{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "7e8eef48",
   "metadata": {},
   "source": [
    "# Lesson 21: Ensemble learning demonstration\n",
    "\n",
    "This notebook demonstrates key concepts and tools for training ensemble models.\n",
    "\n",
    "1. Baseline models\n",
    "    - Logistic regression\n",
    "    - Decision tree\n",
    "2. Parallel ensembles\n",
    "    - Voting ensemble\n",
    "    - Bagging ensemble\n",
    "    - Random forest\n",
    "3. Serial (sequential) ensembles\n",
    "    - AdaBoost\n",
    "    - Gradient boosting\n",
    "    - Stacking ensemble\n",
    "4. Model comparison\n",
    "    - Score comparison\n",
    "    - Confusion matrix comparison\n",
    "5. Model metric optimization\n",
    "    - ROC_AUC optimized thresholds\n",
    "    - F1 optimized thresholds\n",
    "\n",
    "## Notebook set up\n",
    "\n",
    "### Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "463812a9",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n",
    "from sklearn.datasets import make_classification\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import ConfusionMatrixDisplay\n",
    "from sklearn.model_selection import train_test_split, cross_validate, TunedThresholdClassifierCV\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "# Ensemble models\n",
    "from sklearn.ensemble import (\n",
    "    AdaBoostClassifier,\n",
    "    BaggingClassifier,\n",
    "    GradientBoostingClassifier,\n",
    "    RandomForestClassifier,\n",
    "    StackingClassifier,\n",
    "    VotingClassifier,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7012bb8b",
   "metadata": {},
   "source": [
    "### Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "60a2f9f5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>feature_0</th>\n",
       "      <th>feature_1</th>\n",
       "      <th>feature_2</th>\n",
       "      <th>feature_3</th>\n",
       "      <th>feature_4</th>\n",
       "      <th>feature_5</th>\n",
       "      <th>feature_6</th>\n",
       "      <th>feature_7</th>\n",
       "      <th>feature_8</th>\n",
       "      <th>feature_9</th>\n",
       "      <th>...</th>\n",
       "      <th>feature_11</th>\n",
       "      <th>feature_12</th>\n",
       "      <th>feature_13</th>\n",
       "      <th>feature_14</th>\n",
       "      <th>feature_15</th>\n",
       "      <th>feature_16</th>\n",
       "      <th>feature_17</th>\n",
       "      <th>feature_18</th>\n",
       "      <th>feature_19</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>6850</th>\n",
       "      <td>2.056957</td>\n",
       "      <td>0.222830</td>\n",
       "      <td>-1.914398</td>\n",
       "      <td>1.095708</td>\n",
       "      <td>-2.389691</td>\n",
       "      <td>-1.113905</td>\n",
       "      <td>1.296463</td>\n",
       "      <td>-5.769138</td>\n",
       "      <td>-1.742394</td>\n",
       "      <td>-1.717438</td>\n",
       "      <td>...</td>\n",
       "      <td>0.903576</td>\n",
       "      <td>-1.730474</td>\n",
       "      <td>-0.211586</td>\n",
       "      <td>0.402758</td>\n",
       "      <td>-2.171021</td>\n",
       "      <td>1.242654</td>\n",
       "      <td>-0.993714</td>\n",
       "      <td>1.083335</td>\n",
       "      <td>1.735312</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5856</th>\n",
       "      <td>1.776417</td>\n",
       "      <td>-3.711362</td>\n",
       "      <td>-0.042654</td>\n",
       "      <td>-4.450805</td>\n",
       "      <td>0.283605</td>\n",
       "      <td>0.062951</td>\n",
       "      <td>-2.952074</td>\n",
       "      <td>-4.635668</td>\n",
       "      <td>0.133406</td>\n",
       "      <td>-1.665485</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.160329</td>\n",
       "      <td>1.273855</td>\n",
       "      <td>-2.754988</td>\n",
       "      <td>-5.904544</td>\n",
       "      <td>-0.580578</td>\n",
       "      <td>0.952436</td>\n",
       "      <td>-0.566598</td>\n",
       "      <td>-1.590927</td>\n",
       "      <td>0.632844</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2483</th>\n",
       "      <td>-0.017753</td>\n",
       "      <td>1.031828</td>\n",
       "      <td>-2.485858</td>\n",
       "      <td>-0.667876</td>\n",
       "      <td>-1.021137</td>\n",
       "      <td>-0.692445</td>\n",
       "      <td>0.630700</td>\n",
       "      <td>2.300769</td>\n",
       "      <td>-1.097844</td>\n",
       "      <td>0.717599</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.728046</td>\n",
       "      <td>0.328055</td>\n",
       "      <td>0.403882</td>\n",
       "      <td>1.735435</td>\n",
       "      <td>0.162798</td>\n",
       "      <td>-1.306485</td>\n",
       "      <td>-0.760256</td>\n",
       "      <td>0.780826</td>\n",
       "      <td>-1.476193</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4915</th>\n",
       "      <td>-0.495843</td>\n",
       "      <td>2.895735</td>\n",
       "      <td>-2.537299</td>\n",
       "      <td>-2.609338</td>\n",
       "      <td>-0.875950</td>\n",
       "      <td>2.935914</td>\n",
       "      <td>3.253340</td>\n",
       "      <td>6.378967</td>\n",
       "      <td>-0.058042</td>\n",
       "      <td>-1.338881</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.579209</td>\n",
       "      <td>1.383636</td>\n",
       "      <td>0.879382</td>\n",
       "      <td>2.941039</td>\n",
       "      <td>-0.171335</td>\n",
       "      <td>0.508342</td>\n",
       "      <td>-0.704955</td>\n",
       "      <td>1.260240</td>\n",
       "      <td>-2.110639</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4598</th>\n",
       "      <td>0.899913</td>\n",
       "      <td>-0.241683</td>\n",
       "      <td>-1.206857</td>\n",
       "      <td>-4.108075</td>\n",
       "      <td>0.786454</td>\n",
       "      <td>-0.923499</td>\n",
       "      <td>-1.404900</td>\n",
       "      <td>-6.952240</td>\n",
       "      <td>-1.282080</td>\n",
       "      <td>-1.646648</td>\n",
       "      <td>...</td>\n",
       "      <td>3.731664</td>\n",
       "      <td>1.857395</td>\n",
       "      <td>-0.758554</td>\n",
       "      <td>4.081466</td>\n",
       "      <td>2.046172</td>\n",
       "      <td>0.938040</td>\n",
       "      <td>-0.088168</td>\n",
       "      <td>-0.283488</td>\n",
       "      <td>0.257320</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      feature_0  feature_1  feature_2  feature_3  feature_4  feature_5  \\\n",
       "6850   2.056957   0.222830  -1.914398   1.095708  -2.389691  -1.113905   \n",
       "5856   1.776417  -3.711362  -0.042654  -4.450805   0.283605   0.062951   \n",
       "2483  -0.017753   1.031828  -2.485858  -0.667876  -1.021137  -0.692445   \n",
       "4915  -0.495843   2.895735  -2.537299  -2.609338  -0.875950   2.935914   \n",
       "4598   0.899913  -0.241683  -1.206857  -4.108075   0.786454  -0.923499   \n",
       "\n",
       "      feature_6  feature_7  feature_8  feature_9  ...  feature_11  feature_12  \\\n",
       "6850   1.296463  -5.769138  -1.742394  -1.717438  ...    0.903576   -1.730474   \n",
       "5856  -2.952074  -4.635668   0.133406  -1.665485  ...   -1.160329    1.273855   \n",
       "2483   0.630700   2.300769  -1.097844   0.717599  ...   -0.728046    0.328055   \n",
       "4915   3.253340   6.378967  -0.058042  -1.338881  ...   -1.579209    1.383636   \n",
       "4598  -1.404900  -6.952240  -1.282080  -1.646648  ...    3.731664    1.857395   \n",
       "\n",
       "      feature_13  feature_14  feature_15  feature_16  feature_17  feature_18  \\\n",
       "6850   -0.211586    0.402758   -2.171021    1.242654   -0.993714    1.083335   \n",
       "5856   -2.754988   -5.904544   -0.580578    0.952436   -0.566598   -1.590927   \n",
       "2483    0.403882    1.735435    0.162798   -1.306485   -0.760256    0.780826   \n",
       "4915    0.879382    2.941039   -0.171335    0.508342   -0.704955    1.260240   \n",
       "4598   -0.758554    4.081466    2.046172    0.938040   -0.088168   -0.283488   \n",
       "\n",
       "      feature_19  target  \n",
       "6850    1.735312       0  \n",
       "5856    0.632844       0  \n",
       "2483   -1.476193       0  \n",
       "4915   -2.110639       0  \n",
       "4598    0.257320       0  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Generate synthetic dataset, similar to Lesson 20: classification but \n",
    "# more challenging with: more features, redundant features, and noise\n",
    "data = make_classification(\n",
    "    n_samples=10000,\n",
    "    n_features=20,\n",
    "    n_informative=10,\n",
    "    n_redundant=5,\n",
    "    n_clusters_per_class=3,\n",
    "    weights=[0.9, 0.1],\n",
    "    flip_y=0.1,\n",
    "    class_sep=0.5,\n",
    "    random_state=315\n",
    ")\n",
    "\n",
    "# Create dataframe with feature columns\n",
    "df = pd.DataFrame(data[0], columns=[f'feature_{i}' for i in range(data[0].shape[1])])\n",
    "df['target'] = data[1]\n",
    "\n",
    "# Split into train (75%) and test (25%) sets\n",
    "train_df, test_df = train_test_split(df, random_state=315)\n",
    "train_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bfa06f29",
   "metadata": {},
   "source": [
    "## 1. Baseline models\n",
    "\n",
    "Scikit-learn [Logistic Regression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "db77210f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Logistic Regression - AUC: 0.663 (+/- 0.026), F1: 0.305 (+/- 0.016)\n"
     ]
    }
   ],
   "source": [
    "# Logistic regression baseline\n",
    "lr_clf = LogisticRegression(random_state=315, max_iter=1000, class_weight='balanced')\n",
    "\n",
    "lr_scores = cross_validate(\n",
    "    lr_clf,\n",
    "    train_df.drop('target', axis=1),\n",
    "    train_df['target'],\n",
    "    cv=7,\n",
    "    scoring=['roc_auc', 'f1'],\n",
    "    return_train_score=False\n",
    ")\n",
    "\n",
    "lr_auc = lr_scores['test_roc_auc']\n",
    "lr_f1 = lr_scores['test_f1']\n",
    "\n",
    "print(f'Logistic Regression - AUC: {lr_auc.mean():.3f} (+/- {lr_auc.std():.3f}), F1: {lr_f1.mean():.3f} (+/- {lr_f1.std():.3f})')\n",
    "\n",
    "lr_clf.fit(train_df.drop('target', axis=1), train_df['target'])\n",
    "lr_pred = lr_clf.predict(test_df.drop('target', axis=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff1adac4",
   "metadata": {},
   "source": [
    "Scikit-learn [Decision Tree](https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html) implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d8697878",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Decision Tree - AUC: 0.600 (+/- 0.015), F1: 0.307 (+/- 0.025)\n"
     ]
    }
   ],
   "source": [
    "# Decision tree baseline\n",
    "dt_clf = DecisionTreeClassifier(random_state=315, class_weight='balanced')\n",
    "\n",
    "dt_scores = cross_validate(\n",
    "    dt_clf,\n",
    "    train_df.drop('target', axis=1),\n",
    "    train_df['target'],\n",
    "    cv=7,\n",
    "    scoring=['roc_auc', 'f1'],\n",
    "    return_train_score=False\n",
    ")\n",
    "\n",
    "dt_auc = dt_scores['test_roc_auc']\n",
    "dt_f1 = dt_scores['test_f1']\n",
    "\n",
    "print(f'Decision Tree - AUC: {dt_auc.mean():.3f} (+/- {dt_auc.std():.3f}), F1: {dt_f1.mean():.3f} (+/- {dt_f1.std():.3f})')\n",
    "\n",
    "dt_clf.fit(train_df.drop('target', axis=1), train_df['target'])\n",
    "dt_pred = dt_clf.predict(test_df.drop('target', axis=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0b7a6b8c",
   "metadata": {},
   "source": [
    "## 2. Parallel ensembles\n",
    "\n",
    "### 2.1. Voting ensemble\n",
    "\n",
    "Combines predictions from multiple independent models by majority vote (hard voting) or by averaging predicted probabilities (soft voting).\n",
    "\n",
    "Scikit-learn [Voting Classifier](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.VotingClassifier.html) implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8cf9ed05",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Voting - AUC: 0.764 (+/- 0.015), F1: 0.361 (+/- 0.022)\n"
     ]
    }
   ],
   "source": [
    "# Create voting classifier with soft voting\n",
    "voting_clf = VotingClassifier(\n",
    "    estimators=[\n",
    "        ('lr', LogisticRegression(random_state=315, max_iter=1000, class_weight='balanced')),\n",
    "        ('dt', DecisionTreeClassifier(random_state=315, class_weight='balanced')),\n",
    "        ('svm', SVC(probability=True, random_state=315, class_weight='balanced'))\n",
    "    ],\n",
    "    voting='soft'\n",
    ")\n",
    "\n",
    "voting_scores = cross_validate(\n",
    "    voting_clf,\n",
    "    train_df.drop('target', axis=1),\n",
    "    train_df['target'],\n",
    "    cv=7,\n",
    "    scoring=['roc_auc', 'f1'],\n",
    "    return_train_score=False\n",
    ")\n",
    "\n",
    "voting_auc = voting_scores['test_roc_auc']\n",
    "voting_f1 = voting_scores['test_f1']\n",
    "\n",
    "print(f'Voting - AUC: {voting_auc.mean():.3f} (+/- {voting_auc.std():.3f}), F1: {voting_f1.mean():.3f} (+/- {voting_f1.std():.3f})')\n",
    "\n",
    "voting_clf.fit(train_df.drop('target', axis=1), train_df['target'])\n",
    "voting_pred = voting_clf.predict(test_df.drop('target', axis=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f472d779",
   "metadata": {},
   "source": [
    "### 2.2. Bagging ensemble\n",
    "\n",
    "Trains multiple instances of the same model on different random subsets of the training data (with replacement), then averages their predictions to reduce variance.\n",
    "\n",
    "Scikit-learn [Bagging Classifier](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.BaggingClassifier.html) implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "2e226745",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Bagging - AUC: 0.783 (+/- 0.024), F1: 0.244 (+/- 0.034)\n"
     ]
    }
   ],
   "source": [
    "# Create bagging classifier with decision trees\n",
    "bagging_clf = BaggingClassifier(\n",
    "    estimator=DecisionTreeClassifier(random_state=315, class_weight='balanced'),\n",
    "    n_estimators=100,\n",
    "    random_state=315,\n",
    "    oob_score=True\n",
    ")\n",
    "\n",
    "bagging_scores = cross_validate(\n",
    "    bagging_clf,\n",
    "    train_df.drop('target', axis=1),\n",
    "    train_df['target'],\n",
    "    cv=7,\n",
    "    scoring=['roc_auc', 'f1'],\n",
    "    return_train_score=False\n",
    ")\n",
    "\n",
    "bagging_auc = bagging_scores['test_roc_auc']\n",
    "bagging_f1 = bagging_scores['test_f1']\n",
    "\n",
    "print(f'Bagging - AUC: {bagging_auc.mean():.3f} (+/- {bagging_auc.std():.3f}), F1: {bagging_f1.mean():.3f} (+/- {bagging_f1.std():.3f})')\n",
    "\n",
    "bagging_clf.fit(train_df.drop('target', axis=1), train_df['target'])\n",
    "bagging_pred = bagging_clf.predict(test_df.drop('target', axis=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a11de28e",
   "metadata": {},
   "source": [
    "### 2.3. Random forest\n",
    "\n",
    "An extension of bagging that uses decision trees as base models and introduces additional randomness by considering only a random subset of features at each split.\n",
    "\n",
    "Scikit-learn [Random Forest Classifier](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html) implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "dcfadcb8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest - AUC: 0.792 (+/- 0.026), F1: 0.159 (+/- 0.013)\n"
     ]
    }
   ],
   "source": [
    "# Create random forest classifier\n",
    "rf_clf = RandomForestClassifier(\n",
    "    n_estimators=100,\n",
    "    random_state=315,\n",
    "    oob_score=True,\n",
    "    class_weight='balanced'\n",
    ")\n",
    "\n",
    "rf_scores = cross_validate(\n",
    "    rf_clf,\n",
    "    train_df.drop('target', axis=1),\n",
    "    train_df['target'],\n",
    "    cv=7,\n",
    "    scoring=['roc_auc', 'f1'],\n",
    "    return_train_score=False\n",
    ")\n",
    "\n",
    "rf_auc = rf_scores['test_roc_auc']\n",
    "rf_f1 = rf_scores['test_f1']\n",
    "\n",
    "print(f'Random Forest - AUC: {rf_auc.mean():.3f} (+/- {rf_auc.std():.3f}), F1: {rf_f1.mean():.3f} (+/- {rf_f1.std():.3f})')\n",
    "\n",
    "rf_clf.fit(train_df.drop('target', axis=1), train_df['target'])\n",
    "rf_pred = rf_clf.predict(test_df.drop('target', axis=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bd04cae2",
   "metadata": {},
   "source": [
    "## 3. Serial (sequential) ensembles\n",
    "\n",
    "### 3.1. AdaBoost\n",
    "\n",
    "Sequentially trains weak learners where each new model focuses more on the examples that previous models misclassified by adjusting sample weights.\n",
    "\n",
    "Scikit-learn [AdaBoost Classifier](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.AdaBoostClassifier.html) implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "25afd924",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AdaBoost - AUC: 0.684 (+/- 0.028), F1: 0.086 (+/- 0.022)\n"
     ]
    }
   ],
   "source": [
    "# Create AdaBoost classifier\n",
    "adaboost_clf = AdaBoostClassifier(\n",
    "    n_estimators=100,\n",
    "    random_state=315\n",
    ")\n",
    "\n",
    "adaboost_scores = cross_validate(\n",
    "    adaboost_clf,\n",
    "    train_df.drop('target', axis=1),\n",
    "    train_df['target'],\n",
    "    cv=7,\n",
    "    scoring=['roc_auc', 'f1'],\n",
    "    return_train_score=False\n",
    ")\n",
    "\n",
    "adaboost_auc = adaboost_scores['test_roc_auc']\n",
    "adaboost_f1 = adaboost_scores['test_f1']\n",
    "\n",
    "print(f'AdaBoost - AUC: {adaboost_auc.mean():.3f} (+/- {adaboost_auc.std():.3f}), F1: {adaboost_f1.mean():.3f} (+/- {adaboost_f1.std():.3f})')\n",
    "\n",
    "adaboost_clf.fit(train_df.drop('target', axis=1), train_df['target'])\n",
    "adaboost_pred = adaboost_clf.predict(test_df.drop('target', axis=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "78cb3b9b",
   "metadata": {},
   "source": [
    "### 3.2. Gradient boosting\n",
    "\n",
    "Builds models sequentially where each new model is trained to predict the residual errors of the previous ensemble, using gradient descent to minimize a loss function.\n",
    "\n",
    "Scikit-learn [Gradient Boosting Classifier](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.GradientBoostingClassifier.html) implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c97f3753",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gradient Boosting - AUC: 0.736 (+/- 0.023), F1: 0.226 (+/- 0.031)\n"
     ]
    }
   ],
   "source": [
    "# Create Gradient Boosting classifier\n",
    "gb_clf = GradientBoostingClassifier(\n",
    "    n_estimators=100,\n",
    "    learning_rate=0.1,\n",
    "    max_depth=3,\n",
    "    random_state=315\n",
    ")\n",
    "\n",
    "gb_scores = cross_validate(\n",
    "    gb_clf,\n",
    "    train_df.drop('target', axis=1),\n",
    "    train_df['target'],\n",
    "    cv=7,\n",
    "    scoring=['roc_auc', 'f1'],\n",
    "    return_train_score=False\n",
    ")\n",
    "\n",
    "gb_auc = gb_scores['test_roc_auc']\n",
    "gb_f1 = gb_scores['test_f1']\n",
    "\n",
    "print(f'Gradient Boosting - AUC: {gb_auc.mean():.3f} (+/- {gb_auc.std():.3f}), F1: {gb_f1.mean():.3f} (+/- {gb_f1.std():.3f})')\n",
    "\n",
    "gb_clf.fit(train_df.drop('target', axis=1), train_df['target'])\n",
    "gb_pred = gb_clf.predict(test_df.drop('target', axis=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "28a0f3be",
   "metadata": {},
   "source": [
    "### 3.3. Stacking ensemble\n",
    "\n",
    "Trains multiple diverse base models, then uses their predictions as input features for a meta-model that learns how to best combine them.\n",
    "\n",
    "Scikit-learn [Stacking Classifier](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.StackingClassifier.html) implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "193054bb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stacking - AUC: 0.813 (+/- 0.023), F1: 0.522 (+/- 0.019)\n"
     ]
    }
   ],
   "source": [
    "# Create stacking classifier\n",
    "stacking_clf = StackingClassifier(\n",
    "    estimators=[\n",
    "        ('rf', RandomForestClassifier(n_estimators=50, random_state=315, class_weight='balanced')),\n",
    "        ('dt', DecisionTreeClassifier(max_depth=5, random_state=315, class_weight='balanced')),\n",
    "        ('svm', SVC(probability=True, random_state=315, class_weight='balanced'))\n",
    "    ],\n",
    "    final_estimator=LogisticRegression(random_state=315, max_iter=1000, class_weight='balanced'),\n",
    "    cv=7\n",
    ")\n",
    "\n",
    "stacking_scores = cross_validate(\n",
    "    stacking_clf,\n",
    "    train_df.drop('target', axis=1),\n",
    "    train_df['target'],\n",
    "    cv=7,\n",
    "    scoring=['roc_auc', 'f1'],\n",
    "    return_train_score=False\n",
    ")\n",
    "\n",
    "stacking_auc = stacking_scores['test_roc_auc']\n",
    "stacking_f1 = stacking_scores['test_f1']\n",
    "\n",
    "print(f'Stacking - AUC: {stacking_auc.mean():.3f} (+/- {stacking_auc.std():.3f}), F1: {stacking_f1.mean():.3f} (+/- {stacking_f1.std():.3f})')\n",
    "\n",
    "stacking_clf.fit(train_df.drop('target', axis=1), train_df['target'])\n",
    "stacking_pred = stacking_clf.predict(test_df.drop('target', axis=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ce015bfd",
   "metadata": {},
   "source": [
    "## 4. Model comparison"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "118b48f3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Model</th>\n",
       "      <th>AUC Mean</th>\n",
       "      <th>AUC Std</th>\n",
       "      <th>F1 Mean</th>\n",
       "      <th>F1 Std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Logistic Regression</td>\n",
       "      <td>0.662758</td>\n",
       "      <td>0.025827</td>\n",
       "      <td>0.304572</td>\n",
       "      <td>0.016444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Decision Tree</td>\n",
       "      <td>0.599774</td>\n",
       "      <td>0.015110</td>\n",
       "      <td>0.307292</td>\n",
       "      <td>0.025058</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Voting</td>\n",
       "      <td>0.763999</td>\n",
       "      <td>0.014770</td>\n",
       "      <td>0.360892</td>\n",
       "      <td>0.021863</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Bagging</td>\n",
       "      <td>0.783047</td>\n",
       "      <td>0.023869</td>\n",
       "      <td>0.243632</td>\n",
       "      <td>0.033601</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Random Forest</td>\n",
       "      <td>0.791838</td>\n",
       "      <td>0.026019</td>\n",
       "      <td>0.158510</td>\n",
       "      <td>0.012615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>AdaBoost</td>\n",
       "      <td>0.683823</td>\n",
       "      <td>0.028357</td>\n",
       "      <td>0.085703</td>\n",
       "      <td>0.022349</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Gradient Boosting</td>\n",
       "      <td>0.736249</td>\n",
       "      <td>0.022935</td>\n",
       "      <td>0.225941</td>\n",
       "      <td>0.030542</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Stacking</td>\n",
       "      <td>0.812724</td>\n",
       "      <td>0.023151</td>\n",
       "      <td>0.521644</td>\n",
       "      <td>0.018951</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Model  AUC Mean   AUC Std   F1 Mean    F1 Std\n",
       "0  Logistic Regression  0.662758  0.025827  0.304572  0.016444\n",
       "1        Decision Tree  0.599774  0.015110  0.307292  0.025058\n",
       "2               Voting  0.763999  0.014770  0.360892  0.021863\n",
       "3              Bagging  0.783047  0.023869  0.243632  0.033601\n",
       "4        Random Forest  0.791838  0.026019  0.158510  0.012615\n",
       "5             AdaBoost  0.683823  0.028357  0.085703  0.022349\n",
       "6    Gradient Boosting  0.736249  0.022935  0.225941  0.030542\n",
       "7             Stacking  0.812724  0.023151  0.521644  0.018951"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Collect results\n",
    "results = {\n",
    "    'Model': ['Logistic Regression', 'Decision Tree', 'Voting', 'Bagging', 'Random Forest', 'AdaBoost', 'Gradient Boosting', 'Stacking'],\n",
    "    'AUC': [lr_auc, dt_auc, voting_auc, bagging_auc, rf_auc, adaboost_auc, gb_auc, stacking_auc],\n",
    "    'F1': [lr_f1, dt_f1, voting_f1, bagging_f1, rf_f1, adaboost_f1, gb_f1, stacking_f1],\n",
    "    'Predictions': [lr_pred, dt_pred, voting_pred, bagging_pred, rf_pred, adaboost_pred, gb_pred, stacking_pred]\n",
    "}\n",
    "\n",
    "results_df = pd.DataFrame({\n",
    "    'Model': results['Model'],\n",
    "    'AUC Mean': [auc.mean() for auc in results['AUC']],\n",
    "    'AUC Std': [auc.std() for auc in results['AUC']],\n",
    "    'F1 Mean': [f1.mean() for f1 in results['F1']],\n",
    "    'F1 Std': [f1.std() for f1 in results['F1']]\n",
    "})\n",
    "\n",
    "results_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "63c055ca",
   "metadata": {},
   "source": [
    "### 4.1. Score comparison"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a279fc04",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(nrows=1, ncols=2, figsize=(10, 4.8))\n",
    "\n",
    "# AUC boxplot\n",
    "auc_data = [auc for auc in results['AUC']]\n",
    "\n",
    "axes[0].boxplot(\n",
    "    auc_data,\n",
    "    tick_labels=results['Model'],\n",
    "    patch_artist=True,\n",
    "    vert=False\n",
    ")\n",
    "\n",
    "axes[0].set_title('AUC scores by model')\n",
    "axes[0].set_xlabel('AUC')\n",
    "\n",
    "# F1 boxplot\n",
    "f1_data = [f1 for f1 in results['F1']]\n",
    "\n",
    "axes[1].boxplot(\n",
    "    f1_data,\n",
    "    tick_labels=results['Model'],\n",
    "    patch_artist=True,\n",
    "    vert=False\n",
    ")\n",
    "\n",
    "axes[1].set_title('F1 scores by model')\n",
    "axes[1].set_xlabel('F1')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b539a53",
   "metadata": {},
   "source": [
    "### 4.2. Confusion matrix comparison"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6944af4f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA+cAAAIMCAYAAACXCt1JAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjcsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvTLEjVAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAuOhJREFUeJzs3XdYFFfbBvB7d4Fdeq+igNgrBmNXwIY99hoVe2KLidFojIqJ3cTYWzRq7LFEfe3dFDX23kCRWJEive+e7w8+RhcWBUQX4f5d1166Z8/OnBn2mZln5swZmRBCgIiIiIiIiIj0Rq7vBhAREREREREVd0zOiYiIiIiIiPSMyTkRERERERGRnjE5JyIiIiIiItIzJudEREREREREesbknIiIiIiIiEjPmJwTERERERER6RmTcyIiIiIiIiI9Y3JOREREREREpGdMzomI6J2RyWQIDAzM8/cePHgAmUyGNWvWFHibiiJ3d3cEBATouxn0Bln/TidOnIBMJsOJEyfeazt8fX3h6+v7XudJRERvxuSciKiIW7NmDWQyGWQyGf7+++9snwshULJkSchkMrRp00YPLXx7YWFh+Prrr1GhQgWYmJjA1NQU3t7emDp1KqKjo/XdPMqlwMBAyGQyODo6IjExMdvn7u7uH+xvlIiI6E0M9N0AIiJ6P1QqFTZu3IgGDRpolZ88eRKPHj2CUqnUU8vezrlz59CqVSvEx8fj008/hbe3NwDg/PnzmDlzJv78808cOnRIz618t+7cuQO5vOicb3/+/DmWLl2K0aNH67sp71SjRo2QlJQEIyMjfTeFiIgKASbnRETFRKtWrbB161YsWLAABgYvN/8bN26Et7c3IiIi9Ni6/ImOjkaHDh2gUChw6dIlVKhQQevzadOm4ZdfftFT694tIQSSk5NhbGz8wZ5YyYmXlxfmzJmDoUOHwtjY+J3MQ6PRIDU1FSqV6p1MPzfkcrle509ERIVL0TnNTkREr9WjRw9ERkbi8OHDUllqaiq2bduGnj176vxOQkICRo8ejZIlS0KpVKJ8+fL48ccfIYTQqpeSkoIvv/wS9vb2MDc3R7t27fDo0SOd03z8+DH69+8PR0dHKJVKVK5cGb/++mu+lmn58uV4/Pgx5s6dmy0xBwBHR0d89913WmVLlixB5cqVoVQq4eLigmHDhmXr+u7r64sqVarg6tWr8PHxgYmJCcqUKYNt27YByOhtULt2bRgbG6N8+fI4cuSI1vczu2ffvn0bXbt2hYWFBWxtbfHFF18gOTlZq+7q1avRuHFjODg4QKlUolKlSli6dGm2Zcns0n3w4EHUrFkTxsbGWL58ufTZq/cyp6WlYcqUKShbtixUKhVsbW3RoEEDrb89ABw7dgwNGzaEqakprKys8Mknn+DWrVs6lyU4OBgBAQGwsrKCpaUl+vXrp7PreUGYNGkSwsLCdK6HrHL7G5XJZBg+fDg2bNgg/f0PHDgg3fbx999/Y+TIkbC3t4eVlRWGDBmC1NRUREdHo0+fPrC2toa1tTXGjh2bbdo//vgj6tWrB1tbWxgbG8Pb21v6rbxO1nvOX70FJesr6z3i69evh7e3N4yNjWFjY4Pu3bvj4cOH2eaxYsUKeHp6wtjYGLVq1cJff/31xnYREZF+MDknIiom3N3dUbduXWzatEkq279/P2JiYtC9e/ds9YUQaNeuHX7++We0aNECc+fORfny5TFmzBh89dVXWnUHDhyIefPmoXnz5pg5cyYMDQ3RunXrbNMMCwtDnTp1cOTIEQwfPhzz589HmTJlMGDAAMybNy/Py7R7924YGxujc+fOuaofGBiIYcOGwcXFBT/99BM6deqE5cuXo3nz5khLS9Oq++LFC7Rp0wa1a9fG7NmzoVQq0b17d2zZsgXdu3dHq1atMHPmTCQkJKBz586Ii4vLNr+uXbsiOTkZM2bMQKtWrbBgwQIMHjxYq87SpUvh5uaGb7/9Fj/99BNKliyJoUOHYvHixdmmd+fOHfTo0QPNmjXD/Pnz4eXlleNyTpkyBX5+fli0aBEmTJiAUqVK4eLFi1KdI0eOwN/fH8+fP0dgYCC++uornDp1CvXr18eDBw90LktcXBxmzJiBrl27Ys2aNZgyZUou1nreNWzYEI0bN8bs2bORlJSUY728/EaBjJMRX375Jbp164b58+fD3d1d+mzEiBEICgrClClT0K5dO6xYsQITJ05E27ZtoVarMX36dDRo0ABz5szBunXrtKY7f/581KhRA99//z2mT58OAwMDdOnSBXv37s3Tcjdq1Ajr1q3Tek2dOhUA4ODgINWbNm0a+vTpg7Jly2Lu3LkYNWoUjh49ikaNGmmdaFq1ahWGDBkCJycnzJ49G/Xr10e7du10JvFERFQICCIiKtJWr14tAIhz586JRYsWCXNzc5GYmCiEEKJLly7Cz89PCCGEm5ubaN26tfS9nTt3CgBi6tSpWtPr3LmzkMlkIjg4WAghxOXLlwUAMXToUK16PXv2FADE5MmTpbIBAwYIZ2dnERERoVW3e/fuwtLSUmpXSEiIACBWr1792mWztrYW1atXz9V6eP78uTAyMhLNmzcXarVaKl+0aJEAIH799VepzMfHRwAQGzdulMpu374tAAi5XC7OnDkjlR88eDBbWydPniwAiHbt2mm1YejQoQKAuHLlilSWucyv8vf3F6VLl9Yqc3NzEwDEgQMHstV3c3MTffv2ld5Xr15d62+pi5eXl3BwcBCRkZFS2ZUrV4RcLhd9+vTJtiz9+/fX+n6HDh2Era3ta+eRV5nzCg8PFydPnhQAxNy5c6XP8/sbFUJIf7sbN25o1c2MD39/f6HRaKTyunXrCplMJj777DOpLD09Xbi6ugofHx+taWT9G6ampooqVaqIxo0ba5Vn/TsdP35cABDHjx/XuT6SkpKEt7e3cHFxEU+fPhVCCPHgwQOhUCjEtGnTtOpeu3ZNGBgYSOWpqanCwcFBeHl5iZSUFKneihUrBIBsy0BERPrHK+dERMVI165dkZSUhD179iAuLg579uzJsUv7vn37oFAoMHLkSK3y0aNHQwiB/fv3S/UAZKs3atQorfdCCGzfvh1t27aFEAIRERHSy9/fHzExMVpXdnMjNjYW5ubmuap75MgRpKamYtSoUVqDpw0aNAgWFhbZrnKamZlp9SgoX748rKysULFiRdSuXVsqz/z//fv3s81z2LBhWu9HjBgB4OU6A6B1T3VMTAwiIiLg4+OD+/fvIyYmRuv7Hh4e8Pf3f+OyWllZ4caNGwgKCtL5+dOnT3H58mUEBATAxsZGKq9WrRqaNWum1b5Mn332mdb7hg0bIjIyErGxsW9sT340atQIfn5+r716ntvfaCYfHx9UqlRJ57QGDBgAmUwmva9duzaEEBgwYIBUplAoULNmzWx/61f/hi9evEBMTAwaNmyY599zVkOHDsW1a9ewfft2ODk5AQB27NgBjUaDrl27asWQk5MTypYti+PHjwPIGBDx+fPn+Oyzz7QGnAsICIClpeVbtYuIiN4NDghHRFSM2Nvbo2nTpti4cSMSExOhVqtz7BIeGhoKFxeXbMlvxYoVpc8z/5XL5fD09NSqV758ea334eHhiI6OxooVK7BixQqd83z+/HmelsfCwkJnd3JdMtubtV1GRkYoXbq09HkmV1dXrWQNACwtLVGyZMlsZUBGUpZV2bJltd57enpCLpdrdRv/559/MHnyZJw+fTrbPdwxMTFaiZSHh8frFlHy/fff45NPPkG5cuVQpUoVtGjRAr1790a1atUA5LwugIy/78GDB5GQkABTU1OpvFSpUlr1rK2tAWQst4WFhc52xMfHIz4+XnqvUChgb2+fq2UAMrrn+/j4YNmyZfjyyy+zfZ7b32im162/rMuXud51/b2z/q337NmDqVOn4vLly0hJSZHKs/5+8mL58uVYvXo1li9fjjp16kjlQUFBEEJk+21lMjQ0BPBy2bPWMzQ0ROnSpfPdLiIieneYnBMRFTM9e/bEoEGD8OzZM7Rs2RJWVlbvZb4ajQYA8Omnn6Jv374662Qmj7lVoUIFXL58GampqQX+OCqFQpGncpFlkDBdsiZr9+7dQ5MmTVChQgXMnTsXJUuWhJGREfbt24eff/5ZWmeZcjtyeaNGjXDv3j3s2rULhw4dwsqVK/Hzzz9j2bJlGDhwYK6mkVV+lvvHH3/Uui/dzc1N5/3sOWnUqBF8fX0xe/bsbFfu8+N16y8vf+9Xl/mvv/5Cu3bt0KhRIyxZsgTOzs4wNDTE6tWrsXHjxny18+zZs/jiiy8wcODAbGMUaDQayGQy7N+/X2fbzMzM8jVPIiLSPybnRETFTIcOHTBkyBCcOXMGW7ZsybGem5sbjhw5gri4OK0rk7dv35Y+z/xXo9Hg3r17Wldi79y5ozW9zJHc1Wo1mjZtWiDL0rZtW5w+fRrbt29Hjx49Xls3s7137tzRunKYmpqKkJCQAmvTq4KCgrSu1gYHB0Oj0UgDkf3vf/9DSkoKdu/erXXlNrNr8tuwsbFBv3790K9fP8THx6NRo0YIDAzEwIEDtdZFVrdv34adnZ3WVfP86tOnDxo0aCC9z89j0QIDA+Hr6yuNTP+q3P5G36Xt27dDpVLh4MGDWo+0W716db6mFx4ejs6dO8PLy0vnoICenp4QQsDDwwPlypXLcTqZyx4UFITGjRtL5WlpaQgJCUH16tXz1T4iInp3eM85EVExY2ZmhqVLlyIwMBBt27bNsV6rVq2gVquxaNEirfKff/4ZMpkMLVu2BADp3wULFmjVyzr6ukKhQKdOnbB9+3Zcv3492/zCw8PzvCyfffYZnJ2dMXr0aNy9ezfb58+fP5dGu27atCmMjIywYMECrSufq1atQkxMjM7R5d9W1uRq4cKFAF6us8wrn6+2JyYmJt+JXabIyEit92ZmZihTpozU5drZ2RleXl5Yu3at1uje169fx6FDh9CqVau3mn+m0qVLo2nTptKrfv36eZ6Gj48PfH19MWvWrGyPocvtb/RdUigUkMlkUKvVUtmDBw+wc+fOPE9LrVaje/fuSE1Nxfbt23X2BunYsSMUCgWmTJmSrdeCEEL629esWRP29vZYtmwZUlNTpTpr1qzJ9uhAIiIqHHjlnIioGMqpW/mr2rZtCz8/P0yYMAEPHjxA9erVcejQIezatQujRo2S7jH38vJCjx49sGTJEsTExKBevXo4evQogoODs01z5syZOH78OGrXro1BgwahUqVKiIqKwsWLF3HkyBFERUXlaTmsra3xxx9/oFWrVvDy8sKnn34Kb29vAMDFixexadMm1K1bF0DGlfvx48djypQpaNGiBdq1a4c7d+5gyZIl+Pjjj/Hpp5/mad65ERISgnbt2qFFixY4ffo01q9fj549e0pXLZs3bw4jIyO0bdsWQ4YMQXx8PH755Rc4ODjg6dOn+Z5vpUqV4OvrC29vb9jY2OD8+fPYtm0bhg8fLtWZM2cOWrZsibp162LAgAFISkrCwoULYWlpicDAwLdd9AI1efJk+Pn5ZSvP7W/0XWrdujXmzp2LFi1aoGfPnnj+/DkWL16MMmXK4OrVq3ma1rJly3Ds2DF89tln2XpPODo6olmzZvD09MTUqVMxfvx4PHjwAO3bt4e5uTlCQkLwxx9/YPDgwfj6669haGiIqVOnYsiQIWjcuDG6deuGkJAQrF69mvecExEVUkzOiYhIJ7lcjt27d2PSpEnYsmULVq9eDXd3d8yZMwejR4/Wqvvrr7/C3t4eGzZswM6dO9G4cWPs3bs322Bajo6OOHv2LL7//nvs2LEDS5Ysga2tLSpXroxZs2blq521a9fG9evXMWfOHOzduxfr1q2DXC5HxYoVMW7cOK2ENDAwEPb29li0aBG+/PJL2NjYYPDgwZg+fbo0kFZB2rJlCyZNmoRx48bBwMAAw4cPx5w5c6TPy5cvj23btuG7777D119/DScnJ3z++eewt7dH//798z3fkSNHYvfu3Th06BBSUlLg5uaGqVOnYsyYMVKdpk2b4sCBA5g8eTImTZoEQ0ND+Pj4YNasWbkeeO598fX1hY+PD06ePKlVnpff6LvSuHFjrFq1CjNnzsSoUaPg4eGBWbNm4cGDB3lOzjN7jyxbtgzLli3T+szHxwfNmjUDAIwbNw7lypXDzz//LN3TX7JkSTRv3hzt2rWTvjN48GCo1WrMmTMHY8aMQdWqVbF7925MnDjxbRaZiIjeEZnIzQg2RERElGuBgYGYMmUKwsPDYWdnp+/mEBER0QeA95wTERERERER6RmTcyIiIiIiIiI9Y3JOREREREREpGe855yIiIiIiIhIz3jlnIiIiIiIiEjPmJwTERERERER6RmTcyIiIiIiIiI9Y3JOREREREREpGdMzomIiIiIiIj0jMk5ERERERERkZ4xOSciIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50WEr68vfH19C2x67u7uCAgIKLDpESCTyRAYGKjvZhC91oMHDyCTybBmzZo8fa+gt0FE9GEJCAiAu7u7vptBRHrC3KFgMDkvYGvWrIFMJsP58+f13ZQ3OnXqFAIDAxEdHf1O5+Pu7g6ZTCa9TE1NUatWLfz222/vdL5EH6LMbUjmS6VSwcXFBf7+/liwYAHi4uL03cRCI+u2JadXXk80EBUV7dq1g4mJyWu3G7169YKRkREiIyPfOL0nT54gMDAQly9fLsBWEtHrZD0ukMlkcHBwgJ+fH/bv36/v5lEBM9B3A6hgHDp0KM/fOXXqFKZMmYKAgABYWVlpfXbnzh3I5QV37sbLywujR48GADx9+hQrV65E3759kZKSgkGDBhXYfAqzpKQkGBgw5Ch3vv/+e3h4eCAtLQ3Pnj3DiRMnMGrUKMydOxe7d+9GtWrV3sl83dzckJSUBENDwzx9Lz/boLc1b948xMfHS+/37duHTZs24eeff4adnZ1UXq9evffeNqLCoFevXvjf//6HP/74A3369Mn2eWJiInbt2oUWLVrA1tb2jdN78uQJpkyZAnd3d3h5eWl99ssvv0Cj0RRU04koi8zjAiEEwsLCsGbNGrRq1Qr/+9//0KZNG303r8Bzh+KKmUIRYWRkVKDTUyqVBTq9EiVK4NNPP5XeBwQEoHTp0vj555/fe3KekJAAU1PT9zpPAFCpVO99nvThatmyJWrWrCm9Hz9+PI4dO4Y2bdqgXbt2uHXrFoyNjQt8vplX6/OqoLdBudG+fXut98+ePcOmTZvQvn3713av1dc2gOh9a9euHczNzbFx40adyfmuXbuQkJCAXr16vfW88npCj4jyJutxwYABA+Do6IhNmzYViuS8oHOH4oqnN/Tk0qVLaNmyJSwsLGBmZoYmTZrgzJkz2epdvXoVPj4+MDY2hqurK6ZOnYrVq1dDJpPhwYMHUj1d93suXLgQlStXhomJCaytrVGzZk1s3LgRABAYGIgxY8YAADw8PKRuMpnT1HXfSHR0NL788ku4u7tDqVTC1dUVffr0QURERJ6X397eHhUqVMC9e/e0yjUaDebNm4fKlStDpVLB0dERQ4YMwYsXL7LVCwwMhIuLC0xMTODn54ebN29ma3dmV6CTJ09i6NChcHBwgKurq/T5/v370bBhQ5iamsLc3BytW7fGjRs3tOb17Nkz9OvXD66urlAqlXB2dsYnn3yitf7Pnz8Pf39/2NnZwdjYGB4eHujfv7/WdHTdc56b30HmMvzzzz/46quvYG9vD1NTU3To0AHh4eG5XeVUBDRu3BgTJ05EaGgo1q9fr/XZ7du30blzZ9jY2EClUqFmzZrYvXt3tmm8KY513XOemxjQtQ16/vy5dPCgUqlQvXp1rF27VqtO5vx+/PFHrFixAp6enlAqlfj4449x7ty5t1thyDgRaGZmhnv37qFVq1YwNzeXEpHcbm+A3G0riAobY2NjdOzYEUePHsXz58+zfb5x40aYm5ujXbt2uH//Prp06QIbGxuYmJigTp062Lt3r1T3xIkT+PjjjwEA/fr1y3bbSNZ7zvMa21u3bkWlSpWgUqlQpUoV/PHHH7yPneg1rKysYGxsrNUr88cff0S9evVga2sLY2NjeHt7Y9u2bdm+m5SUhJEjR8LOzk7aBjx+/FjnseqJEydQs2ZNqFQqeHp6Yvny5QgMDIRMJtOql9MxeG6OX3N7XF8c8Mq5Hty4cQMNGzaEhYUFxo4dC0NDQyxfvhy+vr44efIkateuDQB4/Pgx/Pz8IJPJMH78eJiammLlypW5OjP1yy+/YOTIkejcuTO++OILJCcn4+rVq/j333/Rs2dPdOzYEXfv3s3WBdTe3l7n9OLj49GwYUPcunUL/fv3x0cffYSIiAjs3r0bjx490upCmhvp6el49OgRrK2ttcqHDBmCNWvWoF+/fhg5ciRCQkKwaNEiXLp0Cf/88490Zn78+PGYPXs22rZtC39/f1y5cgX+/v5ITk7WOb+hQ4fC3t4ekyZNQkJCAgBg3bp16Nu3L/z9/TFr1iwkJiZi6dKlaNCgAS5duiQdEHTq1Ak3btzAiBEj4O7ujufPn+Pw4cP477//pPfNmzeHvb09xo0bBysrKzx48AA7dux47TrI7e8g04gRI2BtbY3JkyfjwYMHmDdvHoYPH44tW7bkad3Th61379749ttvcejQIanXyY0bN1C/fn2UKFEC48aNg6mpKX7//Xe0b98e27dvR4cOHQDkP47fFAO6JCUlwdfXF8HBwRg+fDg8PDywdetWBAQEIDo6Gl988YVW/Y0bNyIuLg5DhgyBTCbD7Nmz0bFjR9y/f/+tr8ilp6fD398fDRo0wI8//ggTExMAud/e5HZbQVQY9erVC2vXrsXvv/+O4cOHS+VRUVE4ePAgevTogdjYWNSrVw+JiYkYOXIkbG1tsXbtWrRr1w7btm1Dhw4dULFiRXz//feYNGkSBg8ejIYNGwJ4820juYntvXv3olu3bqhatSpmzJiBFy9eYMCAAShRosS7WzFEH5iYmBhERERACIHnz59j4cKFiI+P1+qZOn/+fLRr1w69evVCamoqNm/ejC5dumDPnj1o3bq1VC8gIAC///47evfujTp16uDkyZNan2e6dOkSWrRoAWdnZ0yZMgVqtRrff/99jvmCLrk5fs3rcX2RJqhArV69WgAQ586dy7FO+/bthZGRkbh3755U9uTJE2Fubi4aNWoklY0YMULIZDJx6dIlqSwyMlLY2NgIACIkJEQq9/HxET4+PtL7Tz75RFSuXPm1bZ0zZ0626WRyc3MTffv2ld5PmjRJABA7duzIVlej0bx2Pm5ubqJ58+YiPDxchIeHi2vXronevXsLAGLYsGFSvb/++ksAEBs2bND6/oEDB7TKnz17JgwMDET79u216gUGBgoAWu3O/Hs0aNBApKenS+VxcXHCyspKDBo0SGsaz549E5aWllL5ixcvBAAxZ86cHJfvjz/+eOPfXAghAIjJkydL73P7O8hchqZNm2qt6y+//FIoFAoRHR392vnShyU32xBLS0tRo0YN6X2TJk1E1apVRXJyslSm0WhEvXr1RNmyZaWy3MRxSEiIACBWr14thMhdDAiRfRs0b948AUCsX79eKktNTRV169YVZmZmIjY2Vmt+tra2IioqSqq7a9cuAUD873//e+18X6Vrm9a3b18BQIwbN06rbm63N7ndVhAVVunp6cLZ2VnUrVtXq3zZsmUCgDh48KAYNWqUACD++usv6fO4uDjh4eEh3N3dhVqtFkIIce7cOa3tw6v69u0r3NzcpPd5ie2qVasKV1dXERcXJ5WdOHFCANCaJlFxlHlckPWlVCrFmjVrtOomJiZqvU9NTRVVqlQRjRs3lsouXLggAIhRo0Zp1Q0ICMh2rNq2bVthYmIiHj9+LJUFBQUJAwMDkTWNzJo75Pb4NS/H9cUBu7W/Z2q1GocOHUL79u1RunRpqdzZ2Rk9e/bE33//jdjYWADAgQMHULduXa1BV2xsbHJ1b5iVlRUePXpUIN1CAWD79u2oXr26dAXuVVm7tehy6NAh2Nvbw97eHlWrVsW6devQr18/zJkzR6qzdetWWFpaolmzZoiIiJBe3t7eMDMzw/HjxwEAR48eRXp6OoYOHao1jxEjRuQ4/0GDBkGhUEjvDx8+jOjoaPTo0UNrXgqFArVr15bmZWxsDCMjI5w4cUJnV1cA0mB6e/bsQVpa2hvXBZC330GmwYMHa63rhg0bQq1WIzQ0NFfzpKLDzMxMGn05KioKx44dQ9euXREXFyf9liMjI+Hv74+goCA8fvwYQP7iODcxoMu+ffvg5OSEHj16SGWGhoYYOXIk4uPjcfLkSa363bp10+pJk3lV7v79+7me5+t8/vnnWu9zu73J7baCqLBSKBTo3r07Tp8+rXUrysaNG+Ho6IgmTZpg3759qFWrFho0aCB9bmZmhsGDB+PBgwe4efNmvuf/pth+8uQJrl27hj59+sDMzEyq5+Pjg6pVq+Z7vkRFzeLFi3H48GEcPnwY69evh5+fHwYOHKjVU/PVsWhevHiBmJgYNGzYEBcvXpTKDxw4AABvPI5Wq9U4cuQI2rdvDxcXF6m8TJkyaNmyZa7b/abj1/wc1xdlTM7fs/DwcCQmJqJ8+fLZPqtYsSI0Gg0ePnwIAAgNDUWZMmWy1dNVltU333wDMzMz1KpVC2XLlsWwYcPwzz//5Lvd9+7dQ5UqVfL9/dq1a+Pw4cM4cOAAfvzxR1hZWeHFixdag0gFBQUhJiYGDg4OUiKf+YqPj5ful8sM5qzrwcbGJls3+UweHh5a74OCggBk3MObdV6HDh2S5qVUKjFr1izs378fjo6OaNSoEWbPno1nz55J0/Lx8UGnTp0wZcoU2NnZ4ZNPPsHq1auRkpKS4/rIy+8gU6lSpbTeZy5rXhImKhri4+Nhbm4OAAgODoYQAhMnTsz2W548eTIASL/n/MRxbmJAl9DQUJQtWzbbyK0VK1aUPn/Vu/x9GxgYaI01AeR+e5PbbQVRYZZ5Uj9z3JlHjx7hr7/+Qvfu3aFQKBAaGprj/gjIHq958abYzmmfnlMZUXFVq1YtNG3aFE2bNkWvXr2wd+9eVKpUCcOHD0dqaiqAjAtFderUgUqlgo2NDezt7bF06VLExMRI0wkNDYVcLs92bJw13p4/f46kpKS3js38bgNed1xflPGe8yKqYsWKuHPnDvbs2YMDBw5g+/btWLJkCSZNmoQpU6a89/bY2dmhadOmAAB/f39UqFABbdq0wfz58/HVV18ByBgMwsHBARs2bNA5jbzc35JV1lGtMx/3sm7dOjg5OWWr/+rgGqNGjULbtm2xc+dOHDx4EBMnTsSMGTNw7Ngx1KhRAzKZDNu2bcOZM2fwv//9DwcPHkT//v3x008/4cyZM1pXAt7Gq1f+XyWEKJDp04fh0aNHiImJkXZimb/lr7/+Gv7+/jq/87YHuG+KgYLwLn/fSqUy20mC3G5v8rKtICqsvL29UaFCBWzatAnffvstNm3aBCFEgYzS/ibcdxG9G3K5HH5+fpg/fz6CgoIQFRWFdu3aoVGjRliyZAmcnZ1haGiI1atXSyfm9IHbgLzhUcV7Zm9vDxMTE9y5cyfbZ7dv34ZcLkfJkiUBZDxvODg4OFs9XWW6mJqaolu3bujWrRtSU1PRsWNHTJs2DePHj4dKpcpVd/RMnp6euH79eq7rv0nr1q3h4+OD6dOnY8iQITA1NYWnpyeOHDmC+vXrv/YRUW5ubgAy1sOrZ/0iIyNzfZXN09MTAODg4CCdNHhT/dGjR2P06NEICgqCl5cXfvrpJ60Rs+vUqYM6depg2rRp2LhxI3r16oXNmzdj4MCB2aaXl98B0avWrVsHAFIinnlbhKGh4Rt/y28Tx7mJgVe5ubnh6tWr0Gg0Wonx7du3pc/1Kbfbm7xuK4gKq169emHixIm4evUqNm7ciLJly0qjr7u5ueW4P8r8HMjdbWx59eo+PavcHu8QFVfp6ekAMnrUbd++HSqVCgcPHtQaPHr16tVa33Fzc4NGo0FISAjKli0rlWeNNwcHB6hUqncemwVxXF+UsFv7e6ZQKNC8eXPs2rVL696vsLAwbNy4EQ0aNICFhQWAjIPv06dP4/Lly1K9qKioHK/0vCoyMlLrvZGRESpVqgQhhHRfdOZzfqOjo984vU6dOuHKlSv4448/sn2W3zNf33zzDSIjI/HLL78AALp27Qq1Wo0ffvghW9309HSpnU2aNIGBgQGWLl2qVWfRokW5nre/vz8sLCwwffp0nfeJZz7iITExMdtIkZ6enjA3N5e6rb948SLbOsgcJyCnru15+R0QZTp27Bh++OEHeHh4SFe8HBwc4Ovri+XLl+Pp06fZvvPq40ryE8e5iQFdWrVqhWfPnmmNxpqeno6FCxfCzMwMPj4+r1/Ydyy325vcbiuICrvMbcakSZNw+fJlravmrVq1wtmzZ3H69GmpLCEhAStWrIC7uzsqVaoEIG/HDbnl4uKCKlWq4LfffkN8fLxUfvLkSVy7dq3A5kNU1KSlpeHQoUMwMjJCxYoVoVAoIJPJoFarpToPHjzAzp07tb6XeXJ/yZIlWuULFy7Ueq9QKNC0aVPs3LkTT548kcqDg4Oxf//+AluOgjiuL0p45fwd+fXXX6UBF171xRdfYOrUqTh8+DAaNGiAoUOHwsDAAMuXL0dKSgpmz54t1R07dizWr1+PZs2aYcSIEdKj1EqVKoWoqKjXnsFu3rw5nJycUL9+fTg6OuLWrVtYtGgRWrduLd2r6u3tDQCYMGECunfvDkNDQ7Rt21ba+b5qzJgx2LZtG7p06YL+/fvD29sbUVFR2L17N5YtW4bq1avneR21bNkSVapUwdy5czFs2DD4+PhgyJAhmDFjBi5fvozmzZvD0NAQQUFB2Lp1K+bPn4/OnTvD0dERX3zxBX766Se0a9cOLVq0wJUrV7B//37Y2dnl6sy+hYUFli5dit69e+Ojjz5C9+7dYW9vj//++w979+5F/fr1sWjRIty9exdNmjRB165dUalSJRgYGOCPP/5AWFgYunfvDgBYu3YtlixZgg4dOsDT0xNxcXH45ZdfYGFhgVatWuXYhtz+Dqh42r9/P27fvo309HSEhYXh2LFjOHz4MNzc3LB7926oVCqp7uLFi9GgQQNUrVoVgwYNQunSpREWFobTp0/j0aNHuHLlCoD8xXFuYkCXwYMHY/ny5QgICMCFCxfg7u6Obdu24Z9//sG8efOk7ZC+5HZ7k9ttBVFh5+HhgXr16mHXrl0AoJWcjxs3Dps2bULLli0xcuRI2NjYYO3atQgJCcH27dul3i+enp6wsrLCsmXLYG5uDlNTU9SuXTvbvat5NX36dHzyySeoX78++vXrhxcvXmDRokWoUqWKVsJOVJxlHhcAGfeDb9y4EUFBQRg3bhwsLCzQunVrzJ07Fy1atEDPnj3x/PlzLF68GGXKlMHVq1el6Xh7e6NTp06YN28eIiMjpUep3b17F4B2D5nAwEAcOnQI9evXx+effw61Wi3F5qsXD99GQRzXFyn6Gia+qMrpcQeZr4cPHwohhLh48aLw9/cXZmZmwsTERPj5+YlTp05lm96lS5dEw4YNhVKpFK6urmLGjBliwYIFAoB49uyZVC/rY4yWL18uGjVqJGxtbYVSqRSenp5izJgxIiYmRmv6P/zwgyhRooSQy+VajyDK+jgEITIe4zZ8+HBRokQJYWRkJFxdXUXfvn1FRETEa9eJm5ubaN26tc7P1qxZk+2xLCtWrBDe3t7C2NhYmJubi6pVq4qxY8eKJ0+eSHXS09PFxIkThZOTkzA2NhaNGzcWt27dEra2tuKzzz7L9vfI6bFUx48fF/7+/sLS0lKoVCrh6ekpAgICxPnz54UQQkRERIhhw4aJChUqCFNTU2FpaSlq164tfv/9d2kaFy9eFD169BClSpUSSqVSODg4iDZt2kjTyIQsj6fI/O6bfgc5LcPx48cFAHH8+HGdy0YfpqzbECMjI+Hk5CSaNWsm5s+fLz2CLKt79+6JPn36CCcnJ2FoaChKlCgh2rRpI7Zt26ZV701xnPVRarmJASGyb4OEECIsLEz069dP2NnZCSMjI1G1atVsj2DKnJ+uR7XpipnXyelRaqampjl+JzfbGyHevK0g+hAsXrxYABC1atXK9tm9e/dE586dhZWVlVCpVKJWrVpiz5492ert2rVLVKpUSXqUUmZM5/QotdzG9ubNm0WFChWEUqkUVapUEbt37xadOnUSFSpUeKtlJvrQ6cotVCqV8PLyEkuXLtV6TNmqVatE2bJlhVKpFBUqVBCrV68WkydPzvbYs4SEBDFs2DBhY2MjzMzMRPv27cWdO3cEADFz5kytukePHhU1atQQRkZGwtPTU6xcuVKMHj1aqFQqrXo5PUotN8evuT2uLw5kQvBu/A/NqFGjsHz5csTHx+c4yEJxFB0dDWtra0ydOhUTJkzQd3OIiIjoLXh5ecHe3h6HDx/Wd1OIirzLly+jRo0aWL9+/RsHi2zfvj1u3LghPdHkXSiux/W857yQS0pK0nofGRmJdevWoUGDBsU6Mc+6XgBg3rx5AABfX9/32xgiIiLKt7S0NGlgq0wnTpzAlStXuE8negdyOo6Wy+Vo1KjRa+sGBQVh3759BRqbPK5/ifecF3J169aFr68vKlasiLCwMKxatQqxsbGYOHGivpumV1u2bMGaNWvQqlUrmJmZ4e+//8amTZvQvHlz1K9fX9/NIyIiolx6/PgxmjZtik8//RQuLi64ffs2li1bBicnJ3z22Wf6bh5RkTN79mxcuHABfn5+MDAwwP79+7F//34MHjw429OCSpcujYCAAJQuXRqhoaFYunQpjIyMMHbs2AJrD4/rX2JyXsi1atUK27Ztw4oVKyCTyfDRRx9h1apV2c5qFTfVqlWDgYEBZs+ejdjYWGkwialTp+q7aURERJQH1tbW8Pb2xsqVKxEeHg5TU1O0bt0aM2fOhK2trb6bR1Tk1KtXD4cPH8YPP/yA+Ph4lCpVCoGBgTq7j7do0QKbNm3Cs2fPoFQqUbduXUyfPl3rMWxvi8f1L/GecyIiIiIiIiI94z3nRERERERERHrG5JyIiIiIiIhIz3jP+VvQaDR48uQJzM3NIZPJ9N0cIgCAEAJxcXFwcXGBXM7zb/nF+KbCiPFdMBjfVBgxvgsG45sKo9zGN5Pzt/DkyZNsIxoSFRYPHz6Eq6urvpvxwWJ8U2HG+H47jG8qzBjfb4fxTYXZm+KbyflbMDc3BwCEXnSHhRnPcL6tDuWq6rsJRUI60vA39km/T8ofxnfBYnwXDMZ3wWB8FyzGd8FgfBcMxnfBYnwXjNzGN5Pzt5DZVcbCTA4Lcwb/2zKQGeq7CUXD/z9/gV253g7ju2AxvgsI47tAML4LFuO7gDC+CwTju2AxvgtILuObv1giIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50RERERERER6xuSciIiIiIiISM+YnBMRERERERHpGZNzIiIiIiIiIj1jck5ERERERESkZ0zOiYiIiIiIiPSMyTkRERERERGRnjE5JyIiIiIiItIzJudEREREREREesbknIiIiIiIiEjPmJwTERERERER6RmTcyIiIiIiIiI9Y3JOREREREREpGdMzomIiIiIiIj0jMk5ERERERERkZ4xOSciIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50RERERERER6xuSciIiIiIiISM+YnBMRERERERHpGZNzIiIiIiIiIj0z0HcDKO92r7bDtqUOiAo3QOlKSRg69TEq1EjMsX58jAJrZjrhn/1WiItWwME1FZ9NeYxaTeIAAP9ba4u9v9kh7KERAMCtfDJ6ffkMHzeOey/Lo09tAyLQ+fPnsLFPx/2bxljyXQncuWyis279ltHoPvI5XNxTYGAIPA4xwvZl9ji63Uaqc/DJFZ3f/eUHZ2xb6vBOloGKlrzG945f7LF3rS2ePzGChXU6GraJRv/xT2GkEgCAxHg51s52xqn9loiONIBn5SR8/sMjlPdKel+LpDd5iW+3csnoM+YZylRLhFPJNCyb5II/Vtpr1alSOx5dhoajbNVE2DqlI7C/O04fsHwfi0JFRF7iOz0N2LzQEUe22iDimSFcPVMwYMITfOz3ct+8eaED/tlnhYfBShipNKhUMxEDJjxByTIp72uR3ou8xDIANGwTjb5jn8HRNRWPQ5RYNc0Z545ZSJ/nZl9donQKBk18gkofJ8DAUCDklgq/zXbGlVNmBbtwVOxcO2OKrUscEHTNBFFhhpi8KgT1Wsbou1nvTUHHMyDQZ0wYWvSMhJmFGjfPm2LBOFc8CVECABxdU9HzyzB41Y+HtX0aIsMMcWyHNTbNd0B62svr1N4+sej9dRjcyicjNUWG62fMsGKKC8IeGb2rVaETr5wDWLx4Mdzd3aFSqVC7dm2cPXtW303K0YldVlgxxQW9vnqGxQfvoHSlJEzoWRrREbrPs6SlyjC+uyfCHhnhuxUPsPKv2xg15yFsndKkOvbOaej/7RMsOnAHC/ffRfX6cQjs54EHd1Tva7H0wqfdCwye/AQb5jphmH853L+pwrSN92Fpm6azfly0ATbNd8SotmXxWZNyOLTZBqN/fghvn1ipTvfqlbReP31ZEhoN8PdeHsDrS1GO72M7rPDrdGf0+uoZfjl5G1/99BAnd1tj9Uxnqc7Po0vi4p9mGLswFMuO3oa3TxzGdSuDiKeG72ux9CKv8a001uDpf0b4dbozIsN0r2+ViQb3b6iw6FvXd9l0yoOiHN9rZjlj33pbDJ36CL+cuI3WvSPw/QAPBF8zlupcPW2GtgERmLcnCDM234M6Hfi2hyeSE4vO4V1eY7lSzQSMXxKKA5tsMLR5OZw6YIHJvz6AW/mXJyRzs6/+fu19yBUC33TxxPAW5XD/pjG+/y0E1va650sF70OK77xITpSjdOUkDJ/+SN9Nee/eRTx3HRaOT/qHY+E4V3zRpiySE+WYvvE+DJUaAEDJMsmQywXmf+OKwX7lsTzQBa17R6Lf+GfSNBxLpiBw9QNc+ccMQ5uVw4SepWFhk46Jqx680/WhS9HZeufTli1b8NVXX2Hy5Mm4ePEiqlevDn9/fzx//lzfTdNpxwp7tOgZCf/uUXArl4KRsx5BaazBwU02Ousf3GyDuGgFJv8agsq1EuBUMhXV6ibAs3KyVKdO81jUahKHEqVT4eqZgn7jnkFlqsHtCzmfxSoKOg6OwIGNNji0xQb/Bamw4BtXpCTJ4N8jSmf9q6fNcOqAJR4Gq/A0VImdq+xx/5YxKtdKkOq8CDfUetX1j8GVf8zw7D/l+1osekVRj++b501R+eMENO4YDaeSqfD2jYNv+xe4cykjdlOSZPh7nxUGfvcUVeskoIRHKnp//Qwu7inY85vt+1y09y6v8X33iglW/uCCk7uskZYq01nn/HGLjF4IvFpeKBT1+D663QbdRzxHrSZxcHZLRdu+kfi4cSy2L3/Zo2P6xvto3i0K7uWT4Vk5GaPn/Yfnj40QdNVY5zQ/RHmN5fYDw3H+uDm2LXXAw2AVfpvjjOBrxvikX6RU5037agubdLh6puL3RQ4IuWWMJyFK/DrNGSoTDdwrJOucLxWsDy2+8+LjxnEI+OYZ6hejq+WZCj6eBdoPDMem+Y44fdASIbeMMXtkKdg6pqFei4z1e/6EBX76shQunjTHs/+UOHPIEtuW2Wut/7LVkiBXCKyZ5YSnoUoEXzPBtmX28KycBIWBeNerRUuxT87nzp2LQYMGoV+/fqhUqRKWLVsGExMT/Prrr/puWjZpqTIEXTXBRw3jpTK5HKjRMB43L5jq/M6ZQ5ao6J2ARd+6olu1yhjsVx6bFjhArdY9D7UaOLHTCimJclSsmaC7UhFgYKhB2WqJuPiXuVQmhAyX/jJHJe+cuxC/JODVIA4lPVNw/V/dXdys7NJQq0ksDm7WfeBF715Rj+9KNRMQdNUEt/8/GX8aaoRzRy3wcZOM3hxqtQwatQxG/3/2OJNSpcGNs0W3a+bbxzd9CIp6fKel5j12E2IVAABzqxx28h+Y/MRyRe9EXHqlPgBcOGmOit66j2l07atjoxR4GKxE0y4voDRWQ64QaN07Ei/CDYrUiY/C7EOKb8qddxHPTqVSYeuYrjXNxDgFbl8yQcXX7O9NzdWIi1ZI74OuGkOjkaF59yjI5QIm5mo07fQCl/4ygzpd9wn7d6VY33OempqKCxcuYPz48VKZXC5H06ZNcfr06Wz1U1JSkJLy8j6u2NjYbHXepdgoBTRqGayydKmytkvDw2DdV2afhhrh8j9maNzhBaauv4/HIUos+tYV6jQZPh0dJtULuaXCqLZlkZoih7GpBpNWhcCtXNG6Z+1VFjZqKAyA6HDtEHgRYfDae/VMzNXYePEmDI000KhlWPhtCVz801xn3WZdXyApXoG/9/Eqmz4Uh/hu3DEasVEGGN2+DISQQZ0uQ+s+EegxMuPKgomZBhW9E7BxnhNKlX0AK/t0nNhpjVsXTOHizvimD1dxiG9vnzhsX2GPqnXi4eyeikt/meGffVbQaHRWh0YDLJtcApU/ji8yV3fzE8vW9ul4keVWgRfhBrB2SNdZX/e+WoZx3Upj8q8PsDPoOoQGiI4wwIReHoiPKdaHzu/FhxbflDvvIp5t/v/frNOMDjeAjYPurvIu7in4pH8EfvneRSoLe6jEtz1KY8LyUHwx6xEUBsDN8yb47tPSeVvIAlCsr5xHRERArVbD0dFRq9zR0RHPnj3LVn/GjBmwtLSUXiVLlnxfTc03IQAr23R8MechylZLgu8n0egxMgx719lp1XP1TMGSw3ewYO9dtOkTgR+/cEPoXXbFziopXo6hzcphRKtyWDPLCUMmP0G1uvE66/p3j8KxP6yQllKsw0xvikN8Xzllhs0LHTF8+iMsPngHk1aF4OwRC2z4+eUyj10YCiGAnh9VQRv36ti5yg6+7V9Axp8lfcCKQ3x//sMjlPBIxcBGFdHarTqWTHBF826ROcbuom9dEXrbGOOXhr7fhn7gdO+rBYZPf4zoCAOM7lAGI1uXxakDlpiy5kGOB/xUcIpDfJN+2DqlYdqG+/hzjxX2b3x5e5+1fRpGzXmIw1utMaJVWYzu4Im0VBkm/vIAALu1F1rjx49HTEyM9Hr48OF7nb+FTUbXquhw7YGcXkQYwtpe9xlhG4d0lCidAsXLnhsoVTYZUc8Nte6rNDQSKOGRirLVktD/26fwqJSEnVlGKi5KYqMUUKcDVlnWm7VdOl6E53xWXAgZnjxQ4v4NY2xf7oC/9lqh24iwbPWq1IpHyTIpOLCxaN/XW5R8iPG9drYTmnR6gZa9ouBRMRn1W8ag3/in2LLQUbq65uKeih93BGNX8FWsP38DC/cFIT1NBme3onsFOb/xTUXXhxjfVrZqBK4Owa7gq1h39iZW/nUbKlMNnEplj91F35bAv4ctMHtbMOxdik7ymJ9YfhFuAGu7LPXt0/Hiefb6Oe2rvRrEo1bTWMz43A03z5ki+JoJFn3ritRkGZp21X1vLOmPvuObcuddxHPU//+bdZpW9umIeq69vbVxTMPsrcG4ed4U88doD+zaNiACCXEKrJrqgnvXTXD9XzPMHlEKNRrGo8JH7/d2uGKdnNvZ2UGhUCAsTDu5CgsLg5OTU7b6SqUSFhYWWq/3ydBIoGy1RFz6++X9ZhoNcPlvM1TK4V6qSh8n4OkDpVY3uEf3lbBxTIOhUc5ngoQA0lKL7s8jPU2OoKsmqNHg5SNpZDIBrwbxuJmHgfDkcqFzPfr3iMLdK8a4f5P3pulLcYjvlCQ5ZHLt35/8/9+LLD9LlYkGto7piItW4MJJC9T1L7rd/goqvqnwKg7xnclIJWDnnAZ1OvD3Piut2BUiIzE/dcASs7cGw6lU6jtbBn3ITyzfumACr4baPdo+ahSHWzru7c9pX600zjhoynoLgUbIIH+/t58WSx9afFPuvIt4fvafESLDDLSmaWKmRoUaibj1yjRtndIwZ1swgq6Z4KcvS0II7UBWGQsIjXaZRp3xXv6e06Gim33lgpGREby9vXH06FGpTKPR4OjRo6hbt64eW5azjoPDsX+jLQ7/bo3/gpRYOM4VyYlyNO+ecSZ39shS+HX6y8cotekTgbhoBZZOLIFH95T494gFNi9wRNuACKnOr9Odce2MKZ49NELILRV+ne6Mq6fM4NehaJ8d3rHCDi17RqFplyiULJOMETMfQWWiwaH/HxRmzPz/0G/8U6l+t+Fh+KhRHJxKpaBkmWR0GvIcTTq9wLEd1lrTNTFTo1HbGBzYyIHg9Kk4xHedZrHY+5sdTuy0wrP/jHDhpBnWznFG7WYxUm+Z8yfMce64ufT52M5lULJMMpp3i9TVhCIjr/FtYKhB6cpJKF05CYaGArbOaShdOUnr3nyViVqqAwBOJVNRunIS7EsUrYToQ1Ac4vv2RRP8vc8ST0ONcO1fU0zo5QmhAboOfTla9aJvXXFshw3GLQ6FsZkGUc8NEPXcAClJRSeDzGss71xpj5q+seg05DlKlknGp6OfoWy1JOxarX11/HX76lsXTBEfo8CY+Q9RulISSpROwcCJT+BUMhVnjzLxe9c+xPjOi6QEOe5dN8a96xknhZ49NMK968Z4/qhoP+IUeBfxLMPOlfbo8cVz1GkeA/cKSRiz4D9EhhlKT1bJTMzDnxjhl+9dYGmbDmv7NK3HIv571BzlvBLR68tncPFIQZmqiRj980M8e2iI4Ovv90Jbse/f99VXX6Fv376oWbMmatWqhXnz5iEhIQH9+vXTd9N08v0kGjGRBvhtjjNehBugdOUkTNtwX+oWF/7YSOsMj0OJNEzbeA/LA0vgs6blYeeUhvYDw9F12Mude3SEAeaMdEPUcwOYmKvhUTEZ0zbeg7eP7nupi4qTu61haatGnzHPYG2fjvs3jDGhlweiIzI2jvYlUrXOmqtMNBg+/RHsnNOQmizHw3tKzB5RCid3ayfnPp9EAzKB4zu1y+n9K+rx3XPUM8hkAmtmOyPymSEsbdJRp1kMAsa9vCcvIVaB1TOcEfHUEOZWatRvFY1+457CoIgfA+Q1vm0d07H08F3pfZfPw9Hl83BcOWWKsZ3LAADKVU/CnO33pDqfTXkCADi0xRo/fVnqPSwVvaqox3dqigxrZznj6X9GMDbR4OMmsRi7IBRmli9HYt+zNmP8mDGdymrNa/TP/6F5t6Jxgj2vsXzzvClmDnND32+eIWDcMzwJUWJKf3eE3tE+wH7dvjo2ygATepZGwLinmPX7PSgMBULvqBDYz5094t6TDy2+8+LuFRNpvwIAywNLAACadY3C1/P+01ez3ot3Ec+/L7aHykSDL2Y/gpmFGjfOmWJCr9LSOBIfNcp4XHSJ0qnYePGmVnv8XaoDAK78Y46Zw0qhy9BwdBkajpQkOW5dMMF3vUojNfn9XsuWCZG182Pxs2jRIsyZMwfPnj2Dl5cXFixYgNq1a7/xe7GxsbC0tMSLu6VhYV6sOyEUCH8XL303oUhIF2k4gV2IiYlh1y4wvgsLxnfBYHxrY3wXDozvgsH41sb4LhwY3wUjt/HN5PwtMPgLFoO/YHDnXjAY3wWL8V0wGN8Fg/FdsBjfBYPxXTAY3wWL8V0wchvf/MUSERERERER6RmTcyIiIiIiIiI9Y3JOREREREREpGdMzomIiIiIiIj0jMk5ERERERERkZ4xOSciIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50RERERERER6xuSciIiIiIiISM+YnBMRERERERHpGZNzIiIiIiIiIj1jck5ERERERESkZ0zOiYiIiIiIiPSMyTkRERERERGRnjE5JyIiIiIiItIzJudEREREREREesbknIiIiIiIiEjPmJwTERERERER6RmTcyIiIiIiIiI9Y3JOREREREREpGdMzomIiIiIiIj0jMk5ERERERERkZ4xOSciIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPTPQdwOKgnYDPoWBgUrfzfjgNbxyRt9NKBJS4tNwop6+W1F0fP6wLozMjPTdjA9eUvuK+m5CkZCelgzs2aXvZhQZve77wdCU8f224rp56rsJRUJ6WjKwnfFdUFbHOsFYw1TnbclqVtF3E4oEmToZuPjm+OaVcyIiIiIiIiI9Y3JOREREREREpGdMzomIiIiIiIj0jMk5ERERERERkZ4xOSciIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50RERERERER6xuSciIiIiIiISM+YnBMRERERERHpGZNzIiIiIiIiIj1jck5ERERERESkZ0zOiYiIiIiIiPSMyTkRERERERGRnjE5JyIiIiIiItIzJudEREREREREesbknIiIiIiIiEjPmJwTERERERER6RmTcyIiIiIiIiI9Y3JOREREREREpGdMzomIiIiIiIj0jMk5ERERERERkZ4xOSciIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9MxA3w2gvGvX7Ba6trkOG8sk3PvPGovW1sGde/Zv/J5v3fv4bsRJ/HO+FCbPbQIAUCg06NflAmp7PYKTQzwSkgxx6boLVm6qichok3e9KHr3dLMcT9YqkBoBmJYT8BinhnlVobPu811yBE/SDhmZkUDdc2nSe3UiEDpPgajjcqTHAMoSgHMPNZy6at7pclDRlrAtFfHrU6GOEjAsI4flaBWMKit01o34PBGpl9TZypX1FLCdW/Rj+nU6NLqB7s2uwsYiCfce2WD+7/VwK9RBZ91GXiH41P8yStjHwkChwaPnFthytBoOnS37nltNRUnajiSkbU6EiNJA7mkAoy/MoKhkmGN9EadB6i8JUP+ZChGngcxRAaMRpjCoq8z4PFGD1JWJUP+VAvFCA3lZAxiNNIOiYs7TLCo61r+OXo2vwMY8CcFPbDF3R33c+k93PPtUvY8+zS7B1S4WBnINHkZYYvOJajhwvpxWnQ71b6G8azgsTVPQd04nBD2xe1+LQ0XAvQ3GuPurCZIj5LCskA6vCXGwqZaeY/2gtca4v9kYiU8VUFprUKJ5Cqp8FQ+FMnvdO7+Y4PpcM5TpnYjq38a/w6UoHNq2uoPOHW7B2joJ90OssWRFTdwN0h2PbiWj0bvXVZT1jIKjYwKWrfTGzt0VstWztUnEgIBLqPnREyiVajx5aoa5C+oiKNj2XS9OnjA5/8D41rmPzz49i/m/1sOtYHt0ankDM8cdQr/RHREda5zj9xzt4jCk5zlcveWoVa4ySkdZjyis/8ML9/6zgblpCob2+Rfff30Ew75r964XR68iDsjx4EcFSn+nhnlVDZ5uUODm5waosSsNRjnEqcJMoMaul8k4ZNqfP/hRgZizcpSdng6li0D0aTnuT1fAyEHAxld30k/0OkmH0xAzPwVW36hgWFmOhM1piByVCIctplDYZO/8ZDPTGCL95W9NEyMQ3jsRxo2L/sH66zT2vodhnc7gp00NcPOBA7o0vo4fR+xHr8CuiI7Pvu2MTVBi3QEv/BdmhbR0BepV/Q/jep/EizgVzt0qqYcloA9d+tFkpC6Oh9FocygqGSBtaxKSv46ByQYbyKyzx7JIE0geHQOZlRzKHywgs5NDhKkBs5d1U2bFQxOSDuUEc8jsFEg/lIzkr2Jg/Js15Pa6T+AVBU28gjGy/WnM2doQN0Id0c3nKn4eshc9ZnTHC13xnKjC2sMfITTMCulqOepX/g/fdj+BF3HG+PdORjwbK9Nx5b4Tjl4qjfHd/3zfi0QfuIf7lLg6yww1AuNgUy0NQb+Z4O9BVmi+LxIq2+zHf//tUeL6XDN4T4uFbY00xD8wwPnx5oAMqD5OO/mOumaA+1uMYVk+Ldt0iqJGDR5g0ICLWLikFu7ctUP7drcxbcpxDPy8LWJiVNnqK5VqPHtmhr/+KYUhAy7onKaZaQrmzjqEK9cc8d0UP8TEqlDCOQ7x8UbvenHyrFh3a//zzz/Rtm1buLi4QCaTYefOnfpu0ht1anUD+46Xw8GTZfHfYyvMW1UPKSkGaOETlON35DINxg/7E2u318DT5+ZanyUkGeGbGf44+a8HHj21xK1gByxaUwflS0fCwbZon5l7sk4Ox44aOLbXwMQTKP2dGgoV8Hzna8JCBhjZvfLKksTHXpbBvq0alh8LqEoATp01MC0nEH+9WIeaXnyI8a1L/KZUmHxiCJM2hjD0UMDyGyVkKhkS9+jeScstZVDYyqVXylk1ZEpA1aR4n4vt2vga9vxTAfvPlEfoM2v8tKkBklMN0LreHZ31Lwe54K8rHgh9Zo0nERbYdrwK7j+2QTXPsPfcctLlQ4zvtN+TYNBGBcNWKsjdDWA02gwylQxpe5N11k/flwwRq4FyugUUVQ0hd1ZA4WUERZmMWBYpAuo/U2D0uSkUXkaQuypg1N8U8hIKpO/UPc2iorvvNew+XRF7z1bAgzBrzN7aCCmpBmhT+7bO+pfuueDPax4IfW6Nx5GW+P3Pqrj31BbVSj+T6hw4Xw6rD3nj3F3X97UYlIMPMb6D1prAvUsS3Dsmw6KMGh8FxkGhEgjdofvCWeQlQ9h+lIZSbVJgWkIDx/qpKNk6BS+uae+r0xNkODfGAh99HwtDi+JxkafjJ7dx4FAZHD7qif8eWmLhklpISVHAv+k9nfXvBtti5ZqPcPIvd6Sl6T4p2aXTTYRHmGDugrq4G2SHsDAzXLzsjKfPzHXW16dinTEkJCSgevXqWLx4sb6bkisGCjXKeUTi4nUXqUwIGS5ed0alss9z/N6nHa8gOlaFAyfK5VjnVaYmadBogPjEwnc2qaBo0oD4WzJY1nnZ3VwmByzraBB3NeewUCcC51sY4nxzQ9z6wgCJwdqXzi28BKJOypESBggBxJyVISlUBsu67Nb+vn1o8a2LSBNIu6OB8uOXOxuZXAblxwqkXcvdbyrxf2kwbmYIubHszZWLKAOFGuVKReD8nRJSmRAyXLhdApU9ct52viTwUfnHKOkYgyvBTu+uoZRrH1p8izQBzd10KGq+3K/K5DIovA2huaH7RJv671TIKxsi9ed4JHwSgcS+UUhdlwCh/v8DdLUA1IDMKEtsKwH1taJ7hc1AoUZ513Ccv6sdz+eCXFHFLTcnzwS8yz5CKftoXL7n/O4aSvn2ocW3JhWIvmEAh7qpUplMDjjUTUXkZd291mxrpCH6hgGirmYk4/EP5Xj2pxGcGqVq1bv0gxmcfFLhWK/oxvSrDAzUKFsmCpcuv9zXCiHDpStOqFghIt/TrVPrEe4G22LCN39h82/bsGjePrRoHlwQTS5wxfpSSsuWLdGyZUt9NyPXLM1ToFAIvIjRPgv3IsYYJV1idH6nSvkwtPS9iyHffpKreRgapmNgj/M4fro0EpOKbnKe/gKAWpbtyrehLZAUovs7xu4CZaaoYVpWID0eeLJWgWt9DeC1Iw3K/79bwGOcGve+V+BCcyPIDAQgAzwnq2HpXTzOdhYmH1p866KJzjj4ztp9XW4tQ+qD7PeVZ5V6Q430expYfZu9G1hxYmmWDAOFwIsst/5ExRmjlGN0jt8zVaVi+/QNMDJUQ62R4+fN9XH+Nq+qFQYfWnyLGE1GIp2l+7rMRg7Nf7oPujVP1RCX1DBoqoJqtiXEIzVSfo4H0gGjfqaQmcghr2yA1LWJULopILOWQ300BZob6ZCVKLpd2q1MM+I5Ki57PLs5ROf4PVNVCnYFroeRgQZqjQw/bmvAq+SF1IcW3ynRcgi1DCpb7ZPmKlsN4kJ0p1ql2qQg9YUcJz61BgQg0mXw6JaICkMSpToP9yoRfdMQjbdGvdP2FyYWFhm5TnS09nFLdLQKJUvE5nu6zk7xaNPyLnbsqojNWyujXNlIfD7oPNLT5ThyrPTbNrtAFevkPK9SUlKQkpIivY+Nzf+P5H0wVqXhm8//xNyV9REb9+aDc4VCg4kjT0AGgfm/1n0PLfywmFcXMK8uXnmfjssdDBG2VYFSwzMSpaeb5Ii7KkeF+WlQugCxF2QZ95zbC1jVYYJemH1o8Z0bif9Lg4GnPMfB4+j1ElMMMWBGRxgr0+Fd/jGGdTqDJxHmuBzk8uYvU6HyQca3BpBZyWE0xgwyhQwobwgRoUHapiQY9TMFACi/s0DKzDgkdYwCFIC8rAEUTZTQ3Ml5EKriKjHFCH1/7AwTozTULPcYI9ufxpNIC1y6x3j+0H2I8R1+1hC3V5igxsQ42FRPQ3yoAldmmOPWEg0qDk1E4lM5rswwR8NVL3QOEEd5I5MBQcE2WLPOCwBw774N3EvFoHWLICbnH7IZM2ZgypQpept/TJwSarUM1pZJWuXWlkl4EZ39nhYXx1g4O8Rj6tdHpDKZLCNBPLhuDQJGd8TT5xYAMhPz43C0i8eYaS2K9FVzADCwBqAQSI3ULk+LBAxzOTir3BAwrSCQ/DDjvToZ+G+BAuV/TodNo4z1bFpOIOGODE/WKmBVhwdLhZm+41sXuZUMUADqKA2Alwm25oWAwvb1dyVpkgSSDqfBfDD36jHxKqSrZbC20N522pgnISo25xHshZDhcbglACD4kS3cnKLxqf9lJucfIH3Ht8xSDigA8UL7ypqI0kCmY2BHAJDZygEDZCTmmWVuCogoDUSagMxQBnkJBYwXWkEkCYgEDeR2CiRPjoXcpeiekItOyIhnG3Nd8ZzzwLhCyPA4IiOeg57Ywc0xGn2aXmJyXgToO76VVhrIFALJkdqxnBwph8pO9y1oNxaYolS7ZHh0yRgfwrKcGuqkeFycbIEKnyXixQ0DpETKcbSTjfQdoZYh4rwh7m00Rocr4ZAVwTCPjc3IdaystMfNsLJK1pnr5FbUCxX+e2ipVfbfIwvUr/dfvqf5rhTre87zavz48YiJiZFeDx8+fK/zT1crcDfEFh9VfiqVyWQCNSo/xc2g7I8P+e+JJQaObY8h4z+RXqcvlsLlm84YMv4ThEdmnHnPTMxLOMVi7PQWiI0v+l1g5YaAWUWBmH9fhoDQADH/ymFeLXf38go1kBgkk5J5kZ7RLUmWJapk8oxpU+Gm7/jWRWYog2F5OVLPvezCLjQCKefUMKz6+s138tF0iDTApEXxHqUd+P9t53928C7/WCqTyQQ+Kv8EN0J0P3pJF5lMwNCAwfwh0nd8ywxlkJczgPrCy/tJhUZAfTEN8sq6Y1RR1QDisRpC87LXlXiohsxWDpmh9n3mMmMZ5HYKiDgN1OdSoWhQdE+wp6sVuPPIHt7ltOO5ZtnHuB7q+JpvapPLBAwN3nx7EBV++o5vuRFgVTkd4Wdexp3QAOFnjGDrlcOYEkkyyLIMFyEl2wJwqJuGprsi0WRHlPSyrpIxgFyTHVFFMjEHgPR0BYKCbeBV/eVgjTKZgFe1Z7h1O/+PNrx5yx6uWbrFl3CJw/Pnpvme5rvCK+d5oFQqoVTq9yrU9n2VMfazv3Hnvi3u3LNHx5Y3oFKl48DJjGfvfvP5n4iIMsGqLTWRlmaAB4+stb4fn5Cx4cgsVyg0mPzFMZTxiMR3c5pBLtfA2jLjfpe4eCXS1UU0+gG49NYgaKICZpUFzKpo8HS9AuokwKF9xsF30AQFjBwAty8ydt4Pl8lhXk1AVUogPU6GJ2vkSHkKOHbM+NzADLCoqcGDuQrIlWoonQViL8gRvkcO9695AFDYFYb41sWshxFe/JAMw4oKGFaSI2FLGkSygEnrjAP6F1OSoLCXw2KodtsT/5cGVSMDyC2L70Bwr/r9WFWM73MSd0LtcSvUHl38rsNYmYZ9pzMGyvy273FERJtixa5aAIBe/pdxJ9QOj8MtYGSoRp3KD+FfOwg/bWqgz8WgfCoM8W3Y1RgpM+IgL28IRcWMR6mJJAHDVhknxFOmxUJmJ4fREDMAgMEnxkjbkYzUBfEw7GQMzSM1UtcnwrDTy6tH6WdTAQHISyogHquRujQe8lIKGLQq2ifZN5+oiu96nsDth/a4GeqAbj7XoDJKw55/ywMAJvY8hvAYUyzbWxsA0LvJJdx+aI/HkRYwVKhRr9J/aFEzCHO2voxnc5NkOFnFw+7/j4FK/f/965FxJoiKy7mHDelfYYjvsn0TcX68BayrpMO6ahqCfzNBepIMbh0yenic+8Ycxo4aVPkqAQDg7JeKoDXGsKqYLnVrv7HAFM6+KZApAENTActy2seOCmMBIytNtvKiZseuCvh61GkEBdvizl1bdGh3GyqVGoeOZnQ//3rUKURGGWP1bzUAZAwiV6pkzP//XwM7m0SU9ohCUrIhnj7NGI39j10VMXf2QXTrch1//u2G8mUj0Mo/CPMX19bPQr4Gk/MPzIkzpWFpkYyAzpdgbZWEe6E2GD+zufSMcwfbBGg0uT8Yt7NOQL2aGWcYV8zcpfXZ6B9a4MqtojuSqV0LDdJeAP8tUSAtQgHT8gKVlqRLg8SlPJMB8pdXLNLjZAj+XoG0CMDAAjCtJFBlbTpMPF9Os9ysdITOVyBovAHSYwGlM1BquBqOXXi1jfLHuJkhNNECcb+kQB0pYFhWDtufTaRu7epnApBp/77SQzVIvaKGzfz8dwErao5d8ISVWTL6t7kAG4tEBD+yxdeLWuLF/x90O1onQLyy7TQ2SsNX3f+BvVUCUtIM8F+YJaau8cOxC545zYLotQyaqCCiBdJ+TUBqlAbyMgZQ/WgpdWvXhGkgf2X3LXdUQPWjJVIXxSOp3wvI7OQw7GwMw56vJIrxGqSuSIAI10BmLofCxwhGg0whMyjaJ+WOXi4DK7NkDGpxHjYWiQh6bIevlrfCi/jMeI6HRmjH89ed/4KDZUY8hz63wpT1fjh6uYxUp2HlUHzX84T0/oe+RwEAqw54Y9XBmu9nweiDVbJVClJexOPmAlMkR8hhWTEdDVZEQ2WXcRyZ+FSh1bOywmcJgEzgxgJTJIUpoLTRwNk3BZVHJehpCQqPP/92h6VlCnr3vAJr62Tcv2+N7wL9EP3/3dod7BMgXolvW5skLJm/X3rfueMtdO54C1evOWDshGYAMh639v30RujX5zJ6dbuGZ2FmWLayJo6f9Hi/C5cLMiFEsR2lKj4+HsHBGcPo16hRA3PnzoWfnx9sbGxQqlSpN34/NjYWlpaWaNhwEgwMivZZ6veh4fwz+m5CkZASn4ZZ9fYjJiYGFhYW+m6O3hRUfHc/2gtGZkW3i+j7cm9mRX03oUhIT0vGv3smMb4LKL5bHRgAQ1PG99t6tpgnrQpCeloyLmz/jvFdQPE993w9GJvxOuTb2tG7sb6bUCSkq5Nx/OLMN8Z3sf7Fnj9/Hn5+ftL7r776CgDQt29frFmzRk+tIqKCwPgmKroY30RFF+ObirNinZz7+vqiGHccICrSGN9ERRfjm6joYnxTccbR2omIiIiIiIj0jMk5ERERERERkZ4xOSciIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50RERERERER6xuSciIiIiIiISM+YnBMRERERERHpGZNzIiIiIiIiIj1jck5ERERERESkZ0zOiYiIiIiIiPSMyTkRERERERGRnjE5JyIiIiIiItIzJudEREREREREesbknIiIiIiIiEjPmJwTERERERER6ZlBbip5eHhAJpPlacIymQz37t3LV6OIiIiIiIiIipNcJec+Pj55Ts6JiIiIiIiIKHdylZyvWbPmHTeDiIiIiIiIqPjiPedEREREREREepbv5Dw2NhYzZ86Ev78/atSogbNnzwIAoqKiMHfuXAQHBxdYI4mIiIiIiIiKslx1a8/q0aNH8PHxwcOHD1G2bFncvn0b8fHxAAAbGxssX74coaGhmD9/foE2loiIiIiIiKgoyldyPmbMGMTFxeHy5ctwcHCAg4OD1uft27fHnj17CqSBREREREREREVdvrq1Hzp0CCNHjkSlSpV0juJeunRpPHz48K0bR0RERERERFQc5Cs5T0pKgr29fY6fx8XF5btBRERERERERMVNvpLzSpUq4c8//8zx8507d6JGjRr5bhQRERERERFRcZKv5HzUqFHYvHkzZs2ahZiYGACARqNBcHAwevfujdOnT+PLL78s0IYSERERERERFVX5GhDu008/RWhoKL777jtMmDABANCiRQsIISCXyzF9+nS0b9++INtZKAkhAADp6Sl6bknRkBKfpu8mFAkpCRnrMfP3SfmTuf7SEvi7LAjpacn6bkKRkLkeGd9v52V8p+q5JUUD47tgqBnfBSJz/SXHp+u5JUVDuprxXRDS1Rn54pviWybeYgvw33//Yfv27QgODoZGo4Gnpyc6duyI0qVL53eSH5RHjx6hZMmS+m4GkU4PHz6Eq6urvpvxwWJ8U2HG+H47jG8qzBjfb4fxTYXZm+L7rZLz4k6j0eDJkycwNzfXOWp9YREbG4uSJUvi4cOHsLCw0HdzPlgfynoUQiAuLg4uLi6Qy/N15wqB8V3cfCjrkfFdMBjfxcuHsh4Z3wWD8V28fCjrMbfxna9u7ZmuX7+Offv24cGDBwAADw8PtGjRAlWrVn2byX4w5HL5B3Vm08LColD/aD8UH8J6tLS01HcTPniM7+LpQ1iPjO+3x/gunj6E9cj4fnuM7+LpQ1iPuYnvfCXnKSkpGDJkCNatWyfdZw5knKkaN24cevXqhZUrV8LIyCg/kyciIiIiIiIqVvLVZ+abb77Bb7/9hs8//xy3bt1CcnIyUlJScOvWLXz22WdYv349xo4dW9BtJSIiIiIiIiqS8nXlfP369ejduzcWLVqkVV6+fHksXrwYsbGxWL9+PebNm1cQbaS3pFQqMXnyZCiVSn035YPG9UiFEX+XBYPrkQoj/i4LBtcjFUb8XRaMorYe8zUgnKWlJWbOnInPP/9c5+dLly7F+PHjER0d/bbtIyIiIiIiIiry8tWt3d/fHwcPHszx8wMHDqB58+b5bhQRERERERFRcZKrK+dRUVFa78PDw9G1a1d4enpi2LBhKFOmDAAgKCgIixcvRkhICLZs2YLy5cu/m1YTERERERERFSG5Ss7lcnm25wRmfi2ncrlcjvT09IJqJxEREREREVGRlasB4SZNmpQtCSciIiIiIiKiAiKoSFu0aJFwc3MTSqVS1KpVS/z777/6btIH5+TJk6JNmzbC2dlZABB//PGHvptEJIRgfBcExjcVVozvt8f4psKK8f32imp852tAOPowbNmyBV999RUmT56Mixcvonr16vD398fz58/13bQPSkJCAqpXr47FixfruylEEsZ3wWB8U2HE+C4YjG8qjBjfBaOoxne+HqWW6Z9//sHFixcRExMDjUajPWGZDBMnTnzrBlL+1a5dGx9//LH0PHqNRoOSJUtixIgRGDdunJ5b92GSyWT4448/0L59e303hYo5xnfBY3xTYcH4LniMbyosGN8FryjFd67uOc8qKioKrVu3xtmzZyGEgEwm0xogLrOMybn+pKam4sKFCxg/frxUJpfL0bRpU5w+fVqPLSOit8X4Jiq6GN9ERRfjm94kX93ax4wZg6tXr2Ljxo24f/8+hBA4ePAg7t69i88++wxeXl548uRJQbeV8iAiIgJqtRqOjo5a5Y6Ojnj27JmeWkVEBYHxTVR0Mb6Jii7GN71JvpLzffv2YciQIejWrRvMzc0zJiSXo0yZMli8eDHc3d0xatSogmwnERERERERUZGVr+Q8OjoalStXBgCYmZkBAOLj46XPmzdvjoMHDxZA8yi/7OzsoFAoEBYWplUeFhYGJycnPbWKiAoC45uo6GJ8ExVdjG96k3wl5y4uLlLXC6VSCQcHB1y5ckX6/PHjx3wuup4ZGRnB29sbR48elco0Gg2OHj2KunXr6rFlRPS2GN9ERRfjm6joYnzTm+RrQLhGjRrh8OHDmDBhAgCgW7dumD17NhQKBTQaDebNmwd/f/8CbSjl3VdffYW+ffuiZs2aqFWrFubNm4eEhAT069dP3037oMTHxyM4OFh6HxISgsuXL8PGxgalSpXSY8uoOGN8FwzGNxVGjO+CwfimwojxXTCKbHzn5+HoV69eFT/99JNITk4WQggRFRUlmjRpImQymZDJZMLHx0c8fvy4gB7FTm9j4cKFolSpUsLIyEjUqlVLnDlzRt9N+uAcP35cAMj26tu3r76bRsUc4/vtMb6psGJ8vz3GNxVWjO+3V1Tj+62ec55VdHQ0FAqFNEgcEREREREREb1Zvu45z4mVlRXMzc2xceNGNG/evCAnTURERERERFRkFWhynikkJERroAMiIiIiIiIiytk7Sc6JiIiIiIiIKPeYnBMRERERERHpGZNzIiIiIiIiIj1jck5ERERERESkZwa5rVitWrVcT/T58+f5agwRERERERFRcZTr5NzGxgYymSxXdW1tbVGxYsV8N4qIiIiIiIioOJEJIYS+G0FERERERERUnPGecyIiIiIiIiI9Y3JOREREREREpGdMzinXAgIC4O7uru9mENE7tGbNGshkMjx48EDfTSGiLHTth2UyGQIDA/XSnqKI65NI24kTJyCTybBt27bX1uPxQ8Fgcl4IZf64M18GBgYoUaIEAgIC8PjxY303r9DIup5efY0bN07fzdNp+vTp2Llzp76bQcXEkiVLIJPJULt27fcyv4CAgGzbrpIlS6J79+64efPme2nD69y8eROBgYE8cKA8CwkJwfDhw1GuXDmYmJjAxMQElSpVwrBhw3D16lV9N++d27hxI+bNm5fr+u7u7lrbApVKhbJly2LMmDGIiop6dw3NpX379jEBp0Lv2rVr6Ny5M9zc3KBSqVCiRAk0a9YMCxculOrwuLLoyfVo7fT+ff/99/Dw8EBycjLOnDmDNWvW4O+//8b169ehUqn03bxCI3M9vapKlSp6as3rTZ8+HZ07d0b79u313RQqBjZs2AB3d3ecPXsWwcHBKFOmzDufp1KpxMqVKwEA6enpuHfvHpYtW4YDBw7g5s2bcHFxeedtyMnNmzcxZcoU+Pr6shcQ5dqePXvQrVs3GBgYoFevXqhevTrkcjlu376NHTt2YOnSpQgJCYGbm5te2peUlAQDg3d7OLdx40Zcv34do0aNyvV3vLy8MHr0aABAcnIyLly4gHnz5uHkyZM4e/bsO2pp7uzbtw+LFy/WmaC/j/VJ9CanTp2Cn58fSpUqhUGDBsHJyQkPHz7EmTNnMH/+fIwYMQJA4Tqu7N27N7p37w6lUqnvpnzQuPUpxFq2bImaNWsCAAYOHAg7OzvMmjULu3fvRteuXfXcusLj1fVUkBISEmBqalrg0yV6H0JCQnDq1Cns2LEDQ4YMwYYNGzB58uR3Pl8DAwN8+umnWmV16tRBmzZtsHfvXgwaNOidt4GooNy7dw/du3eHm5sbjh49CmdnZ63PZ82ahSVLlkAuf31HxHe5PymsJ+tLlCihtS0YOHAgzMzM8OOPPyIoKAhly5bVY+tyVljXJxUv06ZNg6WlJc6dOwcrKyutz54/f66fRr2BQqGAQqHQdzM+eG/Vrf3x48fYtGkT5s+fj0ePHgEA1Go1oqKioFarC6SB9FLDhg0BZBwsZEpNTcWkSZPg7e0NS0tLmJqaomHDhjh+/LjWdx88eACZTIYff/wRK1asgKenJ5RKJT7++GOcO3cu27x27tyJKlWqQKVSoUqVKvjjjz90tikhIQGjR49GyZIloVQqUb58efz444/I+oQ+mUyG4cOHY+vWrahUqRKMjY1Rt25dXLt2DQCwfPlylClTBiqVCr6+vgXa7fTYsWNo2LAhTE1NYWVlhU8++QS3bt3SqhMYGAiZTIabN2+iZ8+esLa2RoMGDaTP169fD29vbxgbG8PGxgbdu3fHw4cPtaYRFBSETp06wcnJCSqVCq6urujevTtiYmKkdZCQkIC1a9dKXf0CAgIKbDmJXrVhwwZYW1ujdevW6Ny5MzZs2JCtzo0bN9C4cWMYGxvD1dUVU6dOhUajyVZv165daN26NVxcXKBUKuHp6Ykffvgh19t5JycnAMh2Ner+/fvo0qULbGxsYGJigjp16mDv3r3Zvv/8+XMMGDAAjo6OUKlUqF69OtauXZut3ubNm+Ht7Q1zc3NYWFigatWqmD9/PoCM22C6dOkCAPDz85Ni8MSJE7laBiqeZs+ejYSEBKxevTpbYg5k/KZHjhyJkiVLSmUBAQEwMzPDvXv30KpVK5ibm6NXr14AgL/++gtdunRBqVKloFQqUbJkSXz55ZdISkrKNu3c7od13SP9+PFj9O/fH46OjlAqlahcuTJ+/fVXrTqZ95H+/vvvmDZtGlxdXaFSqdCkSRMEBwdL9Xx9fbF3716EhoZKcZPfnic5bQtys58GgEuXLqFly5awsLCAmZkZmjRpgjNnzmjVSUtLw5QpU1C2bFmoVCrY2tqiQYMGOHz4MICMv8/ixYuldZf5ypR1fWYeHwQHByMgIABWVlawtLREv379kJiYqDXvpKQkjBw5EnZ2djA3N0e7du3w+PFj3sdOeXbv3j1Urlw5W2IOAA4ODgBef1wZGhqKoUOHonz58jA2NoatrS26dOmi8/g6OjoaX375Jdzd3aFUKuHq6oo+ffogIiIix/alpKSgTZs2sLS0xKlTpwDovufc3d0dbdq0wd9//41atWpBpVKhdOnS+O2337JN8+rVq/Dx8dE6Jlm9enWxu489X1fOhRAYPXo0Fi1ahPT0dMhkMlStWhWurq6Ij4+Hu7s7vv/++zx1f6I3y/xhWltbS2WxsbFYuXIlevTogUGDBiEuLg6rVq2Cv78/zp49Cy8vL61pbNy4EXFxcRgyZAhkMhlmz56Njh074v79+zA0NAQAHDp0CJ06dUKlSpUwY8YMREZGol+/fnB1ddWalhAC7dq1w/HjxzFgwAB4eXnh4MGDGDNmDB4/foyff/5Zq/5ff/2F3bt3Y9iwYQCAGTNmoE2bNhg7diyWLFmCoUOH4sWLF5g9ezb69++PY8eO5Wq9xMTEZNuA2NnZAQCOHDmCli1bonTp0ggMDERSUhIWLlyI+vXr4+LFi9kOMLp06YKyZcti+vTp0gmGadOmYeLEiejatSsGDhyI8PBwLFy4EI0aNcKlS5dgZWWF1NRU+Pv7IyUlBSNGjICTkxMeP36MPXv2IDo6GpaWlli3bh0GDhyIWrVqYfDgwQAAT0/PXC0jUV5t2LABHTt2hJGREXr06IGlS5fi3Llz+PjjjwEAz549g5+fH9LT0zFu3DiYmppixYoVMDY2zjatNWvWwMzMDF999RXMzMxw7NgxTJo0CbGxsZgzZ062+pnxqFarcf/+fXzzzTewtbVFmzZtpDphYWGoV68eEhMTMXLkSNja2mLt2rVo164dtm3bhg4dOgDIONj19fVFcHAwhg8fDg8PD2zduhUBAQGIjo7GF198AQA4fPgwevTogSZNmmDWrFkAgFu3buGff/7BF198gUaNGmHkyJFYsGABvv32W1SsWBEApH+JdNmzZw/KlCmT53Eb0tPT4e/vjwYNGuDHH3+EiYkJAGDr1q1ITEzE559/DltbW5w9exYLFy7Eo0ePsHXrVun7ud0P6xIWFoY6depIJ8Xt7e2xf/9+DBgwALGxsdmOzWbOnAm5XI6vv/4aMTExmD17Nnr16oV///0XADBhwgTExMTg0aNH0n7dzMzsje1IS0uTtgXJycm4dOkS5s6di0aNGmndipbb/fSNGzfQsGFDWFhYYOzYsTA0NMTy5cvh6+uLkydPSn+jwMBAzJgxQ9rfxsbG4vz587h48SKaNWuGIUOG4MmTJzh8+DDWrVv3xuXI1LVrV3h4eGDGjBm4ePEiVq5cCQcHB2l7A2Qk/r///jt69+6NOnXq4OTJk2jdunWu50GUyc3NDadPn8b169dzvFXzdceV586dw6lTp9C9e3e4urriwYMHWLp0KXx9fXHz5k1pmxQfH4+GDRvi1q1b6N+/Pz766CNERERg9+7dePTokXQ8/aqkpCR88sknOH/+PI4cOSIdV+QkODgYnTt3xoABA9C3b1/8+uuvCAgIgLe3NypXrgwg44Ri5onz8ePHw9TUFCtXriyeXeRFPsyaNUvI5XIxfvx4ceTIESGTycTRo0elz/v27SsaNGiQn0mTEGL16tUCgDhy5IgIDw8XDx8+FNu2bRP29vZCqVSKhw8fSnXT09NFSkqK1vdfvHghHB0dRf/+/aWykJAQAUDY2tqKqKgoqXzXrl0CgPjf//4nlXl5eQlnZ2cRHR0tlR06dEgAEG5ublLZzp07BQAxdepUrfl37txZyGQyERwcLJUBEEqlUoSEhEhly5cvFwCEk5OTiI2NlcrHjx8vAGjVfd160vV6dVkcHBxEZGSkVHblyhUhl8tFnz59pLLJkycLAKJHjx5a83jw4IFQKBRi2rRpWuXXrl0TBgYGUvmlS5cEALF169bXttnU1FT07dv3tXWI3tb58+cFAHH48GEhhBAajUa4urqKL774QqozatQoAUD8+++/Utnz58+FpaVltvhLTEzMNo8hQ4YIExMTkZycLJX17dtXZzyWKFFCXLhwQev7mfP/66+/pLK4uDjh4eEh3N3dhVqtFkIIMW/ePAFArF+/XqqXmpoq6tatK8zMzKRtxxdffCEsLCxEenp6jutl69atAoA4fvz4a9YeUYaYmBgBQLRv3z7bZy9evBDh4eHS69UYyYyDcePGZfuerliaMWOGkMlkIjQ0VCrL7X5YiIz96+TJk6X3AwYMEM7OziIiIkKrXvfu3YWlpaXUhuPHjwsAomLFilrHEfPnzxcAxLVr16Sy1q1bZ5vv67i5uencFtSvXz9bu3K7n27fvr0wMjIS9+7dk8qePHkizM3NRaNGjaSy6tWri9atW7+2fcOGDdM6VnhV1vWZeXzw6jGVEEJ06NBB2NraSu8vXLggAIhRo0Zp1QsICMg2TaI3OXTokFAoFEKhUIi6deuKsWPHioMHD4rU1FStejkdV+ra1pw+fVoAEL/99ptUNmnSJAFA7NixI1t9jUYjhHi5rdi6dauIi4sTPj4+ws7OTly6dEmrfuZx+avHD5nbgj///FMqe/78uVAqlWL06NFS2YgRI4RMJtOaZmRkpLCxsclVTlCU5Ktb+y+//II+ffpg+vTp2a7MAkC1atVw9+7d/EyaXtG0aVPY29ujZMmS6Ny5M0xNTbF7926tM+cKhQJGRkYAAI1Gg6ioKKSnp6NmzZq4ePFitml269ZN68p7Zlf5+/fvAwCePn2Ky5cvo2/fvrC0tJTqNWvWDJUqVdKa1r59+6BQKDBy5Eit8tGjR0MIgf3792uVN2nSROtKdeZZ7k6dOsHc3DxbeWab3mTx4sU4fPiw1uvVZQkICICNjY1Uv1q1amjWrBn27duXbVqfffaZ1vsdO3ZAo9Gga9euiIiIkF5OTk4oW7asdPtA5ro6ePBgtm5uRO/bhg0b4OjoCD8/PwAZXd+6deuGzZs3S13R9+3bhzp16qBWrVrS9+zt7aXut6969Wp6XFwcIiIi0LBhQyQmJuL27dtadVUqlRSHBw8exPLly2FmZoZWrVpp7Rf27duHWrVqad0+YmZmhsGDB+PBgwfS6O779u2Dk5MTevToIdUzNDTEyJEjER8fj5MnTwIArKyskJCQIMU/0duKjY0FoPsqsa+vL+zt7aVXZjfpV33++efZyl6NpYSEBERERKBevXoQQuDSpUsA8rYfzkoIge3bt6Nt27YQQmjtt/z9/RETE5Pt2KBfv37ScQSQ/bggv2rXri1tC/bs2YNp06bhxo0baNeundSNP7f7abVajUOHDqF9+/YoXbq0VM/Z2Rk9e/bE33//Lf29rKyscOPGDQQFBb1V+7PKenzQsGFDREZGSvM9cOAAAGDo0KFa9TIH7iLKi2bNmuH06dNo164drly5gtmzZ8Pf3x8lSpTA7t273/j9V7c1aWlpiIyMRJkyZWBlZaW1Ddi+fTuqV68u9VZ71au3ewAZPVWbN2+O27dv48SJEzpzQF0qVaokbVeAjGON8uXLa21jDhw4gLp162pN08bGRucxSVGXr+T84cOHqFevXo6fm5qaShsryr/MpHPbtm1o1aoVIiIidHbvWLt2LapVqybdW2Vvb4+9e/dK9zq/qlSpUlrvMxP1Fy9eAMi4RwWAzoFaypcvr/U+NDQULi4uWok18LKbaOa0cpp35kHHq/fqvVqe2aY3qVWrFpo2bar1enX+Wdud2caIiAgkJCRolWcd9T0oKAhCCJQtW1brQMze3h63bt2SBuXw8PDAV199hZUrV8LOzg7+/v5YvHixzr8B0bukVquxefNm+Pn5ISQkBMHBwQgODkbt2rURFhaGo0ePAsiIj9zEOZDRnbRDhw6wtLSEhYUF7O3tpYGesv7GFQqFFIfNmzfH4MGDceTIEcTExGD8+PFSvdDQ0BxjM/PzV9uZdcCtrPWGDh2KcuXKoWXLlnB1dUX//v2lg2Wi/Mjct8XHx2f7bPny5Th8+DDWr1+v87sGBgY6u6D/999/UiJqZmYGe3t7+Pj4AHgZS3nZD2cVHh6O6OhorFixIts+q1+/fgCyDyb1puOC/LKzs5O2Ba1bt8a3336LlStX4tSpU9ITHXK7nw4PD0diYmKO9TQajTQOzPfff4/o6GiUK1cOVatWxZgxYwrkcXe5OX6Sy+XZjiPex1MyqGj6+OOPsWPHDrx48QJnz57F+PHjERcXh86dO7/x8aRJSUmYNGmSNCaUnZ0d7O3tER0drbXfvnfvXq6fcDRq1CicO3cOR44ckbqj50bW2AEy4ufVbUxoaKjOWCmO8ZOve84dHByyDYb1qgsXLuj8Q1De1KpVSxqFvH379mjQoAF69uyJO3fuSGfy169fj4CAALRv3x5jxoyBg4MDFAoFZsyYoTVwXKacRlEUWQZwexdymrc+25RV1vttNRoNZDIZ9u/fr7Odr15R+emnnxAQEIBdu3bh0KFDGDlyJGbMmIEzZ87k6j5BooJw7NgxPH36FJs3b8bmzZuzfb5hwwY0b94819OLjo6Gj48PLCws8P3338PT0xMqlQoXL17EN998o3MAuaxcXV1Rvnx5/Pnnn3lalrxwcHDA5cuXcfDgQezfvx/79+/H6tWr0adPH52DxxG9iaWlJZydnXH9+vVsn2X28MppkCKlUpnthJJarUazZs0QFRWFb775BhUqVICpqSkeP36MgICAXMXSm2RO49NPP0Xfvn111qlWrZrW+/e5D27SpAkA4M8//3xnV5QbNWqEe/fuSfvilStX4ueff8ayZcswcODAfE+3MB2rUPFiZGSEjz/+GB9//DHKlSuHfv36YevWra99AsuIESOwevVqjBo1CnXr1oWlpSVkMhm6d++e723NJ598gs2bN2PmzJn47bff3viUikyMnbzJV3LesWNHLFu2DAEBAdJVzsyuD4cOHcKaNWswduzYgmslSQm3n58fFi1ahHHjxgEAtm3bhtKlS2PHjh1a3U/y+8ikzOe06uoOdufOnWx1jxw5gri4OK2r55ndXPX1zNdMmfPP2m4go412dnZvfLSNp6cnhBDw8PBAuXLl3jjPqlWromrVqvjuu+9w6tQp1K9fH8uWLcPUqVMBZO8iRFTQNmzYAAcHB53dbHfs2IE//vgDy5Ytg5ubW67i/MSJE4iMjMSOHTvQqFEjqTwkJCRP7UpPT9e6Aunm5pZjbGZ+nvnv1atXodFotA4EdG1njIyM0LZtW7Rt2xYajQZDhw7F8uXLMXHiRJQpU4bxR3nWunVrrFy5EmfPntW6BSQ/rl27hrt372Lt2rXo06ePVJ71Voy87Iezsre3h7m5OdRqtdSLrCAUVOykp6cDeNkbIbf7aZVKBRMTkxzryeVyrV54NjY26NevH/r164f4+Hg0atQIgYGBUnL+LrYFbm5u0Gg0CAkJ0er18OrI90RvK/Oi3dOnTwHk/Fvetm0b+vbti59++kkqS05ORnR0tFY9T09PnScgdWnfvj2aN2+OgIAAmJubY+nSpflYAt3c3Nx0xkpxjJ98dWufMmUKnJ2d4eXlhT59+kAmk2HWrFlo0KABWrZsiWrVquHbb78t6LYWe76+vqhVqxbmzZuH5ORkAC/PRr169unff//F6dOn8zWPzL/r2rVrtbq9HD58OFsXmlatWkGtVmPRokVa5T///DNkMhlatmyZrzYUlFeX5dWN0fXr13Ho0CG0atXqjdPo2LEjFAoFpkyZku0MnxACkZGRADLuTcw86MhUtWpVyOVypKSkSGWmpqbZNoxEBSUpKQk7duxAmzZt0Llz52yv4cOHIy4uDrt370arVq1w5swZnD17Vvp+eHh4tkeu6drGpKamYsmSJblu1927d3Hnzh1Ur15dKmvVqhXOnj2rta1KSEjAihUr4O7uLt1b26pVKzx79gxbtmyR6qWnp2PhwoUwMzOTugRnxmImuVwuXSHMjMHMk3GMQcqtsWPHwsTEBP3790dYWFi2z/Ny5UdXLAkhpMf9ZcrLfljXPDp16oTt27frPOAODw/PdXtfZWpqWiC3af3vf/8DAGlbkNv9tEKhQPPmzbFr1y6t3gphYWHYuHEjGjRoAAsLCwDZtwVmZmYoU6ZMtn0xULDbAn9/fwDItm1cuHBhgc2Dio/jx4/r3L5kjsOQeYtHTseVCoUi2/cXLlyY7RGonTp1wpUrV3Q+qlHX/Pv06YMFCxZg2bJl+Oabb3K9PG/i7++P06dP4/Lly1JZVFSUzsfAFnX5unJuaWmJM2fO4KeffsK2bdugUqlw8uRJeHp6YvLkyRgzZozOx/HQ2xszZgy6dOmCNWvW4LPPPkObNm2wY8cOdOjQAa1bt0ZISAiWLVuGSpUq6bxPLjdmzJiB1q1bo0GDBujfvz+ioqKwcOFCVK5cWWuabdu2hZ+fHyZMmIAHDx6gevXqOHToEHbt2oVRo0YViseEzZkzBy1btkTdunUxYMAA6REtlpaWuXrmqKenJ6ZOnYrx48fjwYMHaN++PczNzRESEoI//vgDgwcPxtdff41jx45h+PDh6NKlC8qVK4f09HSsW7dOOlDK5O3tjSNHjmDu3LlwcXGBh4dHnh/RQ5ST3bt3Iy4uDu3atdP5eZ06dWBvb48NGzZg+fLlWLduHVq0aIEvvvhCepRa5pXqTPXq1YO1tTX69u2LkSNHQiaTYd26dTkmJenp6dJ9uBqNBg8ePMCyZcug0Wi0evSMGzcOmzZtQsuWLTFy5EjY2Nhg7dq1CAkJwfbt26Wr5IMHD8by5csREBCACxcuwN3dHdu2bcM///yDefPmSb12Bg4ciKioKDRu3Biurq4IDQ3FwoUL4eXlJd2f7uXlBYVCgVmzZiEmJgZKpRKNGzeWnhlLlFXZsmWxceNG9OjRA+XLl0evXr1QvXp1CCEQEhKCjRs3Qi6X5+rWpQoVKsDT0xNff/01Hj9+DAsLC2zfvl3nvd253Q/rMnPmTBw/fhy1a9fGoEGDUKlSJURFReHixYs4cuQIoqKi8rwevL29sWXLFnz11Vf4+OOPYWZmhrZt2772O48fP5a2Bampqbhy5QqWL18OOzs7rS7tud1PT506FYcPH0aDBg0wdOhQGBgYYPny5UhJScHs2bOlepUqVYKvry+8vb1hY2OD8+fPY9u2bRg+fLjW8gDAyJEj4e/vD4VCge7du+d5vWRdR506dcK8efMQGRkpPUotcyBM9tyhvBgxYgQSExPRoUMHVKhQAampqTh16hS2bNkCd3d3aQyJnI4r27Rpg3Xr1sHS0hKVKlXC6dOnceTIEdja2mrNZ8yYMdi2bRu6dOmC/v37w9vbG1FRUdi9ezeWLVumdVI90/DhwxEbG4sJEybA0tKyQC7Ijh07FuvXr0ezZs0wYsQI6VFqpUqVQlRUVPGKn/c4MjzlUuajCM6dO5ftM7VaLTw9PYWnp6dIT08XGo1GTJ8+Xbi5uQmlUilq1Kgh9uzZI/r27av12JPMR6nNmTMn2zSh4xEf27dvFxUrVhRKpVJUqlRJ7NixI9s0hch49NGXX34pXFxchKGhoShbtqyYM2eO9PiFV+cxbNgwrbKc2vTqIxvyu55edeTIEVG/fn1hbGwsLCwsRNu2bcXNmze16mQ+KiU8PFznNLZv3y4aNGggTE1NhampqahQoYIYNmyYuHPnjhBCiPv374v+/fsLT09PoVKphI2NjfDz8xNHjhzRms7t27dFo0aNhLGxsQDAx6pRgWrbtq1QqVQiISEhxzoBAQHC0NBQREREiKtXrwofHx+hUqlEiRIlxA8//CBWrVqV7bEl//zzj6hTp44wNjYWLi4u0iNdkOWxZLoepWZhYSGaNGmSLRaEEOLevXuic+fOwsrKSqhUKlGrVi2xZ8+ebPXCwsJEv379hJ2dnTAyMhJVq1YVq1ev1qqzbds20bx5c+Hg4CCMjIxEqVKlxJAhQ8TTp0+16v3yyy+idOnSQqFQ8LFqlGvBwcHi888/F2XKlBEqlUoYGxuLChUqiM8++0xcvnxZq27fvn2FqampzuncvHlTNG3aVJiZmQk7OzsxaNAgceXKFQEg2286t/thXfvwsLAwMWzYMFGyZElhaGgonJycRJMmTcSKFSukOjntazP3za+2Jz4+XvTs2VNYWVnpfJxbVlkfpSaXy4WDg4Po0aOH1mNWM+VmPy2EEBcvXhT+/v7CzMxMmJiYCD8/P3Hq1CmtOlOnThW1atUSVlZW0t9p2rRpWo+gSk9PFyNGjBD29vZCJpNpPVYt6/rM6fhA12OjEhISxLBhw4SNjY0wMzMT7du3F3fu3BEAxMyZM1+7zohetX//ftG/f39RoUIFYWZmJoyMjESZMmXEiBEjRFhYmFQvp+PKFy9eSPtNMzMz4e/vL27fvi3c3NyyHXtGRkaK4cOHixIlSggjIyPh6uoq+vbtKz32MKdtxdixYwUAsWjRIiFEzo9S0/VoQx8fH+Hj46NVdunSJdGwYUOhVCqFq6urmDFjhliwYIEAIJ49e5bPNfnhkQnBu/GJiIiIiAra5cuXUaNGDaxfv75YPhaK6G2MGjUKy5cvR3x8fI4DyxU1+erW3r9//zfWkclkWLVqVX4mT0RERET0QUlKSsp2W+e8efMgl8u1BtQkouyyxk9kZCTWrVuHBg0aFJvEHMhncn7s2LFsff/VajWePn0KtVoNe3v7N46CTURERERUVMyePRsXLlyAn58fDAwMpMc6Dh48WGs0eSLKrm7duvD19UXFihURFhaGVatWITY2FhMnTtR3096rAu3WnpaWhuXLl2PevHk4fPgwPDw8CmrSRERERESF1uHDhzFlyhTcvHkT8fHxKFWqFHr37o0JEybAwCBf18OIio1vv/0W27Ztw6NHjyCTyfDRRx9h8uTJBfpYyA/BO7nnfOjQoQgNDcXevXsLetJERERERERERU6+nnP+JtWrV8eff/75LiZNREREREREVOS8k+T88OHDMDExeReTJiIiIiIiIipy8nUDzPfff6+zPDo6Gn/++ScuXryIcePGvVXDPgQajQZPnjyBubl5tgHyiPRFCIG4uDi4uLhALn8n59+KBcY3FUaM74LB+KbCiPFdMBjfVBjlNr7zdc95ThO0traGp6cnBg4ciEGDBhX5gHj06BFH36RC6+HDh3B1ddV3Mz5YjG8qzBjfb4fxTYUZ4/vtML6pMHtTfOfryrlGo8l3g4oSc3NzAEDoRXdYmPEM59vqUK6qvptQJKQjDX9jn/T7pPxhfBcsxnfBYHwXDMZ3wWJ8FwzGd8FgfBcsxnfByG185zk5T0pKwoQJE+Dn54e2bdvmu4FFQWbPAAszOSzMGfxvy0BmqO8mFA3/3xemqPdcedcY3wWL8V1AGN8FgvFdsBjfBYTxXSAY3wWL8V1Achnfef7FGhsbY/ny5QgLC8tXu4iIiIiIiIhIW75OJ3l7e+P69esF3RYiIiIiIiKiYilfyfm8efOwefNmrFy5Eunp6QXdJiIiIiIiIqJiJdf3nP/555+oWLEi7O3t0bdvX8jlcgwZMgQjR45EiRIlYGxsrFVfJpPhypUrBd5gIiIiIiIioqIm18m5n58f1q9fjx49esDW1hZ2dnYoX778u2wbERERERERUbGQ6+RcCIHMR6KfOHHiXbWHiIiIiIiIqNjh8wWIiIiIiIiI9CxPyTmfu0hERERERERU8PKUnH/66adQKBS5ehkY5LrHPBEREREREVGxlqcMumnTpihXrty7agsRERERERFRsZSn5Lxv377o2bPnu2oLERERERERUbHEAeGIiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0rNc33Ou0WjeZTuIiIiIiIiIii1eOSciIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50RERERERER6xuSciIiIiIiISM+YnBMRERERERHpGZNzIiIiIiIiIj1jck5ERERERESkZ0zOiYiIiIiIiPSMyTkRERERERGRnjE5JyIiIiIiItIzA303gN6da2dMsXWJA4KumSAqzBCTV4WgXssYfTfrvWkbEIHOnz+HjX067t80xpLvSuDOZZMc6zdsE42+Y5/B0TUVj0OUWDXNGeeOWbxSQ6DPmDC06BkJMws1bp43xYJxrngSopRq9BgZhlpNY1G6chLSU2XoVLFqtvl4NYhD37HP4F4hGcmJchzZao3VM52hUcsKcvGpiGN8v//4DlwTAs/KSbCyTUdcjAKX/jLHqmnOiAozlOp4+8Si99f/1959x0dR538cf+9udje9h1BCEkIvEjoqIqIIJ4qCiF0BvbOhWNCzndhQznKKB/zEcoJ3552oJ1hOUVFsFJUSeieUBBLSe9vd+f0RTG5lUUpgks3r+XjsQzP7ncl3hnx29z3f78xmK6lzpaqrLNqwIlSvPt5a2RmOk3EY4Keo71Nf3z+zOzx66b/b1b57pW49v5N2bQySJF07JUvXTck+rH1luVWXdDj8vR7Nz4dzY/Xeyy2UnxOglG4Vum1aprr0Lj9i+/dfi9N/34zRwf0OhUe5NPiiQt3w4AE5Ag1J0vUDuvl87xg1Pke3T888afvRGBzLa0BSp0pdf1+WOvQsV8u2NZoztbUWvB7n1eai63N14fV5im9bLUnaszVQb70Yr5VLwn1t0lSMnPuxynKrUrpX6PanM8zuyik35OIC3fTofr31QktNGtFJuzYF6ql/7VJETI3P9t36lenB/9ujRf+O1m3DO2nZonA9+sZuJXWuqGtz+aQcXXJDjmY+kKA7L+qoynKrnv7XLtmdnro2AQ5D334Uqf++Gevz96R0q9CT/0jXyiVhmjS8k56+JUmnDy/WjQ8faNgDAL9HfZ/6+l67NFRP3ZykGwd30bQ/JKt1cpUeeW133fPxbav02NzdWrs0VLed30kPX52i8GiXHvnbbgHHgvo+9fX9sxv/dEB5WfbDlr/3cpyuTO3m9diz1alvP4pouJ1Hk/X1B5F69fHWuuaeLM3+bKtSulXo4atTVJjrexz0q/cj9cbTrXTNPVl67Zstuucv+/TNh7WDNT/766db9e+0DXWP6W/vkCQNHuXfJ+qO9TXAGeTRgb0OvfF0K+Vl+z7eOQfseuPpVrr9d510xwWdtHZpqB6bu1tJnSpP5q4cF8K5pNmzZys5OVmBgYEaOHCgfvzxR7O71CD6n1uiCfdnaVAzOtv+s0tvytWif0Xr8/nR2rs9UH+9P0FVFRaNuCrfZ/vRv8/RyiVheu/lFtq3I1B/f66VdqwP0iUT8w61MDT69zn690vxWv5ZhNI3B+nZyYmKia/Rmb+rP77/eL6lFrwWp/QtgT5/z5CLC5W+OVBvvdhS+3c7tX5FqF6f1kqjxucqKMTd0IcBor79kVn1veC1OG1ZHaKDmQ5tWhmi+bNaqEufctkCakc5OvaskNVmaN4zLXVgj1M71gfrvTlxat+9oq4NGhb17X/Mqm9J6je0WH2HlOi1J1of9nsqy20qyLHXPaLiXErqXKXP/h3d0IcAhzSl+n7/1Tj97uo8jbgyX0mdqjT5mQw5gzxH/PvYtDJE3fuX6dxLC9WybbX6nlOic0YXaOua+tHhyBi3olu46h4/LI5Qq+Qq9Tyj9FTtlimO9TVg29pgvf5ka33zQZRqqn3PQv3hiwj99FW49qc7lbnLqXnPtFJlmVVd+padzF05Ls0+nM+fP1/33HOPHn30Ua1evVqpqakaMWKEDh48aHbXcJwC7B517Fmu1d+F1S0zDIvWfBembn19Ty/q2rdca/6nvSSt+iZMXQ8VbcvEasXEu7y2WV5i05Y1wep6hG36YncYqqnyLrvqSqucQYY69qw4wlo4XtS3/2ks9R0W6dK5lxZo08pguV21Hwa2rwuSx2PR8CvzZbUaCg5za9jYAq35LrSuDRoO9e1/zKzvyNga3fVchp69I1FVFb/98fh3V+dp306nNvwYekz7iKPTlOq7ptqi7euC1WdwfWi2WqXeg0u1aVWIz3W69SvT9nXB2nIojB/Y49BPX4ar/3nFR/wdX/0nSiOuzJPFj99Ojuc14FhZrYaGXFIgZ7BHm1f6/vcxU7MP5y+88IL+8Ic/aOLEierWrZvmzJmj4OBgvfHGG2Z3DccpPNotW4BUmOM9taUgN0BRcS6f60TFuVTwi6lHBTkBimpR2z760H9/uc3CnABFt/A9zcaXld+EqWu/Mp0zukBWq6GYljW65u7aa9ii449+Ozg61Lf/Mbu+b3x4vz7YsV7vbdqouNY1emxiu7rnsvc59dBVKZr4QJY+3r1OC7ZuUGzrGj11c/Jx7St+HfXtf8yrb0P3ztin//4jRtvXHfna9p/ZnR6dO6aQUfOTqCnVd3G+TR63RZFx3u8XUbE1KsjxPc363EsLdf29BzRldAeNTEzVhDO6qeeZpbpqsu+TD8sWRai02Kbhl/sePfYXx/MacLSSu1Ro4fb1+nj3Ok3+c4aeuDFZe7f7nulqpmYdzqurq7Vq1SoNGzasbpnVatWwYcO0fPnyw9pXVVWpuLjY6wEci9XfhOn1J1tr8p8z9PHudXrj+y368avas4PG4Ze+4QRQ3zgZ3n25hW4b3kkPXpkij0e676W9kmqnrEfF1eiu5/bpi3ejdMfIjpoypr1qqi2HrktnWntDor7RkC65MVdBoW7Nn9niqNoPuqBIQaFuffFO1EnuWfPUHOp77bJQvT0zXrc/naHZn23V1L+l68fF4XrrxXif7T/7d7T6Dy1WTMsTC6jNWcZOp247v5MmX9hRH/89Vve+tFeJHbnmvFHJzc2V2+1WfLx3IcTHxysrK+uw9tOnT1dERETdo23btqeqqzgGxfk2uV1S5C/OsEXFuo54BrMgJ0BRsb9oH+dSwcHa9vmH/vvLbUbGuZR/8PAbx/ya91+N06Vdeuja/t00rkd3LV9UezOZA3sOv2ssjh/17Z/Mru/i/ABl7nJq9bdhmn5rkgYOK6mbGjtqQq7KSmz627TW2rkhWBt+CNWzdySq9+BSdenTMNPxUIv69k9m1XevQaXq2rdcH+9ep0/2rtXcZZslSbM+3aZ7Z+w97Hf+7qp8/bA4XIW5x/b+j6PT1Oo7PNotq81QYY7330NBrv2Io71vPttS540t0AXX5Ktd10oNuqBIEx88oPkz4+X5xWBNdoZda74L0++uzvO5LX9yPK8BR8tVY9X+3bX3g5k7vZXSNwVp9O9zTmibJ0OzDufH6sEHH1RRUVHdY9++fWZ3CT64aqzavi5Yvc8qqVtmsRjqdVapNq3yPV1t86pg9RrsfYONPmeXaPOha4Wy9jqUlx3gtc3gULe69C7X5iNs89dZlJ9tV3WlVUPHFOpgpl071gcdx3bQUKjvpqEx1bfl0Duo3VE7Kh4YZMjweF8M+PNXJFp5tzUV9d00mFXf//dIG906rJNuPb/28afrUiRJT9+SpHnPtPTadnzbKqUOKtVn/4458R1GgzC7vu0OQx17lmvN9/X3H/B4pLTvQ9XtCDccq6qwymL1nlFlPfSz8YuJVp+/HaPIWJcGDmv8MwJO1PG8Bhwvi6X+/bsxadbfcx4bGyubzabsbO/vrczOzlbLli0Pa+90OuV0Np3RzYoyq9d3eGbtc2jnhiCFRbrUIsG/r29+/9VY3Ttjn7atDdbWNcEa84ccBQZ79PnbtdeH3ffSXuVm2TV3eu1XVix8PU7P/WeHxt58UD9+Ga4hlxSqY88Kzbgv4dAWLVr4epyuuvOgMtOdytrr0Pg/Zikv265li+q/RiWuTbXCIt1q0aZaVpuU0r32Jm/70x2qLLdJki679aBWLgmT4bFo0MgiXT7poJ66JUkejx/f4cME1Lf/MqO+O/cuU+deFdrwY4hKC21qlVyl8X/M0v50R90H/B++DNOYm3J0zd1ZWrIwSsGhbk18IEtZ++zasYGTbw2J+vZfZtR3Tqb3d0lXltV+F/L+PU7lHvB+bsSV+crPDtBPX3nfhA4NpynW96U35ej5uxLVKbVcnXuXa8Frcaost2r4lbXXiD87OVGxLWt0w0O1X517+vnFev/VOHXoUaEufcqVme7Qm8+10sDzi2Sz1W/X45E+nx+tYePyZWsmqe1YXwMC7B4ldqqSJNnthmJa1Sile4Uqy2pHyiVp4oMH9NNXYcrJdCgo1K2hYwrV88xSPXx1ijk7+SuayT+zbw6HQ3379tWXX36p0aNHS5I8Ho++/PJL3X777eZ2rgFsWxusP17Woe7nVx5rI0k6//J8n9O0/Mk3H0YpIsat6+/LUlScS7s2Bunha9rVTUGLa1PtNW1o08oQ/XlSksbfn6UJD2Rpf7pTj9+QrD1b6z9QvzM7ToHBHt35bIZCw93a+FOIHr4mxevu69ffm6XhVxTU/fzyF9skSfeNba91y2vPqPYfWqKrJmfL7jC0a1OQHpuYrJVLwk/m4WiWqG//ZUZ9V1VYNeiCIl03JUuBwR7lH7Rr5ZIwPfVSvGqqa9usXRqmP09K1LjbcjTuthxVVVi1eVWw/nRNiqorGTpvSNS3/zLr/ftoWCyGhl9RoC/eieaE+knUFOv7nEsKVZQXoL8/10oFOQFK6V6hp97aVTetPSfT4TWD6uq7smSxGJr3bCvlZdkVEe3S6ecXacID3tP213wbpoOZDo240r9vBPe/jvU1ICbeVfd5W5LG3ZqjcbfmaO2ykLrX0chYl+77615Ft3CpvMSm9M2BevjqFK3+tvGdZLMYxi8nTzQv8+fP1/jx4/XKK69owIABmjFjht555x1t2bLlsGtdfqm4uFgREREq2Jai8DA+eJ2oEa17md0Fv+AyavS1PlBRUZHCw5t36Ke+Gw/qu2FQ3/Wo78aD+m4Y1Hc96rvxoL4bxtHWd7MeOZekK664Qjk5OZo6daqysrLUq1cvLVq06DcLH0DjR30D/ov6BvwX9Y3mqtmPnJ8Izsw1LM7MNQzOvDcM6rthUd8Ng/puGNR3w6K+Gwb13TCo74ZFfTeMo61v/mIBAAAAADAZ4RwAAAAAAJMRzgEAAAAAMBnhHAAAAAAAkxHOAQAAAAAwGeEcAAAAAACTEc4BAAAAADAZ4RwAAAAAAJMRzgEAAAAAMBnhHAAAAAAAkxHOAQAAAAAwGeEcAAAAAACTEc4BAAAAADAZ4RwAAAAAAJMRzgEAAAAAMBnhHAAAAAAAkxHOAQAAAAAwGeEcAAAAAACTEc4BAAAAADAZ4RwAAAAAAJMRzgEAAAAAMBnhHAAAAAAAkxHOAQAAAAAwGeEcAAAAAACTEc4BAAAAADAZ4RwAAAAAAJMRzgEAAAAAMBnhHAAAAAAAkxHOAQAAAAAwGeEcAAAAAACTEc4BAAAAADAZ4RwAAAAAAJMRzgEAAAAAMBnhHAAAAAAAkwWY3QF/sLRSCrGb3YumL6Bdktld8A+eKmm32Z3wH7trShVWw3nME2WLiTa7C37B8FRL+Wb3wn+8UxqlID4KnTBralezu+AXrO4qab3ZvfAfQ56+UTZHoNndaPLsnx40uwt+wV1WJY397XZ84gQAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkAWZ3AA1j5d9jteK1FirNCVB81woNfyxTbVLLfbZ110jLXo7XuvejVZJlV0xKlc69f7/aDyk5xb0+tS68NF1jr96hqOgqpe8I15wXT9O2zVFHbH/W0P269g9bFN+yXPszQjT35W5auTy+7vnIqEpNvG2zeg84qJBQlzamRWvOi6dpf0ZoXZuo6ErdMGmTevfPUVCwSxl7QzX/7x217OvWJ3Vf0XR9/WYrff5qGxXnOJTQtUxXPL5T7XqV+mzrrrFo0f8laPl7LVSY7VR8SoUufSBd3c8p9Nl+0f8laOEzyTr3hkxd/mj6SdyLU++iKzM0dsI+RcVWK31riF6e3knbNoQfsf1Zww/qutvTFd+6Uvv3BumNF9tr5Xcxdc/fPW2zzr8ky2udld9Ha+qtqZKk0/oV6Jm5aT63feeVfbV945F/N5qvLW+FauPfwlSRY1N0l2oNeKRQsT2rj9h+07xQbft3qMoO2OSM8ihpRIX6TCmUzVnfpjzbplXPRSjzu0C5KywKS3LpzKfzFXtazSnYI/NcdNF2XXbZZkVFVWrXrki9/HJfbdsW47NtYmKRrrtuvTp2zFd8fLleeaW3Fi7sfMRtjxu3STfcsE4LF3bSK6/0OVm7AD8zbsAGXXdmmmJCK7Q9O0bPfTJIGzPjfbYd2nWXJg5eo7bRRQqwebQ3L0JvLUvVJ+s61bVZ+fgcn+u+9Pnp+sfSXidjFxoN+0dFcr5XKEuBW54UhypujZWnc+CRVyh1K/DNfAUsLZOlxC1PvF1VN8XINSCkdnsfF8nx32JZs2tfF91JDlVfHSVX/5BTsTvHpFmPnH/77bcaNWqUWrduLYvFooULF5rdpeOy6eNILX66tQZPztKNH21Vi64Vent8ispyfZ97+eYvrbT63zEa8WiGbv58i/pcnav3bmmnrI1Bp7jnp87g8zL1hzs26l9vdNbkG4YofUeEnnxhhSIiq3y279ojX398bJU+/zhRkycO0fLvWulP039UUrviQy0M/enPP6ll6zI9ef8ATZ44RAezgvXUS8vlDHTVbeeeR9aoTWKpnrh/gCZdf46WfdNKDzyxUikdi07BXjdvTbG+V34Uq/emtdNFd+7VQx+vUULXMs28roeKc+0+23/wfJK+faulrnh8lx5dvEpnX3NAc27qqr0bDn+z2b02VN+91VJtupad7N045c4eka0/3LdD/5qTrDsu76dd20L15CtrFRHtO/R0TS3S/c9s0ufvt9Id4/pp+VexeuSl9Urq4H0SZOX30brmnDPrHs/e363uuc1pEV7PXXPOmVr0XisdyAjU9o1hJ3V/0TTrO/2TIK2cHqnUScW6aEGWorrUaPGNcarI8/1RbNdHwVr9l0j1vL1Yl3ySpTOfytfuT4K1+oXIujZVRRZ9elULWe3SsNdydfF/s9Tv/kI5IzynaK/McfbZe3XTTWv01ls9dMcdI5SeHqlp075WRESlz/aBgS5lZYVq7txU5ef/ygd8SZ065WnkyJ3atSvyJPQcR6Mp1vf53Xfo7hHL9NrX/XTtK2O1LStGM6/7r6JCKny2L65w6o1v+2ji62N05f+N00dpnTV19BKd3n5fXZsRz13v9Xh8wTnyeKSvNqWcqt0yRcA3pQp8NVdV10SpbGaC3O0cCvnTAVkKXb5XqDEU8tABWbJdqng4XqWvJ6pycpw8sfU5yIgNUNXEaJXNTFDZXxPkTg1S0BNZsu458slRszTrcF5WVqbU1FTNnj3b7K6ckB/+FqdeV+QpdVy+4jpWaeS0DAUEebT23Wif7dcvjNagWw+qw9ASRSVWq++1eWp/TrF+eD3uFPf81BlzxU4t+ihRiz9J1L7dYZr1XE9VVtk0/KK9PttffPkurfqhhd7/Vwft2xOmf77WRTu3Reqiy2pHG1u3LVPXHgWa/XxPbd8Spcy9oZr9fE85nG4NOT+zbjtde+Tro/faadvmKGXtD9H8NzuprNSuDl0KT8VuN2tNsb4Xv95Gg67M0pmXH1TrThW6+ukdsge5tewd32fef3g/ThdMytBp5xYoLrFKQ67LUo+hBVr8WhuvdpVlVr1xZ2dd+8x2BUcc4c2tCRtz/T4t+k9rfbGwlfbtCtGsJzqrqsKq4WMO+Gx/ybUZWrU0Wv+Zl6h96SH6x6wU7dwUplFXZXq1q6m2qiDPWfcoLa4/SeJyeT9XXGTX6UNztXhhK0mWk7m7UNOs781zw9Tx8lJ1GFumyA4unf54gWyBHu34j++Rm5w1DrXoU6WUUeUKTXCr9VlVandRuXLXOerabHgtXCEt3Ro0PV+xPasV1ra2XVii+1TtlinGjNmiTz9try++SNHevRGaObO/qqoCNHz4Lp/tt22L0d/+1kvffJOkmpojf/QNDKzRffet0Esv9Vdpqe+Tojj5mmJ9X3PmOi1c1VUfpXVRek60pn98tiprAnRx7y0+26/a3UZfb2mn3blRyiyI0NsrempHdox6JdW/b+WVBns9hnTZrZW72yizwL9nZjkXFKrmgnDVDA+XJ8mhyjviZDgtsn/ue4av/fNiWUrcqpjaUu7uQTLi7XL3DJInpX6Kkev0ELkGhMjTxiFPgkNVE2KkQKtsW3yf0DNTsw7nF1xwgaZNm6YxY8aY3ZXj5q626MCGYLUbVD/iY7FK7QaVKmON7zd8d7VFAU7vs+r2QI/2rQz12b6pCwjwqEPnIqX9VH/ywTAsSlsZqy49Cnyu06V7gdJWxnotW/1DnLp0r21vt9cev+pqm9c2a6qt6t4zv27Z5g3ROvu8/QoNq5bFYujs8zLlcHi0frXvqXdoOE2tvl3VFu1dH6quZxXWLbNapa5nFWrXat8jsa5qq+w+annHSu837rcfaa8e5+ar61n+N2MjIMCjDt1Klbai/hIVw7AobUW0uqQW+1ynS2qR1qzwvqRl1bJodUn1Pj6n9SvUv77+Xq9+uEKT/rRVYRFHniZ8+jm5Cous0ecLW57A3uBoNbX6dldLeRsdanVm/Wwti1VqdWaVctY4fa4T17taeRsddWG8ZJ9Nmd8EKmFI/UhcxldBiulRrW8mx+idM1rro9Hx2vZO45um2ZACAtzq2LFAaWn1Jy0Nw6K0tHh17Zp3QtueNGmVfvqpldLSqGMzNbX6DrC51aVVjn7YlVC3zDAs+nFXgnq2zT6KLRjq3y5DSbGFWrO7lc8W0SHlOqvTXn2wuksD9bqRqjFk3V4lV6/g+mVWi1y9gmTb7DtIB6wol6troAJn5yj0qt0KuWWvHG8XSG7D9+9wGwr4ukSq9Mjd5ddn0piBa86buPICmwy3RSGx3h8aQ2JrlLfT9xt+yuAS/fBGnBIHlCoqqVrpS0O15bNIGX46Cy48slq2AEOF+d7HozDfqbaJvq/ljYqp9Nk+Kqb2hSFjT6gOZgVpws2ba0fhKwI0+oqdiouvrGsjSX9+pJ/uf2Kl5i9aJJfLoqpKm6Y91F8HMv3zRAiOX2mBXR63ReG/qOWw2Bpl7Qz2uU63swu0+PXW6jCwSHFJldqyNFJrFsXI8NSP3P70Yaz2bgjVgx+mnczumyY8qka2AEMFeQ6v5YV5drVt53sKf1RstQoPa+9QVGz99LZV30dr2eI4ZWcGqlXbCo2fvEtPvLxWU67tK4/n8JHx4Zce0Opl0crLbnxv9DBfVYFVhtuioBjvEe2gGLeKd/n+KJYyqlxVBVYturqFDEMyXBZ1urJUp91SP3pUsi9AW/8dqm4TS9TjlmLlrXfop2mRstkNtR/j+74zTV14eLVsNkMFBd61VlAQqIQE3yfkjsaQIXvUvn2B7rxz+Il2Ec1MZHClAmyG8ku9Lw/NLw1ScmzhEdcLcVbp0yn/kCPAI7fHomf+O1g/7Grrs+1FvbaqrMquJZvbNWTXGx1LsVsWj2RE2byWG1EBsmX4vkTAmlUj61qXaoaGqvyJVrLur1Hg7BzJbaj6mvpZxNb0KoXckylVG1KQVRWPtJQnyeFzm2YinB+DqqoqVVXVn/UuLj7+NwEznT81Q588lKg553eVLFJUYpVSL8vT2ncZzT1abrdVTz3UX3c+mKb5ixbJ7aodif9peQuvCa3X/WGLQkNr9NDkM1Rc5NDpg7P0wBMr9cfbztKeXf49LampaYr1fflju/TPBzrqsXP7ymKR4pIqdOa47Lpp8Pn7HXrn8RTd+c8Nsgce4QwyfPp2Uf2o3O7toUrfFqo3Pl2h0/oXaO0P3pcMxcRXqs+Z+frzvd1PdTdxlJpifWf94NT6V8I18NECxfasVsneAP34VKTWzQ5Xz0mH+m9IMT2q1eee2lkfMd1qVLjdrq1vh/ptOD8ZYmPLdPPNq/XQQ0NVU2P77RXQqDTF+pak8mqHrp4zTsGOGvVPydTdI5YpsyBMq3a3Oaztxb23atH6jqp2Ed0OY0hGpE2Vk+Mkm0Wejk5V57nkeK/QK5x7Ehwqnd1WljKP7N+XKvAvB1X+bJtGF9D5Fz4G06dP1+OPP252N7wER7llsRkq+8UNo8py7QqJ831taUiMW+NeSZeryqLyggCFxddoyTOtFJno++ZoTV1xoUNul0WR0d77FxldpYIj3BimIC/Qd/u8+vY7tkbqjgnnKDikRgF2j4oLnXrh1W+1fUukJKllmzKNuixdt157jvam1wbx9B0R6pGap4vGpmv2c6kNuJc4UWbXd2hUjaw247Cbv5Xk2hUe5/uGJWExLt362mbVVFpUWmhXZHy1Fvw5WbGJtbM39q4PVUmuQ09f2LtuHY/boh0/hOvrN1tr1valsjbxz6HFBXa5XRZFxXgfo8iYGuXn+Z49VJDrUORh7atVkHvkN+isjCAV5dvVOrFCa3/wfm746CyVFNq14utY3yvDdGbXtzPKI4vNUEWed8FV5NkUGOt72lraSxFKubhMHcfVzgCJ6lwjV7lFy6dG6bRbi2WxSkFxbkW0955tE5FSoz2f+e8NXouLHXK7LYqK8p7iGhVVqYKC49vvjh0LFBVVpVmzPqtbZrMZ6tEjR6NGbdfFF4+Tx9OsrwRt1Myu78LyQLncFkWHeo/sRodWKK/U98w3qXbqe0Z+hCRpW1as2sUWaMLgNYeF816JB5QcV6gH3x3W8J1vZIxwmwyrZCnwnmVkKXDJE+X7A4sRZZMRYJFs9cNjnrZ2WQvcUo0h2Q8tt1tktLbLkFTV0Snbtio5PiiqDfWNCK80x+DBBx9UUVFR3WPfvn2/vdJJZnMYatWjXLuX1U+TNjzS7mWhSuj963dlDnAaCm9ZI49L2vJZpDoNaxpnGo+Vy2XVjq0R6tUvt26ZxWKoV99cbdng+6vUtmyMUmrfXK9lvfvnaMvGw9uXl9lVXOhU64RSdehSqBXf116r5nTWvrAYv5gC6/ZYZOV+UY2O2fUd4DCUeFqptiyNrFvm8UhblkYqpc+vf82hPdBQVMtqeVwWrfk0RqnDa+970GVQkR75fLUe/nRN3SOpZ4kGjM7Rw5+uafLBXDpU35tClTqw/v4RFouhXqcXaMta37NTtqyNUK+B3veb6H1GvrasjTji74mJr1RYZI3yc34Z+A0NG31AX37UUm4Xb6mNldn1bXNIMd2rdWB5/d+P4ZGyljsV19v3iXFXpUWWX/xJWQ7VrHFoIkxcnyoVp3uf0CvebVdoG/+9IZzLZdP27VHq1av+Wl6LxVCvXtnavPn4ZgCmpcXrllt+p0mTRtQ9tm2L1pIlSZo0aQTBvJEzu75dbpu2HIjTgJT6m4paLIb6t8vUun2+b+jqi9ViyGE7vHYv6bNZmzLjtD27GZwAtteOfAek/c/MH4+hgLQKubv6HlBzdw+UdX+N5KmfIWjNrJEn2lYfzH0xjNrw3sgwcn4MnE6nnE7fIzFmGnhjjj68N1GtTitX69Ry/Tg3TjXlVvW8rPYD+odTEhUWX6Ohf6y9A2RmWrBKsuyK71ahkiy7vnuppQyPdMbNB83cjZNqwfz2uufhNdq+JULbNkXpkst3KTDQrS/+W3ttzz1/Wq283EC9Oaf2q5I+fCdFf569VGOu3KGflsXr7GGZ6tClUDOfqR/tPmvofhUVOpSTHaTklGLddNcGrfiuldb82EJS7XXpmftCdPsf1+pvs7qruNihMwYfUO/+OXr8jwNP/UHAr2oM9T3s95maN6WTknqWKjm1RF+90VrV5TadOa72Q+jcuzspsmWVxty/R5KUviZUhVlOJXQvVWGWUx+/mCjDY9HwmzMkSYGhbrXp7D211RHsUUhUzWHLm7IFf2+re57aou0bw7RtfbguuS5DziC3vlhYe2OdKU9tUt5Bp+a91F6S9ME/E/TM3DUac/1e/fRdjIb87qA6di/RzMdrv/c4MMilq2/draWL41SQ61CrthW64Z6dOrA3SKuWek9pTx1YoFYJlfrsfd838UHj0Bjqu+vEEi29P0axPaoV07Nam98Mk6vCqg6X1p5I//6P0QqOd6vPlNop6glDK7R5bpiiu1XXTWtPeylcCUMr606sdRtfqk+vaqH1c8KUdEGFctc5tP2dEJ3+hO+bnfqLBQu6aMqUFdq+PVpbt0Zr9Ohtcjpd+uKL2q+YmjJlhfLygjRvXu17dkCAW4mJxYf+36OYmAqlpBSooiJABw6EqaLCrj17Ir1+R2WlTSUlzsOWo/FpDPX91rKeemzMEm3KjNPGzBa6+ox1CnLU6KM1te8rj4/5SgdLQjR7ce3nvwmDV2tzZpwyCiJkt7k1qNNejUzdrukfD/baboizWsO679KMz8445ftklqoxkQr6y0G5Ozrl7hwox8IiWaoM1Zxfe3PcwOezZcQEqGpi7cm46gsj5PiwSIFzclV9cYSs+2vkmF+o6ovrT7g75+bJ1S9YnhYBspR7ZP+6VLZ1laqa5nuQzkzNOpyXlpZqx44ddT+np6crLS1N0dHRSkxMNLFnx6bbRYUqyw/QNy+2UllugOK7VujKebsUemhae9F+h9fZd1eVRd+80EoFex1yhHjU4ZxiXfzCHgWG+++Z9u++bKOIyGpd+/utioqu0q7t4Zo65XQVHrqhTFx8hQyj/uza5g3Reu6xvrrups0af/MWZWaEaNqDA7QnvX4kLiqmUr+/Y0PddPcvF7XV23M71T3vdlv12L0DNeHWzZr67A8KCnJrf0aIXpjWWyuXH/2ZVByfpljf/UblqiTPro9eSFRxjkMJ3cp0x983KDyudtpq/n6nLNb6s7w1VVZ98HyScvcFyhnsVo+hBZo4Y5uCI/y3ln359rN4hUfX6LpJ6YqKrdauLaGaekvPupu+xbWqkud/63tthJ59oJuuv32XJty5S5l7gvXknadpz47aGUgej0XtOpVq2MVZCgl3Kf+gU6uXR+kfs1Lk+sXXMI249IA2rQlXRrp/3yG7sWmK9d1uZIWq8guV9tcIVeTYFN21Wue9nqOgQ9Payw7YvN6re95aLItFSpsRofJsm5zRHrUdWqHed9d/q0Bsz2oNnZWr1S9EaO3sCIUluNTvoUKlXOw/J998+fbbREVEVOraa9crOrpSO3dG6pFHzlFhYe17eosWZXWzCyQpOrpCs2fXT1m/7LItuuyyLVq3Lk7333/eqe4+fkNTrO8vNnZQVEilbjn3J8WElmtbVqzu+MeFyi+rndbeMqLkfwd2FWR36f6LvlOL8DJV1QRod26kHvnPufpiYwev7Q7vsUMWSYvWey/3Z64hoaoscsv5zwJZ8l3ytHeq/MlWMqJqY6v1oEseS/17uhEXoPKnWsv5Sq5CbsuQEWNT9SURqh4XWdfGUuhW0PMHZcl3yQixytPOqfJpreTuc+TLDsxiMQyj8Y3nnyJff/21hg4detjy8ePHa968eb+5fnFxsSIiIvTxuhSFhDHl6UQ9c94lZnfBL7g8VVq8e5aKiooUHt58bzrXUPW9ZmMLhVHfJ2xSn4vN7oJfcHmq9WX+POq7ger7ldV9FRTarMcpGsQ/L+cO5w3B5a7SV+ufpb4bqL57TnhKNgffoHGi7Jf678zaU8ldVqVVY2f8Zn0363ekc845R8343ATg16hvwH9R34D/or7RnDEcBAAAAACAyQjnAAAAAACYjHAOAAAAAIDJCOcAAAAAAJiMcA4AAAAAgMkI5wAAAAAAmIxwDgAAAACAyQjnAAAAAACYjHAOAAAAAIDJCOcAAAAAAJiMcA4AAAAAgMkI5wAAAAAAmIxwDgAAAACAyQjnAAAAAACYjHAOAAAAAIDJCOcAAAAAAJiMcA4AAAAAgMkI5wAAAAAAmIxwDgAAAACAyQjnAAAAAACYjHAOAAAAAIDJCOcAAAAAAJiMcA4AAAAAgMkI5wAAAAAAmIxwDgAAAACAyQjnAAAAAACYjHAOAAAAAIDJCOcAAAAAAJiMcA4AAAAAgMkI5wAAAAAAmIxwDgAAAACAyQjnAAAAAACYjHAOAAAAAIDJCOcAAAAAAJgswOwONGWGYUiSyks9JvfEP7g8VWZ3wS+4PNWS6v8+cXx+Pn6l1HeD+PnvEifGZVDfDeHn41dR6ja5J/7B5eb9uyH8fByp7xPz8/FzV1ea3BP/YC2jvhuCu/zo6tti8Apw3DIyMtS2bVuzuwH4tG/fPiUkJJjdjSaL+kZjRn2fGOobjRn1fWKobzRmv1XfhPMT4PF4tH//foWFhclisZjdnSMqLi5W27ZttW/fPoWHh5vdnSarqRxHwzBUUlKi1q1by2rlypXjRX03L03lOFLfDYP6bl6aynGkvhsG9d28NJXjeLT1zbT2E2C1WpvUmc3w8PBG/UfbVDSF4xgREWF2F5o86rt5agrHkfo+cdR389QUjiP1feKo7+apKRzHo6lvTssBAAAAAGAywjkAAAAAACYjnDcDTqdTjz76qJxOp9ldadI4jmiM+LtsGBxHNEb8XTYMjiMaI/4uG4a/HUduCAcAAAAAgMkYOQcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjj3c7Nnz1ZycrICAwM1cOBA/fjjj2Z3qcn59ttvNWrUKLVu3VoWi0ULFy40u0uAJOq7IVDfaKyo7xNHfaOxor5PnL/WN+Hcj82fP1/33HOPHn30Ua1evVqpqakaMWKEDh48aHbXmpSysjKlpqZq9uzZZncFqEN9NwzqG40R9d0wqG80RtR3w/DX+uar1PzYwIED1b9/f82aNUuS5PF41LZtW91xxx164IEHTO5d02SxWLRgwQKNHj3a7K6gmaO+Gx71jcaC+m541DcaC+q74flTfTNy7qeqq6u1atUqDRs2rG6Z1WrVsGHDtHz5chN7BuBEUd+A/6K+Af9FfeO3EM79VG5urtxut+Lj472Wx8fHKysry6ReAWgI1Dfgv6hvwH9R3/gthHMAAAAAAExGOPdTsbGxstlsys7O9lqenZ2tli1bmtQrAA2B+gb8F/UN+C/qG7+FcO6nHA6H+vbtqy+//LJumcfj0ZdffqkzzjjDxJ4BOFHUN+C/qG/Af1Hf+C0BZncAJ88999yj8ePHq1+/fhowYIBmzJihsrIyTZw40eyuNSmlpaXasWNH3c/p6elKS0tTdHS0EhMTTewZmjPqu2FQ32iMqO+GQX2jMaK+G4bf1rcBvzZz5kwjMTHRcDgcxoABA4wVK1aY3aUmZ8mSJYakwx7jx483u2to5qjvE0d9o7Givk8c9Y3Givo+cf5a33zPOQAAAAAAJuOacwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAwA8lJydrwoQJDba9CRMmKDk5ucG2BwAAvBHOAQA4CebNmyeLxVL3CAwMVKdOnXT77bcrOzvb7O75tH//fj322GNKS0szuysAADQ7AWZ3AAAAf/bEE0+oXbt2qqys1Pfff6+XX35Zn3zyiTZs2KDg4GCzu+dl//79evzxx5WcnKxevXp5Pffaa6/J4/GY0zEAAJoBwjkAACfRBRdcoH79+kmSfv/73ysmJkYvvPCCPvjgA1111VWHtS8rK1NISMip7uZvstvtZncBAAC/xrR2AABOoXPPPVeSlJ6ergkTJig0NFQ7d+7UyJEjFRYWpmuuuUaS5PF4NGPGDHXv3l2BgYGKj4/XzTffrIKCAq/tGYahadOmKSEhQcHBwRo6dKg2btzo83cXFhbq7rvvVnJyspxOpxISEnT99dcrNzdXX3/9tfr37y9JmjhxYt10/Hnz5knyfc15WVmZpkyZorZt28rpdKpz5856/vnnZRiGVzuLxaLbb79dCxcuVI8ePeR0OtW9e3ctWrToRA8nAAB+g5FzAABOoZ07d0qSYmJiJEkul0sjRozQWWedpeeff75uqvvNN9+sefPmaeLEiZo8ebLS09M1a9YsrVmzRkuXLq0byZ46daqmTZumkSNHauTIkVq9erWGDx+u6upqr99bWlqqwYMHa/PmzbrhhhvUp08f5ebm6sMPP1RGRoa6du2qJ554QlOnTtVNN92kwYMHS5LOPPNMn/thGIYuvvhiLVmyRDfeeKN69eqlzz77TPfdd58yMzP14osverX//vvv9f777+u2225TWFiY/vrXv2rs2LHau3dv3bEAAKBZMwAAQIObO3euIclYvHixkZOTY+zbt894++23jZiYGCMoKMjIyMgwxo8fb0gyHnjgAa91v/vuO0OS8dZbb3ktX7RokdfygwcPGg6Hw7jwwgsNj8dT1+6hhx4yJBnjx4+vWzZ16lRDkvH+++8f1tef1/3pp58MScbcuXMPazN+/HgjKSmp7ueFCxcakoxp06Z5tbvssssMi8Vi7Nixo26ZJMPhcHgtW7t2rSHJmDlz5hGOIAAAzQvT2gEAOImGDRumuLg4tW3bVldeeaVCQ0O1YMECtWnTpq7Nrbfe6rXOu+++q4iICJ1//vnKzc2te/Tt21ehoaFasmSJJGnx4sWqrq7WHXfcIYvFUrf+XXfddVg//vOf/yg1NVVjxow57Ln/XfdoffLJJ7LZbJo8ebLX8ilTpsgwDH366aeHHYf27dvX/dyzZ0+Fh4dr165dx/y7AQDwR0xrBwDgJJo9e7Y6deqkgIAAxcfHq3PnzrJa68+NBwQEKCEhwWud7du3q6ioSC1atPC5zYMHD0qS9uzZI0nq2LGj1/NxcXGKioryWrZz506NHTv2hPfnZ3v27FHr1q0VFhbmtbxr165efftZYmLiYduIioo67Bp6AACaK8I5AAAn0YABA+ru1u6L0+n0CutS7c3gWrRoobfeesvnOnFxcQ3ax1PBZrP5XG784uZxAAA0V4RzAAAamfbt22vx4sUaNGiQgoKCjtguKSlJUu1Ie0pKSt3ynJycw0ak27dvrw0bNvzq7z2W6e1JSUlavHixSkpKvEbPt2zZ4tU3AABwdLjmHACARubyyy+X2+3Wk08+edhzLpdLhYWFkmqv47bb7Zo5c6bXCPSMGTMOW2/s2LFau3atFixYcNhzP6/78/er/7z9XzNy5Ei53W7NmjXLa/mLL74oi8WiCy644De3AQAA6jFyDgBAIzNkyBDdfPPNmj59utLS0jR8+HDZ7XZt375d7777rl566SVddtlliouL07333qvp06froosu0siRI7VmzRp9+umnio2N9drmfffdp/fee0/jxo3TDTfcoL59+yo/P18ffvih5syZo9TUVLVv316RkZGaM2eOwsLCFBISooEDB6pdu3aH9XHUqFEaOnSoHn74Ye3evVupqan6/PPP9cEHH+iuu+7yuvkbAAD4bYRzAAAaoTlz5qhv37565ZVX9NBDDykgIEDJycm69tprNWjQoLp206ZNU2BgoObMmaMlS5Zo4MCB+vzzz3XhhRd6bS80NFTfffedHn30US1YsEBvvvmmWrRoofPOO6/uhnR2u11vvvmmHnzwQd1yyy1yuVyaO3euz3ButVr14YcfaurUqZo/f77mzp2r5ORkPffcc5oyZcrJPTgAAPghi8GdWAAAAAAAMBXXnAMAAAAAYDLCOQAAAAAAJiOcAwAAAABgMsI5AAAAAAAmI5wDAAAAAGAywjkAAAAAACYjnAMAAAAAYDLCOQAAAAAAJiOcAwAAAABgMsI5AAAAAAAmI5wDAAAAAGAywjkAAAAAACYjnAMAAAAAYLL/B/XQG1yFxFqWAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1040x520 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ncols = 4\n",
