJKTZ8+XR1//PGqe/fuKhgMqkGDBqmTTjpJffjhh20+V4MGDVLjx49v87bUey71r7i4WO2///7qnXfeaXN7IYQQojPWFZ+NHj1ajRw5stX1r776qjrggANUXl6eysvLUyNGjFAXX3yxmj9/fsZ9OxsvAeriiy9Wzz//fDqm22233dRHH33U6rF//PFHNXbsWJWfn69yc3PVwQcfrL788stW29XW1qo///nPql+/fioQCKj+/fur8ePHq5qamvQ2pmmqO++8U40cOTIdG44aNUrddNNNqrGxsd3na+nSpa1uS8UnqX8+n08NGjRIXXLJJaq+vr7V9muLxWLq//7v/9JxSO/evdXvf/97tXjx4g7v98UXX6h99tlH5eTkqL59+6oJEyaoqVOnKiDj+Uu9lt9//73ad999VSgUUoMGDVIPP/xwepumpiY1aNAgtfvuuyvLsjIe5/LLL1e6rquvvvoq47nYddddleu66e1SMWIqllxbW/FfSx999JEaO3asKioqUqFQSA0dOlSdeeaZ6vvvv09v014sfdZZZ6nCwkK1cuXK9X5cpZR65ZVXFKDOP//8dse+Md+n6/Mczpo1SwHqb3/7W4e/Q8txrm3cuHGtnrfOHkcBStM09cMPP2Rcv/Zxx6BBgzL+Dtr6t/YYOhMzp8ZUXV2dcd/vvvuu1XM1fvx4lZeX1+r3nzJlSqu/C6WUevnll9PHh926dVOnnnpqq/fQ+PHjFaAuvfTSjOs7+kxoKTX+1L9QKKT69++vjj76aPX000+3ebyResz2/q3rMcWWQ1NqA+cQCCGEEF3YjTfeyE033UR1dXVG/yshhBBCdB2apnHxxRe3W0kpfrmDDjqImpqadfb63Zq9/vrrHHvssXz66acceOCBrW7P5vv00UcfZcKECSxevDjdAq0rGjx4MDfeeCNnnnlmm7d//PHHnHnmmRmtI4QQHZOetkIIIYQQQgghhNhqPfHEEwwZMoQDDjgg20Np5aOPPuKSSy7p0glbIcSvQ3raCiGEEEIIIYQQYqvz0ksv8dNPP/HWW2/xwAMPdLjIVra01e+0KzruuOMYOnRou7f36tWL4447bhOOSIjNnyRthRBCCCGEEEIIsdU5+eSTyc/P55xzztmgRczEGvfdd1+Ht2+//fbr3EYIkUl62gohhBBCCCGEEEIIIUQXIj1thRBCCCGEEEIIIYQQoguRpK0QQgghhBBCCCGEEEJ0IdLTdhNyXZdVq1ZRUFDQJRucCyGEEEJ0danOXoWFhRJPZYHEs0IIIYQQv4xSinA4TN++fdH19utpJWm7Ca1atYoBAwZkexhCCCGEEJu9xsZGCgsLsz2MrY7Es0IIIYQQG0dZWRn9+/dv93ZJ2m5CBQUFgPeiyEGGEEIIIcT6a2pqkqRhFkk8K4QQQgjxy6Ti2VRc1R5J2m5CqSlkhYWFEuQKIYQQQojNjsSzQgghhBAbx7paTclCZEIIIYQQQgghhBBCCNGFSNJWCCGEEEIIIYQQQgghuhBJ2gohhBBCCCGEEEIIIUQXIklbIYQQQgghhBBCCCGE6EIkaSuEEEIIIYQQQgghhBBdiCRthRBCCCGEEEIIIYQQoguRpK0QQgghhBBCCCGEEEJ0IZK0FUIIIYQQQgghhBBCiC5EkrZCCCGEEEIIIYQQQgjRhUjSVgghhBBCCCGEEEIIIboQSdoKIYQQQgghhBBCCCFEFyJJWyGEEEIIIYQQQgghhOhCJGkrhBBCCCGEEEIIIYQQXYgkbYUQQgghhBBCCCGEEKILkaStEEIIIYTYcMlktkcghBBCCCHEL9MFY1pJ2gohhBBCiA3zySew/fbwzjvZHokQQgghhBDrLxqFP/8ZDj0UHCfbo8kgSVshhBBCCLH+nn0WDjoIli6FW24BpbI9IiGEEEIIITrPNGHPPeGRR+CLL+DDD7M9ogyStBVCCCGEEOvvyCOhtBTOOw/efRc0LdsjEkIIIYQQovMCAfjjH2HAAHj/fRgzJtsjyiBJWyGEEEIIsW7JJLz44prLPXvCnDkwcSIUFGRvXEIIIYQQQnTW99/DrFlrLl9zjXf5sMOyN6Z2SNJWCCGEEEJ0bPp0b+rYqafCyy+vub579+yNSQghhBBCiM4yTbj+ethnHzjtNO8ygN8PRUXZHVs7fNkegBBCCCGE6KIsC+64A26+GWzbq67Nzc32qIQQQgghhOi8WbPgjDNgxgzv8vbbQyLhtUfowiRpK4QQQgghWpszB8aP96aQARx/PDz2mNfHVgghhBBCiK7OtuHuu70KW8uCbt3g0UfhD3/I9sg6RZK2QgghhBAi01NPwcUXe31si4vh4YfhlFNksTEhhBBCCLF5qKmB3/4Wvv7au/zb33prMfTund1xrQdJ2gohhBBCiEwDB3oJ2yOPhCeegH79sj0iIYQQQgghOq+kxOtXW1gIDz7otUfYzAoQJGkrhBBCCLG1c12YP9/r7wVw+OHw+eew336bXXArhBBCCCG2UsuWea28cnPBMOC550DXYcCAbI9sg+jZHoAQQgghhMiiFStg7FjYe29YvnzN9fvvLwlbIYQQQgjR9SnlzQ7baSe45po11w8atNkmbEGStkIIIYQQWyel4JlnvOD2gw+8hRp+/DHboxJCCCGEEKLzysvhqKPg/PMhEoEZM8A0sz2qjUKStkIIIYQQW5uKCjjmGDjrLGhqgn328QLc447L9siEEEIIIYRYN6XghRdgxx3h3XchGIR77oFp0yAQyPboNgrpaSuEEEIIsTWZMgUuvBDq6ryA9uab4a9/9fp+CSGEEEII0dVVV3vx7H/+413eYw949tk16zNsISRpK4QQQgixNfn+ey9hu9tuMHmy1x5BCCGEEEKIzYVpehW1Ph/ccAP87W/ez1uYLe83EkIIIYQQmWIxbxVdgJtugr594aKLtpipY0IIIYQQYgvXMp7t1w+ef977/667ZnVYvybpaSuEEEIIsaVqbPT61h5yiLfQGEAoBJdeKglbIYQQQgixeXjnHRg+HN58c81148Zt0QlbkKStEEIIIcSW6f33vYUZnnkGvv0WPv002yMSQgghhBCi85qa4Lzz4KijYNUqb6GxrYgkbYUQQgghtiSRCPzpTzBmDKxcCcOGwWefedW2QgghhBBCbA4++gh23hmefBI0DS6/HN5+O9uj2qSkp60QQgghxJbis8/gzDNhyRLv8p//DHfcAXl5WR2WEEIIIYQQnRKLwTXXwIMPepe32QYmTYLRo7M7riyQpK0QQgghxJZAKfjrX72E7YABXnB76KHZHpUQQgghhBCd9/HHaxK2F1wAd90FBQVZHVK2SNJWCCGEEGJLoGleovb++73gtqgo2yMSQgghhBBi/Rx1FFx1ldfa64gjsj2arJKetkIIIYQQmyPThOuug+uvX3PdDjvAxImSsBVCCCGEEJuHH3+Egw6C1avXXPfPf271CVuQpK0QQgghxObnp59gr73gllvg1lth4cJsj0gIIYQQQojOsyy46SbYe2/45BO4+upsj6jLkaStEEIIIcTmwrbhtttgjz1g5kzo3h1eegmGD8/2yIQQQgghhOicn3+GffaBG2/04tsTToB77sn2qLoc6WkrhBBCCLE5mDcPxo+Hb7/1Lh9zDDz+OPTqld1xCSGEEEII0RmOA/feC3//u9fqq6QEHnkE/vhHb30GkUGStkIIIYQQXV0sBgceCDU1Xr/ahx6C006T4FYIIYQQQmw+7rsPJkzwfh43zluLoW/f7I6pC5P2CEIIIYQQXV1urte/dswYmD0bTj9dErZCCCGEEGLzcuGFsNtu8NRT8MYbkrBdB0naCiGEEEJ0NUp5rQ8+/njNdeefD+++C/37Z21YQgghhBBCdNry5fC3v4Hrepfz8+H77+Hss6UAoROkPYIQQgghRFeyciWccw689x4MGgSzZkFBgQS2QgghhBBi86AUTJoEl10G4TD06wd/+Yt3my71o50lSVshhBBCiK5AKXjuObjkEmhshFDIC3Tz8rI9MiGEEEIIITpn1So47zx4+23v8n77wRFHZHdMmylJ2gohhBBCZFtlJVxwAbz+und5r71g8mQYMSK74xJCCCGEEKIzlIKXXoKLL4b6eggEvDUZrrgCDCPbo9ssSdJWCCGEECKbVq6EXXeF2lrw++Gmm+Cqq8AnYZoQQgghhNhMXHst3HGH9/OoUV4BwsiR2R3TZk4aSQghhBBCZFO/fjB6NOyyi7cwwzXXSMJWCCGEEEJsXn7/e8jJ8QoQvvpKErYbgRwRCCGEEEJsau+847VA6N7dW2DsqacgN9ebRiaEEEIIIURX19AAn30Gv/2td3nUKFi+HHr2zOqwtiRSaSuEEEIIsak0NcE558BRR3n9vlKKiyVhK4QQQgghNg9Tp8KOO8IJJ8DMmWuul4TtRiVJWyGEEEKITeHDD2GnneDpp73q2n79wHGyPSohhBBCCCE6JxyGCy+EI46A8nIYPBhsO9uj2mJJ0lYIIYQQ4tcUjcJf/gKHHQYrVsCQIfDJJ3DPPbKSrhBCCCGE2Dx88om3BsPjj3uXL7kEZszw2iKIX4X0tBVCCCGE+LXMmQPHHAOLFnmXL7oI/vlPyM/P7riEEEIIIYTorGuvhTvuAKVg0CCYNAkOPjjbo9riSdJWCCGEEOLX0rcvJBLQv7/XFuHww7M9IiGEEEIIIdZP9+5ewva88+Duu6GwMNsj2ipI0lYIIYQQYmOaPx+23dbrW1tcDG+84fX7Ki7O8sCEEEIIIYTohGQSVq2CbbbxLl92Gey1Fxx4YFaHtbWRnrZCCCGEEBuDacINN8DIkd6UsZRdd5WErRBCCCGE2DzMmAF77uktNhaLedcZhiRss0CStkIIIYQQv9Ts2bDPPnDzzeA48OWX2R6REEIIIYQQnWfbcMstXsJ21iyoq/NmkImskaStEEIIIcSGchy4805v1dzp06FbN3jpJXjyyWyPTAghhBBCiM6ZOxf22w+uu85L3h57LPz8M+y2W7ZHtlWTnrZCCCGEEBti4UIYPx6++sq7fPTRMHEi9OmT3XEJIYQQQgjRGa4L998P117r9bEtLoaHHoJTT/XWZxBZ1WUqbe+44w40TeOyyy5LX5dIJLj44ovp3r07+fn5nHDCCVRWVmbcb8WKFYwbN47c3FxKS0u56qqrsG07Y5uPP/6Y3XffnWAwyLBhw3jmmWdaPf4jjzzC4MGDCYVC7L333nz77bcZt3dmLEIIIYTYilRVwddfe6vnTpoE//ufJGyFxLRCCCGE2HxoGrz3npewPeIIr+XXaadJwraL6BJJ2++++47HH3+cnXfeOeP6yy+/nDfeeIMpU6bwySefsGrVKo4//vj07Y7jMG7cOEzT5Msvv2Ty5Mk888wzXH/99eltli5dyrhx4zj44IOZMWMGl112Geeeey5Tp05Nb/Pyyy9zxRVXcMMNN/Djjz+yyy67MHbsWKqqqjo9FiGEEEJsBeLxNT/vv7/XBmHWLDjzTAluhcS0QgghhOj6lIJEwvtZ07x49okn4O23oV+/7I5NZFJZFg6H1fDhw9X777+vRo8erS699FKllFINDQ3K7/erKVOmpLedO3euAtRXX32llFLq7bffVrquq4qKivQ2jz32mCosLFTJZFIppdSECRPUyJEjMx7zD3/4gxo7dmz68l577aUuvvji9GXHcVTfvn3V7bff3umxdEZjY6MCVGNjY6fvI4QQQoguwHWVeuIJpUpLlZo3L9uj2ap11Xhqa4lpu+rzL4QQQohOWLFCqcMPV+qcc7I9kq1aZ+OprFfaXnzxxYwbN47DDjss4/offvgBy7Iyrh8xYgQDBw7kq+becV999RU77bQTvXr1Sm8zduxYmpqa+Pnnn9PbrL3vsWPHpvdhmiY//PBDxja6rnPYYYelt+nMWNqSTCZpamrK+CeEEEKIzcyqVTBuHJx3ntcS4eGHsz0i0QVtqTGtxLNCCCHEFkApmDwZdtwR3n8fXnwRysqyPSqxDllN2r700kv8+OOP3H777a1uq6ioIBAIUFxcnHF9r169qKioSG/TMrhN3Z66raNtmpqaiMfj1NTU4DhOm9u03Me6xtKW22+/naKiovS/AQMGtLutEEIIIboYpbyAdscd4Z13IBiEu+/2FmsQooUtOaaVeFYIIYTYzFVUwLHHeu28mppgn31gxgyQ7/QuL2tJ27KyMi699FJeeOEFQqFQtobxq7rmmmtobGxM/yuTsxhCCCHE5qG6Gk480Vs5t74e9tgDfvwRrrwSDCPboxNdyJYe00o8K4QQQmzGpkzxChD+9z/w++H22+Gzz2DbbbM9MtEJWUva/vDDD1RVVbH77rvj8/nw+Xx88sknPPjgg/h8Pnr16oVpmjQ0NGTcr7Kykt69ewPQu3fvVqvdpi6va5vCwkJycnLo0aMHhmG0uU3LfaxrLG0JBoMUFhZm/BNCCCHEZuDpp+HVV8Hng5tvhi+/hB12yPaoRBe0pce0Es8KIYQQm6mmJrj4YqithV13hR9+gL/9zYtvxWYha0nbQw89lFmzZjFjxoz0vz322INTTz01/bPf7+fDDz9M32f+/PmsWLGCfffdF4B9992XWbNmZayI+/7771NYWMgOzQdW++67b8Y+Utuk9hEIBBg1alTGNq7r8uGHH6a3GTVq1DrHIoQQQogtyBVXwOmnw7ffwnXXeZUJQrRBYlohhBBCdEmFhTBxohfLfvMN7LRTtkck1tcmWhitU1qutKuUUhdeeKEaOHCgmjZtmvr+++/Vvvvuq/bdd9/07bZtqx133FGNGTNGzZgxQ7377ruqZ8+e6pprrklvs2TJEpWbm6uuuuoqNXfuXPXII48owzDUu+++m97mpZdeUsFgUD3zzDNqzpw56vzzz1fFxcUZK/iuayydIavtCiGEEF3UO+8odeSRSiWT2R6JWIfNIZ7akmPazeH5F0IIIbZKDQ1KnXmmUi+9lO2RiHXobDzVpWui77vvPnRd54QTTiCZTDJ27FgeffTR9O2GYfDmm29y0UUXse+++5KXl8f48eO5+eab09tss802vPXWW1x++eU88MAD9O/fnyeffJKxY8emt/nDH/5AdXU1119/PRUVFey66668++67GQs5rGssQgghhNgMhcNen9onnvAuP/gg/PWv2R2T2OJITCuEEEKIX9UHH8DZZ0NZGbz1Fhx9NOTlZXtU4hfSlFIq24PYWjQ1NVFUVERjY6P0AxNCCCGy7aOPvOB22TLv8qWXwm23QW5uVoclOibxVHbJ8y+EEEJ0IZEITJgAjz3mXR46FJ55Bg44IKvDEh3rbDzVpStthRBCCCE2ulgMrrnGq6oFGDwYJk2Cgw7K5qiEEEIIIYTovM8+gzPPhCVLvMsXXwx33ikVtlsQSdoKIYQQYuty0UXw7LPez+efD3ffDQUF2R2TEEIIIYQQnbVokVdw4LowYAA8/TQcdli2RyU2MknaCiGEEGLrcv318PXX8MADcMQR2R6NEEIIIYQQ62fYMLjgAkgk4L77oKgo2yMSvwJJ2gohhBBiyzZjBkybBldc4V0eOhTmzAHDyOqwhBBCCCGE6BTThDvugPHjYdAg77qHHpJ4dgunZ3sAQgghhBC/CsuCf/wD9twTrrwSPv54zW0S4AohhBBCiM3BrFmw995www1w1lmglHe9xLNbPKm0FUIIIcSWZ84cOOMM+OEH7/IJJ8DIkdkdkxBCCCGEEJ1l23DXXV6y1rKge3dvbQZNy/bIxCYiSVshhBBCbDkcx+vr9fe/QzIJJSXwyCPwxz9KgCuEEEIIITYP8+d7rRC++ca7/LvfwcSJ0KtXdsclNilJ2gohhBBiy3HccfDGG97PRx0FTzwBfftmd0xCCCGEEEJ01uefw+GHe4uMFRXBgw/C6adLAcJWSHraCiGEEGLL8Yc/QEEBPPkkvPmmJGyFEEIIIcTmZY89vIVzDz/c62d7xhmSsN1KSaWtEEIIITZfK1bAypWw337e5VNOgcMOk6ljQgghhBBi86AUvPoqHHss+HwQCsG0adCzpyRrt3JSaSuEEEKIzY9SMGkS7LSTt8hYba13vaZJwlYIIYQQQmweysvhyCPhxBO9RcdSSkslYSskaSuEEEKIzczq1d5iDGefDU1NMHgwRCLZHpUQQgghhBCdoxQ8/zzsuCNMnQrBIOTlZXtUoouRpK0QQgghNg9KwUsvecHtm29CIAB33ukt1jBoULZHJ4QQQgghxLpVVsLxx3uLizU0wJ57wvTpcMkl2R6Z6GKkp60QQgghuj7LglNPhSlTvMu77w6TJ3sJXCGEEEIIITYHH3wAJ58MNTXg98ONN8KECV4vWyHWIu8KIYQQQnR9fr+3KIPPB3//O1x7rXedEEIIIYQQm4s+fSAchp13hmefhV12yfaIRBcmSVshhBBCdE0NDV6Fbc+e3uUHH4TLLvOqbIUQQgghhNgcLFwIw4d7P48c6VXb7rWX1+pLiA5IT1shhBBCdD3vvee1Pjj7bK+XLUBxsSRshRBCCCHE5qGpCc49F7bfHr75Zs31BxwgCVvRKZK0FUIIIUTXEYnARRfB2LFQXg7z5kF1dbZHJYQQQgghROdNmwY77QRPPQWu6y2cK8R6kqStEEIIIbqGTz/1+nv961/e5b/8BWbMgNLSrA5LCCGEEEKITolGvRj20ENhxQrYZhv4+GO48spsj0xshqSnrRBCCCGyKx6H//s/uP9+rxXCwIEwaRIccki2RyaEEEIIIUTnfPUVnHEGLFrkXb7wQrjrLsjPz+64xGZLkrZCCCGEyC7Hgddf9xK255wD994LhYXZHpUQQgghhBCdN3Oml7Dt399rizBmTLZHJDZzkrQVQgghxKZnmuDzga571QfPPQcNDXDUUdkemRBCCCGEEJ2TSEAo5P18wQXe+gznnustoCvELyQ9bYUQQgixac2cCXvuCY88sua6/faThK0QQgghhNg8WBbceKO32FhTk3edpsFf/yoJW7HRSNJWCCGEEJuGbcOtt3oJ259+grvvhmQy26MSQgghhBCi82bPhn32gZtu8toh/L//l+0RiS2UJG2FEEII8eubN8+rpv37373KhGOPhe++g2Aw2yMTQgghhBBi3RwH/vlPGDUKfvwRunWDl17y2iII8SuQnrZCCCGE+PW4LjzwAFx7rdfzq6gIHn4YTj3Vm0ImhBBCCCFEV7dwIYwfD1995V0++miYOBH69MnuuMQWTZK2QgghhPj1zJsHEyZ4rRHGjoUnn/RW1BVCCCGEEGJzceONXsK2oMArSDjzTClAEL86SdoKIYQQ4tezww5eH9viYjjvPAluhRBCCCHE5ue++7wihLvugoEDsz0asZWQnrZCCCGE2HjKyuC3v/UWGkuZMAHOP18StkIIIYQQoutTCp56yotfU0pL4eWXJWErNimptBVCCCHEL6cUPPccXHIJNDZCTQ18+aUkaoUQQgghxOZj1Spvdtjbb3uXTzjBa/ElRBZIpa0QQgghfpmKCjj2WG9xhsZG2HtveOYZSdgKIYQQQojNg1Lw4ouw445ewjYY9FohHHZYtkcmtmJSaSuEEEKIDTdlClx0EdTWgt8PN98Mf/0r+CTEEEIIIYQQm4Hqai+effVV7/KoUfDss97aDEJkkRxRCSGEEGLDvPEGnHSS9/Ouu8LkybDzzlkdkhBCCCGEEJ2mFBx5JPzwg1d0cP318Le/ecUIQmSZJG2FEEIIsWGOOgoOOggOPBD+/ncIBLI9IiGEEEIIITpP0+C227yZYpMnw267ZXtEQqRJ0lYIIYQQndPYCHfe6SVoc3PBMOCDD7z/CyGEEEIIsTmYOtWLa1MzxsaMgenTJaYVXY4kbYUQQgixbh9+CGedBWVlEI/Dffd510twK4QQQgghNgfhsFdRO3EiFBTAPvvAwIHebRLTii5Iz/YAhBBCCNGFRaPw5z97K+eWlcGQIXDCCdkelRBCCCGEEJ338cfe2gsTJ3qXzz4bevTI6pCEWBeptBVCCCFE2z7/HM48ExYv9i7/6U9ee4T8/KwOSwghhBBCiE6JxeDaa+GBB7zLgwbBpElw8MHZHZcQnSBJWyGEEEK0NmkSnHOOt6Ju//7w9NNw+OHZHpUQQgghRJfhuoryhjhR0yYv4KNfcQ66rmV7WK1srHFuLr9vWiIBe+wBc+d6l887D+65x2uNsAXZ7F4X0WmStBVCCCE2srUDpz6FIVY3JTavQGrMGCgqQh17LCuvv5XFpg/mV7FNjzwGlOR2/fELIYQQQvyKFlWFmTq7ksXVERK2Q8hnMLRnPmN37MWw0q6TFPwl42wZ01aHk8wsa2BJdTRjP4ePLCXH79ukcW6nk5ShEBx9tLfo2JNPwpFH/qrjyobN5X0oNoymlFLZHsTWoqmpiaKiIhobGyksLMz2cIQQQqzD+p61dl3Fl4tr+GBOFasb4+g6WI4iabkE/ToBn951AynThHfegWOOSV+1dNZCnltm8vWSWhpjFi6KvICfHfsW8vs9+rPf0B6SvBWbnMRT2SXPvxBCeImySV8soy5q0qcoRG7AR8y0Wd2YoFtegLP2H/yrxHnrG5v+knG2TAbWRJKU1cXwGzo79iukb3EuMdNmYWWEpoRFz4LgBsW5G1Ihus4k5YwZkJcHw4d7d0gkvEV0S0o2+DG7krUT6e/MqqA+tmnfh+KX62w8JZW2QgghfrHNIfhpb4ztXZ8KCBdVhamPWxgaDO2Zz+9HDWDb3q2Dn0VVYV78ZgUfzasiZjnkB33kB300xi3ilkNJrp/dB5YQ8hvMXtXIqsb4LwqkNupzPmsWavx4tOnTKXvuFThqHHHT4cFZYWaWNWBokBcyaIrbrGqMU94Q47vldYwd2ZtT9h7YZYLBzeF9KIQQQojNm+sqps6upC5qMrw0H03zYo2CkJ/8oI+FVRHe+7mSIT3yN2ocsq5kZVszvVqOEyCcsDEdl14FQSqaku2Oc0FlE498tJjaSJLehSGSloOrwHFdFlZFyAv6AI36mEl1JInP0Nhnm+7ELafTcW5mrG1iaLoXa+/Rj217tZ3Eap2EziFm2sxe1UhFbROX/fAfut9zp9cS4bPPwOfzqm1DoU49hx1xXUVZfYylNVEAhvTIo38Hs89+jbi05fjjls3y2ji247LXNiUUhPzAr/8+FJuWJG2FEEJskFQgMreiie+X1lEdTpJ03C5XSeq6ii8W1/Dh3EpWNybQNY0cvzfGEX0KmLc63CpwG9GngGnzqlhRFyOWtIkkbZK2w5xVTXy2sIY/7jWAQ0b0SgdfCyrCPPzRQmaWNWK7LgNKcrAdxYr6GLajGNgth4Tlsqw2xh6DShhemv+LAqkNCXLbfG5Mi6Zbbqfw9lvQbYtofhFv/biC+b75VIeTlNXHCRgauQEfleEkluNSEPSRtB2iSZvPF9aQsFzOPiD7Z/Flalj7JJkthBBCtG1DviPLG+Isro7QpyiUTtimaJpGn6IQi6oilDfEGdAtd6OMs2WysndhiHzXRzhh8e2yWsob4hy6fWmrmLZHfoAlNVEGdsulLppkzqowdTEThSLkM8gP+vhxhdtqnAsqwvzjzbksro6QGzBY3ZCgMWHRIz9IYchHXdRkUVUEgLjl0C3XT13EpDKcoE9hDr0KgiyqjvDK92VMGDMCn09v9/dZURsjZqZibZe5q5v4Zmktlxw6nIO3K+0wCd0yWT6wYhkHXT+B7kvnABDr0YuVSyvI6d6NfsU5AHy5uIYXv11BNGkzpEcefYNrEr7tJZlbHu+8/3MFP5c30ZCw0BQU5frZZ0j3NgsYfo24dO2Edb7jY0FlBMdxmbmykV0HaHTLCwK/3vtQbHqStBVCCLHeUoHI9LJ6FlSGsR1Fn6IQ2/Uu2GiVpOujo2rZF79ewUfzq4hbDnkBg4KQn7ygwfzKMK/8UMaAkly27ZWfPlM/q7yR9+ZU4DM0EqZNJOkQ8hsEDIO46bCkJsq97y3gs4XV7DagG8N75fPEp4tZXBMlYToEfToVjUnygwYoL2hqiNl0z/dTFzVpilsA2I7L5wur2X1QMXsN7t7pJFpngtxDt++1zv0s/+pHguecQ++5MwD4dLt9mHja1QzYfgg+12VBVZiYaVOc46c2amLaLnlBA0OHoN/AtF1sx6W8IdZh8nl9qxIAbNvlx7J6aqMm3fMC7D6gpM2Af+3npK2qi035PuyKOjpoGNIjX5K5QgghtlobmliLmjYJ2yE3kNPm7TkBg8rmtQw2hpaVvd3z/Myv8JKvtutiaBrLamL8sLyebXrk0rc4Jx0HzVndxIq6GKbtMr8yTDRpY+gahqYR0x0aYiarG+P8vKqRAd1y04UOT32+lPmVYXrkBcgLecnhuOlQG0kSMHTyQz4qm7yT+ZajqHcckpbLV4trMTQNw9BwlWJ5bQyUxkl79s94PlO/z4raGPUxk4TlkB/yU5ijYdkuq5sS/PPdeXy+oJrGhN0iCR1kSU2Egc0JyKa4hWVa/ObtFzjk+QfxWSbR3AI+uOjvfLLH4SS+rSTkq6E4x4+L4vOFNdREkuQHfTTFLQZ1zyMv6GtVdQxeYn7u6ia+X1bP4uowc1Y30Ri3CRg6vQqDBP06TTGL9+dUUhVOctlhw9O/Y2aVcpDuuQHCSbs5wR7j7AO2abMyuqM4rK3q7ppIEg3oURCkIWaxuDpKSW4gncze2O/DX0qKCDaMJG2FEGIrtqF9pCZ9sYzaSJK6aBKA/KBBVVOcuqjJgG459MwPURNObNQpOetqY9BWteyHcyv5fnk9roLueQEqm6tHXddr567QSFoOg7vnkB/04boK23Uoq4+Bq1Cahq5BfcwiabtoGuT4DSzHobIxybRoJU98toS4aWMYGpbjomkaTsIiato4riLk14kmbXICBrGkxZeLa6iLWcRNG8tR/PWVmewzpDu/3bUvvQpDbb4Oqd+9KWbx/DfLmbe6iajp4LguBTmBjCD3wQ8X4jO0NveV2k/jo4+z7T+uJWAmiARzuW3shby16+G4FiycV8WIPl4gGUm6RJLJ9DjilkvA51Xeuq7CchUluYF2z+KnkuZfL6mlNmriKkVxjp8Dhvfk1H0GtgrgyxviTJtXyZszV1PRlMByXPyGzuDueZy5/+A2k9GbcoriLwk2U/cNJy0iCZv8oI+CkP9XDVg7SmbPrWiitDnIl8pkIYQQW5v2viNnlTcwv7KJfYZ0p29xTvpkM5COAZriFkFDJ2ba6SnpLcVNh6DPIC/QdrplfeOJVGVvjl9n5spGYs1FAkGfgXIVDTETy3HZvk9+xhT5YT3zmV8R5vvldWhoXqyrFDHTxrQVCtA0xVOfLcGn68xb3cTbsyuobIqTsBz8hobP8PrUBn06pqOoi5r0KgzQlLCIJiwvCax78XI06cW2Ib9Or8IQMdNhzupGJn1hc9b+g9MnixdXR5hZVk80aZGwHLrlrUk0Bv0GxTk+yurifGRXc/B2PdMVsXNWN7KiNkbQpzMnYmKuruCGydezy7LZAMzeeT+uPfIvGAP6M8zQ2KY4j9WNcd6eXUE44T2WUl6LiNWNCX5eFaZ7np/coI+AoRNOWuzYt4hZ5Y1ML6tnfkWYhGWjFCRsF5+uAYr6mFftXFoYojaSZEGFdywy5KB8FlVFuPnNOSyoDKNrGj+vasLQNXIDBrl+g9WNCXL8Bn/YcyDvz1lz/BI0dHoWBNljm25s37uw1SLGSqlW1d0BQ8dn6Ngu5DdXQIcTNoU5/k69DzelX3tG3JacEM7+qyeEECIrNuTLs2WCLOQ3WFEXx3YUtY5L0nZwXCirj5HrN8gP+WhK2Px2l76/eEpOR4nZafOq2gi4vWrZgKGjAzkBnZUNcWKmgwZoGrgKdE1RE0ny0bxqCnJ81ERMYkmbuOWiAL8OhSE/ruuC8oLbhO2gA1VhL6kYNW10TSPPbxB2vQpapTRc10vims3PS8J2sBwXPWJi6F4FgqsUleEkr80o572fK9i2dwEDu+VlvA4LKpuY8t1KvltWx+rGOA0xEwX4dJ1ueQHyXIXu09NB7pLqCLe8OYdhpfnk+H3pfQFMnV3JwsomChc2cbOZ4Kshu3PtuEsJl/Yhz9CxHZemhM13S+tIWG6r10EBSVth2haGDrqu4biKxqTF7FWNuEqhATHLoTqc5KVvy/hhRT0J0wG837e80ebV6StZUhPluqO3Z1hpQfr1/XxRNbNWNmK7isKQjx4F3hSvBVVhbn9nHkCrxO2GTFH8VRa96MR9p5fVs6IuRtx0yPEbDOyey24DSn6VRGlHyWzTdvlkQTU5foP9h/Xo1PRAIYQQYkvR3nek5ShWNyZYUh3ho3lVFIT8FOf62a53ASU5ARriVjrBVhMxqYmY7DawOCP+UMrbx079itLT8ltaO55YV7KuX3EOUdMmbjnUR00a4xauq2iIWbhKoRQkbQfQWFoTY2C3PMBLTCYdrw9twnIpyfGhmq+3XRfX9eI6XYOfVzVx/f9m068oh6Tt0LMgRGVTgkjCxrRduuUF8Bs6lmkTTlok6rwqXccFjebkL+D3aRSGfNgu1MVMipsTx5XhJC9+s4JuuV67hupIggUVYRyXdKyXMG0Sthd/N8YsXJWKQ72kcCoJPW+116oMwOcE8CcTRAM53DnmAl7a+VBAo3ckSdxyKc6J0xC30utL0DzO5toNHMelMpwkJ26hUCyphsWVEUpyA8Rtm3DcxFEQTTooIOTXvTZhlktlU4JueUEMQyeSsPhiUTV9i0O8PmMlPy6vw3HBcb3XxzDAcXzkBnwkTYd3Z1ewtDZK0GfQpyhEwtKZXxHmm6V1vPtzBf2Kc/DpesYixoUhPzWRJH1bvKcKQj5Kcv2sakiQFzRIWE7ze8G/zvfhprSu1h6/tM3alt4iTZK2QgixFdrQ6eQtz/TPKm8ibjr4dI2k7eKkAiDXS4o2xr3k6bR5lYzfb5uM/axP0qyjL/r35lRQGPJnBMwFIT+9CxU/rWygONdPzHSojyaJWi5KecEpeAFbwNCwXEVlOEFNRCPkN8jx6ySak7aOC5HmKUW6rmE7LrZSaAqStommQcink7RdbBf8hoblKGxX4SrvOgAdsFwvoemisJw1Aa73E0RMh6U1MfoX56Rfh+16FfD0F0tZ3RDHchQt06hKudRFk4QTNiW5foJ+g9qIie0qEpZLz/wQPkNj9qpG5qxqIH9VGQvzexJJ2qwevDcrTr6FTwfuAppGgePiuoqk7WDZLrbq+P2jANcF23b4YlE1uq5THU6SsFz8BnTPD1LRGKeyKYntKoKGhs8wcJVC1xQJy2F6WT0vfLOck/cayOQvl1MTTrK8NooCCkM+LFdRE/Fe84ElOayojzP5y2WMHt4zo1VCZ6cohpMWZXWx9FS36nCCpOM2HyyF2GNwCdv3KWzzvfhL2i+0bGdRFU7gOIqCkI+k5bCyPkbSdjvcx4ZWDrSXzPYqNaLomkbq6tTBkCxaIYQQYnPUme/Klts0xS0WVYUzviProiZfL6mlsikBeN+NRTk+GmIW782ppCDkZ78h3RjSI5+YaVMTNVnVECex2KFfSQ4FIR8+XaOiKUm3vABjRvZaZzzRmWTd0J757DKgyDvx3RAjZnqJ2IBPx9B0kraD7Sh0XVEbTVJWH6OyMUldzCRmWkSSFq6CmOWC6WI6bnP7Li8ONZrj+NrmtgG24zZX8XozxcLNazv4dY246eA0x9Ku8uLBVMiYOrFfE7HIC+iYtkOfwhCFOX6aEhbT5lbSqyhEn6IceuaFWKpHCCdsKpsSVDWRnkGG8vbtMzRMu3m8zSzHpbC6grJgIcofwKdrXHnsBBIYLC3sBY73O1mOV2yxojZKdSRJyzqEtUNcV3kzyXwGmA5UhJNUhpMowNAg5DPQde+YIGG5mLYJQFMCaiNeIYVS3qy8+ZVhr2VDi8fTAOVCJGkTiJn0KwqyoCrK4qoIR+/Um4VVUeZVhnEcl+55fhriDgsrI+g6dMsLpBcxXlwdoawuRo/8AAOaE/P1MZO46dAYt6gKJzB0jTmrmxhiOcQtt933YVt/L8B6tWro7OyxdbX2WN2QIOTX+fu4HTYo7twaWqRJ0lYIIbYyv2Q6uXem36Y+amHaDgHDC7JcpTAAzdBwHIXtQl5Apynh8OZPqzh1r0HpRNv6nA21bZdXvitjeW2UHvkB5lU0UR+zmisEFNXhJL2LQlSHQ1iOItA8Vcx0XPw+jca4RV3UTCdhgXRyGbzgEs0LtpSmCPkNvEjW+58CzOYMpqE1p1eb7+82R6oR00UDwkmbgKGhFOmkLDSf0W/+uWVwy1o/u8oLvr5dVsfB25Uyd1UjU75fgem0/TparpcIxvICal2jeZE1HUPXcJSiJBQgWLGKA/55LSPL53PMhf+iMrcEx3X5YtCu+HQvQG2K263Gsy4uXvAatVz8htdmwlFegLuqIUHSdkg0P3eWA0o5Gc+J5di8/dNqVtTGaIqbhAyd2kgSTdOIm95radoOdTGTvkUhuucFWFoT5fsVdfQrzk0HlTl+g5DPyJiiqJRKr45s2g4Jy+G1H8tZXB1h7qomLFfRqzDIwG55rKiLeQdLs1ezbe+CVpWvv+TvxXv/rmRZTYRo0quSKQz58ekaeQHDey87LrURs819/JLKgfaS2eGETX3MbD6hYWccDMmiFUIIIbqSziRjF1WFeXd2BbPKG4mZNrkBHzv1K+KIHXunvyvX/j5NWg5l9XH2GNSNgpAXNyysDFMbSWLoGiGfQdJ2MXQdXQPlKkzbYXVjgv4luRSE/AzqlsuKuhjLaqOUN8TRNCjKCbDvkG6c3MbCVC3jiWE981jZEGdWeSOW7dIzP0BdzGJhZQSFd3J3l/7FFIX8fLusltmrGnAdRW3ERNcgL+jHSJ159XocAIqmuMWslY3ezC5NI5L0ErpARixsaN5ddEjPktI0WFEXxafp6SremOWAovk50dOJWqeDgFHhxcaGBrHmPrhzyptY3ZQgnLBYXhPF0CGS8F6HWKoCVgOtRSLYcrzfp7o5iR5L2PR9/SVef+1hJu/+Wx446HRcBXML+7Z6/NqoRX3cBuXSxsSx1u8zyIi3Wx4zRK3MQLzl754qctDwZtrVRd1WsXTq2MFR0BAzyW9uVRAzbf4zYxW1ERPHVRi6Rjhhe0UirmJIj1xi5ppFjHfuV8TqxgSzy5voV5xDQ9xiRlkDcdOhV2GAqrCJBqxuTNAYszh4ROk6F0iLW94xXI7fwG9o2I5a56LS6zt7bO3WHnHTIT/kw2/4sByX+qjJR/OqOGREKQcM79nx67TW50FHC9NtSYUIkrQVQoitzC9Z8TYv4MN1oTqSpDg30Ny7yms5gO4FvZoGtuuStDUKc3xUNCZ5d85qhpUWUBNO8tZPq1jVmKAkN0CPvCCGTptnQxdVhXnlu5W8M7sCRylmlzei6xqlBUFKcgPUhJPELW9hsIqGBJqu4dM1inL9lOQGsGyX6nAS220/GelC+kZd9yZ4JZorGFrcBHiBWXMut9X+NLyEqeUobypUi9t0nfQUtHVxFdRFLT6aV0lFk0kn4kxvnCo1Pq+SVQMCukb/t17liIm3URCPkDT8bL9iLuXb7rOmAthp/XuuD0ulAlWFX1eEfDqRpIPjZu4x9Xipg4TmIg8qwyZV86rXGoMigQLTxad7U8t65AUINfcB+9fHi3GVd5CRF/SxQ+9CDF1jQWWYYT3zsV2vkrQ+ZmI7Lo1xr99afdSkoilBU8JbhGNJdZTF1VGKcwL0KgwSTljUx0xmlWe+Fzf078V7/5bxzuwKkrZDY8xqXpTOwm/o5Pi9hTzqYxb9S3Jb7eOXVg7kBXytktkApuMtIBfwaRi6TsDIXOCtqy1aIYQQYuvUmROXi6rC3P/BQm+qvUpFaBpLq6PMqwhz2WHDAVp9n1Y1JZi7OswPy+vYa5vu+HSNqnAC21XNs6bcdO/XSNIhN2Bgu1DVlGxuL6CYubIRlDc7aJcBxSjltQSojiRZUtM89b1Fq4NUdW+OX+f75d5CvjHTIeDTSVoO8eYEZsjQqUw6fBSpojDHj8/wknlx0yZmeTFe0kpiGBqa5iX4FArHgbDj4LhxfIZO3HKwbZVROJDitBHMarbCVhAMefGN63hFCeAlUOvj6xcXOAqqwklem7GKaHNGNGm3vw+1Vuyt8ApDvl1Wx3A3yl9evovfzP8agH3KZvGocoi5Rpv7Mh3VcWZ5I0vNzmvvEVNPt2krGuJJUFAf8wouUIpAc4Acs7z3XdCnkbS9RG5FY4KmuEVRboAd+xYyfUUDM1c2UBc1aYhZzW2vHPoV5zC8tICcgE55Q5zueYH0omopLRdIywsYNMQsVjXGqYmYaJrG4O657DqgmKBP59tltfy8upFT9hrIfkN7APDl4hpe/GY5K+vjNMZMLFcR9OkkLJuyurZnj7Vs7RE31+pf7DPoWRBkZX2cD+dWsd/QHh3OvFz786DlwnTre0y7OZGkrRBCbGV+yYq3/Ypz6FOUw4yVDZTk+vEbejoQUS16Y6XOGBc0ry773Jcr6JYfYEFFEw1xm8KQj/J6L6gsyQ0wpEcuK+tjvPD1Cs7afzAJy2XyV8tYXhvF0DQc12ttoJqTmq6CpoSFao7J4pZDQcjnXR+3CcctGmLWOqf5t6TjLaAQNdtPla5paZDJBRzXTVcgZNzW2cxrs6TtsqrJXK/7ZCSXXUVeYy3H3nIzu3z/MQAzew/ninFXsLjHALQW43Fp+/dZ38d2lFfBkbBdlOvtt619tnWQ0NFj2y5Ekg6VTQlipkM4afPFohp0XcOva4QCPr5bWoff8Cph5leEUXiLxXnvB4Xb3C5iQVXEe24CBgFDoymhSFoOjXGTHgUBCnL8RBM2g7vrLK+N8sr3ZUwYM6LDvxelFJbjUh1JsLg6krE43qQvvPev43orKzsKdFwsB/yGTtR0vIVE/N6Bleuq9N/c2tW9QLpyeO0Vjtub8hZOWPTID7KiLsq2vXyZi1boGk0xi74luRSEMkPBrrRohRBCiK1TZ05cDumRz4vfrGBmWQMBQ6Mgx4tLLcclHLeYWdbAC18vp3t+MKMSTylFbtCgR36AVY0JFlWG6ZYf8NY1MO309Hxd96bZJ23XWyNB8058Jm2vnVXMtCkMGTQmbBzl9RetaEzwfaW3gFX/khxMW6VbHSQtl8VV4eZkpE0kYYMGCdMh0rwP8NpfKQVx20ta+ptnsVktkpqWAsv2gqqgoZEX9BFO2igXYqaLX3fpIJxtU2r/MctB0xRu8/T1pLvhUWJ9zFqv7duqUj3gx2nc+v5jFMXDmIaPew84jYl7HYdSbSdss6UzT7fCi9PizbPTfFpzzOwqDMOb1We7ELcUFY1xNM07gfDlklpG9imkT3GI7vkBaiMWS2uiADQlbIpz/Izsmc/A7l5i0munEM1IVi6oCHPzG3NYUOUtkBZN2vgMHUPzCj0cYGV9nGjSojg3QNJyqY97ldF7bdMNpWDavCqv3YTj4rjg08EwvFYaEdMhx69TG/FlxKh5AS8er44kKQj5WiVXbdcrwljVEG83udre50FqYbrSgmCbCwJuKYUIEpELIcRWpr0KvJSOkja6rnHYDqV8sqCKVQ1xIsk1U4bSZ8abE5c5ASM95ad7vlcpkOpx6rgu/Upy8BsGK+tjLKwK49M15q0Os7I+SkPMwlGwTfc8VjXEicVdQgEDQ/MWAyivt1FKpQNcW3m9ojRNaz7brVjfGNN2FclOnJXXWrROaKm9u65vqPtLEqgacPjcL7j1vYfpFmvC0g0e3P9kHt3nRBzdaHP/G6sOwW6rBHkjUMDqpmT6squ8/yRRREwvue3ToVuun6RpE7cUEUPDcvwU5/hxmhOrqR5u3sq/Oq5S+H1ehUp1OEmP/ABVkSTJ5Q24KJbXxkBpHLhtjzb/XuqiJouqwpQ3xImbDk9/vpTZ5Y2M2aE378/xEq5De+QxryKM7XhVw4auYzte5Y7XIkIRScJPZfXkh/zUhJPQO7Mavj5msrgqmu7/5dN18gIGP65w26zubVmFYDZXm0eTDsN75ZMT8Np/uHjv1yE9MisTutKiFUIIIbq+X2PF9s62JTpqJ42vl9RiaF4v/ZbVe4F8ncqmBJ8vrGFIz3z6leSgaRp10WT6OzVm2iQthxkrG/DpGtGkk5F4c11wmjOfMbe56lQzWVoTZWV9HKt5OrxlO3y7pJaY5eA3dIpzfESTNvMqmkjaiu55AUYN6kaNnWBVYwLbUWjNs7BSz1TLx7VbXHAUOOuoQEg6CidhZ8wSW9+EbUteNwBvB/avEdh1UkmskX+8/y+OnvcZALN7DeXKcZczv+fgrI1p49Bwmmeq+Q2dhO16SfjmtS9SLNvF1cC2FStqY1Q2JuiRHyBhu/QpClGSF6Awx++ts+G4LK2JUpTjp1vz7LT6WJTZqxoB79jujnfnMnNlI7oGCcvBUQrD8Rak8xkargLHdYiaNpVhk8HdculVGKQ+avHmT6uJmV67sZYFKo4CzVWYtpfgX1Ybo1dRTkZ1q1fwE2JmmVfw05JSikjCpke+N/OyreRqR58Hw3rms6Q6yvzKMD1afAakbCmFCJv36IUQQqy3fsU5DO2Zz+xVjeQHfeudtNlvaA8O2q4n/5u5GtuxvX5Yzbe1/KqsjyZxXCgtDFGc42Pa/BriloOhQ8xUlNXFKcn1UxtJkrRdfLqGpil+WN6Qriyoi3qLWyUsB5oX8DJtp83+VFbLXgcbYH1mUWUvhO2YC+y3fAbdYk3M7TmYK8ddwZxeQ7I9rF+d7UJVZE01h2UrqsMmDXEb1/X6uvl1DdPxAkvbUV7S1vBaYkSTNgnL6/tm5Hq97MJxi59XNRBJehUMqxsT6b+X1GIltZFk8xQtnfL6GP9rTPDjinp0TWNgt1xcV6GaS8Q1NBJWczW2o0g1JfAbGo7rnTR4e9ZqeheFvMXkbIeEpTOrvHX/r4bm6ZdzVzets51C0nZpilusqIs1LyxicOCwHlSGk9RGLQI+g5yAQdz0evW1t2iFEEII0dKG9l1fV6K3s22JvltWT2PMontBoNV24BUPVDUlyAkYDCvNpy6aZEZZgzfLxZdqD6SId9D0VK31c8x0mL6iHstV+DQNQ/cW9ayKeDGvhndSVykI+Lz+9dWRJD+uqPMWs00tKuu23v8vsT4zyzYXBckYBy/+DlvTeWTfP/DwfidhGa2LTTY3MdM7dlLgzVBrcVvLn2PpNTW8RYvjps3SGpucgM7w0jzv5IShEfL7yFeKuqjJ4uoISuUyd3WYyqYEz3y5lDy/j6qmJKub4jiuSyDgQ7Nd/LqWXnzYVYqAT8NSa2ZMVkdMehcGaYpb2K4ibjqtqom9RK/XGs+newvHrWqIM6AkN52A1XWNQ7fvxSfzq6kOm5TkramIjyRscgI++hWHAK3N5Gp5Q5xFVWHygwa1UZOAoacrdgtz/PQpCrG6YU0LifRzuQUVIkjSVgghtjK6rjF2x16saoyzsMoLitcnaZP68p36cwUxU0e1+Apv2T4gZnlT0YeV5vHDikaaEhY+XcMwNG8qeNKbGuZ90YOjFMqBJtcCBX7d648aN20Stjf1vitQXTAwDtgWps8LZG8/6CzKC0t5eo9j0tdtjRwFbvOBmK55C5vprtcuAd0LUB3lVU6btnekFfTpzb1wVfrgrKwuxvBe+ZTk+llYFaFXQYDvl9Wzsj7mTbEM+OhbnOP1nItbzFnVRMhvsF2vApbVRjFtl7jtsvbbN/U2sl2FYWjsNbgbtVFvQbKjd+5D0PBWlG6r/5cKQXU4yQ/L6zls+14A7VYh7DagmAWVEQZ2z+HY3fpREPRW911SE0kfbFc2JQj6DHbqV8SYkete5EwIIcSWYUMrZTe073pnEr2daeNV0RhndWMc03WxbEXIp9LffXHToS6aJJL02hCtakzw1ZJaXKVojFteQiqcJGbaGd/NnWoXpbznzHXBRKE1V0c2hxbeVPfmnTi2wvGDYzksroyApnV6rYKtVct4dkVJH6466jLKinoxq8/wLI9s40m2s8Bwe5zmPmSp4pKY6fL1kjoChkFNBPoUhijK9ZMf8rG6MU5NOEFt1PRi2qjFykSMynASDQimFlzGS86mZg86ylufAuVV/7rNs9Qqwwlipo1S7bd/cBQYSmE6Xn/bpriFU6wyErD7D+3BwduV8vniGq8diLLx6TqlhSGG9MijNmq2Sq66rqKsPsarP67km2W1BA0DQ9fSrfWGlebTLS/Atr0LqI4kWVQdYdteBVtkIYIkbYUQYiuxdmB++j6D+M+PK1lcHcFVUJzjbzNp01ZAv7A6TEPcanWGGNYkbgG65/lJWG5zVYOBhUPSctPTs1N3cJ01gbLmpqoZXBKWt0puV9KVcrb5yRj/N+1JBjZWcNofbkFpOrFADv/a5/fZHlqX0HL6VqqSJrVwmq6BbYOhe1PSvLYJXtWMoyDH7wWe3iIhGkfv3JdP5lfz0bxqyhvjuMpL8vp9OrqmpadjRupi1ISTfL2khmW1cWItFrZrSQN8hvdTXtBHt7wAAZ/OoqoICuhZEOSbpXX0XKuCSCnvhEef4hBVTQnKG+IAHVYl9S0OURM2KQj605W5w0oLGHJQ/kaf1iqEEGLz8EsqZTdkxfbOJnrX1cZrdUOcZbVxwkmbhOmyPBmjW66fbnlBACqaEs09N13ygwa9C0OsqIuRMO3mHqFeQmrtGVadie9cMlsPpBK0utb6pL4NNMZtDMBpubFo0wFLp3PHuw9y1VGX89WgnQF4e8QBWR5V19DyvepV6Kr0om6RZIRuuX6Kc5vbbEE6uZkb8HkzvsJJXAXJVCys0byo75r9pgrOLdfF0DUChk5T3GxVdNDm+BzQdZegzyDZ3L6hZQJW1zVO2WcgCduhvCFOSW6AgpAPn65R0ZRslVxdVBXmxa9X8MmCalY1xkk2VwYXNf+dV4cTRJI2uw4oJsdvsG2vAob0yKMmYm6RhQiStBVCiM1EW8lToFNJl5aBedxyaEpYxJJew3i/T8OnG/QsCHHYDqUZX26p+y2qClMft9A1b1GGGSsaiJtewlbHW6whtRAZ0LwaKtTFLCw3So7fwLQd4qbb5pnalmFsqtLBpTmYkBi3Tfsu/4m73r6f/k1VuGjssXIO3w3YMdvD6vLS7y+Ft9AGGhp4U8x8OgnLa9VRHApQEDKoaEoyY0UDNZEklU1JYqaDUpDj8wLiWNJmtesFoDHTIWnaRE2HGWWNzQuetH4La0DIr+MzNHIDPlxXEU7Y5Aa9BRPilsMe23Tj3Z8riCQcNE1rNZVs214FNMWt9PSzDVlcUNe1zXo1XSGEEBtmQytlYd3tC3oXBplZ1sCnC6sZ2jM/Ha92lOhdUBnhle/LOHa3fuQFfAzpmcfPq5patfGqjST5blk9OX6DYT0KaIxZLKmJUh+zmltwaSRtr/+spmkM6p7HyL5FfL2kliXNbQt8Ouu97kHG78iaWCK1m472t56FlVudXDPONR9P4vTpbwNw8Vcvp5O2on0t1xKpiVrUxax0W46QZlAU8hP06cRNDU3T8OleOzGnuXWX7apWJy50wGluN5e0HVy3c4dhDt7fhKMgaOgcPKK01fHosNICzj5gm/TxaF3UbDO5uqgqzP0fLGTGinrCCa9PtIZXydsQs7AdRb/iHOKmzaKqMEU5fnYfWML5Bw5hdXOsu6UVIkjSVgghNgNtVUMU5/pB0Vzx2n6FRMvAPMevU94QY1lNlKTtnRHdpnseQ0uDlNXHmPzl8nSgnrrfiroYsaRNbTRJbcQkknQyk6waGV/qRnMkG/BpxEyHuOXg0zo/HUimjnUsZCW4+pPJnPXDGwCsKOrFVUddJgnbTnJZUwmeH/RRlOunqsmrQAgnU0lNnepIkvq41/IgabtY1W7z9DFvxd+opdAsb/uY6Z0IyfEbGLqOoXnVtU47VTUKbwXqnECAHvlBkraD6bhoJukFE7bvXci2vQqoj5pETYdIcs1UsqE98/EbGknLTU8/S1Ul5Qd9hBO2t0/bJeDzkr0BQ9/sF2IQQgjR2vq2OOhspezgbnltJkE6al9QFzVZUNnEyvo4T32+hJ753nfWzgOK2k301sdMqsMJfl7VyILKMN1ygxTn+DF0LaONVyxp8+2yOgD22qaEghwfQ0vzaYhbNMZMYqaD4ypCPh3Dp9MzL8iO/Yrolhdkmx55LKmJphOtv6QeQOoJNp49y2Zz99v3M6ihAoBndj+aO0efmd1BbaZcBX4dgobX96AynKS35hUk+HSvNZ2OwqfrBH0+6qJm+r6p4gWrxSpjjlLp9gmdoZS3wFm3HD+zVjbSrzgnfTya+oyyXcVvd+mDAuKW0+rzynUV786qYEFlGF3X8BngNwxcpRNJ2JiOS9R0qIuZFIX8LK+Nsec23Rgzshc+n77FFiJI9C6EEBvgl6yWu74Vs21VQ6xqiPH+nEoA9hxcwpAe+RkVEuP3G0SO30djwuSFr5azoi5Oz4IA8yvCrG5MoKFRkusnbnn9gmzXZXivApbVRHjqsyUcun0v3pldwYraKDHTIZy0iSZtzDbaIax9ltZubmIPGkopL3n1C55rscbu5XO5+637GFK/CoAXdj2C2w46m2hwywxSNrZU7KnhVZnqGpywe19e+a6chrjXc9lLvHqVBg1RE9NZ06/Or2vEzcx96tqa1Z2DhsLBWxJYp+PKG1dBt1w/PkPDdnX8utZqwYRd+xfz7bJab9Xr5r/ZwhxvqujCqkjGtkN75vP1klps16WiKUFjzFs4wqd7LR9G9CkkbspfohBCbEk2pMVBZxb6+nFFPf+cOo+aiNlqv+21L6iLmswoa6ApbhHyG2zTPR+foTF7VSM/r24kkrDpu9aCQKnFwaIJC1cpQj7vO3hVYxyfodOnKERDzKKyKYHtKPy6xshBxUSTDj+VV9EQszAtB13XMVwHWymKcrxWQENL89MtE3ICBkbzY7pSHZB1QSvJlZ89z7nfvYaOYmVhTyYceSlfDt4120PbrFku+F0oCuokbLf52DFIQdBHfcxsnmmmKM71Ebds4paLak725gd9aM2LibmuwnZbr8fQkYKgj74lIUb0LuTnVU2sbkxw1v6DAdr9jFo7yVreEGdWeSOOq8jxe0UIugaGrpMf8hFLOiSbF+QNNi9INm6nPltEC4SOSNJWCCHW07oC5I4Sum1WzOb4QYOGWOuK2SE98ltVQyilqGhMEmg+k1rRlKR/SW66QmL6igb+8eZcAobGkpoolY0JAj7dazTvKpQGIb+BrmmE/BpJy2FlfZzyxgSO4/LN0jpen7HKWw1U1wgaOiG/TsJyvLOurPukq6voMguHbSk05XLr1EcYUr+K1fndufrIS/h0yKhsD2uzkp7GCBgoIkmbt3+qIJK0cZXCr2nNZbgaSdsm2XxGIlURkGp1sPa0yJSo6RDyGxiaRnIdcy/d5tV6w3GL4rwAFU1Juuev6em1qCpMXcykrC7O/MoI+UEfPfID3pQwy23V/2tEnwL+O6Oc+qhXNayU19MsYXlTy2KmzeSvlnU45VUIIcTmY0NbHKxroa+45bCgMkzCcti2V0Gr/Y7fbxBDe+Yze1Vjun2BUopFVRFipo1Ph16FIYpz/WiaRn7Qx8yVDdRETKJJi8Icb4V3pRSLq6I0xixMx1v/YN7qMKGAQUmuH5+uM6JXPmNH9qIuZhGOW7zyfRk/r26ivD6BaTugaRg6BHQdDW8a+MBuuYwa3A2Apri37+pw0lug9Nd7OcR6OHjJ95z/3X8BeGnnMdxyyLlEpABho4hZLgGft5ZIzLSpjXiJWLt5VqSyXFY3JrBcl1y/gQsUBA265wXJDRpEEg6V4QSm6UW6qQX2OuLTYVCPXHbuX0y3vCBKKRZWRfh/36wgbrnUxzr3GRU1baLNs9iCPg1d03CVN4vTb+gU5OgYSZucgMGwXgXkBQy271P4Kz2TXYckbYUQYj2sK0A+ZEQp81aH20zoAm1XzM6txHEUI/oUUJwTwHEV3y6rpbwhxpE79WFRVZiCoI/aqEnA0FEo6mImBc0Vd3VRk6a4haZpVEeSlDfECCe86gfL9lamN22HhL2mYtC0XfyGhk/XiDf38LQs1bwqLjha89lV11usoTGRvedceJSmc9VRlzH+hzf5x6Hn0hTKz/aQNmupBRdW1MW8aloNcgMGtqOIOw6xFv1fU3VIqdYKWosFR1pWmjvpatzOzSWrDicpzgtQkhtg5/5renqlPmdqIybb9S6gvCFOU9xieW2MqqYkh4wo5eS9B2ZMO5u3OkzvwiBJ005XDeuaTre8AHrzghK1EbPNxWGEEEJsXjZ0MTCgw4W+lFIsqAhjO4phPfPTt7fc7wdzqjh8h16saoyn2xdYjktV2Dv5X5DjZ2jPvPSYNE1jSI88qsNJltRE2aW/l8wNJ2xWN3lFDgnLu1+PggC2630/2q5iVWOcWauaiJsOy2uirGpK4Db34tSBgA9Q3rRub4aLYlF1hKJcPxVNSWJJm5jlUB811yx4i7Q3yLZ3t92P53c9kg+H7cVHQ/fM9nC2OA1xm4Bh47peUVDLpKujINFcYRsIQmHQh+0qaqMmEVMjx+9jmx65RBMWS2tjWOuYpGVo0Kcoh90GlFCU652QSfW2/mqJt6DuLv2L058H+UEfvQqCLKr2+lhPGDMCn08HvM+mPL8vvY8cv07U9NZg0TQNVyl0XSM3YJC0HXYdUJyx4NmWSpK2QgjRSesKkKevaODBDxfSpyiHvsWZCd3yhhghv9GqYnZJTZRo0iZu2Xy1xMTfvNpnUY7Xp2dxdZTKxjhBn44DzX2IvKq5gmCIpO3QEEvy2cIaHNelPmZhOc2r1WtQkuOnKa7hKO/xINVPU2G1qCLUNbCbL7dcUVSlNhCbnM+xufirV4j5Qzyx9/EAzO49jKvGXZbdgW1hbFel+zJHEja9CkM0xJzmhRg8ijXJWUXbC0CvWZhEdeqIUAN6F4e44vDtGNm3KF2Rn/qcWVEXw7Zd6uMWtuPi03WKCvyE/Abd8oIM6bEmad9yqmtNOElByI9haBiaRsCnYzreZ0P/klwWVUUob4hvsX2/hBBia9CZFgftfd73K85pVSmb0hS3WN2YoE9xKN2Op639/naXvpy1/+D07LHqSIKE6TCgey7DW7QlSMkN+uiRHySvOfHbpyhEzLSpi5rYjiI3YFCaH8TQdXRNYWiwOpxEAYO651ATSdIQ91ooOC2yr5ajCBje97Ld/PVbEzaZNq8aQ9coCPpI2Dam7aTbGSnA13zyVZoGbRrbVy3h6o8nc9lvr6QhpxA0jb+PvTjbw9qird0Rq+WnhFLe8V7CdLAdF8tRzUlSg7wSP7kBH72LQiyvS6DWUWfrMzR6FgRbfV7YrqIxbrJdrzXHzHXRJIurotTFTOKWzfLaGCiNk/bsz7DSAvoV57BTvyKW1kaJJG1KcgOYTpK45a3NkLRsFBo+Xad/cW7GjLMtmSRthRBbrfXtS9tRgAwQM22qw0l2G1DcqjJh5soGasJJ9tqme/q+ZXXegmC2o1BKw1GKkN/AcryelK7y+s3qGpTkBCgtDOEzNOqjJvVRk1jSIW7axCwXXTMxdC8xq+FVETpRk4Tp9Sty3NbTuVv+nLCbE7Z4U6qDho5rOa361YpNY3j1cu596152qlyMqft4Z8T+rCzqle1hbZEs16vW0TXvZEZZfXy9zlP4de8gUNO8g0DL9Q42O+qOoGte76/B3bxVrVseUJc3xJleVk912Ovflx/y4Q/5sByXSMImabtML6vPOBBPTXUtDPpIWC55ISOdsNU0Db+hE03aGLpGzLSJtqgiFkIIsflZV4uDnIBBZfMiYmvTdY2xO2ZWyuYEDOKmw6LqCD5dY7teBW3Gui33O6J3IUMOyk/Hx//v2xX0LcwBDWoiSQLNPSc1TSNuOvTID3Lsrv34fFE1i6sj1EZNLNslP+Sjd2EOIb9OY8yiLpqkMWFhNcemP5c3oXsdwTC0NYnW1Pds3PaqbNG8RXBdFP7mVXFroiaO610O+rzZZS1PxIpfl+E6XPDNq1z2+YsEXJsJnzzLtUf8OdvD2qK11c5Aw/vb8XrarilAUIBq/vtQCpK2w9KaCFEzxO4DSgj6DRK2224tgq5BwNDpW5zT6vOiKW5hOwrbVTTFLSzXZWZZA3HTIT/kIy8YpDZiMmd1I5O+sNOtEo7YqTfzKsPMLGsgaTnkBHTcpCKctFAKuuX5OWB4D05pMeNsSydJWyHEVmlDFm7oKEAOJ2zCSZuAr3nlzRY0TaMkN8CiyojXU1Z5X14/r24iYTn4mpOtKIWrvC83L5Hr7cdn6ESSNqopTp+iHApDPioaE8SSDqq5vM9vaDiuSk/5Bu/L2O1c0V+a0bwvtLarCcWvS3cdzv3uNa787DmCjk1DKJ/rD7+IlYWl2R7aFq25E8h6M5r7Qxs6NCUclKZh2Z1buKFPcQifobc6oA4nLFbUxnBcl+75wXQQHPQZBPJ0aiNJyupihBNW+j55AR+m7TK7tpG6mEl9zPvcyPEbdMvzo+sahq7juIqgzyAvIOGfEEJszjpqcQAQN50OP++HlRZkVMpWNiUI+gxG9i0i5DcI+Y0277f2fnVdY0C3XPoV5/DJ/Co+nl+FUs3rJ/gMuuUF6FUYpCbitQb7dEGVN8vMsvHrmvfdZugopVheF6Mh5s0uSX0n60B9zELDq9xrL9nqNm8b0HVsx6Eox0/ccnCV1x4s4NPJCxhYjonlyiSyTWFobRn3vHUfu65eAMB7w/fh3gNPzfKotnxthaAaXiV6WwK63nzc520QMHRcF+ZXhulbHKQpbqXXc4DMv508v47f0NE1L0mbOklTG0nw08pG4pbDrJUNhPwGsebS397NxU9J21sHYljPfCrDyXQ7l2GlBRy3Wz9WN8RZVBUhabuAIjfoY7cBJZy1/2D2G9pjq6iwTZGoXQix1dnQhRtSAXI0aQEapuOmqwhMx8W03XTwubaCkA80KG+IsbDKpaopweqmOJajSKaqXDVvejYa+HQN0/aSQoam4Tc0YqZLVTiJoYGug+t4iVVfc1Wfs9a3tKMgabvtJqPaSub6mqdU2+66JsOIjW1wXTl3v30/e5TPBeDDoXtyzdg/U1XQPcsjE+0xdK8y3XIUeQGDUMCgNmxmLFbWloCh0684l5C/9QF1JGkTt5x04NuSpmkE/QbhhE0kuSbZG7e8Kv+qcJLcgFcVYejeiaak7ZATMOhXnEM4YbNz/6Ktov+XEEJsCus7a2tj6ajFgVKK1Y0JdupXRJ/CEGV1sTbHN6y0IF0pm7q9T2GIxz9dss79rv098tH8Kr5cXEtlUwKFd6IxotuUN8SZWQZ+n8aPK+rR0CjK8VMQ8uE39PRCvHUxr02C66p0BSB4VbWO07lFcJUCy3HRNQ2rOfGbH/Rju6mKPy/OXbu4QmxcmnI5+/v/cdWnzxKyTZqCedxw2AX8d+TB3sGO2OQ6escH/Dohn4GrXExb0bsoREPcYkVdDL+hZbQVgTVVu5oGLl7h0TdL68gP+uiZH6Qg5GNeRRilFP1LckhaLpoGdTGToE8nYbmE/DqRhE1podeGxVuA12vnkrQdps2romdBiCE98ojb3jF2wnLoVRikd1Foq0rYgiRthRBbmV+ycEO/4hyKc/x8vrgGHe+Mv8/QKcn1qghM26V7fsBL0K7Fp2uEfAZzVofJDRj4fRqp1G7qSzB9WYFpq/T0rYTtoBs+Aj6NcNwCzevlo2lOuodXW6UHCtqs+NPwprMYOthOc3VCc2Wt7Sh0XWF2plRQbDR5yRj/fe6vlCTChAM53HzoeUzZ6XAJbrs40wEnYVNaGGT3gSVEkjY/JOqIWi4+fa3e0M00oHueH9t1GVaan3Hg67qKSNJG173kbV7AQNfXnARSSpG0XHIDBvnNnzOuq3j/5yoKQ36c1BQ0R2E6Cp+mEU3a3skdXad7fmCr6f8lhBC/tg2ZtbWxdNTiYHVjgm55AbbrXcDjny5pd3ztJZzXtd+1v0cWVIR58MOFNMQtBnbPJRy3CSctInG3eSYZpCaH6Do0xBRBv+71lIfmitiOE0udSbN6U78VIb+PuOU2t0iAgKERtxyiSQvH7VwCWGy4i77+NxM+fRaATwfvxoQjL6WisEeWR7V1WbtFQkfvd9N2CfkMdE3HZ3gnTSIJC9NWaOgU5fiImQ6WozB0jZyADw1FOGnjuC47DCjBcaE6kmRpbZSk5VKU6+fAYT3RNI0ZZQ00xEz05sda1RAnL2hQEPIztKd3LJ5quxJOWnw0t5q6qMm2LXrhghcDd3ScviWTpK0QYqvS2YUbvF6yWkYgu6QmQlUkSdx0MDQozPWjobGqIU5ZXZSCkI/cNqahpSoTinP9mGEXlEJD8xZMWCtyTPUaSo8Jb4GkWMLrR2kYGrl+HV3XCCed9DbrG3zmBgyCPq96N1WNq/ASwLYpCdtNLRrM5dF9TuSgpd8z4cjLKC+SdghdmQ74mnvZ6ppGz/wgpQVBaqMmuSEfMcubfplanAxIV9/6dUDTyQv42Kl/UXqfqYP/n1Y2EDMdkpZLJGlTmh8kP+QjlnQIJ238hk7/khwKgt502NRn2vBe+ViOYlFVhIqmOI1xy5sN4NPxGRrb9y3gpD0GbPREQraqzIQQIps2dNbWxtRei4Od+hWxXe8Cps2rand8h4woZd7qcLsJ3bX3GzB0BnbLZdSgEoI+A9dV6cUz//1DGdXhJL0Lg4T8PgpDfhZXO7iu07q3pgLTcalqStK3KEiiOWELbffiXF/5IR8aOvUxE58OmqajUDgu6SSxJGx/Xc/vdhQnzJ7GU3sew4u7HCEFCJuYQds9bdt73ydt10uq6hqFOX6a4t6MsRy/juUqgn4fPkMnlnRIOi4Jy8Z1FX7dW4BsUPc8AoZOr3iIaMJm5soGuuf56ZYXQNM0dh1QzA/L6qhsXq8lZjm4yk9xbiA9hlTblUjC3uAFFrdkrefwbkKPPfYYO++8M4WFhRQWFrLvvvvyzjvvpG9PJBJcfPHFdO/enfz8fE444QQqKysz9rFixQrGjRtHbm4upaWlXHXVVdh2Zn+4jz/+mN13351gMMiwYcN45plnWo3lkUceYfDgwYRCIfbee2++/fbbjNs7MxYhRNe3pi9t2+escgIGNZEkk75Yyn3vL+DBDxdy3/sLePTjRbz49QocVzF62x70Lc7FtBUx0ybk18kN+BjZt5ABJbksrIoQTlhYjsOqhhg/rqjHcRUleX72HFxCr8IcL1HaPAXMr5OeerL2jC0X0HS8lXFRWI4iYTnUR0yvYpb1Dz4VEEk61MdsTCdzGpqEVZuIUpz403vsump++qon9zqW0/5wiyRsNwOaBgG/QdDv/QWuqIsxq7yRxpiFoWmUFga9lanx/oYV3lQyn641r3rrVdL+98dyHvt4MR/OrWTSF8uYvaqRvsUhhvbIJz/oQykvKbuwMsLKhjjRpE24ufohbnmxTsvPtG55AfYcXMJB25ZyxMjejNmhF2NH9mb7PgUcu1u/jZ5AWFQV5rGPF2d8Vj728WIWVYU36uN0dRLPCrF1WXvWVkHIj6FrFIT8DC/Npy5q8t7PlbibYBr+sNICLjpoKJcfvi1/OXQ4lx++LecfOIR5q8Ptjm9FXYwHP1zIrPIGinP9DOmRT3Gun9mrGpn0xTIWVYUz9nvMbv3oWRCkqinBazPKMz7rV9bHmL2qCdU8YythOaysj1MXtdpMwDrKK0ZIWF7lrqZrXqshbf0Sti3jVZ8G/uZFykI+g4BPQ9c1bNer7Es2z14Tv47eTTX85Yv/l14MoymUz5hzHuHFXY+UhG0WOLQ+Nmzv/a813+jSPLvTcmiMWwQMg34luQQMjVjSQdM0CnJ85Ad86JrXi7pfcQ4+Q2fu6jBfL63j5/JGFldHsBxFbcRkVUOcmkjSa/2HIidgEPLp5PoNehYEiCZtZpQ1UBtJsroxwbBSL/Zd13F60nbWuaCu6yrK6mLMq2iirC62ST6Lf01ZrbTt378/d9xxB8OHD0cpxeTJkznmmGOYPn06I0eO5PLLL+ett95iypQpFBUV8ec//5njjz+eL774AgDHcRg3bhy9e/fmyy+/ZPXq1Zxxxhn4/X5uu+02AJYuXcq4ceO48MILeeGFF/jwww8599xz6dOnD2PHjgXg5Zdf5oorruBf//oXe++9N/fffz9jx45l/vz5lJZ6B8/rGosQYvOwroUbVjfEKauLoWkwtGd+ujLhu2V1rKiNsdvAYrrnh+iWFyScsNN9bUHRGLc5cqfe/FTWyPSyesrqYsRMh9yAQUlugLqoyR6DQ+wxuITltUFipkPMtElYboerfjkumKkqBE1h2gpLbVjCNiXdUUF5K+4WBnXilsJxFTk+jaSt0qvzio2rNFzLHe8+xCFLvmdxt/4cdeYDJP1BlJbV86hiPaQCW3/zctYx06Y2aqVbqFRHkvTID7CqIUHctJsXANNxgJLcAAdt2wOFTlPC4pultUz9eTVFuQF2G1CMpmkM71VA1HSojyapizloGpTmh3CVwt+8OMzkL5dz1v6DW32maZpXKQHe51s4YVHiBNOVuR1Zn6rZrlBl1lVIPCvE1qWzs7Y2VTVYajGwlLK6WLvjA4glvT7ouw0oTsfC7bUJS9oOn8yvbvFZ70t/1s+taMKnayyuitCUMKmPmChUusVXe1wABVHTJT9ooKFhGArXXpO4XVdFbOpkaMCn068oRE3UwnZdLEdRkufF3LZqXYyweaduuhilOP7nadz4wUQKk1EqCnowZefDAXD0theyE5tOe39Dfg3QNFyl0jMtfbpGjl8najo4LvTLDzRX1eusqIuRbK6I1/HWZQgFdExHYTou+v9n78/jLLvqem/8vdaezlxzVc9z0pkTkpAQIjIFIoMzKvpTELwqPlxUohJ49HofQe/gABERrly9iA8CPwVxCgIRmacQMiedpOe5xlNnPntc6/ljnXO6xu7q6qEy7PfrlXR31amzV1fX3vu7v+vz/XwIKWZdnIxN04+ZiH1OVmO+dWCGrGNRaUcopRktuJRbEUGssKSkmLGYqofce6jMjVsHeOWVY3i2dU4Bi7C2tjUXCqH1aToFa8Dg4CB/9Ed/xOte9zpGRkb4+Mc/zute9zoAnnjiCS6//HK+9a1v8YIXvIB/+7d/47WvfS0nTpxgbGwMgP/1v/4Xd955J1NTU7iuy5133sndd9/No48+2jvG61//eiqVCp/73OcAuPnmm3n+85/PBz7wAQCUUmzevJm3ve1tvPOd76RarZ5xLSuhVqvR19dHtVqlVCqdt+9ZSkrKylFK86Ev7+fRE9V5nrbmc4rPPz4BGm6/cmyel+RU3ec/nphk21Cem7YPLiqCY6U4NN3kbS+/BIngz7+0j5lmwIa+rFEn1H2+sW+GUtbm+dsGsaXky09NMtsI8RNFkpjE+YUX5IVJnRLwHEk7UnTuuatKvZ+LFLChL0OtHeLHGiFM6MNyKaMpq0RrfmjPV3n3PR+i328QWDZ//KI38FfP/2FUWtyuOSsdmbTEKZ9agfGGdizJ7nVFrt8yQJQovn2gzGTdN2p6jEVKGGv6cg637BhiuhEy2wo7CdmaqXrA5qEcL989SiNICBNFw4+498AMs+0IKQXrihnWdYJnBnIOeycbXL2xj1960Y5eaMzCa1rX/+vqjX285cU7T2tbcDZF7umuo2dzzNXyTKin0no2JeXZyxPjNd7/xb3sGC5gLXGNm1sTXrbu4p8jp1tfrR3xjf3ThLHi1l3DDBe8eZ+v+xGVVsTbX3EpG/uzvWv9rpF87/7kWpIwTvjq3mlsKYiVYqoeEMZnH2LrdPzfpTAihZU2bbuvcW1B3rXnWJRpxmsB7dCMcCOMsjetac8vw81Z/tvn/5xX7v02AA+s381vvObtHBjatMYrS+m4cOFaEq0UfkeFk3MkY6UMM80ArTVaC/zIfNKxBZaUWFKQKM2tO4e4bH0fB6cbfPvAjMlbECaPRQhj1yeFIOdYrJuzOdQKYvZNNQhixVDeZV1fhqPlNhrjLW1bAkdKcq5F0mkae5bgF79/J5dvKJFzLP75wRM8drK2qvpysaDAbDJ1/bifboKCldZTTxtZT5IkfPKTn6TZbHLLLbfwve99jyiKuO2223qvueyyy9iyZQvf+ta3APjWt77F1Vdf3StwAW6//XZqtRqPPfZY7zVz36P7mu57hGHI9773vXmvkVJy22239V6zkrWkpKQ8M+gGLAzm3Z6NQawUdT/i4eNVolhx1cbSvIYtmJ18z5Icr7Q5UWmzcL+ru/OXdSzueXyCMFFcv2WA9Z3RkfV9WbYO5ZhtReybbJB3LZJEEynNQNbGsy3sJXobmlNqAkvQSfE0L7Stc7uIdw+nNIxXfRqhUfwqpXFskVolnEcGW1X+/J/+B+//lz+i32/w8LpdvPaNf8r/vvnH0obt04C53rOnw5bmGtIJvEYK42k7VnTpyzrkXJtqO6LcNN7X7SghiBKi2DRnpYAnxutM1X0yjsVA3sW1JYmCIzMtvrhnkm8dmOFre6f46t5pZlpR73x3HYsdw/meR1hXyXWy5i97Tds72VgyNGYh3SL30RPVZUdl53I2KrPnGmk9m5Ly7GfuhMNSrEQNdiE53frCRBHECZ4tO5Ni85k7fty91mcdyX2HK3zrwAzfOTDDt/ZP89W90yRK40jjFdutV8+WqNOltaXAscQiscLp0JgAXccS7BgpUPUjmoG55xY8h40DOfqzbur9dZ551RNf5wt/9VZeuffbhNLmD7//DbzuZ/8wbdg+DejWs5boWIPMGZv0Y8WRcou6n9COFLFS5D2bkaLHaCnDTdsH+cFrNrC+L0MzTJhp+BycbpKxLfoyDq5lLEdUZ0ozUbpXk4Jpqs62QmBu+F9sLMUsaIVGrfuiS4a5ZecwN20f4soNfcQa/vXhE7z/i3u569/3Um6FWFKcdU37dLKtOd+seRDZI488wi233ILv+xQKBT7zmc9wxRVX8OCDD+K6Lv39/fNePzY2xvj4OADj4+PzCtzu57ufO91rarUa7Xab2dlZkiRZ8jVPPPFE7z3OtJalCIKAIAh6f67Vamf4bqSkpFwMlgtu2DFcQGvY0D9/lK3cDNg32aARxjSCmPsOlVnfn2PXaIHBvNsLGrt6Yx8C5jUztNY9G4V1pSwzjYDDMy0G8g5CClxbUmnHnabO8mvW2ij6hBBITPNHKdNEis+Dj0FXgRDpbuEsieIkVSacBzZVJ/jM3/wGI60KkbT4sxe+ng++4CeIrTW/BadAz3s2mfOznrEFGvBsSdNPer60sQLjStuh07S9cmM/jhQ8NVHjvkOztKOE/qwNQpIoRTtKSCJFK0yIk4CdI/nexlDGtrA7xeyxapsNfRnaYUKUmGuC0uBISTtMeOhYles29zOYd3tJu80w5rJ1pWXDaF555enHwRYWud3ie7lRWZjro5td8j3nru25QlrPpqQ8d9jYmXp49ESVgmcvUoN1a8KN/UtfI9dyfbaEdqjIOXZHbafnfX5uw7kZxkw3AmaaIUGUUMg4vRHoSivEtc2ms0BgCYEUet69dCV0j+zaFu0oWdSstWCeXVe3zdxT5Aqo+TF7JxrG09YxasFYG9sywRJjbCmr5p1f/ghv+c6nAXh8dDt3vOYOnhjdvsarSoFT6nQF6KQzESbMObMwNyVR5lrlWJpCxiaMFX6ocCzB1qE8WcfiG/tm8KOE4aKLQFBtxwwVJFeuL3F4psmB6SY1P6IkzMRZI4iptCNyrsVg3mWmGRpLv0SRKEHWsXAtiWNJSlmHcjPkyfEaDT9mKG8ax11VrCUF60sZKq1oxTXt+bCtebqG6675E+Pu3bt58MEHqVarfOpTn+KNb3wjX/nKV9Z6WeeF//7f/zu/93u/t9bLSElJWYJdo0V2vKQw78Ksteauf987z0en3Ax48GiFdpjQn3OIlbn5nKi0qfkRu8cKtCPV2/lrRUmvmVFuhuybbPTGoG1LknFtwkQz3QhpBXEnmMjcWKUEvUxtqTBNWq0VGknetWgECVFi1H6r2TRc6ks0ECmNDpKzLrxTluZYaZQ9o9spN2a44zV38Ni6XWu9pOc03aLWEZBxLYQwnn06NpshWddifX+2F/qFAMeCJDFF71w1kdCmUN06lOPhY1UeOFLBjxJsS+JHiowr8GyTvhvFijBKcKQkSjRe58nTsYSxW9AgtAlvCGKFZ0viWBErjWvDSMGh0japugO5gUVKrqWuaSspNldT5J7JG3ytVWZrQVrPpqQ8d+hObZ2ottk7aa6fWdeiHSa9MdwzTTisxfpOVto8crxGlCgm6z7fPjDDQN5bUoSwsT/L0dkW042QZhAzVvJ69wjLElhSEClFEqpe8JdjCXSiV1yT5l2J0pp2pGkGMVIubtLO/b0rjchg7vvHChKliJOEMNH4scIS5mP1SCExoofzIXBIgX/fdRNv/u4/8b9u/nH+7NbXE1ln9stPuTjoBb92hfTxAs8SW5j6LlGaRpBweKaFZ0tq7YgoUVyzqY+qHzHTDEmUph7EZByLTQNZrt5ohAOOLThZ88m5Nn6kaAYxYaywpGAg52BbkpG8y9bhPA8fq1LzI6SA2VbC/UcrXLOxj/1TDSrtiB3D+Z7FwlzBwFDB4423bqMdJSuqac9VUPB09sJd82radV127TIPsDfccAPf/e53+dM//VN+6qd+ijAMqVQq8xQBExMTrFu3DoB169YtSsXtJuDOfc3CVNyJiQlKpRLZbBbLsrAsa8nXzH2PM61lKd71rndxxx139P5cq9XYvHnzSr4tKSkpF4jT7aAppecpEwD2TzZphwkDOYfZVsT24TxZ26LcCplpBDyh4TVXr+P2q9axa7TI0XKLjG1xotJi72STdhj3lAlRoplthsRKkXUspIBWqEBpow5wbZpBfNpmaaJBKIW0LRxLgIDoAshhU4XtuXHroQd5aP2lNLwcCMGv/eBv0nRzhHZa3K41bteHRNPxlDUbFbITwrB5IMe2kTwPHqkQJRq7M66Z8SSt4JT6XGJCxWKt+ewj4xQyxksPzEZKmCgiX+NYkowjkTjU2hGJ1iRz7FXCWKGUxpbQDBOaYYIlBGEMsTZdYilM0V3I2JSbIbV2xHjNZ+tQnrofcbTc6l3Lzjb0ZjVF7tNdZbYWpPVsSspzi+WmtlYy4bAW69s32eBouYVjS27Y2s/Jqk/Tj5cVIUjZtSrQiAXb/JYwTdswStAaghiipNMgFQLBmRW3AjPZFXQ6SklHIqgx91cpFtei4TKGuRqoBwmJ0iRKzfPGBdMITlkdhaDFNSef4pvbrgPgvk1X8qK3/CUTxeG1XVhKj6X8n7sN22SZc0ZxytIkTDRKmamyPeM1EGbSbDDnUMg4RImiEcQ0/JiGHzGQcxgpevRlHfpyDpeMFJhuhhycajLbDJmqhyRaU8zYHJhq4tqSrGMRxwopBJVmyDf3T1P3Y4YKLrsWeNd2BQP7pxpIIVbsC34ugoKne7ju08bTtotSiiAIuOGGG3Achy9+8Yu9zz355JMcOXKEW265BYBbbrmFRx55hMnJyd5r7rnnHkqlEldccUXvNXPfo/ua7nu4rssNN9ww7zVKKb74xS/2XrOStSyF53mUSqV5/6WkpKwd+ybrfOjL+3nfPU/x/i/u5X33PMWHvry/59e40O/2ZNVnuhng2pLZVkTWtbl6Yz83bhvg6o19XL6+xHDB5TXXrO9dyDf2Z9kxnOfREzVaYcxg3sWzJVIIXEuglKIemCJ5uGBUDVnXBiEIOmbw1mmEGRqjMKj5ifHV7fgKPe0u5s9R8kGLP/j8B/jb///v8Nv/8Ze9j8/m+tKG7dOEWBnfZqU1QgpsCY4UbOrPsGkgR6I1j52oUW1HndFROgp7jZTGl8uzBXnXNGMLnlFXSaCUdXBtSefUJO4klo0VM4yWPESnmRt3jh/ECeVWhBDGkqSHMP+zpEAICGJjsWBJExrx8LEqJ6s++6cafOBL+xZdy86G1Xgzns4bfKU+us920no2JeXZz67RIr/ykp28/RWX8raXX8LbX3Epb3nxzjVv2Hbpru/Xb7uEHcMFtgzluP2KMa7c0M/1WwbY0J8j40gjQhhvcNWG0rzmRCtKGC545DsbhkGcoDqbjnGiCTsJuhlbGosES0KnGSTF8lay3Y9X/QS/c+vpjnULAZ4jyLjWaevhhbQ6tkLRgoYtzFfrpqycFx56kM/9n7fyV59+D9vLx3sfTxu2a8dSp8SSk5O6Y4GwxOfiJQyohTDWKZHSPDlRZyjvknFtYqVoBklviuBr+6a592CZ8arP5sEcw3mP/VNN9k7UaUcJGcci6oQVtsKE8ZqPa0mGCy6WLXFtiWOZTRat4Xmb+xnMe2itqbUjphsBtXZExjnlrb1SuoKCk1V/UfZMV1Cwa7SwSFDwTPDCXVOl7bve9S5e9apXsWXLFur1Oh//+Mf58pe/zOc//3n6+vr4hV/4Be644w4GBwcplUq87W1v45Zbbuml277yla/kiiuu4Od+7uf4wz/8Q8bHx/md3/kd3vrWt+J5JgnzLW95Cx/4wAd4xzvewZvf/Gb+4z/+g7/7u7/j7rvv7q3jjjvu4I1vfCM33ngjN910E3fddRfNZpM3velNACtaS0pKyspZC7+Yle6gzVUm3H+kTK0d0Zd1GC1l2DlSAOC+wxVmWyFRovCjhL/+5iF+9gVb2TVaRErBdVv6+fT9x0xifKKMP2yiOt62moJrkShNPueCaKNUgiUhjM+sTOj6EoG5GQsBWpibM6wsbTflwnDzkUf4o8/exZaqUbr5jofQCi3SlvrTCt15oNMQJZq8ayGU8Z62LEkxY2NbkkorxBIWQWIKy+7Yl9sJbwkShWVJhIBixqIZJmQdC501jc5e8xaNJQVZIck4Fo4lSRJl3l9KRgoedT8iiDXDeQ8/Ns1Zu5PiW/MjVMeWZaYR4EemgTtc9NgymOsl465WDbBa1ezTXWV2MUnr2ZSU5y6rmXC4mEgpEMLcS3aOFHp+6oN5j4FtLjU/4kTFpxHE3LhtgB3Dhd7X5l2b4YLHUN7h8EybcisENJ4lyXkWoVJYQjCQdwlinyBSvaBOSwqSRC/ZMF2qThUYUYLA3JtjdXY2XQu96VNWTzb0ufMrf83P3/+vABzuX0chaK3xqlKg89y3gp/zM50L3XNNCvNax5YUPZsgVoSJ4uhsm4wtOTDd7NW+jiUJE8WJapvjlTavvGKMn7hxM39w9x7qfkzOleQzpo5UaJqtGKUxCn9LkPccrtpUYl0py2wr4JFjNWKlKTcD9k82KbfMNKotjQ3gQN49K5ut1drWnA8v3AvNmjZtJycnecMb3sDJkyfp6+vjmmuu4fOf/zyveMUrAHjf+96HlJIf//EfJwgCbr/9dj74wQ/2vt6yLP71X/+VX/mVX+GWW24hn8/zxje+kXe/+92912zfvp27776bt7/97fzpn/4pmzZt4i//8i+5/fbbe6/5qZ/6Kaampvjd3/1dxsfHue666/jc5z43L8zhTGtJSUlZGWvhF7NU0E5XQTeQczg22+Lzj46z4yUmaKfrDXnf4QH+4isHGMw7rO/LMtuKePBohVYY41kSS5jRroPTTT7yjUO9Zslw0WPzYI4wVlTaEc0gxpKS/qxDECXEiWKyHphwIqUJO6q7lRSbauHvF3xNt1iOlhmHSTn/eFHAO776N7zpvn9GojlWGuW3Xv3rfGvrNWu9tJQFCDresQIcR6AUvQK1HST05RzixChx0WBZIJWgmLHQyoS1CGEGRbOOxY7hPEfKLbKuRRgrChnz9YmtiZVJtI4So1aqtiM2D+bYMZynGcYM5FxKGYdWGLF3so4lBcNFl0orphnGWNJswVhSkLEtrtvcx0Q9JIwSBvMeu9cVVxQadibOxZtxtT66zzbSejYlJWUtWKkIYjkbnNlWyP7JJtNNo2778FcPct+hSq8m39ifpT/r8PX900bhp00Tw7YtvFgxkHWxOuG1Oc8y91PVudcmGtuGZAVCOUuYJm+Y6F7oZ8racP2xPfzJZ9/L9tmTAPy/z3s1//0lb6K1jIVSysVlKbGnBAqeZcJuT/Ms2b0y9LJUOg1gKSDr2JSyLtONAFsKys2QnNsxFtGnrL+UNjVytxbOuRbDBZf1fcM4HVHDdCPgG/uniTv2fxpBxrPxbMlkLWBDX5ZtQwX2TTbZP9UkjBP8SFHI2DiWTRgnHKu0iZWmHZ1doO1qBAXPhHBdoRdqh1MuGLVajb6+PqrVajpalvKcZLHa1e6lRA7m3QvmF3O03OJ99zxFf86hmHEWBYSZlHiLO1+1mx3DhV7xu76U4S++eoBHT1TZNZLnvsMVjldaJInG79xgsq7FrpE8jm1xy44h3vLinRyvtHnfPU9Rytg0gphWlJBzTBPka/umEVojO6Mixyptwkj1bpyr8ZIVmCaU0qfG0VK1wcVh1/QR/uIz/42d5WMAfPza2/lvL/0F42Wb8rSke754tiRRGgEEnRNGQs/DNlZmY8e2JJ4j6cvYFLMOlhDU/Yj1fRk29ue4/8gstmWSsy/fUGLfZINqy4Q5tKKEONHkPYv1fVl+9eWXsHUod2rjKoqptCIOzbRwLGmsGiyLmWZgxk47ibyuJdgxUmCo4NEMYrYM5pb066r7EZVWxNtfcelZqwHmbqgFsbFE2DVaeFqqZtN6am1Jv/8pKWvP2YggFtbBMD9o1+3cD6/Z1EcjSHo1OcBd/76Xh45WsASUcg4CwVQ9YLYVsrE/y83bB5mqBzwxUafhx/hRQpioRaPZS02CSU6JESxxqnZNp8bWht/86t/wf33r75FoThSHecerfo2vb3/eWi8rZRlE5z/bEmRsSSNIFlmDLDyX7M7zouJUEzfvWWwfymNJweFyC601ec/G7dS/DT+h3REdaWDXaIGtg1laoeL7Lhnmi09McuX6ErYl0Vpz36FZjlVa1NoRshN4tnkwR861KDdDRksZdo8VOFo2it1yK2RdKYNrm8nUhm+CzwZyLrfsNM/WZysGOJup3qWuj3M5l9r6TKy0nlrzILKUlJTnBkupXeHcFGIrZe4OWrkZdorUUwFhYaw4PtviD+7ew5bBHK4te8XvZeuLnKi2efh4lWOzLRMU1tnmzLkWQwWX6UaIJQX3H5nleKU9T5kgMc0fWwqqfkSUJLi2Rd61aAUxsnPLPFtlrNW56XZ3S7t/XkO7necks9kSfX6dicIgd/7Ar/LlnTeu9ZJSTkO3eNXahKYoDWLOOWMCxMwHLAFSgi01caxohglKQ9Yxnlx+rHj0RJVqO6IVJowUzRjXdZv72T/ZZKbhEynNcN7h5ZeP8robN7FrpMjxSpuXXjbC+v4MDx6t0AjM8KjSmjABRULWtXpFtmsJtIZ1pQzXbx3ga3unyS0zLnYuaoBUNZuSkpLyzGCuCGJdyaOgbGp+xL2HZjheafHm79s+r3G70AYHFgftjpYyrO8zSrO9kw0+/+i4sRxQmhdfOsyBqRblVkiiEjKOxJYSjWbfZIP9002agRmFFsLUxw0/mdcs6gyPLLJLsKVR1iZ6zj36An7vUpYnkjYSzd9fdRvvefl/opYpnPmLUs6JucKbs0XTnSDTNMLFDdvua+YdT5hgPqVNnZv3bLYMZsl5NkqpzjkosKVAYZ7TSxkzKVpuRYwUPS5fV2TfVINjs23KrYDxakC9HXHFhhK2lJRbIYM5F6VMw9O2zPsJIShkbGYaAQcswe51RTSanGt1gnhjbCl7loSOJVZtS3A2tjXPhHDdtGmbkpJyUVhLv5hu0E4ziNg32aDdCQjrriOIEyKlKTdDihmbF2wfoh0lPY/Il102ylefmuL+w7MkSpN1LbKuzWDONc0VrZlpBBwtt6j7EQemEyYbAe3QBDbkXYtIKWrtmEQJo7xzLU5UfYIoWdWN2pICqXUvITcdJbt4bKpOcKzPjBvP5Pv5Tz/+uxwc2EA1+/RSI6YsZu6pttQ5090AkULgdIpM0+Q1I5vtKMGxJFJr6u0YzzG+W0prWqHiu4fK3LB1gN3rChyYFuwaK/L6m7Zw685hDkw3+NCX97N/qsF053rhWJIrN5QIB3OMV9tYUmJbgks2FBnOu8y2zCaTa0kaQcxXnpzi6GyLrCPZPJhftP7TJeOuhKe7N2NKSkrKc525IoihvMsT46cmx2wpOFn1yToWv/2aK3qbbgttcAqevShod+fIKUHF+r4MDx+rgjANjWLGYTDvdbIZFI4luPdAmUMzpvEbJyYVXmmFVtDqKPK6U2TL+c0q5gfpps3ai4uTRAy2qr1gsT+/5Se5b9MVfGPbdWu7sOcQ3cbratXlClBL1LPdcw+MdZ5rSXId6yshBTLR2JZgXV+GjGsTxAkN3zwfN4IEKSUSE5grhaAZJvTnXDYN5Hj4eK0XFnbZWAmlahyrtAnihG3DBWKlcGybgZxDzY96+Q5Ka5SGajti12iBG7cNcmy2zQt2lGiFRqHvdrIlhBDESl0UW4JzsQm7WKRN25SUlIvCWvrFdHfQ7j00w2wzoJBxeoWpUorJWoAlBRv7MjSDhFaYUMqeUgA/OV7n9qvG+MJj42Rdi1LGJMR330MIgedY1P2Ymh9x36FZJms+WUcy1Qiptk8FCXmOhWdbzLbDnnLPZO2eHVGiuwHzaZV7kbBUwq98++/5tW98gl9/7W9y9+UvAuDBDbvXeGUp54ueagFNogW5jqJ263CeXaMFJmsmrGW8GpB1zBjYUMHj0vVFmn7MkXKb7x2e5Yr1JW7ePtSzFlioijpRaaM0JEqxb6rB9o7PbSuI8SPNVD3AkXD/kQoAV2/qY0N/jmYQcXCmyXcPzZJ1LFzb6hW5Bc96WqgBUlJSUlIuHF0RRNaRPHSsOm9yLEo0s82Q/3hikpdcNsqLLhnpfd1yQbsjRY91pSyqk95ezNhkXYtWx0uyO9khhKCUNaPDWmtsW/SUuOY/1bHn0r1JMIlpGJ1umqzjBIQll0+7Tzn/7J46xJ/c/T6kVvzwG95LZDnElp02bNeAbtbC+TQtzbsWUaJRnY6uY0kG8h6IkMvXl3j+tgG+uGeSqXpAO4xR0KklbS4dKzBWyvCdg2UmawH9nUDu7UM59ozXqbRCBJqxYp6BvMsVG/oIYsVUIwCaSKAZxISxYqyUIecav+tmEBImirxr8/LLx9g9ViRjW7SjpHdtmcu5ChHOhqd7uG7atE1JSbkodNWurTBe0i/mQl6Yuztoj52smpENAVEiUVrT9COU1qwrZXBsSdUPmKz7ABQzNutKHg8drVDIWGRcCzTzGrZgilc/THAswTf2TfNPD55gthWilBltzns2jiUpNwMsIcg4sqPCNV9/LiLZuSrdtH974dg5fZQ/+ex7ue7kXgC+79ADvaZtyrODbkiZlB1/6UThRyClZPtwnpxrM14LaIVGdVDzE4RIiJXGjxUDWYerN5aIFfz0zVu4cesgAIdnmnzs24c5Ntvimo3GL7DSjhjIu7iWCXuYboRcu7mfA5NNJuo+R8tNZhohWcfipu0DDBUyAJSyLjdtG+Tfn5jkc49N0J+16e7eWEJw6brimqsBUlJSUlIuHM0wph3FzDajRZNjni0YKbocLbf4xweOM5R3KWacntXNwqBdW0LNj9kzXjNKXUsykHNZ3+eRc8z9Zam6ve7HNIOE4YJHKWNzYKpJEOtFNehS6tqlalXT/E1r2IuBVAm/dO9nePvXP4aXxMxmiuyaOcqe0R1rvbTnLF3brnOle251S8BSxqYdGT/aSCmCOGG0mOGXvn8HL7pkhB+4ah1/8vmnuPdQmVYYIxB4tsSxJD96/SZefvkYH7/3CM0gZjDn8NiJGvumGiRK43We6WdbIYN5j+dtGeDxE1Um6wG2JWgEEVuHcuwaLTKQczhabrN/uk6tGpHJSr721BQnKm36sw4nqz5516IRJGsqRHg624Stqjtyxx13nPbz733ve1e1mJSUlGcuZzL8Xmu/mF2jRV586QjfPVjm2GwbrY3FQM61yLoWtpQcn23TDBMePVEj6zTJ2BINTDcCyq2wMwJmUuEHCy6OZQzTZxohjSBGa/jbbx+m2o6xBB1FL/hhgrI1Rc+h7kc0A/P5hYXr2TRdl/L8Ep03SIve84dUCW+67595x1f/Bi+JqHp5fvcVb+GfrnjJWi8t5SwQGN+87kPhwvNOdiwRlDbhY91xTte26Ms5ONLi/sOzZmQMo5Cda5mQ8yymGgG1IGIo71HKOhyYbvD5Ryd4+FiFR05UyToWUawZLDjEicLpjH8VMjblZsjusSI3bhug0orYM17DsQS7x0qL1AdCQMaWNP2ISFk4ljz1F0k5a9KaNiUl5ZlE3rVRCqYaAcU5k2NdGkFMmGi+uX+amabxluwGlO0YNg2JgmczmHf59oEZXFtSnKPUnaz7HJtt8YrLRxnIe9x3uMzG/iyebfXGloM4oRnEbBvKs2M4x5FyGx2rnmIWTlkQLWzcLqpdmZNmz7kJGVJOz/bycf7k7vdy/YknAfj3nc/nXT/wq0wVBtZ4ZSlnQ8Gz8KOEpKNS750zHb9aKQWx0gghyDoWSsPmwRyjRZebtw9x605jh3G03Gb/VBNLCjb1Z8m6NkppjlXa/M/PPcm7XnUZd7ziUj7+nSN86YnJntXBQM6hlHVoBDEPHq1w3eZ+BvMeN+8YZM/JGi+6ZJSHjlUIIiNommoEPH6ySqUdMZh3uX7LABnH4rETNSwpaIUxn398opcZA+YZ/dKxiy9EeLrahK2qaXvXXXdxyy234LouAF//+te54YYbyGazi24cKSkpz35WkmC71n4x+ybrPHS0Sl/WQXaM1yWCIEmotCJagVHMlbI2owWPehBxYLqJ0pqBnMNl64oopTk00yKIFdV2hCUF7dB8fRgnSGlS57uBYJXOayQQxIqsayGlGRkRYvEozLkqZdMQsvPLltmT/NFn7+LmY48B8OXtN3Dnq97W8/5KWVu6Vwq9zJ+7OAJcx8KSAtcStKMEP1S9MBTHEr2vd22LKFG9d8q7NutKGU5WW1TaEZv6MxydbRMlJlkXNO1I0fAT1vd5TNYD0DBZ8/n8Y8ZzMOtaZB2LvGcxWfeZaQYkWhMlGs8WOJY0Y2SJQgiBbQlKGZtEd49xCq01+yebWAJGix5XbuwzdZeG/qzNRD24YIGOz1bSmjYlJeWZxMb+LOv7sjxwdJacK2mFCksIXNtMcZ2o+NiWwJWCgmejtObeQzM8frLKWDFDpR3RjmKemmhQ9SP6PBsyNr0qtFOYVtoRGjgy0+LJ8Tp516KYdRjKezSDmKxjsb4vw8GZFkLonrqvW9eupKaVnRfanY3TWKXKgwuB0Io33H837/zyX5ONA2pujnff9kt86qqXdxQfKc8kMrY04gKlCJVGdJ47PVtiS0GcKKJY4cdGbJR1LYYLLpsH89x+1TrT1I0VH/n6QSrtkLGih2PJ3iRp3rM4Mtvmo988xF/+3I0M5ly2DOXoyzo8drxGMWPjOSbTpdwM2T/VZCDn4keKgZzHK64Y4/svHebzj06wb7LO4ydr1P2YHR2rscG8B0DBs3ngSIXJuj/nuqHRiPQ6sIBVzyF/5jOfYXR0FIBiscjHP/5xduxIZfUpKc815no1ru/LkHOztMK4F+L1plu39Rq3a+UX0w1tOFpukfdsphsBjcAEMGQc41mplGYgZzNS8BACGn4MaMJOanylHbJ5MIcfJUw3wo6a2OPeQxXiRJFzLRACpRR+bO40WoNOdKc5a0aoLbFYddBV+q1mNKb7tWmde/7ZWjnJzcceo+Fm+f2X/gKfvPb2tLhdYyzMz7lisVrWs428J1Gql4qrgIxjIYTouAiYJqmTtQhiE57iWNL48WmNLQVhZ8TTsSQZ16KUsXnkeI2BnMPWoTxHZ9unjtt9UI4SwlgBAq3ha09NU26GXDJaoO7HJrxMSAbzFjONgCTR1P0IN+8SJQpLSlxL9qYOdo4WmKoFi8ZS635MuRXiOcYD7Nhsm1YnBMa2TCja/UdmL0ig47OZtKZNSUl5piCl4PINRf7xQc2BqSa2ZRo1jmWatmDGohtBwp4TNYQwvrPlZkhfzuGmbYPY0kyK5V2LMNFU2xG2FFhSMlr0iJTm63un2TCQ5bpN/eyfbnK43OTobBvZyYAYzLscnW1RboZIISlmBM0gPiubAwUIDVqYv1eqPrgwCK15zRNfIxsHfG3rddz56l/lRGl0rZf1nGe1P/Kzrai3SeJYklLGphUpip6FApp+TKIUfpSQcSwuGS3wwp3D8561//nhEzxyvIoGTlR9pBBkHclg3iPrWgzlXQ5ON/nCE+McmG6yc6RAwbOZqodM1f1eg7c7LVZrR0zUg97UrJSiY8VS5i++sp+hvMe6JcLIW2FM3Y95xRWjSCHn2SPsm2qmQoQOq2raOo5DGIa9P0dRxKc//Wl+67d+67wtLCUl5enP3ATbS0ZPpc4WM6dCvBZebNfCL+Z4pc0DR2eZrPskSrOuL0O9HdMME6qtCK01Ugpc20ZKQaUVMlkPiBLTvCk3Iz73yDiFjMNYyWMo71JtR1TbIX6UMJB3KWZsphthr2HbZaFPkdJGUTD3ZRozTiaAKFk8Gna6cTHHBksI4kSfNughZWU4SURkmSbZ17Zfz3+97Zf54s7nc6x/3Rqv7LmL1VGld5uyfqjIOBKJUcn6UUKYGA/arCtRWiDRZkRTQ5goMraFFOBHCf1Zh4GcS7kVUm6GHR+tGCEEWmtyrk2kNAXPZn3JI1bG3/r6LQMIYSxV2hhbBKvjSRLGipmWCXXJuxYHpptsHcohhKCYsRnIuUzWfBxLkGiIEkXGtZhpmsTv9f1ZQLN3ssFg3uV112/mnscnFtnJhIkiShLCSBMqhRTRvLHWSitiqhGwZ7yWNm1XSFrTpqSkPB04k81Yl7mTY5WmAq0JI00rTFBKM1z0KDcjbMvcf2xLcGzWx48T/FrCdw6WEUC5FVLK2Li28bG9bH2RiZrPkydrTNZDIqVpBDHHZtuEsZlGcywTPtYKY/pyDpM1cx9VygTtnkldKzD3dM0psYHGWCkkSiGFyYLo1t8p54DW2CohtmyUtPjNV7+d7z94P3/7vFehhVzr1aUAdAQGtiVQSqMUvSkwMJ9zLFObhsmpc0trkJ0pMVsK2pFiU3+G/pzLTCMgTjQjBY+bdgzwksvGuHJ937zryb7JOp954Dh+nNCXdbClQGlohglh4rOuZKZhy82Q8WrQCxIXQrBrtEAjiJlphniWRAtNw4/YO9Vg21B+3tSslCa80HMsRkuLG7Z1P6YexLi2IFYwXJhvB7a+L8O+yUYqRGCVTdvt27fzyU9+kt/8zd/k05/+NI7j8L//9//m/vvv56/+6q/I5Z7b39SUlOcK3QTb9UvsnAkhlr3YXmy/mHoQcaTcIkk0QwUT2FDKOISxohFEHK/4uJZkuGCaseMVn7Ajh+2OboEwPmFlxbqOwiBRZsR5MO/iRwntMCFYKnVhDsYgXmCjUZzaYQ2T5UfJTteLjWJAamzLjJWlRe4q0ZrXPfpF3v61v+Un/3//k+N9RoHw0Rt+cI0X9twl6xgvrqjj2ZV1LBpBhBbdETATJph1BTpMiBJNPTBq9pxr7BDqHcV8JDSjJTOOZQmwLDP+VffNe6zry3DpWBEpBNV2xHDB48du2MhYKUOtHfGJ7xwh41hobZK0LSmYaYQ0oxilNAgoJor1fR5aQ6LVvMTtjCMZr/kEsXnA1oAfK7KOxLMtXEtSbcfzpg6kZJGdTBgrWkFCpDS5jhJibgCNzlhM1RPuO1TmtsvSQLKVkNa0KSkpc1lp8/R8shKbse7aPv/oBLOtkGs2lbj34Cz1IDJ1pRC0OwnuriUZzJl7XpRoWp3chURp2mHC+v4MNT+mGSpsqZlthUzUfB44XKEdJcbTXZp6fbphNrWyjiDrOqA19SBBVNpU/YRWaF7fjuMz/j01nHJhWOJzas6EWprRsHrG6tP8j8/9GQcGN/Gel/8iAEcG1vOxgdes8cpSuvS8nDuOJBnHwpGCih+fmhbT5vzVC84FIWAw5xAmGj9SCDSOLdk9VuSgLblkrMjrb9rCrTuHF127uteQKDaKVqVNnWoJyDqSdqQoN0OKGZOZsK7PmxckPph32T6c54Ejs0zVA+KOpdj6voSXXTa6aGr2dEHkYaIIY9WrgxeSdS0maj7N8MzXlmc7q9pmufPOO3nnO99JJpPhJ3/yJ3nnO9/JfffdR6vV4qabbjrfa0xJSXma0gzjzu7b0vs/WdeMIF+oi61SmqPlFk+M1zhabpnmyRI0/Jh2mOA5stfgEELgORYFz8HtjEf3Z51OkagRdKwHpBlfcWyBFJBozXQ9oNwIqLRCmkHMkXLLKBGSlUlds64k61pYYv5FeDXFqQZCRapKOAdGGrP87394D3/82bvYWJ/izff901ovKQUIY40lzdhnptOwbUcm6KQZxESxIkkUYaLIOhaOJXrnbDtKaIYJni3JORLHEtT8CMeSFDIOfphQa8eMFlwuX9/HjpECVuc4t+4a5m0v38WLdo2Qd20KGZuRoseJik/Bs8g4kplGiBRGgZv1bAbyLp4teHK8wWDBYyDr0upc9w5ON3noaBWlNa4tsKRACPN3aEeKH71+I//3ay7n7a+4lLe8eOciO5mrNvRRaUUcmm4SJ4pNAzkEMJhzWRjo2AzMw/hULeB4pb34m5qyiLSmTUlJ6bJvss6Hvryf993zFO//4l7ed89TfOjL+9k3Wb+gx/zINw7x6Ikq/TmHHcMF+nMOj56o8pFvHJp37K5YIutIDs208RxJKeOahocwKr0w1iRKMdMMOTTTZO9EnXoQEyWaREOtE4hb8CwEmkRpqq2IR49VCWJF3jVWPUpDMzil+2tHxmKh7sdEiWKiHuKHCbY01kUrQWACQZfTN3Ttj1LLr1WiNT/82Jf4wl+9lZce+B4/8+DnGGmU13pVKQuYG75nlOaaRJlNEluaZ8Okcw4k+pR4p/tsmmiYaZrclKG8y0DOZaYRcrLmc/P2Id7+ikt50SUj8xq23Wfmr+6d4uHjFS4bKzKQc2mHCbozEtqz/Apjphoh24fzvPKydewcKXCy6vc8bA9ON3Etwcb+LIN5lx2jeQbzHv/xxOSia2U3iLz79XNxpCCMFYWMTTGzuJfQDhM82yK/TJ/hucSqvgNvetObeOELX8jDDz/M9u3bufHGGwH4p3/6J/7H//gf53WBKSkpT19Ot3sGF/Ziu1JVAhij86xjEUTJvFFjANeWxErRjhTfPVQmSHRP/doNFbOlQAqBkIIoUfg6IWlrMrZp8tQDo+ZbaYUphCBrG9+eROklfW7Plji1RlgVr9nzNX7/Cx9kwK8TWDbv+76f5cM3/ehaL+s5h8AoB7rFKJ1fK62QrGO8rcpNozawLRAIpOw2dgWeI4mUwpGajOMQxAopjU2CRpAojdIaP4opZixGSx47Rgv8zE1beMH2IU52dvK7qqoD0w0+9OX9vetLGCum6oEJDIsUWpv3jZUm40iG8i5RoiFWFF2bwQGXx07WyDmSB47MEsTGlgHM+FnWtRgruozXAu47NMv/9eJd2PbiffSl7GQeP1nlD/51D3U/AmH8zKJE0fBjsq7F7rEi1XaUKhNWSFrTpqSkwNllNJwvztZmrBnGtKOY2WZEO4xZV8oAdPIXYg5NN4k0KGU28tpRQhTP39QPE82x2TbFjI0Upq4NEo0vwHMs/Mj4wi+nho2URndqTtH54EpL0HlJ9ynnlaFmhd//wgd51VPfBOChdZdwx2vuYKowuMYrS1nIwvPKlsYepOqrTiCYCbqeq0XSc15rW6Z9qzR8364hLEtyaKbJT9+0he9f0KyF+c/Mkw2fg5NNqoMRO0cL1I9WqLZN/ehYJsSsEcaMlTK88YXbcF2rFyT+1ESdqXpAK4gpZGyagbEIvG5zPwM5d0lbxNMFkY/XfEaK3pLir27GQ9cj97nOqjspu3fvZvfu3Ys+/s53vvOcFpSSkvLMobt7ttB3ES7sxfZMhfUbX7iVrGP3mhwFz2bLUI5jncCEQsbuNTomOiPLujOCsugGqY1yrzvSrLRGYsZINg7kmeokXgoEYoVd23ZkgoPQnIq877CStN2Uc6e/XeM9X/gQP/jE1wB4dGwnv/Gat/PkyLa1XdhzFA1IDaLjQ9slUqAjhfRjso7Z6NCdB0SBYCDn0AgS/FhhC4G0bJSGfMZGKUU9MA1WgTap1C1znv/wdRv5iRs39R7A59q1LHd9CSIzdlpphZSypjEMxrYBBGMlj3Ulj0o74mWXj3Ky5vPdw7PMtkKyjgmHCGPjxztW9Mi6NsMFo8S9/+gsN20fWvJ7s5SdzKXrisw2Q5phQjOITXhMKcPOkTyOJfEjlSoTzoK0pk1JeW6zmoyG88FKbMb2TtS573CZUtah1o5IEs1UI6CYcXpf49qS6YbCkoJIaYJEE7bNxt3cd7W6f+j4V2bn+NBGicKWJrvdFhCdRg079/fdIN2V1K5pw/bCcPuT3+QPvvDnDLeqRNLiT2/9aT70gp8gkSvVQKdcbGx5SnAjRUdDqxQSyUDWRSU+QWzst7rPphqjVO96VbfDhCcmGly7qY+RQoadIwUAjpZbvWfgdpjw0W+dqmkLns3Jis94tU0x63Dd5n72TzaotCPaobEm6cs6vPWlu3j55WPAqcmvv/vuMR47UcOSEMSn6s7BvLFiWc4Wcbkg8ms29fMTNxb5jycmFzV0T1Z9BvPuPI/c5zKrquhrtdppP18qlVa1mJSUlGcWp9s9u1AX2zMV1g8cqfCef93DcN4l6AQQ7RjOs2UwRxAr4lgx2zajYVIImn6MQODaoJU2KbZzjqc7x4yF6PnRSgl9GYfRgstUPUBphdDzzeNP+30TZhwkVsZgPp7TKU4btheHN933L/zgE18jFpI/v+Wn+MALf7IXQJayNiSw6InO7liIxEqzoc+jGSSUWxHFjM36jsKoFbVJlCLWmpxlGVWt0jTDbsPWTAXESpN1LaQQPG9L/5KKqdNdX563pZ/vHS4Txgk3bRsk45iHoUhpXEtSzNgkWnNouslw0eNNt27j/ffs5YG4gkBjKYu8Z3xos52GajfoYaYZLlrLcmzsz/K8zQM8crzK1SVv3vEB9k42UmXCWZDWtCkpKavNaDhXTtmMLX29bkcJj5+s8Rdf2Y/nWHiWZLoRUmlFDORO1SxBrKi2Q2Kl541ew/y6Ummj0rOlINEaP0pwLUkpa1FpRURKkZhd0RXXo7Yl0VrNEz2cDd2R77lrTlk5fe06f/Rvf0opaLJnZBu/8Zo7eHxsx1ovK+U0CIwa3pUmo0FjbLNc2yJONApNX95luh4Qq1MbL93zox0lFKRRypebIQemm9y8fYh2FM+bEuteLxDwvM39vcDd0WKGybpPK4gJMg4/dO0Gxjt/Ljcjbt01zE/csHnemneNFvmh69bz0PEKA1nzzL2+L4OUp6bETudBe7og8q1DuUUN3bkZDymrbNoODAws+XGtNUIIkmSlrYuUlJRnOsvtnl2oi+3pCuvZVsRk3afux6zfNczGgRytMOaxkzUsKejLOsSJZtNgDksKxqttjpRbWNJ4+BQ8i2rb+GbOLT5NArwZGUMI4kQTJooHjlZphBFxcnaFZhir3ohYkOiFYtsVkSpyz40P3vITXDp9mA++4Cd4ZP0la72clGXoPsRFiaLqJ7hSkHVsMo7VsSsxfnxxosi6Nq4l8VVMK1Iorc2DqTKN1VLWYSjnMNUIuf/ILK+8Yh1SinmhM7V2xL7J+rIP7hv6cxwpt0m0pi/nLlpvO4h7ljBBnJDLWOYY2hRcHUftU68PExzL2CuslLmbZRP1oLdZ1gjiVJmwCtKaNiUl5UzN0wsViHM6m7FyM+T+w7PU/ZihvMdoKUMrjDlWaRPECScqPqMljzjRnKy2aXT8Z+fWhpLF6lYpBEobuyHPFviJYiDvYAmotGPQasWKWAm4ljw1PbYKlmoup6ycarbI777iLVwyfYQ/vfVnCO1UgPBMIONIBvIulhS0QnPuWlJQ7WzIKA3VVgRCY3dSsaPEbKrEStPq+Ek3gpi8Z7N7XZGPfvPwvCmxiZp5Zi5lbWZbIYN5DyEEu0YLNAJT807UfGp+TNGzmaoHDBVcrtvSv2i9+ybr/PODJzgx22ai4mNJwd6MzY7hPJsHcwghzmiLuFwQ+ekauimGVTVtt2/fzuTkJO985zu59dZbz/eaUlJS1pDVpOZezIvtcoW11pp9kw3iRJNzJY4tsaSYN9q2oS/DQN7lwFSTVhifukkKM1oGgrznkKiIMDnVuI1jTTsyJawtJVGSUAtMs6Xo2QgS/Ej1Ck4BWNIoGpZSHiz0r124g7oS0uL27Lj10IP8zIP/xq/+0DtIpEVgu/zKj/7fa72slGUQmPOymyatlKbejrhsfYnnDeSYrPlM1n0aQUQQG6XQtZv7mawFHJ2NCaMExzapuLFSeEKSdy0a4fygriBO5nljB5HiaLnFjdsGlvTpHil6ZF2Lk1Wf9X3ZZS1huuNolWbEYM6l0gpxbUEzTAgTn3WlDJ4tmGmG7B4rcv3mpRuHy3GxN8uezaQ1bUpKylplNCxnM2Zq2jqVdsSO4TzrOhuJxYzDC7YPMlkPaPgxlWZIuRURJQoxZzd/Oe9YWwqKno3SijjRaGGsva7YUMIPE751oEykFjl3nRY/Ss46V2Gh8CCtaVdOPmjx21/6K+655AV8aefzAfjHK1+6xqtKOROy80OvgKwj+eFr15NxHcJEYQtoBIkRJqkGmweyHJxukfNsqu0IIQSWEHS3YQTmvCt4NqNFj9ffuIVHjlcXTYm5tkXOtYhixf6pJgOdANvBjg/tUxM1js22eexElXaUAAKt4TP3H+fho9VeTkzXNmymEVLM2JystNHAeN3n8EyLbUM5rtpYYqYZrXrSa7mGbophVXeePXv28Gd/9mf8wR/8AQ888AB/+Id/yPbt28/32lJSUi4yZxPutZALebFdqITzLLmosK77MbOtEM8xjRrXOjWu0R1tm21FvOGF25BCsOdkjX956AQPHa0QKo0iJpCSrGtRyjo0A9PU1ZixbaVhXcmjP2uzf6pNrBSuBX5kAo7mojGFcayMksG1BGHSHdOW+LFCSgjmCDakND5FKeeXXNjmnV/+a97wwN0AfOeBz/I3N/zgGq8q5Uw4tsCRgiBWeLbEkiYw8NLRPI5tc3imyWwrIogS8+CpFTP1gO3DeVphTMOP8UPz5CkFJInieKVN3rW5bF2RMFHsOVnjK09NLVIl7DkZc/+RWZ6/bbDn06W17l1jRooeg3lvWUuY264Y5Z7HjMXCpeuKuI7ka3unaQQJGUcSxorxWhuAvqzLG1+4bckQsjORKhPOD2lNm5KScrEyGpYSRixlMzZZM82QgZzDrk4TpnsfChPFZWNFnhyvE8QJni0ZzDmcqLSJtLnnObYg7JjSzrWKl8Ko9KJE49oSAfRlbZ44UeNEzcfvfM1Km6iq8/052+mvtEm7Ol5w5GH+6LN/yubqBC/b/11e/Mt/SWCvfFInZe2wBQjLBFvnXAvPsSllHcrNkKcmG8w2A2ZbJmD2sZN10DBWdGmHMWGiQWq0Fni2hSXpTY6++qr1bB3O8Y8PHl80JeZaEtuSSGGU+3U/ptQJxR3Mu1yxvoQtpZk27ahm856zKCfmnscmKTdDhgsO+6c0ibHaJedYBIni4EyT6UbAdVsG0kmvC8SqmraO43DHHXfw8z//87z73e/mmmuu4Zd+6Zf4L//lv9Df33+el5iSknIxWIvU3JWua24juevPM90Ied6W/t7NKYgT2mFMojRjJWO0PpfuaFs7SrCl4CtPTRHGCX1Zh5lmSKIUidIkSlHIOJQyFkqbonZdweNHb9jEtZv7+cjXDzLViKi3IyqtCIFpWNsWRHOmaP1Y9wIfuhYIGVfi2hbSkpQyDuPVNpHqqCHShu1558Zjj/HHd9/FtspJAD56/Wv4+6tfscarSlkJcaxJhMayBJaUZF1JpZXwvcMV2lFCpW3OPVsKHM8mSjRPTDQ4PNvGlaaYTAChzcOr51i4tsSRkn2TDTYP5Ljv0OwiVcL6vixbh3IcmG6yb7LB87e5zLZC9k82mWkGVNsRg3mPS0cKCCmotKJFKlfPtuZZuGwfNsEQDx6pUGkbRVQYw1Ub+/jlF+/sBT2shlSZcO6kNW1KSsrFyGg4nTBi4eSEHymKGZvrtwwwmPcoNwP2TzYpt0JipYgTRbkZEimNFIJ6EJvgTjrTXLFGSNBzXAtsCUKYoKH+nMP6fo+94w3qbcVELexl455tQzVM69cLTibyufMrH+VN3/sXAI72jfFbr/61tGH7DMFslkDWltywZYBD5Sb7phqMFj2enGjQ8GMEmuGiy4a+LI8erxqFLR5DeY/JekAQK6TQeLaF51jgGJuD268aox0lS06hFjM2AzmXyZqPELoT5ms2f4LYKHulML66l44Vlwxg/PT3jjNZMxNiT47X0VqzdSjHbDOiHSUmC0ZrHNuE7O7o1Lwp55dzmvEYHBzkrrvu4j//5//MnXfeya5du/id3/kdfv3Xf/08LS8lJeVisFapucutpatCmKoH/Nsj48y25jeSp5shJ6s+HKlwyViBdpTwyLEKk/UASwocW/K9w7PsHM0zkHN7CrlYaTKW5O5Hxik3Q0aKxi5huhkSJiDRxIkm0WEnybOT6CkFT47XGa8GzDRDXEsan0pMYi6ddE9LLLY+6CIApTX1IEZKQa0dnfU4WcrK8KKA3/jax/hP3/1HJJrjxRHe8epf4xvbrlvrpaWcJUpB049pBubPR2ZbxIlGCsh5NsWMw2DOoR0lHJppUWlF875ed96jlHEYKXporTky26aUDZmstRepEozXV5FyM+TwTItS1uHYbIuGHwOC4YLHpWMFxusBAzmHH7t+I8NFb57K9Ynx2qLieftwga2DeU7W2tTbMZN1nze8cCuXjhVRSqeqhKcBaU2bkvLc5nzazixU1C5McF9KGPErL9k5b6rsE985QsaxKDcDHjxaoR0mFDI2cSI4Vg/wY4UlYSjvUmnHRLEi0UZZa9Zgfu3WppEy//lxRK0VMVnziRbUoWfTuF2NtVfK2XP98T388d3vY8fsCQA+fu0P8AcvfTNNL92wfSYgMM+Sti3oy9qMljwSrRkrZnjoWJW6H9KfcxnKu4z1Zci5NpcrxbcOlDle9XEtsKTEsYzVQcaxiBLNrtECb33JLnaNFjlabi1p79L1ry03A+p+wnQjYP9kg6lGQDOIsS2JZwmu3dy/ZI5DN4Ax0ZpixqHcCilkbDzbIutYhLEiUopmkHDNxj5mW9F5D2tMMayqafu85z1v0T+s1pogCPiN3/iNtMBNSXmGsVapuQt5aqLGp+47zv6pBrFKmKqZRutN20/5SxYzDs/b3A9UQMPhmSZ7JxtEsWIg52BJSSljM1n3mWoE5F2LWjui6kcM5l0+9u3DHCq3KGVsHjpWJVaa0aLLVD3sNVyDWONa4DkWfVlj7t6fczgw3VVAJNT9GCkEVkfVhzCj3HOZ28BVQBBpPEf2/IXSQvfC8If/9n5+eM9XAPi7q2/jPS//Repefo1XlXI22NKcP3HHFNrG+E6rzhimY1nYUjKQdcg4FhO1ADUn1E8CWphNlaAT0uLakkRp+nMulhBU/IhNg4t/LgbzLtdvHeB7h8vsnajTChP6sw6DBY+dIwUG8y5aa/ZONnj4WJW3vHjnvKbrct6IUgqyjsXBqSbTjZBPfe8Y/7FnasUWNCkXhrSmTUlJ6XI+bGe6itp9k3Vm2xESo2zLOPa8CbGFwoi3vLjA5sEcSmmOzrboyznsn2oQRAntMGGwE1g5VWsTJZqBvEu9bTxtHUsykHOYbUcodapx21PeLiBhsSWXYPmG7VIfT2vYC8/O6aP8/d/eiaUV44VB7nzVr/GVHTes9bJSVojAqF0ztiRMNM0g4d6DMxQzDrYlaIYRQ3mP9QMZWkHCnpN1WmFMtTPJaUkzKVbK2EZYpCHn2mwayPIbr9zNpetM3Xg6e5eBnMNoMUPBi3lyvI4fJeQ9m21DefKexaMnajw5USfv2T1LsC5Z10JphSUEdT8iVgrHMu1DIczaiCHj0LN6ON9hjSmGVTVtf+RHfuQ8LyMlJWUtWW1q7mpCy5bji3smeP8X9zJVD3Bt479TaUVkHIuHjlW5brPo3UyEEFwyWqDcDMk6kiBW7BopECvNQ8eqNIMYSwqOllskWmNLE0rmR4p79kwQRAljfVmq7ZAk0UQJ5FxJEGuUMjfFJNG4GZP6/tREw9z0Ch5xkjDbinp+PkqYm6rQel7omAVYNkTxgiRf0Rn91mm5e6H4wC0/yQ3H9/C7r3gL/7HrprVeTspZImGe+kdgRrcKGYd2FBOrBCmMl/RsK0IImG2FzHEnQWH8w+iMpAWxZrIecNm6ItuG84xXfaQQy4bOZB2LbYN5/FgxmHcZyLkUM6cK4dNtZi1XPM80A+49MMNMM2R9X5Yr1pXwY7XmFjTPddKaNiUlZS7nYjvz1HidP//SPo7OtowKLU5ohDHlZkTetRnIOz3LHFh8L5kbjjndCDgw3aTaihgtuWjM5EnNN7XxaMFDKU25FZKxLSMKiBX1IOkOgM1rtgpACHCk+ViYzF97V2V7Vt+rztelFe2FYf/wZv7pihcD8P/c9svUMuno+VrSDZpO1Mp+5mUn6Nq2JH4c0wxjbEtwyVjJ+NpaknoQM364QsGzGcy7zDYV7SgxPrZAFCvqfoQEhJS0Q8WG/gxf3DOJbZnpsDPZu2wezOLZkgPTTTb2Z/Fsi2LGpu7HHJ9t0/TjeUFlXdphQn/WZaTo8eREHUsIokTh2RYASilmmxH9OYd2aEKBz3dYY4phVd/V//pf/+v5XkdKSsoasprU3HMJLVvIU+N13v/FvYzXfNaXMji2pNaOCeIAKaDaihbdTLKuRXU6ou4LLh0r9tZ93eZ+9k3WeeJk3ShfBZSyFmPFDJYUTNd9yq2Imh+TcSy01liWJOPYeDY0gog4Nl6ao8UMhYxN3Y94YrzGQ5EmjJOeaqGrYFCJXtSYtaTAFgItVa8BpYEgUrR1qrI9n+yeOsQNx/fw8eteBcDeka285Jc+TGylhcPZYgNruUduS4EQ9CwQVCdUxXMsBnIuYU1hCUWsICME7Shhpql7xS10FEadrwUTBJgojWdLdo+Z4rY/6zBSzHB0trVs6MzGgRzjNZ9NAzmsJTajltvMWqp4bocJX3lqknIzwrUlzTDmgaNVdo0WuGS0cFEtaFLmk9a0KSnPTs6nsGAlPDVR4z3/uocnJ+q0wxgNFD2bvGfjWgmtMOY7B8qUsg5DcxRt3XvJwnDMDf1ZHEvwnYNlphshUaKxhMCxpLFY8Gz6sg6zrZAgTrCkQGPugZY8ZQ+kMUICRbeJa+6JS7FcA3a5mlUK4xCWcn6wk5hf/O5n+NRVtzFVGADgHa/6tbSefZrQDZrWSpOc4bVSQMaRaK1phTGtwFwTPEtyuNxEK2gECRpNECtyrqYdGju/aE5NqzHiAwFkLE3esxjMu4s2/E9n73LNpj7+4f7j7BwpzHvOL2ZsBvIeJyptZhrBvKCyuQGMt10+xl9/8xAnKz6zTRPI2/BjJhsBSoHSim/uD9g5UqAdpUrbC0F6BUhJSTnr1NzzGVqmlObvv3eUk1WfgWxn5ALIOpKMI4kSTaQU5QU3k3aYmFR4rcjNaSYP5l0uHSvwxHidjC2xLGOMLoVgphHgRyZwLEw0QaRwbYGKkp7EoHujFJ3GaxAljFfaNDrKhaWY+2GBGclWSqOEMA8Ic4rj5TxvU84eSyX88nc+za9//eNYWvH46A4e3LAbIC1wV0nC6oJIVoNkfv6ewJw3dI4vhfmY3dlAKWVtGoFFO0qIlQKMur0RLGiaduRFXT8/1VG1SyEIE8VsPTJF6BWjfPSbh5cNnXn55aP8w/3Hz2ozq8vc4vmBo7M8fqLGbCuilLUZLWawLclU3acRxFy3uf+iWdCkpKSkPBdYjbDgXJq8+ybr/PmX9rN/qg6dCS/XFvixIkxChDBjzc0wZs+JGrfuGu7V2l2F2lLhmFsG8xwtt2gGCQM5l13DeR44WiFWiiBKcCyJZxufyzAxH7MswVDe7VgH+YSxwurcE5UGrfSyN/nuX3eltWrasD1/7Jo+wnvvfi/XjO/jhmOP859+/HdBiLSefRohMKICRKeGPd3PvwZLGJFA1Y9BgCsF/XmXvGcTxYrpZkAzSCh4prZtBPE8EUK3Hu+q5INYU2mFOJbkktHsog3/5exdnpqsLzlR2/W8rfkRM42A2VZIzlscwLhrtMibv28bGUfypScmOTTdxI8UUsBA3sWWEtszpiwf/ebhdHLsArCqq8DAwMAi/6+5lMvlVS8oJSXl4nM2qbnnO7Tsm/un+eKeSZpBRJQkSBGRdSwGcjZZx6KuYvwowbMlYceAq9tI3jlaYKoWLGqqVNoxUaLIurbZoUyg3PI7Ix0Szxa0IrNL2o71kuNdYaxpBTEzzZCqf/r91LkBZN330NrsPEZn2opNWRU7Zo7xJ3e/j+edfBKAe3bdzPHS6Bqv6pnPhXj+cjpN1IWBJ90/mpCGjiK242MrMRsw9SChlLEpZc1m0mDeoxHENAJtUmsRJHr+G2s651/nvSMFniWwpeB4pc2mgVyvCD1d6MyO4QIPHa2ueDNrIbtGi2z7/jx/+PknmWkE2BaMFjNY0hS2bt6l3AzZP9Xgus19BHGSeoGtAWlNm5Ly7GI1woJzmR7r1sUzjQDHkvhRjOtIbGlyD1phjOre2wSUm2FPhNC9l2wezDJV8xdlSxQzNsOFDEHcotaOODDdpBbEtBsJniXQGopZm7xjI6WmFSYUPJuxoocGpurm3pcoPa8+XfbvkjZhLzpSJfyn7/4jv/G1j+ElEVUvzz93LBFSnj44EsZKHrOtmChJEMJMqgaxohnOr0MtYZqsrShhphni2RIhzGZKt570HIvRosfBoEkzjBGIeQpbMO+hNUgJQptnzXZoNmeWs+layt7ldBO1g3mX3WMFntCYUN/p5pIBjLtGi/zOa67g0rEif/rve6kHMZYQzLYi+rIOV27sY9tQjqcm6vz9fcf44es2UMw4F3zC4bnCqpq2d911F2AeWn7lV36Fd7/73YyOpg/LKSnPZFaamns+Q8v2Tdb5+L1HqPkmRMG1LBDGYzdMTPhPEJtUysTVPSP0biP5dddv5p7HJ3j0RJW8a9EIEsJEmREUrQmVouTZ1P3QNHEdSaw08WmaR72iFjg621rUaFqKZUMe0obteUdoxZvu+xfe8dWPkolDam6O37vtl/n0VS8z1U3K047uaNeZXqO0UdUmHRVDPYgpeg6uLXuvyziSUsbBloKaH5MoteRDZlel2z03ldbYlmD3WJHX3bhpXhF6utCZlW5mLcfJms90I+CS0SLt44pYaazOX0cIQSFjM9MIODTdJIg1tXaEUjotcC8iaU2bkvLsYTXCgnOdHptbFx+fbZMoTabnf27sffxImWkSDVGSmE1HQe9ecuO2Qf7xgePzJsfM1wt2juaZqPucqLTxo4TBvMN0wyh0LQkFaRrFM82IOFa0RMShGYVlCTKORYA55kpYaQhZyvlh6+wJ/uTu93Hj8T0AfGnHDdz5A7/KZHFojVeWshDXlsQK4iQhUebcNup2hSMg69kopfBjEzrtSkkQxySd586MY1HMzveMzbs2nm3qyqXOsu4GS1dpqzVoNFGn8F3OpmshZ5qobUeK11y9jh+8dgOtKFl20uDAdINv7Z8BNJsHcniOhdKaMEo4ON0EYKoe8Fgn3Gww56Zhu+eJVTVt3/jGN/Z+/7a3vY0f//EfZ8eOHedtUSkpKWvDSlJzu6FlWSdDrR0RJgrXkr2QnpXeQLqFdTOIGcq7VNsxQZyQdSyyjhkTaUUmBMiP2mRcm+lGQMaZ30iWEvaM1/j8YxOdcC9NFCuiRKO0ws0JZloxjmV8hZpBvGgnE+bfELsF6koatikXEa35y0+/h5fv/y4AX932PO581a9ysjSyxgtLOR1nGp/Unf8JaVQG0hJYAmwpGS66xIlmphHiOZIgSpBSYFuSoYLH+j6P8apP04+phwlKd3xwO+/dPZdtSzKYd5mqB9zz2CRSiF4BebrQmeU2s67aUOLazf3ESnO03FpWSdC9Xm4fyjOYc5ms+7h52SuY40QzVQ8oN0MG8y6f+M4RvntwNi1wLyJpTZuS8uzhbIUF52N6bO51fiDvMt0IOinr5k5kCYFE4zkWWSERQjJR8xnIub161rMtPmePL6mE68+6vY1I0LRDRc6V5B0TcDTbimh1am4hzaSYUgmOkpSyDrbQNBamjp0FacP2wvCCIw/zfz71e+SigIab5d0v+0X+7ppXpAKEpyl+qPCjwAgMJGQcG6UVSefPQZx0LPI0UaSIhDIZDXFCMWOzaSBPsyMq6l5nXFsiMCr4vqxNGCnasepZIvTOvTnTY1II6u2YWjsC9LI2XXNZyUTt7VetY8tQftn36E0UNANynsVA3kV2/h7asxmv+nznoE/etbEkrCtlyLlWGrZ7nkhNUlJSUuZxptTcvGsTxopvH5ihGSbEicK2pPHaGi3gWGJFN5BuYb1jOE8Ym1HnRJlgIdeWOJag4Uf4UcL24Ty/eftuRkuZ5X3Genc3E9Lg2hbtKOZYxSdRGrvz+mAJryA6v4o5abuQFqpPO4TgSzufzwuOPMJ/e+mb+dvrXpUWt88wLHHKumAu3aCU/s7oWJgkDBc8rtnUz7HZNkfLLep+TNa1QGlKWYebtg0ymHf52t5p9vp1+rMOzSAm1rrnLQ3GdmG44HHF+j5sS5x1AblwM2uqHvDQ0Qr/cP/xM47RdkfS2lHCztE89SCi3AwpZGzixPgntsOEDf0Zbtg6QMZJC9yUlJSU1dJtoC70buyyUFhwPqbH5l7nr1hf4vhsm0YQU/CMPUKYaGKlGc06jBQ8rtzYt2h0WCm9pBKu3Ax49HiVI+UWaKi1YyxL0Zd12DhgmiIzhytoYPNAhiBSTNYDIqVJdELS6rZ6Up5uPLzuEqZz/RzrG+Mdr/41jvWNrfWSUpaga6GnoPdgKIQgVgrVsR2JVccrGnO2WUJjWxKtzeTlSNFj50ievZPNXg3oWJKmH5umryXIOha2JfEb4aLnz0SfetzRGMXryWobBbxo1/CyNl1zWelE7XJ0r5Ub+rKdYESFZ1u9z4fKTLoO5hw0xuJwtdaJKYs5L03b03mBpaSkPLtohwlT9YDxms/6UgYnYxMlmqm6T92PGMi53LJz6Iw3kG5hvcHLsmu00AkTCgljRRibsLAgUWwtZfi12y7l+y9dPK7a3fVLlOb2K8Z69ghNP6Z9qEw7jAk7N1GVnPLzcqS5biVK98anu1cx0TGXT8MVnh6M1acZblZ4bN0uAD523av44s7np+raZyinyT/BsgQjRQ9LCuo+jBQzvOnW7UZVEEQ0/JhmEPPJe4+yoT9DKesCsHEgy/6pBkprihmHOFEUsw6VVoTWmq1DOYJEkWjNQMZdVQHZ3czaN1nnc4+Or3iMdu5I2iWjBa7b3M/+ySYzzYCpekA7TBgpurxw13AvTTwtcNeWtKZNSbm4nEsA2EJO590Ip0K/au2IJ8ZrjFd9WmFMwTPTXHMnx2Bl48cLr/M37xjiOwdnaHY8H6NEUfBsihmbrUN5fmKORU+XpZRwfpTwvcOzTNR8EgVZx8JzJHGiaIUJh6ZbVP2IMFZoDcdmjUgB6Cnzwlgbe6DOMcKl/IRSLg5a87L93+VLO29EC0nLzfKTP/M/mSgOooU889enXFQsAU7HNzrqPENqwBZGTRsnuvesqBd8nW1LlIJEaRzLiIlaoeLaTX3sn2oy2wppBDFxYuywdgzmGMy7nKy2qbXN1JnAKHhjdSqrAcCWAs8W+FFComGiHnBgurEqEUL3eguYwMPTXINPNzkWxoooVkghaAQJmwdzFDOdcPGztE5MWZpVNW1/7Md+rPd73/d5y1veQj5/Sk79D//wD+e+spSUlKcdSmnueXyCUtYhUcooCYTZLcx7FuO1AEsKbrti9IwF99zCejDvct3mfvZNNig3g96NbCDn8q5XX8ZLdo8tWdTPVUhIKSllJTONgG8fnGGmERrzduhJars3VSnMDTTQybwAsa7aNhUlPA3Qmh95/Mv83j3/i1qmwA+86c9oejkQIm3YPoNY6IW33OOia4FAMNsMcW2JLSXXbOxn80Bu3rXkifEaliXIe6cexkcKHkMFl3aoCJOEqKN88BzJcMHFsSWJBrczqrraAnI1Y7RLPYhft6WfQ9NGbbGhP8MLd55q2J7L+lJWR1rTpqSsHecSALYUZ/Ju3DvZAA2f+M4RgkRRbUU8NVEn41h4tpw3OTaYd2mHyRmnx5a6zr909wj3HZrlRKWN0hqNphWaMNzlmKuE2ztZ56FjFRp+jGtL3FiRdWUn/0HSCGKaQUKY6J6tV6zM7y0BriN7nu+2MHVvkjZs14yRRpn//rk/47b93+W/vOIt/L/XvxaA8dLwGq/s2c9yE15nQulOo3TOM2HXSi/pPDQu5fcspUAg0NrYHBQyNqWMjWdLZpohl60rECtN3Y85WW2D1ly5ocSmgRx1P+bAdIOHj1ZpBHHPu1Z0/mcLc7zJesj24TxXbigx04xWJULostJr8OkmxyKlCWOF0pqca7FzJD/v2rtS68SU5VlV07avr6/3+5/92Z89b4tJSUlZOedTmbDSY+2favDw8Qq7RgrEQzn2TzYpt0KjJpCSTQNZBnIuWefMl5b1pQzDBZfHT9bYNVJgIOewcySPHyUEUUIQG8+uR4/XSBQ8cbK+6IZyybrCvDG4cjPk3oNlys0QISDrSuJEY0lhbnydDm2kNBnbNEek0POSdHsq3DnK3JSLy1Czwh984c/5gae+BcDBwQ30BQ3TtE152iGFKW6XCytZSYhJ169vphkwkPO4fmsft1+1OORrKRVVMWOzrpRlsu4zYNn4kWLXaIGD000Knt1Ltg3ihFrbvH41BeRqx2gXjqQFsU8QG7/uG7YOMFTwFh4qLXAvImlNm5KyNpxrANhSnM67ce9kg/Gqz7pShoG8ix8p9k82aEcmGGzrUA7Hspiq+zSCmGs39THTDLl6Y98Zp8cWXuenGwFxohgrZdgxkmdjfw5LmnDKj3zj0LJ/t12jRdSVmmOzLVphghTQCmKkFPiRwpam6WvCjTRSmMmwrrWXJQVKa6PwkxI/ViDMJmbNT9Nx14LX7vkq7/nChxjw6wSWja3Sf4eLyVJh0SvBBORqMpZFgPk3s4RRvkoBUprad2GBGysNJCAEGctstCitec2169k73ujUgWYz6PsvGWGmGXKy6gNQyjpct3mATf1ZvvjEJDONANe2ADNJ1p91sC1J3Y/IOjaDeQ/Xtla9yX821+DlJsfKrZB2GBErzUDe5fnbBxnMz69rV7L5lXJ6VvWd+8hHPnK+15GSknIWnG9lwkqPNdnwOTjZpNqKuGSswI3bBqj7cS+MLOtaHJ5pnrHR0H3PA9NNjpRbHJhq0p9zzHhFZyRkrOSxe6zIdw6U+ccHT7Cu5LG+L0MpY1S+jxyv8NREnTBWvdG2fZMN6n6EBhxpSlgpjRKuGcb4WoE24yxhoillHBpBhB/r3u5pLuOgEoXWCWEaRHbRedUTX+f3v/BBhto1Qmlz1/f9DH9x84+TSOvMX5yyLF3F+YUQ2lid99ZLvLfGFLdzle5gCl/BqWI6iCFMYqSAbcM5fvXlu5a8li2lohJCsGu0QN2PmGoEbBrIsq4vw4HpJic6ntZKa+49WO6pqNb3eWddQJ6tV+JcFo6k1doRn/jOETLO0j/XaYF78Uhr2pSUi8/5CABbjqW8G11LgjbhOM/b0g/AnpOzJEqzbSjHkXKbkxWfLYM5+nMOU/WQew+VuXHrAK+8cvEG4nLH3fGSAkdnW3zkGwcRAq7Z2IeUp9S1xYzD3skGn390HPdauSipfd9knY9+8zBHZ1tkHYucY9EKE5JEEXaasa4tieLEWCB0b6zi1K8S07hVmLrW7mRNQNosvJgMtKq85wsf4rVPfh2AR8Z28huveTtPjWxb24WlLEv3NJIYZW6kgOTUedOtWZNOTWuJxU1hpSFOwLIANLPNkCNlE7L9Ky/ZuUhwdWC6wUe+cWjeJlPYeRbePJhn00CWwzPNjn2YuZbYlqDcCqn7MTlvdZv8Z3sNXm5ybKoecKLSIusaz+7BnDvvOFprTlb9FW1+pSxP+jSQkvIM40IoE1Z6rIJnc7LiM141N5zrNvfP202r+9EZGw1z33PLYI7RYoYnTlbZO9kgVpqxkseG/hw7R/IM5FxOVn1mmwFBFDNdD4iV7jReHCxpkjpPVNqMFT0mar5Jsux0jxIFji2xpCDjWESJQiCwgfX9WUDjxwkFiTFUdywKjqQcK7KuRRIkq96hTTk73DjiD//tLn7k8a8AsGdkG3e89g72jKYp7udKt/gUp1HDntP7S4ElIFhgBC3pjKUJENKcl1FnI0R1ghfytiTvWgSxIuj4cx2ZafHJe4/yMzdvWZHvX9a1cCzBQM7FkubXbkhCzY8oZWz6Mg6OLYkSzWTd59hsi1deMXZWBeRKvBJPd/2bO5KmlOa7B2eXHeFNC9yUlJRnM+cjAOx0LLdRNpB3EUJQa0fMtkIKGQfPlmzo10w3QuqB2Ty0LYEjBa++ev1Z1dRSCqQQ1NoxO0cK8xq23b9b1pHc/cg4Dx+vmvq0I7x4xRVj3PO4aaLsGjFrn2wE+FFCd1gsSpKelyV0NmShtzGqtUYgUAoiTDp9KWNjS9mrBVIuPLceepC7/uWPGWlViKTFB275Kf78lp8kttLWy9OdrojHwqjYk2X2OpSeL4QwSnejxFUAiSbETHMGseLfHhlnfV9m0fVkqU0mP0ooZR1u2DqIZ0sm6wGxMu+vtUYpTSuMKbdCwFnVJv9qrsFLTY55tsWtu0bYva7IfzwxuWjC4WTVZzDvrnjzK2VpVn3l+NSnPsXf/d3fceTIEcIwnPe5+++//5wXlpKSspgLqUxYybG01owWM0zWfVpBzP6pJgM5t/e5MzUalnrPYgZcq59KO6IdKYbyHjdsGUBKU1SP13yUhkoroug5DOQdolhxouIjBAzkHWwLvvzUFNONEIEmUWY8xZEaS5gxkkgZ03gw42M7h7N4ro1/uIItoZRxGOvPcGLWp9yOCJOllYMpF4bQssnEIbGQfPAFP8Gf3fp6ImtxYyxlZXTPfqszM6k6T2oX4kc6VsaCpFusgnmQdCwoZBxjd9KxKVGdk2qk4NCOTBd5tmXU8RLoz9pYUvD1vdP4keLN33dqE6pr0xIrzQ9ctY6HjlY4MNXsJeDesnOI264YZbIWcM9jEzwGxImi2o7QGoYKHrZ1alPnbL8XZ/JKPJtG6+lGeNMC9+KT1rQpKReXc5lcWClzN8qeGK8RJIpcp7ERJoo4UTidsBwTYqm5cmMfBc/GEoLpRsBwcbF9zZk43d+t3Ax5csLkN+weK7BxINcTXjw1UacZxmwZzBHGCa0wodGZHuveIxWnGkVSGD/4qNNUUtrcg0Xn7iaFYDDrks/Y2Jag0opohklq/XURqGYKDLRrPDm8hTtec0cvTDfl6Y0U5r9ozu5G97enEz1IAVIbsRCY2ltpc266liDrSo6WW/Oe0RfaHP7y9+/gZOead2oaS1LwbAZyLlN1n8SxmG2FNMOEKFE8cqyCbUletGv4rDf5V3sNXi7MTErB1qHcvOazZ1tcvbGPV155/ieBn2usqmn7/ve/n9/+7d/m53/+5/mnf/on3vSmN7F//36++93v8ta3vvV8rzElJaXDhVYmnOlY3THkRmBuKBM1n0orwrbEihoNy60/UhopBCNFl2aY0AhiSh0Pymor6oUnJFrjRwmzzYhWGBPEiqm6z3DeJdECIUxjB8yNNVQQtWMEpoEkpcCSZkzsqYkG/TnXJMpnXcZKHnsnGlRaIUliGrypIuHCUghaSK2oZQogBL/9yrfyoZtfx0Mbdq/10p7R9Ma7hPmZ11rjSJOC24rO/XFtYeEqME3LRJmHym5IgxYCW0rcrMSJlFEO5GxytmS2HWEJ0TlfdWfNgliZQlFrOF45VeAemG4ssoTZMZznx67fyHDRmzdm9m+PjnNstoVjSbYO5am1I+pBTDNMGCq4rO/Lsq7kUWlFZ3WtPN+N1qXUFWmBe/FJa9qUlIvPuU4unOvxXMuEjkWJxrMFUaKwLclgzqWUdaj7ERlndcfPOhZJYnxpB3IuxYzdEzfsm2zQ8GP6MjaWJZhthbiWZNdInvuPVjhRaTOUc3hyvE4cxSRKk+hOSG6H7m+VhjAGu+tR1Hlt7/NKMVEPyPgxu8fyFDI2QWJsyFLOPxtqk5wojQLw6Lpd/PxP/B73br6K0E4FCOebbpW12p/kpRqwAnNOSSGwhJ43aSk5VVMrpeeFWAsg50iCSPXEPnrOxsr6/gxoaEUxeyfqHK+0CeJkWZvDy9aV5k1jXTJaYNdogelGwOFyCwQIrSl4FonSREnCRD3gwHTjrOrGc7kGLwwz63K6hm7KubGqO+EHP/hBPvzhD/PTP/3T/PVf/zXveMc72LFjB7/7u79LuVw+32tMSUnpcDGUCWc61mDe5brN/Tw1UePYbJuDM01GCt6KGg3LvWe3eBYIEpUQdrYqp+sBjSAGrdHCNIaTjqov61oIIaj7EbUgYUNfBs+WjNd8sq4mSrRJ1e0cI0jAkZqiZ7O+L8NMI6AexNhScGi6yd6JBlGikYBtC9Lby4XllsMP8UefvYvvbbyCX/uh3wJgJt/PTL5/bRf2LKDrI2t1PKgkgmLGphUmnIunXXf0SwijJuiqfWJtigkBuLbo+VLbQphzOTFNXUsKrlpfwrUl3zlYRnYsFYQQ2J0NlTBRzLYiCq7xnt032eCb+6f5t0fHF1nCPHayxsmaz5tu3cbmwdw8Jf/G/izHZ9uUsg59WYe6HzPbCsm7Ntdv7kcLODR9Zv/thZzvRmta4K49aU2bknLxOZ+TC6s5XjFzSr3m5BwafsxoKUMxY5/T8fdN1vnco+McnW0x0wwZyDoM5D12jRawpWC2GRAlCo3g0WNVYmXqVFsK/Chhqh7w+ccmqPnxPOHAcs0phVHX6gWjYZLu+LZGRzH7phrU/dTy60KQC9v831/6P/zEI/fwQ2+8iyc7nrVf3/68tV3YsxTPEp06VM9TxJ4NC0+Dno0Yps5Fm0DqrmetZQkcS2ABoQCV6F4ArxTg2hbNOUEouvM+ni0pZhwsKaj7MZV2yJ6TNb7y1NQZbQ7nigTWlTyyroUUgihJsKQk69qs78uyYzjHTDM660nbC3UNXq6hm3JurKppe+TIEV74whcCkM1mqdfrAPzcz/0cL3jBC/jABz5w/laYkpLS42IqE053rMG8yxXrS/RnXX765i3sHCmsqNGw3HsWMzb9WYej5RaOJQiihOl6mycnaihlboyebXxno07omVYmiEFrKGVsan5EO0xQSmFJiWeZG9vcG3OkYLoRMdWIOg2mEEdK2qHqFccJkMRpVXuhyIY+d37lr/n5+/8VACWeoL9do5ItrfHKnvnM9auzhCkWbSkoZhxsS1D1z34zZ6EaQSnTtNXajHzZUhArhWNZvc0W1xIoDYWMTd61eiqhfMbmx67fxOceG2es6KERTNR8lNIEsUJp876tIKGVsTsqooR/f3xyRZYwc5X8WtNJ2I1p+DHtKCFWiiPlFt88MMO2odyqr5Xnu9GaFrhrS1rTpqRcfC62RcxSx9s2nKPcDDgy26Y/67BtKEcjiFd1fKU039g/zSfuPUIziNk9VuSpyTpNP+ZEpU3Nj1hX8phqhCRKUco6IKAdJdT9iFZo6lkhoB2p0yoIJXQCd01DSQizedrFEuZjSoMtjDCikWY0XBBuOvoof3z3+9hSnQDg+w4+0Gvappx/XEvg2mZiMogVntaEsTpjcPTcpmxXHWt1DKGFhu4QmiUABFJoXEuQcSR1P0EpjZaCbMYmB/ixoh0mxInxrG1HCXJO2G83aNePFXU/ppSxafgxfqT47qHyimrauSKBh49VmGkEDOUd+rI5+vOusQzMOZSyDq5tnfWkbWrT9cxiVZ2ddevWUS6X2bp1K1u2bOHb3/421157LQcPHly005eSknL+uJjKhDMda7wWcO3mfr7/kpFz3tWbbUW0o4TZVogUgm8fmKHejvA7jRwF+LFGozuBR4pYRZ2iVDNVN8qFSHWLWaMm7N6Y52ST9QphTSexHoVrCYK0mr3gXH9sD3/y2feyffYkAB+77lX8t5e+mdYyyvGUs6NbkHZHvDxbUso6VNsxYZwgz3KQrOtLGyRgdxQDUhgVup8okkTj2RLPdhjr85iqBzT8mASjPGiHCSNFj21DeRpBzE3bhti9rsgXHp9gpJjhSLlFmCgSpXtqBU2n0A1jHj5WYeNAlkRpNg5kF12D6n6MZ0seOlrh6GyLdpT0lPxSQMaRHJhqIgV4joVrmxTuyZrPVD3gFXOCyBZ6i52pCZs2Wp89pDVtSsracLEtYhaH6CRsHswxGis8W1JtR/iROuvj75us87lHxvnso+OUmwF9WYco1uwcKTBVD5iomgDfhh8Bipxnk3UsJmqBsT9QGLsgVqYcVNBrEGmMl+Zcco5FEJsmbQK0VitHTFkWLwr4ra/+DW++75+RaI6VRvitV/8639p67Vov7VlNd4IyUQlBrMg4FhsGskzVg3lK14V0P9M9VRzLiBDknMatLSHjWNhSkvcs/FgRxKYZm2iT36A7IbpSCEYKLjONEC2M+KjSPOWHL+dMpB2fbTNtG/sv1xZUWiGXry+dsabdOpTviQS+uneKv/r6AQZyHlN1nxMVnzhpdYK5XbYN5wjiZM2nx1IuHKtq2r7sZS/jn//5n3ne857Hm970Jt7+9rfzqU99ivvuu48f+7EfO99rTElJ6XCuu2Jn05i4EDtwS71nO0q4//AslXbEaCmDawtOVHxaYdLzx5Sd7qvWp4rUJNHY0txwo2ROKj2dNN3O6zUsmoOZqx7UkDZsLzBeHPL2r/8tv3jvZ7C04mRhiDtf9at8dccNa720Zy1JJ7yv2o7QCDxbIIRAoFfcupXS+FhbQuNaki2DeWxLEMWKyZpPOY6pBwkZW1Nuhp3NFGN3kHEkICg3QmrtmGs393P7VWN4tkXWscj0ZTgw1SDpFMFW56RV2mzMOJZgsh6QcSysovEI7FJuhuybbDDbCokS45X7kW8c5LbLx3pK/oJn9zZquoWxaQ6LTiDaKU+0fZP1Zb3F0oL12U9a06akrB0X2yJmqeOtL2V6AUBne/x9k3U+8o1DHJttEcQJY6UMQsBk3WeqEZB3LYQUeLakFSVECYRxTLkRGB94ae5NthSgTxd1NJ+uoKGrup1LPZxjg5SWt+eda088yXvvfh87y8cA+OQ1r+T3X/afaHjpRu7FQGmNJSVWJ7Oh7icMF1zaZX9FWSQSs0niupK+rEPRsxmv+QghGC15CKDhxwSRCQPsPiJGiabaDvFsC9uS+LFiqOARKUUYKYSQxgu3sxEjOqdzpDQ60gwWPHYM53nsZJ0nJ+rkPZvBvLdsTfuzL9jKrtEiUgp2jhTwbIsnx2skSlPIODgZmyjRTNV9ys2AzYO5p8X0WMqFYVVN2w9/+MOoThT1W9/6VoaGhvjmN7/JD/3QD/HLv/zL53WBKSkp81ntrthqGhMXYgeu+56fe2Sch49VeHKyTjtU7BzOsXO0wP6pJg0/Aa2JOh1aKQVxcsrgvUt3VDtZ8AnN0qENy/055cLixSE//NiXsbTiU1e9nHe//BdN+FjKeaebMg2dccnOrkUc6p4KF1Z2DhQ8mzDR9HuSKFE0gpi8ZzHTDIk6Xl9aGy++cjNEKaNAEEArTBDCeOlmXYuxoseOYfNvvnOkwL2HZnA79g1xN9RBa4Qw7xErUCphvNqm2gppBglXbCgBggePVmiHMYWMg2ubv9HB6SaffeQk/TmHk1WfsaJHECs29Gdp+EbJ70eKrGuxoT/L+r4Ms62Ib+yf5nNL+OUu9BZLefaS1rQpKWvLaiYXznY64kzHO93xlzvWUj7qbmciJXE1h2da2FKwZTBHf87h6EyLVhj3xASObcZLko4nrRTLr3+5dq6CNDT3InPT0cfYWT7GRGGQd/7A2/jSzuev9ZKecSy0Eljpc5kt6eSfmHoxTBRRO6IZmA0QW0CUnP79pBRsHMhy2fo+RgoeWmu+uneK2VZEvR0TJQlBrIk7GSp6TjB1lIDSCf225IoNfbz88lHufvgk+6caeEiCyNS+UmqUOjUBJ6Xg6o0ltgzmOVHxafox+6eaaA0PHasuWdN+5BuHejXo+lKGIFJU2hFbBrLIjkTYswVOzuHIbJuxWLG+lAHS6bFnI6tq2kopez8sAK9//et5/etff94WlZKScnrOdlesqwRYTWNiNTtwK7pZCOP10woTco4FQtAMEmZbEf1Zh3aYECWmuHVtc72J1OLSVC1xZ57buEpZG+wkJpbm37WWKfAbr3k7+cjnnktesNZLe05jSVPsnun8yNgm7E8FiVESJMr4Ts+2O97ZgrywiJSmL2sT1BNERyFb9GyaYYIUgpxrc+X6IrOtqOe1dftVYzx2sko7SsjaFpaEdqyIE90JwNZYQuI4pulb9ByOVdr4UYxjW7TDmMG8C0C5aZRN12zsY99Ukw19lgkwm2rgRwlDBRfPFlRaEYN5yVUb+tg8mCPRmoNTDb64Z2JF3mKp4uDZS1rTpqQ8s7iY0xGnO5ZnW4t81KPY+NFO1gK0NpkM7Sgh8hVVPyLpdHG0Bh3p3ni21rDQyXauT31a064tdhITW6Zt8lfP/2GyccBHr38t1Wy6qbsaHHnKls6WnTC9M3yN6Pw/Ut2GKp3G6KlpyzM1gF3L1KUjhQw7hvMIYdS6I0WPciuiGcSEiTJN5Y4CXgOeFBQ9i76sS86VXLmxn//ntVcy0Qj46lNTFD2bUtbteVT7UUIjSHCERkrJYM5mXV/WBCDmPU5U2kzXffwoOW1N261BT9Z8PEcykHNMWG/GxrE6ggo/pj/r4NqSkzWfIE7S6bFnIatOK/rYxz5GkiS84Q1v4Ktf/Sqf/vSnuf766/n5n//587i8lJSU5VjprthcJcBqGxNnswPXLXD3TdaZbYdYQrJzpMDrbtzIpWOleQ3kwbxLKeOQ9yym6gFT9YBWmNCfd9Bowk4KfTtMSLReHIp0mnWczc5tyvnl0qlDvPfu9/HXN/wgn7r6NgC+ue26tV3Us4zV/Hx3VaxJxzdE0PWqFVhC0oqNv6wjBbYl6c+6eHZMECk2DmS5ZDRPrR0zlHfJOhbtKGGi5lPzYxNMZhvVbDtWZF2bsaJHO0o4UTXhCV2vrV2jRX76pi3sHa9zsuYDgpxrmQK0Y7HgOZIw1riWxe51JfZO1hnvhJZtGsgRdgrVrGuxcySPlLKnnv3R6zfytaemOTzTYqYRkHFsNg7k2DmSZzDvAdAOYhIFJ6s+G/vn++Wa75VgfV/mrIMdUp6ZpDVtSsozg3MRIZzvY71498giH/Uj5RZJommGcc+iqxU05gcldW7eCtN06n1o4TTZeflbpJwLlkp4y7c/xWuf+Bo/8nN/QuB4KGnx/lt/eq2X9oxmri3d6Rq2C6fDko5SR3PKB3ouS72PUboaRTvAQN5hth2ZgLCsw2wrpO6b7IcoVghhGsqy0xR2LcnG/iw5z8KPFJeuK+FHiolG0Mtq2XOyjhCawbzLQM6h2o5IKm2EMPVtKevgWhIhBLtGC9T8iImaTztSDObdZWvabg3aDGNcW3L9lgEOTbcot0KaQYwlJaOlDNuGclTbEXvGa3zlyal0euxZyKqatu985zv58Ic/TC6X4/777+eTn/wkt956Kx/5yEeYmJjgzjvvPN/rTElJWSVzE9XPR2PidCrafZN1/s/XD/LURJ1mkBDEJnFzz8ka3zk4w39+2S6eGm/0Gsh1P8axJLLjM3lito2fJFRaxh/z1HS3Ri+4qS/XtOp+vGscn3LxkCrhF7/7Ge742sfwkphf/cYn+McrXtJTJ6ScP1bzo600hLGaVwRLIZBC0l9wUY0QpY0tgmsJrthQ5JFjNdpRwo7hPLECIaGUdYznVxCzvj/LeNVHSNUriDOOxVjReGBLCSerbSwBtXaEUhopBbfuHOaHrt3APzx4nCjRDOUcAI5VfDKOIIgVUhp/sc2DWQoZm+8dKnNwpkm1HZJ1bUZLmXmN2KxrMVHzGSl6vOP23YDm8ZM1do0UzJo7179uaOOG/izjNZ/cMh5g3fc722CHlGcWaU2bkvL04ExTWudDhHA2aznTse47VMazJK0wJkoUzSAmTjRxcup+qIEwrUWfkeycOcqf3P1erju5F4Affvwr/N21r1zjVT27WBgWvRR6wa9gVOhxp4HrdJS6YPIRpICurbPsqto1eJ0AsTBKsIQgTBTlZsADR2aZaYasL2Uot0IjHuiIhvqyDutLGXKejdKaRhBjSUEziNg/1aAZxtx6yTDfOVhmvOazriRwbYnVuV44lsC1JEMFj2LG1JqDeZfdYwXaYUIziGkGEbZlLVvTdq+HGdsi41jcuG3ANJkThWtJihmbRhDTDhPuO1hOp8eepazqSfpjH/sYn/jEJ7j00kvZtWsXH/vYx/jpn/5pPvaxj/Ge97wnLXBTUp5GNMO4pwRYirNpTJxuTGzHcIGPf/sI39g/Q7VlPC8dS5J3LUpZY/L+3i88xcb+LFuGcj2/y4wjOTjdpB3GxIlRFkTJfH3BUoGgnWyyJf1q5ybRL2WfkHL+2VY+zp/c/T5uOPEEAP++8/m86wfeljZszxMCsDobEefyI73wfIgTTawTdDPCsyWxgko7wrMEe07WuWpjH2GimGmGFDwbKYQZH+uoabcO5QijBNcy9ia2JVhX9Mh0QgunGwHlZoQA/uprB/nsIyd5+eVj3LpzmFdds54nJxs8dLRCpR3jOZIoSYhjgRYwkvfYNVpECMFg3uX6rf3MNAN2jhbYOpinmLHnbUQZ2waLvGtj25KffP5mPvKNQ0zUA6QUi4IUX375KP9w/3FaYUwx4yz6Xs19v5RnL2lNm5Ky9qzE8uB8ixBg+Ubx8UqbfZN1ip7NTDPsNUeEEL1jTdUChgseT47Xmaz71P2YobzDeC1IJ72ewQitePN9/8xvffVvyMQhNS/P777iLfzjFS9Z66U969CAa0GcLC+06Ybtzf20bZ0SHiilkeLUc1+iOoHUGJVsd1JzfX+Guh8z247QAqrtkEPTLabqISNFj0tGCzx+0gR9TdaNvYlnSbKuhe40bONEc2K2zVQj5BP3HsGSwlyrhvP4UcJUPcC1Ja5trhdBrChlHXaOFOYJB9qR4rbLRzk80yLnWvTn3NPWtF1F76MnqlwyaoQIve9hR4iwZTDHZM0/r9fHlKcPq3oSmZqa4sorr2TTpk1kMhluuMEkgL/oRS/i6NGj53WBKSkp50Z3d+5cGxNnGhO7elMf//HEBJVWDEJQylgoMGm5SjOQs5luBPhxwu51pgCfbYU0fLM7mKxiDmy5olhDz98o5cIitOIN99/NO7/812TjgLqb5d0v/yX+/urbOtGpKecDjSlo865FK0pWvRkxr+gV4NgSVwoipfC18ewaKWbZMpgjVpqMI3n11eu492CZfVONjn+WYutQjl2jRWwp2Gdb9FuCdqRMkIqAZhBzsurTCmMsCTnXYt9Ug4eOVfjKk1O8dPcoP/OCLfz6bZfw8W8f4dsHy8w0AqJEk3UkW4ZyXL2xv+fzBeBYkqFChkTpRcVtt2i9emMfG/vNBtWZghR3DBd46GiVR09UKXhnfr+UZydpTZuSsras1PLgfIoQusddrlF8YKrJYydqCDR+opHAQN7livUlhgoeWddi32RIzY948FiFum+OKTh/tgYSY2kUJWf2oU85P2yujPPHn72Lm48+CsBXtl/PnT/wq4yXhtd4Zc9egviUDcJCupOTgvlN3Vh11bm6p66VQmBZgihWxnpEg9KKjG0hhaCYcbCkoF6OqbVjvr1/xljy5Rx2jOTZPJhjohYwUfcpZWzqvrne1NoRjSCm0o6whGC82sazLTYPZNk+UqAdGTHArtECOdei3IxQWiFFlulGSCnj4FiCWKl5woHX3biJex6b5NET1TPWtFIKbr9qjBPVNnsnzcbVQiHCDVsH+McHj6fTY89SVtW0HR4eZnp6mk2bNvE7v/M7jI6OAtBsNikWU5+MlJSLxUoCv+buzq22MbHUmJjWGq1hIOdwbLbFYyeqVNqR2TWVxrfHFmB1vC8bgVHjtcOEyZpP3rO592CZiVpA0knmPJuitKs6TFlbrpg8yH/99w8j0Xx967W849W/xonS6Fov61lHVzXQ6M58sdjr62wQgG1L8q5NrLWJxBXg2BY3bx9kqJBBa80DRyp89FuHGco5KK0ZK2aotiMsKXAsQcaxyLsWxyohYyWPvGvTjozCNogVjhQ4tkWYmEbrQM5hqh7y9f3T+HHCm79vO7/z2is4OtviwFSDux8eZ7Luc+2mvnnhUFprxmsBt+wYpB2pZYvWV145Nu8aeKYgxTMVwQvfbznOJck8ZW1Ja9qUlLXjbCwPzpcIAU7fKN4zXqMVxsy2Q4SGRJvgo6lGwPHZNjfvGMSWgv1TDbQyAZ+wMguvs/reYPa+04yGi8dvf+mvuPnoozTcLL//0l/gk9fengoQLgKn+/lW2njSSn1Kcav0qalKMEIdS2iKGQdlK2IFfpTg2RLPligNrSA2k1cCtg3lEAhOVNtIAQemmvRlHXaO5qkHEWGcAJqGH+NHxp9PWiZUMFEaIRT3Hpploh5wzaZ+LhktsHeywbahPG972QZaUdKph2PueWxySeHArtEiUogV16BnEiJ4tsXnHh1Pp8eepazqX+0Nb3gDlUoFgHe96129j3/lK1/h+c9//nlZWEpKyulZaXruSnbnztSYWDiSVm4G7J9sUm6FxMr4eE3XQzSmkRtIcG37/2Pvz+Msu8p6f/y91h7OXHNVV4/pKQMh8zwgEhKJiiMggvfrRQYVjVwhXkGv9+KVC/r7igoik3yvil5FRa6KEqYQIZCBkIQQMnbS3em5ax7OtMe11u+Ptc/pqu7quZKuJPv9egW6T52za3V1772f/azP8/lQ8iWeY8dE2rGit+CQKs1dO6YwBqaa0aKRl8ObUJ2WTUe10PUmYumG7cKv5zw3PLZqCx+99vVMVAf4u0t/BCPk8T+Uc9IYoFKQNCN7NniyoyI4/medBQ99ytjzSkr72oVre9gzG+BJ6/WljcFzHABm20l35HN06yAb+is0woRUaxphyp6ZNgVX0l/xSbWhVnAZ7S0y20qoB9ZuQWkouJLBit99GO+veISxYv9cwFcfG+ftP1jlrMEKZw1apcNf3b2L7ZOtJa9Vb7x6A8BRi9alAhaOFaR4vCL4RAIbnssk85zlJ69pc3LOHCdjebAcIgQ4thChr+Ty7Z0zKK0RxtCKNZWCQ8F1UFrTjFLu3T6VhRIpyp6DFNZL05MSgyFK7Tj2cqhuozSvaJ9Lfu/GX8LRmt+78RfZ1zd6ppfzosOTtlZNM4uDoidZ01tiXX+R7++vM5+Jgzq1rzIgFvxaCoMUkihV9Jc9XCmYD1N8RzLdigkThRDw1HgTpQ3aWN/ZRmjtvV5+9jCbhip8b09CnGpiBYlWuFLgZE+pPaVs8yjR7J1po7Th0g39rO4tsmOyhRCC80Z7un+mrcO1o27qn2wNeiwhgtZmWa6POSsTYcyxrJ9zlpN6vU5vby/z8/P09PQc/wM5OUfhSIWASztOu42NpdIhFzYWotTutm0dqR5xU1hKMfbURIOP3PE0m4eqzAcx39s7RztK8T3HjkHPBST6UGPIlXZMxXUE1YKLIwVz7YSiZ4tcbQypMigDBQeiTDzoZOn2nSL1ZNUFne+bj5I9e4w0pvmfX/tzPviDb+KZgbVnejkvKnwpiLNKVXLsh8FOqeZkPXQhBAXHfr7iOSAEqdJcs3mQZ6Zb9JetDcFsO+bqTYMMVnzu3zXLRD0kVoqBSoEo1aRa28RebbhkQx9vvn4TtaLHtrE6f33PbnZNt2jHKa0opVrykGADGBYUkNoYZtsxF6zpRQrBu37onEVN1RO5Vi23svVUj3cq1+Kc0yevp84s+c8/Z79sCwEAAQAASURBVLl4cqzerS+dJa65qdbsmmrxjhvP5rzRniOuuYdv7J3INXfvTJsP3f4UfWWPWtFjphWzfaLJbDsmzLzYU2Oo+dm9UpssXAji1FAPE1wp8LIbbDtT35E1gJQ2pOb0pmFyngOM4TWP/QcvmXiGD7zybWd6NTnYc8Z3BIk2FBzBSE+RC9b1snOixdMTttF6tBBqCXiuQEqJK+3zpyuhkE2EPT7WQCmDFAIh7MZKktksSGFtvK7aPMBEPaYdpTSjhCjV9JY8ZtsJqdL2uEUPIWwQWqo01YLL2v4yl6zvZfd0u3utOhmWq6ZdjutjPjn23HKi9VSuj87JeZ5xqum5xxsThqMrxi5e30vRtQ3aHRMt5tsJ2himWzGNMO2qXqWwu5+ptrudxhhapAhsgygAiq7DcNVnz0wbrQwLpr0BcIRACrv7ebKjZnZnNi+PnxWM4SeeuJP33f5J+sImfWGTn3vj75/pVb2oiBfIao+n3unYh2idpde6kqLn4CiNI60KwZGCkidxpeyG/7lS4jvShjW0YxCGepCgtaFa9Ch7DkLAXDvhgV2z/OQla/Fdyde3TVIpuFy7eZCJRshDe+ZohTZRO0o1tYLLQMX6ACZK40ob0jDTio/w1zqRa9Wx1LOnwqkc77lMMs/Jycl5IXKylgfLMR2x0Bt3pmWFCEGcUi16OBKmWoYk0QRSMFwtkChNkGhiY+tam06v0cZkClzTVf5FC8bAltsuIWf5GG7O8vtf+Sg/tP0+AL629Wru23DhGV5VjuHQORQrQ5QqnjrYYKIRAVBwBbE6dL45AFkImRB0N1JqBYeekocUMFDx2TcbgLFWJgJrBSYARxvCVGOAdqR4dH+domubvj0lr9u0bUYpxkgSrVFa4zp2EycxUPQcZloxk43olO0HlqumPd3rYz45tnI55abt5z73OT772c+yZ88e4jhe9LXvfve7p72wnJycpTmd9Nxj3RSO5e+1f65NX9ljx2STA/MBrdiqZaNUd8etF46rwKHR7STWeBKklBQcyWhvAaXtzVVmjaWFo9tK627zdyGG46sLO+/LWV4G2vO8/ysf40efugeA749u5b0/9PYzvKqco1F0BWlW9BY8QdF1cRxBmCiCLMTMEeC7DntnA4qupBEkIASreorUijYtO1WayXpErA1uqmk3IqSwjd7essd0M+Zrj48zVCsw04o5Z1WV2XbCVDPGcQROZpUbK00rVsQqZFWtQJAoRnqKuFIctcBd7qbss8GzkWSec2bIa9qcnDPDqVgeHGtjL0013907y3QrZrDic8naPsab0aL3dRrFrShh+0STIE4ZyCx8gjglSu0kWBArxuoh/SWPnqJDO1a0Y92dFBMY2nF6QjZFy2WXkHP6/OiTd/H+r36cgaBOLF0+9AP/iQfWnX+ml5VzGMrAeD1m3ktwpW2SGgO+I9HGZJOdAikgSu0UWH/Zp6foUi15vPOms1nTW2LnVIuPf3074/UQpQ2uIxdMowkKriTOGrez2TPwqt4Sq2oFnhhrZBs0NsirEWrCRFOR0l4DhP18O1OzXrdl6IzbD5yI8GEpTjQQMufMcEpN24985CP8zu/8Dr/wC7/A5z//ed785jezY8cO7r//fm655ZblXmNOTs4Cljs9F05MMbam16HkOUw1I6S0o9ZK2/ESKa06tqOadaW9sXbUstrAqloBKewNdrIRorIdT32YlUGsFjdeHZEZz5sTK3bzpu3ycvNT9/CBr3yMofY8iXT4yHVv4BPX/Aypkw9qrEQk1iLEGIMrrXK27EtasSJK9aHwBmnVt1PNCEdKwkThOZLRngLKGOJUMdOKCVJNwREIIXCwmzTzQUIzUlQLDjsnWxysh90idftEkzBRrO0tMlaPSNKEJNVUfYcotT62a/pKbB6qMFaPntf+Ws/GtTjnuSevaXNyzhynmruw1MbeHU+M8+m7d7FrutWdHim4Dmv6ivRX/K5q7Mbzhxmq+nx3zywzLZvuHqea2XbMRD3qbnpqA3GqmW7HWciuXYMjwcEq/gzHVtF6AhxHYAzdmtl3bWEb5H61zyl9QZ3/9dVP8ONPfguAx0Y28xuvfhdPjmw6wyvLORoGa0kSYX1lpRSUPbuBAoY41TjZpaGn5DFQ8WmECfUopR6kXLO5QjtR+K7EdSRxmuId9vjiCHs+FlyJ70kuWtfHpqEKAGP1iANzAZ2oQc+ROFLQjlMMUC24JErTjhWD1cIJh9c+25ys8CGfHFv5nNJT98c//nE+9alP8cY3vpFPf/rTvPvd72bz5s28973vZWZmZrnXmJOTs4DlTM8Fe6F+YPcM390zw2ClcMTXO4qx2XbCFRsHuGfHNMLYcARtOuELoqs66BSwHbFEZzR7qOIRJJqpZoLOdkqDRB3x/RaWsBK6qkCktT/ovp69xwGOPErOcnDztnv483+1FghPDp3Fb/zYrTy2assZXlXOsbAFrkZIQaXgYrCqAs8xFFxDEQgSDQiiRJGkCiEE6wcqXL2xn/kwZddUC08KCp6kERoQwgaQLXpAVWhtaIQxQerQX/apBwmzrYhq0aPgSlb3CoSAmVZMM1L4jkRIwdq+ItOt+IRCEFcyy30tzjkz5DVtTs6ZZTksD+54Ypw/+NKTNMKEwYr1aD8wFzDbjplrR1x2Vj++I/nGtgm+8thBKgWXfbMBM60YzwmRQtCKFJpDYgFtsgmwrLmaYCi4dg675rvMhynJUTw2O6QGfGFHvrWxuQ0AKo+UeW4xhr//+//GSyZ3kQrJx659PR+97mdJnCPv3Tkri45PdNl38JyOKtZ0bUlsEKBgoOxT8BwMMNmIeHD3LDe9ZBUV36Xiu1QLDmGSkqTW3kBkyt1UaRxHUC26uFJQ8pxu03LrSJVGmNCIEoJY4TmC/rLHbJCgtb1WzAcpW0eq3HLDluetEjWfHFv5nNKTxJ49e7juuusAKJVKNBoNAH7+53+ea665ho9+9KPLt8KcnBchxzIBX670XDjkXfPdPbM8dqBOb8lj32yBLSMVBhY0cDuKsdV9RUZ7ioRJSslzGW+EJCobFcsUBCK7eWpj1+NIgdL2hjBSK9GKEkqeQ2jMESNlC9UKDuC51pOzkw7a8ekUC96fj5o9e9yx9Sq+t/ps7jnrYj58/X8idvPidqXTsRlxJYzUfMJUU/JdZttJt6ELsLq3SE/RI0oVzTChHafccN4qNg1XaMUp9SBh10ybqWZMEOvuSSegq9YNU8X2iRZF32GyEQKCepCwps+Oi5V8lzW9JQquQ7XgEiWKdqJINVy+/sQfxjustHCE5bwW55w58po2J+fMc6IjvUvdB7Q2fPruXTTChA39JYQQHJgLEUJQK7jMByn3PTPDUKXAfJiQasNZg2XOG61xz45pe4/D1pOdQFsDaLW4SDVAmBo70VISeI4NKTtW+9UAQWJvmq7DEVkOFV8QJgZ1nB7uUrZhOSeBEHz4+p/jN7/5N9z6Y7fy/dXnnOkV5RyHzjNhqgwis0AYqfq0Ek3U1N1QMikFfWWfnpJns1SilNV9RSbqIfvnAtb2lbhwbS87p1o0I0WUWos/ow99o5Ln0FtyWd1boh6mjBqDEIKBis+lG/px9gmeHm90z/WzR2r0lTxambXK6y5bjzY25PBM16anQj45tvI5pabt6OgoMzMznHXWWWzYsIFvf/vbXHzxxTzzzDOYfOcwJ+e0OJ4J+KmOki31ff7q7l1MNyMKrqDoSpQ2jNcDGlHCJev7uo3bILYqOWMMA1Wfibom1YaS56CNIc0UBJ3vqLUh1VmxGlufoFasCGKrYmgnijA5st268OqhAHXYvcFZ8LXD359z+lSiNm954PN88prXkTgeqePyuv/0wdwK4XmGgSwdF8q+yzkjVYI4JYgVgbHpun0lj4LnUPIdakWrOPr6tgledvb5SCl4/OA87VhRzLy6jLEHltJanBhjiJVBCk3VcZFCIjCEiWLfbMD6AatWaEYp6wfKXL6hn7F6yEwr4pd/cDNXnDVwUkXtSgxHWK5rcc6ZJa9pc3Kee462CXcsFdfR7gNr+orsmm4xWPGRUhJl/u1SQDvRCAFxommECSprAO2bbmOUoVJwSZWmFSmbPp/5XS5Vo3bXDkw04iNsEWSm3OvQ+aXAbqYeHrwLdvTbd2U2AXOMn1d+KTpprt/1PQppzH9svQqAr5x7HXdsvSqvaZ8ndP7Jp5lwRxvNwXrEQLXAYMXj4HyEIwXVgsdwtUCsNM0wpeS7nLOqRj1IaMUpUgp++MJRnhxv0IhSdFbQOkJkmyWCkudy3upefvrStfzHkxOLajrPsUrTwapPT8EjSBSOhKLnsq6/hAH+9Xv7V0xteirkk2Mrn1P6yb/yla/k3/7t37j00kt585vfzLve9S4+97nP8cADD/Ca17xmudeYk/Oi4URNwE92lOzw4nh1T5GvPDrOnuk2qdbMtmLqYUqsNL1Fl0QZdky26C/bEbOnx5sg4GuPjXdHnQuOwJE2WkFKgeh40xqbptkJDvNda5oQp1nxnI2yHO5bCxxXaZDbIDx7XLPn+3zwi3/K+vlxPJXyJy//eYC8uH2ekirN2HzE+Wt6GKj4uFISq8T6gRU8fFceeq+2D64H5oLu6FMztE3e/or1+jPG4Lh2nAxj1UYCGKkVUAbqQUKYpChtaIQpOyebDFYL9JY8tgxXEAKaUcrlZw1w2fr+k1LMruRwhOUY6805s+Q1bU7Oc8uJbsJpbdg72+aZqRYH5wLu3TlNnGrW9JUW3QfufWaKIFaM9FihgTIGpXU2Pm26Ke/zYYq99dlx632ZCu/gfEi2L0mY6EVTXMfyq134eidobOGHOl8/Vu2aaEh0PjO2nJTikN/+xl/xnx+6jZlSDze/5WNMVvuBvKZ9LjnWuXOydI4Tppr9swECg5TQU3TpKbmEqcLVkpGeIluGq3iOIEp0t8m4daTGO286m898ew93Pj3JRD0kVgZXCkZ6irzinGHeePUGNg9V8V3JHU+Ms38uwBGCoudw0bo+XvXSVWweOjQNMNmI+NIjY8y2V15terLkk2Mrn1O6cn3qU59CZzeYW265hcHBQe655x5+4id+gl/+5V9e1gXm5LxYOFkT8BMdJVuqOB6q+jyyf56ZVozShmrRY02f5OB8yFyQ4ruSfbNt+koeY/Mh0+2Y9f0lBqoFrjjL48HdM8y1E8oFh3NHa8y0Yg7MBUSJxlngPStkFkZmOs1bqyroWBxIQbd5dCxVQ86zRzEJefedf8NbHvw3APb1jHDvWRef4VXlnA4SuwFSD1MwmlRrNIZ2rKgVXQbKXvf6YoyhGaYMVQs4ku7oU7XgUvLsrnvBc9Bak2gwHbmPsT5jniuZa8SUfYeq6xHEinakCBJNI0i5cG0vniN5eqLJQMXn3NEaf/7NnSesmH0+hCOcalJvzsogr2lzcp47TnQTbvtEg898ew/ffmaGuXZMI0wA2DxcZbhWxJGiex+475lpEqUJIkWtJG2wEIJYaRuSm41RCwNKg8FOhzUjRT1MQBy9wXT4a8d6nzkslUws+FrOc8MV+x7jj277MBvnDgLwhfN+gJZfPMOrevEhsFZ5AsNyPd4JoOQKlBGZv61AABev66VS8PAdSa1oW1tPTzSPaDJuHanx33/sfN4022b7RIMdEy2KvuScVTWu2DDArpkWn/jGDnZMNgmSFAyM9BS56fwRrtsy1K3p1g+U0dpwx+MTzLZXbm16MuSTYyufU2raSimR8pBK5w1veANveMMblm1ROTkvRk7EBPzp8QYP7J6hp+R1GwPHGyX7y7t2sX+uzUDZZ6hSwJGCxw7M8/iBOj0lr/v9Cq5kbV+J6WbETDuiGSa0wjTz+LKqhEQZhmsFrto0yPf3zbFzsslEPaLsOxQ9J/PLtMpaARgNcfZrNyuSXZGNumBv6DLzKcqL2ueeS/c/yR998UNsmdkPwGcuvpkP3PBWWoXcZP5MsRzKhE59bICH9s6zfbJFR8gTp5o41RQ8Q7JglGxtXxEQXVVCreixYbDMzskmqeqcn/bBV2deKL4rqQd21GxVrUAhuwY0I6tAcB3B9okmviO5cG0v547W+I8nJ05KMft8CUc42aTenJVDXtPm5Dx7LJz0KnsOX35k7LibcFrDR/7jaR7eO4cUgp6SRzu2kxzPTLVox4prNg8yUPERQvCS0Ro7J9tMNCMqBQfflRRcSTNM0MI2agUcas6aQyG2490aVhKlVnig9dHtCJZ6uXMsnb3Bye6PqdLL1rDKOTaFNOY3vvl/eNv9/4rEcKA2xLt/5Ne5a9OlZ3ppL0rsJoZZ1tyRwarPSE8RpayC/rzVPTy4e5bHDzS4fusg5YK15Do4HzJQ9rlwXS9PTTQWbaRLKUiU5omDDR7ZP087Trl7+zRfeXScyWaE0obVvUXWZPXpwfmQLz06xmhvcVF9+nypTU+GfHJsZXNKTdvvfe97XHLJJUe8Pjs7yzve8Q7+9m//9nTXlZPzouN4JuBhonjsYJ0/v3MnBU+ekELtM/ft4YFdMwhhbzCulAyUfSoFh1hponTx0FbJdyj7DlMtgcputqnSeNJh/1xAM1Jcsr4PMLSiBAxoY/AkRIlGA0p32juHilsD3cK1c9Hp3OOiVKPyovY55/UPf5U/+MpHcYxmrDrAb/3wf+EbW64408vKOUU6IYDGLH6ojJShV0iu3TrA05NN9sy0OVgPaSeKsu8y0lNk81CFqWbMhsESjShh70yb1T1FLl3fz0wrYrwedlX0rpQgBVGqSVKNNtBTcilkabsFz4EsjHDLcIUoNbzx6g1ctr6fP//mzpNWzObhCDnPNnlNm5Pz7HD4pJfShr0zAeeNVo8pTphtRTw13sB3JNWCQztRGOwESJAoplsR2ycaXLlxACEElaLHWYMlJhoRe2YDBis+taLLdDM6VHs6AtNJm+94s2eTX6nSeK5ECAdH2s3NND3+9qnABpcVfQelNKm2NXGt6FJwJdOthFxn++xTikM+/ze3cs70HgA+e+FN/K8bf5FGoXKGV/bipjtVyWK7ESmOb4e3FJ0sFePBbDumUnC5cmM/T441OTAf4kpBwXVY3VsEA//y3SN9ZgE+/LWneWqsgeqENRh4YNcsniu56byRrqfrserTF2ptmk+OrVxOqWl7ww038IUvfIHrr7+++9rnP/953v72t3PhhRcu2+Jycl5MVHyXgiOZqId4ruyOeQghmGlFPLh7lmaYMlDxWNWztEJtof/XY/vn+cqjB3GloL9awHMkidJMNEJo2IK1FaVEqabo2YivZhizZzYgztQGA2WPqVZMogwmVkDM0+N15tspB+cjpBB4AjxXErYSjpfZ4kBXSqgNlF2JFIJWnHYbt3l5+9xw34YLiFyPL59zHf/zpl+mXqye6SXlcHL//g8voRY2bLuKXQP1MGH7ZJPL1vchgfFGRLXgctmGfjxHsH2yRT1ISLXmo/+xvVvgnjta46uPj1F0HTzHQRmNKyTaaECSKgNaM1QpHGG3MNJTZONQhd3TbXpKHgfr4SmpEvJwhJxnm7ymzclZfpayQdg/22amFbFtHCoFj4GKv+gzJd/hmamY8XpIECsSZZjL7k1BrFDKUPAkaWrYOxuwti9gTV+JIFZsGKjwuivWc9vDB9k13SJOFZ4rUZlHrcAgHYHQhzY4PQciZT3aXQlDVY9mpLOpsRNEgEDgSMlQ1WOyEREkmji1QUdJXtU+6wR+kfvXn09/WOe3fvgd3LH16jO9pJwMg1Wed2y7Oq+dCo0wpR6kFH2JK+1zcl9fiTBRvP7KDYz2FplqRHzxkYPMtpMjJrr2zwWEScrDe+fwHUGt5OE5klaUEtZD4lTw6IF5fmDrEM1IESuN70hGewpH1KenW5seLYhxJZBPjq1MTukp54Mf/CA/+qM/ymc/+1muuuoqbrnlFm677Tb+8A//MPf/ysk5RYIkZaoZs2OyScmXeI7DQNln83CFHRNN5oKEzUMVVveWEEIsOU72D9855P81FyREqWKw4tOrQbp2B9KvSMbrIY4UGA0zrZjBqk8jSNgz0+6qEpSGA/MhUkDBlUSpHUd5cqxBK1Y2mCi7wUw1E5IT2DZV0L1bG6ARKarZzVcKc0LHyDk1HK24bvfDfGvTZQDs7l/Dq976cfb1rjrDK8s5FkezS/Ck9ZSNUoMxR4amLGzexqlm+0STgivZMlKl6DlMNCIbsiAF9SChp+ixYaBM2Xe7Be5T4w3KvsPGoQoz7Zj5ICFVBi/bUGrFKe04JUoVJd9ZZLewZbhKmOhu0XqqqoQ8HCHn2SavaXNylpejeZH3lX16Sx7NMGXHZJP+cv+ia3oQK6SQ1MOYuSDGyaY3Cq6LUoYwVYSpwhiIteaB3bOszqbIrt0yyOsvX8/rLl3Hd/fO8uiBOt/cNsF9O2esBZcBz5F4niBMtFXZmUN32CBVHJzXVgWozSJ14NEQYBPotaHs22mToufiu5KiJ3GkYOdU+1n5Gb/YOXdyF41CmQM9IwB84Ia38sGX/2fmSj1neGU5HTqq2v6yRyGrO02mdE9SgxBZ7slhnznak6A2hvF6QK3ksa6/TK3o0oxSip6tOdf2lbjj8QlmWgmjPYXsvE+pFV3OHqny8L45njhYp+RKBquHBASOFPiug1aa3dNtBFMEic2DcKWkr2TDexfWp6dTm55oEGNOzkJOqWn7tre9jZ6eHl772tdSrVa56KKLeOSRR9iwYcNyry8n50XB9okGf33PbgBqRZdUGaSA8XrAWN3aEvRXfLaO1BbdGDoKte/umeWBXTM8PdHs+n81w4TYwGw7IVWwrr9EKSsq+8oe9TDFdwVDVZ+Jhh2B7gaIYW+0YaK7abpSClqxQQqB1gbfFWhtd06VVie1c+osuFEnSmfHWK6fZs7hbJnayx9/8U+45ODTvOGNv8+3N1wEkDdsnwcYrEJBCvA8iS8F7dg2Sau+w1Q7IVkwxnlYDkpmaSBINYzNh6TacMGaHnpKLq+7Yh0P75nHlYJzVtlCsRGmxEqzqlbg6YkGM62YG88bIUisnUqc6swv0KEdJ9zxxCRzbauyd51Dyb39ZW9REMT+ueCUVAl5OELOs01e0+bkLC9H83usFV0GKwX2z7WZaUY0wpSekr0fdBodW4ar7J5qobWhXHBxs2u770mCRKGMvSf6rsSVgp1TLWpFj3NHa13PyoGKz46JJgXXYVVPgX1zAcZAlCoSJagVXFJjCBOFjw3P9R2BlBKlrerWAOhje3J27rOp0sRK4Cm7zsvW9zEfJnx39+yz8NN9cSO14pe+8y+8666/5YF15/P//Oz7MULS9ku0yTdvVxKutDVcI7Z1JdmGiFZ0G7auzJ7/jtPAtRskECQKKSUj1cIRzdH9cwEP7Z1lthWza7rVbboOlH22jFQouA6NMGVgsLzouuQIgRQCI6ERJozNh6zqLeI5LonSHJwPkEIw1Yhg1H7mVGvTEw1izMk5nFOeJ3z9619PrVbjda97Ha973evy4jYn5xRZqEi4dEMfs+2E7RNNZtsxQlgVnDFw2fq+I0bJAIqeZM90iyhrpgxWfIJE4TiSog9hrAiSlJlWxBrPqnQ9R+IIGKgWGK4WODDXUd4eKlAXWh0oQ9cLzGS30CQ12c3IHNcWYSECq6pLlSZKDZ7j0F90mGzG9muepBHlHdzlQGrFWx74PL/5zf9DQSXMFyrUolz1sdLxJJk/tP2970pENl6WKuvpVfIcImXwpSTmkDe1ZEHDNrMuAcAYakWPIFY8NdFkQ3+Z4WqBqWbEmr7SoutOqjSuYx+I54OEyWbEmr4ysLjZKgRcvL4XKQRBoljdW2S4ViBMNE9PNBcVraejSsjDEXKebfKaNidn+TjaZIUQgi0jFebDmOlmzGw7plxwaEcpO6daVAou54xW+dZTE7jSyZRyAgNdX3WjOvc4O+21eaiC60i2jTW44VyruvzKo+PsmWmTpppYWe91bUAY0BjmwwRramC6gWPKgC8g0va4xtichs6daqky12B9Nn1XZv62Dn1Fl6fGG+ybDboZDznLw6aZ/fzxbX/CZQe2AdD2SpSSiPZRJnhyzixSCjuVKSSJMlSypmaibHdWYjdMOlZeRz0OgIBEgSMNzSjhO7umeeygy7mjPd0684mDdZ4aa+B1rQ/cri1gI0oYrPh2Mu2w7+W7kpInmWzY5+1a0aPgWttA35G4jlXNP7x3juu2DHUbsSdbmx5tAuF42Q45OXCKTdtbb721++tLLrmEX/mVX+Gee+5hYGAAgD/5kz9ZntXl5KxwlsOT5nBFwkDF58qN/V3F22wr5vv75kj10o3MyYZVK/iupFb0EEJ0dw0LrlUNJEpTD1MGEoWUgtlWQq3o85aXbeLhPbM0wgQpbKNoYXjYoj9r9oJe+PtTsDMQ2J+bwY5395c9zlvdw6P75wkSxUDFJ9UhQR65e1qcNXuAP7rtw1y5/3EAvr75ct7zw/+FidrgGV5ZzvFwpEAicIXBYPAdQaSsRx7Y0bJ2bBVHh6sSOr1QR4AUAoF9qHSloORJEHBwLuTys/qpFlw7bpo4PLJ/niBOqRY9vKJLogz1ICFIVHZ9Ki3ZaL38rAFueskqbn/cFq27p9tLFq2nq5jNwxFyni3ymjYnx7JcPovH8nscqBQ4d1WNJ02DIFY8sm+eqWaMyAQAX3rkIO1UUy44zLRiWpE1/El1prZzrPDg4nX9bBmudkekO56TAA/tnWWyEdKOFFFqKHoOcapQ2tawtpw2FFyJNNYnN1aGdqLR2tga9Tj+mws3SDF2asw2lqyCN9UG1wGdHt9mIefYCKP5hQf/nXff+TeU0oi6X+b3bvpl/u8FrzxU9OSsOKLMK/qqTf08cbBJO05tmK3QpMrQeax1MkmtWuIYvgMV3yNMNalWGG3Psbl2QpQaGmEC2GvXA7tmSbVhoOJ3m64dW8CZVsyUia3NQZjSV/K6Na0QgrInSbVBAEprlNak2mY0lH2Xs0cq7JhsHZG7cDK16dEmEDprOFq2Q04OnGLT9qGHHur+2vd9Xv7yl7N792527959xD/CnJwXKsvlSbOUIkFkFgcAvSWP7ZNNDs6HixonxtimyvaJJo4UuFLiSoiyYrHjd1ktuDZkSGlmg4SK51DwJD+wdYjXX76eWtHlS4+No5RGCHVc1eyJ+Hwdi44yASHoK9vUec+V3TG4guPgOhKRWTPknDw/8/3b+b2vfZJyEtH0S7zvlb/IZy/6oby4fZ7gOZKhqk+QaNpRSq3k4SeKuSDJAlQEuqt5tw+yXpbyVy64pMpej5qxIla2CK2VXBCCZpjiOoIrNg5YNYEj2TZWJ4hTBip+9/pScAW1okMzcghizVPjDdb0lZZstG4dqbF15PhF6+kqZvNwhJxng7ymzXmxsVRzdudUs3ttDhKFNobVvUVufMkqrl+gLjsR1vaV2Dxc4f5dM6ztK1FwnW6wrjGGING8+sI1XLCuh3+8fy9CwOahKpWCy0Q95Pv77Ca+JwWpsBMmAGSbkf1ljy3D1W6dvNATXWvD7qkWYZISKyta6C06pFrSCFPCBXZCqdI40oaI9XiCZqRI1PHtviTguaK7kZooTbXoUfGtsq/k24ZRnObhuqfLUGuWP/u3P+TaPY8A8M2Nl/KeH/kvHOwZPsMryzkRtIa5VsJVmwbYPxcw24pRRjPXtpkItYKLEFAPFcIcNrkpwHcdHGktwiq+Z20VtGa4p4jRsHOyxd/ft4c3XbeRyUbI6t4i80GSTagdaspWCg6TjZgNA2Xm2jHTrZha0cVzJI0wYbwRW5svVzDZiJgNEnpLHqt7S9m1xmXXVOuI3AU48dr0VLMdcnLgFJu2X//615d7HTk5zysWetKM9hSoapf5IOGbT0/w6IE5fu7qs064yD1eAmWYKDYMlKlkoxOre4vW/2b/POP1yCpXgXqYMh8kGAxpphRIszGvoivpKfuct6pGrDRr+0q88eoNSCnsuIg2xKnGEYsT6JfidAtQg204l32H/rKf7XA6xKlmPky642ieI4jzYLJTIpUO5STing0X8e4f/fXcu/Z5RjNSSBEjpWDdQJmL1vZy57YJ23z1XRqxwsEGBEoBUWrPGkcKtNJgBImyCt0EcKWk6NlR076KR3/Z5yWjPaztKzFcK3LfMzOM1ApHKGlbkWL9QIn+sseGgQpTzeiojdYTLVpzxWzOSiOvaXNeTCwlOOgreUw0I5Q2lDzJbCtmshnx8N457tw2yQ3njvBz12w47sZapxn8xFid3dMtdk+32TbWoFJwGa4WWNNXJEg0AxWfHzrfTmgYAxev6+vef1b1FCi6knqYMFj2GKoWaMaKyUaMJw1hauvaasHpft+Fnuj375phohGRaE2UdILFJL4rF9WvbuahmSjDfJDgiAWj2sdBc6iR7AgQQjJaK6KNoVr0qAfxcWvpnBOjUagw1Jqj5RX5/Rvewt9d8iO5AOF5QMePVgjYNxcQpJorNvbjO1ZB/93dM1b9bgxxanAkFIREcyhPpeQKKr5LPUzwHGthEKaaWslnoGztAsfrEffunOH6s4eIlObc0R4e2T/PTCummjVlbUiuwnUEP3PFOu7fNctT4w070ZrqbshuyZes6yvhSEkjSvEdyeahCgMVn0aYLJm7cDIc73n/aNkOOTlwik3b+fl5lFLd0bEOMzMzuK5LT0+e3JjzwmWhJ81gxefJsSZj9YD5dkKi7I3mqbEmP33pWn74wtHjFrkdr8dH9s8z2mNItMHP0tkBDs6HXLahn5vOH+H2xya4a/skjx+oEytNwZX0ljxaUWr/wzY7hRBoY1DaECbgu4I+YW+iV20c4OYLDq1rtFbEkbZB6snj//mXowhVGpTSTDYjNvSXKbnWuzfV0Ft0mG7GGAyugDSveo+PMYw2phnrGQLgX156A41ChTu2XokRJ/CXmrPiSJTBxTZfZ9oJ9Si1ioA4RRjrVdspih1pwwFLnoMUgkrBjov2Fj16Sh4DlSK1kktf0WW8EXHRur5uo/SKjf18+dGDdsRMsKDATSn5DueuqjEfJPzUpWuoFb1labTmitmclURe0+a8WFgqBKcVJdy1Y4ogVly8rofHDtRpx4pq0aWvVGSqab8epoq3vGzTUWvaTjP4ob22IZIqQ1/Zo7fk0YxSdk23GK+H3HDeCD939QYKrrPkqHAzUriupOQ51CNFX8VaKgSxYj6wzRulrT3PmswDveOJHsSKf3loP63YTpyBdfFKYk0Q60W9vo4dmOw0b0+y1uxYhiUapNAgDGmqUVrRzEa2c06N4eYM0+VetHSIXJ9f+8n3ELgF9vSvPtNLyzkBrBJdEqc2aDpMNPvnAoKnFS8/e5hq0QYBamODyYzJ+vBC4AibbRIk2gp3BN3n3TDVeI5goHzI2qC37DHdjJhu2E2nIE7ZOlxhrB4y205oRtaSoSNYeOV5q7h+6xBffmSM7++bY9tEg6LnsHl1mTC1tW9v0aVadJlpxeycatFf9o6Zu3CinE62Q07OKTVt3/CGN/DjP/7j/Oqv/uqi1z/72c/yb//2b3zxi19clsXl5KxEOp40JU/y8L555oOYVpSijaHgOcRZM/IbT01wsB4eMwlSa8Pe2TaJUjw13uC7e2YpeZKC51AruJQLLhsGyl1Fm1bwhe8fwHUEa/oqVAoOqTLMBZmnj4FIGQqOTcPsDHkpDeWCi5A20EFr2DvTphWnPLR71qpfgfg5MN5yhf1ekTJoo9g13WbvbJsotWNqe2cj4o55Wc5xGW7O8Adf/jMuHN/Bq97yMeZLNRCCr5199ZleWs5pUPId1vSVqAcJj+6fI0g0rhQ2iEGASe3JapXrZMUvOK7gZ65Yz3eemWHPTJtGmNKIWoBt7p6zqrbIO/Ylq3s4Z7TGbCumFStaUYojJSM9RbYMV/AcSZhoakXviEbrcvkf5uScSfKaNufFwNFCcEAgsT6O33p6GmUMniNpx4qSZxWqceZt/pVHx9j8iiNDcjrN4KlGyNh8iNbQW3KJUqt0vWhtLyXfYf9cwGDFZ/NQlacmGotGhY0xNMKUsXpAnKlxp1sx80FMM0oz1Z4gybIe7t89w9BUgZLvMFgtsHm4wv/3rZ08cbCOzrwpF/ZhMxvbxaVl9ptT0QZ08x+y++9TY02ktBufscrtvU4JY/ipx7/B793+ST5xzc/wyWteB8C24Y1ndl05J4wroVJwiRLrUCuxgdkIwUwr5p4dU5w1WCZMNNoYUmVIsh0UKRVS2Kkx17Enp1Iao21zt7fsMVD2KC1SoloLlG/vnGHvTMBMK6K35DFQ9nnJaI1ywcWTgrF6uEiw8Ks3VHlg9wx/fucOBisFRnuLzLZjvrd3rqvSLRdcxush398/z7r+8jFzF06E0812yHlxc0oSrPvuu48bbrjhiNdf8YpXcN99953wcf7gD/6AK6+8klqtxsjICD/1Uz/Ftm3bFr0nDENuueUWBgcHqVarvPa1r2V8fHzRe/bs2cOrX/1qyuUyIyMj/OZv/iZputgP5Bvf+AaXXXYZhUKBrVu38ulPf/qI9XzsYx9j48aNFItFrr76ar7zne+c9FpyXvi04pQgSTkwF9KOU5Syu4Vl36XgSiqegzKGRBmmmxFffWwcvUSC7PaJBu//wuO85dP386lv7mTfbJt6mDAfWNXswfmQepDwyvNGbMNWG/75oX3MBwnr+kv0lDwcKTHGYLQ5pCDIdi5jZXcvPUfgOnaUek1vifuemeE3P/cwv/V/v8/v/PMj/PFXtnWDyJ4LlMmUD8oQJZpWlKC1oVZ0GKz6pFrZoIglEj5zFvNjT3yTr/7FLdy04376gjpXZKFjOc9PJLYh60pBT9FlpmkTtsfqkT0fshPCGDvGmWoW+X+l2liFkdL0lDx8xxbKovP4uOC9Whv2zrRpRAmbhyoMVwtcs2mAqzYNcu3mQa44q5/+ss/B+ZCtI9Ujdv63TzT4xDd28KHbn+IjdzzNh25/ik98YwfbJxrP8k8pJ2d5WY6aNq9nc1Y6RwvBiZWmHSvasaIVq+4YMsB0K2b/XMh4I+LAXMhtj4xx946pRcftNIP3TLeZaETZZmHC2HxEM0yZbsWM1yOGqgW2DFe7YT4LR4VnWhEP7JrlG09NcM+OacYbIXtnAtpRSjNUtOOUdmzHmwuegyMFShn2zgZsn2iyZ7rF//7WTr70yEEaQYLrCFx59IdckU2esYx1psKqbtuJJttTfc7q6hcCg605Pvmvv8+Hv/DH9EYtbtz+HaReKpoqZyXjCGhHqZ0WcwSFLOBLa0OSKg7MB9y/a5Y4VaQaEmXPRSlBIFDaPhsKAbWCy7mra5R9h7LvsKa3uKhha4xhqhERp4bJZsR5o1UGqwXCRHNgPuCpiQZBnDLeiBisFhY1RKW02TEFz2GkpxMEXuCS9X2M1IqE2fNpmCg2D1WPKcA6GTrZDhes6WWunbBrqsVcO+HCtb3L9j1yXpicktI2iqIjikiAJEkIguCEj3PnnXdyyy23cOWVV5KmKf/tv/03XvWqV/H4449TqVQAeNe73sVtt93GP/3TP9Hb28uv/dqv8ZrXvIa7774bAKUUr371qxkdHeWee+7h4MGD/Of//J/xPI/f//3fB+CZZ57h1a9+NW9/+9v5u7/7O+644w7e9ra3sXr1am6++WYA/vEf/5Fbb72VT37yk1x99dV8+MMf5uabb2bbtm2MjIyc0FpyXth0VGVj8yGtSDHZiCi4krm0Y3ievQ/wpKQZptRWuUsmQW6faPDhrz3Nw3vnaIQpniNsknuismAEl0vX9zHbjrnr6SnOHql2PyeETdGNEoXSholmRDuxqbhgdzl7Sx6p0iQaiq61PphrxcwFMQfnAsbqIb4j0VrTihXa2OaPJ21fR52gr9fJ0imgO8dWWNVvPbTXE9lOusVuztHpb8/zv776CX5s210APLpqC7/x6nflaoQVwOkE9Wmg5EoqBTuWlSiDMta/z8NudHTe11ERdXzzPMeOlRng7u1TDFYL3PzSVTQjRaw0viOpFhy2T7b4zH17GCj77JxqEaaKONVMNiJaieLskSol36GZbRwttfO/1IhtO0559MA8B+aDvPDMeV6xHDVtXs/mrHSOFoLjSUE7C7DFGFwpSDU2jExngZdCUPYlM62Iv//OHlb3FrvX+P1zAQ/tnWWiEdKObYAZxhBqaMYpAgjjeWpFl/6Kz0wrZvtEg1W9RYaqPk+M1ZltxdSDlFasEAhcKQjTLMndGJLYIIV9aFUG1g+U2TRc4dF980y1YuLpNr5j/xzGgJPVskvVsSb7n2e71HSEXWvO8fnhbXfzga98jMGgTiId/vT6N/KJa34GLZ3jfzhnWZEcOm88KfBcSSs+8eZ5ou1GiBRWzLSqVmA+TGwAmYayK/EciesI2nFs61nT+b6ZMAGQWPGCKyXnrq4xPh8y3YyolbyujVc9SGjGiv6Sz8XrepFSUil4bJ9oMtuKmG7GPGkavGzrEFdsGqDgOmhtuvXsUh6zA5UC/Rt9GmHKbDsmSBRvvn4jGwYry/YzzrMdck4FYczxsuKP5IYbbuCCCy7gz/7szxa9fsstt/D973+fb33rW6e0mMnJSUZGRrjzzjt5+ctfzvz8PMPDw3zmM5/hda+zIxJPPvkkL3nJS7j33nu55ppr+NKXvsSP/diPceDAAVatsmE7n/zkJ3nPe97D5OQkvu/znve8h9tuu41HH320+73e8IY3MDc3x5e//GUArr76aq688ko++tGPAjaZcP369bzjHe/gt37rt05oLcejXq/T29vL/Px87pH2PGNhcEOQpGwbazJRDxntLTDdSih5Tpbqbgvdsu9Q8hyu2NhPI0x5x41nc96o/TvX2vDxr2/n375/gChRtmnrSlwpMAbasR0D6yl4eK4gjBVbV9VwheDxsTqtKM0u7IJU6SyISJGozPQdW7AqbW+aUtqFFTwbNjEfJEhhtXdGGyKlMcZ0m6WdYLNng8yyqKtsOHx8Lef43PT0ffzBl/+M4fYcqZB89Lqf5aPX/iypkxvXrwQW/ps+lX/fq3sLjFSLzAUxBdfh4LxtGvWUPCYbIUvVzlJA2XOQUrCqx6oMrtk8wJq+I31j9860eGjPHBsGy2wZrlL2XdpxytPjTephwnDN7147toxUed1l6zln9FADVmvDJ76xg0cPzB82YmtVD09PNLlwbS9v/8EteQGa86yxnPXUs1HT5vVszkpj70ybD93+FH1lb1EIznw75guPHCSIFYnS9BStrUGiTGbJo5FSsHGwgjaG0d4iV28a7F7jHz84zzv/4XsoZXAl7JppgwHXsb7rkdIobSh7DuWCQ5hoRmoFVvcWUdqwbbxJqjTFbEzYkdCKbBNZCnAdgdJ2AsWVAm1gw0AZzxHsnLLWP1II+sse+2aDbqNUCvtfLgRYufQGDX7va5/kpx6/E4AnhjfyG6++lcdXbT7DK3vxIYG+kovnOswFMVpbq4+1fUXG6iGN6OiN26VqXVfCYKXAhoESB+YCZtoJPUWXnpLLRD0m1daztiNG6FSLUlgRAsDqvhLXbR7ilS8Z4V8e2s9TYw2U6WzH2OyWONVcvWmA9QOHmqodq5U9My12TwdsHi7jOpKi67BluMrNF6zqTrDm9WzOmeZE66lTesp///vfz0033cTDDz/MjTfeCMAdd9zB/fffz1e/+tVTWzE2DALohkE8+OCDJEnCTTfd1H3Peeedx4YNG7qF5b333suFF17YLXABbr75Zn7lV36Fxx57jEsvvZR777130TE673nnO98JQBzHPPjgg/z2b/929+tSSm666SbuvffeE15LzguT7RMN/uJbz7BzqknFd+kr+WwZLjM2H3BgLsR1BKm24V9xan0ni66D0oZ6cGTa5P65gEf2z9uUXt+hHqY42Y1CCGxqZWhDzUZ7ioSp5qnxBnNZ0JnnCFwjMAhipXGF9fnq7FB2mqKGzIpAWbWCzsbfHCHoK7nWUxZb7DqOwKR2pOvZvC11lIELf59zcvzotrsYbs/x1OAGbv2xW3l0dOuZXtKLGlfQtfOAxf+mT/bft8A+mNajhL6yx1w7IUgUjhS0Y4XvSJTSHF46F1yJ70kGKwW2jlR5aM8sjjxyMNQYw4G5kCBRrO0rdR/ca0WPSzf08d3ds8wHKbWCizKGyXrE7Y+PIyWLVFVLjdiCvQ6t7i0uOV2Qk7NSeTZq2ryezVlpHC0EJ8kaqlF2r+k0bIWwljtSCjxH0oxS1g+U2TxU4enxBg/snqGn5LFzskkQK2oFl/kg6dajxuhuDaqBWCmStqbou7SihG3jCa4UNEJrkdUIU2tbIATGGDxHIIUgVTZUV2mDdmzjdudUkzRr9lhVrcANF6tbdabey3QLeb25AhlpzvAj2+5GCcknrnkdH7nujcSud/wP5iw7niuoljxrPWI82lEKGPbNBThC4DnWyuBoSKzFgdF2ijLV0IwSZloO82FK2XfoLXlMtxISZXAcQa/n0AhTEm1sHetIDNbnFgEXr+/jzS+zk1tnDZb58qNjPLJ/nnZsxVGjvUWemWwdIVAQQpBqw1g9ohklVsnaX15yIiz3mM15vnBKTdvrr7+ee++9lw9+8IN89rOfpVQqcdFFF/EXf/EXnH322ae0EK0173znO7n++uu54IILABgbG8P3ffr6+ha9d9WqVYyNjXXfs7DA7Xy987VjvaderxMEAbOzsyillnzPk08+ecJrOZwoioiiqPv7er1+vB9Dzgqjo4q986kp4tTaCDhS0Ft0WdVTZKIZkaQapa3Ruu9aj9mJRojnSB7aM8fWVTWCBfK4RpQw3Y6IUo0nHQSQam0bqwbCJCVVBkfCZDPGaBvQVfSsnUGqQRuNNlaBoI1V3HYQwt40FyKyncsoNSgM9Sil4EhbEGftXpHtldZKLklqaB/r7pzznOJohcrGxP7nTb/Mrv41/PnVryVy/TO8shcvnQK1VnBphOkpq9M7CgWBfeCMUs1MK2GmZQMOlQaBIUpUFsRiP+AIuoENw9UC6/pLbBmpEiY6S9c+ckX1MOHgXGCDUlKrsO88uM+2EyabEY0w5WVbhxjpKS5Z4B5txLZDyXcYr4e04iPHzXNyViLLXdPm9WzOSqQTgrN/rs3D++boL/vUijYwKNWGwWqBoifZPxuQKI3rCHxH4kjbACn7DluGK0Sp5rGDdf78zp0UPEmYCQISpUiUbb60I0VszKJN+kSBEIZBX5IokwWdORRcQTu2jV1fQslzCBKV1aS25tbZgSSQKN2dOBGAA93m7uGN2TTv1K44FtazTw+fxe/c/GtsH1zP99ace4ZX9uJGZ8+xL13dS5QqvvX0FM1IZVYF4LsS3xekWpNmWS6d00uKQ7YjnivQqT0XW7FmohHhCBgo+8wHKe04pafokmqD40h6Sh71IEEgkBL6ywWKnkN/2eddN53DWZktwdaRGr96mKWANoY//drTi+wNwAoUtk80aYYpvSWPvrKPIwW1oke14PL0RJOvPjbO5qFq12O2M007Xg8puA4Xru3thoDn5KwETnme9pJLLuHv/u7vlm0ht9xyC48++ih33XXXsh3zTPMHf/AH/N7v/d6ZXkbOafBPD+7l9scnUFp3EygTbZhpW4VAX8mzdgieQ6o1YWKLVt+VVHyHStGeYn997y7efP1GAP71u/vZPxNQD1NaoU2ZbcYaJ9sZ7IyKBInBSVOKnvXuKWbhC80oRRuRBZCBytQMrrSFa3KYiVZnf9BzJFFqPRSS1I6xlT2IU0OirKJCAK4QtgpOnpufcc7RKccB/+3rf8lIa5Zf+unfASGoF6t85Po3numlvWjpeNYWXIHGevadzoPhwqK3r+LjOYIgURCD64DvSRJtCFPdbezKbPNICigXHC47q58NA2WMMXx/Zp7hms94PWS0p4jMFLczrYiH984x3owoew6P7J9n/5wNGesvWw+wVBnKvsRz5VEL3KU8wBYSxOqI6YKcnJXOcta0eT2bs5Ipeg6TjYgnDzbQGHqLrvUmL7hcsq6Xp8ab3PfMTLapZ4NtByo+V27sB+DB3bM0w5SBiseqnhJj820MhvlAobTONgIXN2zhUFNnpp3gO5JKwSFONYk6pIaFTN0rBI60Pu7amK7NgTK2WdT1dAdiBb7UpEKwxF5lzgri2t0P8wdf/ii//uP/lYezJu3nLrzpOJ/KeS4wxjBej9B6Dm1sqG3nFDaANgajjPW+xircF05ndqz1Em3r4yg9tIkSpjYTJkqtCAlsBkMQK1wpKPkOgxUfbeDS9f204pSL1vWxvn+xglZKwfqBcjdjphWlDFUL7Jlpcc6qQ9MDjTBlthUhsJtRteKhenSpibDcYzbn+cCyPlU1Gg1+/dd/HYDe3l4+9KEPndDnfu3Xfo0vfOELfPOb32TdunXd10dHR4njmLm5uUWKgPHxcUZHR7vvOTwVt5OAu/A9h6fijo+P09PTQ6lUwnEcHMdZ8j0Lj3G8tRzOb//2b3Prrbd2f1+v11m/fv2J/EhyVgBpqvmH7+wlVor+stcdNy5IgScF9TBFacNQxae/4rNjokWUakqeQ2/ZY3VviS3DFfrLPk9PNPn7+/YQJNqG9/SVCKeamWLA+n1pzBGhBfZ+YUejPUdQcB2S1MYQNSP7/TsfKboOviu7r3eO1bEkCLIbcOf97VgRp9b7dmGd24wSjmFdlPMccdXeR/ngFz/MWXNW+XTJwadyJcIKoKOK7RazJ9GwFdhzeqFCofN60XNY12ctB+baCWGi6Sn6CAFz7aT7vRTguzYoxnclRVcyVg8RAh7bXydRhsGqx96ZgLH5iAvW9lD0HB7cPct0K8Z3JEM1n5LvMtkIaUYpW4erzLZjCp5EG/CdQ9YKhxe4RxuxteszHJwPuXBtL2v7llbi5uQ8XziVmjavZ3NWKp0AyT0zbYqeQ8V3CJSmHiocmVItemyfbLGmr8i5o1X2zbSRUlL2Xa7c2M9Axef+Z2aYCxI2D1VY3Vtitp2wfy7MBAN2EkxypOK1gwGSrHHjSIkx+pAHvLA2B0prHClJte4G47pSIIRtBCttNy8XbpbG2to8SGFDjZb6/nmGwpmjmIS8586/5s0P/jsA77z7M7z5Z/JNoJWEMlbFvn8uJEr0EdNjSmebOFnImFzwjCk6hTGAMUgh8RxIlSFVVpmbYs/bkmeDblOl0dkmjOdK2pm6frwRctZg5ai2BIszZhT1MGGyETHVjLhwbS/lgstsO2Y2SBiq+mwZrhxh5VXyHcbmrdXXwiZtbumVs5I5pabta17zmiVfj6KIL3/5y/zzP/8zxWLxuMcxxvCOd7yDf/mXf+Eb3/gGmzZtWvT1yy+/HM/zuOOOO3jta18LwLZt29izZw/XXnstANdeey0f+MAHmJiY6Kbi3n777fT09HD++ed33/PFL35x0bFvv/327jF83+fyyy/njjvu4Kd+6qcAO952xx138Gu/9msnvJbDKRQKFAqF4/4cclYm3907y9h8QNF1MIc5vcpsZ7AZpZw9UuVnr1rPPz+4v9uwLbgOteKhhsaqms83n56kt+RxwZpehmo+jTBhz3QbrW06vDpstEsIeyMr+w5zgaadaIpZYSqwN1RjwJOH3qu06Y6RCcBzMl9bfeSYmMHuiDqHVbJ5w/bMUkgi/uu3/g9vvf/zSAz7eob5zR99Z96wXSF0RsXizCv6ZPbhDfZ88x2bjL1w7Gy0p0AjVCBguFpg31xAlCoKntMdPVMmO+cdScETjPYUSbVh52STfbMBZd/h0g29rOkrc2CuzaMH6nx3zxzGGIJEsaa3SKIMcarwHcFAluRt/QGtXcKq3tIiVQIstjzojNjmHmA5LxSWo6bN69mclYzWhq88Os6emTazrchuCpY9Bh1JnCrG6tb64rotg8wHKQXXwXcdPFfy0jU1esseB+dDds+06Sv7bB2pMdtO+N7eOYLYqt2SVDGf3cMW1pSdzcqFXrNJqkiU07XuSpEYrZFCoIzBE3bN3U1RAVXPIUo1kNn6HLZj2rEYW7ix6iwII8sbtmeGy/Y/wR/d9iE2zx4A4G8v+RF+/4a3nOFV5RyOAYJY21yUJb6mFqh+rFWg/b3AigkwHaFRNrkp7Ht6yj4lTzJej0iVph2n3VrWPqcKfEcw37Z+2Gv6Sl07rsPpbDzNtGJKnmS2FTPZjJhv28btbCthbX+JkucwVPE5d1WNgcqR962DcwG7pgP+/jt7rP3gYQFlOTkrkVNq2v7rv/4rr3/96ymVFitpgsCmXP/kT/7kCR3nlltu4TOf+Qyf//znqdVqXS+t3t5eSqUSvb29vPWtb+XWW29lYGCAnp4e3vGOd3Dttdd2gxJe9apXcf755/PzP//z/OEf/iFjY2P89//+37nlllu6Bebb3/52PvrRj/Lud7+bt7zlLfzHf/wHn/3sZ7ntttu6a7n11lt505vexBVXXMFVV13Fhz/8YVqtFm9+85u7azreWnJOj+64wwoZTZhuxWigUnAIEo3j2QKzgyvtruRAxWfzUJWi77BpqJqNfhxiphXxyL559s0GNGNFM1IMlH3OGqww1YgJEmV3HAHfoavoVZlvEFi7hShRRIkCBFofaswm2XhKnKSUfZeGOdTM9RwHlSirPoAjlLwc9lp3o3RZfoI5J8tFB5/iT77wJ2yd2QfAP1z0Kt7/yrfRLOS7vyuJzkOhzs6zkzlhEm2Vsq60ntQlz8EARd8GMgCUPZeC61DwHBKV+VdLQcG1adzDtSK+I7hm8yCpNnzr6UmGawWu3TzYtUNYP1BhbV+JO5+aZO9sQK3g0o4VylgvwWQ+ZKDqd1UJYaIYqCytSjjc8iD3AMt5IbEcNW1ez+YczkqqaffPBWyfaNCOUsLE1q2d63zRcxntEUw2IjxH8s6bzqadKKYaEd/bM8fOqRa7plqEiaJWdLl8Qz/9ZY/7d80SxGn3WKkq0oxaR9SZdrR6cS83VRCnmrLvECcKbQwF6WCM6TZrywUXpQ1BrBCYbLPSWvOorAZe2Iw9vNEksLVwkhe0ZwQ/TXjXXX/HL33nn3GM5mB1kPf8yH/hm5svP9NLyzkKnclMOP7z4EJrhM6mjDJWiOA5VinvuZLVmbhgUsQ23FAdqqERtvYd7SnSilI8V7K+v8zmoeoR36+z8TTTihmseDy8b94GIBZd+ssek40YZQyVgsMbrtrAI/vmeexAfVF2A8B0M+L+XbOUPIc1vSUqBXfJ/IacnJXGKdsjfOQjH+kqATqMjY3xT//0Tyd8jE984hMAvOIVr1j0+l/91V/xC7/wCwB86EMfQkrJa1/7WqIo4uabb+bjH/94972O4/CFL3yBX/mVX+Haa6+lUqnwpje9ife9733d92zatInbbruNd73rXfzpn/4p69at43//7//NzTff3H3Pz/7szzI5Ocl73/texsbGuOSSS/jyl7+8KMzheGvJOXUWjjuEqVoRu16DFR/fkd1R5CBR+K7EyZQA7UjhSskPnjtCregt6fM404r43t45phoRUggGyz6uI5hohNCEWsll/UCJ+SBhLPOgDBPFXDvJUjoNzTDNVLT2mFLYwIaFibgGaCeGSCVWOSsAIQgTlY07Z6paacdY1GHjYwXHeuHmdmBnDqkVH/rCH7NlZj/j1QF+64ffwde3XHmml5WzgIU+epWsoaqykUx9nAdDJztfhYDBaoEo1USpDXtRurPRYgdLgyR7EMbgFBy0NhRcJwuE0SitGegt01Oy6idtYNNQpduw7TAXpNTDlCBWrOopUiu6JMqgdESUaOaDxKqQlGZdf5lKwaW/vDjc7miWB7kHWM4LidOtafN6NmchK62mbcUps0FCM0qpFt0jNuZ81wbp7phsIYTgvNEeGIXrtgx1r/H1IOHv79tD0ZPWM7IdUy163WOZzHvWkYJ4CYWAI23YkcE2WAWGkucwFySkSttsCOwEyEXr+hiq+uyfDZkNIvbNhhgMsbK5D4kyONn3ShfcfEX2fTxHEKdHWo4tF660E2x5P/jo/Oi2u/iV+z4HwP+94JX83o2/RL14ZDMuZ2WwlH2I6xyZk7IQ37Wht0YfynuQ0l4LCq5DpeDgOZKxehsDDFd9JhoxYM9dleW4TDYi1vSVOGdVlWemWl2v2YXsn7N2BqM9RbaNNQhitWjzqb/idUMRH9tf51Xnj3JwPlw0EdaOUr6zawaAqzb101Oyz+tL5TfktWzOSkMYczKOfBbHcRgbG2N4eHjR6+Pj46xZswal8vnqpajX6/T29jI/P09PT8+ZXs6KYeG4w+reImXf7np1Rm1PZtdrOZUNaap586fv56mJBsNVn7l2SpApAkRWPJ43WuMffvFapBR84hs7ePTAPFuHKzQjRZQqnjzYYK4dkRrrIbuur0TBczLD95B2rFjbVyJKNbunW1m4kMB1BK04JUoOjWB3GrWH+2E64tCOpe/IzOgdCq5DmGhSbUPHlLFj2Yk6VMh2ju0fZhqfc2a4au+j/Nz3vsTv3vR25kv5Tu+ZZqGI1ndtlnWYGlwBZd9FSjKFkW3etpOltz0cwaJwli0jVYquw8H5gHamFCj7Dv1lj3qoaEQpvQWX8UZErJRVFWWjZwXPYV1/mUs39NFf9vnunlkmGxE3vWQV7gIvWmMM9++aZf9cm5lmxLr+Mr1ZQ9YYw3Qrprfksb6/RJhq/tPVG/jKY+Pd6/Dhlge5+iBnJbGc9VRe0548eT17dJazpj0dFtbD9SDhk3fuYOdkk+FaEXlY0zZKVVaPFnnPj7zENm0PO9a+2TZ/edcunplusqa3yP27Zumv+EghumntQWLFDK0oxWADicAGihnsxIhWBi2g4rt2ek3Ye5JSBj+zFiv7DiO1IhsGy/SXPf7irmdsbZtZBXVUfiKriaWw92vHEZQ9x4Z3xuq0QkKPRT6VdgIYw4e+8Md86dzr+eo5S9uu5KwcvMMatAufPZdCCih60lqaaE2YHLL8Gu0tcv6aHvbNBsRJyr65ECGsDUI7VoueYyX2OnH92cOcO1pj11SLd9x4NueN9iy6ho3Nh/zj/XsZrhb4zq4Zip6k4Drd9WhjmG3HXLCmFykE7/qhc4hS1d08i1JFqux17NzRGusHKkf8mRphwlw74V0/dE7ub5vznHGi9dQpKW2NMdx4442USiV6enrYtGkTL3/5y7nuuutOecE5L04WjjucPVLt7pidyq7XcisbXFfyC9dv5A++9CSTzZjBsk9f2Y4Y18OU0arHr7xiK65rGyU3X7CKJ8bqfOXxcZTSBIlmrh3jSsFwrcC6/hLNMMXP3l/2XeaDhP1zbVJtrRbCBAqZqtZ3JCVPYoyxY9NZYepIkdkmZDuaZKNnBvpKHtPtGFfA+v4yM62IqVZsb5DG+gx1tmlsIyl7XeUN2+caRyt++b7/y3yxyt9d+qMAfGf9BXxn/QVneGU5HTT2PLH+XdaWRAC+J5HSnquRsp7Uo70lJhoh80Ha/bzAFqOOlDb12sqLCOIUN/PRskErNpG3v1Jgw6DLfTtnmGpF9JZc2rFgrh3T6QdrFO04pREkTDVjBqsFip5DkChqC5q2XSWU79L2FUGi6MnGxIQQ1Iou7ShlLki4etMgL9s6zJq+0jEtD1bSuG9OznKR17Q5y8Vy1rSnw+H1cMGRNIKUdqyJU0XRO/T4Z4yd6Oote/SV/K4NzlLHmmpG7J1ps2emTaoMcWo9MGdbMdoYRnuL1IOEdgwY+/NwHNkVDpQLLlXfYaDiobRgshlRdO390Sp17ebmbDthpFbgB88Z4mNf39Gtm1NtbOhRVrF26lljQGRK3mZsmzOL506Wl7xePpKzJ3fzG3f9Lb/xo++iVSiDELzrx//rmV5WzgkgsBOXKpu4LDhWLa8XfH3xdCa4jp02cyWsqhVxHRuEHSvNlWf1s3agzP7ZNvvmApSGasFZUnhkgCg1PHZgnp6itQYrew53PT3J7Y+P88xUC200UkimmhFxqq31grP4OmXXIqkVXWZaMa045bzRnkUTYZ3G75q+pRuyC/MbcnJWGqfUtP3d3/1dwIY0TE9Ps3PnTv7xH//xiHGbnJzjsW+2zff3z1H2rJ/jwvCuw1PLj7XrdaSyobQsHjU3vsSOE3767l3smm6RKI3nSM5f3cObrtvY/fpCokRTDxPCOCVKDa4jmAtSNgxWUBrG6iFxqjNlQ4rS9s9qVbaGVGtia11LwbFGYE7mF+Q4kp6Cy0w7zkzcrcKhMx02F8SkypAY++vR3hJSCBphSqQUibK7mp0bpyALIoO8Cn0O2TK9lz++7UNccvApArfA17ZexXht6EwvK2cJHAla2caqyixGfMdaFfSXC0w2I+IsxGu46hMmiiQLKSu4kqLn4Ht2dyWIFb0lr+tfWy44nDNaZetIjcvP6qfoOnzqmztIlMYYmGrGJKlV+XWuAXFq2DXVZv9cyA+cPcS7bjqH2x8f59ED81QL9pbeCFPGGyFBnOJIwfr+EokyzLRiqkUXz5FoY2wK+Ei1Gxx2LMuDlTbum5OzXOQ1bc5ysVw17elwtHp4shmhtGHfbMi6/hK+K0mUphmmFD2Hsudy9qraIhuc7RMN/vKuZ9g/F9Bf9tk4WGGw4vHo/jqzYUIQKwarPv1lHwMMVQtUfId2bLMaOpMonRpVa0OUagqua+26KDAXJMSp7vq3+67EdwRTzZh/fegAk42IjYNltIbJRkiQaoKsMes5gigTHdiNzUOFbK6Pf26QWvG2+/+V3/jW/6GgUg7UhnnfTb90ppeVswQdWw8AV1i/Z4F9LhRC4Lmiew5qVGbTt0AVmz2PVosevuvQDBN6Sh5XbBpgTW+JfbNtnhxrEirNo/vmGatH3SZt53n38EfNzvPoZCPi2ztnuPmlo/zV3c/w1cfHmW8nIKxHbsW3+TKTzYiKbxvGHaVtZ+NppKeY5T8cymCQUnSvtRXfpeQdaWXY4fD8hpyclcRpNW0XEkUR/+N//A/+6I/+iPe9731Uq1VuvfXW015gzguX7RMN/vbePTy6b56iL/Ech4Gyz5aRSjft8UR2vZZT2bCUkuzGl6ziB88e5rt7Z5luxQxWfC5b399V2C5cw3yQUC04gKG/7DPTinAEhInikX3zbBwq0Y4UYaqy8SpBwbWN10hl3pg6G0cxEBpNX8nF9R0mmglJqklce8fVBtRh7iZaG1wpiLVhPkjpKSkGqgXaicJkP0LNYnVC3qx97hBG85YH/o3f/ObfUExj6oUKv3vTLzNeHTzTS8s5Ciobw0yUoehJiq5LlGqqBQ/fsb58AOP1kDjb1Kn6EkcKfM8qC1wpcB3J5Wf188arNjBQ8bvegrWCx9q+EjunmvzZHdvZOdViuFag5El2zwbWHkFZVXzRFWhjMmWt5oFdM9yzc5KL1/eyf67NQ3vnaEcpjSilFSXMtBIqvssFa3vpLXnsmGgx045pRSnaWN/un7tqw6Km68ICt8OztSmWk7MSyGvanOVguWra0+FY9fBlG/oJYsX+uYDJRtT1se0te5Q9lw2D5e4GXudYn/n2Hh7YPYsE9s8GuI6kv+xz1aYBto037NRZb5GRWoGH983TilLiVLMu2ygMEkU7SlFaI6Wkt+hSLnq0opRnplt4WQhDEKtuuJiUUPIkc+2E8XpA0XfxXQcpBOu9Ms1I0YpT28DNU8bOKBtn9vNHX/wwV+x/AoA7tlzJJ6553RleVc7R0NrWkjKzGREGyp5ESkFP0aPkW2uRRpigjCHJOrzWK1rQX3aRUhKnGqUUBljVU2RNr93oCRLNqy8c5cJ1vfzl3bsQAqoFl0aYchT3MAyHJtqmmxE7Jxvsmg6YDxIb2Os6pNoQJBpPQiNSJKmdGh2uFUi1bdiWfJfNQxXG6tERGQwd1vaV2DJc7YocFm7MHi2/ISdnpbBsWwmFQoHf/d3fpVKpYLIxl5yco9FpAuybbVP0neziCRONkEaUcMn6PgYqhRPa9eqYk6/uLR6hjDlZte6xlGRXbTp6Y21hMm+Ualb1FAGbjlsPE8AwF8Q8st8GhdWKnlUWSGuDkGqDzhR1nmtvrGAbRs1I4WRFtDIwF6Yc7fRSkIWnGRwpGZuPMEZ3QyGW+lgePvbcsG5ujD/+4oe5eu+jANy56TLe88P/hbGeXGG7klEGPAllz6FccBDCnq9z7YTpllW2S2nwHZeRniLnj9ZoxYp6kFArekgJZc/lonW93HzB6JLNzTTVfPb+veyZaeNI2+RtRookURh9qKh1HUmqDK6UeNJQD1P+9Gvb+cGzhxBSMFEPaWQWLBXfxRg7drZzssmlG/q5YmO/Vd2n9sH9qo2DXLfl2P/+Vsq4b07Oc0le0+acDMtZ054Ox6uHL17fR1/ZY6RWYKIRo42mr+Rz9qpa1wanw907pvj6tgm0gf6K3/W8nGiEzLQi1vXbjIZzVtWYbEQkSjPbShmpFdk4VMZzJA/umqFpDELYqZN1A2W2DFd5cPcsUarR2tr1pJn1UMejNk6tBLARpjasKC2gjWGmlRAkKrMTy8/JM4Uwmp//7m389jc+TSmNaPgl3nfjL/FPF95k/xJzVhSdvxHPAc9xkFLgZOdXog09vt206Zz/T4/XeXqiiRQ2y8Fz7Pk7XCugtG1uzocJtYLHuatqNKNDnt0/9NJV3P7YBPXACo1cKekre8wHCelRHjgdR+AASWrYO9umESY4QlDOrqO+sU1b13HoKdq61k4NBFQKLkPVAmv7iky3YgYq/qLNp4VIKbj5glUcmA8WBZQtzG842mdzcs40p101hGFIsWgbVJVKZUnFQk7OQhY2AS5a20ucGiYb9mI5UPGZacXsmGzRlyWjH2/XqxWnhKmi7C/9nuMpG7Q23LNjis98Zw+tKGXzUIU1hZNTkrVi6x85247xHJus60g7ahKlCpWNb6XK4HsO9TAlzVLnfU+SRqkNPTKQKGsIrzL/gk5Agyft+Nexnh21gkArhqo+Zd/hwFyIzEZLlF4qGzTnuaAnbHLbp3+d3qhFyyvy/le+jb+/+Oa8uD0FjheOsNyUPEnJc/BdSTmzH4iVIkrszJfrQNH1MiWsw0hPif6yx1PjTTYMlvipS9d21bRLFYLbJxp89v59fOnRMaJUMd9OmGzEOFIQJTZIpROIprS9jthARNvIjVNFrDTbD7YIEsVVGweoFF18x46+fm/vHJPNiMcP1rl606D1H2wnrOsvc/MFxy9Ol2tTLCfn+UBe0+acLMtd03aOuX8uoBElNMOUasGlVjz6faTDidTDRc/h56/bSK3gHdWfXGvDHU+MEyQ2QDfVhjDRJKkmiFPmg5SD8wFF12Gw5FHwbYDYVD1ix2SD/XNtqgWXVqyoFlx6Sj4XrOlh/UCZepAw0QhxhPXNXJioYLJR6iiredNMuLBvtt1t7vqO7IYedcir2+eWd9zzj9x6198BcPdZF/HuH3kn+3tHzvCqco5GyZOkGgpZPRultoYc7bWhhCM9Bc4brVEuuBycC5huxpQ8l1U9LvUgQUpBteASJIo41UgB/WWfTUMV6mFClOpu/kHBdXh6okEzSlHGUJDguw5FT9OMFpuWOALKvoM21h9bA1PNhDBR3Y0vsLWm70qCRDNc86kHCW+8agOTjZiD8wE2zkEsymA4GltHarz5+o3HzG/IyVmJnFLTVmvNBz7wAT75yU8yPj7OU089xebNm/kf/+N/sHHjRt761rcu9zpzXkAsbAJIKdk6UqUZpV2/xXLBNlm/v3+edf3l4+56VXyXontqHjXbJxp8+dExvvjIQaZbMf0ljzg1bB2pMlDxT1hJNtWI2DMTcGAuAOwIcyedV2XhX51bVStWeI441HzSVs3ne5Io0YdUBh1EFogkBXKBf+3hiM57jd09nW8nKG2oljxE9n1zzgz1YpVPX/4TXLP3Ef7rj76TvX2jZ3pJz1ueK0cPAbiO4GVbB1nbV2b7ZJOXjNYwwJNjDVZVC3x37xy+a4tgzxHMthN2TDa54qx+1vQVmWrE1AreUZuZHXXW7ukWSh86txOlrfdXx8YEe51IsUlmJgskJAsSDBKdjbwJJpoRVwyWuw3WSzf08/iBeSbqEU8crNNf9k+qOD3dTbGcnJVOXtPmnA7LXdN2pr4e2jvLnpk2QawoeQ4bBstcur7/mD7iFd+l4EjG6wG+6+A7cpGvbqcePtZ9SWvDA7tneGq8iRCCfXMBiTKZulWhzKGQrzjVfP2pSQqupFb06Kv4RIm1L5huxWht8Mo+L1ndw4ZBm9Y+G1gvXFjceD38vt6xJlIGZtoJniPoK/kYDHFmMdZR5uZN2+eWz1zyw7z++7fzqat+mv9z2asx4tmMfcs5HQQQK22vRZm1VpRqpBBct3WIm186ypMHG2yfaPDYgTr75gJ8R3Llxn7W9pc5MNfm0f112pH1jK0VPbaOVPnpy9ZQ8Y/c+PnqY2M8vHeOqcxDez5I8F3VfSZeiJ8JlLQ5dB5XPEkrSgkShesIvCxg1xGQGIMyton70rW9vPzs4VMKxz1WfkNOzkrllJq273//+/nrv/5r/vAP/5Bf/MVf7L5+wQUX8OEPfzgvcHOOyeFNgIGKzyXr+9g+0WS2HXcLw01DFf6fa846bmPhVD1qOgELOyab1MOU/rJHwXWYbIQ0ozQbZ/OPqyTbPtHgi48cJExSUg1C2HEQpQ2JOlSUdopKm6JrcAQYI2jFdvzLdSSuFCR6cSGbdq0SzDGLUsOhm16QmQf5jm08tSJ11GZvzrOAMbzmsf/gsVVb2Da8EYCPXP8GPizemBe3ZwgXONG2osAWkyM9Rc4b7ckUpSUeO1DHdSRnj9SIlcZxBD3ZpkicaqQU7JtpM1T16SvZULJjKfw76qwtQxWeHGsQJdYKJdVmkf9Xt1GdmV0LQVep7wqBKwXKGHrLHjOtmEaY0lOyG1gDFZ+rNw/wxME6r79yPRes6T2p4vR0NsVycp4P5DVtzumwnDVtZyNvz3SbiUaIUoZa0SVKrNo0SvUxp79aUcq+2YBd0y3Kmap2oFJg60iV/vLxlb6dhvGDu2d44mCdZphggN6iR6o0qbJFZpaVi5sFjUWp/W/DQBm34oMxTDSjrOFsU983Dtr6eXw+JEj0cZushiwINNvATJUhiBM0NixJSvCkIExNbvP1LDPcnOEnHr+Tv7jqpwGYqvRzwy/9OYlzZE2Qs3JwBFngn0Bpa73VX/bwnAKeK/jhC0Z5xTkjrB8oMRfEPLxvjjjVVH2H3TNtEm0YrhZ41fkjPHKgzqahCm++fhPr+8tHnRy77ZGDNKMUhLUVa0Qp7XjpM9RkIieTPaAWPOuxPRsk1hM7VrhFa6Fim7XQDlN6M5XvUhkMJ8rpfDYn50xwSk9Zf/M3f8OnPvUpbrzxRt7+9rd3X7/44ot58sknl21xOS9MlmoCDFR8rsz8FmfbMUGsePP1mzgr25k/FqfiUbMwYCFONbOtmCC2irn+sk8Qp+yYbNJf7j+mkmxh42W4WmC6lXSN3gVHNlm9TBWgTafoNUhjU+lTpQ4ZDy1gYdP3eBgyg/nsQ0pDkiriNFfZPlcMtWb5/a98jFc9/W0eG9nMT/3nPyZxPJR0zvTSXlQIoOgJHGk9aJXSJ+Sp4Ah7TSkXXC7d0M9ckLBjosVUK2K6GSGFIE41q/uKuFLSCBMaYUo9TAhjuzkyF8T0lnw7LtqIYAlh9UJ1ltaGVBnCROFIGw6TKk16WNKuzFT3Bvs/OvuDljwHoyGIUxJliFIFHHqYChNNf7nABWt6T7pIzYMbcl7o5DVtzumwXDVtp56cbkakWqO0YbDqI4QdTZ5pxaRKM92Ml5z+uuOJcT5yx9McmA8IEmUbpr5DO1LMtCJGasUjwsYW0m0Yz7SZaUW0okNp77NBApl1gcSqXzup856bWX4pxb4srKzkOfQUPaJEEySKsfmAPTMFnplqsX2ieUKqWIMVLXSm0uxrgr6ShwTCVBGmuRrhWcUYfuKJb/K+2z9BX9jkYM8wXzzvZQB5w/Z5gBTQXykwXPVphIqi51D2JfUgoRFp/vE7e7nrqSkmmhGNMEEIwWDFYz6wz8/bJ5oMVn1Ge0qs7i1QD1KkEEtePzrXryhRrOsvsW28SaKtwlZI0z2XF56xsdJWtS8ElYLDmt4iiTb0lqwAIUw1hcwqJUwUjhAIAdduHmB9f95wzXlxcUpN2/3797N169YjXtdakyTJaS8q54XN0ZoAQghqRZexesjF6/tO6oK8daTGm67dyOce3MuOySbaQF/JO+oY8MKAhYrvUHAlQghasSJWEQOVQ4o1ITiqkqzTeKkVXXYpw5q+IhP1yPoFLWgQSWwjt1J0aUcpC+vMsufQihXqOHPfnWL5eH2nhYpaY6Ad21HrnGefH3nyLj7w1Y8zENSJpctt570MnStrzwgG21Ss+g5F32H/bHDcz3TGLQerPldtHKS35PK9vXMEsbIp2yWPVBvG5gOrJMCwdyYgVZokG/ES2CZpqmKkFHzxkYOM9haPuAYtVGfNtKLuFo/B4AiJ7zoIpTHmkOq24/cnBQhpFbYlV/LQ3jnacUo7VnhS8MTBBo6UDFT8026s5sENOS908po253RYrpp2UT053aZa9BYdq1p0u37kh09/PTXW4CN3PM1YPWRtb4lUa6aaCa3I+k2GqWJVT5E3Xbe00rfTcNkz02a2FdHIfCxtW84Qq0PlqVpQY8ZKE+tDqllHClxHUA8TZlqRVcQKmGsnTDYiNKD0yRWkQtiUewDPlawbKPHEgZQk1yI8qwy05/lfX/04r952NwDfH93K04Prz/CqchbSqVk7teFCXAk9JY/hqo8rJUXPMF4P6Cm6CCHZ0F9mdW+Bu3ZMEcSKi9b1kihNPVCk2lDyHMJU2zBA1aYeJAxW/aNOj3WuX2v6SgxWfJ4ab5Io032+ltqgNfiuIEy0XbMBI6DgCtb0FTlnVY1nplokSnfrzDBWBNhrS61sn+vfePWGvObMedFxSk3b888/n29961ucddZZi17/3Oc+x6WXXrosC8t54fJsNAG2TzS4/fFxJhtRZj0gGa4Vuen8kSMK1EUBC/0lpBCUQ5dWnFJ0Je1EMdPKblhJylyQHrXh0Wm81AouQZxSLXqs65NMtyKrTNCdPzM4wnrzuI4kzopWg22+9JU8ZttJ5lh59N7tiZa6nWN0VXnHOW7O6dEbNHjf7Z/kJ5+4E4DHRzZx66tv5cmRTWd4ZS92BO1EEaW6W9wKc8gHr7MRIgR4rmBNb5FqwePlZw8hhODu7dPMtRP6yy6tyD70Akw0ItpxSj1MiVPdtS5xRHZsbUjQlFzJTGtpVdRCdVac2gK2UvBQ2pBqjc7k8n1ln3qYkKQG1xXd71X2JP3VAs0wYbYdM1LzrWpBCKabEd/dM8u5q6oEiT7txmoe3JDzQiavaXNOh+WqaTv1ZE9mReAVFz+ieY71enSkoB2n3eaJ1obPPbiXyUbEaE+BgudQwKHsu0SJz0w7obfkMVD2KHlLP/btnwvYPtGgHaWEiaan5DMXpKTCoI3B1ZokKyA790wBVoCwQEHXaSClShMraxXWX/YYb0Q0olPrsjrCzq2lBhpBwiP76oSpyuvZZ5FXPXUvH/jKxxhuz5FIhz+77g18/JqfIXVyG6QzjRSHxDmdqaulnu8qvkuqDM9MtfEd0RUXVIse/SWPs1fVbDMVW7senA9pR4pEaQpZ6FesFEGiSFLNfJDQiBIOzAVLesAuFCIYA31lj5mWIcmsvIQQSGmfre05DQVH0Fv2u6GFz0y12TRUYaoRY4zNgyl6DuWCw7q+MtdtGeTmC0bzmjPnRckpXX3f+9738qY3vYn9+/ejteaf//mf2bZtG3/zN3/DF77wheVeY84LkOVsAnRGumZaMat7i6ztL9OOU/bOtvnre3Yf4f21fy7g4HxIpeCSarsLOFDxaMcps+0EYwztKKXpSr69c4aXZmvq3JwWpvrunGwyVQ95ohkz2YiYacW4UuBIgZfdWTVZgSusd4824IosfMwRDFcLIGAuSKyS9jQq0Y4Sd/H4SV7aPpucNXuAf/q79zDSmiUVko9f8zP82fVvyEfHVgDWU9o2QJWGgifACMq+k3lAW78uKWG+nfLWl23h8YN1Ht5XZ7Yds3u6BcBsO6ZScHlprUhvyaOZjZvW2zGOtF5hHdcAR9rNGQFMtWIuWNuzpCf2QnVWX8nFlQINVAsO2jgEiaLiOwxUfNqxolyWXL6hj1gZ9s22wUCQKJTSOEKQKBjtKVHx7WenmxFPGnj1haMnXeR2rnELAxry4IacFyp5TZtzuixHTdvZyFNa4zoyU6kdur4mSuNkeQkLp786CjffFfjuIRsmIQRF32VQCtqxYj5MjqqSa8Ups0FCM0qpFl0wtknsOYY4USzUmzsSXEfaejbzk+1MgDjC+pwrY5DS1rwz7QRzGmFhphM+BCQaTKpyBcKzyG99/S95+3f+GYBtQxu49dW38tjokZMIOZbnUhAjOdIq72hDmq6E4Z4ijTClGaa0Eo0UUCu63cyWqWZEqg09ZY+5dmxrZWNoRina2A0TkX2XVqxox4pPfGMH6/vLbBmuLgpFXCREUJqS57B5qMJ0MyZIFToL5fZdScl37DWs4NKKUrvR5DnMtGKmmjGXbejFcwUbByv8yIWj9JQ8agUvrzlzXtScUtP2J3/yJ/n3f/933ve+91GpVHjve9/LZZddxr//+7/zQz/0Q8u9xpwXKMvRBFjoKXv2SLU7SlYrelQLLk9PNI9QubVi68kzXC0w1YzwKwvG14UtELUBV0pcZ/FaDk/1nWlGViGbjYCBTXDXWZGZHTKTIJgsFV7jugKpobfs4UhBI7Sm7QVHECXmqDfh4/48TuEzOafH3t5V7O5bTb1Q4Tde/S4eXnPumV5STobrCDt+ldnZam2Qwm6qxKmm6Nmi0RhDy1Gs6SviOoL7d81kxawtPH3PxXcEz0y1uGR9H5es7+M7z0xzUEPRETgO+I6k4Eo8R2aNXEMzSphpxyTKsGOyuej6tlCdtW+2TaXg0o4V7cQ+kBZc0W0QO9J6Gq4fqNBT8jhrsMxjB+pMTTTtw7CA4VqBl67ppb/sHfJRTBQ/fvGabmr3idC5xu2YbBKmiqLrLCrO8+CGnBcaeU2bsxycbk3b2ch7ZP8c/WWPyUaEX7GetsYYmmHKcK1AI0y5aN2h6a9WnKKMtfGyKrnF/vmeI4nTBCnkkjZfWhvqQUIQpTSjhFrRRQpByXPsBJrvEqkElRW1xthgXFcKtDSYzPfWdSRh2vFUtzVxclgg76nQsROTHPK5zXn2uPesi/nF+/+VT131Gj70sv9E7OYChGPxXDRsBVD2JWGiT1jYY4TAdxzW9tmGbDyrrc1X0aW/bP9O/Wz6U2ADcH1XEiaqq5LX2CZuOxHWy1ZAlCh6Sx6PHphfFIq4UIiwqlbAdSSe67BhsEyUamZaMb0lL9t8smGCm4cqbJ9sMtOKqRZdygWX8XrIIwcE6/rL/Py1xw8jPx5LiRDyxm/O85FTnnP4gR/4AW6//fblXEvOi5DTTW9cGOazMCAHyBLfi0eo3Cq+S8lzKPUVacUp062YdpRijKG3aBW3ANdsGuDc0RrbJ1t89bFxtIa/vvdQqm8QKZqRHdESHBq37ow1dzBYdUDJk1QLLmGsiFNj7RKkVf2t6StlwRFJ93idz+asPK7a+ygPj55N5BXQ0uFXf+q3qBcqRF7hTC8tZwEdz6wOiQLPsQ+oCHt+7ZttW6VB0ePup6e4Z+cM80FCxXfQGvrLHoNV+/c604rZMdnkirP62TBQYtd0i4GqTzNU+K5V2B76XppEGZ440EAIwV/e9Qzf2DbBjS9ZxfVbhpBSdNVZX35kjMlGTBCneFLacDHfAWEtG8Cm6lYL9mF8oFLgvNU1ppohWgtGegpcv2UQKe337yl5lAsOu6Zatgl8ghw+tVD2S7Tj9IjiPCfnhUZe0+YsB6ebZt7ZyGtlm3XTzZiCJ4kShZs1WAari+0WKr5Lf8mjESTMBwl+RS6qh+NUE6earSPVI2y+Opt02ycaTDQippsxUWroz0aWo1QRJgqlDU42lp0aEFkIQ8cWCCGo+k4mTDDZvdCw3INenqTr8Z6zPFSiNudN7uLBdecDcOfmy3nFL32KvX1LJKjmnBF8VyyyRDiauteRVqTguQKlDNOtiKFqgSi16vzBisdckNIIU6teLbr0l30OzAW4QpAIuyFjDESp7n4v38EKHIBWpDAYVtUKbJ9s8tkH9vLuV52H68ru9WusHlLxHebaCdWiQytS9JU9zhqs8MTBeVIlWNVbZP1AmWrRZcdEKxM4KMJEs3moyn+6ZsNp15vHEyHk5DyfOC1zmgceeIAnnngCsJ5gl19++bIsKifnRFnoobMUJd9hvB4uGglbuBt48bpeHj/YYLIRWYVtkuJIycbBMuet7uk2fp8ebzDXjrupvmkWvqCMwZcClY1fH0sh24gVzSxd3ibuClqxZtd0i55WTG/JpRmlCwKJclYapTjkPXd+ml/47hf431f8JO+/8RcBmKwOnOGVPf841saEwKp0TueBr3MMzWL/r1gZCo5VrjpSEMaKMNVobbjvmWmiRDHaU0QArajNZDOmXHAp+y7Vok3wrgcJUWroKXk4CKoFGyboSqt6jVOdBZWBMvb77Z9t89R4gzu3TXLDuSP8XFaQbh2p8as3VLl4Qx9//509NMOEkZpV/CptaIQpgzXbNN4+2er6JSapIVVQK7m8dE1vt2HbIYjVUQMUl+JUphZycl5I5DVtzplmoc3CQ3tn2TvTphGmlH2H9QNlLtvQf4TdQse+ZqoVE6XKqtYKLtpAmKTMtBPW9pV47eVrF127F27SjfYU2ThY5sB8wHQzZj5IKPsOXmbH0LmH+o7AdyWGrLmjDEJaNW/Jcyh5DkII+ksuB+bCZf3Z2GmZZT3ki56r9zzCB7/4YfqCBj/81o9yoGcEIG/YriAcAcZ6FeBIgcnOR5l1boueAASOlJR9QTPSVAsusbI+tEXXYU1fMctOiYiVYaYdUyva0MQtwxVruQW0Y0WiDL4ruuFmyhiEgFQbakWXRGke2jNHlGqCJGX3dBuM4PVXrjvi+jXZjJhsKFb3FTl3VY1EGYJYUyt5bBm2deZApUD/Rn/RhNibr994UhNiS5GLEHJeaJxS03bfvn288Y1v5O6776avrw+Aubk5rrvuOv7hH/6BdevWLecac3KOSsV3KTiS8XqA7zr4juzeiGDpxsVCNcN0K6av7CGyVmmsDL5rfz3bThio+JR8h2emYuqh/f2u6TYGm4abKoMSR/ehdaUdP1HaEGUjYp60zWRHCIyAdqSohynugmI6b9iuPC7b9wR//MU/YdPsQQA8ndKdTc85aY71b9wc7w3HofM3kiiD71ivWM8h85O2Y59aG5LUBjMAhKli72yAAEq+S6Xgsra/xO7pNvvnAjb0l3EcQZgotk822TBQZrDq88CuWYSwBXQQp0gpaIYpShuKnvUlLHqCWsmjv+Iz2Yi5a8cUYap4y8s2sXWkhpSCHzh7mNW9xa4qoBGmFFyHi9b18aqXrgJY5JfoO5Itw1WA7qhb9+dnDAfnw6MGKC7FqUwt5OS8EMhr2pyVxEKbhUaU0AxTqgWXWnFpT8eFNS3AdDNi72w7U8hCwZUMVX3G6xFbhzOv2QWbdIMVj+/unmXndKs7mZIom+3gu05mfSDwHcm6/iK9JZ841bTilL0z7azJA7Wig+86NKOUA/NhN7gsZ+VRSCLe/c2/4c0P/BsSw76eEQZb892mbc6ZY6GKduEUJ9rgO5KSJxYJkYQQFFxJb9mnv+wz247pKboYA7NBwgXreqkVXB49MM9sOyFRhof2zDI2H7C2r0SQaC5a18vYfEgzTBGZrUnBkfiupB0rUmVwpK2nZ9sJQkB/xadSKDDdjHn84Dx/dXfabYR2rl9PjNV54JkZJhuRnQI4St0qhKBWdBmrh1y8ro91/adXY+YihJwXIqfUtH3b295GkiQ88cQTnHuu9W/ctm0bb37zm3nb297Gl7/85WVdZE7O0QhixVTTjiyXfQfXkfSXfbaOVOkve0dtXHR2Az9z3x6++tg4UWq9fAYqHrWiRytK+d7eOS5Z34fnCKSQpFoTJJq5trVTSJXd8jfHKEylENSKHqnWxK0kC2sQDFV8mpFiPkytD62xO6leFj6Rs3IopDHv+tbf8kvf+RckhgO1Id79I7/OXZvyVPHlwLNuAMSHKWg0px7w4AjbS1caotTgZJsnUghW9xVJlSFIFM0wtaohA54jSJQh1ZqD8yFr+0qUfZc1fUWmmjGNMEUZjdLw0jW9/MwV69Aa/n/BE+yYaHa9wIJEkWqD5whKnovvSqvczYrG/opHGCv2zwVHFI3H80Q8/GtBkvLX9+w+rcTyDqcytZCT80Igr2lzVgqH+y+et6rnhK7hC2vavTPtrAlimyKuFDx2oM57//XR7pRHwXXYMdmk5Em+vXOafXMhShs7BeZYf0ul7f05Sq3CruxJmmFqU99dyUwrJu2EcBo4MB9lijydWxisYC45sI0/vu1DbJnZB8BnLr6Z37/hrTQL+WbsmWDh1NnhNa/BBld39tGlFBjImrd2g6RacBntLbF5qMK6/jIP7p5lomHDtiu+i0Tw8L552nFKreThAJ4U7J5uM1GPeOV5I9xw3gh/dfczjPYV7YZNknatTowxeK71s51sxrhSMlIrIqUgShVFz2HrcJXxRrSopl0/UGb9QJmbzlv1rNWtRyMXIeS8EDmlpu2dd97JPffc0y1uAc4991z+7M/+jB/4gR9YtsXlPDs8m6bcz6Xh9/aJBn997y4Q0FNySVKbjDlRD5lpRYzUimwYLB/1BrB5qMpA2WfzcIWBisd8kDBSKyClxBjDTCtm+0SD3pLHQMXjoT1zNKMG002bsCnEoYbt0ZpLxtg0+XZ8yBsoVoaJRowUdtSs6DloA/XABpoVHIhO3IYy51nk7MndfOzz/y/nTO8B4J8uuIn/dePbqBerZ3hlLxyMAd+TpIcFLCz0dnayc+1EA/pSA8JYpTvCKvLbcYorbYO2t+ThORLfsT7TB+atclUbTclzaUaK6VbEWq9kN12U4fw1Pcy0Ys5f08Nvvupcds20uP3xcaQQFDxJnGp6ih6uI5hqxpQ8STOyKdoH5kIGKh4l38VzJE2T0l/2lywaj+WJuNTXTjexvMPC5N/Ow/5CTtZuISfn+UJe0+YsxXMdYHO6/oudmvaswTLVgsP2iRapsqPIrhSLpjxece4IQaI4MBswNh91g8UEdIUDGggSW7sGsSJJNVGqmWzG3VrWETYNfsNAmShVVmGrDC6Qb++tMIzh1m/9Lbd8+59wjGa8OsB7fvi/8I0tV5zplb2gOBGxwcJG7eFN2sNJszf50j53KmUo+g5r+0rsm20jpaQZpWyfbFEtemwZqVAPY8bqIWv7iuyfa1EPElwJgxWfi9f14TmSKLXigYFKgWaU8vR4E88RrOm3QoUoUWhtSLUN/wsThRRQKcDB+ZD+skuQaEZ6ivSUPKQUJ1zTLlfdejRyEULOC5FTevpav349SZIc8bpSijVr1pz2onKePZ5NU+7n0vB74ejDpev7mG3HXSNzIQyNULGqB9503dGTJ/fPBeycarF1pEqiNN/bO8dsO6FatI0V35Xsnm6zfqDEM1MtxushidKkWWdpYel+9Bu0IUzUojCgLLOBILUJmiUhMdoq+ISwPpU5K4Nmocxoc5rJSh+/ffM7+NrZV5/pJb3gSA04C7q1IvtPQ6ZMp+vf1TnnTkSM3nmLMYIgUXiuQ3/WrJ1p2cCW0d4CXqbATbRBCkF/uUCqI+pBQm/Rw3WtsmE+SDhrsMLrr1jPrplW1ytrw0CZc1fVmGiEWQEa4buSsu8SpVZx24pTYmW9cqUUuNLauMy04tMuGk83sbzDQq/vasFdpE44FbuFnJznC3lNm3M4z3WAzXL4L3Zq2s1DVbaNNVDaMFgtLDnl8eDuWdJU8cx0y3pWYsOHEm2v9537cKKMnXpRhrLv0FP0CJKUIBMiaGNDN6dbEd07tDG4LqR5P2RlIQTFNMYxmn85/xX8z5t+mflS7um5nJzodJjBZpuc6CSZwJ6bSZBS8hzW9tpasrfs4zuSVGnqQcJT4w3OX12jv1LAkZKS57JrukXRc1jVU2TLcIWBSicw2aPoOWyfaHQDeQcqPgXPwXccZloxjSghzNT1jhT4rrViaIQJ9TDJjmntB06mEbpcdevRyEUIOS9ETulf6wc/+EHe8Y538LGPfYwrrrA7dA888AC//uu/zh/90R8t6wJzlo9TKQpPVGnwXBt+Hz76sNDIPFY2KTdVVjV3NBbuxDlScMn6vm7jtxWlCCHwHMF4PSJIFGv6ikw0IhKVdovV45FqmAuSriJ3oU9Rx1y+GSYkqUZjlQs5Z5aRxjQTtUEADvYM80uv+R22DZ3FbLn3DK/shUuSHtoIkdkoWCdwpHOedf7/ZCyEk8x7pOxJNo9UKXkOk42IasFhqhXRCFNqvS5FVzIXJPSXPXpKLq4jODAf0E4UYVsxWPG5auMgN1+wis1DVT7xjR1HeGWt7i2xbzYgUZqia32s7QaNoORJgkQx3Yq7xbMrxbIVjaeTWL7wGB1fxGdzbC0nZ6WR17TPb5ZbEftc17PL5b/YqWmr2mWmHVMtLt58Wzzl0WDPTJtmmOI4AqXpjo45UthpMuzmqZOFghqjkdKxI9PZMQ32HteK0m796rnWTuxU7Y1ylg9HK/qCBtOVPgD++OU/z71nXcTXt1x5Zhf2AmU5/71L7Pl3+HF9V1Dy3ez6ZG0Rdk62GG+E7Jtt01fyuHbzIDe9xNZzf3HXTjYNVm1+y2EFtM1sSRBBwureovWddSUl32G1WyCdtRYJidL0luymT6wMrmODyioFp+tNe7KN0OWoW49GLkLIeSFySk+Lv/ALv0C73ebqq6/Gde0h0jTFdV3e8pa38Ja3vKX73pmZmeVZac5pcSpF4YkqDZbT8PtEi++lRh+EEPSU7M0j1ZpdU61j7vgdvhN3eOM3jFMe2D1LPbC7iTY109CO7chIZ1y7g5MtU5nDRl8M+I5t4HYaT1GqMVnTNsobtSsCqRW/9J1/4V13/S2//NP/vTsy9u0NF53hlb3w0dgCVQp7zhyeEN3xqE31sT2kj0aiob/ss66/RDNKqQcJEmtJUnIdpLQWB1IIYqWRUtBX8hiuFRisFvi5qzZw3ZYhpBTsnWkv6ZVlk28ThqoFq+x1JI0oJYwVZd/BkYJ6kFAteGweqjBWj1Zc0bgw+ffZGlvLyVlp5DXt85flVsQud4DNidS0y+W/2KlpG2FCqjWes/gRL04VWsNsO+ap8QYCO0Wi///s3XecnFd1+P/PvU+Zvr2oW9W9Y9ypNjEthJDQIZQkwDcEAiGEJL8QkhACCTWFAGnGpEASAiHE9GqwwdjYuDdJVpe2l6lPvb8/7szsrrSSdtVWks/79ZKtmZ2dudqdcp7znHvOrHYICuwJR938JmM/czW26rbSiEnNgcOSfFfTiO3XvGZCx9XNz+zD/pTE8bBhdCcf/spHSJTDi1/5F6TaIXB9SdieQAc7ceGoww/cna8ttFYQxQn7pht05jw29Bfozvu4jqKn6DM03eCll6/m6WcOoLWtfu0vZnEddcB7C9hEq1aQGMNZyzq4b/cU41V7widNDUGc4iiFdh0uO6ObvVMBo5WAnrwHChpRaosfsu5JlQiVIgRxOjqipO1HP/rReV/84uS12KBwMZUGxyrgPFzwPTv4na5HZBx9VFsf5jsT10r8jlUC7to52e5fu2eyhkE1t2grDHYyfBjbPpyljIOrFWFi8B1IjGK6MTthrPB060PY2CEPEsmeNNaN7+bDN3+ES/c8AsANj94mfb5OMK1Aadu8tl3JPqtvtO8oHGUIk7kHi61qoNbl+V5WqTH8bOckxYzLYIetmK+HCYmBOKnTV8pw0apOgsgwVrVTbnsKGZ66qZ8bzl825+C/GsbUo5hi4jJaCfAd2+ogTGx1f2feI0pSzhwskfcdto3WmG7EeI7Ga07fHquGJ23QeLy3rQlxspGY9tR0PCpij+UAm4UmlI+k/+J8yeBWTPuTbWM4ShElKRnXAaAWxOyerKMU7NtWpx6m7ZOksztytXrJKzPThshujbafyWGSopVq387RNuFjWxgpktQQxinKgNMcaCZOLJ0mvP7OL/HOW/6FTBIxlSmwcWwnj/avXeqlPaHYIbu29dbsmPVQOzUPV52eGqhGhlwYc/HqLgDu3DbBeDWgHMTEqeELd+0mjFPOW9HJ8mb7gkNVnG4YKDIyHZD1NBev7mLzcIWJmm3dFSUpxYxDzncZ7MjRW8zys52TVMOYfMYlSmImarZ/7skW00oRgjjdLCppOz09DcCLXvSiQ96uo6PjyFckjovFBIWLrTQ4Fg2/Dxd8P/PsAR7eW24HvxlHM1oJGW32tD2SrQ8HOxO3d7LOHdsmCJOUQsahHqcEUUKS2m1jGU9jQrtdxDQrDupR0uy/qdqDzFxtt6M1ohTPURQzLlnPpRrETNbCw//SxHGnTMprfvp/vOv7N5GLA6b9PH96/Rv4/PnXLfXSTgka0M1qmsVS2Eqc1pTp2IBODBnP9nwNk5Q0tf30EkNzS5bGZ6Y6vXXg6DaTu/O1g/YdO3is3Ij57iPDYAyNOLXbPbFV+ZO1gB3jmgtXdeK5io0DRV52+RquaVbXzjZSDtg+VufRoYp9bEfTnfdZ1pHFdTT1MMHRmv5ihvV9Bdb11tkyWmGyFhInKa6jT/qg8XhuWxPiZCEx7anrWFfEthyrATaLSSgvtv/ioZLBN5xvJ7XvnWwwUQ3pL9khQ3smG3b9nqYe2g9dp9kwPjUckFRqdixCA65jCxbi5g4zrRVamTkVt0nSHM6L/Uz3HUUYG6myPcHWTOzlQ1/5KJfvehCA7617Eu96zlsYKvUt8cpOHUfS1mP/7/FdKPoejoYgSpluTpduxaxazT+bYaGPW23E3LNzEt91COKEMEqYbsQkxvCtB/fxg0dHOHdFB9du7Ofs5aVDVpz+8qWr+eaDQ9y/Z4pNA0WevLabciNmvBZy365JktSwvDNHqdlupZXYHS43aEQp9SjholVdJ2VMK0UI4nSyqKRtV1fXgqoRkkQmKZ1sFhMULrbS4Ggbfh8u+L57xyR//e3HWN6ZY0XXTPA7WgnZO9UAJtk0UDzs1of5KhP2PxO3b6rOttEarqPYOFBi20iVarMiz9UQpilxYCsN0nRWhV9q+3i5jq39Mxg8RzPYmWWsEmKaPYHCJKIexu1taBLQLp1VU0N88Csf46od9wHwgzMu5l3PfSt7OgaWeGWnjhQw6ZE9lx1tWx60+nY5qjl0LIVGmpJxFPXEtE+MYCBN7Wuv6Dss78qyZ7JOI0ptD1zmX0SaKtsz2hgagb1t63aK1lTclF0TNRTwi5esOujW2s3DZb563z7iJCVJUvpKGeIURsoNyo2IjKvZN91gfV+BUta+33XlPc5bXmLraJVzlnXw+mvXsao7L0GjEEtMYtpT17GsiJ3tWAywWWxCeTH9FxeSDH79tWvJeprvPjzMjvEaQfPMaE/OZU85wAA5V+E4ulkVaz9nk1nJWw3NwgOF0pogTnG1IogNYZLa7d3GVv65jsJ1FL7rEsR2oJ/MZzjBjOFVP/sqf/DdfyIfBVT8HH/2jF/lcxfdsLhBAE9QatZfjqQFl1K2Oh1m4thaaGejOKpZkT6ruEFrhUmNTd6m9rUGM0UM82kNLlPKnlTZN93Ab+7gqkcJqYGM4+C5ikaU8thwBUerAwqf5qs41Zo5id18xgE8XEcTJQnr+/Lt96Wegs9lZ3Rx7+4p1vcVed01a0/qmFaKEMTpYtHtET7/+c/T09NzPNYijqPFBIWPDpcXVWlwtA2/DxV8g/3gGykHXLK6qx1El7Iel6zpgh2TAExUQ4am7Vaw81d0cuHqTuLUsGOsigEeGSpzx9Yxto/VqMUJBc/lgpWdPPuCZXPOxP3gsREeG9qGBnaM15hsxARRglK29UGrj63r2g+C1hWdORfX0QSJbeDlakU1ShieDsh6DqGxlbquYwc87N8PVxK4J955+7Zw1Y77qHkZ/vwZv8q/XvwcCW4XwdMzPZsXc3zWrsxJIZ51XasXdIIhSW3lDthqIKXAUa22JC7nLC/x5mds5L/v2sXX7x9qV7rP91go2q1KWuvVyiaJPceefEHZKvkwNjz3wmWs6yvOuZ80NeycqPGvP97O7skaT17Xzb27ppisRRSzLl15j5FyCApKGftesGuixu7JOqOVkEoQk/cczlsx0zNXCLH0JKY9NR1pRezhesweiwE2C00o75yooZXdrXbh6k52Tx66/yKwoGTwm562gT983rmctazEZ27bxpaRKkoZhispcWx3gWmtUNjdYFGS4jj2s6mVWHI1zevAxIlNMs0aCtr6HG39TO3W7MP+2sRx4qYJL7vn6+SjgB+tuYB3Pvdt7OocXOplnfQcBWt6sjTClKFyiO/amPBITjr4riKK7ZCuFPu6UMoQm7lzGpSyLbtonvCwBQyKjKOYbswfy8LM7Ie87xDFKXFiSNKExBi85gRf+54FrtZUw5hqEDNWCXlkX5k3PnU9e6dtgUEliClmXTKuQxzbY+enndnPndsmGCk32sfTT9nYx1A5YKwa4bvOnPelVd15XnnlGtb0Fo7gJy+EWKxFJ22vueYaBgakEu1Us5im3IutNDjaht+HCr7LjZhyENsPw/2aACml2DRYZKIa8vIr1tCR8xgtB/xsxyRfvGs3oxXbMD2IUqYaEUFzOFAh4+K7msfHqjw8VOZt129i40CJIE744eZRpoOIwY4MGddOyS03IuLm1i/7YWu3iXlOc3BRnFIOYzuMwXHsGVOjbO+hMEFj6CrY4URhnBAnByZoJWF7Yug0IdW2x9vXz7qav3jaa7j5rGvZ0b18iVd26onTWdUJi7T/SYo5WzJn9dGzf2lOsTYGDdSjmEoj5vIzerhyXS/V4C5u3TxKNZxbDecoe/JEa0Ujmnm0VjsF19F2SraCJLVVRJO1gMdHq3OStq2tqPfumuS+PVPkPIcoNqzrKzBaDhmvhfb7HUXW1fzK1Wt5ZG+Z7z4yTD1KKGRc1vUVWNGZZe9Ugxtv3XbMp48LIY6MxLSnpiOpiF1Ij9ljMcBmIQnlzcMVbrz1cabrcXstXTmP5Z1ZJmvRvNVwO8aq3Lt7krzntIf/tBK3+1cXB3HCvbum6Mp79BV9SjmPcj2ybRIMxIlBazt0rNV3tlUpqLBJWTv407YootmXs12Jq+zwMTAEkZnTV953mNN7XhwnxqBNSqodYsflt5/3dq7efi83Pen5mPYkOXEwBd9hQ38BhWGsUgMgjs28A8AORQHr+wv05l3u3jlFGNvjxfnyvlpB0bfHIPUoIUoMxYyLo20P6v3nNMx+jFaVbc53mrMfbGzcW/CpBjFaaeI0RTXnrzhaMVGL2DhQYvNwhb3TDYI44bsPj7TfA8M4JYhSMp7GdzUZR9NfynDZuh7OWdbByq4cW0cr0hdWiJPAEQ0iE6emhTblPpJKg6Np+H2o4DtMUsLmWUDfOTAIsdUUKR05D1crvnr/PsarITlPM1YNqDYixqoxQZKQcTVJamhEMYVMhigx3LNzks/evoPff845fP3+IapBTHfOQyubkO3K+YxVQirNfkStKfZdeZ+unMe+qQbJrLOo4TzbKMthSiWs42hbLSjFCEvAGH7hwe/xW7d+lpe88i8YLXQD8IkrX7zECzt1zT6AWyin+Vai1EzS1zD/lNyW1IDfPFkSNSc4PDpc5v/70v1ctrabx4bKdOY9tDLUwrTdZ08pO9gsTc2cINjT9mBz5mAXwCZ3wzhlrDLTb3r2VtSc75DzHAoZh+Fyg3IQcdGqLs5ySs3toorRSsAFqzoZKQes6c2zsitHxnXaB9fGmCPutSiEEMJabJy6mB6zRzvA5nAJ5b2TdXaO11AKNvQX22vZO9WgO+/xoktX0lfKzKkE3jxc5l9/tIP7d02R9TWe1uQzLiu6cvQXM5Sybru6uNyI+O7DI82K3BJj1cgOJMuC6wSY5iDcOE7b1bOze9NrbLuvnO/iaMAYJusRYQJ+s62R22yZsP+ZWwME0k3kuOuvTPDnX/8bHupfx0ee+moAHu1fK8PGFkgBWU/x8NA0UWyTqa6DPaGxyIM0A6RJyj27yhhjd4TNrjFqxbm6+fdqmKBnvV8lxrCslGXvZG3m9dSqZDf2tYmZSdqmqWkeowLKoJWiFqa0olzV3Dlm2/UZHK2ohTEP7Z3m+4+OtN8DG5HDT7ePM1mP6M57XLqmm6znsHOiTjUcYX1fAa1Vezfqzokaj49WAVjfV2BVt7QcEOJEkqTtE8xCmnIfaaXBkTb8PlTw7TUTKb1Fv90jcrZWNUXec/jyPXsZr4Zs7C/w0+2TBFFKZ85juBLawNQYsr5DPUqZrEWs7s4xXA750dZx7twxzpaRCuv7CoSxYaTcwC/4+K4NjOtRggZ816GYdVnTk2eqFlKPk3k/nPdnBx5BLLUHJ1xvdZI/+8bf8ZxHbwPg137yRT7wjNcv8aqemBJjq1wxC6/SVdg+X4pm3y9lq4Nu3zrKPbsmmW7EDHZkqIcJHTmH6UZke+2lEBl7UDo7Bo9S0KndImoTqTbITZK0uTbDw/umyXsOX7tvX3srarkRN6vrNT0Fh/FqyNbRKped0Y1SinIjIus5VBoxW0eqbOgvHnDAfjS9FoUQQliLiVOPZGjZ0QywOVRMm6Yp9++ZxnM0F67sRGt9wFru3TXFm562ob32Hz42wr//ZAdjFTvh3dWaqXrErsk6m4cr9BZ9lnfkWNZpd4hVgrjdnqGYcenO+4yUG3TlXLKuphalJM1+mgBK2970pvn5bNsXGdI0AeyOFM+xw0HD1CaTomYfBddp9gCV0PaEed5DP+DPvvF3dDfKXLP9Hj592QsYz3cu9bJOKQYYr84ckSUG0oR2onOx9kwFNOL0gLi2dW+Oag4ga8adrgLHcQnjhDBKGSkH5HwXpW0BQJSkdmdmamNZu9PTVtCGcWpjUe3QiGLGKgFxaq9zmu8ZYZKSGruaJDX4jubObRPt90CAh/ZOkKSGNd05JmoR28ZqXHZGN5sGige8J86utj3YLgUhxPG1qKStUmpBQxvEyW0hTbmPtNLgSBp+Hyr43jfdoL+UIT/P0IfZ1RQG2kFqJUgYr4UUsy71MCFODY6CRpwSpzEGQyOypQAdOY+pesjmYftBtCKTY+NAkUoQM16199GZdZmoRsRpSk4r+oo+QZzMJIObpC/tyeeGR27jfd/4OH21KSLt8NdXv0yqa5dQqwK2VclwuNdLxrG9bxMDvqdwlbIVQs3BftVKiDGm+ccOeyhlXGqh3XbWOqHSqlhoHVs2YoOrbSCbGINCkWDwHYcv/mw3N9+/j4yj2TlR5+xl9iC/lJ05+O0p+BSzLuPVsL1NtfVeVMy4x2T6uBDi+JKY9tS20Dj1SIeWHekAm0PFtFtGKkRxyiVrutoJ24OtJYgTvnb/Pr5y317GqiFdWZdqmFAOArxmK7MwTmhECfum6+ycqPGscwcpZmc+g5RS7Zh2sh7TW8rQmGwQp/Yz03NsMUKYpO2WB3FqCBJDUE+AA8tmZ+9mieND75QRx05XfZr3fuMT/PzDPwDggYH1/Pbzf1sStkdovjZ10REO0GvMKs81zBwPtu4tbcanHVl7/GgMNKKE1NgZDo6jeNnla3h4b5m87/D4WJUtwxXicKbHrcaQpjb+dR0oZBxqUUI1tK/dIE7xm9tBPccWJCSpoVyPWNOXZ2S60X4PnK5HTNRCilkPrfWceLaj2aZl9vvQQncpCCGOn0UlbY0xvPa1ryWTyRzydl/4wheOalHi5HA0lQZH8ljzBd8XrurixZeV+M7Dw/NWU3TnPS5c1cmDe6cZrwW2H1g9ap51dKmTYIwhSptnO7XB1Zo4SamFtses62iyrtPeztZT8Ll4dRebhytM1EISY8h6mkZk6Mi5NOKUNDWkxuAqCPfvwSmWXEejwp9885P84oPfA+Ch/rX8zvPezgODG5Z2YU9whkNPx92falb96OZ2L5QibbYiSFF4jqERwZ6pBkE0M+zPd+yQlTA1lHyHKEmbw1MMYfPxbQVDiqNttVCUgFaGPRN1Mp6DVoqRcoBSUMh49BT8OSd08hmXKEmYqIXsm56p7MrMei850unjQojjT2LaU99C4tQjHVp2tOuaL6Zd31fEGFjRNX8yuLWW1lbmXRM1gjhlsCODQrFvOiCMUxzPAWXbGNTDtN0+THFge4b9Y9pCc5CR49i2QEopegs+jShluhEtaHt4a6inODGu23w7H/ja39BfnSRWmo9f9VL+9uqXEDkHxhhi4Y51sc3BBkzb4097TRCnZD2HDs8jNSlRYgsHHh+tsrIrx97pBk/Z2Me63gIP7p1mz1SDRmgH/TXnjRElKbXAHoMa3awUNhDEBt9TOMq+tmthSpCkrO0rsHO81i6ACpOUOEnxmjtYPUdTDWLCZgV9u9VKEPHdh0YWtUtBCHF8LOqo8TWvec3xWoc4SR1ppcGROFTwfUZv/oDgd0VnFgN84a7djNdCtgxXKddj1vTkcbWdiGvSlDSdqQSIEkOapmgNWU/bKjlHc+nqLoamA+7bPcmyDpuQPWvQbiEJk5RHh8qEcUpX3sd17IdrLUgWlYASJ85v/Pi/+MUHv0eiNJ+48pf566tfTuhKcHsycJivdmd+raF9M4NRDEqD7ygyriZNDGEcoZQNOiElSVJibFLW0TaQ9VyHwY4M0/WYiVpInNhhE63vi9MUR2tWdWfpyPlEScpENbRJ2WrIlpEK3fnuOQe/w+UGjSilHiVctKqrXdmVpuaop48LIY4/iWlPD4eLU49kaNmxMF9Ma4zhY9967JBrmb2VeWVXjt2TdTKuTbS6jiKT2gRtHKd210makveznL2sxEQtQsEBn0E9BZ8nr+1muh5x/54pJqohZw6UUI4i7zkMFH0+f9fuRffzFMdfT22Kv/nfvyQfBTzau4Z3PO/t3Ld801Iv67RwrBK2s1rRHvR+FYZKENmY1LHVsI1Y0dFsWzJeDekrZujOe2weqbK8M8t15wwyPN3gwb1TjFZC+goZlDJM1eyJqCBOmz2mFRjbq9okKdpxUEoRxgnTjYjvPzLCzvE6Oc9hdU8Bv9lKoRLEuFoRp7Y3buvkT+s9sdKIj2iXghDi2FtUhHLjjTcer3UIARw8+N4/+B0pB3z1vn1M1Ox2jeWdWcr1iF0TdcqNGKVgeLrBdD2cs3UrNc0p9AYqjRitFKWMh+Nqzl5e4hsP7uPeXVP4rp2kWcq45H2Xs5Z18MyzB/i/e/fw3YdHqAaRJGxPYn9z1Us5e3gbH7v2FfxsxVlLvRwxy0JeNro5eMzMGr5gjK2i1UrjasWKjizbJ2qkxvbkqoUJlSCmEaXEadr+vqzv0Ff06cz5dOd9uvOe3QYWxBSbrRQ8R7O6J09nzgcg4zoMdmQpBzHVwPYMa7VBcLVibW+O1KScs6yD11+7jlXd+XaVwbGYPi6EOP4kpn1iOJLhusfK/jHtoU7qpWnKlpEK/aUM20YrrOzOAapdhJAY2/Inn3FoRClKK0yzB1AQJeyZssndWpQc9DNoqBxw5mCJrOewd6rBxv4iE7WQH2+bYKoRHfN/vzh64/lO3veMX2X15D4++pRXEbj+Ui9JNDnNWFUrW+16KI145nsqQYxS9oSJndeiWN6ZY7IW8aJLV3LPzim2jFQI4ga+o+ktZuktZtnQV+D7j41Qb56w8Rzbw9ZzFI7joLCzYAoZlyBOcbXDGT15OrM+j49UuWPbRLvathbanWKeVkSpoSdvCxZmvydKuy8hTh6yP1OcMlrBb5oavv3gMBO1uds1Vnbn2DleY8d4zZ7BjA5eCZsaqMcpRd8hNobvPjzM3TsmcbStSIjihCBO2NuI6S9leObZA6zsyvHQnmkMhrzvUAttk3jZIbb0rtxxLy+6/zu86zlvxShNNZPntS/5k6VelphHa4puytwhDa3+s76r8RxFPUxwNGhjqwjyzdeqMdBb9FnWlWXLWJVixiVpXtdT8AnilMlaSLkRkc94FHynXdGklKIz75P1HMqNmLW9eX62c5KMqylmXNtiwRgcpfBdzUApw97JOmPVkB3jVSpBzGglpBLE5D2H81Yo2wtwnqGMRzN9XAghxLFxMp1IO9ha9k7WuX/PNFGcUgli9k41mG7EbOgvkPccRioBRd8mZZLEtHtoKqA779OR89g3FaAVjJQDrtnQxw3nDfK/P9vLw3vL+C505zOcv6KDi1Z3UYsSbr5nL3fvnGB4OqDSiEhTiWZPBvmwzu9979N85exr+PGaCwH4t0ueu8SrEvvTQN5ThKkiSRd+MKibQ8la7za1IGFVT57+UobtY1X6Shn+39M3tIuUpusRn719B90Fn0aUMFWLSIG855CmEBo7ELvgaMLEUG7EBLEtXvAdh5/tnGSwI8tZy4r8bOcUP3hslIxrH93VyvbCdTUGuGPbOAOlLGt689LuS4iTjLzKxClnvqES49WAx0erZH0HgGqY2ImbNJNCYKfONz9UVfO6nrxPNYj42+88hkGR8zW+o+nMeWzqztNX8Nk3HfDDx0a5b9cE28druNomhKWf19LLRg3e9f2beN1PvwzAnSvP5T8v+rklXpU4HKXslGplbPLWdWwPLkeBdjRRs+fXmu4clTBhrBJQDRPyvsPq3jwb+gsMlwNyrsOFqzoZq4QMle0EXVdr1vUVqEcpuyfrKKWIEtMOUo0xVIKYwc4sxZyL6yiUVuyaaDR739ptYjnPoSvvUci4OFrzyL4ycWooZFzW9RVY0Zll71SDG2/dNu8ghhPZE1wIIcTBnUwn0vZfy+bhCjvHa3iO5pI1XRQzHpO1iB3jNR4freI6inIjZqIaYpQhjGyP9kRD1nPoKfj4rsbV4GjNdx8a5jsPDXH74+NM1iKS1FDIOGRch9FKyBfu2m23VkcJO8bq1MIY31HtAWMS2i6dy3Y9wIdu/hhrJ/fy9K13ct2vf1L61p6kDOC6Dh6K6cbB95Dp2bfXUMi4+K4DxlANEzw3ZX1fnkZkE6A5z5kTN6bGECQped9luh4Rpwbf1Whtj1mT1PbFTYxt/xckhihJcB2FcWCiFjFZjxitZDhrsMQ9uyapR9CRdXELPklqcLSNv8uNmMEOeM3VZ0i7LyFOMpK0Faec/YdKGGPYMlylHiYs68iSGsOO8RpJmpJxNbUoRWEnbzqoZjN4CKOUchgTxSnVIMF1IEldUgNjlZCJWsS1G/vI+5pvPLCPSmjbLgBSYXsSuHT3Q3zo5o+yfmIPAP928bO5+exrl3hVYqFm95o2iSFVBuVqso6mO++xsjvP+37hfLRW/HDzKD/dPkG5EeFqDSjOW9FJ1nPwXQ3NygXaU60V/SWfqVoIwEQ1pL/kE6eGSiMm5zus78uzc6JOMeMyWYtoxKk9sNUOiTFUw5haGOE4mpVdGXK+w6ruPBnXbmdTSmGMOeQghhPZE1wIIcTBnUwn0lpr2TlR48ZbH0cpuHBlJ1prjDGUMi7D0w1SA6Wcy+ruLGPVmIlq0N5BFieG2DHsm2rgaEVnzmNVd46vP7gPY2zv976iT5gYxqoBP3hslFLW5ZrmkKNto1WqQUTOdzijN890I6YSLLTjvDiWMnHIO275F37tjv9BY9hd6uf3nv0WSdiepDR2ZkIQGbQ27b6wLa1KWtdRuFqhlT32zHqaapigkgStNKWcS85zcLVm71SDFZ1Z/vdne7h/zxS1MCbvu6zpyRPGqT254tr2YHFihw96jibnO5gwIUlN+/XraDssLONqEmMI45Sh6QaOtnMczl/ZSX8pg+/YXWaV5hCyMLYDynKeTQ+dTLsUhHiik6StOOXMHipRzLjsmaqzd6pOrlllG6cGpRRJ2txCZkw7yPVduxU7TFNSYLpum8Kn2KFHjch+4CUGqhN1vvXgPrIZl9FqAM1+uHVJ2C4pP454263/xhtv/wKOSdlb7OX3nvNWvr/+SUu9NLEIs+sSPK0wGBpRSpLG9HdkuGZDH2t6C2iteEVvgZc9eQ07J2zlEcDanjw33baNbz08bPtP5zw8x1bVDpcDwomUy9d205Hz+N4jI+yaqFPIuPSXMqzozDJWjVjZmaPSiJmsR+R9x07xVQZH2QPeyXpMZ07TkXU5o694wPYwGcQghBCnjpPtRNrQdIOd4zUGS/sN+lH288XFEER2UGZXzmWyFti+la1BRlFCNYjblXKP7LMVuR05F8/1GC4HRElKpRETp4a0EfPAnin2TNbZOVGnHMRMN2Ima7aCTyptT7wL9z7Kh2/+KJvGdgLwHxc8iz+77tcoZwpLvDJxMAZIUugtaAwaTEpUjUiahQOOAs9VKBRhYij4mp6iz0Wrunho7zS1MKGYcdBaMVG1wwEHO7JsHqmyd3KUxBhaY3i3jlTRCoIoZUN/gc68x3Q9ph7ZoYVpaujKeSitqDRiMNCZ88h4tsbXVQrHc6gGMXsnGmR8zbKOLN2Fmd7IHTkb28Zpak/mzOpRezLtUhDiiUyStuKUkKamXR2R8xzW9xe4fes4lSBiuBwwXg3J+Q451yE1hrjZB8EY2yPTxIYktdenqSFJ7Rbt1oeraSZuw8SQcTW+hjA2DJVDTDlEqebUeyMB7VL7yP99mOc/8kMA/vu8Z/An17+R6WxxiVclFmP2wAYN5DN2wFea2tYF45WQZ549MOfs/dbRSjtobMQJGUezY7xGlKT4TutQs/n/Zu+SzrzPHzznHJ55zhjffmiIvVMNHGVrIC5Y2ckFqzoZvi2wVbVBQpimRHFKI7XvF6WMx/KOHAm0hzfsTwYxCCGEWIzNw2W+fv8Qd+2Y4IE903RmXTYPu3QVfBylmK5HrOjMUm7ETDUiJmoRlSDGoNDK2ARsatBaUczYz6YwNUzXQ8LYYIgZr9rBYm7zc9R1bIXeluEqvlcnSgxR88M4TKTCdimcM7yVL/zL7+CalOFCN7/37LfwnY2XL/WyxGG0UqpD5QhPN9t8aYXbDEETY4hTu8NTYQt+lnVkWdOTp5R1uW/XFLsm69TD2A4XbM5JGa+Gdg5DzsNz7ADCcj2iHCaAIu9revI+cWKI4oRKaAdqZ32HjKMZ1Yqsq20LBWPaJ4KUssfC0/WIzryHo+f/dx2sR+3JtEtBiCcqSdqKk14ruG0la7Kuw3Qj5N7dk80KBIiTlGrDUFUxCkVXzqOnkGGiFhIlKVoptLbDjhJjP8CKGYdalOJpRWJSTPMMaZIajLIVu63ckmZmgr1YWn9/xYu4bPeDvOdZb+LrZ1691MsRR6EVN7ZOsmil6MzbM/7TsyZZbx4uc+Ot2xivhizvzJL3cwxN19k71SDr2UFjjShtVhxpBjtzLOvIMFmL2Dvd4Cmb+rlmQ98BAeejw2V8V3Ppmm62jdYYqwYEsQ2Oewo+Zw0W2TcdoJWSQQxCCCGO2ubhMv/8w23snqyRcex25+FKSHWsRmoMrtakxtCd9+nKe2Q9h8GODPfsmsQ0Y9UU+0enhnqYUMy41IOYKLFxa5ykaK2b8bFN8mYcRdRsS5SEB+/BKU6ch/rX8Z2Nl9Nwff7oWW9iMtex1EsSi2Cw7fKUAqMMJlVkPE3RdTDGECYpQZQQpYbBjlY1vW2lUPAdPEcz2JFhQ2+Rbzy0DwP0lzJkXLtzNOM6+EVNPN0gSVPW9BZIDIxUArTWrOvNsqonTzHj8vhoBc/R9BZ96mFKPbIDxpzmPJcgSXEczdnLSuybDihlvUX1qD3ZdikI8UQjR5nipDZfsmb3RJU7Hp+gESeUMi5RardVR82tXWCYrEf0FDyynqYWpmjVPDParJQtZh26Cxlqk3VMcyK9PRtqJ/PuL20mesWJt2lkO+cNb+V/znsGAPcuP5OnvvGfCF3p9XUqc5qvSQz0lXyKGReneXJlz2SDsartR5umhq/fP8R4NWTTQLEdZPquQ963gXHed7l4TZEoMfiOppR1SYyZs81rvoCz1Wol6zlctrabcsP29WrdRyWI6cql9Jey7JyoySAGIYQQRyxNDf9++w7u3DaOUlALY9vCILVteRylUMrebqwSMN2IWdubZ7QSEMSpbYOgDWkz55oaCBJDUo/ae03cZpLGVaBQNv41hiA2M33ksSdNpQ7hxHLShNf+9Mt8/vzrmMqVQCl+8wXvknj2FJaCbZ+XQNaz/WvBnjzJuBrdvM1IOaA77/Ho0DTlRoyrFR05jwtWdjJdj9uvyYlaRM5zZlXJ2mKGsUrIVRt6+ZWr1vLQvmnufHyckXJAmKQkqeHSNd1MVEOS1DBYskVL9SglMqa9s3RFR4ZXXbmWrz+4T3rUCnGKkaStOGntn6wB24P24X0VosQOGdNa4Rg7QTdKEsLmDq9GnLJnKmgOJQLd7A+GMigD3XmPUta1fYLaVX5zt23PJoHtiafThF+74394xw/+BYCHBtbxSP9aAAlwTwOmeYSpFGRd3W4/UG5Etlqg2W9r92SdLSM2uJydMPWbFUqJgb1TdVZ0Z1nRmWvfph7Eh62AXdmVa0/G3TRQbPf1suubSchef+4AN922XYJcIYQQR+y2LaN89+FhjDF05X3KjQhjTLtowNGQpHZwQmwgjRJ2jlVpxHYOgzZm3jg1bl7XkXEI4tRun05sirZVrLD/t0lce2JtGNvJh2/+KBfvfZQL9j3G237+nYDEs6cLA812XZqBUhbXUThKEcYp47WQNT059kw22DVRb1bPZ9nQX6CnkGG6YYsLPNfGlmGckvGcWfdu40utbPHB6p481589OGf32PKOLHFq+OaDQ9TCmN6ijzEQxGm7DcNTzxrg2k19rOjOSo9aIU4xkrQVJ1wcp9y1c4KxakhvwefS1d247oENdmYnayZqEZuHKwxPN9gzVW9vp56KY7KupuA7TNVT9g9DDaCMrTZwHTvtPTYwUgmZrNkPsVZAe7CEbXvdspvshDljYg8fvvmjXLb7IQC+s/4yJrISSJxOWlW2nqtwtH39p2nKWDXkrMESl67uBqAaxjTihLw/t5I1ShLbA6wWkqaGn26bYE9ngw0DBbrz/oIqYBc6GVcGMQghhDgaaWr41oPD1KKE1d054sTYnWBa4yhDFDfnLhhbRKCVTeRWopngs1WY0JrHoNRM2y8NZD1NbOwVcdq8DfO395Kk7YmhTMrr7vwyv3vLTWTjkOlMgVvWXWp/IbKF77SgsK/XOIVGnOBoRd53McZQbsR05jxeedUZjJZD/umHW1nXW6QrP9OeoDvnkXWdZnsu7JAxV6OUPW6dqkV05nzW9c0Mp5tv99grrljDcDng0X1lKkFCa9ZD1nM5c1mJV1yxBq2V9KgV4hQkSVtxQn37oSE+fes2to1ViZIUz9Gs7S3w2mvWct05g3Nu20rWNCKH+3ZPUQ9jPFfZgUXGntGMUxuk1ps9g+ZjgCg1uI4i4yqI7TReXFjWkWWkHNCYJyPbSiOnzb9Lzvb4UyblVXd/hd//3o3ko4Cyn+O9z/x1/vPCZ0lwewpq/cb2f2W6aqbliOfYTZrlRsRYNaQj6/Gaq9e2T+S0WhjM7ik7Xg24Z9cUYL8/IsV1FEPTdcaqAQPNgQ8LqYBdaEJWglwhhBBHavdknb1TdYoZlzi1w4rSZpWtqxTaVTQi286rlHGpNqvjDqaV3G1xHagECWmazsSvze9X2DhWxo2dWKsm9/Ghr3yMK3feD8Atay/hXc95K3s7+pd4ZeJYchUYu5mTpFnlHsQJlUaM62pWd+fozPp0Zn36i7YKd/bOsdgYPEcxWU8wBvZONqg0YkpZ2wIwNYar1vewuvvQPWU3DpR42/Wb+Nr9+7hv9xS1MCHvO1y4sosbzp9bYCA9aoU4tUjS9jSUpuakTCx8+6Eh3v/Vhyk3InoLfrui7dHhMu//6sMAcxK3Bd8l42ge2TdNPYzpKfiEsU30pqSERmGwCdg4TQ8a3LZ62dabvW2h2QcsTpmuRzgaPA2tYgYFuNr+xaTNoLc5EVSSt8eRMfzjf7+X67bcAcBtay7knc99G7s7B5Z4YWKxmi8XgGY/vbmvm56CR1feY2g6wmAYrYR4juaswRKvuXruCZzZLQxaU7K3DFephwmDHRmYtrdzlMIoKDdiBjvgNVetXXAF7EITshLkCiHE6eNExsvVMEZr6C9mGK0EFHwHrdSc1gWtPpizY9rWIKHZ2t/THCymgCiBsNkSwXdsfJwms4aWHeS+xPFx1fZ7+Mf/fi+FqEHVy/K+Z/4q/37Rs6UA4RTQqpxNzcxrc3Zcu/9tfU8TRqntHaugFiYopegvZXC1HXbb2vU1O55VStkihJ2TuFrhO9rGysowWQ+ZbkT0FDNcuKqTZ5w9wKPD5cO+T20cKPEbUmAgxGlHkranmc3D5XbFWCNOyLoOG/qLB5xhO9HiOOXTt26j3IhY051DN7dDl7K2tcGOiTo33baNp23qb1fYrezK0V/Kcvvj4wyUMiil8Ju9L+N62N7qtX+V7OyPpdkfsCk2GRunM+0QphpxO0nb2o6mgIyjCVPTbCU/U60gCdvjSCl+svo8rt5+L+9/+mv5l0ufh1EHts0QJ56vYTHDpluvwdaxSTuBC3TlPS5b20tq4LIzPC5b10POdw7aKmX/FgbFjMtoNcB3NRO1iM68z0WrOvEcTZikhHFCFKdM1kMe3je94IBVErJCCPHEcSzi5cUkfQu+S85zyXU5VMOEShiTcRX1yBAntvIWbNK2HiUkswbkzkcxN96dnZCN9svMakBrcLRCpbZnuyRvj68HBjdQzuS5f3ADv/O8t7Oza9lSL0nMMl8S1tXNAgAMWs20I3Gxr62DvWS0UhilcDD0FnwuWt1JznMoN2J6i5k5u75mx7PLOjI8NlRhuh7hew4ru/PkPE0lSNrVun0Fj4FSli/ctXvB71MSzwpx+pGk7Wlk83CZG2/dxng1ZHlnlryfoxbG3L9nij1TdV53zcIrz46lNDV87cG9PDZcpiM7d/o6gNZ26NDjo1Xu3DHOyq58OwC+9Iwuvnb/XsqNCJpbqYtZh4ka0NxWNu+Zz2Z13+yvK5r9a2dtF2tVK7iOwjQjWFcrOvI+1SCmFsaoQ3xQi6PTXxmnq17msf4zAPiHJ/8iN591LbskuF1yrdeH7ygcrdAmbR742RYlStlq9dZryjCzVdPR4DmOrRZKDRnXDgxztKKv6JNxNZsGSwvuBzu7hcFdO8aZrkd05jwGOrJs6C/S0xxaBjBcbnDvrik+9f0tZDznpDlxJYQQ4uRwLOLlxSZ9Z+8auWhVJ1tGqqTGUAkSgjgFo/Acm1CN4rlR58GqbZPm10pZB2NUc4t1NGcGQ9aFvO/iOppGmIADjqOYqh+6/YJYJGO4evs93HbGRaAU09kiL37lX7Krc0AKEE4BCtqzT6J45nKSGpRWuEBqTPu1ZRt7zQwg68y5ZFyHrrxPmhqSFC5c1XVAnDs7nr139yQ7J2qzBpMV6c57lBsxYZIyWm7wyFCFraMVNvQXT5rjeiHEiSdJ29NEmhq+fv8Q49WQTQPFdmK0lPUoZlweG67wjQeGWN9XPKotEovdStYKam95dISJWkgYJ9TDlJ6CR27WVPec7zA83eDfb9+Bq3U7AO4r+qzszhHGKdUwoRrEOErRkXWpOYo0SSnvVwLYaoew//AFhU0mRenM1pdWEFyPZiLXODVE0412AkqrmfuRyoRjxBh+/qFb+NNvfpKxfCfPe+1fEXgZUu1IwvYkooAkNTZoxb62WideojjFOPZGyoDrajqzHklz8IICMq5DmBjyvoujFau7c7zwkpVcs7GP1d35Rb0XtVoY3Lm9m099fys9BY/lnbk5J4HGqyF3bZ+w1Q2FDAMdWQlwhRBCtB0sXi5mXAZLGTaPVPivO3fxzp87a94huTA36busI0MxdZluRPxk2xi7J2u8/tp1B3zWzN41MlYNOXtZkY0DBfZM1nl4b5lqEBOmhvqswWMGm5TVaia23T8M9RxFlNiemMWM1xxAlNLslEAh45H1nGblXkpqDF6iMMYmdONUBu0erb7qBO/7+se54bEf847nvp3/vuA6AKmuPck5yhYTpMaQphDFpn0io5h1eNIZ3dy9Y5J6lJL3NN0FnzhJGamENgbGvjZXd+cZ7MhyRm+BZ1+wjP5S5pDHyK149pbHRvjHH25l/X6DyTpyHsYYNg+XaUQJK7ty7ZkOx/q4XghxapCk7Wli92SdLSN2+vn+laxKKZZ3Ztk8XGH3ZP2It0wstqpgdlDbW/TwHY1BUQ1jgjiht+jjORpHKcqNiHqUMjwdcO6KjvbZxB3jdWphQinrccHKTqLUEMQJ9++aYqCUYc9Ug0YcEs0TcLYC3BZHQ3MwZ7s9wsHEs5qMtab0OlqRNr9JcrdHrrs2xXu/8Qme/8gPAdjVOUBPfZq9ngxmWEoOM0NKWs/51MwaZNKsdM/5LgXfZbIWEKYpcWLIeg7r+wqcv7ITUNy3e5JdE3XqYUKcpsRJSinjk/Uc7tw2wUQ1OqLKV60Vl53Rwx1rJrh/z9Scr7UC3Ml6xPq+Asua74US4AohxKnpePScnS9eHq+GbB6uMFELaUQJ28dqGGN4yZNXH/A5NTvp21vweXif/b44SXG1Yu9Ug5zn8PvPOYe90405a99/8GUQJ3TmfF5xxRmcv6qD3RN1Pn3rVh4brrUfL2kWDGgFOd/2zmy3TgAazYBVY4sbUjMTp6rmehtRQrkRY4y9VmmFVqad2BVH7jkP/5A/+8bf0VufJtQunY3KUi9JHMLsAblaKdb05KgECdO1ELQ9AdJX8HjmOYP0FjJMVCO2jFQIE0MtSljZmcV1NEPTDcI4bZ/suXJ9D8++YPmC41qtFRv6iwzMM5gM7HyG0UpIIWMreOf8G47Rcb0Q4tQhSdvTRDWMacQJeT8379dzvsNQM3g8EovdSrZ/JUOapty9Y5LJWkTW00w3YqpBTMZzwKRUw5S873LZGd04TqvfrceZg3aC7nQ9Yt90gxVdOZLUUI8TwjilFiY4WpHxFEFsSFMzp3H8bHE60wNsIUnX1u1sxaAmNRAnRvraHoVnPfZj/vxrf0t/bZJIO/ztVS/l41e9hNiRt6KjsdDn9MG4WlHwHKpRDM1ErdaKtPliap20WNGVZVlHlvFKQGygt+DTkfXJ+Zqr1ve2e1U/dVM/U7WQn+6YpBbGrOzKsnGgRCHjHXXl6/49bpd3ZtuV+tvHanTnPTbOqp4CCXCFEOJUc7xmNOwfL49XQ362c5J6GFPMehQyDmOVgAf3TnPjrdsO+JxqJX1znuaeXVP2+zIuqaMIk5RaI+Gr9+9juhGTpGbetR9s8OW3HxpiupHO2SXWkhoIm9V9Ctof/K2dY6YZoyps/1pobd02JLG9T0dr4tTgNCt301ZCGJnZsFid9TJ/+s1P8gsPfR+ABwfW8Y7nvZ2HBtYv8coEzAwS21+rOlYz08or70NHtsBELaSUdbl6Qx+9xQwAF6zqpBomDJcbTNci6kFCECeEzRYm9SjhkeEKaMU5KzoX9d60/6Dd2XFrECdUgph1fQVK2QOPkY72uF4IcWqRTMlpouC7ZF2HWhi3t1DMVg8TMq5DwV/8r3x2AnZjf4FKkDBRC/EdzYa+PPftmebffryD112zllXNLc+toHZZR4ad4zW2jlZt8Jim1Ou2N2aqDb7rUI/s9utCxmGqEdFTyLQfWynFpoEiO8ZrrOkpMFoJmKiFpClkPBuVFrNe88M5IVWGxKQkia0MbFUnpGZmi9l8Vbntx6NZkZvOHfAACtWKjqXMdtEyUcCff+Pj/NL93wHgkb41/PbzfpsHlm1c4pWdevxmzy3P0USJwdV2aEIYH9kJBa1onwjxXc1g0WeykZD3HKI0pRbEOFrTW/RwHU0p47B3yuC7muWdOUpZl9FKyM92TbFpoEjOd6iHCUPlEEcrVnbluWRNF0C7T9dgKcO+6eCIK1/3r1Yamm7QiFJKWZdL13TPeQ9pkQBXCCFODcdzRsPseLmYcdk8XKEexvQUfJRSBHFC1nPZ2F9kqHzg51Q1jKmFEfumAiZrIXlPM1IOaDRbDyRpSi1MuXXzKDect4wVmfnXvv/Jw9bA3jhN6c7ZVkOVIJ7TuiBJmzEttGNR0/yP46h24UI6qw3YYMljvJagNUzXI7SCamjmtFqQhO3iXL3tZ3z05o8wWBknVpq/u/LF/M01LyNyDjz+EkujI+syVY/nPWRztWKg5FMJEkYrATnfYaDkoxQ86YzudsIWoKdgq2jv2TnJo0MVoiS1Jz405H0HhSJKUh7aW+b9X30YgOvOGVzQGg9WhFAPE3ZP1sl7Divm2UELR3dcL4Q49cgr/TRxqLN1xhj2TjW4YGUnK7vmr8Q9lNlVBXdun2xvA0uMIW62C9g8VGHPVJ0LV3Zxw/mDxKlhtBKwZbjC9vEaUZLiOXaAUZoaDDYxGsQJ3XkPBWRdzZaRKt15f876c75DxtW88JIVlLIe5UbE/9y9h588PgqApxVaK4oZqDRiO+1TzWzvLmZc6mFMYsBVh866tioPmHUrA81/7yG/VRxC4Pr0VyZIlObvL38RH732lYSuBLeL0ZVzCeIEhSJOoR6lzcnSpj0MoVUts5hzC7MrEUoZB60dzl6W56zBElFqqIUx+6YajFVDxisB1aA5JHBNFyu68tTCmDA2TNcjdozXyLiajOuwpjdHnKas6ckzUQvZPFRhqBwQpSme1nTlXO7akR5x5ev+1UrT9YjP3r6DrOfMe3sJcIUQ4uR3vGc0zI6XB0sZJmohxaztJ2mModKIGejI0pHz0FrN2aGRpoYH90zz4J4yo9UArRTDSYpSioLvkHU1U40EYwzTjYjJekhHzlvQ2u/aOcG2sSr9RZ9yI6YcxJQyLrVmy4NW3Dxfa68E2h/m7XhAgecqqqEhSBKiICVOZ2KDVowstQiLFzsu/ZUJNves4h3Pezv3rDhrqZckZlFAOYjbcXHG1eR8W2XeiBJSA1P1mI6sy1PP6ucZZw3SlfcOGkN2531cR9GRc+1uziCmq/n+YIytts26inI94qbbtvG0Tf0H7Ye9v/mKEDKuw+Vre9jQX2TvVANjzDE9rhdCnHrk6PU0caizdXunGvQUfH7uvMEjCnCrYcxoJWCsGhJECcWsR9zs29WIEjKuppBxyHtOu5LgglWd7BirMlGLSI0h52mS1Fb0+a7GcyBKYHV3jivX9fCT7ZNoZbeplRsxHbmZhF4r2VLKeu3kju9qHh+t8MhQhUackmt+yCbG4GhF0XdIDKSh/WD1HMV4NaKxgEli+ydtAaLmNnEJbBeuENQwSlHzc6AU73rOW1k+Pcpdq85Z6qWd9Ga35mj9v6/oU27YqgAAV9uqm9ntQGbaedh2IVqB72jqC5gyks+4GBSuq9k4UKIjZyfY4rucNVii3Ii4e+ckA6UsT17b3W6FUMp6XLKmi0eHypzRW+AXLm6eXAki/vY7m2lECXdsG2esEs55/UzUQvZON3ho7/QRtyvQWrW/N00Ndzw+cVxOXAkhhDgxjveMhtnx8uaRCo0ooZBx7HbkRkzOd9jQX0ApNWeHxubhMv/24+1886FhRsoNwtjgNntuOVpRj1KMse0Isp7GGHh8pMrq7jxKqcOufawaEiUpSrtESUojspW7ti/t3GpYBfiuIk5MO4m7f3jrOwqtFGGS0GiuDeZOvW8ltUDi28Ppr0wwUuwG4Cerz+cNL/pDfrD2YgLvwJ09Yum04uZWtbmjoZh18RxNkhoyzSr7JDU89cx+3vuCC3BdfcgYcrpuK+u78z57JusUMm77eFope0zaiA09BY/HR6vctXOCy9f1LnjNB2uZsnW0wo23bjvmx/VCiFPPwk4DiVNC62zd+Ss6maxFbButMlmLuGBl51FtJct5DqOVkErDbh/zHcVkLSQ1ho6sa7dWRymdeY9NA0XGKiE337OHKE6pRQlRklIJEmpR3BxsZIgS6Mp7aKXQWtOd9wmilDhJCGdNRmglWzYOFOckWzYOlHjHDWcyUMpSDxNqYUQQpShsb07P1YSx7XdbC+3jZlzNYj7aHDXTIqGVGNPNFgsKyHnaBuziAFfsuI+v3vgW/vA7/9S+bm9HvyRsF2h2whZsgtZz7ImJVhzZU/TJefqAAy1XQ5iYWQdlZkHP+0aU0FfIMFDMYIzhjm0T/GjrGLdvHePHW8f46Y5JNHDO8o52wrZFKcWKrhwj5aB9cqWU8cg4mp/tnGRoOsDQrHbwNBlXkxjDVD3mWw8N2d65R6l1IN5T8HlsuEK5ERGnKeVGxGPDFQlwhRDiFDDTc3b+upKcbxOsR9PqphUvn7u8kyQ1jFUCGlHKQEeWi1d3tVvstIoGRsoBH/3mo/zvPXuZqIbt2C9q9oRV2LYI1SAGYxNDvqttxWxjZp2HWntP3iNOU3aN12nEKcVm5S7YhG3rU7IVh7Yc7BMtTe2gsSgxeA64aqYXbqt1Quu+xcHlwgZ//M1P8r2//3XWju9uX/+tTVdIwvYkNDua9BxF3ncJ4pSpRsR0808jSgkTw2Bnrl0Re6gYcvNIBddR9JcypMbu8JzNUYrUGFzHtkoYq4aLXnerCOHsZR2s7rGtBo/Xcb0Q4tQjlbanmUMNODhS9juN7emKHYRQj1KbBG1mkFr3bqe1u/xs5wQGgzG2v5brKoyxvTfjxOA6tt9QbAxRatg4UGS8GlBu2AFjcZoe9mzi2cs6effzz+XPbn6QiVoIzd6c1TCmHNj+Xp7WzaoHhzoJnmPvI272/ToUY6CUdUmNoRok+C6UMi6VMCFOmhW9eY+RarSon+WhHjbrqvZ2tVbibT66eUeGWRWXC17F8ZOJAt55y2d4/Z3/i8bw1MfvohRUKWcKS720JZdxIVjEMebsLYw538VzFGGzqrwRpbhas7wrx1Q9YrIWtqdJ+65DagxhnJKa1mTpA597mmbrEc+e4PAdh3NXdlAPE77/6AhaKbryHr6rmK5FNOK02e9v/mfa/j1jV3bl6Cv6fOuhGkorct7M+4Vu/vEdzf27ptg5UeOM3qN/jhxsm9kFKzv5ufOObniNEEKI4+94zmiYbeNAid951plMNyIe2VdmfX+eFZ35dqzZKho4f0UH33loiLu2TxBECTnfwXNc4mpE0PycDRPTngCfdW1FXzHrohRzChEOtvbNw2Vu3zJGLUypRwm+Bt91cR1FnNq2Ys2PcnKeph6l7UFIB4sToxRMmhLG0JF1UC6UG0m7CCE9GYLGk9ylux7iw1/5COsm9gLwzC138s89K5d4VacHd9ackWNd5e1qW2zjO5quvMdopVnFjj1O9V1Fkhp+tnOCzcPldmx4sBjy3BUdZD2HjKNxtCJKDZlZx6SJMWhlK989R9Nb8I/Zv+V4HNcLIU49krQ9Dc3eMnws1KKEvmKGsWb7AkfbM4qpgSBO8V1NzneImnuztIJKEFP0XYoZlygxth/XrN5Zyn5y4iqF72hKWZeBUpaBDts/dttodUHJluvPHWTvVJ2Pf3cLE7XAtkdQNoGU9WyVg1aa1KSY5oeqDaIPHSJowHVsosl3HcK4QZwYJutxs8rWwXH0Ye9nzn2qZqVDc8yvYWaoRJLaLTy6OWEibAYX8927xn5BNQMe31F4vqYSJGQ8TT1MlySBe9GeR/jIzR9lw/guAD534c/xZ8/8NSqZY/dcPBm1wqZDPRMcBXnPIU0SImN/Z1lPgwHP1RhjmKzNDExoVdrannSa7oLP0HRAOYgAhVb2pMeGgRK9eY9vPbiP0VqEoxSDHRlynmbneINaNDPEpPV8sgP5FL6r6czZyp4IWykbxyl53yHnOSgFtdAOIVvRnWeg5POTxyd4ZGiavmLfAdtW9z8Y1VqxbqBIYgyeUSQGHGxwG8YpvuvQlfeYakQ8Plo9JklbkABXCCFOZcdzRsNsm4fLfP3+oWb7r4Ch6QbLO2uctaxE1nPaRQMXrOziA197CIX9PPabydlixiFO0nYStOA7JOlM24FixkEpje/oQ669NXRt10TN7l5rzlBoRDEmskmmVpGBpyHrOcSpIYgPHX/OjifCxNCTt0UHZp4A8Vj2tXUVHGZpJ71MHPL2H/4bv/6TL+KYlD2lPt71nLfyg3WXLvXSTkoLiYNn08oeY81+Di/kOZhttpuNkmYf54PIebaNQBAn1EM7ZLeYcdpD+4LEzlZIU8M3HhhibU+Bvc2ig4Lv8sanrp9zeXlHlk/dspV7d03SmXUZr0XteSrG2GKmvKeZrsectazEpau7F/iTWJhjfVwvhDj1SNJWHFbBd+krZugr+uybCtg33SBOUozRFLMuxYwDqHZgWm7EpKmhmHVwI0U1TOxWkmYlbCWIiRM7uMgmagyPDVdY05vnNVefQc5zF5xsSVPDdD3m4lWd5DOaB/aUqQQxyzuyTAd2W5oNbm1vMM9ReFrRiGYSY75jz8hGqf173rctH3K+g1a2FUSS2r5hLjDYlWVZR5ZKI2bnRP2wgUbGsUlulA2+FQpj87Z4rkIZOxwiNXZ6cCsPfLCZaa1BUxh7H3VjaMQ2KLliXS+PDJXZO9k4YYlbL4l4662f4zd+/F84JmWo2MO7nv0WvrfhySdoBSeWo+b2jlOA0+zRvD+F3Z7VV/Sbf9c0IoPWioEOn6xrk6PVIKHou0zWIxLT2tKYNhOrLlO1mSoBFHTkXMI4ZftYjXroU43tSZQ4SRmthJSyLj0Fj7RiSJo98VwH6qFduOMo8r5jK9ODhGLW5fyVnQxNN9BKcc3GPsCePGidVAHYOV5n72SD6XpEZ36mkuBgB6MrOnMUMy4au0UzNClaKQoZl568D8pQDY7lb8eSAFcIIU5Nx3NGQ0srUTpeDVnTk2eglOWRfdPsnWowUgk4c7DEpWu6+bnzBtk1UWeqFtGR9wjKIamxcUDGdch6aXuwUSO2O2BWdueaJ/kjVnXnyPkO5UY079pnD11b2ZVj90SdUtZlaDqgHtqTrhpDztfNAgib8LIFEdFBd4zt3xc/jFPKQYrraLSywWNiZmKZxcaLGsi6djBanM6cZHaaCW3HHD6p3Lqf/R97qQejnb9vMx+++SOcNboDgM+ffx1/et2vM50tLuGqTj4OcxOnzXqUBT2XUgNhPHc34f5twQ54PGVnL0SJHaw3n1brkChJyXgOjTihEkTkMx6OtsUDYWL7O6/qsSeH7toxwV9+/RFGKwGNOCHrOmzoL3LD+YOcvayjfd+t96SBTnt8Od2I8V2NSW0Vje1p6/Oaq9cueAiZEEIslCRtxWHNrnp40hldlBsxd++cZLIWMVDymahFDHRkKGVdjDFM1SNyvp2w2Z33CZOgnYDyHYXn2DOTWikyrmaqHh/x9uXWwIoV3TlKWY9CxufHW8fYOlYliW0rBjCkRjUnAze3xjiqnYg1s9oMlLL2g70SJKQpFHMOXc3G85UwxnNsJWs9TJhsRO0hEQcLVrKuophx6S74RHHKUDnAGLt9xtWKRpSQGEXeU83Oo4YosveiNajm2nwHXK0Jk9Q215+dOGz+P05SHtw7Ddim+7VwpsryUOYLmhejo1HlFT/7Ko5J+Z9zn8Z7rn8TU7mTaxt6q8o5NYurAGn1NW4dKIH9uc+uJEkBkub9AxhbQeBq+5xrtRHpynlcsb6XdX0FPv/TXYyUA0LXVpx25j3yfo5rurJcvq6XiXrE/969ix3jNcaqEUlqK8gN4GvF8g57ILhvqsGeyTpRnJBxNMq165ysR3has6onh+9o9k03mpXxEa5WuI59LrW2cF6xroe879iqb5NSyLg48xwQn7WsxEglYPNIhTMHS4c9kF7fV6CvmKEaxAzmfVJjcJpVvgBD0w26ch7r+6R9hhBCCOt4trqZnSjdNFBstvWCvmIf03Xbv3JDf5E3PGU9rqvZNVHHKLvVOudpqmHSbvdTyLgkaUoQG7Kuw5rePBev6mTLSI2879Cd99k+dvCdY7OHrhkDrqPJeg5nL/MZKdvqX0cr1vUX2DNRZ7IeEyQGT6t2a6zZJ5FbvXZb1b+tJFhi7L+7lHGohilpajCpTV45eub2C2kvX8rYXWZZ36Ur7zFRi4iTlLg532J22k21/qgDB6Vp5k/QLbZq81i7fvPtnDW6g5F8F3/w7N/km5uuPGb3fapVIe9fpOBq5j2ucJs7BZPU4DYrTsKDlMK2WhfM91wz2Bh29te0slXsQbM1Xyt+9LQttoGZkwZZz85LiJpDwbxQE8QJcZKSpDPt/AY7MlywspMgTnl0qEwjSjhzsETez1EL4/ZQ7dl9Y2e/J/mO5tGhCvUwRjeHX585WOI1V6/lunMGF/UzFkKIhZCkrTisuZN2qyzvzHLuig7u2j7Bjok6XTmPtb15KkHM3qkGfaUMZxrDRC2kHiX0FOwU+lpo+9VqpVjdnePNz9zIuSs6jmr78szAilZ1X3NjmrHr9hxFmBh8rejIeTiObg6J0EzVI1xsu4S8o/FdRRClNMzMoLG+YoYgTkmMoTtvBytNNWJ2T9ZJU0PBd1HKVg7b/re2mtgYW025vCNHPYpZ05PnNVet48v37uan2yeoBvbsb2psZW/Os+0NwiRtJwe1UuQyunmdTf5FSdoOxPevpEgMDE8H5H17MKH3K9Vt/XT1PEFY2KyUaAVeUTwzyGrOc6GZnFZpgtF2q9FYoYvffe5v4SUxXzvrmoP+rmavxm8m7qODHCEcy6BdYYcEOI4iqxXlIJn3fh01c+DialjemaUaJIChEtjoM2mtV9kqg1b/YcdRdGRd+os+6/uLXLa2l/NWlOgrZdg5XgdsAnNVt+2Xd8W6Xj7/051sGamQGujKeWwaLLUP6DYPl7lr+wQTzV6yGUfb14c968BUI0Jru8WrFsaAIe+7rOjK4ijVHJ6Q0J33OWuwyN6pgKec2ceX7t7DrokavqtQStNT8Dl3eYmeQobHhitsGCgyMh0ctJdg1rOB6fq+IqOV4LAH0qu681y5vpdvPjhEpRFRynl4zYRxuW6f/1es72VVt1TFCiGEmHG8Wt3MTpTObr2glKIz73PmYImRcsDe6Qare/Ks6yvQlfOZqkcUfIdamFANbUWeo2zM5DuQzzh0ZD2mGwlXbejl+nMHDrtzbHYMq5UtdBgp25OgnTmP6UZEnBjbx97RzXYRtKtuZ9NqpjVDK682J79mDFprCr6iGsZECShtdwBFzV01YWJ3fh0s9ir4mv5SFgVMNWIUdo7FZDWc0+9eY3cZJa3WaGqmCAEg70JrPtvs5NsBaz5BdJqQarv3/uNXvQQ/ifiHJ/8iE/nOY/cY2BP68SmStd2/6rVVPNq+vvk7Vcr+8R2N52sGOjK4WrF9rEYjSjCziiUyDji6eSzTjp+hI+vRiFLixM5OiJp9oEtZl8GOrB0KVovIurr9PNNK4SjTnufgOqrZnkThaDij18bb+ybr7WMNV2sGOzJsHCjSnfe5dfMocWLY2F9sx7ylrEcx4/LYcIVvPDDE+r5i+3Xbek96wcUrmKqHbBuropSiv5jh0tXdUmErhDhuJGkrFmT/qocgTljdk2cgTtsJ0EaUcsHKTq4/d4BvPjDMjx8fI45TJupRc2K8QzHrkHEdnn5mPy+8eOVRB9+zB1YUMy5bhqsYAxv6C0SJYbIWMVYN6cq7zcFKPo6227xdx/Yayzb71kZJSiVIcHRz+3YxQ5ikjFdDtFIs68iR9x06gpipWoRS0FPw2T1Rp5rGRIlpB9TG2CA7JSVKDBsGily7qY9rN/Wxc6LG/92zh8/8aBuOUvR3ZNtThvdM1EmSGM9ROFqzcaBIV97jwT3ldpIXZgJzm5CcSdoCVMIU3z3wLPbswtzZZ/uzvoNpnhL3XE3cHIDmu6CVJohsj1xHY6s3h7fzof/7KH9/+Yv4+jnXkhr47qYr5vYsZibxmmITnChwtWo27Dc086AHnFWHZr/e5vcupPLjUBxNu2eywZBM1KiFNujzXYXnaPK+QyNMoNmqoDvvs7wzx55J29OqlHGZbETt4FArUI6m6DvkXEUjgUvXdPEHzzmHNb2FOc/rDf0HVgSduazE7z3nnHkPRltVQElquHpDL7dtGaOY9fCaJyGGpm0/gXIjphrEKAUdWb9ZBeTY6l4cXEczUQ15fKzGFet6ecXlZ3D52l4+/t3NjFUDlnfm6C9laEQJjw1X6Cn4/PKlq/nmg0OH7CV46Zpu3vCUuf2+DnYgrbXiFVesYbgc8Oi+cnOKtn2GOFpz0YoSr7hijfSbFUIIcYDj0ermwJP9c+0/VHN1d56zB4t8+5FhW6FqIDaGIEptwgo4f2UHv3X9mQx0ZBeVXN5/6NrGgSKVIGa8GlLI2CRiLUrYOV7DczQdWbe5+8c0T+IbnOaaWgNpD3ZSOuc7xM0ZEwoYLGUo5TzK9ZDJekTWd0hDMMa26dr/fnKeYmN/kWqY0Jn3QMH5KzvJew7fe3iIMEmbsySSmZ1Gjh1e6mqN70KjOXOhEc8kZ6OlyNI26TTh1+/4Is9+5Ee85JUfIHI8IsfjL5/22mP+WCm0h8cdrQ19OSZqEdNBTJxARgOHKLgAmxxHKZLEtH/2TrMWwHUUed+lEcY0mgcThmbyXdsh0o62xSAKu93P13bQrVa2wjbjOizrzOA6mkojZk1vwZ7UwK5t53idehQTNr9HK9NsZ2DnK0CM8mzbrpznECQJZ/TkUcrGvVnPYaCUYd9UnSCeKaJwlP1TyngoZY+9fMfBcxRXrOsjNSl3bp9gZVeOjOtQytrYdqoWsneqwfKuLB25uUUKSimWd2bZPFxh92R9zntQ6z1pNXnOX9l1TH6fQghxOJK0FQs2X9XD8o7svMkbrRR7puqMVUJW9eRtL6HUUG7E9BZ9bjh/2TFJ1Mxu3TBYyjBeCylmXbTW+MpO890wUCDrOoxUAkYrAeet6ODJa3tZ0ZVl90Sd4emGrVxUmp6CRxjZD/hqEOE6NkhwHdWeDFzM2NYDGoXTrIBtRPbr9cj2lqW59We8GrGqK8cvX7q6/e9d3Z1npBKglGJFdw5H2zOznTmPoq95bMQmnrtyDucs76C/lGVlV547t42zdbRKzoEwhlTNDFbTSrWnGIP9+sF+uvtvm2tEM5WnUTJ7z5Jt4WBcbYN8k/LqH3+Bd3z/M2TjkHfe8hm+f+41NLBDMlQz+AN7Rl4pW+WsgIyn6cx7GANT9QgFFHzoK2XAwJ4pO+it1a9XMRPQL6a/2XwVunnPYWVXjnzGpRpEtv+aag0Ds8nNepTSiFPbp82zg7gA+3xIEhqhbZehm/8WV9vvXd6ZI+tphqYb7J6o2+f+Ap/XBzsY3X+7ZM538RxNpnkGv6fo0wgTVvfmaUT2IOnCVV1sGakwXrXPf8+xv7PJesT6gWK7bcGZy0q85bqN7ZMv823b1JrD9hJ0Xb3gA+mNAyXedv0mvnb/Pu7bPUUtTMj7Dheu7OKG849um6sQQgixGPsnSve3/1DNraMVwmYiqREmoEGnEKa23c95Kzt47wvP58zBjgPu63D2H7rWU/C5eHUXm4cr7JuqUwliMLZ/7kBHhqynmapFJAmct6KTUtZh60iN4XKdODXzxkoa8F2NdjR9eY9KkJAvZXjyuh4Abt08arOoxm4vj1ODap68d5tFAY5WrOrKs36gyEN7y9TDBM9x6Mn7lIOIapRSbP4sHZ2Sa/bMj1MbAwZxSkfWxctqKkHcjhVhJm6bHTnNzuMqoOg7VMPkmFbhbpzczQf/72NcsvshAF7w8A/48oXXYYw5bonkxd7tfPFv1lWcv7KbYsbh6w/soxzEdrisozCkpImheRiCVjZWjZOUWpTa9hnNigitoSPjUo/tDj5bXOPjhdFMtbUxlLIeYZwSJWnzRIE9Dlrbk2OsGmGMoRomdORc29c5hYGOLMs6Mkw15zScv6KLyVrIT7dPsG2siqub1bRK4WubMA5jQzWIMaRkPRffs88Vp3nbYtYlTlJW9+Sphwn7pgOUot2aI07tEYQBilmP1d0FbjjftirYNx0wXg1Z3umQGEM9iNk8UsF1FGcNdhwwWBcOPHkjhBBLSZK2YlHmSzTNl7zZvzK3FsZkXIcLVx19P7L91zPTuqFCPYopZDIEcUKlEZPzXS5Y2UV33mOiZreyvOKKM3jqpv52VeP+1Y67Jmq87ysPkfdsP9tixuWn2ycYLjfwC3ZbT9Z1KGRc22JBK0pZD9/VxElKI04IIoOjYU1Pgbdct4kzl83tYbZ3qkEh4xKn9nYtjuOwvCPH7skajdgQJTYQ8V3NQEeWSpjQV/B5dLiMQtv+UY6ygV1zEuvhEpyt/mKObg3ScJiqR80KEnt/XmpI0pR8ziefcenYvZM/+uKHeNL2+wD4wfpL+fjL30UplyWsBrYC1TS3TynVTFqb9qTjqzb0cuZAkfv2TJNxNFtHq+0WEo7WrO8vMDwdMNncMp+Y1lYy+zuOE7Ow/rzNrYoaSIzB1ZpCxkFpqIUxO8cbgO2PVcy67SmyJk1JEoPraZ68tpuxathOgA6Usuwcr7XbJviuQ6l5cJXzbXK3M+8xVgnZOlplTe/R9Wc92HZJv+CjlK0Mrpq4nVjuLvis6clTytpK8/FaSDWISQ30FnxecfmaOa+3w205PR69BDcOlPiN47DNVQghhFiM/ROl8+0oaQ3VnL3z5fpzBtgyXGWoHBCnKa62J+7PW9HJxnl21CzEfEPXOnIuZw0W2Dddpyvvc/6KErUwZaIWUQsTcp5DxoOLVnXy0iev4XN37ODme/cQxuGc3UC6WV27ojPHUDlgqmbbO6zuybOh38Ypm4cqVIME33GoN6epZlxNom0lb5raNgfr+gs8eW0P3XmffZMNto5VWd9XoJR1GS43bCyqsSeSXadZ4OAQJoZKEKGMoTPnUgkSilkP3zFM1GJcbQf1GqPa8ayjIGy2f2idwI/TFN+zO7+OtFa1/XMh5ZV33sy7vvdpcnFA2c/xZ896A5uf9UIG6yG7JxtH+AiHf/zFrn327We3MKuFtho7n/HwXYdKEBElBt/RRCaFZhyc82yyvxok6EZErlnJOl0P2zMeWgn2SmCTk67jsKonw0Q1phbFZF1tE6IpFDIunTmfS9d0sao7x53bJ9g6WuWswSJnL+sgSu0aihmHzSPVOS23eosZrljXQxCneI5GYRipBISx7bHclfeoBBGNyNBTcDlneYl9Uw12TtgWY76jGezMtZ+7P946xlglBGNnMbRayXVmPa47Z4BXXDET+84X0567ooOs55D15m9psP/JGyGEWEryTiSOm+PVj2y+x3ndNWv5zzt2sn2sxlgltNtoOrJs6C/SU7BT7j1H01+017XWMF8SelV3ngtXdnH/nqn2NpoNAwXKQcRYJSBOYVlnlhWdWe7cPgFKceGqDsqNmJFygyg1dBc012zs49eesu6A6otqaM+K9xczjFYC/IKec9DQaiGxqjtPnKRsG7XVkFes62HjQJF7d02CgdjYLWcKe6ABNqBrVQg4zUFmrW1uBjvQzHNsJUWaQkfWZWVXnp0TNarNs/U536GQcRiaDvC05vm3/S9v/sqnyAZ1qn6Ojz7njXzzqufjOg7djqIexc3tb3bAVc5zqIQJythBFev6CqzsyrFltMaq7jxPO7OfL969m/5ihsTYAK+UdRmvBty2ZYztozUSbIDqO9putUoOf6bbBuS28jnnac7oLbC6O88jQ2XGKiFxcyvhxv4CGwaKjJZDxqt2WiwZl3pkv762t8Dqnnw7ARonCY6j8DQs78rRk/fxXb3fmflj95w+1HbJYtYlNfZgZrIW0l/KkG8GlD2FDN1rfcqNmCBO2D1Z5/K1vVy9oe+AxzjcltPj8do9HttchRBCiMWYL1F6sKGaO8dr7Z0vpaxHTyFDuRETNnvAgm3Dtf8W6sWY70RpnBjynsMlq7tY3VPAGHPA407UInK+wx8+71w2DRT55Pe3kKaG2BgcZXuLnruig568z7axKj95fJzugu1zH8QpP90+wWQ9orvgc+nqLiZqIQ/tLROnKSs6MkQpTDciegu+bYOQcakEMa6r6ch6uNpWQtp4yLaWyvkuXTmPybrtx99KyDkKgtjGLss7s2QcRTWotAdSpamxw8qaU18N4DmQ81zKQUwQG1zX1gTQmj2g7M6nMDHtLf8Zx8a8s0/yt5KlnqPYVB/jj774Ya7Ydg8At669iP/veW/DrF7NM9Z2c/vWcRzVaA8oPpRWAcRCbkvz8TGGcJHlto6yVbBxakjSZruxZj+yM5qD73ZN1tk6UqUSxIRRwkQtwmDa61vdk6ev6DNaDtkzVWO6Yb+QpilaO0Rpq6WbIWegK+dz9fo+UmPYM9VgtBwwVA7IeZoLVnayvCtHNUxwtaaU9fBcB60VXVmXepiweaQ6b8utjpzHYEeWkXKD7rxHLUwo+LbII4xsMrcz77Ouz76W1vTkufSMbkbKARO1kAtXdqKbuxOvXN/LY/vKbB2r0uv5rOnOs2GgyPXnDnL1hr458erBdop+6patCzp5I4QQS02StuK4OlGJmo0DJX73hrMBxYN7p9jYX6Qj57U/hBfzATx/5YPHpoEi9++eJm0OMlNK8XPnDmKAyVrUDDY8lndmue6cQa7ZL2hoKfguOc8h15WlGsZztrRHScpENaSUdXnb9ZtY11eckzTbOlphtBLYwDpOcbVpV6YqpdAaXNPsaWXA9zSquQUp6+o57QvC1JD1XLRWtk9Us0dbEKd05Ty68z4X7HqId3zxYwBsPudJfOXt7yMdXMWmiRo7x2tUwoTuQoYoSSn4Luv6C3TnfSZqIVtHqsRJa+hb3K7UzLgOX7t/H66j6M767Z9LbzHL8y9cwS2PjfDA7ik8R2OwVba+q+n1HcZr0bzb1hSQ922Cd2V3jk0DJd78jA1s7C+xa6LG1tEqO8drfPuhIc5b0YnraNb2zj0IqkcJt20ZZfNIhTMHS1y8pssOI5mqs7avyNaRCknKAQlbYwxTtYjOnM+6vqOrsoVDb5ecqNpq5N6CzxXr+jh7eYnvPDw858BTKZioRazqznPD+YNHnGiVJKsQQojT0UJ3lOzf/1YpNaf/ZZymDJeDo95CvX9Sad9Ug/+4YycruvKHfVytFddu6ucnj4/jORrP1e2T4a1Ypa+Y4fyVnazvKzJSbvDA3mkqjZj1fQU2DpToKfgMdGTZNFDk3t1TrOsr8Lpr1hFEKd980P6MhssBGdfhqvW9nLWsxMN7y2wZqdCIUnKeQy1KGCj5FDIeWc9huNygGthZAFnPDoCarEV2XUBH3mOqFtm5Bs1/lzGtaljwHVsFmRiHME5IEoNq9vZS2F1GxaxLuR4zVQ8xzS39jlK06nFLWRdXKyZqEXFq+IMv/zVXbLuHupfhb5/9Bv7l4mdTyHqs67GxWz2y1ZXECVrZ5K/WCpPaWLvdCxbIeA5BfPiWDa2kcZwsJLV74PdqZYf2ZlwH39Vcu7GPl12+mv/4yU66Cz5aa9b02CKFVkwbxgl7Jht4rsYYw/q+AvmMS2/Bx3MVa3oKxGnKA7unqYQJjoaOnEt3waMja58LL718NRv7S+3n5Gg54Gc7Jtk6Wm0Xk1y1Ye5zYb7X0f4tt9b25RmvBu1B1peu6SJO7THaeSs7+X9PX0/B9w449rnx1m3tYdg538FzNF2FDFd15Xjehcs5Z1nHIYsL5otpF3ryRgghlpokbcVpw3U1L3nyKm68NWaoHKC1OuIP4IMF9L9w8UouXN1JfynTDiaARVUkzk7KXbSqk60jtfaWdlsp6tihZRv7D7ifjQMlfvMZm9g6UuWhvdMEUYrjaHzH9lJtDaYqZlzCJGVVcy2VRkwtSqiHCUpBV9ajGiXNCa6GnO+yrCPLWDVkqhYyUgkY7Mhx7gufxe7yG3A3rCfz/97MbzYD292TdcpBRKVhB8CNVUPu2TnJ1pEq5UZEwXd54Tw/q1ZLioNtTQTb23fjQIm1vTmU1kRxypaRMjnfox4b/DQljFuJc3uGPkmNHdChFFes7eWVV85si1rTW2BNb4Gd4zXu3jFJPUooOfqAgyCl4MzBEuv7CoxWQoLY/s6v3tDH9ecM8rmf7OCbDw0x1kyqt5Ls5UZMagxXre9hdffRJzkPtl3y7GVFto4q1g8UecXla9qVBGf05o9pKwMhhBDidLeQHSWL7X97NGYnlVon9xf6uCu7cmwcKHH/nik2dRYPOUT0rp0TfOr7W+kpeCzvzM25rdaaDf1FJmsRWtk++BsH5v8ZPeOsgfb19+yc5O9v2cpIxW67z/kOuea282LG5fJ13azqyvPjx8eJmifz+woZakGCo2w1brOA0s5SwO6aChNDKeviOz6NOKWYcdqVthes7KKn4DNcbvDDx0aoBAmuo/EcTcnRM4OqHE3GU8QJvOf6N/Inacrf/fLbGV++Bq/coKeYaQ8vNtgWDXFqW3cZUvt8UAqUaSdeZ+9KC2MbW4MdBpfM+j21B9c2f8QLafPV+r5WijfF7porZlwuPaOb33zmRuLUECRpe6cVzE3sx2lKLUz4hUtW8ti+ypyk+xXrern+nEG+8cA+XK3pznmgFd05r/39jw1X+NaDw2x8Wmkm0bkMrt7Qd9jnwuFabu0/yHq6Ebdj7YPFrQc7JjvalnvHox2YEEIcD5K0FaeVY/kBvJgt4oupSJydlBurhpy1rEjS3IY2UQtZ1ZXnFVesOWji98xlJT704ov4jX/7KXunGrha2UqF1AaSec8h3xywlfddMp5DR9ajEsRMN2I0NvjLxLaPWavSN+M5rA2nefWX/pZvvvq3eOFzL7HVwtd9akH/3msOEswd6t8/39ntVV15NvQV2TvdYGN/kXIjZudE3fbb1QqlNKWsQ1/Rx9Uag6ESJGzoL6KU7V01X1/ZhfSxax3UzDdc7xVXrmG4EvDoUJlyY6aqxtGKi1Z38fJD/M4W62DP4yvW9R7wPD5RbUiEEEKI08nhdpQspv/tsbTYx11oywfX1XTkPDKeZrAjd8BJczhwANPBfkazrz97WQd9xQyfvnUb28aqjFdDFNCV97lwVSfnrejCGDOnR7/rKHqLPlnXYetolbCZxOvIuoSpTUznPI2nNb3FDLUgJus5VIKYwY4sZ/Tm2+vvLWZAheQ9Ozw40xwa96Tbvsbgvu187CmvZHlXhq7V5/A3F/8tOV+zWmu0grMGS/QUMkzXI7KuQ+DZHWdRkqCVpuA5RElKkBi7g83VFHwXpaC/lEErxb6pOpP1GJTBwe5+a1GquUMLMFFK0pz/4GlFPZ5bfdtK8hYyjh0wHNkBXArDz507yKuuOoONAyV2jtcWdDLhnGUdXH/24AHx4e7JOltHq2wcKM77/cs7s2werhzQ9mMhz4X5LGaQ9WLu41jEuhJDCyFOBZK0FaedY/kBfLy2iM935vlgSbn5nL28g//veefy199+jL1TdVyt8V1FEDuEsaG7YHtMVRq23xhAGKes6c6TmpTHx2qs7yuwob/IlpEqE7WQS2//Fm/974/SVZvm2q+5FH/nBYv6Ny3mZ3W45DrAjbdu47HhCss6MnTlPPZO1UGBSaG/mCGfcTHGMF4NGWgGzheu6mTVQapdF3NQc7Dhem+7fhNfu28f9+2eohbF5D2XC1d1csP5y475GfnFPI+llYEQQghxbC2m/+1SP+5CixaOV/XwdecM8rRN/dy1c4Kxakh33uOOx8d5cG8ZYwxKqXaP/rGqnTOwvCvH2YNFHAd2jtdxlMJzHbx2azObRD5rsMijQxV2TdbpL/ps6C+0E7bFjIOrNV05n2LGoRomdFYn+e3//AhX3v09UqV46KJr+KVX/xJXr+9rJwpznsOX79nDA3umMcZW9PYWMtTDhKznUG4YlDI4WqG1JkhsG4jL1/XwlDP7+OztO0lSQ2RS+ooZ+juypKlhrBIwVo1odnJAYXv1pim4jiajbB/fM3pz7JsOqTQi4ubwXa0h77uc0ZOnFiV05TxWdedoxCmvv3ZduyBhMUn9+eLD/dt+7G//xP2xsNBB1ou9j2NBYmghxMlOkrbitHQqfAAfbXL5unMGWd2T4/N37mbLSIXU2Mmpo5WQjpxHf9HnkaEKQ9MBYChmXZZ1ZhiphO0hEr6ruarL8NSb3sfFt30DgMb5F1D8yAeP47/cOty/f/bBh+9qtFKUMh5KQS1MSIwhiOx2ONfR9BYPf/B0tJXYGwdK/MYzTtwZ+VPheSyEEEKcrpZqC/WRPO5C4srjWT3suprL1/W2L/cVM+ybDua0eto0UOD+PdOkqcJ3NNONhOeev4JNg0XueHxiTjwbNlspTDdiugs+cWroyHnNgbppO4l95jL7s5iqRZz/k2/zxs9+kK7KJLF2+I/nvJZf+NWf56lnDgBzE4XPPn8Ze6cac/qtjlVtr+Bi1gUD1TAhSQ05z+WZZ/fzG8/YSMZ1uHfn1AE9hAG2j9e45dFh6mGKUjQHiBk8R9NT8OktZhgpN+gtZjijt8Bjw7Z1QZSk5D2Hwc4s9Sgh7zucs7zEWDXiolVdcwoSjvZkwols+yGEEOLoKdMaOy+Ou+npaTo7O5mamqKjo2OplyNOE2lq5gTo9TBpD44YrQSMVkIUht5ihr5iho0DxfbggMzXbubFf/9eOqfGSLXD5NveQc/73wu+f/gHPgFm/9taQxB+tmuSneM1aqENalf35Ll0TfeiDp72/5nJVighhDh1SDy1tJ6IP/+lihuOx+NuHi5z463bGK+G8yb8XnfN2mOWjN48XD5gV9mG/gIXre6ib56ZB4faQj87vm3d18aBok1QTk4QvfmtnPWtLwGwa+UGvvGuv+DalzyLMwcP/hzdf31BnBJEKb6rSFJDGKes6MrxgotXtGdNpKnhE9/bYnsID8ztITxVC/nGg0MMdvis7ytRj2KMga68R9ZzAcPOiXp7fsJoJWD3RJ3xWoijFJ15j/5ShhWdWepResjfx3w/29bP41C/v0Ot3xjDY8MVLljZyZuetkFiYyGEOI4WGk8tadL2lltu4YMf/CA//elP2bt3L1/84hd54Qtf2P66MYb3vOc9/MM//AOTk5Ncc801fOITn2DTpk3t24yPj/OWt7yFL3/5y2it+aVf+iX+6q/+imKx2L7Nvffey5vf/GbuuOMO+vv7ectb3sLv/u7vzlnLf/3Xf/Hud7+bbdu2sWnTJv7iL/6C5z73uYtay+E8EYNcsTRmB745z0EBtSiZGxx/9nPoV7wcgOjMs3E+cxP6isuXduGH0fp3zR6CVsp6knQVQognkJMxnnoixbQn489fLM6RJvyOxLFMPM97XyaFCy+EBx/EaM34b/wWtT94NysHuxb0OIdLFs+33oMlvvdMNtg7Vacj63HJmq6DJkRnz0/IeQ7bx6p85+Fh9k41cJQi6y08AXskP9sTmbgXQggxv4XGU0u676FarXLRRRfx+te/nhe96EUHfP0v//Iv+eu//mtuuukm1q1bx7vf/W5uuOEGHnzwQbLZLACvfOUr2bt3L9/85jeJoojXve51vOENb+Df//3fAfuD+Lmf+zmuv/56PvnJT3Lffffx+te/nq6uLt7whjcAcNttt/Hyl7+c97///Tz/+c/n3//933nhC1/IXXfdxfnnn7/gtQhxsljItnr9wl+A886D5zwH773vhVPgeSztAoQQQpyMJKYVp5ITOYDpWMZu89+XA7/7u/Dnf4666SZ6r7yS3nm/e+H3ebj1Hqx9xYWrOnnxZav4zsPDi5qfcEZvgWs39i/693GkP9ulavshhBDiCJiTBGC++MUvti+naWqWLVtmPvjBD7avm5ycNJlMxnz2s581xhjz4IMPGsDccccd7dt89atfNUops3v3bmOMMX/3d39nuru7TRAE7du8613vMmeddVb78kte8hLzvOc9b856rrjiCvPGN75xwWtZiKmpKQOYqampBX+PEMdMpWLMBz9oTBzPXFevL916hBBCiCNwssdTp3tMe7L//MUTwHe/a8y3vz1zOU2XJKZNktTsGKuah/ZOmR1jVZMkqTHGmMeGps3ffvsx8/bP3W1+41/vNG//3N3m4995zDw2NH3C13goB1u/EEKI42+h8ZReunTxoT3++OPs27eP66+/vn1dZ2cnV1xxBT/60Y8A+NGPfkRXVxeXXXZZ+zbXX389Wmtuv/329m2e+tSn4s/q0XnDDTfwyCOPMDEx0b7N7Mdp3ab1OAtZy3yCIGB6enrOHyGWxA9/CBddBO98J3z4wzPXS0WNEEIIcVyd6jGtxLPipFGrwdveBs94BrzqVTA+bq9Xakli2lal69nLOljdk29Xxm4cKPH/nr6Btz/rTN5y3Sbe/qwzedPTNpx0FawHW78QQoiTx0mbtN23bx8Ag4ODc64fHBxsf23fvn0MDAzM+brruvT09My5zXz3MfsxDnab2V8/3Frm8/73v5/Ozs72n9WrVx/mXy3EMdZo2ETtU58KW7bA6tVw6aVLvSohhBDiCeNUj2klnhUnhR//GC65BP7qr+zln/958LylXdMhSEJUCCHEsXDSJm1PB7//+7/P1NRU+8/OnTuXekniieTOO22C9kMfAmPgda+D++6D/SpwhBBCCCEORuJZsaSCAP7gD+Caa+DRR2HFCvjqV+FTn4LSyVW5KoQQQhxrJ23SdtmyZQAMDQ3NuX5oaKj9tWXLljE8PDzn63EcMz4+Puc2893H7Mc42G1mf/1wa5lPJpOho6Njzh8hToh//Ee48kp46CFYtgy+/GX453+Gzs6lXpkQQgjxhHKqx7QSz4olMz0NT34yvP/9kKa2JcL998Ozn73UKxNCCCFOiJM2abtu3TqWLVvGt7/97fZ109PT3H777Vx11VUAXHXVVUxOTvLTn/60fZvvfOc7pGnKFVdc0b7NLbfcQhRF7dt885vf5KyzzqK7u7t9m9mP07pN63EWshYhTipXXgmOAy97mQ1un//8pV6REEII8YQkMa0QR6ijAy64APr74QtfgH/5F2g+14UQQognhBM0GG1e5XLZ3H333ebuu+82gPnIRz5i7r77brN9+3ZjjDEf+MAHTFdXl/nSl75k7r33XvMLv/ALZt26daY+azros5/9bHPJJZeY22+/3fzwhz80mzZtMi9/+cvbX5+cnDSDg4Pm1a9+tbn//vvN5z73OZPP582nPvWp9m1uvfVW47qu+dCHPmQeeugh8573vMd4nmfuu+++9m0WspbDkWm74riJImNuuWXudY88sjRrEUIIIY6jkzGeeiLFtCfjz1+cRh580Jh9+2Yuj48bMzS0dOsRQgghjoOFxlNLmrT97ne/a4AD/rzmNa8xxhiTpql597vfbQYHB00mkzHXXXedeWS/RNTY2Jh5+ctfborFouno6DCve93rTLlcnnObe+65x1x77bUmk8mYlStXmg984AMHrOU///M/zZlnnml83zfnnXeeufnmm+d8fSFrORwJcsVx8cgjxlx5pTGOY8xPfrLUqxFCCCGOq5MxnnoixbQn489fnAbi2JgPfciYTMaYF7zAmDRd6hUJIYQQx81C4ylljDEntrb3iWt6eprOzk6mpqakH5g4emkKf/M38Pu/D/W63UJ2003wwhcu9cqEEEKI40biqaUlP39xzG3eDK99Ldx6q738nOfAf/0XFApLuiwhhBDieFloPHXS9rQVQhzC44/DddfB295mE7bPepbtXSsJWyGEEEIIcSpIU/i7v4OLLrIJ22IR/uEf4OabJWErhBBCAO5SL0AIsUif/jS85S1QqdiA9kMfgje+EZRa6pUJIYQQQghxeEND8KpXwbe+ZS8/4xnwz/8Ma9cu6bKEEEKIk4kkbYU41dRqNmH7lKfAjTfChg1LvSIhhBBCCCEWLp+HLVsgl4O/+At485tByyZQIYQQYjZJ2gpxsjMGRkZgYMBeftOboLcXfvmXwXGWdm1CCCGEEEIsxMiIjWG1hlIJ/uM/oLMTzjxzqVcmhBBCnJTkdKYQJ7PhYfilX4Irr4Ry2V6nNbz0pZKwFUIIIYQQp4b/+A84+2z45CdnrnvykyVhK4QQQhyCJG2FOFl94Qtw3nnwxS/Crl3wgx8s9YqEEEIIIYRYuNFRW2zwspfB+Dh89rN2F5kQQgghDkuStkKcbCYm7GCGX/olG+heeCH85Cfw3Ocu9cqEEEIIIYRYmP/9Xzj/fPjP/7Q7xN7zHvjOd2R4rhBCCLFA0tNWiJPJV78Kv/ZrsGePbYPwe78Hf/RHkMks9cqEEEIIIYQ4vKkpeNvb4NOftpfPPRc+8xl40pOWclVCCCHEKUeStkKcTP7+723C9qyz4Kab4IorlnpFQgghhBBCLNxjj8G//IutqH3nO+FP/gSy2aVelRBCCHHKkaStEEstTW1VLdjhDOecA+9+N+RyS7suIYQQQgghFmJ2PHvZZfBXfwWXXAJXX7206xJCCCFOYdLTVoilUqvBW98Kr3nNzHWDg/Dnfy4JWyGEEEIIcWq45Rbbu/b++2eue/ObJWErhBBCHCVJ2gqxFG67DS6+GP7mb+Bf/xXuvnupVySEEEIIIcTC1evwjnfA058ODz0Ef/iHS70iIYQQ4rQiSVshTqQgsMPFnvIU2+9r5Ur42tfs9jEhhBBCCCFOBT/5CVx6KXzkI2AM/Oqv2mFjQgghhDhmpKetECfKXXfBr/wKPPCAvfwrv2L7fXV1LemyhBBCCCGEWJAwhD/9U/jAByBJYNky+Md/hOc9b6lXJoQQQpx2JGkrxIkQRfCiF8H27TAwAJ/6FLzwhUu9KiGEEEIIIRbuM5+B973P/v3lL7etvnp7l3ZNQgghxGlKkrZCnAieB5/4BPzzP8Pf/R309y/1ioQQQgghhFic170OvvxleNWr4MUvXurVCCGEEKc16WkrxPGQJPDBD9ohYy3PeQ78139JwlYIIYQQQpwaHn7YJmjrdXvZceBLX5KErRBCCHECSKWtEMfaY4/Ba18Lt90GHR3wrGfB4OBSr0oIIYQQQoiFSVM7e+EP/gAaDVi9Gt7//qVelRBCCPGEIklbIY6VNLWtD373d201QqkEH/uY7WErhBBCCCHEqWDrVtsG4ZZb7OUbboA3v3lp1ySEEEI8AUnSVohjYft2eP3r4TvfsZef+Uzbv/aMM5Z2XUIIIYQQQiyEMfD3fw/veAdUq1AowEc+Ar/+66DUUq9OCCGEeMKRpK0QR2tsDC6+GCYnIZ+Hv/xL+H//D7S0jBZCCCGEEKeI97wH3vte+/enPhVuvBHWr1/aNQkhhBBPYJJVEuJo9fbaLWRXXw0/+5ndPiYJWyGEEEIIcSr59V+HZcvgox+F735XErZCCCHEEpNKWyEWyxj43Ofgsstg0yZ73fvfD65rJ+oKIYQQQghxshsagv/5H3jjG+3l1attP9tcbkmXJYQQQghLygGFWIyREXjJS+AVr4DXvhaSxF6fyUjCVgghhBBCnBo+/3k47zx405vga1+buV4StkIIIcRJQ5K2QizU//wPnH++DXJd107SNWapVyWEEEIIIcTCjI3By18OL36x/ftFF8GKFUu9KiGEEELMQ9ojCHE4ExPwW78F//Iv9vJ558FnPgOXXrq06xJCCCGEEGKh/u//bN/affvsDrHf/31497vB95d6ZUIIIYSYhyRthTiURx6B666D3bvtcLHf/V344z+27RCEEEIIIYQ4FbzznfChD9m/n322LUB48pOXdk1CCCGEOCRJ2gpxKOvWwcAA5PNw001w1VVLvSIhhBBCCCEW54orQCl4+9vhz/5MetcKIYQQpwBJ2gqxvx//2LY+8H3754tfhL4+KBSWemVCCCGEEEIcXrUKDz8MT3qSvfzLvwwPPADnnLO06xJCCCHEgskgMiFa6nVbfXD11fC+981cf8YZkrAVQgghhBCnhltvhYsvhmc/G4aGZq6XhK0QQghxSpGkrRAAt98Ol1wCH/sYGAPDw/b/QgghhBBCnAoaDTt/4SlPgc2bIZuFnTuXelVCCCGEOEKStBVPbEEAf/AHtrr2kUdg+XL4ylfgE5+wfb+EEEIIIYQ42d15p22F8MEP2sKD174W7rsPLrtsqVcmhBBCiCMkPW3FE9cDD8DLX24DWoBXvhL++q+hp2dp1yWEEEIIIcRCGAN//Me2tVeSwOAg/P3fwwtesNQrE0IIIcRRkqSteOLKZGDLFujvh09+El70oqVekRBCCCGEEAunFOzbZxO2L30pfPzj0Nu71KsSQgghxDEgSVvxxDI6Cn199u8bN8J//zdceikMDCztuoQQQgghhFiIJIHpaejutpc/9CE7dOwXf3Fp1yWEEEKIY0p62oonhiSBD38Y1qyBW26Zuf7Zz5aErRBCCCGEODU8+ihcey285CUzQ3NLJUnYCiGEEKchSdqK09+WLfD0p8Pv/A7U6/Bv/7bUKxJCCCGEEGLh0tTOXrj4Yvjxj+EnP7FDdIUQQghx2pKkrTh9GQOf+ARceCH88IdQLNrBDJ/85FKvTAghhBBCiIXZtg2uuw5+67dsAcL119tBumefvdQrE0IIIcRxJD1txelp50741V+Fb37TXn760+HGG2Ht2qVclRBCCCGEEAtjDPzTP8Hb3w6VCuTztn/tm95kB5AJIYQQ4rQmSVtxevrud23CNpeDD3wAfvM3QUthuRBCCCGEOEWEoZ3JUKnYPraf/jRs2LDUqxJCCCHECSJJW3H6SNOZxOyrX20HNfzKr8CZZy7tuoQQQgghhFgIY+wfrSGTgc98xg7RfdvbwHGWenVCCCGEOIGk9FCcHv7zP+1ghvFxe1kp+LM/k4StEEIIIYQ4NQwPwy//sm2B0PLkJ8M73iEJWyGEEOIJSJK24tQ2NgYvexm89KV2IMOHP7zUKxJCCCGEEGJxvvAFOP98+/8//dOZQgQhhBBCPGFJ0lacur78ZTjvPPiP/7DVB3/0R/Ce9yz1qoQQQgghhFiYiQl41avgl34JRkbgggvgBz+Anp6lXpkQQgghlpj0tBWnnqkp29fr05+2l885x/b7uuyypVyVEEIIIYQQC/fVr8Kv/Rrs2WN72L7rXbYAIZNZ6pUJIYQQ4iQgSVtx6nn3u23CVinb4+u974VsdqlXJYQQQgghxMLs3QsvfCGEoZ3BcNNNcOWVS70qIYQQQpxEJGkrTj1//Mdwzz3w538O11yz1KsRQgghhBBicZYvt0Nz9+yB970P8vmlXpEQQgghTjLS01ac/H7wA3jrW8EYe7mnB77/fUnYCiGEEEKIU0OtZtt73XHHzHXvfCd89KOSsBVCCCHEvCRpK05ejQb8zu/A054Gf/M38LnPLfWKhBBCCCGEWJwf/Qguvhj+6q/gNa+BOF7qFQkhhBDiFCBJW3FyuuMOuPRS+PCHbYXt618Pz33uUq9KCCGEEEKIhQkC+P3fh2uvhcceg5UrbWWtKx3qhBBCCHF4EjGIk0sY2v5ef/7nkCSwbBn8wz/A85+/1CsTQgghhBBiYe6+G37lV+D+++3lV7/aVtp2dy/tuoQQQghxypCkrTi5vOxl8MUvzvz9b/8WenuXdk1CCCGEEEIs1J13wlVX2TYI/f3wqU/BL/7iUq9KCCGEEKcYSdqKk8tv/Rb88Ifw8Y/Di1+81KsRQgghhBBicS691M5k6OqCT3zCJm6FEEIIIRZJkrZiaT3yCDzwALzoRfby054Gjz8OhcLSrksIIYQQQoiFSBJbTfvqV0OpBFrDl74E+TwotdSrE0IIIcQpSgaRiaWRpvCxj9lJuq9+NWzePPM1SdgKIYQQQohTwebN8NSnwpvfDO9858z1hYIkbIUQQghxVCRpK068xx+HZz4T3v52aDTsRN1cbqlXJYQQQgghxMKkqW3nddFFcNtttsL28suXelVCCCGEOI1IewRx4hgD//AP8Nu/DdWqrUD48IfhDW+QSgQhhBBCCHFq2LEDXv96+Pa37eVnPANuvBHOOGNp1yWEEEKI04okbcWJkabwghfAzTfby095Cnz607B+/ZIuSwghhBBCiAX71rfsLIZy2e4U+8u/hN/4DdvHVgghhBDiGJLoQpwYWttJupkMfOQj8L3vScJWCCGEEEKcWi64AHwfrroK7rkHfvM3JWErhBBCiONCKm3F8TM0ZKsQNm60l//wD+FVr4Izz1zadQkhhBBCCLEQxsCPfgRXX20vDw7CD38ImzaB4yzt2oQQQghxWpPTwuL4+Pzn4fzz4SUvgSiy1/m+JGyFEEIIIcSpYXQUXvpSuOYa+O//nrn+7LMlYSuEEEKI406StuLYGh+HV7wCXvxiG+gmia24FUIIIYQQ4lTxpS/BeefBf/0XuK4dPiaEEEIIcQJJ0lYcOzffbKtrP/tZW33wh38Id9wBq1Yt9cqEEEIIIYQ4vMlJeM1r4IUvhOFhm7i9/XZ4+9uXemVCCCGEeIKRnrbi6NXr8Ja3wD/9k7189tlw001w+eVLuy4hhBBCCCEW6jvfgV/5/9u796go632P4x8uMoAwgJmQCRsLNBXvGlGZmRRqpu3aJ3eRmbeyNPXoSbtqp21bu+jxkpZtFdJcara3djNXhor3Sygoaoil6S7BygvgXfmdP2YzOQo6qDkzzPu1Fmsxz/ObZ77z+zWrj1+eeZ4npJ9+st1c7Pnnpf/9X9uNdAEAAK4xmra4cgEB0o4dko+P7SyEUaOkoCBXVwUAAAA479QpW8M2Ls52AkLZzccAAABcgMsjVNLkyZMVGxurwMBAJSYmasOGDa4uyTWOHbOdYSvZLoWQni5lZkpjx9KwBQAAcHNk2v84ePD33zt0sF3mKzubhi0AAHA5mraVMG/ePA0ZMkQjR47Upk2b1LRpU6WkpOjAgQOuLu3aWrNGatbMds3aMvHxUps2LisJAAAAziHTynbywZAhtrNq9+37fftf/ypVr+66ugAAAP6Dpm0ljBs3Tn379lXPnj3VsGFDvf/++woODtaMGTNcXdq1ceKENHy4rTmbny998olUUuLqqgAAAFAJXp9pN2yQmjeX/u//pEOHpIULXV0RAADABWjaOunUqVPKyspScnKyfZuvr6+Sk5O1du3acp9z8uRJFRUVOfx4rKwsqVUr6a23pNJS2111c3KkkBBXVwYAAAAnVTbTVqk8e+qU7ZtiSUlSXp50ww3Sl1/abqgLAADgZmjaOunXX3/V2bNnFRkZ6bA9MjJSBQUF5T5n9OjRCgsLs/9ER0dfi1KvrtOnpddek267Tdq2TapVy3Y2Qnq6FB7u2toAAABQKZXNtFUiz0q2kw1at5beeMN2AsJjj0m5uVKnTq6uDAAAoFw0bf9AL774oo4cOWL/2Xfu9bI8RWGh7atjZ85I//VftsZt166urgoAAADXQJXIs5I0a5a0ZYtUs6btEl+zZ0s1ari6KgAAgAr5u7oAT1GzZk35+fmpsLDQYXthYaGioqLKfY7FYpHFYrkW5V1dxkg+Prbf69SRpk61Pe7WzbV1AQAA4IpUNtN6bJ6VHDPt3/5m+wbZyy/bvjkGAADg5jjT1kkBAQFq2bKlMjIy7NtKS0uVkZGhpKQkF1Z2le3cKd1xh7R48e/b/vpXGrYAAABVgFdk2rNnpXHjpHvvtf0uSUFB0oQJNGwBAIDHoGlbCUOGDNE//vEPffjhh9qxY4eeeeYZHT16VD179nR1aVeutFSaNElq1kxau1YaMsS2DQAAAFVKlc60338vtWsnDR0qZWRI8+e7uiIAAIDLwuURKqFbt2765ZdfNGLECBUUFKhZs2ZavHjxBTdy8Dh79ki9eknLltket28vTZ8u+dLTBwAAqGqqZKY1xnZJr//5H+noUSkkRBo7lm+LAQAAj+VjjDGuLsJbFBUVKSwsTEeOHJHVanV1ObZwO3269N//LZWUSMHB0ttvS/360bAFAABuye3ylJdxy/nft0/q3VtassT2uG1bKS1NqlvXtXUBAACUw9k8xZm23mzFCqlvX9vvd9whpadLcXEuLQkAAAColCeekJYvlwIDpTFjpOee4wQEAADg8WjaerO2baWePaWGDW1n2/r5uboiAAAAoHImTpQGDJA++ECqX9/V1QAAAFwV/Anam/zyi+2rY4WFv2+bMcN27S8atgAAAPAE8+fbrldbpnFjKTOThi0AAKhSONPWWyxYID39tK1xW1TEnXQBAADgWX77TerfX5o3z3bCQfv2UrNmrq4KAADgD8GZtlXdoUNS9+7SQw/ZGraNG0svveTqqgAAAADnff65lJDwe8P25Zdtl/gCAACoojjTtipbvNh2OYSff7bdjGH4cGnkSMlicXVlAAAAwKUdOWK790Jamu1xgwbSzJlSq1aurQsAAOAPRtO2qpo9W3r8cdvv9epJH34o3Xaba2sCAAAAnHXmjC2/fved5OMjDR0q/e1vUmCgqysDAAD4w3F5hKqqSxfpppukQYOkzZtp2AIAAMCz+PvbrmF7883SihXS22/TsAUAAF6DM22rqtBQKSdHCglxdSUAAADA5Xn2WalnT6l6dVdXAgAAcE1xpm1VRsMWAAAAnszXl4YtAADwSjRtAQAAAAAAAMCN0LQFAAAAAAAAADdC0xYAAAAAAAAA3AhNWwAAAAAAAABwIzRtAQAAAAAAAMCN0LQFAAAAAAAAADdC0xYAAAAAAAAA3AhNWwAAAAAAAABwIzRtAQAAAAAAAMCN0LQFAAAAAAAAADdC0xYAAAAAAAAA3AhNWwAAAAAAAABwIzRtAQAAAAAAAMCN0LQFAAAAAAAAADdC0xYAAAAAAAAA3Ii/qwvwJsYYSVJRUZGLKwEAAPBM5CjXIs8CAABcmbIcVZarKkLT9hoqLi6WJEVHR7u4EgAAAKDyyLMAAABXR3FxscLCwirc72Mu1dbFVVNaWqqff/5ZoaGh8vHxcXU5lVJUVKTo6Gjt27dPVqvV1eXgMrCGno3182ysn+djDd1HWXS1Wq0el6eqAk/OsxKfZU/H+nk21s/zsYaejfVzH8YYFRcXq3bt2vL1rfjKtZxpew35+vqqTp06ri7jilitVj7cHo419Gysn2dj/TwfawhvVxXyrMRn2dOxfp6N9fN8rKFnY/3cw8XOsC3DjcgAAAAAAAAAwI3QtAUAAAAAAAAAN0LTFk6xWCwaOXKkLBaLq0vBZWINPRvr59lYP8/HGgJVA59lz8b6eTbWz/Oxhp6N9fM83IgMAAAAAAAAANwIZ9oCAAAAAAAAgBuhaQsAAAAAAAAAboSmLQAAAAAAAAC4EZq2cMrkyZMVGxurwMBAJSYmasOGDa4uqcpZsWKFHnjgAdWuXVs+Pj5auHChw35jjEaMGKEbbrhBQUFBSk5OVn5+vsOYgwcPKjU1VVarVeHh4erdu7dKSkocxmzZskVt2rRRYGCgoqOj9dZbb11Qy/z583XLLbcoMDBQjRs31qJFiypdi7cZPXq0WrdurdDQUNWqVUsPPvig8vLyHMacOHFC/fv313XXXaeQkBA9/PDDKiwsdBizd+9e3X///QoODlatWrX0/PPP68yZMw5jli9frhYtWshisSguLk7p6ekX1HOpz6wztXiT9957T02aNJF5i0MCAAASz0lEQVTVapXValVSUpK++uor+37WzrOMGTNGPj4+Gjx4sH0bawiAPPvHI896NvKs5yPTVi1kWsgAlzB37lwTEBBgZsyYYbZt22b69u1rwsPDTWFhoatLq1IWLVpkXn75ZfOvf/3LSDILFixw2D9mzBgTFhZmFi5caHJyckyXLl1M3bp1zfHjx+1jOnToYJo2bWrWrVtnVq5caeLi4syjjz5q33/kyBETGRlpUlNTTW5urpkzZ44JCgoyU6dOtY9ZvXq18fPzM2+99ZbZvn27eeWVV0y1atXM1q1bK1WLt0lJSTFpaWkmNzfXZGdnm06dOpmYmBhTUlJiH9OvXz8THR1tMjIyzLfffmtuu+02c/vtt9v3nzlzxiQkJJjk5GSzefNms2jRIlOzZk3z4osv2sf88MMPJjg42AwZMsRs377dTJo0yfj5+ZnFixfbxzjzmb1ULd7ms88+M19++aXZuXOnycvLMy+99JKpVq2ayc3NNcawdp5kw4YNJjY21jRp0sQMGjTIvp01BLwbefbaIM96NvKs5yPTVh1kWhhjDE1bXNKtt95q+vfvb3989uxZU7t2bTN69GgXVlW1nR9yS0tLTVRUlHn77bft2w4fPmwsFouZM2eOMcaY7du3G0lm48aN9jFfffWV8fHxMT/99JMxxpgpU6aYiIgIc/LkSfuY4cOHm/r169sfP/LII+b+++93qCcxMdE8/fTTTtcCYw4cOGAkmczMTGOMbY6qVatm5s+fbx+zY8cOI8msXbvWGGP7h46vr68pKCiwj3nvvfeM1Wq1r9mwYcNMo0aNHF6rW7duJiUlxf74Up9ZZ2qBMREREWbatGmsnQcpLi428fHxZsmSJaZt27b2gMsaAiDPXnvkWc9Hnq0ayLSeh0yLMlweARd16tQpZWVlKTk52b7N19dXycnJWrt2rQsr8y67d+9WQUGBwzqEhYUpMTHRvg5r165VeHi4WrVqZR+TnJwsX19frV+/3j7mrrvuUkBAgH1MSkqK8vLydOjQIfuYc1+nbEzZ6zhTC6QjR45IkmrUqCFJysrK0unTpx3m7ZZbblFMTIzDGjZu3FiRkZH2MSkpKSoqKtK2bdvsYy62Ps58Zp2pxZudPXtWc+fO1dGjR5WUlMTaeZD+/fvr/vvvv2CeWUPAu5Fn3QN51vOQZz0bmdZzkWlRxt/VBcC9/frrrzp79qzDB16SIiMj9d1337moKu9TUFAgSeWuQ9m+goIC1apVy2G/v7+/atSo4TCmbt26FxyjbF9ERIQKCgou+TqXqsXblZaWavDgwbrjjjuUkJAgyTZvAQEBCg8Pdxh7/tyWN69l+y42pqioSMePH9ehQ4cu+Zl1phZvtHXrViUlJenEiRMKCQnRggUL1LBhQ2VnZ7N2HmDu3LnatGmTNm7ceME+Pn+AdyPPugfyrGchz3ouMq1nI9PiXDRtAeAq69+/v3Jzc7Vq1SpXl4JKqF+/vrKzs3XkyBF98skn6tGjhzIzM11dFpywb98+DRo0SEuWLFFgYKCrywEAwOORZz0XmdZzkWlxPi6PgIuqWbOm/Pz8LrgDYGFhoaKiolxUlfcpm+uLrUNUVJQOHDjgsP/MmTM6ePCgw5jyjnHua1Q05tz9l6rFmw0YMEBffPGFli1bpjp16ti3R0VF6dSpUzp8+LDD+PPn9nLXx2q1KigoyKnPrDO1eKOAgADFxcWpZcuWGj16tJo2baoJEyawdh4gKytLBw4cUIsWLeTv7y9/f39lZmZq4sSJ8vf3V2RkJGsIeDHyrHsgz3oO8qxnI9N6LjItzkfTFhcVEBCgli1bKiMjw76ttLRUGRkZSkpKcmFl3qVu3bqKiopyWIeioiKtX7/evg5JSUk6fPiwsrKy7GOWLl2q0tJSJSYm2sesWLFCp0+fto9ZsmSJ6tevr4iICPuYc1+nbEzZ6zhTizcyxmjAgAFasGCBli5desHX9lq2bKlq1ao5zFteXp727t3rsIZbt251+MfKkiVLZLVa1bBhQ/uYi62PM59ZZ2qBbd5OnjzJ2nmA9u3ba+vWrcrOzrb/tGrVSqmpqfbfWUPAe5Fn3QN51v2RZ6smMq3nINPiAq6+Exrc39y5c43FYjHp6elm+/bt5qmnnjLh4eEOdyPElSsuLjabN282mzdvNpLMuHHjzObNm82PP/5ojDFmzJgxJjw83Hz66admy5YtpmvXrqZu3brm+PHj9mN06NDBNG/e3Kxfv96sWrXKxMfHm0cffdS+//DhwyYyMtJ0797d5Obmmrlz55rg4GAzdepU+5jVq1cbf39/884775gdO3aYkSNHmmrVqpmtW7faxzhTi7d55plnTFhYmFm+fLnZv3+//efYsWP2Mf369TMxMTFm6dKl5ttvvzVJSUkmKSnJvv/MmTMmISHB3HfffSY7O9ssXrzYXH/99ebFF1+0j/nhhx9McHCwef75582OHTvM5MmTjZ+fn1m8eLF9jDOf2UvV4m1eeOEFk5mZaXbv3m22bNliXnjhBePj42O+/vprYwxr54nOvdOuMawh4O3Is9cGedazkWc9H5m26iHTejeatnDKpEmTTExMjAkICDC33nqrWbdunatLqnKWLVtmJF3w06NHD2OMMaWlpebVV181kZGRxmKxmPbt25u8vDyHY/z222/m0UcfNSEhIcZqtZqePXua4uJihzE5OTnmzjvvNBaLxdx4441mzJgxF9Ty8ccfm3r16pmAgADTqFEj8+WXXzrsd6YWb1Pe2kkyaWlp9jHHjx83zz77rImIiDDBwcHmz3/+s9m/f7/Dcfbs2WM6duxogoKCTM2aNc3QoUPN6dOnHcYsW7bMNGvWzAQEBJibbrrJ4TXKXOoz60wt3qRXr17mT3/6kwkICDDXX3+9ad++vT3cGsPaeaLzAy5rCIA8+8cjz3o28qznI9NWPWRa7+ZjjDHX7rxeAAAAAAAAAMDFcE1bAAAAAAAAAHAjNG0BAAAAAAAAwI3QtAUAAAAAAAAAN0LTFgAAAAAAAADcCE1bAAAAAAAAAHAjNG0BAAAAAAAAwI3QtAUAAAAAAAAAN0LTFgAAAAAAAADcCE1bAAAAAAAAAHAjNG0BVDlPPPGEHnjgAVeXAQAAAFwW8iwAgKYtgCph27Zt6tatm+rUqaNZs2bpiy++UGhoqDp27KglS5a4ujwAAADgosizAIBz0bQF4PEWLFigpk2b6uTJk/roo4/0yCOPqEOHDvrqq68UFRWl++67T5MnT7aP37hxo+69917VrFlTYWFhatu2rTZt2uRwTB8fHy1cuFCSZIzRE088oSZNmujQoUNKT0+Xj49PuT+xsbGSpNdee03NmjWzH+/UqVOKi4uTj4+PDh8+LEl68skn9eCDD1b4upK0b98+PfLIIwoPD1eNGjXUtWtX7dmzx+E5M2bMUKNGjWSxWHTDDTdowIABTr0PZ+uUpFWrVqlNmzYKCgpSdHS0Bg4cqKNHj1a4JhMmTFBMTIwsFosiIyPVp08fHTt2TJK0Z88e+fj4KDs72+E5sbGxGj9+vP3xuHHj1LhxY1WvXl3R0dF69tlnVVJSYt9/NeavvGOkp6crPDzc/tjZOUpLS1P9+vUVEBBg/+9h8ODBFc4RAABAGfIsefZy5488C1RdNG0BeLzBgwfr7rvv1sKFC3X33XcrKChIFotFd955p9LS0vTkk09q2LBh9lBWXFysHj16aNWqVVq3bp3i4+PVqVMnFRcXl3v8gQMHas2aNfr6668VERGhbt26af/+/dq/f7/Gjx+vOnXq2B9v3Lix3GO8++67KiwsrNT7On36tFJSUhQaGqqVK1dq9erVCgkJUYcOHXTq1ClJ0nvvvaf+/fvrqaee0tatW/XZZ58pLi7OqffhbJ3ff/+9OnTooIcfflhbtmzRvHnztGrVKocwfb5bb71V8+fPV35+vj755BNlZGTonXfeqdT79/X11cSJE7Vt2zZ9+OGHWrp0qYYNG+b0852Zv8tR3hx999136tOnj3r16qVdu3Zp//79SkpKuuzXAAAA3oU8S54tD3kW8G40bQF4tMLCQu3du1edO3eucEyXLl107Ngx5ebmSpLuuecePf7447rlllvUoEEDffDBBzp27JgyMzMveO4rr7yiBQsW6JtvvlFUVJQkKSgoSFFRUYqKilJYWJj8/Pzsj6+//voLjnHw4EGNGjVKw4cPd9geFBSk48ePV1j3vHnzVFpaqmnTpqlx48Zq0KCB0tLStHfvXi1fvlySNGrUKA0dOlSDBg1SvXr11Lp163L/Gl7e+3C2ztGjRys1NVWDBw9WfHy8br/9dk2cOFEzZ87UiRMnyj1WUlKSEhMTFRMTo/r168tqters2bMVvtfyDB48WO3atVNsbKzuuecejRo1Sh9//LF9/9WYv8qqaI62bNkiPz8/DR8+XDExMYqKilJAQMBlvQYAAPAu5FnybEXIs4B3o2kLwKOVBYmyryqVp2xfYGCgJFsw7tu3r+Lj4xUWFiar1aqSkhLt3bvX4Xnvvvuu3njjDdWvX9/+NbHL8frrr6tdu3a68847HbYnJCRo3bp12r17d7nPy8nJ0a5duxQaGqqQkBCFhISoRo0aOnHihL7//nsdOHBAP//8s9q3b3/R13f2fVRUZ05OjtLT0+01hISEKCUlRaWlpRXWLkmzZ89W9erVFRkZqfj4+AuC4e233+5wzPPn/5tvvlH79u114403KjQ0VN27d9dvv/1mX88rnb8yX3zxhUMd/fr1q/Qc1a1bV6dPn9b8+fNljKnw+QAAAOcjz5JnybMAykPTFoBHi4iIUGJiombOnFnuNanOnDmjqVOnqk6dOkpISJAk9ejRQ9nZ2ZowYYLWrFmj7OxsXXfddRd8xWjDhg1atGiRcnNzNXXq1MuqLz8/X9OmTdObb755wb5evXqpdevWuummm+wB61wlJSVq2bKlsrOzHX527typxx57TEFBQU7V4Mz7uFidJSUlevrppx1qyMnJUX5+vm6++eYKX7dLly7avHmzPv30U61fv14LFixw2D9v3jyHY9auXdu+b8+ePercubOaNGmif/7zn8rKyrJfx61sna50/sq0a9fOYf/rr79e6Tlq3bq1Xn/9dfXs2VOBgYEKCQnRypUrK5wbAACAMuTZSyPPkmcBb+Tv6gIA4EpNmzZNnTt3VoMGDdS7d2/t3r1bx44d09///nfNnDlTBw4c0MKFC+Xn5ydJWr16taZMmaJOnTpJsl3c/9dff73guOPHj1fHjh01ZcoU9ezZUx07dlRMTEylahs+fLj69OmjuLg4/fvf/3bYFxQUpG+++UaFhYX264/Fx8fb97do0ULz5s1TrVq1ZLVayz1+bGysMjIy1K5duwprcOZ9XKzOFi1aaPv27RVeW6wioaGhCg0NVb169bRs2TLNmTNHqamp9v3R0dEOx/T3//1/SVlZWSotLdXYsWPl62v7++K5XyWTrs78SVL16tUd6qhVq1a54y42R5LtGmszZ85U79699Ze//MXhvQIAAFwMeZY8S54FcD7OtAXg8RISEpSXl6eXXnpJ+fn52rFjh3bt2qW1a9eqV69eysvL01133WUfHx8fr1mzZmnHjh1av369UlNTy/0rf40aNSRJDz/8sDp16qQ+ffpUqq5du3Zp+fLlGjFixEXHRUZGKi4u7oIQmZqaqpo1a6pr165auXKldu/ereXLl2vgwIH2kPXaa69p7NixmjhxovLz87Vp0yZNmjSpUu/jUnUOHz5ca9as0YABA5Sdna38/Hx9+umnF71xQ1pamnJycvTjjz/qs88+05w5c9S8efOLzsO54uLidPr0aU2aNEk//PCDZs2apffff7/csVcyf8661ByV3cm4RYsWeuGFFxQXF+f0mSMAAADkWfIseRbA+WjaAqgSLBaL+vXrp48++kidOnVS27Zt9fnnn2vYsGEX3Exh+vTpOnTokFq0aKHu3btr4MCBFf41usy7776rnJwcffDBB07XdPToUb388sv2kFlZwcHBWrFihWJiYvTQQw/Zz7w4ceKE/S/tPXr00Pjx4zVlyhQ1atRInTt3Vn5+fqXex6XqbNKkiTIzM7Vz5061adNGzZs314gRIxy+/nW+tWvXqkOHDqpXr56ee+45paam6tVXX3X6vTdt2lTjxo3Tm2++qYSEBM2ePVujR492+vmSc/PnrEvN0ZgxY5Sfn6/p06dX6rgAAABlyLPk2fORZwHv5mO4wjQAAAAAAAAAuA3OtAUAAAAAAAAAN0LTFgAAAAAAAADcCE1bAAAAAAAAAHAjNG0BAAAAAAAAwI3QtAUAAAAAAAAAN0LTFgAAAAAAAADcCE1bAAAAAAAAAHAjNG0BAAAAAAAAwI3QtAUAAAAAAAAAN0LTFgAAAAAAAADcCE1bAAAAAAAAAHAjNG0BAAAAAAAAwI38P567tRDUaYPKAAAAAElFTkSuQmCC", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "# Визуальное сравнение факта и прогноза\n", "fig, axes = plt.subplots(1, 2, figsize=(14, 5))\n", "\n", "axes[0].scatter(y_test, baseline_pred_test, alpha=0.5)\n", "axes[0].plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()], 'r--')\n", "axes[0].set_title('Классическая нейросеть')\n", "axes[0].set_xlabel('Фактические значения')\n", "axes[0].set_ylabel('Предсказания')\n", "\n", "axes[1].scatter(y_test, searched_pred_test, alpha=0.5)\n", "axes[1].plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()], 'r--')\n", "axes[1].set_title('Нейросеть с архитектурой, найденной DE')\n", "axes[1].set_xlabel('Фактические значения')\n", "axes[1].set_ylabel('Предсказания')\n", "\n", "plt.tight_layout()\n", "plt.show()\n" ] }, { "cell_type": "code", "execution_count": 18, "id": "32e5b8dc", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "=== КЛЮЧЕВАЯ ИНФОРМАЦИЯ ДЛЯ ВЫВОДА ===\n", "Зависимая переменная: price\n", "\n", "Число исходных признаков: 5\n", "Число признаков после кодирования: 7\n", "Фиксированная архитектура: (64, 32)\n", "Архитектура, найденная дифференциальной эволюцией: (80, 96)\n", "alpha найденной модели: 1.8274889467365341e-06\n", "learning_rate_init найденной модели: 0.00042874921144929587\n", "Использовано log1p-преобразование цели: False\n", "\n", "Метрики моделей на тестовой выборке:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
modelr2_testrmse_testmae_testr2_valrmse_valmae_valn_original_featuresn_processed_featureshidden_layersalphalearning_rate_init
0Классическая нейросеть0.57400249449.05878539483.8251530.57263849821.04593539964.57365457(64, 32)NaNNaN
1Нейросеть с архитектурой, найденной DE0.57320149495.53441839527.8663140.57299549800.26001039955.01779157(80, 96)0.0000020.000429
\n", "
" ], "text/plain": [ " model r2_test rmse_test \\\n", "0 Классическая нейросеть 0.574002 49449.058785 \n", "1 Нейросеть с архитектурой, найденной DE 0.573201 49495.534418 \n", "\n", " mae_test r2_val rmse_val mae_val n_original_features \\\n", "0 39483.825153 0.572638 49821.045935 39964.573654 5 \n", "1 39527.866314 0.572995 49800.260010 39955.017791 5 \n", "\n", " n_processed_features hidden_layers alpha learning_rate_init \n", "0 7 (64, 32) NaN NaN \n", "1 7 (80, 96) 0.000002 0.000429 " ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Лучшая модель:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
modelr2_testrmse_testmae_testr2_valrmse_valmae_valn_original_featuresn_processed_featureshidden_layersalphalearning_rate_init
0Классическая нейросеть0.57400249449.05878539483.8251530.57263849821.04593539964.57365457(64, 32)NaNNaN
\n", "
" ], "text/plain": [ " model r2_test rmse_test mae_test r2_val \\\n", "0 Классическая нейросеть 0.574002 49449.058785 39483.825153 0.572638 \n", "\n", " rmse_val mae_val n_original_features n_processed_features \\\n", "0 49821.045935 39964.573654 5 7 \n", "\n", " hidden_layers alpha learning_rate_init \n", "0 (64, 32) NaN NaN " ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Топ-10 конфигураций по результатам эволюционного поиска:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
hidden_layersalphalearning_rate_initrmse_valmae_valr2_val
0(80, 96)0.0000020.00042949520.45809139914.3894770.570645
1(128,)0.0000930.00019249584.79976840046.1962760.569528
2(120,)0.0000460.00111249650.09369539974.6919250.568394
3(64,)0.0002860.00428149655.42698539866.6043770.568301
4(88,)0.0001630.00010149671.96210739922.9692430.568013
5(112, 16)0.0084730.00023749693.86517440039.1385920.567632
6(64,)0.0000040.00029549699.37875439982.0231500.567536
7(64,)0.0000130.00153849703.25761939942.6915780.567469
8(120, 96, 8, 56)0.0013320.00071149712.48979540074.5857690.567308
9(112,)0.0007750.00020649712.82364140067.8064040.567302
\n", "
" ], "text/plain": [ " hidden_layers alpha learning_rate_init rmse_val mae_val \\\n", "0 (80, 96) 0.000002 0.000429 49520.458091 39914.389477 \n", "1 (128,) 0.000093 0.000192 49584.799768 40046.196276 \n", "2 (120,) 0.000046 0.001112 49650.093695 39974.691925 \n", "3 (64,) 0.000286 0.004281 49655.426985 39866.604377 \n", "4 (88,) 0.000163 0.000101 49671.962107 39922.969243 \n", "5 (112, 16) 0.008473 0.000237 49693.865174 40039.138592 \n", "6 (64,) 0.000004 0.000295 49699.378754 39982.023150 \n", "7 (64,) 0.000013 0.001538 49703.257619 39942.691578 \n", "8 (120, 96, 8, 56) 0.001332 0.000711 49712.489795 40074.585769 \n", "9 (112,) 0.000775 0.000206 49712.823641 40067.806404 \n", "\n", " r2_val \n", "0 0.570645 \n", "1 0.569528 \n", "2 0.568394 \n", "3 0.568301 \n", "4 0.568013 \n", "5 0.567632 \n", "6 0.567536 \n", "7 0.567469 \n", "8 0.567308 \n", "9 0.567302 " ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Последние эпохи обучения классической нейросети:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_loss
15160.214724
16170.215112
17180.214927
18190.214795
19200.214660
20210.214485
21220.214921
22230.215026
23240.214648
24250.214526
\n", "
" ], "text/plain": [ " epoch train_loss\n", "15 16 0.214724\n", "16 17 0.215112\n", "17 18 0.214927\n", "18 19 0.214795\n", "19 20 0.214660\n", "20 21 0.214485\n", "21 22 0.214921\n", "22 23 0.215026\n", "23 24 0.214648\n", "24 25 0.214526" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Последние эпохи обучения модели с найденной архитектурой:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_loss
31320.215046
32330.215172
33340.214974
34350.215040
35360.214977
36370.214963
37380.214892
38390.214782
39400.214849
40410.214744
\n", "
" ], "text/plain": [ " epoch train_loss\n", "31 32 0.215046\n", "32 33 0.215172\n", "33 34 0.214974\n", "34 35 0.215040\n", "35 36 0.214977\n", "36 37 0.214963\n", "37 38 0.214892\n", "38 39 0.214782\n", "39 40 0.214849\n", "40 41 0.214744" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Первые 10 фактических и предсказанных значений на тесте для классической нейросети:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
y_truey_pred_baseline
0170835.035713221375.864995
1126913.469998132902.709126
2246611.883092255392.811125
3244250.462969263089.260988
4271127.650112274352.654928
5189289.951199205015.181918
6343510.576500279901.749737
7203461.591080198159.696005
8285903.247682305626.569686
9224868.958375216284.347016
\n", "
" ], "text/plain": [ " y_true y_pred_baseline\n", "0 170835.035713 221375.864995\n", "1 126913.469998 132902.709126\n", "2 246611.883092 255392.811125\n", "3 244250.462969 263089.260988\n", "4 271127.650112 274352.654928\n", "5 189289.951199 205015.181918\n", "6 343510.576500 279901.749737\n", "7 203461.591080 198159.696005\n", "8 285903.247682 305626.569686\n", "9 224868.958375 216284.347016" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Первые 10 фактических и предсказанных значений на тесте для модели с найденной архитектурой:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
y_truey_pred_de
0170835.035713215656.937397
1126913.469998135838.216645
2246611.883092252148.546000
3244250.462969257517.641597
4271127.650112272991.652689
5189289.951199204442.641467
6343510.576500275099.596427
7203461.591080198134.278779
8285903.247682311642.805088
9224868.958375215296.762920
\n", "
" ], "text/plain": [ " y_true y_pred_de\n", "0 170835.035713 215656.937397\n", "1 126913.469998 135838.216645\n", "2 246611.883092 252148.546000\n", "3 244250.462969 257517.641597\n", "4 271127.650112 272991.652689\n", "5 189289.951199 204442.641467\n", "6 343510.576500 275099.596427\n", "7 203461.591080 198134.278779\n", "8 285903.247682 311642.805088\n", "9 224868.958375 215296.762920" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "# Финальная ячейка с ключевой информацией для вывода\n", "print('=== КЛЮЧЕВАЯ ИНФОРМАЦИЯ ДЛЯ ВЫВОДА ===')\n", "print('Зависимая переменная:', TARGET_COL)\n", "print()\n", "print('Число исходных признаков:', X.shape[1])\n", "print('Число признаков после кодирования:', len(feature_names_processed))\n", "print('Фиксированная архитектура:', fixed_hidden_layers)\n", "print('Архитектура, найденная дифференциальной эволюцией:', best_config['hidden_layers'])\n", "print('alpha найденной модели:', best_config['alpha'])\n", "print('learning_rate_init найденной модели:', best_config['learning_rate_init'])\n", "print('Использовано log1p-преобразование цели:', use_log1p_target)\n", "print()\n", "print('Метрики моделей на тестовой выборке:')\n", "display(results_df)\n", "print('Лучшая модель:')\n", "display(best_model_row)\n", "print('Топ-10 конфигураций по результатам эволюционного поиска:')\n", "display(search_history_df.head(10))\n", "print('Последние эпохи обучения классической нейросети:')\n", "display(baseline_history.tail(10))\n", "print('Последние эпохи обучения модели с найденной архитектурой:')\n", "display(searched_history.tail(10))\n", "print('Первые 10 фактических и предсказанных значений на тесте для классической нейросети:')\n", "display(pd.DataFrame({'y_true': y_test.values[:10], 'y_pred_baseline': baseline_pred_test[:10]}))\n", "print('Первые 10 фактических и предсказанных значений на тесте для модели с найденной архитектурой:')\n", "display(pd.DataFrame({'y_true': y_test.values[:10], 'y_pred_de': searched_pred_test[:10]}))\n" ] }, { "cell_type": "markdown", "id": "283bec8d", "metadata": {}, "source": [ "Итог\n", "\n", "Зависимой переменной выбрана price. Сравнивались две нейросети: модель с фиксированной архитектурой (64, 32) и модель с архитектурой, найденной методом дифференциальной эволюции ((80, 96)).\n", "\n", "На тестовой выборке лучшая модель — классическая нейросеть: R² = 0.5740, RMSE ≈ 49449, MAE ≈ 39484. Модель с найденной архитектурой показала практически тот же результат, но немного хуже (R² = 0.5732, RMSE ≈ 49496, MAE ≈ 39528).\n", "\n", "Следовательно, эволюционный поиск архитектуры не дал улучшения качества, и лучшей следует считать нейросеть с заранее заданной архитектурой." ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "name": "python" } }, "nbformat": 4, "nbformat_minor": 5 }