    "nrows = 2\n",
    "\n",
    "fig, axes = plt.subplots(nrows=nrows, ncols=ncols, figsize=(ncols*2.6, nrows*2.6))\n",
    "axes = axes.flatten()\n",
    "\n",
    "fig.suptitle('Model Comparison - Normalized')\n",
    "fig.supylabel('True Label')\n",
    "fig.supxlabel('Predction')\n",
    "\n",
    "for i, (model_name, pred) in enumerate(zip(results['Model'], results['Predictions'])):\n",
    "\n",
    "    disp = ConfusionMatrixDisplay.from_predictions(\n",
    "        test_df['target'],\n",
    "        pred,\n",
    "        normalize='true',\n",
    "        ax=axes[i],\n",
    "        colorbar=False\n",
    "    )\n",
    "\n",
    "    axes[i].set_title(model_name)\n",
    "    axes[i].set_xlabel('')\n",
    "    axes[i].set_ylabel('')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "99654169",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1040x520 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ncols = 4\n",
    "nrows = 2\n",
    "\n",
    "fig, axes = plt.subplots(nrows=nrows, ncols=ncols, figsize=(ncols*2.6, nrows*2.6))\n",
    "axes = axes.flatten()\n",
    "\n",
    "fig.suptitle('Model Comparison - Raw Class Counts')\n",
    "fig.supylabel('True Label')\n",
    "fig.supxlabel('Predction')\n",
    "\n",
    "for i, (model_name, pred) in enumerate(zip(results['Model'], results['Predictions'])):\n",
    "\n",
    "    disp = ConfusionMatrixDisplay.from_predictions(\n",
    "        test_df['target'],\n",
    "        pred,\n",
    "        ax=axes[i],\n",
    "        colorbar=False\n",
    "    )\n",
    "\n",
    "    axes[i].set_title(model_name)\n",
    "    axes[i].set_xlabel('')\n",
    "    axes[i].set_ylabel('')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5e23b03a",
   "metadata": {},
   "source": [
    "## 5. Model metric optimization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "3c2598f1",
   "metadata": {},
   "outputs": [],
   "source": [
    "num_samples = 1000"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fa9ee276",
   "metadata": {},
   "source": [
    "### 5.1. ROC_AUC optimized thresholds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "62d69883",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LogisticRegression new threshold: 0.476\n",
      "DecisionTreeClassifier new threshold: 0.010\n",
      "VotingClassifier new threshold: 0.226\n",
      "BaggingClassifier new threshold: 0.132\n",
      "RandomForestClassifier new threshold: 0.113\n",
      "AdaBoostClassifier new threshold: 0.449\n",
      "GradientBoostingClassifier new threshold: 0.126\n",
      "StackingClassifier new threshold: 0.532\n"
     ]
    }
   ],
   "source": [
    "model_thres_results = {}\n",
    "\n",
    "for model in [lr_clf, dt_clf, voting_clf, bagging_clf, rf_clf, adaboost_clf, gb_clf, stacking_clf]:\n",
    "    TunedThreshCV = TunedThresholdClassifierCV(estimator=model, scoring='roc_auc') \n",
    "    TunedThreshCV.fit(X=train_df.drop('target', axis=1)[:num_samples], y=train_df['target'][:num_samples])\n",
    "    \n",
    "    model_thres_results[model] = TunedThreshCV.best_threshold_\n",
    "    print(f'{type(model).__name__} new threshold: {TunedThreshCV.best_threshold_:.3f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "6cd6e93f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "roc_auc optimized -  Logistic Regression 0.476\n",
      "roc_auc optimized -  Decision Tree 0.01\n",
      "roc_auc optimized -  Voting 0.226\n",
      "roc_auc optimized -  Bagging 0.132\n",
      "roc_auc optimized -  Random Forest 0.113\n",
      "roc_auc optimized -  AdaBoost 0.449\n",
      "roc_auc optimized -  Gradient Boosting 0.126\n",
      "roc_auc optimized -  Stacking 0.532\n"
     ]
    }
   ],
   "source": [
    "preds = {}\n",
    "\n",
    "for model_name, model in zip(results['Model'], [lr_clf, dt_clf, voting_clf, bagging_clf, rf_clf, adaboost_clf, gb_clf, stacking_clf]):\n",
    "    print(\"roc_auc optimized - \", model_name, round(model_thres_results[model], 3))\n",
    "\n",
    "    preds[model] = model.predict_proba(test_df.drop(['target'][:num_samples], axis=1))[:, 1] > model_thres_results[model]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "6b91913f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA+cAAAIMCAYAAACXCt1JAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjcsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvTLEjVAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAyGhJREFUeJzs3Xd0FFUbBvBns0l203shBNLoHUPvPdJDb0pCV7pSBJWOIKACAlJEAWkqRUB6RxEQ6b0EQiRAQnrve78/8u2YzW5gEwIbkud3zp6cnbk7c2cy78y8M3fuyIQQAkRERERERERkMEaGrgARERERERFRScfknIiIiIiIiMjAmJwTERERERERGRiTcyIiIiIiIiIDY3JOREREREREZGBMzomIiIiIiIgMjMk5ERERERERkYExOSciIiIiIiIyMCbnRERERERERAbG5JyICkwmk2HmzJn5/t2jR48gk8mwfv36Qq9TcXHy5EnIZDKcPHnS0FV5a3l6eiIwMLBQp1nQbf5VzJw5EzKZrNCnFxkZWWjTfBWvoz4tWrRAixYtXlouv3G2cOFCVKpUCSqV6tUqSACAVatWoWzZskhLSzN0VYqE3PuX9evXQyaT4dGjR2+0Hq9j30lE+mFyTvSWUx+8ZTIZTp8+rTVeCIEyZcpAJpOhU6dOBqghEZCRkYFvv/0WdevWhZWVFSwtLVG3bl18++23yMjIKPB0z5w5g5kzZyI2NrbwKvuWUSeY+nyo4OLj47FgwQJ88sknMDL67/Qp9zq2trZG8+bNsW/fvjyndfPmTbz33nsoXbo0FAoF3NzcMGDAANy8eTPP3zx48AAjRoyAt7c3lEolrK2t0bhxYyxduhQpKSkFWqbJkydDJpOhT58+Osert63t27frHD969Gid21VWVhbWrVuHFi1awN7eHgqFAp6enhg0aBAuXLgglQsMDER6ejpWr16tV30DAwMhk8lQo0YNCCG0xstkMowePVqvaRERFUXGhq4AERUOpVKJLVu2oEmTJhrDT506hdDQUCgUCgPVjAqiWbNmSElJgampqaGr8sqSkpLQsWNHnDp1Cp06dUJgYCCMjIxw8OBBjBs3Djt37sS+fftgYWGR72mfOXMGs2bNQmBgIGxtbTXG3b17VyOJKgwpKSkwNi5ah87KlStj48aNGsOmTp0KS0tLfPbZZwaqVfHz448/IjMzE/369dMa17ZtWwwcOBBCCISEhGDlypXo3LkzDhw4AD8/P42yO3fuRL9+/WBvb48hQ4bAy8sLjx49wg8//IDt27fj559/Rrdu3TR+s2/fPvTq1QsKhQIDBw5EtWrVkJ6ejtOnT2PSpEm4efMm1qxZk6/lEUJg69at8PT0xO+//46EhARYWVnlf8XkkpKSgu7du+PgwYNo1qwZPv30U9jb2+PRo0f49ddfsWHDBvz7779wd3eHUqlEQEAAvvnmG4wZM0bvC0jXr1/Hzp070aNHj1eub1H2/vvvo2/fvjx+E5UgResMg4gKrEOHDti2bRu+/fZbjeRhy5Yt8PX1LTJNWOnFUlNTYWpqCiMjIyiVSkNXp1B8/PHHOHXqFJYtW6ZxV+vDDz/EihUrMHr0aEycOBErV64s1Pm+jhPaovg/cXFxwXvvvacx7Msvv4Sjo6PW8FelUqmQnp5eJNfD67Zu3Tp06dJF57JXqFBBY1336NEDVapUwdKlSzWS8wcPHuD999+Ht7c3/vjjDzg5OUnjxo0bh6ZNm+L999/HtWvX4O3tDQAIDg5G37594eHhgePHj6NUqVLSb0aNGoWgoKAX3qXPy8mTJxEaGorjx4/Dz88PO3fuREBAQL6nk9ukSZNw8OBBLF68GOPHj9cYN2PGDCxevFhjWO/evbFw4UKcOHECrVq1eun0zczMUKZMGcyePRvdu3d/bS1CMjMzoVKpDHqBVC6XQy6XG2z+RPTmsVk7UTHRr18/REVF4ciRI9Kw9PR0bN++Hf3799f5m6SkJEyYMAFlypSBQqFAxYoV8dVXX2k1F0xLS8NHH30EJycnWFlZoUuXLggNDdU5zSdPnmDw4MFwcXGBQqFA1apV8eOPPxZ4uWJjY/HRRx/B09MTCoUC7u7uGDhwoMbFhufPn2PIkCFwcXGBUqlEzZo1sWHDBo3pqJ9z/+qrr7BixQp4e3vD3Nwc7dq1w+PHjyGEwJw5c+Du7g4zMzN07doV0dHRGtPw9PREp06dcPjwYdSqVQtKpRJVqlTBzp07NcpFR0dj4sSJqF69OiwtLWFtbY327dvj6tWrGuXUTUZ//vlnfP755yhdujTMzc0RHx+v81nY+/fvo0ePHnB1dYVSqYS7uzv69u2LuLg4qUxmZibmzJkDHx8fqSnpp59+qvVMp3pZTp8+jXr16kGpVMLb2xs//fRTgf5PeQkNDcUPP/yAVq1a6WxuOmrUKLRs2RJr167V2KbUzVM3b96MihUrQqlUwtfXF3/88YdUZubMmZg0aRIAwMvLS2pWrH4+M/dzk+pHQE6fPo2xY8fCyckJtra2GDFiBNLT0xEbG4uBAwfCzs4OdnZ2mDx5slYs5HwmVL1N6dOE/O+//8a7774LGxsbmJubo3nz5vjrr7+01sfp06dRt25dKJVK+Pj46N3ctyBiY2OlFgc2NjYYNGgQkpOTtZZX/X+oWrUqFAoFDh48CED/WF+2bBmqVq0Kc3Nz2NnZoU6dOtiyZUuB6qPv9q1LaGgo/P39YWFhAWdnZ3z00Ud6P+scHByMa9euoU2bNnqVr1y5MhwdHfHgwQON4YsWLUJycjLWrFmjkZgDgKOjI1avXo2kpCQsXLhQGr5w4UIkJibihx9+0EjM1cqVK4dx48bpVa+cNm/ejCpVqqBly5Zo06YNNm/enO9p5BYaGorVq1ejbdu2Wok5kJ1sTpw4Ee7u7tIwX19f2NvbY/fu3XrNw8jICJ9//jmuXbuG33777aXl83t8WLJkibR93bp1S+oT4d69e3jvvfdgY2MDJycnTJs2DUIIPH78GF27doW1tTVcXV3x9ddfa0w7PT0d06dPh6+vL2xsbGBhYYGmTZvixIkTL6177mfO1XXR9cm5r1OpVFiyZAmqVq0KpVIJFxcXjBgxAjExMRrTF0Jg7ty5cHd3h7m5OVq2bPnCRyuI6PXjnXOiYsLT0xMNGzbE1q1b0b59ewDAgQMHEBcXh759++Lbb7/VKC+EQJcuXXDixAkMGTIEtWrVwqFDhzBp0iQ8efJE4+7G0KFDsWnTJvTv3x+NGjXC8ePH0bFjR606hIeHo0GDBtIJvZOTEw4cOIAhQ4YgPj5e58naiyQmJqJp06a4ffs2Bg8ejHfeeQeRkZHYs2cPQkND4ejoiJSUFLRo0QJBQUEYPXo0vLy8sG3bNgQGBiI2NlbrpHXz5s1IT0/HmDFjEB0djYULF6J3795o1aoVTp48iU8++QRBQUFYtmwZJk6cqJVs3L9/H3369MEHH3yAgIAArFu3Dr169cLBgwfRtm1bAMDDhw+xa9cu9OrVC15eXggPD8fq1avRvHlz3Lp1C25ubhrTnDNnDkxNTTFx4kSkpaXpvFOTnp4OPz8/pKWlYcyYMXB1dcWTJ0+wd+9exMbGwsbGRvpfbdiwAT179sSECRPw999/Y/78+bh9+7bWiWxQUBB69uyJIUOGICAgAD/++CMCAwPh6+uLqlWr5ut/lZcDBw4gKysLAwcOzLPMwIEDceLECRw8eBBDhw6Vhp86dQq//PILxo4dC4VCge+++w7vvvsuzp8/j2rVqqF79+64d+8etm7disWLF8PR0REAtJKe3NTrb9asWTh37hzWrFkDW1tbnDlzBmXLlsW8efOwf/9+LFq0CNWqVcuz7k5OTlrNyTMyMvDRRx9p/A+PHz+O9u3bw9fXFzNmzICRkRHWrVuHVq1a4c8//0S9evUAZDfVbdeuHZycnDBz5kxkZmZixowZcHFxefFKLqDevXvDy8sL8+fPx6VLl7B27Vo4OztjwYIFGuWOHz+OX3/9FaNHj4ajoyM8PT31jvXvv/8eY8eORc+ePTFu3Dikpqbi2rVr+Pvvv7UuGupTn/xs3zmlpKSgdevW+PfffzF27Fi4ublh48aNOH78uF7r6syZMwCAd955R6/ycXFxiImJgY+Pj8bw33//HZ6enmjatKnO3zVr1gyenp4ad8J///13eHt7o1GjRnrNWx9paWnYsWMHJkyYACD74u6gQYMQFhYGV1fXAk/3wIEDyMzMxPvvv5+v373zzjs6L1blpX///pgzZw5mz56Nbt265Xn3PL/Hh3Xr1iE1NRXDhw+HQqGAvb29NK5Pnz6oXLkyvvzyS+zbtw9z586Fvb09Vq9ejVatWmHBggXYvHkzJk6ciLp166JZs2YAsvsqWLt2Lfr164dhw4YhISEBP/zwA/z8/HD+/HnUqlVL7+Xu3r07ypUrpzHs4sWLWLJkCZydnaVhI0aMwPr16zFo0CCMHTsWwcHBWL58OS5fvoy//voLJiYmAIDp06dj7ty56NChAzp06IBLly6hXbt2SE9P17tORFTIBBG91datWycAiH/++UcsX75cWFlZieTkZCGEEL169RItW7YUQgjh4eEhOnbsKP1u165dAoCYO3euxvR69uwpZDKZCAoKEkIIceXKFQFAjBw5UqNc//79BQAxY8YMadiQIUNEqVKlRGRkpEbZvn37ChsbG6lewcHBAoBYt27dC5dt+vTpAoDYuXOn1jiVSiWEEGLJkiUCgNi0aZM0Lj09XTRs2FBYWlqK+Ph4jXk6OTmJ2NhYqezUqVMFAFGzZk2RkZEhDe/Xr58wNTUVqamp0jAPDw8BQOzYsUMaFhcXJ0qVKiVq164tDUtNTRVZWVka9Q0ODhYKhULMnj1bGnbixAkBQHh7e0vrJve4EydOCCGEuHz5sgAgtm3bluf6Uv+vhg4dqjF84sSJAoA4fvy41rL88ccf0rDnz58LhUIhJkyYkOc88mv8+PECgLh8+XKeZS5duiQAiI8//lgaBkAAEBcuXJCGhYSECKVSKbp16yYNW7RokQAggoODtabr4eEhAgICpO/qWPHz85O2HyGEaNiwoZDJZOKDDz6QhmVmZgp3d3fRvHlzjWnm3uZzGzlypJDL5dK6VqlUonz58lrzTE5OFl5eXqJt27bSMH9/f6FUKkVISIg07NatW0Iul4v8Hq6rVq2qVXe1GTNmCABi8ODBGsO7desmHBwcNIYBEEZGRuLmzZsaw/WN9a5du4qqVau+sK761ic/23fz5s01ll+9n/j111+lYUlJSaJcuXIacZaXzz//XAAQCQkJWuMAiCFDhoiIiAjx/PlzceHCBfHuu+8KAGLRokVSudjYWAFAdO3a9YXz6tKliwAg4uPjRVxcnF6/ya/t27cLAOL+/ftCCCHi4+OFUqkUixcv1iin3g/ltd8ZNWqUxrb50UcfvTTedRk+fLgwMzN7abmAgABhYWEhhBBiw4YNWscHAGLUqFHS9/weH6ytrcXz58815qnePocPHy4NU+8fZDKZ+PLLL6XhMTExwszMTGO/k5mZKdLS0jSmGRMTI1xcXLS2+dz7F/U+S9f+TQghIiIiRNmyZUX16tVFYmKiEEKIP//8UwAQmzdv1ih78OBBjeHPnz8XpqamomPHjhr7pk8//VQA0FgGInpz2KydqBjp3bs3UlJSsHfvXiQkJGDv3r15Nmnfv38/5HI5xo4dqzF8woQJEELgwIEDUjkAWuVy3wUXQmDHjh3o3LkzhBCIjIyUPn5+foiLi8OlS5fytTw7duxAzZo1tTpHAiDdKdm/fz9cXV01OmkyMTHB2LFjkZiYiFOnTmn8rlevXtJdZgCoX78+AOC9997TeFa/fv36SE9Px5MnTzR+7+bmplEfa2trDBw4EJcvX0ZYWBiA7Ged1R2RZWVlISoqCpaWlqhYsaLOdRAQEAAzM7MXrgt1nQ8dOqTV1FdN/b/6+OOPNYar747lfi61SpUqGnfwnJycULFiRTx8+PCFdcmPhIQEAHhhR1PqcfHx8RrDGzZsCF9fX+l72bJl0bVrVxw6dAhZWVkFrtOQIUM07rTVr18fQggMGTJEGiaXy1GnTp18rYuffvoJ3333HRYuXIiWLVsCAK5cuYL79++jf//+iIqKkmIiKSkJrVu3xh9//AGVSoWsrCwcOnQI/v7+KFu2rDTNypUra3UoVlg++OADje9NmzZFVFSU1v+hefPmqFKlivQ9P7Fua2uL0NBQ/PPPP69cn/xu3znt378fpUqVQs+ePaVh5ubmGD58+EvrBQBRUVEwNjaGpaWlzvE//PADnJyc4OzsjDp16uDYsWOYPHmyRl31iYWc4+Pj46VlL4yO2nLavHkz6tSpI92FtbKyQseOHV+5aXtB62tnZ4eUlJQ89226DBgwAOXLl8fs2bN19twO5P/40KNHjzxb3uRs1aPeP+Teb9ja2mrtQ+VyudSSRqVSITo6GpmZmahTp06+j4k5ZWVloV+/fkhISMBvv/0mdai5bds22NjYoG3bthqx6evrC0tLS6k5/dGjR6VWZDn3h/lt4UZEhYvJOVEx4uTkhDZt2mDLli3YuXMnsrKyNE5GcwoJCYGbm5vWSVTlypWl8eq/RkZGWs0zK1asqPE9IiICsbGx0rOUOT+DBg0CkP3sX348ePAA1apVe2GZkJAQlC9fXqtX7tzLoZYz8QH+S3rLlCmjc3juZ/TKlSun1YSyQoUKACA9F6hSqbB48WKUL18eCoUCjo6OcHJywrVr1zSeD1fz8vJ64TKqy3z88cdYu3YtHB0d4efnhxUrVmhMT/2/yt3s0dXVFba2ti9dF0D2SXLuZc4tIiICYWFh0icxMTHPsurtS52Y6JJX0lK+fHmtshUqVEBycjIiIiJeWMcXyc828LJ1oXblyhV88MEH6Nevn0ZCdv/+fQDZF2Byx8XatWuRlpaGuLg4REREICUlRecy5461wpJ7PdjZ2QHQ3uZzb5/5ifVPPvkElpaWqFevHsqXL49Ro0bl2Xz5ZfXJ7/adU0hIiM7YLax127VrVxw5cgT79u2TngtOTk7W2C/pEws5x1tZWcHa2lqv3+RHbGws9u/fj+bNmyMoKEj6NG7cGBcuXMC9e/cKPO2C1ledXOencze5XI7PP/8cV65cwa5du3SWye/x4UX7Yl37DaVSKT1Ok3N47hjasGEDatSoAaVSCQcHBzg5OWHfvn06jwf6+vzzz3H8+HFs2bJF4/h8//59xMXFwdnZWSs+ExMTpdhUL3vufY6Tk5MUe0T05vGZc6Jipn///hg2bBjCwsLQvn17rddLvS4qlQpA9h3ovHr8rVGjxhupy4vk1fNtXsPzuiPzIvPmzcO0adMwePBgzJkzB/b29jAyMsL48eOl9ZTTy+6aq3399dcIDAzE7t27cfjwYYwdOxbz58/HuXPnNDpY0vcEt6DLXLduXY2T2hkzZkidpOWmPgm+du1ans9WXrt2DQA07s6+TvnZBvT5/8fExKBHjx6oUKEC1q5dqzFO/f9etGhRnstvaWmpd8dkhUnf/3/u7TM/sV65cmXcvXsXe/fuxcGDB7Fjxw589913mD59OmbNmlWg+hjife0ODg7IzMzM83Vj7u7uUmdxHTp0gKOjI0aPHo2WLVuie/fuALKTtlKlSknbe16uXbuG0qVLS4mum5sbbty4UWjLsm3bNqSlpeHrr7/W6rwMyL6rrv7fqHumz+s96snJyRq911eqVAlAdv8J+XmWOiYmBubm5nrvC9UGDBggPXvu7++fr9/q8qL569o+9dlmN23ahMDAQPj7+2PSpElwdnaGXC7H/PnztToM1NeuXbuwYMECzJkzB++++67GOJVKBWdn5zxbQbysTw4iMiwm50TFTLdu3TBixAicO3cOv/zyS57lPDw8cPToUa2TzTt37kjj1X9VKhUePHigcZfp7t27GtNT9+SelZWld4/GL+Pj4/PSk1IPDw9cu3YNKpVK4+5I7uUoLEFBQRBCaCQI6jtNnp6eAIDt27ejZcuW+OGHHzR+Gxsbq3WXJb+qV6+O6tWr4/PPP8eZM2fQuHFjrFq1CnPnzpX+V/fv35eSYiC7o77Y2NhCWxebN2/WOFlXv/JJl/bt20Mul2Pjxo15dqz2008/wdjYWOskU33XOad79+7B3NxcOsE0RKKWk0qlwoABAxAbG4ujR4/C3NxcY7z6jpa1tfUL48LJyQlmZmY6lzl3rBlafmPdwsICffr0QZ8+fZCeno7u3bvjiy++wNSpU/P1SrZX2b49PDxw48YNrdjVd92qk87g4GC9LjKOGDECixcvxueff67RYVmnTp3w/fff4/Tp02jSpInW7/788088evQII0aMkIZ16tQJa9aswdmzZ9GwYUO96vsimzdvRrVq1TBjxgytcatXr8aWLVuk5Fy9TvNaT3fv3tVY7+p437RpU746hQsODtb4n+pLffdcfdEytzd9fNBl+/bt8Pb2xs6dOzW2PV3rXx/37t1DQEAA/P398emnn2qN9/HxwdGjR9G4ceMXXmxQL/v9+/c19uERERF6txgiosLHZu1ExYylpSVWrlyJmTNnonPnznmW69ChA7KysrB8+XKN4YsXL4ZMJpN6fFf/zd3b+5IlSzS+y+Vy9OjRAzt27NCZUBekGXKPHj1w9epVnb0wq+9MdOjQAWFhYRoXIjIzM7Fs2TJYWlqiefPm+Z7vizx9+lSjPvHx8fjpp59Qq1YtqZdjuVyudbdv27ZtWs+v50d8fDwyMzM1hlWvXh1GRkbSXdcOHToA0P7ffPPNNwCgs4f9gmjcuDHatGkjfV6UnJcpUwaDBg3C0aNHdb7HfNWqVTh+/DiGDBmicfcfAM6ePavxTObjx4+xe/dutGvXTrpjpX7OMjY2thCWLP9mzZqFQ4cOYevWrTqbxPr6+sLHxwdfffWVzub/6riQy+Xw8/PDrl278O+//0rjb9++jUOHDr2+BSiA/MR6VFSUxjhTU1NUqVIFQghkZGTka76vsn136NABT58+xfbt26Vh6lea6UOdFF+4cEGv8sbGxpgwYQJu376tkTROmjQJZmZmGDFihNa6iY6OxgcffABzc3PpFYEAMHnyZFhYWGDo0KEIDw/XmteDBw+wdOlSver1+PFj/PHHH+jduzd69uyp9Rk0aBCCgoLw999/AwBKlSqFWrVqYdOmTVoxdvHiRZw7d046RgDZ8T5s2DAcPnwYy5Yt05q/SqXC119/rfUqzkuXLhW4N/r33nsP5cqV02qJAbz544Mu6n1VzmPC33//jbNnz+Z7WomJiejWrRtKly6NDRs26Lw42bt3b2RlZWHOnDla4zIzM6X/Y5s2bWBiYoJly5Zp1C13fBHRm8U750TFUF5NTXPq3LkzWrZsic8++wyPHj1CzZo1cfjwYezevRvjx4+X7vjVqlUL/fr1w3fffYe4uDg0atQIx44dQ1BQkNY0v/zyS5w4cQL169fHsGHDUKVKFURHR+PSpUs4evSo1nvDX2bSpEnYvn07evXqhcGDB8PX1xfR0dHYs2cPVq1ahZo1a2L48OFYvXo1AgMDcfHiRXh6emL79u3466+/sGTJkkLvSKlChQoYMmQI/vnnH7i4uODHH39EeHg41q1bJ5Xp1KkTZs+ejUGDBqFRo0a4fv06Nm/e/MIk9mWOHz+O0aNHo1evXqhQoQIyMzOxceNGKVECgJo1ayIgIABr1qxBbGwsmjdvjvPnz2PDhg3w9/eXOil70xYvXow7d+5g5MiROHjwoHSH/NChQ9i9ezeaN2+us3lttWrV4Ofnp/EqNQAaJ+HqDuM+++wz9O3bFyYmJujcubOUtL9O169fx5w5c9CsWTM8f/4cmzZt0hj/3nvvwcjICGvXrkX79u1RtWpVDBo0CKVLl8aTJ09w4sQJWFtb4/fff5eW6+DBg2jatClGjhwpJRFVq1Z9aVPoN03fWG/Xrh1cXV3RuHFjuLi44Pbt21i+fDk6duyY79h8le172LBhWL58OQYOHIiLFy+iVKlS2Lhxo1ZLh7x4e3ujWrVqOHr0KAYPHqzXbwIDAzF9+nQsWLBAanJdvnx5bNiwAQMGDED16tUxZMgQeHl54dGjR/jhhx8QGRmJrVu3ajxD7OPjgy1btkiv8ho4cCCqVauG9PR0nDlzRno1mD62bNkivUZTlw4dOsDY2BibN2+WOsv85ptv4Ofnh1q1aiEwMBBubm64ffs21qxZg1KlSmHq1Kka0/j666/x4MEDjB07Fjt37kSnTp1gZ2eHf//9F9u2bcOdO3fQt29fqfzFixcRHR2Nrl276rUMucnlcnz22WdSfwc5venjgy6dOnXCzp070a1bN3Ts2BHBwcFYtWoVqlSp8sL+OnSZNWsWbt26hc8//1yrpYCPjw8aNmyI5s2bY8SIEZg/fz6uXLmCdu3awcTEBPfv38e2bduwdOlS9OzZE05OTpg4cSLmz5+PTp06oUOHDrh8+TIOHDjwyi28iOgVvMGe4YnoNcj5KrUXyf0qNSGESEhIEB999JFwc3MTJiYmonz58mLRokUar1URQoiUlBQxduxY4eDgICwsLETnzp3F48ePdb5WKjw8XIwaNUqUKVNGmJiYCFdXV9G6dWuxZs0aqYy+r1ITQoioqCgxevRoUbp0aWFqairc3d1FQECAxiucwsPDxaBBg4Sjo6MwNTUV1atX15q2ep45X20kRN6vCtK1XtXr8NChQ6JGjRpCoVCISpUqaf02NTVVTJgwQZQqVUqYmZmJxo0bi7Nnz2q93ulFrynK/Sq1hw8fisGDBwsfHx+hVCqFvb29aNmypTh69KjG7zIyMsSsWbOEl5eXMDExEWXKlBFTp07VeCVczmXJLXcdC0taWppYvHix8PX1FRYWFsLc3Fy88847YsmSJSI9PV2rPP7/SqRNmzaJ8uXLC4VCIWrXrq3zlVdz5swRpUuXFkZGRhqvHcrrVWq5Y0X9qqSIiAiN4Tlf25SzXuptXv0/yuuT0+XLl0X37t2Fg4ODUCgUwsPDQ/Tu3VscO3ZMo9ypU6eEr6+vMDU1Fd7e3mLVqlVS/fJDn1ep5V5eXa9tUv8fdNEn1levXi2aNWsmLbePj4+YNGmSiIuLK1B99N2+dW3HISEhokuXLsLc3Fw4OjqKcePGSa+Xetmr1IQQ4ptvvhGWlpZarz180TqaOXOmzulfu3ZN9OvXT5QqVUpad/369RPXr1/Pc/737t0Tw4YNE56ensLU1FRYWVmJxo0bi2XLlmktf16qV68uypYt+8IyLVq0EM7Ozhqvljx37pzo1KmTsLOzE8bGxqJ06dJi6NChIjQ0VOc0MjMzxdq1a0XTpk2FjY2NMDExER4eHmLQoEFar1n75JNPRNmyZbWOO7roikkhsrcLHx8fnf+LVzk+CJG//YMQ2dteztcHqlQqMW/ePOHh4SHtx/bu3SsCAgKEh4eHxm9zH1Nzx0BAQECe+5vcrz5bs2aN8PX1FWZmZsLKykpUr15dTJ48WTx9+lQqk5WVJWbNmiUdq1q0aCFu3Lihte8kojdHJkQBejsiIiqBPD09Ua1aNezdu9fQVSn2ZDIZRo0apfXYBZGhxMXFwdvbGwsXLtR4fRYVXFpaGjw9PTFlyhSMGzfO0NUhIjI4PnNORERE9BI2NjaYPHkyFi1apPOtC5R/69atg4mJidY77omISireOSci0hPvnL85vHNOlD9xcXF5vvJMTd1pJRERFU3sEI6IiIjoLTdu3Dhs2LDhhWV4P4aIqGjjnXMiIiKit9ytW7fw9OnTF5bR5730RERkOEzOiYiIiIiIiAyMHcIRERERERERGRiTcyIiIiIiIiIDY3JOREREREREZGBMzomIiIiIiIgMjMk5ERERERERkYExOSciIiIiIiIyMCbnRERERERERAbG5JyIiIiIiIjIwJicExERERERERkYk3MiIiIiIiIiA2NyTkRERERERGRgTM6JiIiIiIiIDIzJOREREREREZGBMTknIiIiIiIiMjAm58VEixYt0KJFi0KbnqenJwIDAwttegTIZDLMnDnT0NUgeqFHjx5BJpNh/fr1+fpdYe+DiOjtEhgYCE9PT0NXg4gMhLlD4WByXsjWr18PmUyGCxcuGLoqL3XmzBnMnDkTsbGxr3U+np6ekMlk0sfCwgL16tXDTz/99FrnS/Q2Uu9D1B+lUgk3Nzf4+fnh22+/RUJCgqGrWGTk3rfk9cnvhQai4qJLly4wNzd/4X5jwIABMDU1RVRU1Eun9/TpU8ycORNXrlwpxFoS0YvkPi+QyWRwdnZGy5YtceDAAUNXjwqZsaErQIXj8OHD+f7NmTNnMGvWLAQGBsLW1lZj3N27d2FkVHjXbmrVqoUJEyYAAJ49e4a1a9ciICAAaWlpGDZsWKHNpyhLSUmBsTFDjvQze/ZseHl5ISMjA2FhYTh58iTGjx+Pb775Bnv27EGNGjVey3w9PDyQkpICExOTfP2uIPugV7VkyRIkJiZK3/fv34+tW7di8eLFcHR0lIY3atTojdeNqCgYMGAAfv/9d/z2228YOHCg1vjk5GTs3r0b7777LhwcHF46vadPn2LWrFnw9PRErVq1NMZ9//33UKlUhVV1IspFfV4ghEB4eDjWr1+PDh064Pfff0enTp0MXb1Czx1KKmYKxYSpqWmhTk+hUBTq9EqXLo333ntP+h4YGAhvb28sXrz4jSfnSUlJsLCweKPzBAClUvnG50lvr/bt26NOnTrS96lTp+L48ePo1KkTunTpgtu3b8PMzKzQ56u+W59fhb0P0oe/v7/G97CwMGzduhX+/v4vbF5rqH0A0ZvWpUsXWFlZYcuWLTqT8927dyMpKQkDBgx45Xnl94IeEeVP7vOCIUOGwMXFBVu3bi0SyXlh5w4lFS9vGMjly5fRvn17WFtbw9LSEq1bt8a5c+e0yl27dg3NmzeHmZkZ3N3dMXfuXKxbtw4ymQyPHj2Syul63nPZsmWoWrUqzM3NYWdnhzp16mDLli0AgJkzZ2LSpEkAAC8vL6mZjHqaup4biY2NxUcffQRPT08oFAq4u7tj4MCBiIyMzPfyOzk5oVKlSnjw4IHGcJVKhSVLlqBq1apQKpVwcXHBiBEjEBMTo1Vu5syZcHNzg7m5OVq2bIlbt25p1VvdFOjUqVMYOXIknJ2d4e7uLo0/cOAAmjZtCgsLC1hZWaFjx464efOmxrzCwsIwaNAguLu7Q6FQoFSpUujatavG+r9w4QL8/Pzg6OgIMzMzeHl5YfDgwRrT0fXMuT7bgXoZ/vrrL3z88cdwcnKChYUFunXrhoiICH1XORUDrVq1wrRp0xASEoJNmzZpjLtz5w569uwJe3t7KJVK1KlTB3v27NGaxsviWNcz5/rEgK590PPnz6WTB6VSiZo1a2LDhg0aZdTz++qrr7BmzRr4+PhAoVCgbt26+Oeff15thSH7QqClpSUePHiADh06wMrKSkpE9N3fAPrtK4iKGjMzM3Tv3h3Hjh3D8+fPtcZv2bIFVlZW6NKlCx4+fIhevXrB3t4e5ubmaNCgAfbt2yeVPXnyJOrWrQsAGDRokNZjI7mfOc9vbG/btg1VqlSBUqlEtWrV8Ntvv/E5dqIXsLW1hZmZmUarzK+++gqNGjWCg4MDzMzM4Ovri+3bt2v9NiUlBWPHjoWjo6O0D3jy5InOc9WTJ0+iTp06UCqV8PHxwerVqzFz5kzIZDKNcnmdg+tz/qrveX1JwDvnBnDz5k00bdoU1tbWmDx5MkxMTLB69Wq0aNECp06dQv369QEAT548QcuWLSGTyTB16lRYWFhg7dq1el2Z+v777zF27Fj07NkT48aNQ2pqKq5du4a///4b/fv3R/fu3XHv3j2tJqBOTk46p5eYmIimTZvi9u3bGDx4MN555x1ERkZiz549CA0N1WhCqo/MzEyEhobCzs5OY/iIESOwfv16DBo0CGPHjkVwcDCWL1+Oy5cv46+//pKuzE+dOhULFy5E586d4efnh6tXr8LPzw+pqak65zdy5Eg4OTlh+vTpSEpKAgBs3LgRAQEB8PPzw4IFC5CcnIyVK1eiSZMmuHz5snRC0KNHD9y8eRNjxoyBp6cnnj9/jiNHjuDff/+Vvrdr1w5OTk6YMmUKbG1t8ejRI+zcufOF60Df7UBtzJgxsLOzw4wZM/Do0SMsWbIEo0ePxi+//JKvdU9vt/fffx+ffvopDh8+LLU6uXnzJho3bozSpUtjypQpsLCwwK+//gp/f3/s2LED3bp1A1DwOH5ZDOiSkpKCFi1aICgoCKNHj4aXlxe2bduGwMBAxMbGYty4cRrlt2zZgoSEBIwYMQIymQwLFy5E9+7d8fDhw1e+I5eZmQk/Pz80adIEX331FczNzQHov7/Rd19BVBQNGDAAGzZswK+//orRo0dLw6Ojo3Ho0CH069cP8fHxaNSoEZKTkzF27Fg4ODhgw4YN6NKlC7Zv345u3bqhcuXKmD17NqZPn47hw4ejadOmAF7+2Ig+sb1v3z706dMH1atXx/z58xETE4MhQ4agdOnSr2/FEL1l4uLiEBkZCSEEnj9/jmXLliExMVGjZerSpUvRpUsXDBgwAOnp6fj555/Rq1cv7N27Fx07dpTKBQYG4tdff8X777+PBg0a4NSpUxrj1S5fvox3330XpUqVwqxZs5CVlYXZs2fnmS/oos/5a37P64s1QYVq3bp1AoD4559/8izj7+8vTE1NxYMHD6RhT58+FVZWVqJZs2bSsDFjxgiZTCYuX74sDYuKihL29vYCgAgODpaGN2/eXDRv3lz63rVrV1G1atUX1nXRokVa01Hz8PAQAQEB0vfp06cLAGLnzp1aZVUq1Qvn4+HhIdq1ayciIiJERESEuH79unj//fcFADFq1Cip3J9//ikAiM2bN2v8/uDBgxrDw8LChLGxsfD399coN3PmTAFAo97q/0eTJk1EZmamNDwhIUHY2tqKYcOGaUwjLCxM2NjYSMNjYmIEALFo0aI8l++333576f9cCCEAiBkzZkjf9d0O1MvQpk0bjXX90UcfCblcLmJjY184X3q76LMPsbGxEbVr15a+t27dWlSvXl2kpqZKw1QqlWjUqJEoX768NEyfOA4ODhYAxLp164QQ+sWAENr7oCVLlggAYtOmTdKw9PR00bBhQ2FpaSni4+M15ufg4CCio6Olsrt37xYAxO+///7C+eaka58WEBAgAIgpU6ZolNV3f6PvvoKoqMrMzBSlSpUSDRs21Bi+atUqAUAcOnRIjB8/XgAQf/75pzQ+ISFBeHl5CU9PT5GVlSWEEOKff/7R2D/kFBAQIDw8PKTv+Ynt6tWrC3d3d5GQkCANO3nypACgMU2ikkh9XpD7o1AoxPr16zXKJicna3xPT08X1apVE61atZKGXbx4UQAQ48eP1ygbGBioda7auXNnYW5uLp48eSINu3//vjA2Nha508jcuYO+56/5Oa8vCdis/Q3LysrC4cOH4e/vD29vb2l4qVKl0L9/f5w+fRrx8fEAgIMHD6Jhw4Yana7Y29vr9WyYra0tQkNDC6VZKADs2LEDNWvWlO7A5ZS7WYsuhw8fhpOTE5ycnFC9enVs3LgRgwYNwqJFi6Qy27Ztg42NDdq2bYvIyEjp4+vrC0tLS5w4cQIAcOzYMWRmZmLkyJEa8xgzZkye8x82bBjkcrn0/ciRI4iNjUW/fv005iWXy1G/fn1pXmZmZjA1NcXJkyd1NnUFIHWmt3fvXmRkZLx0XQD52w7Uhg8frrGumzZtiqysLISEhOg1Tyo+LC0tpd6Xo6Ojcfz4cfTu3RsJCQnSthwVFQU/Pz/cv38fT548AVCwONYnBnTZv38/XF1d0a9fP2mYiYkJxo4di8TERJw6dUqjfJ8+fTRa0qjvyj18+FDveb7Ihx9+qPFd3/2NvvsKoqJKLpejb9++OHv2rMajKFu2bIGLiwtat26N/fv3o169emjSpIk03tLSEsOHD8ejR49w69atAs//ZbH99OlTXL9+HQMHDoSlpaVUrnnz5qhevXqB50tU3KxYsQJHjhzBkSNHsGnTJrRs2RJDhw7VaKmZsy+amJgYxMXFoWnTprh06ZI0/ODBgwDw0vPorKwsHD16FP7+/nBzc5OGlytXDu3bt9e73i87fy3IeX1xxuT8DYuIiEBycjIqVqyoNa5y5cpQqVR4/PgxACAkJATlypXTKqdrWG6ffPIJLC0tUa9ePZQvXx6jRo3CX3/9VeB6P3jwANWqVSvw7+vXr48jR47g4MGD+Oqrr2Bra4uYmBiNTqTu37+PuLg4ODs7S4m8+pOYmCg9L6cO5tzrwd7eXquZvJqXl5fG9/v37wPIfoY397wOHz4szUuhUGDBggU4cOAAXFxc0KxZMyxcuBBhYWHStJo3b44ePXpg1qxZcHR0RNeuXbFu3TqkpaXluT7ysx2olS1bVuO7elnzkzBR8ZCYmAgrKysAQFBQEIQQmDZtmta2PGPGDACQtueCxLE+MaBLSEgIypcvr9Vza+XKlaXxOb3O7dvY2FijrwlA//2NvvsKoqJMfVFf3e9MaGgo/vzzT/Tt2xdyuRwhISF5Ho8A7XjNj5fFdl7H9LyGEZVU9erVQ5s2bdCmTRsMGDAA+/btQ5UqVTB69Gikp6cDyL5R1KBBAyiVStjb28PJyQkrV65EXFycNJ2QkBAYGRlpnRvnjrfnz58jJSXllWOzoPuAF53XF2d85ryYqly5Mu7evYu9e/fi4MGD2LFjB7777jtMnz4ds2bNeuP1cXR0RJs2bQAAfn5+qFSpEjp16oSlS5fi448/BpDdGYSzszM2b96scxr5eb4lt9y9Wqtf97Jx40a4urpqlc/Zucb48ePRuXNn7Nq1C4cOHcK0adMwf/58HD9+HLVr14ZMJsP27dtx7tw5/P777zh06BAGDx6Mr7/+GufOndO4E/Aqct75z0kIUSjTp7dDaGgo4uLipIOYelueOHEi/Pz8dP7mVU9wXxYDheF1bt8KhULrIoG++5v87CuIiipfX19UqlQJW7duxaeffoqtW7dCCFEovbS/DI9dRK+HkZERWrZsiaVLl+L+/fuIjo5Gly5d0KxZM3z33XcoVaoUTExMsG7dOunCnCFwH5A/PKt4w5ycnGBubo67d+9qjbtz5w6MjIxQpkwZANnvGw4KCtIqp2uYLhYWFujTpw/69OmD9PR0dO/eHV988QWmTp0KpVKpV3N0NR8fH9y4cUPv8i/TsWNHNG/eHPPmzcOIESNgYWEBHx8fHD16FI0bN37hK6I8PDwAZK+HnFf9oqKi9L7L5uPjAwBwdnaWLhq8rPyECRMwYcIE3L9/H7Vq1cLXX3+t0WN2gwYN0KBBA3zxxRfYsmULBgwYgJ9//hlDhw7Vml5+tgOinDZu3AgAUiKufizCxMTkpdvyq8SxPjGQk4eHB65duwaVSqWRGN+5c0cab0j67m/yu68gKqoGDBiAadOm4dq1a9iyZQvKly8v9b7u4eGR5/FIPR7Q7zG2/Mp5TM9N3/MdopIqMzMTQHaLuh07dkCpVOLQoUManUevW7dO4zceHh5QqVQIDg5G+fLlpeG5483Z2RlKpfK1x2ZhnNcXJ2zW/obJ5XK0a9cOu3fv1nj2Kzw8HFu2bEGTJk1gbW0NIPvk++zZs7hy5YpULjo6Os87PTlFRUVpfDc1NUWVKlUghJCei1a/5zc2Nval0+vRoweuXr2K3377TWtcQa98ffLJJ4iKisL3338PAOjduzeysrIwZ84crbKZmZlSPVu3bg1jY2OsXLlSo8zy5cv1nrefnx+sra0xb948nc+Jq1/xkJycrNVTpI+PD6ysrKRm6zExMVrrQN1PQF5N2/OzHRCpHT9+HHPmzIGXl5d0x8vZ2RktWrTA6tWr8ezZM63f5HxdSUHiWJ8Y0KVDhw4ICwvT6I01MzMTy5Ytg6WlJZo3b/7ihX3N9N3f6LuvICrq1PuM6dOn48qVKxp3zTt06IDz58/j7Nmz0rCkpCSsWbMGnp6eqFKlCoD8nTfoy83NDdWqVcNPP/2ExMREafipU6dw/fr1QpsPUXGTkZGBw4cPw9TUFJUrV4ZcLodMJkNWVpZU5tGjR9i1a5fG79QX97/77juN4cuWLdP4LpfL0aZNG+zatQtPnz6VhgcFBeHAgQOFthyFcV5fnPDO+Wvy448/Sh0u5DRu3DjMnTsXR44cQZMmTTBy5EgYGxtj9erVSEtLw8KFC6WykydPxqZNm9C2bVuMGTNGepVa2bJlER0d/cIr2O3atYOrqysaN24MFxcX3L59G8uXL0fHjh2lZ1V9fX0BAJ999hn69u0LExMTdO7cWTr45jRp0iRs374dvXr1wuDBg+Hr64vo6Gjs2bMHq1atQs2aNfO9jtq3b49q1arhm2++wahRo9C8eXOMGDEC8+fPx5UrV9CuXTuYmJjg/v372LZtG5YuXYqePXvCxcUF48aNw9dff40uXbrg3XffxdWrV3HgwAE4OjrqdWXf2toaK1euxPvvv4933nkHffv2hZOTE/7991/s27cPjRs3xvLly3Hv3j20bt0avXv3RpUqVWBsbIzffvsN4eHh6Nu3LwBgw4YN+O6779CtWzf4+PggISEB33//PaytrdGhQ4c866DvdkAl04EDB3Dnzh1kZmYiPDwcx48fx5EjR+Dh4YE9e/ZAqVRKZVesWIEmTZqgevXqGDZsGLy9vREeHo6zZ88iNDQUV69eBVCwONYnBnQZPnw4Vq9ejcDAQFy8eBGenp7Yvn07/vrrLyxZskTaDxmKvvsbffcVREWdl5cXGjVqhN27dwOARnI+ZcoUbN26Fe3bt8fYsWNhb2+PDRs2IDg4GDt27JBav/j4+MDW1harVq2ClZUVLCwsUL9+fa1nV/Nr3rx56Nq1Kxo3boxBgwYhJiYGy5cvR7Vq1TQSdqKSTH1eAGQ/D75lyxbcv38fU6ZMgbW1NTp27IhvvvkG7777Lvr374/nz59jxYoVKFeuHK5duyZNx9fXFz169MCSJUsQFRUlvUrt3r17ADRbyMycOROHDx9G48aN8eGHHyIrK0uKzZw3D19FYZzXFyuG6ia+uMrrdQfqz+PHj4UQQly6dEn4+fkJS0tLYW5uLlq2bCnOnDmjNb3Lly+Lpk2bCoVCIdzd3cX8+fPFt99+KwCIsLAwqVzu1xitXr1aNGvWTDg4OAiFQiF8fHzEpEmTRFxcnMb058yZI0qXLi2MjIw0XkGU+3UIQmS/xm306NGidOnSwtTUVLi7u4uAgAARGRn5wnXi4eEhOnbsqHPc+vXrtV7LsmbNGuHr6yvMzMyElZWVqF69upg8ebJ4+vSpVCYzM1NMmzZNuLq6CjMzM9GqVStx+/Zt4eDgID744AOt/0der6U6ceKE8PPzEzY2NkKpVAofHx8RGBgoLly4IIQQIjIyUowaNUpUqlRJWFhYCBsbG1G/fn3x66+/StO4dOmS6NevnyhbtqxQKBTC2dlZdOrUSZqGGnK9nkL925dtB3ktw4kTJwQAceLECZ3LRm+n3PsQU1NT4erqKtq2bSuWLl0qvYIstwcPHoiBAwcKV1dXYWJiIkqXLi06deoktm/frlHuZXGc+1Vq+sSAENr7ICGECA8PF4MGDRKOjo7C1NRUVK9eXesVTOr56XpVm66YeZG8XqVmYWGR52/02d8I8fJ9BdHbYMWKFQKAqFevnta4Bw8eiJ49ewpbW1uhVCpFvXr1xN69e7XK7d69W1SpUkV6lZI6pvN6lZq+sf3zzz+LSpUqCYVCIapVqyb27NkjevToISpVqvRKy0z0ttOVWyiVSlGrVi2xcuVKjdeU/fDDD6J8+fJCoVCISpUqiXXr1okZM2ZovfYsKSlJjBo1Stjb2wtLS0vh7+8v7t69KwCIL7/8UqPssWPHRO3atYWpqanw8fERa9euFRMmTBBKpVKjXF6vUtPn/FXf8/qSQCYEn8Z/24wfPx6rV69GYmJinp0slESxsbGws7PD3Llz8dlnnxm6OkRERPQKatWqBScnJxw5csTQVSEq9q5cuYLatWtj06ZNL+0s0t/fHzdv3pTeaPI6lNTzej5zXsSlpKRofI+KisLGjRvRpEmTEp2Y514vALBkyRIAQIsWLd5sZYiIiKjAMjIypI6t1E6ePImrV6/ymE70GuR1Hm1kZIRmzZq9sOz9+/exf//+Qo1Nntf/h8+cF3ENGzZEixYtULlyZYSHh+OHH35AfHw8pk2bZuiqGdQvv/yC9evXo0OHDrC0tMTp06exdetWtGvXDo0bNzZ09YiIiEhPT548QZs2bfDee+/Bzc0Nd+7cwapVq+Dq6ooPPvjA0NUjKnYWLlyIixcvomXLljA2NsaBAwdw4MABDB8+XOttQd7e3ggMDIS3tzdCQkKwcuVKmJqaYvLkyYVWH57X/4fJeRHXoUMHbN++HWvWrIFMJsM777yDH374QeuqVklTo0YNGBsbY+HChYiPj5c6k5g7d66hq0ZERET5YGdnB19fX6xduxYRERGwsLBAx44d8eWXX8LBwcHQ1SMqdho1aoQjR45gzpw5SExMRNmyZTFz5kydzcffffddbN26FWFhYVAoFGjYsCHmzZun8Rq2V8Xz+v/wmXMiIiIiIiIiA+Mz50REREREREQGxuSciIiIiIiIyMD4zPkrUKlUePr0KaysrCCTyQxdHSIAgBACCQkJcHNzg5ERr78VFOObiiLGd+FgfFNRxPguHIxvKor0jW8m56/g6dOnWj0aEhUVjx8/hru7u6Gr8dZifFNRxvh+NYxvKsoY36+G8U1F2cvim8n5K7CysgIA+O18DyYWpgauzdsvbLW3oatQLGRlpOLy73Ol7ZMKRr3+Qi55wtqSdzBeVbcK1Q1dhWIhExk4jf2M71fE+C5cjO/CwfguHIzvwsX4Lhz6xjeT81egbipjYmHK5LwQGJsoDV2FYoVNuV6Nev1ZWxrB2ooH91dlLDMxdBWKh/+/X4Xx/WoY34WL8V1IGN+FgvFduBjfhUTP+OYWS0RERERERGRgTM6JiIiIiIiIDIzJOREREREREZGBMTknIiIiIiIiMjAm50REREREREQGxuSciIiIiIiIyMCYnBMREREREREZGJNzIiIiIiIiIgNjck5ERERERERkYEzOiYiIiIiIiAyMyTkRERERERGRgTE5JyIiIiIiIjIwJudEREREREREBsbknIiIiIiIiMjAmJwTERERERERGRiTcyIiIiIiIiIDY3JOREREREREZGBMzomIiIiIiIgMjMk5ERERERERkYExOSciIiIiIiIyMCbnRERERERERAbG5JyIiIiIiIjIwJicExERERERERkYk3MiIiIiIiIiA2NyTkRERERERGRgTM6JiIiIiIiIDIzJOREREREREZGBMTknIiIiIiIiMjAm50REREREREQGxuSciIiIiIiIyMCYnBMREREREREZGJNzIiIiIiIiIgNjck5ERERERERkYMaGrgDlX/rOFKRvTYGIVsHIxxjK8RaQVzHRWTZjfypS5ydqDjQFrI45Sl8Tmkbq/K3iQ3OY9jcvtHoXRd2b3ED/Vldhb5WCoKcOWLyjMW7/6/zS37WuHYTZAcfwx3VPTP3BTxo++N0LaFP7AZxtE5GRZYS7j52wZn9d3ApxeZ2LQcXInnWO2L7SGdERxvCukoKRc5+gUu3kPMvv/N4J+zY44PlTU1jbZaJpp1gMnvoMpkoBAEhONMKGhaVw5oANYqOM4VM1BR/OCUXFWilvapEMpnNgJHp++Bz2Tpl4eMsM331eGnev6N6neVRIxcBJYShXIxmuZTKwarobflvrpFGmWv1E9BoZgfLVk+HgmomZgz1x9qDNm1gUKibyE9+TepTDtbOWWsPrtY7DnI3BAIDT+22w7ycH3L9ujoQYY3x3+C58qhX/2AbyF9/t+0ehTa8YeFRMBQAEXTfDuvmlNMrbOmZgyGfP4Ns8ARY2WbhxzhIrPi+Np8GKN7I89PYrzPjOzADWLyiFf45b41mIKSysVajdNAFDPn0KB9fM170oBlfY8a00z8KQz56hoV88rO0yEfbYFLt/cMS+jY46p2lIvHP+lsk4loa05UlQBJrDfK0t5OXkSJ4QD1WMKu8fWchgscv+v882e83ROcftsodyiiUgA4xbFO8DUuvaQRjjfxY/HvTF4K96IOiJPb75YB9sLV98YuNqn4DRXc/hygNXrXGPn9vgmx2NMXBhL4z8tivCoq2w+IP9sLUoGSdL9GpO7rbFmlluGPBxGFYcugvvKin4rL83YiN1X0c9vtMWP84rhQEfh+H7U3fw8dePcWqPHdZ9WUoqs3hCGVz6wxKTl4Vg1bE78G2egCl9yiHyme4LesVF8y4xGD7jKTZ/44pRfhXw8JYSX2x5CBuHDJ3lFWYqPPvXFD/OK4WocN3rW2muwsObSiz/1P11Vp2KqfzG97S1wdh65Yb0WX3iDozkAk07xUllUpONULVeEoZ8+vRNLUaRkN/4rtEoESd22WJyLx981KUcIp6aYN7WB3BwVZcXmPHjI5TySMfMQV4Y1a4CwkNN8OUvD6Awy3pzC0ZvrcKO77QUIwRdN0f/8eFYcegepq8NRugDBWYEer/JxTKIwo9vYMTMp6jTIgELx5TFsOaV8Nv3Thj1xRM0aBenc5qGxOQcwIoVK+Dp6QmlUon69evj/Pnzhq5SntJ/SYFJZyVMOioh9zKGYqIlZEoZMval5v0jGWDkYPTfx17z364xzsEImafTIa9tAiM3+WteGsPq0+I6fj9bGfvPV8KjcDss2tYMaenG6FT/Tp6/MZKpMOO9Y/jhQB08jbLWGn/kUnlcuOeOp1HWCA6zx7e7GsLSLB0+blGvc1HoBd6m+N65xgnv9o+CX99oeFRIw9gFoVCYqXBoq73O8rcuWKBq3SS06h4L1zLp8G2RgBb+Mbh7OftqcVqKDKf322Lo589QvUESSnul4/2JYXDzTMPenxze5KK9cd2HR+LgFnsc/sUe/95X4ttP3JGWIoNfv2id5e9dNcfaOW44tdsOGekynWUunLDOboXAu+VFRnGOb2u7LNg7Z0qfS39YQWmmQrPOsVKZNj1j8N7H4ajdLFHnNIqr/Mb3gtEe2LvBEQ9vmuFxkBKLJ5SBzAio3SQBAFDaOx1V6iRj2RR33LtqjtAHSiyb4g6FUqBlt9g3uGSUU0mObwtrFb785QGad4lFmXJpqOybjFFfhOL+NXM8Dy3eF9cLO74BoEqdZBzZZo9rZy0RHmqKA5sd8PCWGSrWyrtloqGU+OT8l19+wccff4wZM2bg0qVLqFmzJvz8/PD8+XNDV02LyBBQ3cuE3Pe/oJQZySCvYwLVzRc0cUkRSOwZjcQe0UiZGo+s4LzLqqJVyDybDpNOxfuuubE8CxXdI/DPvdLSMCFkuHDPHdU8w/P83SC/i4hJNMPevyvpNY+ujW4jIcUUQU+LdyJUVL1N8Z2RLsP9a+Z4p+l/J9lGRkDtpom4ddFC52+q1EnC/WvmuPP/ZPxZiCn+OWaNuq3jAQBZWTKosmQwVWi2rFEoVbh5Xrs5XXFhbKJC+RrJuPSnlTRMCBku/2mFKr5F70BMBVPc4zu3Q1vt0bxrDJTmL2gpVwIURnwrzFQwNhZIiM2+q2limr1O09P+uzAnhAwZ6TJUrZtUiLUnfTG+tSXFyyGTCVjYFN/WHK8jvgHg1gVzNGgX9/+76QI1GyWitHcaLp6yyntCBlLik/NvvvkGw4YNw6BBg1ClShWsWrUK5ubm+PHHHw1dNS0iTgVkQevOt8zOCKoo3cFsVFYO5RRLmM23hvJzKwiVQPKHcVA91x3YGQdSAXMZjJsV7+Tc1iIVxnKB6AQzjeHRCWawt9bdBL2G1zN0anAXC35p9sJpN6oSgiMLfsCJRWvRp/k1jP+uI+KSzF74G3o93qb4jo+WQ5Ulg62TZrMtO8cMxETobhbXqnssBk58hgn+5dChbE0ENqyCGo0S0W9s9smLuaUKlX2TsGWJK6LCjJGVBRzbYYfbFy0QnUfT7eLA2j4LcmMgNtd6i4k0hp1T8X9Wr6Qo7vGd053L5nh0xwzv9td956gkKYz4HvLZM0SFm+DSn9kXKR8HKREeaoLBU5/B0iYTxiYq9B71HE5uGbB30d2Ull4vxrem9FQZfvjCDS38Y2BhVXwv0L2O+AaA7z4vjX/vKbHl0i3sC7mGuZsfYsWnpXHj76J3o6JEJ+fp6em4ePEi2rRpIw0zMjJCmzZtcPbsWa3yaWlpiI+P1/gUdfJqJjB5Vwl5eWMY1zaB2RfWkNnKkLFHdzP4zP1pMGmrgEyhu1lnSWWuSMe0905gwS/NXppoXwpyQ+CinvhgqT/O3SmDOYFHX/ocOxW+khDfV89Y4udlLhg9LxQrDt3F9B+Ccf6oNTYv/q8DwsnLQiAE0P+daujkWRO7fnBEC/8YyEr03p/ediUhvnM6tNUeXpVTXtg5JOmn9+hwtOgai9lDPJGRlr0jzMqUYfYQT5T2ScOO2zex58F11GyUiPPHrCBUPB960xjfmjIzgC9GeAICGPNl6Jut3FtGV3wDQNfBkajkm4zpAZ4Y/W4FfD/bDaPmPUHtpgkvmJphFN9bJ3qIjIxEVlYWXFw0e9J2cXHBnTvazx3Pnz8fs2bNelPV0yKzMQLk2U3Pcz4NLmJUMHLQ70xbZiyDvLwxVKHad84zr2ZA9W8WlLOKXhOPwhabpERmlgz2VppJs71VCqLjtZPv0o7xcHNIwIKhB6VhRrLs3rBPfb0G/ef1wZOo7OdQU9NN8CTSBk8ibXAzxAU/f7YVnRvcwcajtV/jElFub1t8W9tnwUguEBuh+SxZTKRJnleLNyx0ReseMWg/IPtqu1flVKQmG2HppDLoNy4cRkaAm2c6vtoZhNRkIyQlGMHBJRNfjPBAKY+0175MhhIfLUdWJmCba73ZOWbqdReDir6SEN9qqclGOLnbDgMnPXudVXxrvEp89/zgOfqMeo4pfXwQfFvzWB903Rwj21aEuVUWTEwE4qKNsXTvfdy7xpZvbxrj+z/qxDz8iSkW/hpUrO+aA68nvk2VKgROCcPsIZ44fyy7v6jg22bwrpqCnh9E4PKfRSvv4b2TfJg6dSri4uKkz+PHj9/o/GUmMhhVMEbWxf+azQiVQNbFDBhV1e+EU2QJqB5mQaYjmc/YmwqjisaQlyv+J6+ZWXLcDXVCnfJPpGEymYBvhSe48Uj7tWch4bZ478teCFzUU/qcvukp3SUPj827WYyRDDAxLr7PBxUXho5vE1OB8jWScfn0f9uSSgVcOW2JKr66n3lMSzGCzEhoDDP6/3ehORhKcxUcXDKRECvHxVPWaOj3dt1ZyI/MDCPcv2au0RmMTCZQq0kibl0s3q+HJN3exvhW++N3W2Sky9C6e8zrruZboaDx3Wvkc/QfH47PBnjj/rW8yyUnyBEXbQw3rzSUr5mMs4fYAWRRV1zjW52YPwlW4MtfgmBtX/zPJV9HfBsbC5iYCqhyXddQZUHrHKooKP5Z2As4OjpCLpcjPFyzA7Dw8HC4umq/JkuhUEChMOyz2KZ9zJA6LwHySsYwqmyMjG2pECkCJh2UAICUuQkwcjSC4oPsDijS1iVDXtUYRu5yiASB9K3JUIVlQdlJ8yqRSFIh82QaFKP067iiOPjlZHV81v8k7jx2wq1/ndG7+XUoTTOw7++KAIDPBxxHZJwFVu2tj/RMYwSHafa4mZhiCgDScKVpBgLaXsLpG56IjDeHrUUquje9CUebJJy4UvxffVHUvI3x3X14BL4aXxYVaiajYu1k/Pa9E1KTjdCub/ad8YVjy8LRNQODP82+wt6gbTx2rnFCuWopqPROMp4Em2LDolKo3zYO8v83r7lw0gpCAGV80vAk2BRr55RGmXKpaNeneL9BYOcaR0xc8hj3rprj7mVzdBsWAaW5Cod/zo7XSUv/RWSYCdbNz37tnLGJCmUrZLcmMDERcCiVAe+qKUhNMsLTR9nbhdI8C25e6dI8XMukw7tqChJi5Yh4YvqGl7BkKwnxrXZwqz0a+cXpPDGPj8ne9tSv/3v8IHsZ7ZwzYO9cfPtXyG989x71HO9PDMOCUWUR/tgUdv9/NjglyQipydk7y6adYhEXZYznT0zgVTkVH8x+grMHbXCpCHYYVdwxvrMT8znDvBB03Qyzf3oIVZYM0c+z49zKNgsmpkUvqSwshR3fyYlyXD1jgWHTniE91QjhoSao0TAJbXrGYM0sN4MtZ15KdHJuamoKX19fHDt2DP7+/gAAlUqFY8eOYfTo0YatXB5MWisgYlVI+yEZIloFo3LGMP/KWuokToRnIefjUSJBhdSFiRDRKsissu+8m6+0hdxL81+fcSwdEIBJm+LdEVxOxy6Xg61FKoa2vwB762Tcf+KICas7ICYx+4qbi10ihND/WTOVSgYP51i0H3QYNpapiE9S4va/Thj5bRetxJ5ev7cxvlt0zT45/GlRKcREGMO7agq+2PxQahYX8cQURjkavfQfHwaZTGD9wlKICjOBjX0mGrSNQ+CUMKlMUrwc6+aXQuQzE1jZZqFxh1gMmvIMxsX7TSw4tccONg5ZGDgpDHZOmXh40wyfDfBCbGT2gjuVTte4iu7gkomVR+5J33t9GIFeH0bg6hkLTO5ZDgBQoWYKFu14IJX5YFb2u6UP/2KHrz8q+waWitRKQnwDwOMgBW6et8S8rUE6p3nusI3Gtjf/Q08AwHsfh+H9iWE6f1Mc5De+Ow6MhKlCYNraEI3pbPzaBZu+zk727F0yMGLmU9g6ZiL6uTGObrPDliXaLeno9WN8A5Fhpjh3OLvVxsi2mm8IWrg9CDUbFd/XJ76O+J7/oQcGf/oMnywPgZVtFp4/McX6BaWK5GtlZULkbvxYsvzyyy8ICAjA6tWrUa9ePSxZsgS//vor7ty5o/WsS27x8fGwsbFBp0ODYWLBuyav6tmycoauQrGQmZGKCzs/R1xcHKyttd/FXpIURnzH3POGtRWfAHpVfm61DF2FYiFTZOAkdjO+wfguShjfhYPx/R/Gd9HB+C4c+sZ3ib5zDgB9+vRBREQEpk+fjrCwMNSqVQsHDx58aeATUdHH+CYqvhjfRMUX45tKqhKfnAPA6NGji2wzGSJ6NYxvouKL8U1UfDG+qSRiWw8iIiIiIiIiA2NyTkRERERERGRgTM6JiIiIiIiIDIzJOREREREREZGBMTknIiIiIiIiMjAm50REREREREQGxuSciIiIiIiIyMCYnBMREREREREZGJNzIiIiIiIiIgNjck5ERERERERkYEzOiYiIiIiIiAyMyTkRERERERGRgTE5JyIiIiIiIjIwJudEREREREREBsbknIiIiIiIiMjAmJwTERERERERGRiTcyIiIiIiIiIDY3JOREREREREZGBMzomIiIiIiIgMjMk5ERERERERkYExOSciIiIiIiIyMCbnRERERERERAbG5JyIiIiIiIjIwJicExERERERERkYk3MiIiIiIiIiA2NyTkRERERERGRgTM6JiIiIiIiIDIzJOREREREREZGBMTknIiIiIiIiMjAm50REREREREQGxuSciIiIiIiIyMCYnBMREREREREZGJNzIiIiIiIiIgNjck5ERERERERkYMaGrkBxkOgXBWOZiaGr8db76+lRQ1ehWIhPUMFup6FrUXx8+LghTC1NDV2Nt16Kf2VDV6FYyMxIBfbuNnQ1io2ejVrC2Ijx/aqWhewydBWKhcQEFepWNXQtig/Gd+FgfBcOfeObd86JiIiIiIiIDIzJOREREREREZGBMTknIiIiIiIiMjAm50REREREREQGxuSciIiIiIiIyMCYnBMREREREREZGJNzIiIiIiIiIgNjck5ERERERERkYEzOiYiIiIiIiAyMyTkRERERERGRgTE5JyIiIiIiIjIwJudEREREREREBsbknIiIiIiIiMjAmJwTERERERERGRiTcyIiIiIiIiIDY3JOREREREREZGBMzomIiIiIiIgMjMk5ERERERERkYExOSciIiIiIiIyMCbnRERERERERAbG5JyIiIiIiIjIwJicExERERERERkYk3MiIiIiIiIiA2NyTkRERERERGRgTM6JiIiIiIiIDIzJOREREREREZGBMTknIiIiIiIiMjAm50REREREREQGxuSciIiIiIiIyMCYnBMREREREREZGJNzIiIiIiIiIgNjck5ERERERERkYEzOiYiIiIiIiAyMyTkRERERERGRgTE5JyIiIiIiIjIwY0NXgPKvc2Aken74HPZOmXh4ywzffV4ad6+Y6yzbuH0s+o59DjfPNBibAE+CTbFjlROO7bCXykxY/C/a9YnR+N2FE1b4bID3a12OomDPOkdsX+mM6AhjeFdJwci5T1CpdnKe5RPj5Fj/pSv+OmCLhFg5nN3T8cGsJ6jXOgEAsPErV2z6xlXjN+4+qfjhzzuvdTmoeEvano7ETenIihYwKWcEmwlKmFaV6ywb+WEy0i9naQ1XNJLD4Rvd+4mSoluzm+jb9hrsrVPwINQeS39thNshzjrLNqsVjPf8rqC0UzyM5SqEPrfGL8dq4PD58m+41lScdOrzGD0CQmDnmI7ge5ZY+WVF3Ltho7Nso9bP0WdIMEqVSYGxiQpPQszx20YPHN9bCgAgN1Zh4OgHqNskEq7uKUhKMMaVv+2xbml5REco3uRiGcQfG1xxbE1pxEeYonTlJPSc9RCetRLzLJ8cJ8feRR64etAByXHGsCudhh7Tg1G1Vfb5jyoL2L+4LP75zQkJESawcUlH/Z7P4Tc2FDLZm1oqKm4KM+ZLsvzE+9I+1RB0TnsdV2kZjQ/X337dVX1lJTo5/+OPP7Bo0SJcvHgRz549w2+//QZ/f39DV+uFmneJwfAZT7FsijvuXDJHt2ER+GLLQwxpWhFxUSZa5RNijbF1qQseBymQmSFD/TbxmLD4MWIjjXHxlLVU7p/jVvj6ozLS94z04n8kOrnbFmtmuWHMl6Go9E4SfvveCZ/198YPf96BrWOmVvmMdBmm9vWBrWMGPl/zCA6lMvA81AQW1pqJkEfFFHz5ywPpu1wuXvuykLa3Mb51STmSgbilabD9RAmTqkZI+jkDUeOT4fyLBeT22o2f7L80g8j8b5tTxQlEvJ8Ms1ba+4eSpJXvA4zqcQ5fb22CW4+c0avVDXw15gAGzOyN2EQzrfLxSQpsPFgL/4bbIiNTjkbV/8WU908hJkGJf26X0TEHepPexvhu5heGYRPvYfncyrhz3Rr+Ax5jzsrLGN61EeKiTbXKJ8QZ4+e1XggNtkBGhgz1m0Xio1m3EBttiktnHKBQqlCuUgK2rvHGw7uWsLTOxAef3MWMpVcwrn99Ayzhm3Pxd0f8NtcLfb54AI9aCTj5oxu+e78qpp24BCvHDK3ymekyrHivKiwdMjBk5R3YuKYj+okCZtb/HeuPrHTH6U2ueO/r+yhVIRn/XrPE5knlobTOQotBz97k4pV4b2N861LYMV9S5Tfeh66+g6wceUxSrDG+fLc2aneMepPVLrAS3aw9KSkJNWvWxIoVKwxdFb11Hx6Jg1vscfgXe/x7X4lvP3FHWooMfv2idZa/dtYSZw7a4HGQEs9CFNj1gxMe3jZD1XpJGuUy0mWIiTCRPolxxf+6zc41Tni3fxT8+kbDo0Iaxi4IhcJMhUNb7XWWP/SzPRJi5ZjxYzCq1kuCa5l01GiYBJ+qqRrl5HLA3jlT+tg4aN/FpNfvbYxvXRK3psO8qwnMO5nAxEsOm08UkCllSN6rfUACACMbGeQORtIn7XwWZApA2br4x/SL9G51HXv/qoQD5yoiJMwOX29tgtR0Y3RsdFdn+Sv33fDnVS+EhNnhaaQ1tp+ohodP7FHDJ/wN15x0eRvju9v7/+LgztI4stsNjx9aYvncSkhLlaOd/1Od5a9fsMfZ4854HGyBsFBz7N5SFsH3LVG1diwAIDnRGJ998A7+POyCJyEWuHvdBt/Nr4jyVRPg5Jqqc5rFxYm1bmjYNxwNej9HqQop6DPvAUzNsnD2V90tYc796oLkWGMM//4OvOsmwKFMGso3iId7lf9aygVftEL1ttGo1joGDmXSULtjFCo1jUHIFcs3tVj0f29jfOtS2DFfUuU33i1sM2HtnCF97vxpC1OzLNTuGPmGa14wJfpsrX379mjfvr2hq6E3YxMVytdIxs/L/9sYhZDh8p9WqOKbd1Ps/wjUapKIMj5p+PELzSYyNRom4pdrN5EQJ8fV05ZYv9AVCTHFd/PISJfh/jVz9B39XBpmZATUbpqIWxctdP7m3GEbVPZNwvJP3XH2kA1sHDLRslsMeo96DnmOFsZPgk3Rr3ZVmCpUqOybhMFTn8HZXXciRa/P2xbfuogMgYy7KlgG/HeFXWYkg6KuHBnXVXpNI/n3DJi1NYGRWfFvDZMXY3kWKpSNxKbDtaRhQshw8U5pVPV6nvcP/yuNdyo+RRmXOKza5fry4vTavW3xbWysQrnKCfj1B09pmBAyXDlnj0o1YvWYgkDNejFw90zCuiXl8ixlYZkJlQpITCi+x+/MdBkeX7dE25Gh0jAjI6Bikzg8umSl8zfXj9jB850E/DrNG9eP2MPSPgO+XSPR9sNQGP3/+O3lm4AzW13w/KESzt6pCL1ljocXrNHt80dvYKkop7ctvnV5UzFf3BUk3nM7+4sL3ukcCYW5fudNhlZ8996vQVpaGtLS0qTv8fHxb3T+1vZZkBsDsRGa/7aYSGOUKZeWx68Ac6ssbLl0CyamKqiyZFj2aWlc+uO/DfrCSSv8dcAGYf+aopRnOgZNeYYvNj3E+M7loVIVzxP6+Gg5VFky2DppJs12jhl4HKT7Wb1nIaa48pclWnWLwdxND/EkWIHln7ojK0OG9yZk302r9E4SJi5JgbtPGqKfm2DT166Y0K08Vp+4A3PLt2OnUFIZOr51UcUKIAtazdeN7GRIf/TyFhnpN7OQ+UAF20+Vr6uKbwUby1QYywVi4jWbr0cnmKGsS2yev7NQpmPHvM0wNclClsoIi39ujAt33F9zbel1MHR8W9tlQG4sEBOl2ZQ1NsoUZbyS8vgVYG6ZiY1H/oSJiQoqlQwr5lXE5XO6m7eamGZh0PggnDrgipSk4nt6lxRjAlWWDNa5mrNaOaYj/IHuZ3kjHysRfVaJOl0j8MH6W4h4ZIZfP/dGVqYMHcY/BgC0HRmK1EQ55rZ6BzK5gMiSodOkENTtFvHal4lejaHjW5c3EfMlQUHiPadHVyzx7K4F+i8Mel1VLHTFd+/9GsyfPx+zZs0ydDXyLSXRCCPbVoDSQoXaTRIwYsZThIUocO1sdlOtU7vtpLKP7pgh+JYSG87dQY1GibhyWr+rUiWBEICtQybGLXoMuRwoXyMFUWEm2L7SWUrO67ZKkMp7V0lFpdrJeL9eFfyxxxbv9tf96AEVDW9rfL9I8u8ZMPYxyrPzOHqx5DQTDJnfHWaKTPhWfIJRPc7haaQVrtx3M3TVKJ/e1vhOSZJjdO/6MDPPQs360Rg24T7CQs1w/YLm41dyYxWmLroOmQxY/kUlA9W26BIqGawcMtDvyyAYyYGy1ZMQF2aKY6tLS8n55b2OuLDLCQHf3kOpCskIvWWBHbO8/t8xHBP0ouxtjW9d9I150s+5X1zgVinphZ1FFjUl+pnz/Jo6dSri4uKkz+PHj9/o/OOj5cjKBGydNDsrs3PMRExE3tdZhJDh6SMFHt40w47Vzvhzny36jMn7ucmwfxWIjZLDzTO90Ope1FjbZ8FILhAbodlJVkykCeyctDuDA7KfIy/tnabRhL1s+VREPzfJswM9S5ssuHun4emj4t9z7tvO0PGti5GtDJADWdGarS5UMQJyhxfvvlUpAilHMmDepWR3BAcAcYlKZGbJYGedojHc3ioF0fF592AvhAxPImwQFOqAX47VwKnLXnjP78prri29DoaO7/gYE2RlymDnoHlctXVIR3SkdsdQakLI8OyxOR7etcJvP3ngr6PO6D3kkUYZdWLuXCoVn42oXazvmgOAhV0GjOQC8ZGa+7aESFNYO+k+b7FxToeTV4rUhB0AXMolIz7CFJn/P37vmueJth+GwrdLJNwqJaNe9wi0HPIUh79ja5miztDxrcvrjPmSpCDxrpaWbISLvzuiYZ+3q68YJuf5oFAoYG1trfF5kzIzjHD/mjlqN/nv7qxMlv0c+a2L+r8iychIwMQ07x7EHUulw9ouC9HPi+8B3sRUoHyNZFw+/V9HLyoVcOW0Jar46m5uVKVuEp49UkCVI08KfaiAvUtGnuszJckIT0NMYe/MZ86LOkPHty4yExlMKhoh/Z//mrALlUDaP1kwqf7i3XfqsUyIDMD8XSbnmVly3PvXEb4Vn0jDZLLs58hvBuvuUEYXmUzAxJiPp7yNDB3fmZlGCLpthZr1/2tBJZMJ1KofjTvXbPWejsxIwMTkv21QnZi7lU3GpyPeQUJc3if9xYWxqUCZ6om499d/TVpVKuDeXzbwfCdB52+86sQjMkSpcfyOCDaDtXM6jP9//E5PMYIs127VSC4giunjfcWJoeNbl9cV8yVNQeJd7fI+R2SmG711j6YwOX/L7FzjiPb9o9GmVzTKlEvFmC9DoTRX4fDP2c1dJi39F4Om/vfKjz6jw/FOswS4lk1DmXKp6DHiOVr3iMHxndlN2ZXmWRg67SkqvZMEF/d01GqSgJnrHuFpsCkunizeTdq7D4/AgS0OOPKrHf69r8CyKe5ITTZCu77ZO9KFY8vix3n/dZzXaWAkEmLlWDmtNEIfKPD3UWv8/K0LOgf+1/vjmlluuHbWAmGPTXHzH3PMGuwFuRHQoluM1vyJ9GHZzxRJezKQvC8DGcFZiFuYBpEqYN4xO+mOmZWC+O+0+5xI/j0DymbGMLLhiSUA/Hq8Ojo1vot369+Dh2sMJvQ9DTNFBvafrQAA+DTgBIZ3PS+VH+B3BXUqhaKUQzw8XGPQp/U1+NW/j8PnS27HPPRqfttYFu92f4rWnZ+ijFcSRn1+BwqzLBzZlX2cmTD3BgLH/vdcZO/BwajdIAqupZNRxisJ3QaGoFXHMJzY9997zj/96hrKV4nHoqnVIDcSsHNIg51DGoyL+UWklkOf4szPrvh7uxPC7pvh1898kJYsR4Ne2R08/vRReexZ4CGVb/peGJJjjbFjpheeP1TixjE7HF7hjmYD/ztfqtYmGoeXu+PGMTtEPVbg6kF7nFhbGjX93o7XL1HRU9gxX1LlN97Vzv7ijBrtomBhp7tFbFFVfG+N6iExMRFBQf8FRXBwMK5cuQJ7e3uULVvWgDXL26k9drBxyMLASWGwc8rEw5tm+GyAF2L/39zDqXS6xpVhpbkKo+eFwrFUBtJTjfD4gQILx5TFqT3ZyblKJYNX5RS07RUDC+ssRIUb49IpK2xY6IqM9OJ97aZF11jERRnjp0WlEBNhDO+qKfhi80OpWXvEE1MY5VgFzqUz8MWWB1g9szQ+aFMRjq4Z8B8agd6j/uvtOfKZCeaP9ERCjBw2DpmoWjcJS/begy1fp/bGvY3xrYtZWxOoYgUSvk9DVpSASXkjOCw2l5q1Z4UJQKZ5Ip4ZokL61SzYL9V+f3dJdfyiD2wtUzG400XYWycjKNQBE5e3R0xCdqsjF7skjTtkZqYZ+LjvX3CyTUJahjH+DbfB3PUtcfyij6EWgXJ4G+P7j0OusLbLwPsjH8LOMQ0P71ph+sjaiI3OfuzJyTVVoxNWpVkWRn56B44uaUhPM8LjYAt89VlV/HEo+40BDs5paNgy++Lwim1/a8zrkyHvFOtnVH07RyIxyhj7vimLhAhTlK6ShJE/3YT1/zt5jXmqgMzovxZtdm7pGPnTLeyc44X577rC1iUNzQc9Q9sP/+sButesYOz7uix+neaNxEgT2Liko3H/MLw7zvBNpEuatzG+dSnsmC+p8hvvABD+wAwP/7HBqE03DFHlVyITQuTdvrmYO3nyJFq2bKk1PCAgAOvXr3/p7+Pj42FjY4MW6ApjGZuOvqpDT68YugrFQnyCCnYVHiIuLq5INO0ylMKK777HBsDUsvg3FX3dHnxZ2dBVKBYyM1Lx997pjO9Ciu/WjkNgbMT4flVLLuwydBWKhcQEFepWDWd8M76LFMZ34dA3vkv0nfMWLVqgBF+bICrWGN9ExRfjm6j4YnxTSVa82y0TERERERERvQWYnBMREREREREZGJNzIiIiIiIiIgNjck5ERERERERkYEzOiYiIiIiIiAyMyTkRERERERGRgTE5JyIiIiIiIjIwJudEREREREREBsbknIiIiIiIiMjAmJwTERERERERGRiTcyIiIiIiIiIDY3JOREREREREZGBMzomIiIiIiIgMjMk5ERERERERkYExOSciIiIiIiIyMCbnRERERERERAbG5JyIiIiIiIjIwJicExERERERERkYk3MiIiIiIiIiA2NyTkRERERERGRgTM6JiIiIiIiIDIzJOREREREREZGBMTknIiIiIiIiMjAm50REREREREQGZqxPoT/++KNAE2/WrFmBfkdERERERERUkuiVnLdo0QIymUzviQohIJPJkJWVVeCKEREREREREZUUeiXnJ06ceN31ICIiIiIiIiqx9ErOmzdv/rrrQURERERERFRivXKHcM+ePcPVq1eRlJRUGPUhIiIiIiIiKnEKnJzv3r0blSpVgru7O9555x38/fffAIDIyEjUrl0bu3btKqw6EhERERERERVrBUrOf//9d3Tv3h2Ojo6YMWMGhBDSOEdHR5QuXRrr1q0rtEoSERERERERFWcFSs5nz56NZs2a4fTp0xg1apTW+IYNG+Ly5cuvXDkiIiIiIiKikqBAyfmNGzfQu3fvPMe7uLjg+fPnBa4UERERERERUUlSoOTc3Nz8hR3APXz4EA4ODgWuFBEREREREVFJoter1HJr2bIlNmzYgPHjx2uNCwsLw/fff49OnTq9at2KPPWz9pnIAMRLCtNLxSeoDF2FYiE+MXs95uwLgvJPvf4ykjIMXJPiITMj1dBVKBbU65Hx/Wqk47cq3cA1KR4SefwuFIk8fhcKxnfhYnwXDn3jWyYKsAe4e/cuGjRoAE9PT/Tq1QvTpk3DxIkTYWJigtWrV0MIgQsXLsDT07NAlX9bhIaGokyZMoauBpFOjx8/hru7u6Gr8dZifFNRxvh+NYxvKsoY36+G8U1F2cviu0DJOQDcvHkT48aNw4kTJzSuALRo0QIrVqxA5cqVCzLZt4pKpcLTp09hZWUFmUxm6OrkKT4+HmXKlMHjx49hbW1t6Oq8td6W9SiEQEJCAtzc3GBkVOC3JZZ4jO+S5W1Zj4zvwsH4LlnelvXI+C4cjO+S5W1Zj/rGd4GTc7WYmBgEBQVBpVLB29sbTk5OrzI5eg3i4+NhY2ODuLi4Ir3RFnVcj1QUcbssHFyPVBRxuywcXI9UFHG7LBzFbT0W6JnznOzs7FC3bt3CqAsRERERERFRiVTgNjMRERGYOHEiqlSpAnNzc5ibm6NKlSqYOHEiwsPDC7OORERERERERMVagZLzmzdvonr16vjmm29gY2ODXr16oVevXrCxscE333yDGjVq4MaNG4VdVyoghUKBGTNmQKFQGLoqbzWuRyqKuF0WDq5HKoq4XRYOrkcqirhdFo7ith4L9Mx5ixYtcPPmTezfv1+rSfv58+fRoUMHVK9eHSdOnCi0ihIREREREREVVwW6c37+/HmMGzdO57Pm9erVw7hx4/D333+/cuWIiIiIiIiISoICJefOzs5QKpV5jlcqlXB2di5wpYiIiIiIiIhKkgIl5+PHj8fKlSsRFhamNe7p06dYuXIlxo8f/6p1IyIiIiIiIioR9HqV2jfffKM1zNLSEuXKlUO3bt1Qrlw5AMD9+/exa9culCtXDq/4+nQiIiIiIiKiEkOvO+cTJ07U+ly/fh3JycnYvHkzZs2ahVmzZmHLli1ITk7GtWvXMHHixNddd9LDihUr4OnpCaVSifr16+P8+fOGrtJb548//kDnzp3h5uYGmUyGXbt2GbpKRAAY34WB8U1FFeP71TG+qahifL+64hrfeiXnwcHB+f48fPjwddedXuKXX37Bxx9/jBkzZuDSpUuoWbMm/Pz88Pz5c0NX7a2SlJSEmjVrYsWKFYauCpGE8V04GN9UFDG+Cwfjm4oixnfhKK7xXaBXqdHboX79+qhbty6WL18OAFCpVChTpgzGjBmDKVOmGLh2byeZTIbffvsN/v7+hq4KlXCM78LH+KaigvFd+BjfVFQwvgtfcYrvAnUIR0Vfeno6Ll68iDZt2kjDjIyM0KZNG5w9e9aANSOiV8X4Jiq+GN9ExRfjm15Grw7hdLl27RqWLVuGS5cuIS4uDiqVSmO8TCbDgwcPXrmCVDCRkZHIysqCi4uLxnAXFxfcuXPHQLUiosLA+CYqvhjfRMUX45tepkB3zk+ePIl69eph7969cHNzw8OHD+Ht7Q03NzeEhITA0tISzZo1K+y6EhERERERERVLBUrOp0+fDm9vb9y9exfr1q0DAHz66ac4ffo0zpw5g9DQUPTu3btQK0r54+joCLlcjvDwcI3h4eHhcHV1NVCtiKgwML6Jii/GN1HxxfimlylQcn7p0iUMGTIE1tbWkMvlAICsrCwA2Z0cjBgxAtOmTSu8WlK+mZqawtfXF8eOHZOGqVQqHDt2DA0bNjRgzYjoVTG+iYovxjdR8cX4ppcp0DPnxsbGsLKyAgDY2trCxMREo/t/b29v3Lp1q3BqSAX28ccfIyAgAHXq1EG9evWwZMkSJCUlYdCgQYau2lslMTERQUFB0vfg4GBcuXIF9vb2KFu2rAFrRiUZ47twML6pKGJ8Fw7GNxVFjO/CUWzjWxSAr6+vmDhxovS9Ro0aokePHtL3rl27Ci8vr4JMmgrZsmXLRNmyZYWpqamoV6+eOHfunKGr9NY5ceKEAKD1CQgIMHTVqIRjfL86xjcVVYzvV8f4pqKK8f3qimt8F+g959OnT8ePP/6IR48ewdjYGBs2bMCgQYPg4+MDAHjw4AHmz5+PTz755JUuHBARERERERGVBAVKzjMyMhAfHw97e3vIZDIAwKZNm7Bjxw7I5XJ06tQJgYGBhV1XIiIiIiIiomKpQMk5ERERERERERWeAvXWTkRERERERESFR6/e2lu1apXvCctkMo3XBBARERERERGRbnol5yqVSnq2XF9sLU9ERERERESkHz5zTkRERERERGRgfOaciIiIiIiIyMCYnBMREREREREZGJNzIiIiIiIiIgNjck5ERERERERkYEzOSW+BgYHw9PQ0dDWI6DVav349ZDIZHj16ZOiqEFEuuo7DMpkMM2fONEh9iiOuTyJNJ0+ehEwmw/bt219YjucPhYPJeRGk3rjVH2NjY5QuXRqBgYF48uSJoatXZOReTzk/U6ZMMXT1dJo3bx527dpl6GpQCfHdd99BJpOhfv36b2R+gYGBWvuuMmXKoG/fvrh169YbqcOL3Lp1CzNnzuSJA+VbcHAwRo8ejQoVKsDc3Bzm5uaoUqUKRo0ahWvXrhm6eq/dli1bsGTJEr3Le3p6auwLlEolypcvj0mTJiE6Ovr1VVRP+/fvZwJORd7169fRs2dPeHh4QKlUonTp0mjbti2WLVsmleF5ZfGj13vO8/LkyRP88ccfeP78OXr06AF3d3dkZWUhLi4ONjY2kMvlhVXPEmn27Nnw8vJCamoqzp07h/Xr1+P06dO4ceMGlEqloatXZKjXU07VqlUzUG1ebN68eejZsyf8/f0NXRUqATZv3gxPT0+cP38eQUFBKFeu3Gufp0KhwNq1awEAmZmZePDgAVatWoWDBw/i1q1bcHNze+11yMutW7cwa9YstGjRgq2ASG979+5Fnz59YGxsjAEDBqBmzZowMjLCnTt3sHPnTqxcuRLBwcHw8PAwSP1SUlJgbPxKp3MvtWXLFty4cQPjx4/X+ze1atXChAkTAACpqam4ePEilixZglOnTuH8+fOvqab62b9/P1asWKEzQX8T65PoZc6cOYOWLVuibNmyGDZsGFxdXfH48WOcO3cOS5cuxZgxYwAUrfPK999/H3379oVCoTB0Vd5qBdr7CCEwYcIELF++HJmZmZDJZKhevTrc3d2RmJgIT09PzJ49O187cdLWvn171KlTBwAwdOhQODo6YsGCBdizZw969+5t4NoVHTnXU2FKSkqChYVFoU+X6E0IDg7GmTNnsHPnTowYMQKbN2/GjBkzXvt8jY2N8d5772kMa9CgATp16oR9+/Zh2LBhr70ORIXlwYMH6Nu3Lzw8PHDs2DGUKlVKY/yCBQvw3XffwcjoxQ0RX+fxpKherC9durTGvmDo0KGwtLTEV199hfv376N8+fIGrF3eiur6pJLliy++gI2NDf755x/Y2tpqjHv+/LlhKvUScrmcN2YLQYGatS9atAhLly7FxIkTceTIEQghpHE2Njbo3r07duzYUWiVpGxNmzYFkH2yoJaeno7p06fD19cXNjY2sLCwQNOmTXHixAmN3z569AgymQxfffUV1qxZAx8fHygUCtStWxf//POP1rx27dqFatWqQalUolq1avjtt9901ikpKQkTJkxAmTJloFAoULFiRXz11Vca2wSQ/QzX6NGjsW3bNlSpUgVmZmZo2LAhrl+/DgBYvXo1ypUrB6VSiRYtWhRqs9Pjx4+jadOmsLCwgK2tLbp27Yrbt29rlJk5cyZkMhlu3bqF/v37w87ODk2aNJHGb9q0Cb6+vjAzM4O9vT369u2Lx48fa0zj/v376NGjB1xdXaFUKuHu7o6+ffsiLi5OWgdJSUnYsGGD1NQvMDCw0JaTKKfNmzfDzs4OHTt2RM+ePbF582atMjdv3kSrVq1gZmYGd3d3zJ07FyqVSqvc7t270bFjR7i5uUGhUMDHxwdz5sxBVlaWXnVxdXUFAK27UQ8fPkSvXr1gb28Pc3NzNGjQAPv27dP6/fPnzzFkyBC4uLhAqVSiZs2a2LBhg1a5n3/+Gb6+vrCysoK1tTWqV6+OpUuXAsh+DKZXr14AgJYtW0oxePLkSb2WgUqmhQsXIikpCevWrdNKzIHsbXrs2LEoU6aMNCwwMBCWlpZ48OABOnToACsrKwwYMAAA8Oeff6JXr14oW7YsFAoFypQpg48++ggpKSla09b3OKzrGeknT55g8ODBcHFxgUKhQNWqVfHjjz9qlFE/R/rrr7/iiy++gLu7O5RKJVq3bo2goCCpXIsWLbBv3z6EhIRIcVPQlid57Qv0OU4DwOXLl9G+fXtYW1vD0tISrVu3xrlz5zTKZGRkYNasWShfvjyUSiUcHBzQpEkTHDlyBED2/2fFihXSulN/1HKvT/X5QVBQEAIDA2FrawsbGxsMGjQIycnJGvNOSUnB2LFj4ejoCCsrK3Tp0gVPnjzhc+yUbw8ePEDVqlW1EnMAcHZ2BvDi88qQkBCMHDkSFStWhJmZGRwcHNCrVy+d59exsbH46KOP4OnpCYVCAXd3dwwcOBCRkZF51i8tLQ2dOnWCjY0Nzpw5A0D3M+eenp7o1KkTTp8+jXr16kGpVMLb2xs//fST1jSvXbuG5s2ba5yTrFu3rsQ9x16gO+fff/89Bg4ciHnz5iEqKkprfI0aNXDgwIFXrhxpUm+YdnZ20rD4+HisXbsW/fr1w7Bhw5CQkIAffvgBfn5+OH/+PGrVqqUxjS1btiAhIQEjRoyATCbDwoUL0b17dzx8+BAmJiYAgMOHD6NHjx6oUqUK5s+fj6ioKAwaNAju7u4a0xJCoEuXLjhx4gSGDBmCWrVq4dChQ5g0aRKePHmCxYsXa5T/888/sWfPHowaNQoAMH/+fHTq1AmTJ0/Gd999h5EjRyImJgYLFy7E4MGDcfz4cb3WS1xcnNYOxNHREQBw9OhRtG/fHt7e3pg5cyZSUlKwbNkyNG7cGJcuXdI6wejVqxfKly+PefPmSRcYvvjiC0ybNg29e/fG0KFDERERgWXLlqFZs2a4fPkybG1tkZ6eDj8/P6SlpWHMmDFwdXXFkydPsHfvXsTGxsLGxgYbN27E0KFDUa9ePQwfPhwA4OPjo9cyEuXX5s2b0b17d5iamqJfv35YuXIl/vnnH9StWxcAEBYWhpYtWyIzMxNTpkyBhYUF1qxZAzMzM61prV+/HpaWlvj4449haWmJ48ePY/r06YiPj8eiRYu0yqvjMSsrCw8fPsQnn3wCBwcHdOrUSSoTHh6ORo0aITk5GWPHjoWDgwM2bNiALl26YPv27ejWrRuA7JPdFi1aICgoCKNHj4aXlxe2bduGwMBAxMbGYty4cQCAI0eOoF+/fmjdujUWLFgAALh9+zb++usvjBs3Ds2aNcPYsWPx7bff4tNPP0XlypUBQPpLpMvevXtRrly5fPfbkJmZCT8/PzRp0gRfffUVzM3NAQDbtm1DcnIyPvzwQzg4OOD8+fNYtmwZQkNDsW3bNun3+h6HdQkPD0eDBg2ki+JOTk44cOAAhgwZgvj4eK1WjV9++SWMjIwwceJExMXFYeHChRgwYAD+/vtvAMBnn32GuLg4hIaGSsd1S0vLl9YjIyND2hekpqbi8uXL+Oabb9CsWTONR9H0PU7fvHkTTZs2hbW1NSZPngwTExOsXr0aLVq0wKlTp6T/0cyZMzF//nzpeBsfH48LFy7g0qVLaNu2LUaMGIGnT5/iyJEj2Lhx40uXQ613797w8vLC/PnzcenSJaxduxbOzs7S/gbITvx//fVXvP/++2jQoAFOnTqFjh076j0PIjUPDw+cPXsWN27cyPNRzRedV/7zzz84c+YM+vbtC3d3dzx69AgrV65EixYtcOvWLWmflJiYiKZNm+L27dsYPHgw3nnnHURGRmLPnj0IDQ2VzqdzSklJQdeuXXHhwgUcPXpUOq/IS1BQEHr27IkhQ4YgICAAP/74IwIDA+Hr64uqVasCyL6gqL5wPnXqVFhYWGDt2rUls4m8KACFQiHWrFkjhBAiMjJSyGQycezYMWn8qlWrhFKpLMikSQixbt06AUAcPXpUREREiMePH4vt27cLJycnoVAoxOPHj6WymZmZIi0tTeP3MTExwsXFRQwePFgaFhwcLAAIBwcHER0dLQ3fvXu3ACB+//13aVitWrVEqVKlRGxsrDTs8OHDAoDw8PCQhu3atUsAEHPnztWYf8+ePYVMJhNBQUHSMABCoVCI4OBgadjq1asFAOHq6iri4+Ol4VOnThUANMq+aD3p+uRcFmdnZxEVFSUNu3r1qjAyMhIDBw6Uhs2YMUMAEP369dOYx6NHj4RcLhdffPGFxvDr168LY2Njafjly5cFALFt27YX1tnCwkIEBAS8sAzRq7pw4YIAII4cOSKEEEKlUgl3d3cxbtw4qcz48eMFAPH3339Lw54/fy5sbGy04i85OVlrHiNGjBDm5uYiNTVVGhYQEKAzHkuXLi0uXryo8Xv1/P/8809pWEJCgvDy8hKenp4iKytLCCHEkiVLBACxadMmqVx6erpo2LChsLS0lPYd48aNE9bW1iIzMzPP9bJt2zYBQJw4ceIFa48oW1xcnAAg/P39tcbFxMSIiIgI6ZMzRtRxMGXKFK3f6Yql+fPnC5lMJkJCQqRh+h6Hhcg+vs6YMUP6PmTIEFGqVCkRGRmpUa5v377CxsZGqsOJEycEAFG5cmWN84ilS5cKAOL69evSsI4dO2rN90U8PDx07gsaN26sVS99j9P+/v7C1NRUPHjwQBr29OlTYWVlJZo1ayYNq1mzpujYseML6zdq1CiNc4Wccq9P9flBznMqIYTo1q2bcHBwkL5fvHhRABDjx4/XKBcYGKg1TaKXOXz4sJDL5UIul4uGDRuKyZMni0OHDon09HSNcnmdV+ra15w9e1YAED/99JM0bPr06QKA2Llzp1Z5lUolhPhvX7Ft2zaRkJAgmjdvLhwdHcXly5c1yqvPy3OeP6j3BX/88Yc07Pnz50KhUIgJEyZIw8aMGSNkMpnGNKOiooS9vb1eOUFxUqBm7c7OzlpNenO6ePEiypYtW5BJUw5t2rSBk5MTypQpg549e8LCwgJ79uzRuHIul8thamoKAFCpVIiOjkZmZibq1KmDS5cuaU2zT58+Gnfe1U3lHz58CAB49uwZrly5goCAANjY2Ejl2rZtiypVqmhMa//+/ZDL5Rg7dqzG8AkTJkAIodV6onXr1hp3qtVXuXv06AErKyut4eo6vcyKFStw5MgRjU/OZQkMDIS9vb1UvkaNGmjbti3279+vNa0PPvhA4/vOnTuhUqnQu3dvREZGSh9XV1eUL19eenxAva4OHTqk1cyN6E3bvHkzXFxc0LJlSwDZTd/69OmDn3/+WWqKvn//fjRo0AD16tWTfufk5CQ1v80p5930hIQEREZGomnTpkhOTsadO3c0yiqVSikODx06hNWrV8PS0hIdOnTAvXv3pHL79+9HvXr1NB4fsbS0xPDhw/Ho0SOpd/f9+/fD1dUV/fr1k8qZmJhg7NixSExMxKlTpwAAtra2SEpKkuKf6FXFx8cD0H2XuEWLFnBycpI+6mbSOX344Ydaw3LGUlJSEiIjI9GoUSMIIXD58mUA+TsO5yaEwI4dO9C5c2cIITSOW35+foiLi9M6Nxg0aJB0HgFonxcUVP369aV9wd69e/HFF1/g5s2b6NKli9SMX9/jdFZWFg4fPgx/f394e3tL5UqVKoX+/fvj9OnT0v/L1tYWN2/exP3791+p/rnlPj9o2rQpoqKipPkePHgQADBy5EiNcuqOu4jyo23btjh79iy6dOmCq1evYuHChfDz80Pp0qWxZ8+el/4+574mIyMDUVFRKFeuHGxtbTX2ATt27EDNmjWl1mo55XzcA8huqdquXTvcuXMHJ0+e1Gqdm5cqVapI+xUg+1yjYsWKGvuYgwcPomHDhhrTtLe313lOUtwVKDnv3r07Vq1apbFS1f/Aw4cPazzbRwWnTjq3b9+ODh06IDIyUmfzjg0bNqBGjRrSs1VOTk7Yt2+f9KxzTrkvmqgT9ZiYGADZz6gA0NlRS8WKFTW+h4SEwM3NTSOxBv5rJqqeVl7zVp905HxWL+dwdZ1epl69emjTpo3GJ+f8c9dbXcfIyEgkJSVpDM/d6/v9+/chhED58uU1TsScnJxw+/ZtqVMOLy8vfPzxx1i7di0cHR3h5+eHFStW6PwfEL1OWVlZ+Pnnn9GyZUsEBwcjKCgIQUFBqF+/PsLDw3Hs2DEA2fGhT5wD2c1Ju3XrBhsbG1hbW8PJyUnq6Cn3Ni6Xy6U4bNeuHYYPH46jR48iLi4OU6dOlcqFhITkGZvq8TnrmbvDrdzlRo4ciQoVKqB9+/Zwd3fH4MGDpZNlooJQH9sSExO1xq1evRpHjhzBpk2bdP7W2NhYZxP0f//9V0pELS0t4eTkhObNmwP4L5bycxzOLSIiArGxsVizZo3WMWvQoEEAtDuTetl5QUE5OjpK+4KOHTvi008/xdq1a3HmzBnpjQ76HqcjIiKQnJycZzmVSiXdNJo9ezZiY2NRoUIFVK9eHZMmTSqU193pc/5kZGSkdR7xJt6SQcVT3bp1sXPnTsTExOD8+fOYOnUqEhIS0LNnz5e+njQlJQXTp0+X+oRydHSEk5MTYmNjNY7bDx480PsNR+PHj8c///yDo0ePSs3R9aHrhq2dnZ3GPiYkJERnrJTE+CnQM+ezZs3CiRMnUKtWLTRt2hQymQwLFizAtGnTcPbsWdSuXRuffvppYde1xKlXr57UC7m/vz+aNGmC/v374+7du9KV/E2bNiEwMBD+/v6YNGkSnJ2dIZfLMX/+fI2O49Ty6kVR5OrA7XXIa96GrFNuuZ+3ValUkMlkOHDggM565ryj8vXXXyMwMBC7d+/G4cOHMXbsWMyfPx/nzp3T6zlBosJw/PhxPHv2DD///DN+/vlnrfGbN29Gu3bt9J5ebGwsmjdvDmtra8yePRs+Pj5QKpW4dOkSPvnkE50dyOXm7u6OihUr4o8//sjXsuSHs7Mzrly5gkOHDuHAgQM4cOAA1q1bh4EDB+rsPI7oZWxsbFCqVCncuHFDa5y6hVdenRQpFAqtC0pZWVlo27YtoqOj8cknn6BSpUqwsLDAkydPEBgYqFcsvYx6Gu+99x4CAgJ0lqlRo4bG9zd5DG7dujUA4I8//nhtd5SbNWuGBw8eSMfitWvXYvHixVi1ahWGDh1a4OkWpXMVKllMTU1Rt25d1K1bFxUqVMCgQYOwbdu2F76BZcyYMVi3bh3Gjx+Phg0bwsbGBjKZDH379i3wvqZr1674+eef8eWXX+Knn3566Vsq1Bg7+VOg5NzGxgbnzp3D119/je3bt0OpVOLUqVPw8fHBjBkzMGnSJJ2dClHBqRPuli1bYvny5ZgyZQoAYPv27fD29sbOnTs1mp8U9JVJ6ve06moOdvfuXa2yR48eRUJCgsbdc3UzV0O981VNPf/c9Qay6+jo6PjSV9v4+PhACAEvLy9UqFDhpfOsXr06qlevjs8//xxnzpxB48aNsWrVKsydOxeAdhMhosK2efNmODs762xmu3PnTvz2229YtWoVPDw89IrzkydPIioqCjt37kSzZs2k4cHBwfmqV2ZmpsYdSA8PjzxjUz1e/ffatWtQqVQaJwK69jOmpqbo3LkzOnfuDJVKhZEjR2L16tWYNm0aypUrx/ijfOvYsSPWrl2L8+fPazwCUhDXr1/HvXv3sGHDBgwcOFAanvtRjPwch3NzcnKClZUVsrKypFZkhaGwYiczMxPAf60R9D1OK5VKmJub51nOyMhIoxWevb09Bg0ahEGDBiExMRHNmjXDzJkzpeT8dewLPDw8oFKpEBwcrNHqIWfP90SvSn3T7tmzZwDy3pa3b9+OgIAAfP3119Kw1NRUxMbGapTz8fHReQFSF39/f7Rr1w6BgYGwsrLCypUrC7AEunl4eOiMlZIYPwVq1g5k32H8/PPPceXKFSQlJSElJQU3btzA9OnTmZi/Ji1atEC9evWwZMkSpKamAvjvalTOq09///03zp49W6B5lCpVCrVq1cKGDRs0mr0cOXJEqwlNhw4dkJWVheXLl2sMX7x4MWQyGdq3b1+gOhSWnMuSc2d048YNHD58GB06dHjpNLp37w65XI5Zs2ZpXeETQkhvK4iPj5dOOtSqV68OIyMjpKWlScMsLCy0doxEhSUlJQU7d+5Ep06d0LNnT63P6NGjkZCQgD179qBDhw44d+4czp8/L/0+IiJC65VruvYx6enp+O677/Su171793D37l3UrFlTGtahQwecP39eY1+VlJSENWvWwNPTU3q2tkOHDggLC8Mvv/wilcvMzMSyZctgaWkpNQnO/eYQIyMj6Q6hOgbVF+MYg6SvyZMnw9zcHIMHD0Z4eLjW+Pzc+dEVS0II6XV/avk5DuuaR48ePbBjxw6dJ9wRERF61zcnCwuLQnlM6/fffwcAaV+g73FaLpejXbt22L17t0ZrhfDwcGzZsgVNmjSBtbU1AO19gaWlJcqVK6d1LAYKd1/g5+cHAFr7xmXLlhXaPKjkOHHihM79i7ofBvUjHnmdV8rlcq3fL1u2TOsVqD169MDVq1d1vqpR1/wHDhyIb7/9FqtWrcInn3yi9/K8jJ+fH86ePYsrV65Iw6Kjo3W+Bra4K9CdczKcSZMmoVevXli/fj0++OADdOrUCTt37kS3bt3QsWNHBAcHY9WqVahSpYrO5+T0MX/+fHTs2BFNmjTB4MGDER0djWXLlqFq1aoa0+zcuTNatmyJzz77DI8ePULNmjVx+PBh7N69G+PHjy8SrwlbtGgR2rdvj4YNG2LIkCHSK1psbGz0eueoj48P5s6di6lTp+LRo0fw9/eHlZUVgoOD8dtvv2H48OGYOHEijh8/jtGjR6NXr16oUKECMjMzsXHjRulESc3X1xdHjx7FN998Azc3N3h5eeX7FT1EedmzZw8SEhLQpUsXneMbNGgAJycnbN68GatXr8bGjRvx7rvvYty4cdKr1NR3qtUaNWoEOzs7BAQEYOzYsZDJZNi4cWOeSUlmZqb0HK5KpcKjR4+watUqqFQqjRY9U6ZMwdatW9G+fXuMHTsW9vb22LBhA4KDg7Fjxw7pLvnw4cOxevVqBAYG4uLFi/D09MT27dvx119/YcmSJVKrnaFDhyI6OhqtWrWCu7s7QkJCsGzZMtSqVUt6Pr1WrVqQy+VYsGAB4uLioFAo0KpVK+mdsUS5lS9fHlv+1959x0dR538cf2/J7qZ3SiAQehMCRsGCItLOgg3s9xPBggp4NjyVUywolrsTCyfYsCN3ND0LKhyIvVBFpIYSSCO9l92d3x+RxCWJUkIm2byej8f+weS7k+8M+8nse+Y733nnHV155ZXq0aOHrr76aiUmJsowDO3atUvvvPOOrFbrYd261LNnT3Xp0kV33XWX9u/fr7CwMC1atKjOe7sP9zhcl8cff1wrV67UoEGDdMMNN6h3797KycnR2rVrtXz5cuXk5BzxfkhKStKCBQt0xx136OSTT1ZISIhGjx79u+/Zv39/9d+CiooKbdiwQXPnzlVMTIzPkPbDPU7PmDFDn332mQYPHqxbbrlFdrtdc+fOVXl5uZ588snqdr1799ZZZ52lpKQkRUVF6ccff9TChQs1efJkn+2RpFtvvVWjRo2SzWbTFVdcccT75dB9NGbMGM2aNUvZ2dnVj1I7OBEmI3dwJKZMmaKSkhJdfPHF6tmzpyoqKvT1119rwYIFSkhIqJ5Dor7vleeff77efPNNhYeHq3fv3vrmm2+0fPlyRUdH+/yeqVOnauHChbr00ks1YcIEJSUlKScnR++//77mzJnjc1L9oMmTJ6ugoEDTpk1TeHh4g9zKfPfdd+utt97SiBEjNGXKlOpHqXXo0EE5OTktq36OZor38ePH/+Hr0EdO4PAdfBTBDz/8UOtnHo/H6NKli9GlSxfD7XYbXq/XeOyxx4yOHTsaTqfTGDBggPHBBx8Y48aN83nsycFHqT311FO11qk6HvGxaNEio1evXobT6TR69+5tLF68uNY6DaPq0Ue33367ERcXZwQEBBjdunUznnrqqerHL/z2d0yaNMlnWX19+u0jG452P/3W8uXLjdNPP90IDAw0wsLCjNGjRxubN2/2aXPwUSkHDhyocx2LFi0yBg8ebAQHBxvBwcFGz549jUmTJhlbt241DMMwkpOTjQkTJhhdunQxXC6XERUVZQwdOtRYvny5z3q2bNlinHnmmUZgYKAhiceqoUGNHj3acLlcRnFxcb1trr32WiMgIMDIysoyNm7caAwZMsRwuVxGu3btjEceecR45ZVXaj225KuvvjJOOeUUIzAw0IiLi6t+pIsOeSxZXY9SCwsLM4YNG1arFgzDMHbu3GmMHTvWiIiIMFwulzFw4EDjgw8+qNUuIyPDGD9+vBETE2M4HA6jb9++xrx583zaLFy40Bg5cqTRqlUrw+FwGB06dDAmTpxopKWl+bR76aWXjM6dOxs2m43HquGw7dixw7j55puNrl27Gi6XywgMDDR69uxp3HTTTcb69et92o4bN84IDg6ucz2bN282hg8fboSEhBgxMTHGDTfcYGzYsMGQVOszfbjH4bqO4RkZGcakSZOM+Ph4IyAgwGjTpo0xbNiw6sfgGkb9x9qDx+bf9qeoqMi46qqrjIiIiDof53aoQx+lZrVajVatWhlXXnmlz2NWDzqc47RhGMbatWuNUaNGGSEhIUZQUJAxdOhQ4+uvv/ZpM2PGDGPgwIFGRERE9f/To48+6vMIKrfbbUyZMsWIjY01LBaLz2PVDt2f9X0/qOuxUcXFxcakSZOMqKgoIyQkxLjooouMrVu3GpKMxx9//Hf3GfBbH3/8sTFhwgSjZ8+eRkhIiOFwOIyuXbsaU6ZMMTIyMqrb1fe9Mjc3t/q4GRISYowaNcrYsmWL0bFjx1rfPbOzs43Jkycb7dq1MxwOh9G+fXtj3Lhx1Y89rO9vxd13321IMp5//nnDMOp/lFpdjzYcMmSIMWTIEJ9l69atM8444wzD6XQa7du3N2bOnGk8++yzhiQjPT39KPdk82MxjCO/Gz8hIaHWGQyPx6O0tDR5PB7FxsYqODj4mB/DAQAAADRX69ev14ABA/TWW2+1yMdCAcfitttu09y5c1VUVFTvxHL+5qiGtdc3O2llZaXmzp2rWbNm8axZAAAAtBilpaW15l2aNWuWrFarz4SaAGo7tH6ys7P15ptvavDgwS0mmEvSUV05/yO33HKL9uzZow8//LChVw0AAAA0OQ899JDWrFmjoUOHym63Vz/W8eDcGQDq179/f5111lnq1auXMjIy9Morryg1NVUrVqxoUSe3jsuEcImJiXrzzTePx6oBAACAJue0007TZ599pkceeURFRUXq0KGDHnzwQU2bNs3srgFN3rnnnquFCxfqxRdflMVi0YknnqhXXnmlRQVz6ThdOR87dqy++OKLOh87AgAAAAAAfB3VlfOHH364zuV5eXlavXq11q5dq3vuueeYOgYAAAAAQEtxVFfODz5/9lCRkZHq0qWLrr/+et1www0t65l0AAAAAAAcpeMyrL2l8Hq9Sk1NVWhoKCci0GQYhqHCwkLFxcXVeyINf4z6RlNEfTcM6htNEfXdMKhvNEWHW99HPKy9tLRU06ZN09ChQzV69Ohj6mRzl5qaqvj4eLO7AdQpJSVF7du3N7sbzRb1jaaM+j421DeaMur72FDfaMr+qL6POJwHBgZq7ty56t279zF1zB+EhoZKkh783+lyhRyXie9blGWTTje7C37B7SnXl2v/Uf35xNE5uP/2rE1QWAhXMI7Vxd37mt0Fv+BWpb7UR9T3MaK+Gxb13TCo74ZBfTcs6rthHG59H1WiTEpK0qZNm46qY/7k4FAZV4idcN4A7HaX2V3wKwzlOjYH919YiFVhoRzcj5XdEmB2F/zDrzeiUd/HhvpuWNR3A6G+GwT13bCo7wZymPV9VJ/YWbNm6d1339XLL78st9t9NKsAAAAAAAC/OuzLvatXr1avXr0UGxurcePGyWq1auLEibr11lvVrl07BQYG+rS3WCzasGFDg3cYAAAAAAB/c9jhfOjQoXrrrbd05ZVXKjo6WjExMerRo8fx7BsAAAAAAC3CYYdzwzB08Klrq1atOl79AQAAAACgxWGWBAAAAAAATHZE4ZzZIwEAAAAAaHhHFM7//Oc/y2azHdbLbufRYgAAAAAAHI4jStDDhw9X9+7dj1dfAAAAAABokY4onI8bN05XXXXV8eoLAAAAAAAtEhPCAQAAAABgMsI5AAAAAAAmI5wDAAAAAGCyw77n3Ov1Hs9+AAAAAADQYnHlHAAAAAAAkxHOAQAAAAAwGeEcAAAAAACTEc4BAAAAADAZ4RwAAAAAAJMRzgEAAAAAMBnhHAAAAAAAkxHOAQAAAAAwGeEcAAAAAACTEc4BAAAAADAZ4RwAAAAAAJMRzgEAAAAAMBnhHAAAAAAAkxHOAQAAAAAwGeEcAAAAAACTEc4BAAAAADAZ4RwAAAAAAJMRzgEAAAAAMBnhHAAAAAAAk9nN7gCO3O53nNo5z6nyLKvCenjU574SRfbz1Nu+ssCiLc+4lL7cocp8iwLjvOp9T4lan+mWJG2d7dL2fwX6vCe4k0dDPyg4rtvRFIz+01aNvehnRUWUKnl3pP718kBt3RFTZ9vTB+3VFWN+UlzbQtltXu1PC9Oi93trxeedq9vcOfkrjTw72ed9P66L07RHhh3X7YD/eH9ejBa+0Eo5B+zq3LtUt8zYr54DSups666U3n2utZb/J0pZ6QFq36Vc101L1clDC33aZaUF6JVH2+qHlWEqL7UqLqFcdz69V90TSxtjk0wz+tosjb05U1GxbiVvDtS//tZOW9cH1dv+jPPzNO7udLVuX6H9u5xV++x/YXW2vfXxfTrvmmzNeSBOS16OPV6bgBbgSGp+6piu2vhNSK3lA4fl65E3dx3vrjZpR1Lvp5+TpytuzVRcQrnsAdL+XQ4tmhOrFYuiGrnX8CdHUsuStPilWH34erQyUx0Ki3TrjPPzNOHeNDlchiSppMiq159sq68/Dldetl1d+pTq5kf2qUd//z52S0dWzx27l+maqenq2q9EbeIr6zwunzCoSJfeckDd+pYouo1bD05I0DfLwhtjU44Y4byZSf04QJufDFTf6SWK6OvWrjdd+n5iiM76oEDOaKNWe2+F9O31IXJGG0p6ukiu1oZKU60KCPVtG9rVo0Ev13yht7aAT8aQ03frxvE/6rm5g7RlW4wuPv8XPfrACl035QLl5wfWal9Y5ND8RX2Vsi9cbrdVg07apzsnf628fJfWrI+rbvfD2jj94/nTqv9dWckAFRyeVe9F6MWH4jTl8X3qeWKxlrwUq2lXddYrX2xRRIy7VvvXnmir/y2O1G1PpSi+a7l+XBWqh6/rpKff266ufasO3oV5Nt1xYTf1O61QM95KVkS0W/uTnQoJr/+Enj8YckGubpyequfuaa8ta4N08Q0H9Og7ybrujB7Kzw6o1b73ScW691979OrMtvruszANvThX01/drUmjumnPVt+/B6f9KV89k4qVldYC/lDiuDrSmr//5V1yV1qq/12Qa9fNw3vojPPzG7PbTc6R1nthnl3zn2mtlB1OuSstGjS8QHc+naK8LLvWfF73CTng9xxpLf9vcYRefayt7vjHXvU+uUT7dzr199s7yGKRJj6YKkl6+s547d7q0t3P7VFU60r9b1GU7rm8q15atUUxbSsbexMbzZHWszPQq7S9Dq3+ILx63x3KFeRV8s8ufTI/StNf3X2ct+DYkBokzZ49WwkJCXK5XBo0aJC+//57s7tUr+TXXYofW674iysU2tWrvtNLZHVJKYsddbZPWeJQZYFFJz1bpKgTPQpq51X0yW6F9fT9Ym6xGXLF1rwckbWDvr+5ZPRmLfusmz79X1ft3RehZ+eeovJym0advbPO9ht/bqOvv+uglP3hSssI1dIPeyl5T6T69Mr0aVdZaVVuXmD1q6jY2Ribg3o0p/pe/GKs/nRVtkZdkaOO3ct16xP75Az06pP5dV/NWbEoSldMydTAYYVq27FCo8dl6+SzC7Robs0Z43/PbqWYuArdNStFPQeUqE2HCiWdVai4hIrG2ixTXHJjlpa9E6VPF0Rp73aXnv1re5WXWjTqypw62190/QH9uDJUC19opZQdLr3xVFvt+ClQF47P9mkX3aZSt8zYrycmdZTbbalzXWg8zam+63KkNR8W6VFUK3f1a+3qULkCvTpzdF7jdryJOdJ63/hNiL5eFq6UHS6l7XFq6SuxSv4lUH0GFjdyz/F7mlN9H2ktb/4xWH1OLtbZl+SpTXzVcfmsi3K1dV3V1eHyUou+/ChC1/8tTX1PKVa7ThX6v7vSFZdQrg/eiG7MTWt0R1rP2zYE6eVH4vT5e5GqrKj7uPzjyrCqUQhN9Gr5b7X4cL5gwQLdcccdmj59utauXavExESNGjVKmZmZf/zmRuatkPI32xR7as0ZOItVij2lUrkb6r6Ck77SochEtzbNCNKnZ4br8wvDtP1Fl4xDLpoV77Xps7PC9b9RYVp7d5BKU/37S6fd7lG3Ljlau7FN9TLDsGjdxrbq3ePAYazBUP++aYqPy9emza18ftLvhAwtmPdvvfzce5py43cKDSlv4N7jcDWn+q6ssGj7xiCdeEZR9TKrVRpwRpE2rwmu9z0Op9dnmdPl1c/f1wx7/fbTcHVPLNGMGxN0Wd8+umVEd330tn8P3bQHeNWtX4nWfhFavcwwLFr3Rah6J9U9xLBXUonW/aa9JK35PFS9kmq+rFsshu5+dq8WvhCrPdtcx6fzOGzNqb7rcjQ1f6hP5kdpyIW5cgV5/7ixnzqaevdlqP/gQsV3Kdem72rfMgBzNKf6Pppa7n1SsbZvDNKWX8N42h6HflgRppOHVd1S6vFY5PX88THe3xx7PTd/LT6c//Of/9QNN9yg8ePHq3fv3pozZ46CgoL06quvmt21WiryLDI8FjmjfQvVEW2oPKvu/8qSfValfeqQ4ZUGvlCkbjeVKvk1p7bPrfliGdnPrcRHizVobpFOuL9Epftt+vqaULn9+ARyWGi5bDZDeXm+w1Vz81yKjKj/Xp6goAotfXu+Pvz323pk2v80++WBWruhZkj7j+vi9NSzp+uv00folTcHqG+fDD16/wpZrS33i5OZmlN9F+TY5PVYFBHrO1QtMqZSuQfqPvmWNKRQi16M1f5kh7xeac3nIfrqowjlZNa0T9vr0AdvxCiuU7keeydZ54/L1gv3t9dn/448rttjprAoj2x2Ke+Q/ZabZVdkbO3hhZIUGetWbtYh7Q/YFdmqpv1lkzLl8UhLX6l7Xgo0ruZU33U5mpr/rS3rgrR7S6D+dFXdV5NaiqOpd0kKCvVo6faf9OGejXrkjV2a/bc4rV0dWm97NK7mVN9HU8tnX5Kna+5K050XddW5HRJ17am91e+0Il15a9XJh6AQr3olFeudWW2UnW6XxyOtWBSpX9YEKyfDf2+pOtp69if++797GCoqKrRmzRrde++91cusVquGDx+ub775plb78vJylZfXXAUtKGgGE6Z5JUeUoX4PlshikyL6eFSWUaad81zqfkuZJKnVGTUf9rAeUmS/Iq0YEa7UZQ51GOPfQ1+PVGlpgG658zy5XG4N6JeuieN/VHpGiDb+XHUF/vOvOlW33b03Urv2ROr1F5aqX58Mrf+prVndbpFaQn3f/Mg+zbqrg64/s5dkkeI6lmvk5dn6ZEHNkDfDK3XrV6oJ96ZJkrr2LdXuLS59+GaMRlyWa1bXm52ufUt00fVZmjSquyT/HlnUHLSE+v4jn8yPUqdepb874RTqV1pk1S0jussV7NWAwYWaOD1V6XucdU64h8bVEup7w9cheve51pr82D71PLFEqbudeuH+dnr76da6+vYMSdLdz+3RP+/ooKtOPEFWm6GufUt01kW52r6x/olN0fy16CvnWVlZ8ng8at26tc/y1q1bKz09vVb7mTNnKjw8vPoVHx/fWF2VJDkiDFlshsqzff/bKrItcsbUfWXWGetVSIJHFlvNspAuHpVnWeWtJ3cHhBkK7uhR8V7//XgUFDrl8VgUcchV8siIMuXm1Z4M7iDDsCg1PUzJu6O06P3e+uKbjrr8kk31tk/PCFVevlNxbQvrbYPjo7nVd1iUR1abobwDvpOd5GYF1Hu2OCLaowfn7dJ7Ozbqze836+UvtsgV7FWbDjVfUqJaudWxe5nP++K7lSlzf+1JVfxFQY5NHrcUcch+i4xx13sVI/eAXZGHTNoTGetW7q+jEPoOKlZEjFtv/bBZH+3doI/2blCb+ErdMD1Vr3+3+fhsCOrV3Oq7LkdT8weVlVi16r1Ijboy+3fbtQRHU+/Sr8fz3U4l/xyoRXNb6YsPI3T5lIzj3V0chuZW30dTy68/2UbDxuTqnKtz1KlXmU4/J1/j703Tguday/vrV/q4hAr9ffEOvbdjo9768Wc999F2uSstatvRf2+XPNp69if+m76Og3vvvVf5+fnVr5SUlEb9/VaHFN7bo6xvaz6chlfK+i5AkYl1F3/UALeK91pl/Ca7F+22yRnrlbXuOeTkLpZKUqxyxfrvpHBut03bd0ZpQL+aP/IWi6H+/dK1eevhPxbJajEUEFD/kPWY6GKFhZYrJ7f+wI+mwez6DnAY6tavROu+rLlq4/VK678MUe+k37/HxOEyFNO2Uh639OVHETp1VM1Vg94nFytlp++khPuTnWrVzn9nenVXWrV9Y5AGDK45KWaxGOo/uEib19R9xeGXNUHq/5v7BSXpxDML9cuv9wsuXxSpm4Z1180jal5ZaXYtfKFqRl40bWbXd12OpeZX/zdClRUWDbuE0S9HU+91sVoNBTj893uPPzO7vo+mlstLrbJYfT9v1l//bRzyMXQFeRXd2q3CPJvWfB7mc4z3Nw1Vz81ZyzgFUY+YmBjZbDZlZPieKc3IyFCbNm1qtXc6nXI6zZ15u/O4Mq2/L1jhfTzVj1LzlErxF1ddBl93b5BcrbzqdXvVlbKOl5dr9zsu/TwzUAlXl6t4j1U7XnKp09U1Z902PxWo1mdVKjDOq7JMi7bNDpTFJsWd699D2hf/t7fumvKVtu2I1tbtMbp49C9yOd369H9dJElTb/1KWdmBmvf2iZKkyy/5Sdt3Ris1PVQBdq8GJu3XsCHJeu7FQZIkl6tSf75so778toNycwPVtk2hrr9mrVLTQ7VmXVy9/cDx0Rzr+5IbD+jvt3VQ98QS9RhQoiUvxaqsxKqRV1TdU/rkrR0U06ZSE+6rGqK+ZW2QstID1KVPqbLSA/TWP9rI8EqX3ZL5m3Vm6vYLumv+s6105ug8bV0XpI/eitZtT+0zZRsby+IXY3TXrBRt2xCkreuqHsXiCvLq03erJsOb+sxeZaUHaN7MqttNlr4cq6cW7dCYiZn6fkWYhlyYp279SjVrantJUmGuXYW5vodMt9ui3MwA7dvJ5HCNrTnWd12OtOYPWjY/SqeNyldYlH8/EvFwHWm9Xz45Q9s3Bil1t0MBDkMDhxVo2JhcPXdvezM3A79qjvV9pLV8yogCLX4xVl1PKFXPE0u0f5dDrz/VVoNG5Mv262jXH1eFyjCk+C7l2r/LoZcfaaf4rmUaebl/j5g50nq2B3jVoXtVrgkIMBTdtlKd+5SqrNiq1N1VnwtXkEdxnWpyTZv4CnXuU6rCPJsO7K/naqVJWnQ4dzgcSkpK0ooVK3TRRRdJkrxer1asWKHJkyeb27l6xJ1TqfKcUm173qXyLKvCeno0cG6RnDFVp9lK06yy/OZ2yMC2hga9WKifnwjS6oudcrX2qtOfy9X1upphrmUZVq2dGqzKPIscUYaiTnTr9HcK5Yzy7zPIn3+VoPCwMl1z5QZFRpQqeVekpj1ytvJ+fcZ5bExx9dAiSXI53Zp8w/eKiS5RRYVNKfvD9eQzg/X5VwmSJK/Xok4dczVi6E4FB1UqOzdQa9e31evz+6vSbaujBziemmN9n3VhnvKz7XrjqbbKPWBX5z6levTt5OphcQf2O2T9zXininKLXn+irdL2OhQY5NXJwwp097N7fJ5h3qN/qR54ZZfmzWyrt59uozbxFbrp4f0628+vuH3+fqTCoz26Zmq6ImPdSv45UNOu7qS8rKphh7HtKnzqe/OPwXp8UkeN+2u6rr0nXam7nHpoQkKtZ5yjaWiO9V2XI615SUrZ4dTP34fosfk7TOhx03Sk9e4K8mryY/sU07ZSFWVWpex06skpHfT5+/47UWZz0hzr+0hr+arb0mWxGHrtybbKTg9QeJRbp4zI17X31IzoLC6wad7MtspKC1BohEenn5un8fekye6/d6VJOvJ6jm7t1gufbav+96U3H9ClNx/Qhq+DdffYrpKk7omlempRzaOSb3qo6nnony6I1D9u79AIW3X4LIZx6OCJlmXBggUaN26c5s6dq4EDB2rWrFn697//rS1bttS61+VQBQUFCg8P1+PfD5ErpEWf52gQH1w3xOwu+AW3u0yrfnhM+fn5CgsLM7s7pmqI+s7d1llhodwBdKxGxfU3uwt+wW1UapXeo75FfTcl1HfDoL5rUN9NB/XdMA63vlt8orz88st14MABPfDAA0pPT1f//v21bNmyPyx8AE0f9Q34L+ob8F/UN1qqFh/OJWny5MlNdpgMgGNDfQP+i/oG/Bf1jZaIsR4AAAAAAJiMcA4AAAAAgMkI5wAAAAAAmIxwDgAAAACAyQjnAAAAAACYjHAOAAAAAIDJCOcAAAAAAJiMcA4AAAAAgMkI5wAAAAAAmIxwDgAAAACAyQjnAAAAAACYjHAOAAAAAIDJCOcAAAAAAJiMcA4AAAAAgMkI5wAAAAAAmIxwDgAAAACAyQjnAAAAAACYjHAOAAAAAIDJCOcAAAAAAJiMcA4AAAAAgMkI5wAAAAAAmIxwDgAAAACAyQjnAAAAAACYjHAOAAAAAIDJCOcAAAAAAJiMcA4AAAAAgMkI5wAAAAAAmIxwDgAAAACAyQjnAAAAAACYjHAOAAAAAIDJCOcAAAAAAJiMcA4AAAAAgMkI5wAAAAAAmIxwDgAAAACAyexmd8AfLBsYIbslwOxuNHufpL5hdhf8QkGhV5Hdze6F/3g6p7NcFdT3sfIO7m92F/yC110mffOe2d3wG2MuGiO7zWl2N5q967Z9bHYX/EJJoUerTjS7F/5j8D+ul83hMrsbzZ53aa7ZXfALnpJy6co/bseVcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExmN7sDOHKjr83S2JszFRXrVvLmQP3rb+20dX1QnW3PuSpbwy/NVcceZZKkHT8Fat7Mtj7tXUEeXTctTaeOKlBYpFvpKQ6990qMPnwzplG2x0zvz4vRwhdaKeeAXZ17l+qWGfvVc0BJnW2njumqjd+E1Fo+cFi+Hnlzl9yV0mtPtNUP/wtT2h6HgsO8GnBGoa67L1XRbdzHe1PgJ/bNt2vvawGqyLIopIdX3e+tUFhfb51t1453Ke9HW63l0We4lfivcklSRZa042mHcr6xyV1oUUSSR93vrVBQR+O4bkdTcMGoLbp09CZFRZRq554ozX51oLbujK2z7TnDtmnEmTuVEJ8nSdqeHK1X5w/waT944B6dP2KrunXOUVhouW6aOlo790Q1xqbAT5w/ervGXrpFkVFlSk6O0AuzT9S2rdF1tj3t9H26/MrNiosrkt3u1f79oVq8sIf+tyKhus0dd32nESN3+7zvxx/a6P5pQ47jVjQNm98K1U+vhKv0gE1RPSt06v3Zik2sqLd9eYFVa/4Zqd2fBak8z6aQdm6dcl+24s8qlSStfTZC656P9HlPeKcKjf1k/3HdDviPy5I2adyg9YoOKdG2jGg98elg/ZzWus62Z/dI1nWnrVV8ZL7sVq/25obrze8S9eGmHtVtooJL9Jeh3+rUTikKcVVo7d62evLTwdqbG9FIW2SeoI9yFLwkR9Y8tyoTnCq8oY0quwfW295S5FHI2wfk+rZA1kKvPK0CVDChtSpOqvreHvBziYKXZCtgZ5lsuW7l3tNe5aeENtbmHJEWfeV89erVGj16tOLi4mSxWLR06VKzu/SHhlyQqxunp+rtf7bRpFHdlbzZpUffSVZ4dGWd7fudVqSVSyN096VddPsFXXUgNUCPzd+p6DY17Sc+mKqTzirUk1M66IYhPbXkpVhNenS/ThmZ31ibZYpV70XoxYfidPUd6Zr9yVZ17l2qaVd1Vl5W3ees7n95l+av31T9mrtyi6w2Q2ecX7Wfykut2vFTkK66LUOzP9mmB17epX07nZp+befG3Cz8qjnWd8Yym7Y/5VDCTZU6+d+lCunu1fqJLlVk192+76wynb6ypPo1cEmJLDZDrUZ6JEmGIW38i0ul+6zq92y5Tv53qVxtDa27wSVP3eeg/MaQU3dp4jU/6K2Fibr5r6OVvCdSM6ctV0RYaZ3tE3una+VXnTT1oVH6y9/O1YHsID3+t88UHVlc3cbldGvTltZ6+e0TG2szUI/mWN9nDtmrGyeu19tv9dGUW0ZqV3KEZjz2ucIjyupsX1jo0IL5vXXHX4brlol/0mefdNIdd32vE5PSfNr98EMbXXX5BdWvJ2ae2hibY6rkD4P13cxoDZicpwuXpiqqZ4WWXddGpdl1f631VEjLrm2jwv12DXs2U2M/2afBM7IU1Nrj0y6iW4Wu/Gpv9ev8+Wl1rg/HV3Os75G9dujOYV9p7pcn6apXx2pbZrT+dcUHigyq+2CbX+rUy1+dqHGvX6LLXr5M723sqQfPX6lTO+39tYWhp8csU/uIAt228Bxd+cpYpRWEas5V/5UroO7v/P7C9WWBQl/NVNEVMcr6Zye5E1yKfGivrHn1XOiqNBT14F7ZMiuUd3d7Zc3urPxb2sgbXfN93lLmlbuTUwUT6z5Z0pS06HBeXFysxMREzZ492+yuHLZLbszSsnei9OmCKO3d7tKzf22v8lKLRl2ZU2f7JyZ31Aevxyj550Cl7HDp6TvjZbFKAwYXVrfpfVKJPvtPlDZ+E6KMfQ59/Ha0kjcHqkd///72vvjFWP3pqmyNuiJHHbuX69Yn9skZ6NUn8+u+EhYW6VFUK3f1a+3qULkCvTpzdJ4kKTjMq8cX7NSQC/IU37VcvZJKNOnRfdq+MUiZ+wIaccsgNc/6TnkjQHFj3Iq72K3gLoZ6PFAha6Ch1CV1f34CwiVnjFH9yv3GJqtLajWy6gBWuseigo029bi/XGEneBXcyVCP+yvkLbco42P/Hjg15vzN+nhFN32yqpv27o/QMy+dqvIKm0YN3VFn+8efO1P//bSndu6JUkpquP455zRZLNKAvunVbZZ/0UVvLUrU2p/iGmszUI/mWN8Xj9mqjz/urM8+7ay9e8P13DMnqbzcrpGjdtXZ/qeNrfT1V+2VkhKmtLQQvbe0u3Ylh6vPCVk+7SorbcrNDax+FRU5GmNzTLVpXph6XFao7mOKFNm1Uqc/nC27y9C2hXVfCdu2KFTl+VaN+FeGWieVK7S9W20Hlim6l++VdqvNUFCsp/rliqp71BKOr+ZY338euEGL1/fW+xt7KjkrSo9+PERl7gBdlLilzvZr9rbTym2dtSs7UvvywjX/h37anhmtAfFVx5wOUfnq1z5Djy47U5vTWmlPTqQe+/hMOe1undN7e2NuWqMLei9bJSMjVDosQp54pwpubiPDaVXgirw62weuyJOl0KO8e+NV2StIntYOVZ4QLHcnV3WbiqQQFV3dSuWnhDXSVhw9//529gfOOeccnXPOOWZ347DZA7zq1q9E7z7fqnqZYVi07otQ9U46vCDtDPTKbjdUmFfzX7/5xyCdMjJfn7wbpex0uxJPK1a7zuWaM91/v4BWVli0fWOQrpicWb3MapUGnFGkzWuCD2sdn8yP0pALc+UKqv/gXVxgk8ViKDjcU28bHB/Nrb69lVLhZqs6XldzRtxilaJO8ahgw+GdR01dHKDWf3LL9utdK95fv3danTVtLFbJGmAob61VcWMaqvdNi93mUffO2Xp3ad/qZYZh0dqf4tS7+4HDWofT6ZHd7lVhCwg6zVFzq2+73aNu3XL173d7VS8zDIvWr2utXr2yfued1a3Vv3+m2scX6tVXfG/N6NcvU/P/vVRFhQ5tWN9Kr7/WV4WFznrW0/x5KqSsn53qN7FmdJ/FKsWdVqrM9XVv994VQWo1oFxfPxStPSuC5YryqMv5Rep3Y76sv7kzqGBPgOYPjpfNYajVgHKddGeOQuI4fje2ZlffVo96tT2gV7+pGVVlyKLvdrVTv3YZh7EGQwMT9ishKk/P7D1FkuSwVX3uKty237SyqMJjU//4dC3Z0LtBt6HJqDQUsLNMxWN+c2ut1aKKxGAFbK175Jvr+0JV9gxU2Nx0Ob8vlDfcrrIzwlR8SbRkszRSxxtOiw7nzU1YlEc2u5R3wPe/LTfLrviu5Ye1juumpSk7I0Brv6i5d/pff2unvzy5T++s3Sx3peT1WvTM1Pba9F3t+6v9RUGOTV6PRRGxvkODImMqlbLjj7/UbFkXpN1bAnX7P1LqbVNRZtErj8bprItyFRzK2Xf8vspciwyPRY5o33vBHdGGSnb9cTgv+Mmq4h1W9Xq45m9BUCdDzrZeJc9yqMcD5bIFVV2dL8+wqiLLfz+T4WHlstkM5ea5fJbn5rkUH3d4t+tcf/UaZecEcpUcDSIsrKLqM5l7yGcy16X28QX1vi8oqEJvzf+vAgI88notmv1cktatbVP98zU/ttVXX7ZXRnqw2sYV6drxP+mRR1frjtuGyev1z8GRZbk2GR6LAmN8Q3NgjEf5yXWPMipMsSvtW7u6XFCsUS+lq2BPgL5+KFpet0UnTsmTJMUmluvMxw8ovFOlSg7Yte75CH1wVZwu+WCfHCH+P0cHjl5kUJnsVkM5xb73RGcXBykhOq/e94U4y/XJlDcUYPPKa1g0c9kZ+m53vCRpd3aE0vJDNGXod5rx8RCVVtj154Eb1SasWDEh/juy1VrolsUreSN859PxhNvk2Fd31rFlVMrxU4lKzwxT7v3xsqdXKmxuuuQxVHxF3fPMNGWE8yNQXl6u8vKaD0ZBQf0H1KbosskZOuvCPE0d20WV5TUH7QsnZKlnUokeGJegzH0O9T2lWJMe26/sjACt+6JpTpZgtk/mR6lTr9J6J49zV0qPTkyQDGnK4/sat3M4Ks29vlMX2xXczeszeZw1QOr7dLm2THfoi8HBstgMRZ7iUfRgtwy+a9br8gt/0lmn79JdD45SZWXtCffQ/DTX+i4tDdCkm0cq0OVW/wEZumHieqWlheinjVUj6D5f1aG67e7dEdqVHKF5b3yofv0OaP36pn9vZWMxDItc0V6d/kiWrDYp5oQKFWfY9NMr4dXhPH5IzVW5qJ6Vik0s14Kz4rXr42D1uLTIpJ7jcDTX+i4ud+iKVy5TYEClBiXs053Dv9a+vDCt2dtObq9Ndy76k6aft1Kr73hVbq9F3+1qry93dJDFwgHchyF5w20quKWtZLPI3TVQ1uxKBS/NJpz7u5kzZ+qhhx4y7fcX5NjkcUsRsb4TIkTGuJV74Pf/K8felKnLJ2Xqnsu7aNcvNWf2HC6vrr0nXQ9fl6DvV1Tdh7Hrl0B17lOqsTcd8NtwHhblkdVmKO+A71n23KwARcb+/szqZSVWrXovUtdMrXuimIPBPGO/Q0/+ewdXzZsJs+s7INKQxWaoItt3CFZFdu2r6YfylEgZy+zqPKn2TMVhfbwauLBM7sKqofOOKOnHq1wK7e2/n8v8Aqc8HosiD5loKzKiTLl59c/2KkljR2/SFRf9pL8+MlK79jITu78wu74LChxVn8nIQz6TkWXKzXHV866qUJmWWnUcTk6OVHyHAl1+xS/V4fxQ6ekhys9zqm27Qr8N565Ijyw2Q6VZvifOSrNsCoytewh6UKxbVrt8hrBHdKlU6QG7PBWSrY67V5xhXoUnVKpgD3PGNHVm13duiUtur0VRwb7DrqODS5RdXPfTlKSqYeopueGSpG2ZMeoUk6sJp63Tmr3tJEm/pMfqilcuU4izXAE2r3JLAvXGuEXanN78Aufh8obaZVgla55vLdvyPfJG1p11vJF2GTb5DGF3t3fKluuRKg0poHkNbffPMU/Hyb333qv8/PzqV0pK/UOajwd3pVXbNwb5TOZmsRjqP7hIm9fUX/yX3pKpq27L0LSrO2v7Rt92druhAIch7yHf070eyWL13zNzAQ5D3fqVaN2XNUP3vV5p/Zch6p1U/DvvlFb/N0KVFRYNuyS31s8OBvP9u5x6fMEOhUVxr1pzYXZ9WwOk0N5e5X73m/vLvFLutzaFJf5+kM781C6jQmpzfv0nluyhVcG8ZI9FBT9bFXO2/3423R6btiVHa8AJNSfQLBZDA05I0+Zt9X+pueyCTfrzmI2677ER2pbs/4+SbEnMrm+326bt2yPVv3/N/acWi6H+/TP0yy+H/1mzWqSAgPprNyamRKFh5crJ/v2TUM2ZzSHF9ClX2jc1JzUMr5T6TaBa9a972GvrE8tVsNcu4zd/SvN3ByiolbvOYC5JlcUWFaTYFdTKf/9W+gvT69tr0y9psRqUUDNS0vLrfeQb9x/+STKLpeZe898qKncqtyRQHSLz1LvtAa3a1qlB+t0kBVhU2cUlx8bffBf3GnJsLFZlj7r/rlX0DJQ9rVLy1uQWe2qFPJH2ZhfMJa6cHxGn0ymn09xJVha/GKO7ZqVo24YgbV0XpItvOCBXkFefvlt1hWfqM3uVlR6geTPbSpIum5Sp/7srXU9M6qCMFIcif73HurTYqrISm0qKbNrwdbBuuD9NFWVWZewLUL9TizV8bK5efMi/77W85MYD+vttHdQ9sUQ9BpRoyUuxKiuxauQVVTPfP3lrB8W0qdSE+3yvkC+bH6XTRuXXCt7uSumRGzppx0+BeviNZHk9FuVkVpVYaIRHAQ7/PdnhD5pCfcdfU6lfpjkV2sersL4epbwZIE+pRXEXVdXt5vsccrYy1OU237kSUpfYFXO2RwERtdeZ+YlNAVGGXG0MFW23avsTDsWe7VH0af79hXPRB71196QvtS05Wlt3xOjic3+Ry+nWJ6u6SpLunvSFsnKC9Or8JElVQ9mvuWy9Zj57ptIzQxQZXnUFpLTMrrLyqitnocHlahVTrOioqttZ2v96/3pOXqBy8/03DPmDplDfSxb10J1Tv9P27VHauiVaF12yVU6XW599UvVF+86p3yo7O0ivvdpPknTZFZu1fVuU0lJDFBDg1ckDU3X28N16/tmqz6zLVamr/+9nffVFe+XkBiqubZEm3LBBqakhWrumTb398AcnjC/Q6r/GKOaECsX2K9em18PkLrWo+5iqixefT41RUGuPTr6r6iR6z6sKtPmtMH07I0q9/69A+bsDtGFOhPpcUzMHxXePR6nD2SUKiXOrJNOmtc9GymqVOp/PkPamrinU91vfJ+rh0f/T5rRYbUptrasGblRgQKXe29hTkvTI6BXKLAzWc6uqJnybcOpa/ZwWq3154XLYPBrcZY/OO2GbZi47o3qdw3vuVG6JS+kFoeoWm62pI77Sqm0J+nZXvCnb2FhKLoxW+DOpquzqUmW3QAX/N0eWMq9Kh0VIksJnpcoTbVfR/1WNICr5U6SCPspV6MsZKjkvUva0CgUvzFLJ+TWj3yylXtnSakYX2jIrZE8ukzfUJm9s0xod06LDeVFRkXbsqHmszq5du7R+/XpFRUWpQ4cOv/NO83z+fqTCoz26Zmq6ImPdSv45UNOu7qS8rKoPVmy7Cp+r4OddkyWH09D9L+/xWc+b/2itt/5RdfCeeXNHTbgvTX99fo9CIzzK3O/Qa0+01QdvRDfadpnhrAvzlJ9t1xtPtVXuAbs69ynVo28nVw9rP7DfIeshY0tSdjj18/chemx+7ccxZaU79O2nVcOTbhnR0+dnTy7cocTTOMA3puZY363/5FFlToWSZweoIsuh0J5eJc4pk+PXC2tlaVbJ4nsVvXiXRflrbeo/t+5ZTMuzLNr+lKNqeHysobaj3Uq4yb+fkSpJn3/TSRFhZRp32XpFRpRq5+4o3ffYcOX9GqJbxRTLMGrOqJ8/YqscAV5Nv3OVz3re+E+i3vxPf0nSqSelaOqkr6p/9rfbV9dqg8bRHOt79ecdFB5erj9fs0lRkWXamRyh+6cNUd6vExe2alXi85l0uTyaNGWNYmJKVVFuU0pKqJ564hSt/rxq+7xeizp1ytfwEbsVHFypnGyX1q5tozde6+v3cyV0Pq9YZTlWrXk2UqUHbIruVa5Rr2QoMKbq72NRml2W3xy/Q9p6NOrVdH33WJSWjG6noNYe9bkmX/1urAnnxek2rbojVmW5NrmiPGqdVKbR/0lVII9Ta3TNsb4//aWrIoNKdfOZPyg6uERbM2I0acH5yvl1WHubsCJ5f1vfjkrd96cv1Cq0SOVuu3ZnR+hv7w/Tp790rW4TG1KsO4d/pejgUmUVBemDn3roxS+TGn3bGlvZ4DBZ890KnX9A1lyPKjs5lTu9g7wRVbHVdqBS+s0FcW9sgHKnxyv01QzF3JYnT5RdJedHVc3W/quAHaWKun9v9b/DXq16WlPp0HDl/6VpXYy0GEbLnRZo1apVGjp0aK3l48aN02uvvfaH7y8oKFB4eLjO0oWyW5rWWZfm6JPU9WZ3wS8UFHoV2T1Z+fn5Cgtr+s9zPF4aqr4f+Ha4XCHU97FafvPpZnfBL7jdZVr9zQzqu4Hq++zeU2W3+e9jxxrL+MUfm90Fv1BS6NF1J66nvhuovvtMfEw2R/1zOuDweIfXvo0TR85TUq5frnzyD+u7RV85P+uss9SCz00Afo36BvwX9Q34L+obLRkTwgEAAAAAYDLCOQAAAAAAJiOcAwAAAABgMsI5AAAAAAAmI5wDAAAAAGAywjkAAAAAACYjnAMAAAAAYDLCOQAAAAAAJiOcAwAAAABgMsI5AAAAAAAmI5wDAAAAAGAywjkAAAAAACYjnAMAAAAAYDLCOQAAAAAAJiOcAwAAAABgMsI5AAAAAAAmI5wDAAAAAGAywjkAAAAAACYjnAMAAAAAYDLCOQAAAAAAJiOcAwAAAABgMsI5AAAAAAAmI5wDAAAAAGAywjkAAAAAACYjnAMAAAAAYDLCOQAAAAAAJiOcAwAAAABgMsI5AAAAAAAmI5wDAAAAAGAywjkAAAAAACYjnAMAAAAAYDLCOQAAAAAAJiOcAwAAAABgMsI5AAAAAAAms5vdgebMMAxJkluVkmFyZ/xAQaHX7C74hYKiqv148POJo3Nw/5UXu03uiX9wu8vM7oJfcLvLJVHfx6r6+O0pN7kn/qGk0GN2F/xCaVHVfqS+j83B/eep4LjTELwl/J1sCJ6Swzt+Wwz+Ahy1ffv2KT4+3uxuAHVKSUlR+/btze5Gs0V9oymjvo8N9Y2mjPo+NtQ3mrI/qm/C+THwer1KTU1VaGioLBaL2d2pV0FBgeLj45WSkqKwsDCzu9NsNZf9aBiGCgsLFRcXJ6uVO1eOFvXdsjSX/Uh9Nwzqu2VpLvuR+m4Y1HfL0lz24+HWN8Paj4HVam1WZzbDwsKa9Ie2uWgO+zE8PNzsLjR71HfL1Bz2I/V97Kjvlqk57Efq+9hR3y1Tc9iPh1PfnJYDAAAAAMBkhHMAAAAAAExGOG8BnE6npk+fLqfTaXZXmjX2I5oiPpcNg/2IpojPZcNgP6Ip4nPZMPxtPzIhHAAAAAAAJuPKOQAAAAAAJiOcAwAAAABgMsI5AAAAAAAmI5wDAAAAAGAywrmfmz17thISEuRyuTRo0CB9//33Znep2Vm9erVGjx6tuLg4WSwWLV261OwuAZKo74ZAfaOpor6PHfWNpor6Pnb+Wt+Ecz+2YMEC3XHHHZo+fbrWrl2rxMREjRo1SpmZmWZ3rVkpLi5WYmKiZs+ebXZXgGrUd8OgvtEUUd8Ng/pGU0R9Nwx/rW8epebHBg0apJNPPlnPP/+8JMnr9So+Pl5TpkzRPffcY3LvmieLxaIlS5booosuMrsraOGo74ZHfaOpoL4bHvWNpoL6bnj+VN9cOfdTFRUVWrNmjYYPH169zGq1avjw4frmm29M7BmAY0V9A/6L+gb8F/WNP0I491NZWVnyeDxq3bq1z/LWrVsrPT3dpF4BaAjUN+C/qG/Af1Hf+COEcwAAAAAATEY491MxMTGy2WzKyMjwWZ6RkaE2bdqY1CsADYH6BvwX9Q34L+obf4Rw7qccDoeSkpK0YsWK6mVer1crVqzQqaeeamLPABwr6hvwX9Q34L+ob/wRu9kdwPFzxx13aNy4cTrppJM0cOBAzZo1S8XFxRo/frzZXWtWioqKtGPHjup/79q1S+vXr1dUVJQ6dOhgYs/QklHfDYP6RlNEfTcM6htNEfXdMPy2vg34teeee87o0KGD4XA4jIEDBxrffvut2V1qdlauXGlIqvUaN26c2V1DC0d9HzvqG00V9X3sqG80VdT3sfPX+uY55wAAAAAAmIx7zgEAAAAAMBnhHAAAAAAAkxHOAQAAAAAwGeEcAAAAAACTEc4BAAAAADAZ4RwAAAAAAJMRzgEAAAAAMBnhHAAAAAAAkxHOAQAAAAAwGeEcAAAAAACTEc4BAAAAADAZ4RwAAAAAAJMRzgEAAAAAMBnhHAAAAAAAkxHOAQAAAAAwGeEcAAAAAACTEc4BAAAAADAZ4RwAAAAAAJMRzgEAAAAAMBnhHAAAAAAAkxHOAQAAAAAwGeEcAAAAAACTEc4BAAAAADAZ4RwAAD+UkJCga6+9tsHWd+211yohIaHB1gcAAHwRzgEAOA5ee+01WSyW6pfL5VL37t01efJkZWRkmN29OqWmpurBBx/U+vXrze4KAAAtjt3sDgAA4M8efvhhderUSWVlZfryyy/1wgsv6KOPPtKmTZsUFBRkdvd8pKam6qGHHlJCQoL69+/v87OXXnpJXq/XnI4BANACEM4BADiOzjnnHJ100kmSpOuvv17R0dH65z//qffee09XXnllrfbFxcUKDg5u7G7+oYCAALO7AACAX2NYOwAAjejss8+WJO3atUvXXnutQkJCtHPnTp177rkKDQ3V1VdfLUnyer2aNWuW+vTpI5fLpdatW2vixInKzc31WZ9hGJoxY4bat2+voKAgDR06VD///HOdvzsvL0+33367EhIS5HQ61b59e11zzTXKysrSqlWrdPLJJ0uSxo8fXz0c/7XXXpNU9z3nxcXFuvPOOxUfHy+n06kePXro73//uwzD8GlnsVg0efJkLV26VCeccIKcTqf69OmjZcuWHevuBADAb3DlHACARrRz505JUnR0tCTJ7XZr1KhRGjx4sP7+979XD3WfOHGiXnvtNY0fP1633nqrdu3apeeff17r1q3TV199VX0l+4EHHtCMGTN07rnn6txzz9XatWs1cuRIVVRU+PzeoqIinXHGGfrll180YcIEnXjiicrKytL777+vffv2qVevXnr44Yf1wAMP6MYbb9QZZ5whSTrttNPq3A7DMHTBBRdo5cqVuu6669S/f3998sknmjp1qvbv36+nn37ap/2XX36pxYsX65ZbblFoaKieffZZjRkzRnv37q3eFwAAtGgGAABocPPmzTMkGcuXLzcOHDhgpKSkGO+++64RHR1tBAYGGvv27TPGjRtnSDLuuecen/d+8cUXhiTj7bff9lm+bNkyn+WZmZmGw+EwzjvvPMPr9Va3u++++wxJxrhx46qXPfDAA4YkY/HixbX6evC9P/zwgyHJmDdvXq0248aNMzp27Fj976VLlxqSjBkzZvi0Gzt2rGGxWIwdO3ZUL5NkOBwOn2UbNmwwJBnPPfdcPXsQAICWhWHtAAAcR8OHD1dsbKzi4+N1xRVXKCQkREuWLFG7du2q29x8880+7/nPf/6j8PBwjRgxQllZWdWvpKQkhYSEaOXKlZKk5cuXq6KiQlOmTJHFYql+/2233VarH4sWLVJiYqIuvvjiWj/77XsP10cffSSbzaZbb73VZ/mdd94pwzD08ccf19oPXbp0qf53v379FBYWpuTk5CP+3QAA+COGtQMAcBzNnj1b3bt3l91uV+vWrdWjRw9ZrTXnxu12u9q3b+/znu3btys/P1+tWrWqc52ZmZmSpD179kiSunXr5vPz2NhYRUZG+izbuXOnxowZc8zbc9CePXsUFxen0NBQn+W9evXy6dtBHTp0qLWOyMjIWvfQAwDQUhHOAQA4jgYOHFg9W3tdnE6nT1iXqiaDa9Wqld5+++063xMbG9ugfWwMNputzuXGIZPHAQDQUhHOAQBoYrp06aLly5fr9NNPV2BgYL3tOnbsKKnqSnvnzp2rlx84cKDWFekuXbpo06ZNv/t7j2R4e8eOHbV8+XIVFhb6XD3fsmWLT98AAMDh4Z5zAACamMsuu0wej0ePPPJIrZ+53W7l5eVJqrqPOyAgQM8995zPFehZs2bVet+YMWO0YcMGLVmypNbPDr734PPVD67/95x77rnyeDx6/vnnfZY//fTTslgsOuecc/5wHQAAoAZXzgEAaGKGDBmiiRMnaubMmVq/fr1GjhypgIAAbd++Xf/5z3/0zDPPaOzYsYqNjdVdd92lmTNn6vzzz9e5556rdevW6eOPP1ZMTIzPOqdOnaqFCxfq0ksv1YQJE5SUlKScnBy9//77mjNnjhITE9WlSxdFRERozpw5Cg0NVXBwsAYNGqROnTrV6uPo0aM1dOhQTZs2Tbt371ZiYqI+/fRTvffee7rtttt8Jn8DAAB/jHAOAEATNGfOHCUlJWnu3Lm67777ZLfblZCQoD//+c86/fTTq9vNmDFDLpdLc+bM0cqVKzVo0CB9+umnOu+883zWFxISoi+++ELTp0/XkiVL9Prrr6tVq1YaNmxY9YR0AQEBev3113Xvvffqpptuktvt1rx58+oM51arVe+//74eeOABLViwQPPmzVNCQoKeeuop3Xnnncd35wAA4IcsBjOxAAAAAABgKu45BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZIRzAAAAAABMRjgHAAAAAMBkhHMAAAAAAExGOAcAAAAAwGSEcwAAAAAATEY4BwAAAADAZP8Pe2NHTtX0+CcAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1040x520 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ncols = 4\n",
    "nrows = 2\n",
    "\n",
    "fig, axes = plt.subplots(nrows=nrows, ncols=ncols, figsize=(ncols*2.6, nrows*2.6))\n",
    "axes = axes.flatten()\n",
    "\n",
    "fig.suptitle('Model comparison - Optimized Threshold (ROC_AUC) Normalized')\n",
    "fig.supylabel('True label')\n",
    "fig.supxlabel('Predction')\n",
    "\n",
    "for i, (model_name, model) in enumerate(zip(results['Model'], [lr_clf, dt_clf, voting_clf, bagging_clf, rf_clf, adaboost_clf, gb_clf, stacking_clf])):\n",
    "\n",
    "    disp = ConfusionMatrixDisplay.from_predictions(\n",
    "        test_df['target'],\n",
    "        preds[model],\n",
    "        normalize='true',\n",
    "        ax=axes[i],\n",
    "        colorbar=False\n",
    "    )\n",
    "\n",
    "    axes[i].set_title(model_name)\n",
    "    axes[i].set_xlabel('')\n",
    "    axes[i].set_ylabel('')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "6479f4e5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA+cAAAIMCAYAAACXCt1JAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjcsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvTLEjVAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAxoZJREFUeJzs3XdUFFcbBvBnWWCXjjRpSlOjWCP2CjZi70ZjoqBRE3svKfaSaGyxGxM11sQSNfauMWrsvYKAihEp0hFY9n5/8O3osqCAmEV4fufs0b1zd+bOMO/MvDN3ZmRCCAEiIiIiIiIi0hsDfTeAiIiIiIiIqLhjck5ERERERESkZ0zOiYiIiIiIiPSMyTkRERERERGRnjE5JyIiIiIiItIzJudEREREREREesbknIiIiIiIiEjPmJwTERERERER6RmTcyIiIiIiIiI9Y3JORMWeTCbD5MmT8/y70NBQyGQyrFmzpsDbVFQcP34cMpkMx48f13dT3lvu7u4ICAgo0HHmd51/G5MnT4ZMJivw8UVFRRXYON/Gu2iPr68vfH1931gvr3E2e/ZslC9fHmq1+u0aSACA5cuXo3Tp0khNTdV3U4joPcfknIgKhTVr1kAmk0Emk+HUqVM6w4UQKFWqFGQyGdq0aaOHFhIB6enp+PHHH1GzZk1YWFjA3NwcNWvWxI8//oj09PR8j/f06dOYPHkyYmNjC66x7xlNgpmbD+VffHw8vv/+e4wbNw4GBi8PA7MuY0tLSzRu3Bh79uzJcVw3b97Ep59+ChcXFygUCjg7O6Nnz564efNmjr8JDg7GgAED4OnpCaVSCUtLS9SvXx8LFy5ESkpKvuZp7NixkMlk+Pjjj7Mdrlm3tm7dmu3wwYMHZ7teZWRkYPXq1fD19YWNjQ0UCgXc3d0RGBiICxcuSPUCAgKQlpaGFStW5Kv9REQahvpuABHRq5RKJTZu3IgGDRpolZ84cQKPHz+GQqHQU8soPxo1aoSUlBQYGxvruylvLSkpCa1bt8aJEyfQpk0bBAQEwMDAAPv378ewYcOwfft27NmzB2ZmZnke9+nTpzFlyhQEBATA2tpaa9jdu3e1kqiCkJKSAkPDwnUIUKFCBaxbt06rbMKECTA3N8fXX3+tp1YVPb/88gtUKhV69OihM6x58+bo1asXhBAICwvDsmXL0LZtW+zbtw/+/v5adbdv344ePXrAxsYGffv2hYeHB0JDQ/Hzzz9j69at2Lx5Mzp27Kj1mz179qBr165QKBTo1asXKlWqhLS0NJw6dQpjxozBzZs3sXLlyjzNjxACmzZtgru7O/78808kJCTAwsIi7wsmi5SUFHTq1An79+9Ho0aN8NVXX8HGxgahoaH4/fffsXbtWjx8+BCurq5QKpXo3bs35s2bhyFDhvAEEhHlW+HaMxNRsdeqVSts2bIFP/74o1bysHHjRvj4+BSaLqz0ei9evICxsTEMDAygVCr13ZwCMXLkSJw4cQKLFi3C4MGDpfIvv/wSS5YsweDBgzF69GgsW7asQKf7Lk5IFca/ScmSJfHpp59qlX333Xews7PTKX9barUaaWlphXI5vGurV69Gu3btsp33cuXKaS3rzp07w9vbGwsXLtRKzoODg/HZZ5/B09MTJ0+ehL29vTRs2LBhaNiwIT777DNcu3YNnp6eAICQkBB0794dbm5uOHr0KJycnKTfDBo0CEFBQa+9Sp+T48eP4/Hjxzh69Cj8/f2xfft29O7dO8/jyWrMmDHYv38/5s+fj+HDh2sNmzRpEubPn69V1q1bN8yePRvHjh1DkyZN3nr6RFQ8sVs7ERUqPXr0QHR0NA4dOiSVpaWlYevWrfjkk0+y/U1SUhJGjRqFUqVKQaFQ4IMPPsAPP/wAIYRWvdTUVIwYMQL29vawsLBAu3bt8Pjx42zHGR4ejj59+qBkyZJQKBSoWLEifvnll3zPV2xsLEaMGAF3d3coFAq4urqiV69eWicbnj17hr59+6JkyZJQKpWoWrUq1q5dqzUezX3uP/zwA5YsWQJPT0+YmpqiRYsWePToEYQQmDZtGlxdXWFiYoL27dsjJiZGaxzu7u5o06YNDh48iGrVqkGpVMLb2xvbt2/XqhcTE4PRo0ejcuXKMDc3h6WlJVq2bImrV69q1dN0Gd28eTO++eYbuLi4wNTUFPHx8dneC3v//n107twZjo6OUCqVcHV1Rffu3REXFyfVUalUmDZtGry8vKSupF999ZXOPZ2aeTl16hRq1aoFpVIJT09P/Prrr/n6O+Xk8ePH+Pnnn9GkSROtxFxj0KBB8PPzw6pVq7TWKZlMhsGDB2PDhg344IMPoFQq4ePjg5MnT0p1Jk+ejDFjxgAAPDw8pG7FoaGh0jy+es+55haQU6dOYejQobC3t4e1tTUGDBiAtLQ0xMbGolevXihRogRKlCiBsWPH6sTCq/eca9ap3HQh/+eff/DRRx/BysoKpqamaNy4Mf7++2+d5XHq1CnUrFkTSqUSXl5e77S7b2xsrNTjwMrKCoGBgUhOTtaZX83foWLFilAoFNi/fz+A3Mf6okWLULFiRZiamqJEiRKoUaMGNm7cmK/25Hb9zs7jx4/RoUMHmJmZwcHBASNGjMj1vc4hISG4du0amjVrlqv6FSpUgJ2dHYKDg7XK58yZg+TkZKxcuVIrMQcAOzs7rFixAklJSZg9e7ZUPnv2bCQmJuLnn3/WSsw1ypQpg2HDhuWqXa/asGEDvL294efnh2bNmmHDhg15HkdWjx8/xooVK9C8eXOdxBwA5HI5Ro8eDVdXV6nMx8cHNjY22Llz51tPn4iKL145J6JCxd3dHXXr1sWmTZvQsmVLAMC+ffsQFxeH7t2748cff9SqL4RAu3btcOzYMfTt2xfVqlXDgQMHMGbMGISHh2td3fj888+xfv16fPLJJ6hXrx6OHj2K1q1b67QhIiICderUkQ7o7e3tsW/fPvTt2xfx8fHZHqy9TmJiIho2bIjbt2+jT58+qF69OqKiorBr1y48fvwYdnZ2SElJga+vL4KCgjB48GB4eHhgy5YtCAgIQGxsrM5B64YNG5CWloYhQ4YgJiYGs2fPRrdu3dCkSRMcP34c48aNQ1BQEBYtWoTRo0frJBv379/Hxx9/jC+++AK9e/fG6tWr0bVrV+zfvx/NmzcHADx48AA7duxA165d4eHhgYiICKxYsQKNGzfGrVu34OzsrDXOadOmwdjYGKNHj0Zqamq2XdnT0tLg7++P1NRUDBkyBI6OjggPD8fu3bsRGxsLKysr6W+1du1adOnSBaNGjcI///yDWbNm4fbt2/jjjz+0xhkUFIQuXbqgb9++6N27N3755RcEBATAx8cHFStWzNPfKif79u1DRkYGevXqlWOdXr164dixY9i/fz8+//xzqfzEiRP47bffMHToUCgUCixduhQfffQRzp07h0qVKqFTp064d+8eNm3ahPnz58POzg4AdJKerDTLb8qUKTh79ixWrlwJa2trnD59GqVLl8bMmTOxd+9ezJkzB5UqVcqx7fb29jrdydPT0zFixAitv+HRo0fRsmVL+Pj4YNKkSTAwMMDq1avRpEkT/PXXX6hVqxYA4Pr162jRogXs7e0xefJkqFQqTJo0CSVLlnz9Qs6nbt26wcPDA7NmzcKlS5ewatUqODg44Pvvv9eqd/ToUfz+++8YPHgw7Ozs4O7unutY/+mnnzB06FB06dIFw4YNw4sXL3Dt2jX8888/OicNc9OevKzfr0pJSUHTpk3x8OFDDB06FM7Ozli3bh2OHj2aq2V1+vRpAED16tVzVT8uLg7Pnz+Hl5eXVvmff/4Jd3d3NGzYMNvfNWrUCO7u7lpXwv/88094enqiXr16uZp2bqSmpmLbtm0YNWoUgMyTu4GBgXj69CkcHR3zPd59+/ZBpVLhs88+y9Pvqlevnu3JKiKiXBNERIXA6tWrBQBx/vx5sXjxYmFhYSGSk5OFEEJ07dpV+Pn5CSGEcHNzE61bt5Z+t2PHDgFATJ8+XWt8Xbp0ETKZTAQFBQkhhLhy5YoAIAYOHKhV75NPPhEAxKRJk6Syvn37CicnJxEVFaVVt3v37sLKykpqV0hIiAAgVq9e/dp5mzhxogAgtm/frjNMrVYLIYRYsGCBACDWr18vDUtLSxN169YV5ubmIj4+Xmua9vb2IjY2Vqo7YcIEAUBUrVpVpKenS+U9evQQxsbG4sWLF1KZm5ubACC2bdsmlcXFxQknJyfx4YcfSmUvXrwQGRkZWu0NCQkRCoVCTJ06VSo7duyYACA8PT2lZZN12LFjx4QQQly+fFkAEFu2bMlxeWn+Vp9//rlW+ejRowUAcfToUZ15OXnypFT27NkzoVAoxKhRo3KcRl4NHz5cABCXL1/Osc6lS5cEADFy5EipDIAAIC5cuCCVhYWFCaVSKTp27CiVzZkzRwAQISEhOuN1c3MTvXv3lr5rYsXf319af4QQom7dukImk4kvvvhCKlOpVMLV1VU0btxYa5xZ1/msBg4cKORyubSs1Wq1KFu2rM40k5OThYeHh2jevLlU1qFDB6FUKkVYWJhUduvWLSGXy0VeDzsqVqyo03aNSZMmCQCiT58+WuUdO3YUtra2WmUAhIGBgbh586ZWeW5jvX379qJixYqvbWtu25OX9btx48Za86/ZTvz+++9SWVJSkihTpoxWnOXkm2++EQBEQkKCzjAAom/fviIyMlI8e/ZMXLhwQXz00UcCgJgzZ45ULzY2VgAQ7du3f+202rVrJwCI+Ph4ERcXl6vf5NXWrVsFAHH//n0hhBDx8fFCqVSK+fPna9XTbIdy2u4MGjRIa90cMWLEG+M9O/379xcmJiZ5+g0R0avYrZ2ICp1u3bohJSUFu3fvRkJCAnbv3p1jl/a9e/dCLpdj6NChWuWjRo2CEAL79u2T6gHQqZf1KrgQAtu2bUPbtm0hhEBUVJT08ff3R1xcHC5dupSn+dm2bRuqVq2q83AkAFK34b1798LR0VHrIU1GRkYYOnQoEhMTceLECa3fde3aVbrKDAC1a9cGAHz66ada9+rXrl0baWlpCA8P1/q9s7OzVnssLS3Rq1cvXL58GU+fPgWQea+z5kFkGRkZiI6Ohrm5OT744INsl0Hv3r1hYmLy2mWhafOBAwd0uvpqaP5WI0eO1CrXXB3Lel+qt7e31hU8e3t7fPDBB3jw4MFr25IXCQkJAPDaB01phsXHx2uV161bFz4+PtL30qVLo3379jhw4AAyMjLy3aa+fftqdTuvXbs2hBDo27evVCaXy1GjRo08LYtff/0VS5cuxezZs+Hn5wcAuHLlCu7fv49PPvkE0dHRUkwkJSWhadOmOHnyJNRqNTIyMnDgwAF06NABpUuXlsZZoUIFnQeKFZQvvvhC63vDhg0RHR2t83do3LgxvL29pe95iXVra2s8fvwY58+ff+v25HX9ftXevXvh5OSELl26SGWmpqbo37//G9sFANHR0TA0NIS5uXm2w3/++WfY29vDwcEBNWrUwJEjRzB27FittuYmFl4dHh8fL817QTyo7VUbNmxAjRo1UKZMGWn8rVu3fuuu7fltb4kSJZCSkpLjto2I6E2YnBNRoWNvb49mzZph48aN2L59OzIyMrQORl8VFhYGZ2dnnYOoChUqSMM1/xoYGOh0z/zggw+0vkdGRiI2Nla6l/LVT2BgIIDMe8PzIjg4GJUqVXptnbCwMJQtW1bnqdxZ50Pj1cQHeJn0lipVKtvy58+fa5WXKVNG537icuXKAYB0r7Narcb8+fNRtmxZKBQK2NnZwd7eHteuXdO6P1zDw8PjtfOoqTNy5EisWrUKdnZ28Pf3x5IlS7TGp/lbaQ64NRwdHWFtbf3GZQFkHiRnneesIiMj8fTpU+mTmJiYY13N+qVJTLKTU9JStmxZnbrlypVDcnIyIiMjX9vG18nLOvCmZaFx5coVfPHFF+jRo4dWQnb//n0AmSdgssbFqlWrkJqairi4OERGRiIlJSXbec4aawUl63IoUaIEAN11Puv6mZdYHzduHMzNzVGrVi2ULVsWgwYNyrH78pvak9f1+1VhYWHZxm5BLdv27dvj0KFD2LNnj/Te9uTkZK3tUm5i4dXhFhYWsLS0zNVv8iI2NhZ79+5F48aNERQUJH3q16+PCxcu4N69e/ked37bK/7/bAc+rZ2I8ov3nBNRofTJJ5+gX79+ePr0KVq2bKnzeql3Ra1WA8i8Ap3TE3+rVKnyn7TldeRyeZ7KRZYHguXGzJkz8e2336JPnz6YNm0abGxsYGBggOHDh0vL6VVvumquMXfuXAQEBGDnzp04ePAghg4dilmzZuHs2bNaD1jK7QFufue5Zs2aWonQpEmTpIekZaU5SXLt2jVUq1Yt2zrXrl0DAK2rs+9SXtaB3Pz9nz9/js6dO6NcuXJYtWqV1jDN33vOnDk5zr+5uXmuH0xWkHL798+6fuYl1itUqIC7d+9i9+7d2L9/P7Zt24alS5di4sSJmDJlSr7ao48EztbWFiqVKsfXjbm6ukoPi2vVqhXs7OwwePBg+Pn5oVOnTgAyT/Y4OTlJ63tOrl27BhcXFynRdXZ2xo0bNwpsXrZs2YLU1FTMnTsXc+fO1Rm+YcMG6W+jeTJ9Tu9RT05O1np6ffny5QFkPj8hp/U9O8+fP4epqWmut4VERFkxOSeiQqljx44YMGAAzp49i99++y3Hem5ubjh8+LDOweadO3ek4Zp/1Wo1goODta4y3b17V2t8mie5Z2Rk5PqJxm/i5eX1xoNSNzc3XLt2DWq1WusqVdb5KChBQUEQQmglCJorTe7u7gCArVu3ws/PDz///LPWb2NjY6WHluVX5cqVUblyZXzzzTc4ffo06tevj+XLl2P69OnS3+r+/ftSUgxkPqgvNja2wJbFhg0btA7WNa98yk7Lli0hl8uxbt26HB+s9uuvv8LQ0BAfffSRVrnmqvOr7t27B1NTU+mhb/q+0qZWq9GzZ0/Exsbi8OHDMDU11Rqu6XFiaWn52riwt7eHiYlJtvOcNdb0La+xbmZmho8//hgff/wx0tLS0KlTJ8yYMQMTJkzI0yvZ3mb9dnNzw40bN3RiN7fLVpN0hoSE5Ook44ABAzB//nx888036NixozTNNm3a4KeffsKpU6fQoEEDnd/99ddfCA0NxYABA6SyNm3aYOXKlThz5gzq1q2bq/a+zoYNG1CpUiVMmjRJZ9iKFSuwceNGKTnXLNOcltPdu3e1lrsm3tevX5+nh8KFhIRo/U2JiPKK3dqJqFAyNzfHsmXLMHnyZLRt2zbHeq1atUJGRgYWL16sVT5//nzIZDLpie+af7M+7X3BggVa3+VyOTp37oxt27Zlm1Dnpxty586dcfXq1Wyfwqy5mtaqVSs8ffpU60SESqXCokWLYG5ujsaNG+d5uq/z5MkTrfbEx8fj119/RbVq1aSnHMvlcp2rfVu2bNG5fz0v4uPjoVKptMoqV64MAwMD6aprq1atAOj+bebNmwcA2T5hPz/q16+PZs2aSZ/XJeelSpVCYGAgDh8+nO17zJcvX46jR4+ib9++Wlf/AeDMmTNa9+g/evQIO3fuRIsWLaSrrGZmZgAyT3zow5QpU3DgwAFs2rQp29sTfHx84OXlhR9++CHb7v+auJDL5fD398eOHTvw8OFDafjt27dx4MCBdzcD+ZCXWI+OjtYaZmxsDG9vbwghkJ6enqfpvs363apVKzx58gRbt26VyjSvNMsNTVJ84cKFXNU3NDTEqFGjcPv2ba1XhI0ZMwYmJiYYMGCAzrKJiYnBF198AVNTU+kVgQAwduxYmJmZ4fPPP0dERITOtIKDg7Fw4cJctevRo0c4efIkunXrhi5duuh8AgMDERQUhH/++QcA4OTkhGrVqmH9+vU6MXbx4kWcPXtW2kcAmfHer18/HDx4EIsWLdKZvlqtxty5c3VexXnp0qUCfRo9ERU/vHJORIVWTl1NX9W2bVv4+fnh66+/RmhoKKpWrYqDBw9i586dGD58uHTFr1q1aujRoweWLl2KuLg41KtXD0eOHEFQUJDOOL/77jscO3YMtWvXRr9+/eDt7Y2YmBhcunQJhw8f1nlv+JuMGTMGW7duRdeuXdGnTx/4+PggJiYGu3btwvLly1G1alX0798fK1asQEBAAC5evAh3d3ds3boVf//9NxYsWFDgD1IqV64c+vbti/Pnz6NkyZL45ZdfEBERgdWrV0t12rRpg6lTpyIwMBD16tXD9evXsWHDhtcmsW9y9OhRDB48GF27dkW5cuWgUqmwbt06KVECgKpVq6J3795YuXIlYmNj0bhxY5w7dw5r165Fhw4dpIeU/dfmz5+PO3fuYODAgdi/f790hfzAgQPYuXMnGjdunG332kqVKsHf31/rVWoAtLpDax4Y9/XXX6N79+4wMjJC27ZtpaT9Xbp+/TqmTZuGRo0a4dmzZ1i/fr3W8E8//RQGBgZYtWoVWrZsiYoVKyIwMBAuLi4IDw/HsWPHYGlpiT///FOar/3796Nhw4YYOHCgdJKpYsWKb+wK/V/Lbay3aNECjo6OqF+/PkqWLInbt29j8eLFaN26dZ5j823W7379+mHx4sXo1asXLl68CCcnJ6xbt06np0NOPD09UalSJRw+fBh9+vTJ1W8CAgIwceJEfP/99+jQoQOAzOcorF27Fj179kTlypXRt29feHh4IDQ0FD///DOioqKwadMmrWd8eHl5YePGjfj4449RoUIF9OrVC5UqVUJaWhpOnz4tvToyNzZu3Ci9RjM7rVq1gqGhITZs2CA9LHPevHnw9/dHtWrVEBAQAGdnZ9y+fRsrV66Ek5MTJkyYoDWOuXPnIjg4GEOHDsX27dvRpk0blChRAg8fPsSWLVtw584ddO/eXap/8eJFxMTEoH379rmaByKibP3Xj4cnIsrOq69Se52sr1ITQoiEhAQxYsQI4ezsLIyMjETZsmXFnDlztF75JIQQKSkpYujQocLW1laYmZmJtm3bikePHmX7WqmIiAgxaNAgUapUKWFkZCQcHR1F06ZNxcqVK6U6uX2VmhBCREdHi8GDBwsXFxdhbGwsXF1dRe/evbVe4RQRESECAwOFnZ2dMDY2FpUrV9YZt2aar77aSIicXxWU3XLVLMMDBw6IKlWqCIVCIcqXL6/z2xcvXohRo0YJJycnYWJiIurXry/OnDmj83qn172mKOur1B48eCD69OkjvLy8hFKpFDY2NsLPz08cPnxY63fp6eliypQpwsPDQxgZGYlSpUqJCRMmaL0S7tV5ySprGwtKamqqmD9/vvDx8RFmZmbC1NRUVK9eXSxYsECkpaXp1AcgBg0aJNavXy/Kli0rFAqF+PDDD7N95dW0adOEi4uLMDAw0HqtWk6vUssaK5pXeUVGRmqV9+7dW5iZmem0S7POa/5GOX1edfnyZdGpUydha2srFAqFcHNzE926dRNHjhzRqnfixAnh4+MjjI2Nhaenp1i+fLnUvrzIzavUss6vZvm8+lo6zd8hO7mJ9RUrVohGjRpJ8+3l5SXGjBkj4uLi8tWe3K7f2a3HYWFhol27dsLU1FTY2dmJYcOGif379+fqVWpCCDFv3jxhbm6u89rD1y2jyZMnZzv+a9euiR49eggnJydp2fXo0UNcv349x+nfu3dP9OvXT7i7uwtjY2NhYWEh6tevLxYtWqQz/zmpXLmyKF269Gvr+Pr6CgcHB61XS549e1a0adNGlChRQhgaGgoXFxfx+eefi8ePH2c7DpVKJVatWiUaNmworKyshJGRkXBzcxOBgYE6r1kbN26cKF26tM5+h4goL2RC5OMpQURE9N5yd3dHpUqVsHv3bn03pciTyWQYNGiQzm0XRPoSFxcHT09PzJ49W+u1e5R/qampcHd3x/jx4zFs2DB9N4eI3mO855yIiIiomLCyssLYsWMxZ86cbN+6QHm3evVqGBkZ6bzjnogor3jlnIiomOGV8/8Or5wT5U1cXFyOrzzT0Dy0koioqOED4YiIiIioUBg2bBjWrl372jq8rkRERRWvnBMRERFRoXDr1i08efLktXVy8156IqL3EZNzIiIiIiIiIj3jA+GIiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50RERERERER6xuSciIiIiIiISM+YnBMRERERERHpGZNzIiIiIiIiIj1jck5ERERERESkZ0zOiYiIiIiIiPSMyTkRERERERGRnjE5JyIiIiIiItIzJudEREREREREesbknIiIiIiIiEjPmJwTERERERER6RmT8yLC19cXvr6+BTY+d3d3BAQEFNj4CJDJZJg8ebK+m0H0WqGhoZDJZFizZk2eflfQ2yAier8EBATA3d1d380gIj1h7lAwmJwXsDVr1kAmk+HChQv6bsobnT59GpMnT0ZsbOw7nY67uztkMpn0MTMzQ61atfDrr7++0+kSvY802xDNR6lUwtnZGf7+/vjxxx+RkJCg7yYWGlm3LTl98nqigaioaNeuHUxNTV+73ejZsyeMjY0RHR39xvE9efIEkydPxpUrVwqwlUT0OlmPC2QyGRwcHODn54d9+/bpu3lUwAz13QAqGAcPHszzb06fPo0pU6YgICAA1tbWWsPu3r0LA4OCO3dTrVo1jBo1CgDw77//YtWqVejduzdSU1PRr1+/AptOYZaSkgJDQ4Yc5c7UqVPh4eGB9PR0PH36FMePH8fw4cMxb9487Nq1C1WqVHkn03Vzc0NKSgqMjIzy9Lv8bIPe1oIFC5CYmCh937t3LzZt2oT58+fDzs5OKq9Xr95/3jaiwqBnz574888/8ccff6BXr146w5OTk7Fz50589NFHsLW1feP4njx5gilTpsDd3R3VqlXTGvbTTz9BrVYXVNOJKAvNcYEQAhEREVizZg1atWqFP//8E23atNF38wo8dyiumCkUEcbGxgU6PoVCUaDjc3Fxwaeffip9DwgIgKenJ+bPn/+fJ+dJSUkwMzP7T6cJAEql8j+fJr2/WrZsiRo1akjfJ0yYgKNHj6JNmzZo164dbt++DRMTkwKfruZqfV4V9DYoNzp06KD1/enTp9i0aRM6dOjw2u61+toGEP3X2rVrBwsLC2zcuDHb5Hznzp1ISkpCz54933paeT2hR0R5k/W4oG/fvihZsiQ2bdpUKJLzgs4diiue3tCTy5cvo2XLlrC0tIS5uTmaNm2Ks2fP6tS7du0aGjduDBMTE7i6umL69OlYvXo1ZDIZQkNDpXrZ3e+5aNEiVKxYEaampihRogRq1KiBjRs3AgAmT56MMWPGAAA8PDykbjKacWZ330hsbCxGjBgBd3d3KBQKuLq6olevXoiKisrz/Nvb26N8+fIIDg7WKler1ViwYAEqVqwIpVKJkiVLYsCAAXj+/LlOvcmTJ8PZ2Rmmpqbw8/PDrVu3dNqt6Qp04sQJDBw4EA4ODnB1dZWG79u3Dw0bNoSZmRksLCzQunVr3Lx5U2taT58+RWBgIFxdXaFQKODk5IT27dtrLf8LFy7A398fdnZ2MDExgYeHB/r06aM1nuzuOc/NeqCZh7///hsjR46Evb09zMzM0LFjR0RGRuZ2kVMR0KRJE3z77bcICwvD+vXrtYbduXMHXbp0gY2NDZRKJWrUqIFdu3bpjONNcZzdPee5iYHstkHPnj2TDh6USiWqVq2KtWvXatXRTO+HH37AypUr4eXlBYVCgZo1a+L8+fNvt8CQeSLQ3NwcwcHBaNWqFSwsLKREJLfbGyB32wqiwsbExASdOnXCkSNH8OzZM53hGzduhIWFBdq1a4cHDx6ga9eusLGxgampKerUqYM9e/ZIdY8fP46aNWsCAAIDA3VuG8l6z3leY3vLli3w9vaGUqlEpUqV8Mcff/A+dqLXsLa2homJiVavzB9++AH16tWDra0tTExM4OPjg61bt+r8NiUlBUOHDoWdnZ20DQgPD8/2WPX48eOoUaMGlEolvLy8sGLFCkyePBkymUyrXk7H4Lk5fs3tcX1xwCvnenDz5k00bNgQlpaWGDt2LIyMjLBixQr4+vrixIkTqF27NgAgPDwcfn5+kMlkmDBhAszMzLBq1apcnZn66aefMHToUHTp0gXDhg3DixcvcO3aNfzzzz/45JNP0KlTJ9y7d0+nC6i9vX2240tMTETDhg1x+/Zt9OnTB9WrV0dUVBR27dqFx48fa3UhzQ2VSoXHjx+jRIkSWuUDBgzAmjVrEBgYiKFDhyIkJASLFy/G5cuX8ffff0tn5idMmIDZs2ejbdu28Pf3x9WrV+Hv748XL15kO72BAwfC3t4eEydORFJSEgBg3bp16N27N/z9/fH9998jOTkZy5YtQ4MGDXD58mXpgKBz5864efMmhgwZAnd3dzx79gyHDh3Cw4cPpe8tWrSAvb09xo8fD2tra4SGhmL79u2vXQa5XQ80hgwZghIlSmDSpEkIDQ3FggULMHjwYPz22295Wvb0fvvss8/w1Vdf4eDBg1Kvk5s3b6J+/fpwcXHB+PHjYWZmht9//x0dOnTAtm3b0LFjRwD5j+M3xUB2UlJS4Ovri6CgIAwePBgeHh7YsmULAgICEBsbi2HDhmnV37hxIxISEjBgwADIZDLMnj0bnTp1woMHD976ipxKpYK/vz8aNGiAH374AaampgByv73J7baCqDDq2bMn1q5di99//x2DBw+WymNiYnDgwAH06NED8fHxqFevHpKTkzF06FDY2tpi7dq1aNeuHbZu3YqOHTuiQoUKmDp1KiZOnIj+/fujYcOGAN5820huYnvPnj34+OOPUblyZcyaNQvPnz9H37594eLi8u4WDNF7Ji4uDlFRURBC4NmzZ1i0aBESExO1eqYuXLgQ7dq1Q8+ePZGWlobNmzeja9eu2L17N1q3bi3VCwgIwO+//47PPvsMderUwYkTJ7SGa1y+fBkfffQRnJycMGXKFGRkZGDq1Kk55gvZyc3xa16P64s0QQVq9erVAoA4f/58jnU6dOggjI2NRXBwsFT25MkTYWFhIRo1aiSVDRkyRMhkMnH58mWpLDo6WtjY2AgAIiQkRCpv3LixaNy4sfS9ffv2omLFiq9t65w5c3TGo+Hm5iZ69+4tfZ84caIAILZv365TV61Wv3Y6bm5uokWLFiIyMlJERkaK69evi88++0wAEIMGDZLq/fXXXwKA2LBhg9bv9+/fr1X+9OlTYWhoKDp06KBVb/LkyQKAVrs1f48GDRoIlUollSckJAhra2vRr18/rXE8ffpUWFlZSeXPnz8XAMScOXNynL8//vjjjX9zIYQAICZNmiR9z+16oJmHZs2aaS3rESNGCLlcLmJjY187XXq/5GYbYmVlJT788EPpe9OmTUXlypXFixcvpDK1Wi3q1asnypYtK5XlJo5DQkIEALF69WohRO5iQAjdbdCCBQsEALF+/XqpLC0tTdStW1eYm5uL+Ph4renZ2tqKmJgYqe7OnTsFAPHnn3++drqvym6b1rt3bwFAjB8/Xqtubrc3ud1WEBVWKpVKODk5ibp162qVL1++XAAQBw4cEMOHDxcAxF9//SUNT0hIEB4eHsLd3V1kZGQIIYQ4f/681vbhVb179xZubm7S97zEduXKlYWrq6tISEiQyo4fPy4AaI2TqDjSHBdk/SgUCrFmzRqtusnJyVrf09LSRKVKlUSTJk2ksosXLwoAYvjw4Vp1AwICdI5V27ZtK0xNTUV4eLhUdv/+fWFoaCiyppFZc4fcHr/m5bi+OGC39v9YRkYGDh48iA4dOsDT01Mqd3JywieffIJTp04hPj4eALB//37UrVtX66ErNjY2ubo3zNraGo8fPy6QbqEAsG3bNlStWlW6AveqrN1asnPw4EHY29vD3t4elStXxrp16xAYGIg5c+ZIdbZs2QIrKys0b94cUVFR0sfHxwfm5uY4duwYAODIkSNQqVQYOHCg1jSGDBmS4/T79esHuVwufT906BBiY2PRo0cPrWnJ5XLUrl1bmpaJiQmMjY1x/PjxbLu6ApAeprd7926kp6e/cVkAeVsPNPr376+1rBs2bIiMjAyEhYXlappUdJibm0tPX46JicHRo0fRrVs3JCQkSOtydHQ0/P39cf/+fYSHhwPIXxznJgays3fvXjg6OqJHjx5SmZGREYYOHYrExEScOHFCq/7HH3+s1ZNGc1XuwYMHuZ7m63z55Zda33O7vcnttoKosJLL5ejevTvOnDmjdSvKxo0bUbJkSTRt2hR79+5FrVq10KBBA2m4ubk5+vfvj9DQUNy6dSvf039TbD958gTXr19Hr169YG5uLtVr3LgxKleunO/pEhU1S5YswaFDh3Do0CGsX78efn5++Pzzz7V6ar76LJrnz58jLi4ODRs2xKVLl6Ty/fv3A8Abj6MzMjJw+PBhdOjQAc7OzlJ5mTJl0LJly1y3+03Hr/k5ri/KmJz/xyIjI5GcnIwPPvhAZ1iFChWgVqvx6NEjAEBYWBjKlCmjUy+7sqzGjRsHc3Nz1KpVC2XLlsWgQYPw999/57vdwcHBqFSpUr5/X7t2bRw6dAj79+/HDz/8AGtrazx//lzrIVL3799HXFwcHBwcpERe80lMTJTul9MEc9blYGNjo9NNXsPDw0Pr+/379wFk3sObdVoHDx6UpqVQKPD9999j3759KFmyJBo1aoTZs2fj6dOn0rgaN26Mzp07Y8qUKbCzs0P79u2xevVqpKam5rg88rIeaJQuXVrru2Ze85IwUdGQmJgICwsLAEBQUBCEEPj222911uVJkyYBgLQ+5yeOcxMD2QkLC0PZsmV1ntxaoUIFafir3uX6bWhoqPWsCSD325vcbiuICjPNSX3Nc2ceP36Mv/76C927d4dcLkdYWFiO+yNAN17z4k2xndM+PacyouKqVq1aaNasGZo1a4aePXtiz5498Pb2xuDBg5GWlgYg80JRnTp1oFQqYWNjA3t7eyxbtgxxcXHSeMLCwmBgYKBzbJw13p49e4aUlJS3js38bgNed1xflPGe8yKqQoUKuHv3Lnbv3o39+/dj27ZtWLp0KSZOnIgpU6b85+2xs7NDs2bNAAD+/v4oX7482rRpg4ULF2LkyJEAMh8G4eDggA0bNmQ7jrzc35JV1qdaa173sm7dOjg6OurUf/XhGsOHD0fbtm2xY8cOHDhwAN9++y1mzZqFo0eP4sMPP4RMJsPWrVtx9uxZ/Pnnnzhw4AD69OmDuXPn4uzZs1pXAt7Gq1f+XyWEKJDx0/vh8ePHiIuLk3ZimnV59OjR8Pf3z/Y3b3uA+6YYKAjvcv1WKBQ6Jwlyu73Jy7aCqLDy8fFB+fLlsWnTJnz11VfYtGkThBAF8pT2N+G+i+jdMDAwgJ+fHxYuXIj79+8jJiYG7dq1Q6NGjbB06VI4OTnByMgIq1evlk7M6QO3AXnDo4r/mL29PUxNTXH37l2dYXfu3IGBgQFKlSoFIPN9w0FBQTr1sivLjpmZGT7++GN8/PHHSEtLQ6dOnTBjxgxMmDABSqUyV93RNby8vHDjxo1c13+T1q1bo3Hjxpg5cyYGDBgAMzMzeHl54fDhw6hfv/5rXxHl5uYGIHM5vHrWLzo6OtdX2by8vAAADg4O0kmDN9UfNWoURo0ahfv376NatWqYO3eu1hOz69Spgzp16mDGjBnYuHEjevbsic2bN+Pzzz/XGV9e1gOiV61btw4ApERcc1uEkZHRG9flt4nj3MTAq9zc3HDt2jWo1WqtxPjOnTvScH3K7fYmr9sKosKqZ8+e+Pbbb3Ht2jVs3LgRZcuWlZ6+7ubmluP+SDMcyN1tbHn16j49q9we7xAVVyqVCkBmj7pt27ZBqVTiwIEDWg+PXr16tdZv3NzcoFarERISgrJly0rlWePNwcEBSqXyncdmQRzXFyXs1v4fk8vlaNGiBXbu3Kl171dERAQ2btyIBg0awNLSEkDmwfeZM2dw5coVqV5MTEyOV3peFR0drfXd2NgY3t7eEEJI90Vr3vMbGxv7xvF17twZV69exR9//KEzLL9nvsaNG4fo6Gj89NNPAIBu3bohIyMD06ZN06mrUqmkdjZt2hSGhoZYtmyZVp3Fixfnetr+/v6wtLTEzJkzs71PXPOKh+TkZJ0nRXp5ecHCwkLqtv78+XOdZaB5TkBOXdvzsh4QaRw9ehTTpk2Dh4eHdMXLwcEBvr6+WLFiBf7991+d37z6upL8xHFuYiA7rVq1wtOnT7WexqpSqbBo0SKYm5ujcePGr5/Zdyy325vcbiuICjvNNmPixIm4cuWK1lXzVq1a4dy5czhz5oxUlpSUhJUrV8Ld3R3e3t4A8nbckFvOzs6oVKkSfv31VyQmJkrlJ06cwPXr1wtsOkRFTXp6Og4ePAhjY2NUqFABcrkcMpkMGRkZUp3Q0FDs2LFD63eak/tLly7VKl+0aJHWd7lcjmbNmmHHjh148uSJVB4UFIR9+/YV2HwUxHF9UcIr5+/IL7/8Ij1w4VXDhg3D9OnTcejQITRo0AADBw6EoaEhVqxYgdTUVMyePVuqO3bsWKxfvx7NmzfHkCFDpFeplS5dGjExMa89g92iRQs4Ojqifv36KFmyJG7fvo3FixejdevW0r2qPj4+AICvv/4a3bt3h5GREdq2bSvtfF81ZswYbN26FV27dkWfPn3g4+ODmJgY7Nq1C8uXL0fVqlXzvIxatmyJSpUqYd68eRg0aBAaN26MAQMGYNasWbhy5QpatGgBIyMj3L9/H1u2bMHChQvRpUsXlCxZEsOGDcPcuXPRrl07fPTRR7h69Sr27dsHOzu7XJ3Zt7S0xLJly/DZZ5+hevXq6N69O+zt7fHw4UPs2bMH9evXx+LFi3Hv3j00bdoU3bp1g7e3NwwNDfHHH38gIiIC3bt3BwCsXbsWS5cuRceOHeHl5YWEhAT89NNPsLS0RKtWrXJsQ27XAyqe9u3bhzt37kClUiEiIgJHjx7FoUOH4Obmhl27dkGpVEp1lyxZggYNGqBy5cro168fPD09ERERgTNnzuDx48e4evUqgPzFcW5iIDv9+/fHihUrEBAQgIsXL8Ld3R1bt27F33//jQULFkjbIX3J7fYmt9sKosLOw8MD9erVw86dOwFAKzkfP348Nm3ahJYtW2Lo0KGwsbHB2rVrERISgm3btkm9X7y8vGBtbY3ly5fDwsICZmZmqF27ts69q3k1c+ZMtG/fHvXr10dgYCCeP3+OxYsXo1KlSloJO1FxpjkuADLvB9+4cSPu37+P8ePHw9LSEq1bt8a8efPw0Ucf4ZNPPsGzZ8+wZMkSlClTBteuXZPG4+Pjg86dO2PBggWIjo6WXqV27949ANo9ZCZPnoyDBw+ifv36+PLLL5GRkSHF5qsXD99GQRzXFyn6ekx8UZXT6w40n0ePHgkhhLh06ZLw9/cX5ubmwtTUVPj5+YnTp0/rjO/y5cuiYcOGQqFQCFdXVzFr1izx448/CgDi6dOnUr2srzFasWKFaNSokbC1tRUKhUJ4eXmJMWPGiLi4OK3xT5s2Tbi4uAgDAwOtVxBlfR2CEJmvcRs8eLBwcXERxsbGwtXVVfTu3VtERUW9dpm4ubmJ1q1bZztszZo1Oq9lWblypfDx8REmJibCwsJCVK5cWYwdO1Y8efJEqqNSqcS3334rHB0dhYmJiWjSpIm4ffu2sLW1FV988YXO3yOn11IdO3ZM+Pv7CysrK6FUKoWXl5cICAgQFy5cEEIIERUVJQYNGiTKly8vzMzMhJWVlahdu7b4/fffpXFcunRJ9OjRQ5QuXVooFArh4OAg2rRpI41DA1leT6H57ZvWg5zm4dixYwKAOHbsWLbzRu+nrNsQY2Nj4ejoKJo3by4WLlwovYIsq+DgYNGrVy/h6OgojIyMhIuLi2jTpo3YunWrVr03xXHWV6nlJgaE0N0GCSFERESECAwMFHZ2dsLY2FhUrlxZ5xVMmull96q27GLmdXJ6lZqZmVmOv8nN9kaIN28riN4HS5YsEQBErVq1dIYFBweLLl26CGtra6FUKkWtWrXE7t27dert3LlTeHt7S69S0sR0Tq9Sy21sb968WZQvX14oFApRqVIlsWvXLtG5c2dRvnz5t5pnovdddrmFUqkU1apVE8uWLdN6TdnPP/8sypYtKxQKhShfvrxYvXq1mDRpks5rz5KSksSgQYOEjY2NMDc3Fx06dBB3794VAMR3332nVffIkSPiww8/FMbGxsLLy0usWrVKjBo1SiiVSq16Ob1KLTfHr7k9ri8OZELwbvz3zfDhw7FixQokJibm+JCF4ig2NhYlSpTA9OnT8fXXX+u7OURERPQWqlWrBnt7exw6dEjfTSEq8q5cuYIPP/wQ69evf+PDIjt06ICbN29KbzR5F4rrcT3vOS/kUlJStL5HR0dj3bp1aNCgQbFOzLMuFwBYsGABAMDX1/e/bQwRERHlW3p6uvRgK43jx4/j6tWr3KcTvQM5HUcbGBigUaNGr617//597N27t0Bjk8f1L/Ge80Kubt268PX1RYUKFRAREYGff/4Z8fHx+Pbbb/XdNL367bffsGbNGrRq1Qrm5uY4deoUNm3ahBYtWqB+/fr6bh4RERHlUnh4OJo1a4ZPP/0Uzs7OuHPnDpYvXw5HR0d88cUX+m4eUZEze/ZsXLx4EX5+fjA0NMS+ffuwb98+9O/fX+dtQZ6enggICICnpyfCwsKwbNkyGBsbY+zYsQXWHh7Xv8TkvJBr1aoVtm7dipUrV0Imk6F69er4+eefdc5qFTdVqlSBoaEhZs+ejfj4eOlhEtOnT9d304iIiCgPSpQoAR8fH6xatQqRkZEwMzND69at8d1338HW1lbfzSMqcurVq4dDhw5h2rRpSExMROnSpTF58uRsu49/9NFH2LRpE54+fQqFQoG6deti5syZWq9he1s8rn+J95wTERERERER6RnvOSciIiIiIiLSMybnRERERERERHrGe87fglqtxpMnT2BhYQGZTKbv5hABAIQQSEhIgLOzMwwMeP4tvxjfVBgxvgsG45sKI8Z3wWB8U2GU2/hmcv4Wnjx5ovNEQ6LC4tGjR3B1ddV3M95bjG8qzBjfb4fxTYUZ4/vtML6pMHtTfDM5fwsWFhYAgLBL7rA05xnOt9XzgZ++m1AkpCel4VDnddL6SfnD+C5YXRo00XcTigSVOg0non5lfL8lxnfBqjf/c303oUjISHuBez9PZXy/JcZ3wWo0q6++m1AkZKS9wK0N094Y30zO34Kmq4yluQEsLRj8b8vIzFjfTShS2JXr7TC+C5ahAeO7IDG+3w7ju2DJFUp9N6FIYXy/HcZ3wZIbM74L0pvim2ssERERERERkZ4xOSciIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50RERERERER6xuSciIiIiIiISM+YnBMRERERERHpGZNzIiIiIiIiIj1jck5ERERERESkZ0zOiYiIiIiIiPSMyTkRERERERGRnjE5JyIiIiIiItIzJudEREREREREesbknIiIiIiIiEjPmJwTERERERER6RmTcyIiIiIiIiI9Y3JOREREREREpGdMzomIiIiIiIj0jMk5ERERERERkZ4xOSciIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeGeq7AZQ718+aYctSB9y/boqYCCNM+jkE9VrGAQBU6cCa751w/qgl/g0zhpmlGh82TEDfr57A1lEljWPjwpI4d9gSD26awNBYYPud6zrT8XeuplM2YWkofDvEvqtZ0zuRrEbaqmRk/JUK8VwNg7KGMB5qDnkFIwCA6kQq0nemQH1PBcQLKH8uAXlZ3dDJuJGOtJ+SoL6dDhjIYFDGEMq5VpApZP/1LNF75nXxDQDPIw3x8wxnXDxhgaQ4OSrVScSg6Y/h4pkm1Ul7IcPKKc44vqsE0lNl8PFNwJBZj1HC/uU2oLjFd7c+IajX5Blc3ZOQlmqA21et8cvCsggPMwMAmFum49Mvg1G9TjTsHV8g7rkxzhy3x7qlXkhONNIaV7O2T9Dx0zC4uCUjOUmOU4dKYul3FfQxW/SeeVN8/zC8NA79bqP1Gx/feMzc+ED6Hv9cjqXfuOCfQ1aQGQANWsXiy2nhMDFTAwAeBSnw43hXPLynRFKCHLYl0+HX8Tk+HfkUhtqrcpHiYJ6I4b5nUd/zIZSGKjyKtcLEvX649dQBhgYZGNzwHBp4PYSrVTwSUo3xT5grFp6og8hEM2kcbiViMcLvDKq5PIWRPAP3I22x5K9aOP/QRY9zRu+LgohvjbRUGYa1LocHt0yw9OBdeFVKySx/IcOP40vh/jUTPLyvRO1m8Zi8OuTdzpie9fc9jwG+F7XKQqOs0XlxdwCAa4k4DG9xBtVKP4WRYQbOBJXC7L0NEJNkKtW3NHmBsS1PoeEHYRBChiO3PPHD/vpISSvcG0Um5++JF8kG8KyYAv8eMZja10NrWGqKAYKum+KT4RHw9E5BYpwcyya6YFKAJxbvvyfVU6XJ0KhtLCrUSMKBTbY5TmvU/Ieo4RcvfTe3zCj4GSpEUr9PhDpEBcXXFpDZyaE6+AIvRsbB5NcSMLCXQ7wQkFcxgmETBdJmJ2Y7jowb6XgxJg5GPU2hGG4OyAF1kApgXk658Lr4FgKY0scDckOByasfwNRcje0r7TH+4zL46cQdKE0zD86XT3bBucOW+GZFKMwsM7Dka1dM7euO+buCtMZXnOK7UvXn2P1bKdy7aQm5oUDvwUGYsewSBnSqh9QXctjap8LWPhWr5pfDwwdmKOn0AoO/vg1b+1TMHFNVGk/HT8PQ8bMw/DK/LO7csILSJAMlnVP0OGf0PnldfGvU8IvHqPkPpe9GxkJr+PeD3RATYYRZm4OhSpdh7sjSWDCmFCYsDQMAGBoJNOvyHGUqJ8PcKgMPbppgwZhSUKtl6DPh33c3c3pkoUjFmk934MJDZwza0hrPk01QukQc4l8oAABKQxXKO0Zh5Wkf3H1mC0tlKsY1/RsLO+3DJ792kcazqMtehD23Qr/N7ZCqkqNnjWtY1HkvWq/siehXDvSJslMQ8a3x83Rn2Dqm48EtE61ytVoGY6Ua7ftG4tQe6wJre2EX9KwEBv7aVvqeoc48qFYapWPJZ3twL8IWX6zNHP5lk/OY/8k+BKzqBCEy603vdAR2FskY9GsbGMrVmNT+GL5pewJfb2v2389MHrBbO4AlS5bA3d0dSqUStWvXxrlz5/TdJB01myQgYNxT1H/lbJyGmaUa3/0WjMbtYlGqTCoq+CRj0IzHuH/NFM8evzw71GvMU3TqHwmP8i9eOy1zywzYOKikj7Ey+41IUSBSBTJOpsL4SzPIqxnDwFUO4z5mMHCRQ7UjczkZ+SthHGAGuY9xjuNJW5wIo84mMP7UFAYehjAobQjDJkrIjJmd69v7Ht/hDxS4fdEMQ757jA+qpaBUmVQM+e4xUl/IcOwPawBAUrwBDmyywYDJ4ajWIBFlq6Rg5LyHuHXBHLcvah9cFqf4nji4Og7/6YyHD8wRcs8C8yZVhIPTC5T1zjw5ERZsjhmjq+LcSXs8fWyKq+dtsHZxGdRuFAkDeeZJD3OLdHw2MAhzv62I4/ud8PSxKULvW+CfEw76nDX6v/c9vjWMjIVWXFpYvzxp9vC+AheOWWLE3IcoXz0ZlWonYeD0xzix0xrRTzOvsTi5pcG/ewy8Kr5ASdd01PWPR5NOz3HjH7OcJvne61PnMiLizTBxbxPc+LckwuMscSa0FB7HWgEAEtMU+OK3tjh4pwzCYkrg+hNHzDrUEBWdIuFokQAAsDZJgZtNHH45+yHuR9ri4XNrLDxRBybGKpSxi9Hn7BGKR3xrnD9qgYsnLNBvYrjOMKWpGkO/e4xWPWNg46DSGV5UZagNEJ1oKn1ikzNPWlQr/RRO1gmYvMMPQc9sEfTMFpP+8IO3cyRqemQuP3e756hf9hGm7WqMG+ElceWhE2bva4AWlYJgZ5Gkz9l6o2KfnP/2228YOXIkJk2ahEuXLqFq1arw9/fHs2fP9N20t5IUL4dMJmBmlferYou/dkHXipUwpFVZHNhkA1F0j92BDAFkQDeJVgAZ19NzNQrxXA31LRVkJQyQ8uVzJLWPQsqQWGRcy93v6d0pCvGdnpa5bhor1FKZgUHmzv7meXMAwP1rplClG+DDhi97dpQumwoHlzTcvqh9cF6s4jsLM/PMg5qEuJy7tJlZpCM5yRDqjMzd44d1omFgANg6pGL5ttP4df9JTPj+GuxKvv4kJ717RSG+Na6dMUe3yhXRt0F5/DjeFfExcmnY7QtmMLdSoVzVl701qjdMgMwAuHM5++Q7PMQYF45Zokrd7Ht7FQWNy4Ti5lMHzGl/AMcGr8ZvAVvQqeqt1/7GXJEGtQASUjOvrsemKBESbY22le7BxCgdcpkaXardQnSSCW49tf8vZoNyUFziG8i8dW3BmFIYuygMCpNitFN+g9I2cdg/6lfsHLYB0zsdhqNV5kk1I3kGBIA01cvlmKoyhFrIUK10Zk+hKqUiEJ9ijNtPXp5IP/fAFWohQ2WXwr0OFfvkfN68eejXrx8CAwPh7e2N5cuXw9TUFL/88ou+m5ZvaS9k+HmGM3w7PIeZhfrNP3hFrzH/4uvlYZi1ORgNWsVh0Veu2Pmz3Ttqqf7JTA1gUNEQaWuToY7KgMgQUB18AfVNFUR07pad+knmCZC01UkwbGsC5RwrGJQzxIsRsVA/Kj5nOAujohDfpcq8gINLGn6Z5YSEWDnS02T4bbEDov41RkxE5lWzmGeGMDJWwzzLyThr+3TEPHt591Jxi+9XyWQCA0bfxc3L1ggLNs+2jqV1Gnr0C8G+ba5SmaNrCmQGAh/3CcHKH8phxpgqMLdKx4xlF2FomLftKxWsohDfAFDDNx5jFobh+9+D0ffrf3H9jDm+/tQTGf8P55hIQ1jbau9L5IaAhbVKK74BYHjbsmjjUQV96nujUu1E9Brz9L+ajf+cq3U8un14Ew+fW+HL39vg98sVMa7pKbStdCfb+sZyFYb7nsG+W2WRlKbpCSdD/9/aorxDFE6PWIVzo1fisxpXMfD31lICT/pRXOJbiMz70lt/Fq11Aq64u/G4JCbv8MPg9a3x3e5GcC6RgFWBO2FqnIbrj0viRZoRhjY/C6VROpRG6Rje4gwMDQTszJMBALbmyYhJ0r49IENtgPgUBWz/X6ewKtb3nKelpeHixYuYMGGCVGZgYIBmzZrhzJkzOvVTU1ORmpoqfY+Pj9epo2+qdGDGAHdAAEO+e5zn3/ccESH9v0zlFLxINsCWZQ7o8HlUAbaycFF8Y4nU7xKQ0ikGkAMGZQ0hb6qA+m4uE+v/H58btVPCqJUSACAvZ4SMi2lQ7X0B4wHZJwL0bhWV+DY0Aib+HIJ5I0uji3dlGMgFPmyYgJpN4vN81bs4xrfGwAl34FYmEaMDa2Y73MRMhSk/XsbDB2bYsMJTKpfJACMjgeWzy+Py2cxndXw/oTI2HDqBKjVjcOlM8Ti5UdgUlfgGoPVARo8KL+DhnYKAut64dtpcqzdMbny1PBQpSQZ4cNMEq6Y7Y+syB3QbVLivEuWXgUzg5lN7LDpZBwBw55k9ytjFoGu1W/jzRnmtuoYGGZjT/iBkAGYcbPTKEIGvmv+FmGQTBG7ogBcqQ3Sqchs/dtmHT9Z2RlRS0b0toDArTvG982c7pCQa4OMhETmPpBg6HVRa+n9QhC2uhztgz/ANaF4xGDsvV8C4Lc0xofVf6F77OtRChgPXy+D2EzvpfvP3WbG+ch4VFYWMjAyULFlSq7xkyZJ4+lT3bPOsWbNgZWUlfUqVKvVfNTVXNIl5RLgxZm0OzvNV8+yUr56MqH+NkZb6/q/sOTFwkcNkkTVMD9jBZIsNTFaWAFSAgbP8zT8GILPNDCMDd+1zXQZuhlBH8MqavhSl+C5bJQXLDt/F9jvXsOnKDczc+ADxz+VwKp15MGLjoEJ6mgES47TX2dhIo9fen1Yc4hsAvhx3B7UaRmJ8vxqIfqbUGW5iqsK0JZeQnGyIaSOrIkP1ctf4PCrzCtvDBy8P0uOfGyM+1hj2juzari9FKb6zcnJLg5WNCk9CM6/c2tirEButvX/JUAEJsYY68e3gkg63cqnw6xiLPl/9i/VzHaUrdEVNZKIpHkSV0Cp7EG0NJ0vtExqZifkhOFklYsBvbV+5ag7UcgtHI68wjNvVHFfCnXAnwh4zDzXCi3RDtKt09z+ZD9JVnOL7yt8WuH3RDG3cq6JlqaoIrJf5FpDBLcthzrDSrxtVsZL4QoGwaCuUssk88XI2uBTa//gJms/pjaazAzDxj6awt0zC4+eWAIDoRFPYmGn3RJAbqGFpkoroxML9oMdinZzn1YQJExAXFyd9Hj16pO8mSTSJeXiIAt/9FgRLm4LZGwffNIG5tQrGiqJ/D4zMRAYDOzlEghoZ59Mgb5DzA+C0fudkAJmdAdQPtZe5eJwBA0eG2PuiMMe3hpmlGta2GQh/YIz7V01R1z9zJ1W2SjIMjdS4fOplL41HQQo8CzdGBZ+cH3xS9ONb4Mtxd1C3yTNMGOCDiCcmOjVMzFSYvuwSVOkGmDq8GtLTtE9w3LpiDQBwdX+5HM0t02FpnYZn/+qOjwqn9yG+NSKfGCH+uRw2DpnPLalQIwmJcYa4f+3l+nbllAWEGij/Yc7xrVYDKpUMooieI74S7gh3m1itMjebODyJf7kd1CTmpUvEYsDmtoh7oX1yzsQw8+SGOsvVNiEyb4Wh98P7HN8Dpz3GssN3sexQ5mf6usxXrH21PBQB44rmmxbyw8Q4Ha428YjKkljHJpsg8YUCNT3CYWOWgpN33QEA1x6VhKVJGso7RUp1a3qEw0AmcD28cD/QtVh3a7ezs4NcLkdEhHZXkoiICDg6OurUVygUUCj0cw9SSpIBnoS8nPbTR8YIvmECC2sVbEqmY1o/DwRdN8HUXx9AnSGT7kOzsM6QXtnw7LEREmIN8SzcCOoMIPhG5o7e2SMVJmZqnD1oieeRhqjgkwwjhRqXTlpg848O6PJFpG6DihDVuTRAAAal5BDhGUhblgiD0nIY/r+LuohXQx2hhojKTL7FQxUyAMhsDGBgawCZTAaj7iZIW50MgzKGMChjCNX+F1CHqaCYaqnHOSveikp8O7im4+SfVrCyzYCDSxpCbiuxfKIr6n4UBx/fzIejmFmq4d8jBisnu8DCOgNmFpmvUqvgk4QKPpn3VhXH+B444Q58Wz7F1BFVkZJkiBK2mT0NkhINkZYqh4mZCjOWXoJCmYE5X1eCqZkKpmaZB+txz42hVssQ/tAMZ47ZY8CYu1g03RvJiYYIGHIfj0PNcO1CiddNnt6hohLfFiUysH6uIxq0jkUJBxX+DTXGqunOcPZIleK7dNlU1PCLx4LRpTDk+8fISJdhyTcuaNw+FraOmevr0e0lIDcU8KiQAiNjgXtXTbF6lhMat3teZN9zvv58Vaz99A/0rXMRB++UQSWnCHSpegtTDzQGkJmY/9DhICqUjMSQra1gYCBga5a5PYxLUUClluPqk5KIf6HA9NZHsOLvGkhVGaJT1VtwsU7AX8Fu+py9Yq04xbeDazqAlw8QVpplnk1zdkuDvfPL8rB7CqjSDJDwXI7kJAPpGF7zLvSiZniLMzh51w3/xpnD3iIZA3zPQ62WYf/1MgCAttXuICSqBGKTlKhcKgKjP/obG89UQVi0NQAgNKoE/r5fCt+2O4GZuxvC0ECNsa1O4eCNMohKKNy3qxTr5NzY2Bg+Pj44cuQIOnToAABQq9U4cuQIBg8erN/GZXHvqinGdikjfV8x2QUA0LxbDD4d9RRnD2a+OmRgc+37rGZvDULVepldvH79wQmHfreRhg1s8YFWHbmRwJ9r7LBisgJCAM7uaRgw+Qla9ox+p/Omd4lqpK1MgohUQ2ZhAHljYxj3M4PMMPNMuurvNKTNSpCqp075/9MiA0xh3CczwI26mUKkAWmLEiES1DDwMoRynjUMXHLXNZ4KXlGJ79ELHiImwggrJrsgNiqzG2uzrjH4ZLj2QcsXkzPPCE/r5470VBlq+CZg8KyXz50ojvHdplvm/M9edVGrfN7Eijj8pzPKlI9H+SqZr7/55c+/teoEtGogXRn/4dtK6D/6Lib/eBlCLcP1iyXw7aDqWt3f6b9VVOJ7yKxHCLmtxKEtHkiKl8O2pArVG8ej99inWj1axi0Ow5KvXTG+mxdkBkCDVrEYOP3lK5cM5AK/L3FA+IPM+HZwTUO7wCh06ld0T77dfOqAkX/4Y2jjfzCg/kWEx1lg9tH62HurHADAwTwJfmVDAQBb+mzR+m3fje1w4ZELYlNMMHBLawxpdA4/9dgFQwM1gqNsMGz7R7gXyedJ6Etxi+/c+PZTL0Q8ftmjU3MMf+DJlbefiULIwTIRM7schpXJCzxPNsGVh44IWNVRep2au10sBjf7B1YmqXgSa4Ff/qqODWeqaI3jm+1NMa7VKSzrtRtCyHDktgfm7Gugj9nJE5kQxelFOrp+++039O7dGytWrECtWrWwYMEC/P7777hz547OvS5ZxcfHw8rKCs/vecLSggdpb6tTUHN9N6FISE9Kw96PfkZcXBwsLYv3lXvGd+HR6sMW+m5CkaBSp+HIs1WMbzC+C5Oq3w/UdxOKhIzUF7i97CvGNxjfhYnP5C/13YQiISPtBa6v/vqN8V2sr5wDwMcff4zIyEhMnDgRT58+RbVq1bB///43Bj4RFX6Mb6Kii/FNVHQxvqm4KvbJOQAMHjy40HWTIaKCwfgmKroY30RFF+ObiiP29SAiIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50RERERERER6xuSciIiIiIiISM+YnBMRERERERHpGZNzIiIiIiIiIj1jck5ERERERESkZ0zOiYiIiIiIiPSMyTkRERERERGRnjE5JyIiIiIiItIzJudEREREREREesbknIiIiIiIiEjPmJwTERERERER6RmTcyIiIiIiIiI9Y3JOREREREREpGdMzomIiIiIiIj0jMk5ERERERERkZ4xOSciIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50RERERERER6ZqjvBhQFHctVhqHMSN/NeO/Jy5jruwlFgiojVd9NKFK6+rWAoYFC381472VEhOu7CUVChkjXdxOKFO6/C4arB+O7IKjUqbit70YUIYzvguHo+a++m1AkqNSpuJ6LerxyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50RERERERER6xuSciIiIiIiISM+YnBMRERERERHpGZNzIiIiIiIiIj1jck5ERERERESkZ0zOiYiIiIiIiPSMyTkRERERERGRnjE5JyIiIiIiItIzJudEREREREREesbknIiIiIiIiEjPmJwTERERERER6RmTcyIiIiIiIiI9Y3JOREREREREpGdMzomIiIiIiIj0jMk5ERERERERkZ4xOSciIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50RERERERER6xuSciIiIiIiISM+YnBMRERERERHpGZNzIiIiIiIiIj1jck5ERERERESkZ0zOiYiIiIiIiPSMyTkRERERERGRnhnquwGUP5VqJ6LrwEiUrZwMW0cVJvdxx5n9VgAAuaFAwLh/UbNJApzc0pAUb4DLf1ng55lOiIkwksbh4pmKft8+gXfNJBgaCYTcVuLX2U64etpcX7P1n+vW8x7qNXoCV7dEpKUa4PYNG/yyvCLCH1lkU1tg6uwzqFHnGaZ9VQtnTjlLQwYMvQbvytFw90jAwzBzDOnb5L+bCSpyuvYOQj2/iP+vl3Lcvl4Cqxd9gPCHL2Nz1rKzqOITo/W7vdtLYcl3laXvZSvEImDwXZQpHwcI4O4ta6xeVB4h9y3/s3nRt9dtKwGgfstYtO4VjbKVU2Bpk4Evm5fDg5smWuNo2TMafh2fo0zlFJhZqNGpfCUkxcv/61mhIqIg1skS9un4/Nt/Ub1RAkzN1XgUrMDmhQ44tdf6P54b/en62X3Ua/wvXN0S/r+dtMHqZd7a28lFf6NK9Wit3+3d4YYlc6pK3wcMvw7vyjFw80zAozBzDAnw/a9mgYqgt41vC2sVPhv9FNUbJ8LBOQ1xMYY4vd8Ka2c7Ijmh+Ox3un56L/v4fuX4fNaiU6jyYdb4dseSHzLju1nLhxjx9eVsx/9Jm48QF6t4dzPwFor1lfOTJ0+ibdu2cHZ2hkwmw44dO/TdpFxTmqrx4KYSi79y1RmmMFGjTOUUbFxQEoP8y2Lq5+5w9UrFlDUhWvWmrn0AA7nAuK5eGPxROTy4ZYKpv4aghH36fzUbelepWhR2/+GBkV80wtcj60NuKDBj7mkolCqduh26BkNAluO4Du11w8mjLu+yuZQH73N8V64egz1b3DCqbz18M6QWDOVqTF90Tme93P9HKXzasqn0+WVReWmY0kSFqT+eR+RTJUYG1sOY/nWRkmSIaT+eg1yu/q9nSW9et63UDL95zgw/z3TKeRwmalw4boHNixzeVTMpj97n+C6IdXLMjw9RyusFJgd4YECTcvh7rxW+WhEGr0rJ76rZhU7lalHYs90do/o3xDfD68LQUI3p88/obid3uuHTti2kzy9LvHXGdXBPaZw84qxTTvpRnOPbpmQ6bEuq8NNUJwxo8gF+GF4KNXzjMXLuo3fZ7EKn8ofR2LPdA6MGNMI3I+rB0FBkH9+73PBpO3/p88vSl/F98oiL1rBP2/nj4j8OuHbZttAm5kAxv3KelJSEqlWrok+fPujUqZO+m5MnF45Z4sKx7K9+JSfIMaG7l1bZkq9dsGjffdi7pCEy3BiWNiq4eqVh/qhSCLmdecbulxlOaBcQDffyL/A80ii7URc5E8fU0/o+b2Z1bP5zH8p+EIsbV+2kcs8ysej0cRCG9ffFhh37dcaz4scqAAAr69tw94p7t42mXHmf43visFpa3+dNrYJNB4+gTIV43LxsI5W/eCHH8+jsdzCu7omwtErH+hXlEPUsM8Y3riqDpZtOwcEpBf8+Nnt3M1CIvG5bCQBHtmUuz5KuaTnW+WOVPQCgSt3Egm0c5dv7HN8FsU5610jGovEuuHvFFACwaWFJdOoXibJVUhB8w7RgG1xITRxVV+v7vBkfYtOeAyjzQRxuXrWVyl+kyvE8RpnjeFYsyOxtZGWdCo8y8e+msZQnxTm+w+6aYFo/d+n7v2EKrPneCWMXPYSBXECdkfNFoqJEJ75nfohNu/ejzAexuPnK8fmLFznHd1qaHGkxL3sbWFqnokr1SCz87sN30+gCUqyT85YtW6Jly5b6bsZ/wswyA2o1kBSXuZLGx8jxKEiBZl2f4/51E6SnGaD1Z9F4HmmI+9dM3jC2osvMPLPXQEK8sVSmUKgwduJFLF1Q9bU7eCpcilJ8m5lnnilOjNM+aeb30RP4tQzH82gFzv3lgM0/l0VqamaMh4eZIy7WCC3aP8Lvq8vAQC7Qot1jPHxgjoh/i2+MU9FQlOI7P25dMEXjdrE4d8QSiXFyNGoXC2OlwLVidFtaVmZmmfvvxPgs28nmj+HX4jGexyhw7u+S2Ly6HFJTi/Xhb6FX3OM7KzPLDCQnGhSbxDw7L+PbWKtcO74dsXlNzvHd9KNHSH0hx9/HCncvGW6d8iA1NRWpqanS9/j49+MMq5FCjb5f/4vjO6yRnKg5gyTD+I89MemXUOy4fwNCDcRGGeLrnh5IjCueq4VMJjBgyHXcvGaDsJCXZz37DbmB2zdscPZUzl0M6f1XWONbJhPoP/IWbl4pgbAHL++1OnHAGc+emiA6UgGPMgkIHHwXrm5JmDHOBwCQkmyICV/UwTdzLqJ7nyAAwJNHZvh2aC2oM4r1HU1UDBXW+M6vGQPc8dXyUGy9dROqdCA1xQBT+rrjSWjh7ar5LslkAv2H3cTNq9r77xOHXPDsqSmio5TwKBOPwC9vwbV0ImZ8Ves1Y6P3TVGL71dZ2qjwyfAI7Ftv++bKRZRMJtB/6A2d4/MTh1xfxrdX3Mv4/jr7+G7ROgwnDrsiLa1w37tfPLOwfJo1axamTJmi72bkidxQ4OsVYYAMWDT+1ftfBAbPDEdslCFGdSyDtBcyfNQjBlPWhGJoq7KIeVY8urW/auCIq3DziMfowY2kstr1/0XV6pEY0tdPjy2j/0Jhje8vx96Em2cixvSvo1W+f0dp6f9hwZaIiVZg1tJzcHRJwtNwMxgrMjDsm2u4da0EZn9TDQZygU49QzB5/nmMCKiPtNTCvXMiKkiFNb7zq/fYf2Fuqca4bp6IjzFE3Y/i8PXyUIzqWAahd4pfz5gvR12Dm2c8xnzZQKt8/y536f9hDywRE6XArEVnpO0kFQ1FLb41TM0zMO3XEDy8p8S6uY76bo7efDny//E9sKFWuU58Rysx68fTcHROwtMn2vFdvmIMSnskYu50n/+iyW+Fl0/yYMKECYiLi5M+jx4V7oczZCbmoSjpkoYJ3T1fuWoOVGuQiFrN4jHrSzfcOm+GoOumWPyVK9JeyNCsW8xrxlo0fTn8KmrVi8D44Q0QHfnywKZq9Ug4OSdhy549+PPoTvx5dCcA4Ktp5/Ddwr/01Vx6BwpjfH8x+iZqNXiGCQNrI/rZ6w+4796wBgA4l8p8IJSv/xM4OKVgwdQquH/bGndvlMCcb6vB0TkFdRpFvOumExUqhTG+88vJLRXt+0Rj3shSuHLKAg9umWDDPEfcv2aKdgHRbx5BEfPFyGuoVS8CE4bU09p/Z+furRIAAGeXpP+iafQfKUrxrWFiloEZGx8gJSmzV0yGqnh2af9ixDXUqvcUE4bWz318u+rGt3/bMATfs0LQXet30cwCxSvneaBQKKBQvB9dxjSJuYtHGsZ28ULCc+0/tcIk82nN6iwPbVYLGQyKVfwLfDn8Guo2/BfjhzVAxL/aZ9q2bCiHA7vdtcqWrT2KnxZXxj+ni+9ZzKKocMW3wBejb6Gu71NM+LIOIp68+QFPnuUyu/HFRGXOg0KZASFkEOJlHbUAhABkBiK7URAVWYUrvt9OTvvvjIziFtsCX4y8jrqNnmLC4Ho6++/seJbNfGBrTA4P0qT3U1GKbyDzivmMjQ+QnibDpAAPpKcWx2upAl+MuI66jf7FhCH13yq+lSYqNGgSjrXLdd/UUBgxOX9PKU0z4Ozx8kmPjqXS4FkxBQmxcsREGOHbn0JRpnIKJvbygIFcSK9HS4iVQ5VugNsXzZAYJ8eYhY+wYX5JpL4wQMue0XAslYZzR4rPO5AHjrgG32aPMPWrOkhJNkQJmxcAgKREI6SlZT4BMruHwEVGmGhtKJxcEmFiokIJm1QoFGp4lokFADwMtYRKVRw3qvQ2Bo69icb+TzBttE/memmbeS9dUqIh0lLlcHRJgq//E1w47YD4OCN4lElAvxG3cf2SDUKDMuP38j926DPkDgaOvYk/f3eHzECga69gZGTIcO1C8bl37XXbyshwY1hYq2Dvkg7bkpnbyFJemduA588MpbdWlLBPRwkHFZw9Mv8OHuVTkJwkR2S4ERJiuRulvHnbdfJRkBLhD4wxbPZj/DTVGfHP5aj3URyqN0rExF4eepknfRg46joaN3+MaeNrZbv/dnRJgm/zx7hwpiTi44zhUSYe/YbewPXLtggNfvneaSeXRJiYZqCEbSqMFRnSAf7DEAvuvynP3ja+Tc0zMHPTAyhM1Jg9xB2m5hkwNc8AAMRFG0KtLh5X0AaOuobGzR5j2oTa2ce38//j++z/49srLtv4BoBGTcIhlwscO1hKH7OSZ8X6qCIxMRFBQUHS95CQEFy5cgU2NjYoXbr0a36pf+WqpmDOtmDp+xdTngAADv5WAuvnOqKuf+ZVtGWH72n9bkxnL1w7Y474GEN8/YknAsb/i+9/D4bcSCDsrhKTA93x4FbxuV+tTcfMd7/PXnRKq3zezA9xeL9brsczbOxlVPnwZXfCxb8cBwAEdGuOZ095X5s+vM/x3brLQwDA9yv+0SqfP6UKDu9xhSrdANVqRaN9j1AolRmIjFDi72OO2PzLy1coPg4zx5RRPvjk8yD88PNpCLUMwfcsMXFYTTyPLj5vHXjdtnLuiNKo0yIeoxe87AL51fLMZb9ubkms//89fq17ReOzUS9vBZi7I3N8PwwvhUO/v3y1Hf133uf4ftt1MkMlwzefeaLvV/9iytoQmJip8STEGD8MK4XzR4vPyfXWnUIBAN8vOa1VPn9GNRzeWzpzO1kjCu27PcjcTj4zwd/HnbB5TTmt+kPHX0WV6i/334vWnAAABHZuhmdPi8dr6Qqb4hzfZSqnoIJP5u1pa87c0Rp3r1oVEPFY+2nlRVXrjqEAgO8X/61VPn/Ghzi8rzRUKgNUqxGJ9t2CX4lvZ2xeW05nXC3ahOH0CWckJb4fz9OSCSGKUx8oLcePH4efn+6Dvnr37o01a9a88ffx8fGwsrKCL9rDUPZ+/MELM3mZ4nPG/11SZaTiyIMfERcXB0vL4nOgllVBxXczly9gaFB0usvpi+pxuL6bUCSoRDqOYyfjm/vvQsXQI/cnsylnKnUqDocuZnwzvgsVQ093fTehSFCpU3E4ZNEb47tYXzn39fVFMT43QVSkMb6Jii7GN1HRxfim4ow30xARERERERHpGZNzIiIiIiIiIj1jck5ERERERESkZ0zOiYiIiIiIiPSMyTkRERERERGRnjE5JyIiIiIiItIzJudEREREREREesbknIiIiIiIiEjPmJwTERERERER6RmTcyIiIiIiIiI9Y3JOREREREREpGdMzomIiIiIiIj0jMk5ERERERERkZ4xOSciIiIiIiLSMybnRERERERERHrG5JyIiIiIiIhIz5icExEREREREekZk3MiIiIiIiIiPWNyTkRERERERKRnTM6JiIiIiIiI9IzJOREREREREZGeMTknIiIiIiIi0jMm50RERERERER6xuSciIiIiIiISM8Mc1Pp5MmT+Rp5o0aN8vU7IiIiIiIiouIkV8m5r68vZDJZrkcqhIBMJkNGRka+G0ZERERERERUXOQqOT927Ni7bgcRERERERFRsZWr5Lxx48bvuh1ERERERERExdZbPxDu33//xdWrV5GUlFQQ7SEiIiIiIiIqdvKdnO/cuRPly5eHq6srqlevjn/++QcAEBUVhQ8//BA7duwoqDYSERERERERFWn5Ss7//PNPdOrUCXZ2dpg0aRKEENIwOzs7uLi4YPXq1QXWSCIiIiIiIqKiLF/J+dSpU9GoUSOcOnUKgwYN0hlet25dXL58+a0bR0RERERERFQc5Cs5v3HjBrp165bj8JIlS+LZs2f5bhQRERERERFRcZKv5NzU1PS1D4B78OABbG1t890oIiIiIiIiouIkV69Sy8rPzw9r167F8OHDdYY9ffoUP/30E9q0afO2bSv0NPfaq5AOiDdUpjcSGan6bkKRoFJnLsdXnwVBeSfFtzpNzy0pGlQiXd9NKBJUyFyOjO+3w/13AVNz/10QNPsbxvfbYXwXMMZ3gchtfOcrOZ8xYwbq1KmDmjVromvXrpDJZDhw4ACOHj2KFStWQAiBSZMm5WfU75WEhAQAwCns1XNLiogH+m5A0ZKQkAArKyt9N+O9pYnv4//+oueWEOlifL8d7r8LWKi+G1C0ML7fDuO7gIXouwFFy5viWybyeXru5s2bGDZsGI4dO6Z1BsDX1xdLlixBhQoV8jPa94parcaTJ09gYWEBmUym7+bkKD4+HqVKlcKjR49gaWmp7+a8t96X5SiEQEJCApydnWFgkO+3JRZ7jO/i5X1ZjozvgsH4Ll7el+XI+C4YjO/i5X1ZjrmN73wn5xrPnz9HUFAQ1Go1PD09YW9v/zajo3cgPj4eVlZWiIuLK9QrbWHH5UiFEdfLgsHlSIUR18uCweVIhRHXy4JR1JZjvrq1v6pEiRKoWbNmQbSFiIiIiIiIqFjKd5+ZyMhIjB49Gt7e3jA1NYWpqSm8vb0xevRoREREFGQbiYiIiIiIiIq0fCXnN2/eROXKlTFv3jxYWVmha9eu6Nq1K6ysrDBv3jxUqVIFN27cKOi2Uj4pFApMmjQJCoVC3015r3E5UmHE9bJgcDlSYcT1smBwOVJhxPWyYBS15Zive859fX1x8+ZN7N27V6dL+7lz59CqVStUrlwZx44dK7CGEhERERERERVV+bpyfu7cOQwbNizbe81r1aqFYcOG4Z9//nnrxhEREREREREVB/lKzh0cHKBUKnMcrlQq4eDgkO9GERERERERERUn+UrOhw8fjmXLluHp06c6w548eYJly5Zh+PDhb9s2IiIiIiIiomIhV69Smzdvnk6Zubk5ypQpg44dO6JMmTIAgPv372PHjh0oU6YM3vL16URERERERETFRq6unI8ePVrnc/36dSQnJ2PDhg2YMmUKpkyZgo0bNyI5ORnXrl3D6NGj33XbKReWLFkCd3d3KJVK1K5dG+fOndN3k947J0+eRNu2beHs7AyZTIYdO3bou0lEABjfBYHxTYUV4/vtMb6psGJ8v72iGt+5Ss5DQkLy/Hnw4MG7bju9wW+//YaRI0di0qRJuHTpEqpWrQp/f388e/ZM3017ryQlJaFq1apYsmSJvptCJGF8FwzGNxVGjO+CwfimwojxXTCKanzn61Vq9H6oXbs2atasicWLFwMA1Go1SpUqhSFDhmD8+PF6bt37SSaT4Y8//kCHDh303RQq5hjfBY/xTYUF47vgMb6psGB8F7yiFN/5eiAcFX5paWm4ePEimjVrJpUZGBigWbNmOHPmjB5bRkRvi/FNVHQxvomKLsY3vUmuHgiXnWvXrmHRokW4dOkS4uLioFartYbLZDIEBwe/dQMpf6KiopCRkYGSJUtqlZcsWRJ37tzRU6uIqCAwvomKLsY3UdHF+KY3ydeV8+PHj6NWrVrYvXs3nJ2d8eDBA3h6esLZ2RlhYWEwNzdHo0aNCrqtREREREREREVSvpLziRMnwtPTE3fv3sXq1asBAF999RVOnTqF06dP4/Hjx+jWrVuBNpTyxs7ODnK5HBEREVrlERERcHR01FOriKggML6Jii7GN1HRxfimN8lXcn7p0iX07dsXlpaWkMvlAICMjAwAmQ85GDBgAL799tuCayXlmbGxMXx8fHDkyBGpTK1W48iRI6hbt64eW0ZEb4vxTVR0Mb6Jii7GN71Jvu45NzQ0hIWFBQDA2toaRkZGWo//9/T0xK1btwqmhZRvI0eORO/evVGjRg3UqlULCxYsQFJSEgIDA/XdtPdKYmIigoKCpO8hISG4cuUKbGxsULp0aT22jIozxnfBYHxTYcT4LhiMbyqMGN8Fo8jGt8gHHx8fMXr0aOl7lSpVROfOnaXv7du3Fx4eHvkZNRWwRYsWidKlSwtjY2NRq1YtcfbsWX036b1z7NgxAUDn07t3b303jYo5xvfbY3xTYcX4fnuMbyqsGN9vr6jGd77ecz5x4kT88ssvCA0NhaGhIdauXYvAwEB4eXkBAIKDgzFr1iyMGzfurU4cEBERERERERUH+UrO09PTER8fDxsbG8hkMgDA+vXrsW3bNsjlcrRp0wYBAQEF3VYiIiIiIiKiIilfyTkRERERERERFZx8Pa2diIiIiIiIiApOrp7W3qRJkzyPWCaTab0mgIiIiIiIiIiyl6vkXK1WS/eW5xZ7yxMRERERERHlDu85JyIiIiIiItIz3nNOREREREREpGdMzomIiIiIiIj0jMk5ERERERERkZ4xOSciIiIiIiLSMybnlGsBAQFwd3fXdzOI6B1as2YNZDIZQkND9d0UIsoiu/2wTCbD5MmT9dKeoojLk0jb8ePHIZPJsHXr1tfW4/FDwWByXghpVm7Nx9DQEC4uLggICEB4eLi+m1doZF1Or37Gjx+v7+Zla+bMmdixY4e+m0HFxNKlSyGTyVC7du3/ZHoBAQE6265SpUqhe/fuuHXr1n/Shte5desWJk+ezAMHyrOQkBAMHjwY5cqVg6mpKUxNTeHt7Y1Bgwbh2rVr+m7eO7dx40YsWLAg1/Xd3d21tgVKpRJly5bFmDFjEBMT8+4amkt79+5lAk6F3vXr19GlSxe4ublBqVTCxcUFzZs3x6JFi6Q6PK4senL1nvOchIeH4+TJk3j27Bk6d+4MV1dXZGRkIC4uDlZWVpDL5QXVzmJp6tSp8PDwwIsXL3D27FmsWbMGp06dwo0bN6BUKvXdvEJDs5xeValSJT215vVmzpyJLl26oEOHDvpuChUDGzZsgLu7O86dO4egoCCUKVPmnU9ToVBg1apVAACVSoXg4GAsX74c+/fvx61bt+Ds7PzO25CTW7duYcqUKfD19WUvIMq13bt34+OPP4ahoSF69uyJqlWrwsDAAHfu3MH27duxbNkyhISEwM3NTS/tS0lJgaHhWx3OvdHGjRtx48YNDB8+PNe/qVatGkaNGgUAePHiBS5evIgFCxbgxIkTOHfu3Dtqae7s3bsXS5YsyTZB/y+WJ9GbnD59Gn5+fihdujT69esHR0dHPHr0CGfPnsXChQsxZMgQAIXruPKzzz5D9+7doVAo9N2U91q+tj5CCIwaNQqLFy+GSqWCTCZD5cqV4erqisTERLi7u2Pq1Kl52oiTrpYtW6JGjRoAgM8//xx2dnb4/vvvsWvXLnTr1k3PrSs8Xl1OBSkpKQlmZmYFPl6i/0JISAhOnz6N7du3Y8CAAdiwYQMmTZr0zqdraGiITz/9VKusTp06aNOmDfbs2YN+/fq98zYQFZTg4GB0794dbm5uOHLkCJycnLSGf//991i6dCkMDF7fEfFd7k8K68l6FxcXrW3B559/DnNzc/zwww+4f/8+ypYtq8fW5aywLk8qXmbMmAErKyucP38e1tbWWsOePXumn0a9gVwu54XZApCvbu1z5szBwoULMXr0aBw6dAhCCGmYlZUVOnXqhG3bthVYIylTw4YNAWQeLGikpaVh4sSJ8PHxgZWVFczMzNCwYUMcO3ZM67ehoaGQyWT44YcfsHLlSnh5eUGhUKBmzZo4f/68zrR27NiBSpUqQalUolKlSvjjjz+ybVNSUhJGjRqFUqVKQaFQ4IMPPsAPP/ygtU4AmfdwDR48GFu2bIG3tzdMTExQt25dXL9+HQCwYsUKlClTBkqlEr6+vgXa7fTo0aNo2LAhzMzMYG1tjfbt2+P27dtadSZPngyZTIZbt27hk08+QYkSJdCgQQNp+Pr16+Hj4wMTExPY2Nige/fuePTokdY47t+/j86dO8PR0RFKpRKurq7o3r074uLipGWQlJSEtWvXSl39AgICCmw+iV61YcMGlChRAq1bt0aXLl2wYcMGnTo3b95EkyZNYGJiAldXV0yfPh1qtVqn3s6dO9G6dWs4OztDoVDAy8sL06ZNQ0ZGRq7a4ujoCAA6V6MePHiArl27wsbGBqampqhTpw727Nmj8/tnz56hb9++KFmyJJRKJapWrYq1a9fq1Nu8eTN8fHxgYWEBS0tLVK5cGQsXLgSQeRtM165dAQB+fn5SDB4/fjxX80DF0+zZs5GUlITVq1frJOZA5jo9dOhQlCpVSioLCAiAubk5goOD0apVK1hYWKBnz54AgL/++gtdu3ZF6dKloVAoUKpUKYwYMQIpKSk6487tfji7e6TDw8PRp08flCxZEgqFAhUrVsQvv/yiVUdzH+nvv/+OGTNmwNXVFUqlEk2bNkVQUJBUz9fXF3v27EFYWJgUN/nteZLTtiA3+2kAuHz5Mlq2bAlLS0uYm5ujadOmOHv2rFad9PR0TJkyBWXLloVSqYStrS0aNGiAQ4cOAcj8+yxZskRadpqPRtblqTk+CAoKQkBAAKytrWFlZYXAwEAkJydrTTslJQVDhw6FnZ0dLCws0K5dO4SHh/M+dsqz4OBgVKxYUScxBwAHBwcArz+uDAsLw8CBA/HBBx/AxMQEtra26Nq1a7bH17GxsRgxYgTc3d2hUCjg6uqKXr16ISoqKsf2paamok2bNrCyssLp06cBZH/Pubu7O9q0aYNTp06hVq1aUCqV8PT0xK+//qozzmvXrqFx48ZaxySrV68udvex5+vK+U8//YRevXph5syZiI6O1hlepUoV7Nu3760bR9o0K2aJEiWksvj4eKxatQo9evRAv379kJCQgJ9//hn+/v44d+4cqlWrpjWOjRs3IiEhAQMGDIBMJsPs2bPRqVMnPHjwAEZGRgCAgwcPonPnzvD29sasWbMQHR2NwMBAuLq6ao1LCIF27drh2LFj6Nu3L6pVq4YDBw5gzJgxCA8Px/z587Xq//XXX9i1axcGDRoEAJg1axbatGmDsWPHYunSpRg4cCCeP3+O2bNno0+fPjh69GiulktcXJzOBsTOzg4AcPjwYbRs2RKenp6YPHkyUlJSsGjRItSvXx+XLl3SOcDo2rUrypYti5kzZ0onGGbMmIFvv/0W3bp1w+eff47IyEgsWrQIjRo1wuXLl2FtbY20tDT4+/sjNTUVQ4YMgaOjI8LDw7F7927ExsbCysoK69atw+eff45atWqhf//+AAAvL69czSNRXm3YsAGdOnWCsbExevTogWXLluH8+fOoWbMmAODp06fw8/ODSqXC+PHjYWZmhpUrV8LExERnXGvWrIG5uTlGjhwJc3NzHD16FBMnTkR8fDzmzJmjU18TjxkZGXjw4AHGjRsHW1tbtGnTRqoTERGBevXqITk5GUOHDoWtrS3Wrl2Ldu3aYevWrejYsSOAzINdX19fBAUFYfDgwfDw8MCWLVsQEBCA2NhYDBs2DABw6NAh9OjRA02bNsX3338PALh9+zb+/vtvDBs2DI0aNcLQoUPx448/4quvvkKFChUAQPqXKDu7d+9GmTJl8vzcBpVKBX9/fzRo0AA//PADTE1NAQBbtmxBcnIyvvzyS9ja2uLcuXNYtGgRHj9+jC1btki/z+1+ODsRERGoU6eOdFLc3t4e+/btQ9++fREfH6/Tq/G7776DgYEBRo8ejbi4OMyePRs9e/bEP//8AwD4+uuvERcXh8ePH0v7dXNz8ze2Iz09XdoWvHjxApcvX8a8efPQqFEjrVvRcrufvnnzJho2bAhLS0uMHTsWRkZGWLFiBXx9fXHixAnpbzR58mTMmjVL2t/Gx8fjwoULuHTpEpo3b44BAwbgyZMnOHToENatW/fG+dDo1q0bPDw8MGvWLFy6dAmrVq2Cg4ODtL0BMhP/33//HZ999hnq1KmDEydOoHXr1rmeBpGGm5sbzpw5gxs3buR4q+brjivPnz+P06dPo3v37nB1dUVoaCiWLVsGX19f3Lp1S9omJSYmomHDhrh9+zb69OmD6tWrIyoqCrt27cLjx4+l4+lXpaSkoH379rhw4QIOHz4sHVfkJCgoCF26dEHfvn3Ru3dv/PLLLwgICICPjw8qVqwIIPOEoubE+YQJE2BmZoZVq1YVzy7yIh8UCoVYuXKlEEKIqKgoIZPJxJEjR6Thy5cvF0qlMj+jJiHE6tWrBQBx+PBhERkZKR49eiS2bt0q7O3thUKhEI8ePZLqqlQqkZqaqvX758+fi5IlS4o+ffpIZSEhIQKAsLW1FTExMVL5zp07BQDx559/SmXVqlUTTk5OIjY2Vio7ePCgACDc3Nyksh07dggAYvr06VrT79Kli5DJZCIoKEgqAyAUCoUICQmRylasWCEACEdHRxEfHy+VT5gwQQDQqvu65ZTd59V5cXBwENHR0VLZ1atXhYGBgejVq5dUNmnSJAFA9OjRQ2saoaGhQi6XixkzZmiVX79+XRgaGkrlly9fFgDEli1bXttmMzMz0bt379fWIXpbFy5cEADEoUOHhBBCqNVq4erqKoYNGybVGT58uAAg/vnnH6ns2bNnwsrKSif+kpOTdaYxYMAAYWpqKl68eCGV9e7dO9t4dHFxERcvXtT6vWb6f/31l1SWkJAgPDw8hLu7u8jIyBBCCLFgwQIBQKxfv16ql5aWJurWrSvMzc2lbcewYcOEpaWlUKlUOS6XLVu2CADi2LFjr1l6RJni4uIEANGhQwedYc+fPxeRkZHS59UY0cTB+PHjdX6XXSzNmjVLyGQyERYWJpXldj8sROb+ddKkSdL3vn37CicnJxEVFaVVr3v37sLKykpqw7FjxwQAUaFCBa3jiIULFwoA4vr161JZ69atdab7Om5ubtluC+rXr6/Trtzupzt06CCMjY1FcHCwVPbkyRNhYWEhGjVqJJVVrVpVtG7d+rXtGzRokNaxwquyLk/N8cGrx1RCCNGxY0dha2srfb948aIAIIYPH65VLyAgQGecRG9y8OBBIZfLhVwuF3Xr1hVjx44VBw4cEGlpaVr1cjquzG5bc+bMGQFA/Prrr1LZxIkTBQCxfft2nfpqtVoI8XJbsWXLFpGQkCAaN24s7OzsxOXLl7Xqa47LXz1+0GwLTp48KZU9e/ZMKBQKMWrUKKlsyJAhQiaTaY0zOjpa2NjY5ConKEry1a3dwcFBp0vvqy5evIjSpUvnZ9T0imbNmsHe3h6lSpVCly5dYGZmhl27dmmdOZfL5TA2NgYAqNVqxMTEQKVSoUaNGrh06ZLOOD/++GOtK++arvIPHjwAAPz777+4cuUKevfuDSsrK6le8+bN4e3trTWuvXv3Qi6XY+jQoVrlo0aNghBCp/dE06ZNta5Ua85yd+7cGRYWFjrlmja9yZIlS3Do0CGtz6vzEhAQABsbG6l+lSpV0Lx5c+zdu1dnXF988YXW9+3bt0OtVqNbt26IioqSPo6Ojihbtqx0+4BmWR04cECnmxvRf23Dhg0oWbIk/Pz8AGR2ffv444+xefNmqSv63r17UadOHdSqVUv6nb29vdT99lWvXk1PSEhAVFQUGjZsiOTkZNy5c0errlKplOLwwIEDWLFiBczNzdGqVSvcu3dPqrd3717UqlVL6/YRc3Nz9O/fH6GhodLT3ffu3QtHR0f06NFDqmdkZIShQ4ciMTERJ06cAABYW1sjKSlJin+itxUfHw8g+6vEvr6+sLe3lz6abtKv+vLLL3XKXo2lpKQkREVFoV69ehBC4PLlywDyth/OSgiBbdu2oW3bthBCaO23/P39ERcXp3NsEBgYKB1HALrHBflVu3ZtaVuwe/duzJgxAzdv3kS7du2kbvy53U9nZGTg4MGD6NChAzw9PaV6Tk5O+OSTT3Dq1Cnp72VtbY2bN2/i/v37b9X+rLIeHzRs2BDR0dHSdPfv3w8AGDhwoFY9zYO7iPKiefPmOHPmDNq1a4erV69i9uzZ8Pf3h4uLC3bt2vXG37+6rUlPT0d0dDTKlCkDa2trrW3Atm3bULVqVam32qtevd0DyOyp2qJFC9y5cwfHjx/X6Z2bE29vb2m7AmQea3zwwQda25j9+/ejbt26WuO0sbHJ9pikqMtXct6pUycsX75ca6Fq/oAHDx7UureP8k+TdG7duhWtWrVCVFRUtt071q5diypVqkj3Vtnb22PPnj3Svc6vynrSRJOoP3/+HEDmPSoAsn1QywcffKD1PSwsDM7OzlqJNfCym6hmXDlNW3PQ8eq9eq+Wa9r0JrVq1UKzZs20Pq9OP2u7NW2MiopCUlKSVnnWp77fv38fQgiULVtW60DM3t4et2/flh7K4eHhgZEjR2LVqlWws7ODv78/lixZku3fgOhdysjIwObNm+Hn54eQkBAEBQUhKCgItWvXRkREBI4cOQIgMz5yE+dAZnfSjh07wsrKCpaWlrC3t5ce9JR1HZfL5VIctmjRAv3798fhw4cRFxeHCRMmSPXCwsJyjE3N8FfbmfWBW1nrDRw4EOXKlUPLli3h6uqKPn36SAfLRPmh2bclJibqDFuxYgUOHTqE9evXZ/tbQ0PDbLugP3z4UEpEzc3NYW9vj8aNGwN4GUt52Q9nFRkZidjYWKxcuVJnnxUYGAhA92FSbzouyC87OztpW9C6dWt89dVXWLVqFU6fPi290SG3++nIyEgkJyfnWE+tVksXjaZOnYrY2FiUK1cOlStXxpgxYwrkdXe5OX4yMDDQOY74L96SQUVTzZo1sX37djx//hznzp3DhAkTkJCQgC5durzx9aQpKSmYOHGi9EwoOzs72NvbIzY2Vmu/HRwcnOs3HA0fPhznz5/H4cOHpe7ouZHdBdsSJUpobWPCwsKyjZXiGD/5uud8ypQpOHbsGKpVq4aGDRtCJpPh+++/x7fffoszZ87gww8/xFdffVXQbS12atWqJT2FvEOHDmjQoAE++eQT3L17VzqTv379egQEBKBDhw4YM2YMHBwcIJfLMWvWLK0Hx2nk9BRFkeUBbu9CTtPWZ5uyynq/rVqthkwmw759+7Jt56tXVObOnYuAgADs3LkTBw8exNChQzFr1iycPXs2V/cJEhWEo0eP4t9//8XmzZuxefNmneEbNmxAixYtcj2+2NhYNG7cGJaWlpg6dSq8vLygVCpx6dIljBs3LtsHyGXl6uqKDz74ACdPnszTvOSFg4MDrly5ggMHDmDfvn3Yt28fVq9ejV69emX78DiiN7GysoKTkxNu3LihM0zTwyunhxQpFAqdE0oZGRlo3rw5YmJiMG7cOJQvXx5mZmYIDw9HQEBArmLpTTTj+PTTT9G7d+9s61SpUkXr+3+5D27atCkA4OTJk+/sinKjRo0QHBws7YtXrVqF+fPnY/ny5fj888/zPd7CdKxCxYuxsTFq1qyJmjVroly5cggMDMSWLVte+waWIUOGYPXq1Rg+fDjq1q0LKysryGQydO/ePd/bmvbt22Pz5s347rvv8Ouvv77xLRUajJ28yVdybmVlhbNnz2Lu3LnYunUrlEolTpw4AS8vL0yaNAljxozJ9qFClH+ahNvPzw+LFy/G+PHjAQBbt26Fp6cntm/frtX9JL+vTNK8pzW77mB3797VqXv48GEkJCRoXT3XdHPV1ztfNTTTz9puILONdnZ2b3y1jZeXF4QQ8PDwQLly5d44zcqVK6Ny5cr45ptvcPr0adSvXx/Lly/H9OnTAeh2ESIqaBs2bICDg0O23Wy3b9+OP/74A8uXL4ebm1uu4vz48eOIjo7G9u3b0ahRI6k8JCQkT+1SqVRaVyDd3NxyjE3NcM2/165dg1qt1joQyG47Y2xsjLZt26Jt27ZQq9UYOHAgVqxYgW+//RZlypRh/FGetW7dGqtWrcK5c+e0bgHJj+vXr+PevXtYu3YtevXqJZVnvRUjL/vhrOzt7WFhYYGMjAypF1lBKKjYUalUAF72RsjtflqpVMLU1DTHegYGBlq98GxsbBAYGIjAwEAkJiaiUaNGmDx5spScv4ttgZubG9RqNUJCQrR6Pbz65Huit6W5aPfvv/8CyHld3rp1K3r37o25c+dKZS9evEBsbKxWPS8vr2xPQGanQ4cOaNGiBQICAmBhYYFly5blYw6y5+bmlm2sFMf4yVe3diDzCuM333yDK1euICkpCSkpKbhx4wYmTpzIxPwd8fX1Ra1atbBgwQK8ePECwMuzUa+effrnn39w5syZfE3DyckJ1apVw9q1a7W6vRw6dEinC02rVq2QkZGBxYsXa5XPnz8fMpkMLVu2zFcbCsqr8/LqxujGjRs4ePAgWrVq9cZxdOrUCXK5HFOmTNE5wyeEkN5WEB8fLx10aFSuXBkGBgZITU2VyszMzHQ2jEQFJSUlBdu3b0ebNm3QpUsXnc/gwYORkJCAXbt2oVWrVjh79izOnTsn/T4yMlLnlWvZbWPS0tKwdOnSXLfr3r17uHv3LqpWrSqVtWrVCufOndPaViUlJWHlypVwd3eX7q1t1aoVnj59it9++02qp1KpsGjRIpibm0tdgrO+OcTAwEC6QqiJQc3JOMYg5dbYsWNhamqKPn36ICIiQmd4Xq78ZBdLQgjpdX8aedkPZzeNzp07Y9u2bdkecEdGRua6va8yMzMrkNu0/vzzTwCQtgW53U/L5XK0aNECO3fu1OqtEBERgY0bN6JBgwawtLQEoLstMDc3R5kyZXT2xUDBbgv8/f0BQGfbuGjRogKbBhUfx44dy3b7onkOg+YWj5yOK+Vyuc7vFy1apPMK1M6dO+Pq1avZvqoxu+n36tULP/74I5YvX45x48blen7exN/fH2fOnMGVK1ekspiYmGxfA1vU5evKOenPmDFj0LVrV6xZswZffPEF2rRpg+3bt6Njx45o3bo1QkJCsHz5cnh7e2d7n1xuzJo1C61bt0aDBg3Qp08fxMTEYNGiRahYsaLWONu2bQs/Pz98/fXXCA0NRdWqVXHw4EHs3LkTw4cPLxSvCZszZw5atmyJunXrom/fvtIrWqysrHL1zlEvLy9Mnz4dEyZMQGhoKDp06AALCwuEhITgjz/+QP/+/TF69GgcPXoUgwcPRteuXVGuXDmoVCqsW7dOOlDS8PHxweHDhzFv3jw4OzvDw8Mjz6/oIcrJrl27kJCQgHbt2mU7vE6dOrC3t8eGDRuwYsUKrFu3Dh999BGGDRsmvUpNc6Vao169eihRogR69+6NoUOHQiaTYd26dTkmJSqVSroPV61WIzQ0FMuXL4dardbq0TN+/Hhs2rQJLVu2xNChQ2FjY4O1a9ciJCQE27Ztk66S9+/fHytWrEBAQAAuXrwId3d3bN26FX///TcWLFgg9dr5/PPPERMTgyZNmsDV1RVhYWFYtGgRqlWrJt2fXq1aNcjlcnz//feIi4uDQqFAkyZNpHfGEmVVtmxZbNy4ET169MAHH3yAnj17omrVqhBCICQkBBs3boSBgUGubl0qX748vLy8MHr0aISHh8PS0hLbtm3L9t7u3O6Hs/Pdd9/h2LFjqF27Nvr16wdvb2/ExMTg0qVLOHz4MGJiYvK8HHx8fPDbb79h5MiRqFmzJszNzdG2bdvX/iY8PFzaFqSlpeHq1atYsWIF7OzstLq053Y/PX36dBw6dAgNGjTAwIEDYWhoiBUrViA1NRWzZ8+W6nl7e8PX1xc+Pj6wsbHBhQsXsHXrVgwePFhrfgBg6NCh8Pf3h1wuR/fu3fO8XLIuo86dO2PBggWIjo6WXqWmeRAme+5QXgwZMgTJycno2LEjypcvj7S0NJw+fRq//fYb3N3dpWdI5HRc2aZNG6xbtw5WVlbw9vbGmTNncPjwYdja2mpNZ8yYMdi6dSu6du2KPn36wMfHBzExMdi1axeWL1+udVJdY/DgwYiPj8fXX38NKyurArmVeezYsVi/fj2aN2+OIUOGSK9SK126NGJiYopX/OTnEe+BgYFv/GR95QTlnuZVBOfPn9cZlpGRIby8vISXl5dQqVRCrVaLmTNnCjc3N6FQKMSHH34odu/eLXr37q312hPNq9TmzJmjM05k84qPbdu2iQoVKgiFQiG8vb3F9u3bdcYpROarj0aMGCGcnZ2FkZGRKFu2rJgzZ470+oVXpzFo0CCtspza9OorG/K7nF51+PBhUb9+fWFiYiIsLS1F27Ztxa1bt7TqaF6VEhkZme04tm3bJho0aCDMzMyEmZmZKF++vBg0aJC4e/euEEKIBw8eiD59+ggvLy+hVCqFjY2N8PPzE4f/1959h0dRrn0c/2167wXSCL0JAYMgqCCKoigeCxbUY0RFpIpioSgqoHBsoIAHbOBrOSIq6FGwwEFBRVGa0ltIICG9hySbZOf9Yw/hrAmCEphk8/1c115ezM7O3jPunWfu53lmZtUqh+3s2rXL6Nu3r+Ht7W1I4rFqqFeDBw82vLy8jNLS0hOuc+eddxru7u5GTk6O8euvvxr9+vUzvLy8jOjoaGP69OnGG2+8UeuxJd9//71x/vnnG97e3kZUVFTNI130u8eS1fUotYCAAOPSSy+tlQuGYRj79+83hgwZYgQFBRleXl5Gz549jc8++6zWepmZmcawYcOMsLAww8PDw+jSpYuxaNEih3U+/PBD4/LLLzciIiIMDw8PIy4uzhgxYoRx5MgRh/Vee+01o1WrVoarqyuPVcMp27dvnzFy5EijTZs2hpeXl+Ht7W106NDBuO+++4wtW7Y4rJuUlGT4+vrWuZ0dO3YYAwYMMPz8/IywsDBj+PDhxtatWw1JtX7Tp9oO19WGZ2ZmGqNHjzZiY2MNd3d3o1mzZsall15a8xhcwzhxW3usbf7feEpKSoxbb73VCAoKqvNxbr/3+0epubi4GBEREcbQoUMdHrN6zKm004ZhGJs2bTIGDhxo+Pn5GT4+Pkb//v2NH374wWGdGTNmGD179jSCgoJq/j89/fTTDo+gqqqqMsaOHWuEh4cbFovF4bFqvz+eJzo/qOuxUaWlpcbo0aONkJAQw8/Pz7j22muN3bt3G5KMWbNm/eExA/7XypUrjbvuusvo0KGD4efnZ3h4eBht2rQxxo4da2RmZtasd6Lzyvz8/Jp208/Pzxg4cKCxa9cuo0WLFrXOPXNzc40xY8YY0dHRhoeHhxETE2MkJSXVPPbwRH8rHnnkEUOSMW/ePMMwTvwotboebdivXz+jX79+Dss2b95sXHTRRYanp6cRExNjzJw503j55ZcNSUZGRsZfPJKNj8Uw/vzV+PHx8bV6MKqrq3XkyBFVV1crPDxcvr6+p/0YDgAAAKCx2rJli7p376533nmnST4WCjgd48eP18KFC1VSUnLCG8s5m780rf1EdyetrKzUwoULNWfOHJ41CwAAgCajrKys1n2X5syZIxcXF4cbagKo7ff5k5ubq7ffflsXXnhhkynMJekvjZyfzKhRo5SSkqLPP/+8vjcNAAAANDhPPfWUNm7cqP79+8vNza3msY7H7p0B4MS6deumiy++WB07dlRmZqbeeOMNpaena/Xq1U2qc+uM3BAuISFBb7/99pnYNAAAANDg9OnTR19//bWmT5+ukpISxcXF6cknn9SUKVPMDg1o8AYNGqQPP/xQr776qiwWi84991y98cYbTaowl87QyPmQIUO0bt26Oh87AgAAAAAAHP2lkfNp06bVubygoEBr167Vpk2bNHHixNMKDAAAAACApuIvjZwfe/7s7wUHB6t169a65557NHz48Kb1TDoAAAAAAP6iMzKtvamw2WxKT0+Xv78/HRFoMAzDUHFxsaKiok7YkYaTI7/REJHf9YP8RkNEftcP8hsN0anm95+e1l5WVqYpU6aof//+Gjx48GkF2dilp6crNjbW7DCAOh06dEgxMTFmh9Fokd9oyMjv00N+oyEjv08P+Y2G7GT5/aeLc29vby1cuFCdOnU6rcCcgb+/vyQpZVO8Avzo4TxdCR/fZXYITsFWXq7DT86o+X3iryG/69d17buaHYJTqDIq9Z0+J79PE/ldv/rMucfsEJxCtbVce16fRn6fJvK7fg05v2ndLf1MqTKs+jb/vZPm91+6IVxiYqK2bdv2lwJzJsemygT4uSjAn+Q/XS5eXmaH4FSYynV6yO/65WZxNzsE52GQ36eL/K5frp603/WJ/D495Hf9cnPxMDsE52Cz/+dk+f2XfrFz5szR+++/r9dff11VVVV/ZRMAAAAAAOC/TnnkfO3aterYsaPCw8OVlJQkFxcXjRgxQuPGjVN0dLS8vb0d1rdYLNq6dWu9BwwAAAAAgLM55eK8f//+eueddzR06FCFhoYqLCxM7du3P5OxAQAAAADQJJxycW4Yho49de2bb745U/EAAAAAANDkcJcEAAAAAABM9qeKc+4eCQAAAABA/ftTxfntt98uV1fXU3q5uf2lp7QBAAAAANDk/KkKesCAAWrXrt2ZigUAAAAAgCbpTxXnSUlJuvXWW89ULAAAAAAANEncEA4AAAAAAJNRnAMAAAAAYDKKcwAAAAAATHbK15zbbLYzGQcAAAAAAE0WI+cAAAAAAJiM4hwAAAAAAJNRnAMAAAAAYDKKcwAAAAAATEZxDgAAAACAySjOAQAAAAAwGcU5AAAAAAAmozgHAAAAAMBkFOcAAAAAAJiM4hwAAAAAAJNRnAMAAAAAYDKKcwAAAAAATEZxDgAAAACAySjOAQAAAAAwGcU5AAAAAAAmozgHAAAAAMBkFOcAAAAAAJiM4hwAAAAAAJNRnAMAAAAAYDI3swPAqfntR18tfSVCe3/zUV6mu554I1l9riysef/t55vpm0+ClJ3uLncPQ226lGnYxCPqcO7RmnUO7/fUa9OjtONnX1VVWtSyY5nueCRD3S4oqfV9RXmuGnlZe+Uc8dBHO3+TX2D1WdnPs63FtE1yz7fWWl5wQaRyhrQ8vsAw1PzVXfLdVagjd7VTaZcQSZL/hixF/utAndtOnpaoan/3MxI3nMf7cyP0/YogHdrnKQ8vmzr1OKq7p6Qrtk1FzTrWcotefSpK33warMoKixIvLtbYmYcVHF5Va3t/lLvWCovenR2p/3wUovxsN4VEVOm2BzI0cGjeWdlXs7m4GLp9QoYuvT5fweGVys1019dLQ/TenEhJFknShNkpuvymfIfP/bLGX1Nub21CxHAGJ2u//9dLj8ZoxdthGvFUmq4fnl2zfO+v3nrj6Sjt2eojF1dDFw4q0Ign0+Xta5Nkz/tZY1ooeae3ivNdFRhapd4DCzVs0hH5+tvOyn6aIcKvROP7/agLWqXKy61KhwoCNXVlf+3IiJAk3XfBz7qiwz418y9Rpc1FOzLCNW9dL/12JLJmGy2CC/TAxevVLTpD7q7V2psdqvnf9dTPqdFm7RYakZPld1mpi954urnWfxmoonw3NYu16m93Z+vqO3Jr1kk/6KHXpkVp+wY/VVotSuxfpNEz0mra+IxDHnpvdqS2fO+n/Gx3hUZW6pLr8zX0/ky5exhnfZ/PhpvuTlGfAdmKaXlU1nIX7dwaqDdnt1baQZ+adcZM3a3u5+cpJNyq8qOu2rE1UItmt9LhZN+adRJ65envY5IV37ZU5WWuWv1pM731ckvZqhvu+DTFeSNRftRFrTqXaeDQPE27u2Wt96NblWv004fVvIVVFeUuWvZquCYNba1FP+xQUKj95HxqUktFt6zQP5buk6eXTcteC9fUO1pq8fqdColwPMl/cUKcWnYsV84Rj7Oyf2Y59GAXWWzH/7B5HClT9IKdKu0W4rBe4LcZksVS6/Ml3cJ0tEOQw7KIf+2XS6WNwhyn5Nf1fhp8Z47adTuq6ipp8azmmjy0tV77dpe8fOwn1QuejNaGVQF6bOFB+QZUa/6UGE27O16zP91Xa3t/lLtPj4hXQY6bHnghVVEtrcrLdJNhq/27dlY3jc7S1Xfk6PnxcUrZ7aW2CWWa8GKqSotc9cmb4TXr/fwff73wYFzNvyutTecYof6drP0+5vuVgdq10VehzRw7jHMz3DTxltbqd02BRj99WEdLXLRgarSeHx+nx187KEmyuEi9BxbqzkePKDC0SunJnpo3OUbFBW6a9ErKmdw90/h7Vmjxbcv1S2qURi+9Svll3ooLLlRRuWfNOil5gZq56iIdLgiQl1uVbj9vq/5502ca/Oqtyi/zliTNvWGFUvIDNXzJNaqoctVtib9q7vUrdNVrtym31OdEXw9IOnl+L3wySlu+99cjc1MVGWvVpm/9NXdSjEIjK9V7YJHKj7po8tDWatWpTP9Yam/T33q2uaYmtdRLn+2Vi4t0aJ+nbDbp/n8cVlTLCh3c5aU5D8eq/KiL7n0i/Wzv8llxTo8CffZ+tPZsC5Crq6Gk+/fr6YVbNOLaXqooc5Uk7dvhr28+j1TWEU/5B1bptpHJmrFwq+66ordsNotativRtFd+1fuvtdALkzsqNLJCYx7fIxcXQ2+80MbkPTyxhtttcBbNnz9f8fHx8vLyUq9evbRhwwazQ6rlvEuKdeejGbrgBL3tl1xfoHP7lqh5C6vi25fr3ifTdLTYVck77I1PYa6r0g546aYxWWrVqVzRray6a8oRVZS56uAuL4dt/futUJUWuWrIfVlnfL/MZvNzV3WAR83Ld0e+rGGeKmsdULOOR1qpgr85oqxbWtX6vOHh4vB5w8Uin71FKuoVcTZ3A3+goef3M+8d0OU35ym+fblady7XhDmpykrz0N5f7blbWuSiL/8VohFPpqnbhSVq27VMD76Yqh2/+GnnRscTxz/K3Z/X+Ou3H/00/e0DOrdviZrFWtWpx1F17ll6VvazIejUo1TrvwzUhtWByjzsqe8+D9Kmb/3VvttRh/UqrRblZ7vXvEoK6cduqBp6fksnb78lKeeIu155LFqPzk+R2+9+bj+tCpSbm6ExzxxWbJsKte9WpnH/OKzvPg9SWrK9E84/qFqDk3LVLqFMkTGV6n5RiQYn5WjbT751fJtzuKvXZmUW+Wrqyku0LSNSaYUBWn8wVocLAmvWWbmznX5KiVFaYYD254bo+f9cIH9Pq9qG20ctg7zL1CKkUG/+1F17s0OVmh+kl9aeL2+PKrUJaxozihoyZ8jvHb/46rIb85TQx97uDro9V606lWn3Fnv7vX2DrzIPeWjCnFS17Fiulh3L9fBLKdq71UdbvvOzf0f/Yj0055ASLy5W8xZW9R5YpCH3Zen7lYF1fqczmDoyQas+aa7U/b5K3uOnFx/rqIioCrXtVFyzzhcfRmnbxiBlpXtr/05//d+8VopoXqGIqHJJUt8rMpW8x0//WtBSRw75aNsvwXrzxda6+pY0efvUnnnYUDT54nzJkiV68MEH9cQTT2jTpk1KSEjQwIEDlZXVeAvTSqtFK94JlW9AtVp1KpMkBYRUK6Z1uVYtDVH5URdVV0mfvx2qoLBKte1aVvPZlD2eem92Mz38UoosTe3XUWWT/8YcFfeMqBklt1ir1eztfcq+IV7VASefRRDwc7Zs7i4qSQg909HiFDTG/C4tsvcI+wfZZ7zs/dVHVZUu6n7R8ctP4tpWKCLaqp0bj594nyx3f/wqUG27HtXSVyJ067mddNeFHfTqU1GqKGs6o8I7fvFVtwuLFd3K3nC36lSmzj1L9fMaf4f1uvYu0ZKt2/T62p0aO/OQ/IMbbiPelDXG/K6LzSY9Oy5OQ0ZmKb59ea33KysscnM35PI/ee3hZZ9Vs32DX53bzM1w0/crg9S1d+3L1pxFvzYHtT0zQs9d86XWjF6kJUlLdX3XHSdc382lWjck7FBRuYf2ZNvb6IIyLyXnBmlw5z3ydq+Uq8WmIQk7lFvqrR0Z4SfcFs48Z8nvTj1K9eNXgco54i7DkLZ876e0A55K7GcvMiutFskih+np7p6GLC4nzm9JKi12rTlPaAp8/eztcPEJOss9vat12bVHdOSwl3Iy7LNn3D0MWSscT4isFS7y9LKpzf8U+Q1NUyu/annxxRc1fPhwDRs2TJ06ddKCBQvk4+OjN9980+zQ/rQfvw7Q39p00eCWXbXstXDNfH+fAv87pd1ikWYt2a/927x1bdsuurplgj5+NUJPv3ugJrmtFRbNHBWvex5PV0RMpZm7Ygq/3/LlUlalop7HG+Sw5Skqi/erucb8ZAJ+ylZJYpgMjyafWg1CY8tvm01a8ES0Op9XovgO9pP0vCw3uXvYat33ISi8UnlZ9kbqVHL3SIqHtv/sq4O7vTT1jYO676k0ffd5kOZOijmzO9WALJkXoW8/Cdbr3+7S5we3aP6Xu7Xs9XCtWXY8v39ZE6Dn7m+hR29urTeebq4u55fo6bcPyMXFOa/ra8waW36fyAfzI+Tqaujau3PqfD/hwhLlZ7tr6SvhqrRaVFzgqjefiZKkmr8Bx8wc2ULXtOqqW889Rz5+1Xrg+UNnPH6zxAQV6aZu25WaH6iRS6/WB1s669FLv9Pgzrsc1uvb+qDWj39NP094VX/v8avu+2CwCv47pV2y6N4lg9UhMkc/jH9dGya8qr+ft1Wjll6l4grP2l+Ks8ZZ8nvUjDTFtSvXbYmddVWLBD12WyuNfuawupxvn7XWIbFUXj42vfF0lMqPWlR+1EWvTYuSrdpSK7+PSUv20CdvhmvQ3+v+m+FsLBZDIx7dp+2bApWyz7HD4qqb0/TRT2u1bMNa9bgwT1OGd1NVlf0cfOP3IerYrVD9rsyUi4uh0IgK3XrfQUlSSHjt+001FE26grBardq4caMGDBhQs8zFxUUDBgzQ+vXra61fUVGhoqIih1dD0u2CEr3y9W7N/nSvelxcXHN9qSQZhjRvcoyCwqr0wrJ9evnzPepzRaGeuLOlcjPt6yya2Vxxbcp16Q35f/Q1Tivgpywd7RCk6kD7CLnPtjx57y1SznXxp/R5r4PF8sgsU1EvetsbgsaY3/Mmxyhll7cm/fPPXSN6Krlr2OyddBPnpahD96PqeWmx7n0yTauWhjSZ0fO+gwt0yfX5mjW6hUZf0V7Pj4/TkPuyNODG49NXv/00WD9+HaiDu7y1/ssgTU1qpfbdj6prH+cdgWyMGmN+12Xvr95a/nq4HpqTWtdtTSRJ8e3L9dCcFH20MELXtO6qod06q1msVcHhlbU+M+KpNM37creeXHRA6SkeWviU897UzMViaGdmmOauO1+7ssL10dZO+vjXTrqxm+Po+c+p0bpp8U26453r9H1yrJ675iuF+By7lMXQ5MvWKa/UW8Peu1a3vX2D1uxtqZdvWKkw36ZzyU9D4yz5LUmfvBmmXRt99NTiA5r3xW4Nn5qu+ZNjtGmtvcgMCq3WYwsP6qevA3Rt2666rn0XlRa5qk2Xo3XOgss54q4pt7VW36sLNOi2pnHpxagpe9SiTalmPdKp1ntrPo/U2Bt76JE7uyvtoLcmvbBN7h72wYzN60P05outNebx3fpk47d67d8/6ud19lkzRgO+T2aTvpAuJydH1dXVioyMdFgeGRmpXbt21Vp/5syZeuqpp85WeH+al49N0S2tim5pVcfEoxp2QUd98a8Q3TI2S1u+89OGVQH6cOdvNXdubdv1sDat7ahVH4To5rFZ2vKdvw7u8tKVsUH2Df53oOjGc87R0HGZuuPhDHN27Cxwy6uQ955CZQxrV7PMZ2+R3HPL1Wryzw7rNlu0R+Wt/JU2prPD8oAfs1QR7aOK2BNPQ8LZ09jye97kaP30dYBeWLZP4VHHR79DIqpUaXVRSaGrw+h5QbZ7zY0cTyV3QyKrFNqsUr4Bx1ukuLblMgyLco64K7pVw+1Fri/DH0+3j55/GixJOrjLWxExVt0yJlOrltY9OyYj1VMFua6Kiq/Qlu/861wHZ19jy+8T+e0nPxXkuOn28463J7Zqi157KkrLXwvX/22wF5qXXF+gS64vUH62m7x8bLJYpI9fDVfzFhUO2wuJqFJIRJXi2lbIP6haE65rq1vHZyg00vkuzcgu8dGB3GCHZQdygzSgneMTVMoq3XWoIFCHCgL125Fm+nT4e7q2yy69+dO56hmXpr6tU3TRy3ep1GrvmH/m63CdH39Y15yzW2/+dO5Z2x8c5yz5XVFm0eJZzTX1jYPqNcDeYdCqU7kObPfWhwsidG5fe6dv4sXFWrx+pwpzXeXqJvkFVuuWhM5qHueY37kZbnrkxtbq1KNU9z/nvLNi/tfIyXvUs1+uHrmzu3IzvWq9f7TETUdL3JSe6qNdWwP0wffr1OfSHH270v7bWfZ/cVr2f7EKCbeqpMhNkVHlGjb+gI4c9q61rYaiSRfnf9akSZP04IMP1vy7qKhIsbGxJkb0xwybVPnfay0qyuz/dfldL5yLxdCxm5U//nqyrOXHV9i9xUcvPhinF5btVVS8c5+4B2zIUrWfu0o7HW/o8y+NUtH5jjd2i3v2V+Vc20KlnR1PCCwV1fLbkqvcq+KExsms/DYMaf6UaP3wRaCe+3CfmsU55lrbrkfl5m7T5u/8dNFV9hvOHNrnqaw0D3VMtI/snErudj6vVOv+HaSyUpeaxy8d3u8pFxdDYc2bxmUsnt42GYbjUKOt2vKH99cIa25VQHC18jJ5+kJj1lDb7wE35OncixyvfZx8aytdekO+Lr+59qjYsUcrffmvELl72mpO7uti/Ldtr7Q65yTJLWnNFB9c4LCsRUih0ov+uIPcRYY83Owdnd7u9uNp+93fBcOwT6VF49BQ87uqyqKqSpdal0W5uBp1jtweuxR1y3f2TrvzLz8+AyDniLseubG12nYp04TZqbXO552PoZGT96r3JdmaeFd3ZaadQjFt0X+v3//9wbUoL9t+mUq/QZnKOuKp/Tsbbmd7ky7Ow8LC5OrqqszMTIflmZmZatasWa31PT095elpzjVIZaUuSk8+/t0Zhzy0f5u3/IOqFBBSrfdeilTvywsVElmpojw3fbooTDkZ7rpocIEkqWNiqfwCq/Xc/XG67YEMeXoZWvluqDIOeajnpfbk/30BXphn/3nEta1w2uecS5Jshvw3ZKv4vHDJ9XgDfewO7L9XFeypqlDH3ju/zbmSzVBxj7AzHi5OTWPJ73mTY7RmWbCeXHRA3n62mmvMfP2r5eltyDfApoFD8/Tqk9HyD6qWr7/9UWodE0vVMdE+NfNUcrf/dfl6d3akXnggTn9/6IiK8tz0+owoXX5Lnjy9m8ZJ6I9fB+iWcZnKSnNXym4vtT6nTNffm6Wv3rdPc/PyqdbtD2bouxVBys9yU/N4q+6Zkq70g57a+G3DbcibosaS39Ift98RMZUKCHFsX93cpOCIKsW2OT5q9smbYerUo1TevjZtWuuv16dH6a7J6TX5vWG1v/Kz3dW+21F5+dqUsttLr0+PUufz7HeIdkbv/JKgt25bprvP36ivdrXROc0zNaTrDk37qp8kydu9Uvecv1Hf7ItXTqmvgrzLdUv3bYrwL9XXu1pLkramR6qo3FMzBq3Wwh96qKLKTdcn7FB0YLHW7W9h5u41ac6U3117l+i16VHy8EpTZIxVv67306oPQ3TvE2k1n/ny/RDFtS1XYGiVdm701T+nRuu6e7Nr/gbkHHHXw0PaKCLaquFT01WYe7x8+/2jkJ3FqCl7dPGgLE27/xyVlboqONR+LEpL3GStcFWzmDL1HZilTetDVJjnrrDICt14d4qsFS41U9cl6YY7U7Xx+xDZbBZdMCBbN96dqlkPdZatAT9GtkkX5x4eHkpMTNTq1at17bXXSpJsNptWr16tMWPGmBvc7+zZ6qNHhhx/Jt/CJ+3XkV12U57GzTqkw/s8NX1pvIry3OQfXK12CUf1wrK9NXd+DQyt1tPv7dfiWc316E1tVF1pUYv25XpyUbJad659d9imxHtPodzzrad1rXjAT1kq7RIim3eTTqkGpbHk92dv2Tt0Hr6hrcPyCbNTa0bO7nsyTS4WQ9OHx6uywqIeFxdrzMzDf+p7vH1tmvn+fr3yWIzGXtFe/sFV6ntNge585Ej97Egj8MpjMUp65IjGPHNYQaFVys1014p3wvTubPv0N5vNopYdy3XZjcnyDahWbqabNn0boLeea+a0o4+NVWPJb+mP2++H5qSe0jZ2b/HR2y80U3mpi2LaVGjcs4c0YMjxe0x4/LfDfeGT0aq0WhQeZdUFVxbq5jGN687Wf8b2jAg9uHygxvX9SSP6bFRaob+e/c8FWrHDfnlatc2ilqEFuuacrxTkXaaCci9tPxKhYe9dq/259stYCsq8NerDqzT2og167ZZP5eZi0/6cEN3/8RXak01nu1mcKb8n/fOg3nymuf4xJk7FBW6KiLbqzkeP6Oo7cms+c3i/pxbNbK7iAldFxlo1dFymrr83u+b9TWv9lZ7sqfRkT92W6HhJ5ZfpW87sDprk6lvsz29/dtEWh+UvPtZBqz5pLmuFizonFuhvfz8kv4AqFeR6aNvGIE34e6IK844PrPW4MFc3D0+Ru4dNybv9NH1cF/3yXcN+opLFMIymMWRyAkuWLFFSUpIWLlyonj17as6cOfrggw+0a9euWte6/F5RUZECAwOVv6eVAvw5cTtdrZfcZ3YITsFWXq7UiY+psLBQAQEBJ/+AEyO/G46B0d3NDsEpVBmV+sZYTn6L/G5IEp4dZXYITqG6olw7X5lMfov8bkgGdbnE7BCcQpXNqtV5i0+a301+mO/mm29Wdna2pk6dqoyMDHXr1k1ffPHFSRMfQMNHfgPOi/wGnBf5jaaqyRfnkjRmzJgGN00GQP0gvwHnRX4Dzov8RlPEXA8AAAAAAExGcQ4AAAAAgMkozgEAAAAAMBnFOQAAAAAAJqM4BwAAAADAZBTnAAAAAACYjOIcAAAAAACTUZwDAAAAAGAyinMAAAAAAExGcQ4AAAAAgMkozgEAAAAAMBnFOQAAAAAAJqM4BwAAAADAZBTnAAAAAACYjOIcAAAAAACTUZwDAAAAAGAyinMAAAAAAExGcQ4AAAAAgMkozgEAAAAAMBnFOQAAAAAAJqM4BwAAAADAZBTnAAAAAACYjOIcAAAAAACTUZwDAAAAAGAyinMAAAAAAExGcQ4AAAAAgMkozgEAAAAAMBnFOQAAAAAAJqM4BwAAAADAZBTnAAAAAACYjOIcAAAAAACTUZwDAAAAAGAyinMAAAAAAExGcQ4AAAAAgMnczA7AGVzXrovcLO5mh9Hote+cb3YITqGqukKpZgfhRIZcOVhurp5mh9H4GQfMjsA5GIbZETgV2u/6EROTYnYITqHKVqGdZgfhRMjv+uEaVG12CM7BOLXjyMg5AAAAAAAmozgHAAAAAMBkFOcAAAAAAJiM4hwAAAAAAJNRnAMAAAAAYDKKcwAAAAAATEZxDgAAAACAySjOAQAAAAAwGcU5AAAAAAAmozgHAAAAAMBkFOcAAAAAAJiM4hwAAAAAAJNRnAMAAAAAYDKKcwAAAAAATEZxDgAAAACAySjOAQAAAAAwGcU5AAAAAAAmozgHAAAAAMBkFOcAAAAAAJiM4hwAAAAAAJNRnAMAAAAAYDKKcwAAAAAATEZxDgAAAACAySjOAQAAAAAwGcU5AAAAAAAmozgHAAAAAMBkFOcAAAAAAJiM4hwAAAAAAJNRnAMAAAAAYDKKcwAAAAAATEZxDgAAAACAySjOAQAAAAAwGcU5AAAAAAAmozgHAAAAAMBkFOcAAAAAAJjMzewA8Nec06tEN47KVtsuRxXarEpP3hWv9V8E/s8ahu54OFNX3Jorv4Bq7fjFVy9PjFF6sqckKTLGqlsfyFS3C0oUHF6p3Ex3/efjYP3rpQhVVTadPpubbtmpPhceVkxssawVrtq5I1Rvvt5VaYcDHNbr0DFHScO2qX2HXNlsFh3YH6THJvWV1eqmLl2z9I8Xvqlz+/ePHqC9e0LOwp7Amdx022716ZuumLgSWStctHNbqN5c2Flph/xr1mkWVaJ7Rm1T5y65cne3aeOGSP3zpa4qyPeSJEU0K9XQO3Yr4dxsBYeUKy/HW//5OlZL3m6vqqqmk+O/99ZPO9QstrLW8k8Xh+rDVyL0fxt21vm5Gfe20LrPgs5wdGgKTtZ+3z4hQxf/rUDhUZWqtFq07zdvLZrVTLs3+9as4x9UpVEz0tTrsiIZNum7FUH65+NRKj/qasYumeLGpH3q0z9TMS1K7O33b8FaNLe90lL9atYZM/E3deuZq5CwcpWXuWnnr0FaNK+DDqccXyfhvBz9fcQetWhdrIpyV63+PEZv/bOdbNVN9+8k/rqTn58fN27WYV11R64WTI3SstfDa73v7mHTS5/vVevO5Rp5WTsd2O59psNvMG4anqo+A3IU06pM1nIX7dwSoDdfaKm0gz6SJL/ASt0+JkXn9slXePMKFea7a/3qUL39cryOlhwvb1fsWFtr27MmdNDalRFnbV/+rCb9l2ft2rUaPHiwoqKiZLFYtHz5crNDOmVePjYd2O6leZNj6nz/ptHZ+ttd2Zo7MUb3X91W5Udd9Mx7B+TuaZMkxbYpl4uLoZcejdG9/dtr4ZNRuurvuRo2KeNs7obpzumarc8+baMHx12qKRP7ydXN0NOz1srTq6pmnQ4dczR95jpt2hip8WMH6P4xA/TvT9rKZlgkSTt3hOq2mwY7vL5Y0VJHjvhq755gs3atyWvM+X1OQo4+W9ZKD47spykTLpSrm01PP/99ze/S06tKTz//gwxDmvTAhXpoTF+5udn0xMwfZbEYkqTYuBK5uBia+3w3jUwaoFfnddGga5KVNHy7mbtmunFXttMtCZ1qXhNvbiVJWvfvIGWnuzu8d0tCJ/3fc5E6WuKin//jf5It42xqzPl9svY77YCn5k+J1ohL2mnCtW2UcchDM/91QIEhx9ulR+elqkX7ck26pZWmJrVUl14lGv/c4bO1Cw1Cl3Pz9PnSFppwdx89Nran3FxtmjF3g0P7vW9XoGZP76r7bu6rx8edJ4tFmj53g1xc7H8nW7Yt0lOzf9HG9eEa9/cLNWtyd/W6KFPDRu82a7cg587vY/pcUagOiaXKOXLicdK7Hzui3Az3+g6xUTinR6E++1eUHhzaTVPu6WI/P3/9N3l6V0uSQsOtCg236vXnWmnk3xI1e3I79bgwX+On76m1rRcnt9Ntfc+vea1fHXa2d+dPadLFeWlpqRISEjR//nyzQ/nTflkToLeeba4f6uyNM3TtPdn610uRWv9loJJ3euvZcXEKjaxUnysK7Z//JkAvPBCnTd/6KyPVUz9+FagPF4TrgisLz+6OmGzq5L5a9VVLpaYEKvlAkF587jxFRB5V27b5NevcO3KLPl3WRkuXdFRqSqDSDgdo3dpYVVXaRyiqqlyVn+9d8yoq8tT5vdO16st4SRZzdgyNOr+nPnKBVn3RQqkHA5S8P1AvzkxURLMytW1XIEnqdE6uIpqV6sWZiTp4IFAHDwTqhZmJats+XwnnZkuSNm6I1OxZidr8S6Qyjvjqpx+a6+MlbdSnb7qJe2a+wjw35We717x6DShSerKHfl3vK5vN4vBefra7+lxZqLX/DmpSI5KNQWPO7z9uv6U1y4K1eZ29bU7Z46VXn4ySb4BNLTuVSbJ3rp93SbFmT4jV7s2+2r7BT688Fq1+fytQSGTtWSHOaur9PbXq8xilHvBX8t4AvTitqyKal6tNx6Kadb5YHqftm0OUdcRH+3cH6v8WtFNEs3JFND8qSbpowBEl7/PXv95oqyOHfbVtc6jenNtBVw1JkbdP1Ym+GmeYM+e3JIU2q9SoGWn6x+gWqqqq+zyxR/8iJfYr1mvTos5UqA3a1BFdtGp5M6Xu81Xybj+9OLmdIqIq1LZTsSQpZZ+vnh7fSRu+CVXGIW9t/SlYb70Ur179c+Xiajhsq7TYTfk5HjWvSmvDLn+b9LT2K6+8UldeeaXZYdS7ZnFWhUZWadO64yM9R4tdtWuzjzomHtW3n9Q9muvrX63igqZ9Aurraz+xKS72kCQFBpWrQ8c8rVndQs/PWa3mUSU6fChAb715jnZsrz0FSZLO750u/wCrvvqy5VmLG7U5U377+jn+Lt09bJJhUeX/XIJitbrIsFnUuUuutmyse7qWr2+VSoo8znzAjYSbu02X3JCvjxeGq66OtDZdjqrNOeWaf5IREJx9zpTff8TN3aZBt+eqpNBFB3bYp7R27FGq4gJX7f3Vp2a9Tev8ZdikDt2P/mFR4Mx8/ezFdElh3SONnl5VumzwYWWkeSsn034s3T1ssv7uRN1a4SpPL5vadCjUb5tCz2zQqJMz57fFYuiRl1P14T/DlbLHq851gsIqNf65w3rqrnhVlDXsQvJs8fW3j5gXnyC/JfvfgKMlbrJVO7bnIx/bp3HT9ijjkLdWfNBcX38cqYY8eMb/cScUEmFvoAqyHfteCrLdFBJRd696VHyF/nZXjla83XQbIovF0IiRW7R9W5hSDtpPbpo1L5Uk3XbHdn25spUen9RX+/YGaeaz3yoqurjO7Vx+5QFt2hip3ByfOt8H/gyLxdCIMb9q+68hSkm23wth1/YQlZe76q4R2+XpWSVPryrdM2qbXN0MBYeW17md5tElGnz9fq34d/xZjL5h63NFkfwCqvXVB3XfF+KKoXlK2eOpHb/41vk+cKb0GlCk5Xt/07+Tf9N1w7M16ZbWKsqzt+kh4VUqyHVs323VFhUXnLiNd3YWi6F7H9yh7VuClXLA8RKUq25I0YfffKmP136lxN7ZmjKmZ819Nzb9GKaOXfLV7/J0ubgYCg0v19B79kqSQsIqzvp+wPndNDpL1dXS8jdONLXa0ENzDunzt0MdOuCaMovF0IiJ+7V9Y4BS9tXdHgcEVWroyFStXNrMYfnbL7fQrAc7aso9XfX912Ea/fheXXN7w55B2KRHzv+siooKVVQc/2NdVFT0B2s3HqHNKvX0uwe09rMgrXyv6Rbno8ZuUov4Qj30wCU1y1z+e/3uys9b6ev/joQf2B+sbt2zdPnAZC1+s6vDNkLDjurcxEzNmtH77AWOetFQ83vUA1vVomWxHhrbt2ZZUaGnnnmip8Y8uFXX3LBfhs2ib/8To727g2QYtXuDQ8PKNP3ZH/TdN9H68jNmdBwzcGiufl4ToLzM2j3xHl429b8uX+/NiTQhMtS3hprfJ7Lle1+NuqydAkKqdOVteZqyMEXjrmqjwtymef3pyYx8ZLtatCrRw/eeX+u9NV9EafOGMAWHVeiG2w5o0jOb9dDw3qq0umrzT+F6c24HjZ64TROe3KrKShe9/0YbndM9XzabCTuCv6Sx5HebLkd17T05Gj2wnU40cvu3u3Pk7VetJXMb7g3LzrZRj+9Ti7aleuj2bnW+7+1bpacWbFPqfh+9O7+Fw3v/WnD83wd2+snLu1o3DDukT9+JPpMhnxZGzv+EmTNnKjAwsOYVGxtrdkh1ysuy97kEhTteLxUUXqW8LMeGPSSyUs8u3acdv/jqpYeb7tTNkWM2qWevdE18+GKHEe+8PPuUo9QUx2mCh1IDFB5xtNZ2Lh+YrOIiD/24vmleI9SYNcT8Hnn/VvXsnaGJ4y9UbrbjXVo3/xKpu2+9XLdeO0i3/G2Qnn+6h0LDypSR7tjTHhJapllz1mnn9hC9/Hz3sxl+gxYRbVX3i0r0xXt1j5pfdFWBPL0NrVrK0xacQUPM7z9SUeaq9IOe2rXJV7MnxKq6yj6TQ5Lyst0UFOrYvru4GvIPqt3GNwX3PbRdPS/M0qRRvZSbVftu1kdL3ZV+yFfbN4fomYnnKia+VH0uzqx5f/l7rXTTJZfpzmv6a+jlA/TjWnuHXEYao5aNRWPJ7y69ShUUVqV3ft6hFalbtSJ1q5rFVmr4E+l666cdkqRuF5SoY+JRfXbwV61I3apFP9ifIDJv5R49NCfVzPBNMXLKPvXsl6uJd3ZVbqZnrfe9fao0/dVtOlrqquljO6v6JE+j2f2rv8KbW+Xm3nB73yjO/4RJkyapsLCw5nXo0CGzQ6pTRqqHcjPd1P3C49Ouffyq1aH7Ue3ceLyxCW1Wqec+3Ke9v/nohQdi6xxxc36GRo7ZpN4XpGnSIxcrM8PP4d3MDF/l5HgrJsaxFzY6plhZWb9vuA0NGHhQq1e1UDWPYGl0GlZ+Gxp5/1b1vihdk8ZfqMyME0+rLir0VGmJhxK6ZysouEI/ft+85r3QsDL946V12rsnWLNnJTbRHK/b5bfkqSDHTT+tCqjz/YFD8/TjVwEqzGOCmTNoWPn951lcJHdP+0yunb/4yj+oWm26HO8g7nZhiSwu0q7NTamgNHTfQ9vV++IMTR7VS5npp7DvFkOyGHKvdWJuUV6Ol6wVrup3ebqyMry0f3fTvHa/MWos+b3qo2Ddd2k7jbzs+CvniJs+/Ge4ptxqf3LIK49Ha+SA4+8/9nf78mfua6HF/2j2R5t3MoZGTtmn3gNyNOmuBGWm1e548/at0ozXf1NVpUXTRnc+pRu9tepYquJCtwb92GjOOv4ET09PeXrW7rUxg5dPtaJaWmv+3SzWqlady1Rc4KrsNA8tfz1cQ+/PUlqypzJSPZT0SIZyM91rbhRzrDDPSvPQa9OiFPg/vfD52U2n533U2E26+JJUTXviApUddVNwsP1uuKWl7rJa3SRZ9NEH7XV70nYdOBCkA/uDNOCyg4qJLdbT0/o4bCuhe5aaNy/VlytbmbAnOF0NKb9HPbBVF196WNOmnK+yMjcFh9ivIy8tcZfVar9p42VXpig1xV+FBR7q2DlPI8b+quVL29Q8Cz00rEyzXlqnrAwfvfHKOQoMOj7lLz+v7pvQNBUWi6HLb87TqqXBtW4cI9nvwdHl/FI9fjuXADiLhpTff9R+F+W56tb7s7T+K/vlFgEhVbpmWI7CmlVq3b+DJEmH9nnp5//4a/zzhzX30Ri5uhsaPeOwvv0kqM5LNJzVqEe2q9/AdE1/KNHefofa/8aVlrjJWuGqZlFHddFl6dr8U7gK8z0UFlGuG5P2y1rhqp9/OH5D1+tvP6CN68NkGBb1uThDQ5L2a9bk7rLZ6MxsLBpLfmeneag437H0qqqyKD/LXYf329vl7DTHm7aWl9q3lZ7iqZwjTeeGrqMe36eLr8rStDGdVVbqquAw+3EoLXaVtcJV3r5V9keredn03KMd5ONXLR8/+03jCvPcZbNZ1PPiXAWHWrVra4CsVhd1752vm4en6qPFDXumcJMuzktKSrRv376afycnJ2vLli0KCQlRXFyciZGdXLuEMj330f6af9/3lP3mBl8tCdYLD8Tpg/nh8vKx6f5nD8svoFrbf/bVlNtaqbLC3lN0bt9iRbeyKrqVVe9t2uGw7YFRCWdvR0x29TX2Y/jsC984LH/xufO06iv7ifkny9rJw6Na9963Rf7+Vh04EKQpj/ZVxhHHUfaBVyRrx/ZQHT5U90gczq7GnN9XX5ssSXr25XUOy1+cea5WfWG/fio6tlhJw7fLP8CqrAwfLXmnvZZ90KZm3e49shQdU6romFK9/dEXDtsZ1O+6M7wHDVv3viWKjKnUl+/XfY+NgbfkKeeIuzZ+y7PNG6rGnN9/1H6/PDFGMW0q9PiNBxUQUq3ifFft2eqjCde1cbiz8z/GxGn002ma9cF+GTbpuxWBeuWxhnsN5Zlw1RD7FN9/LPzJYfnsp7pq1ecxslpd1Llbvv52y0H5BVSqIM9T2zaH6KG7e6sw/3gh16NPtm4etk/u7jYl7w3Q9IcStXE91/uayVnz+4UHGnbsDcnVQ49Ikp79v18dlr84uZ1WLW+mNp1K1CHBPkP4zS9/dljnzgE9lZXupeoqi66+NV3DJx6QxWIoPdVbrz3bSl8sba6GzGIYhnHy1ZzTN998o/79+9danpSUpMWLF5/080VFRQoMDNTF+pvcLE2nt/pMce3c3uwQnEJVdYVW73xehYWFCghouh0F9ZXfl7a+X26uDaNHvjGr3nvA7BCcQpVRqW/0CflN+92guMU0rY6BM6XKVqFVaQvIb/K7QXEN4hKP+lBlWLW64O2T5neTHjm/+OKL1YT7JgCnRn4Dzov8BpwX+Y2mrOFeDQ8AAAAAQBNBcQ4AAAAAgMkozgEAAAAAMBnFOQAAAAAAJqM4BwAAAADAZBTnAAAAAACYjOIcAAAAAACTUZwDAAAAAGAyinMAAAAAAExGcQ4AAAAAgMkozgEAAAAAMBnFOQAAAAAAJqM4BwAAAADAZBTnAAAAAACYjOIcAAAAAACTUZwDAAAAAGAyinMAAAAAAExGcQ4AAAAAgMkozgEAAAAAMBnFOQAAAAAAJqM4BwAAAADAZBTnAAAAAACYjOIcAAAAAACTUZwDAAAAAGAyinMAAAAAAExGcQ4AAAAAgMkozgEAAAAAMBnFOQAAAAAAJqM4BwAAAADAZBTnAAAAAACYjOIcAAAAAACTUZwDAAAAAGAyinMAAAAAAExGcQ4AAAAAgMnczA6gMTMMQ5JUpUrJMDkYJ2BUV5gdglOo+u9xPPb7xF9Tk982fpf1odqoNDsEp1Al+3Ekv08P7Xc94+9kvaiyWSWR36eL/K5fhmE1OwSnUGWcWn5TnJ+G4uJiSdJ3WmFyJE5ip9kBOJfi4mIFBgaaHUajdSy/v01eYHIkQG3k9+mh/a5naWYH4FzI79NDftezArMDcC4ny2+LQffcX2az2ZSeni5/f39ZLBazwzmhoqIixcbG6tChQwoICDA7nEarsRxHwzBUXFysqKgoubhw5cpfRX43LY3lOJLf9YP8bloay3Ekv+sH+d20NJbjeKr5zcj5aXBxcVFMTIzZYZyygICABv2jbSwaw3Gkx/30kd9NU2M4juT36SO/m6bGcBzJ79NHfjdNjeE4nkp+0y0HAAAAAIDJKM4BAAAAADAZxXkT4OnpqSeeeEKenp5mh9KocRzREPG7rB8cRzRE/C7rB8cRDRG/y/rhbMeRG8IBAAAAAGAyRs4BAAAAADAZxTkAAAAAACajOAcAAAAAwGQU5wAAAAAAmIzi3MnNnz9f8fHx8vLyUq9evbRhwwazQ2p01q5dq8GDBysqKkoWi0XLly83OyRAEvldH8hvNFTk9+kjv9FQkd+nz1nzm+LciS1ZskQPPvignnjiCW3atEkJCQkaOHCgsrKyzA6tUSktLVVCQoLmz59vdihADfK7fpDfaIjI7/pBfqMhIr/rh7PmN49Sc2K9evXSeeedp3nz5kmSbDabYmNjNXbsWE2cONHk6Boni8WiZcuW6dprrzU7FDRx5Hf9I7/RUJDf9Y/8RkNBftc/Z8pvRs6dlNVq1caNGzVgwICaZS4uLhowYIDWr19vYmQAThf5DTgv8htwXuQ3Tobi3Enl5OSourpakZGRDssjIyOVkZFhUlQA6gP5DTgv8htwXuQ3TobiHAAAAAAAk1GcO6mwsDC5uroqMzPTYXlmZqaaNWtmUlQA6gP5DTgv8htwXuQ3Tobi3El5eHgoMTFRq1evrllms9m0evVq9e7d28TIAJwu8htwXuQ34LzIb5yMm9kB4Mx58MEHlZSUpB49eqhnz56aM2eOSktLNWzYMLNDa1RKSkq0b9++mn8nJydry5YtCgkJUVxcnImRoSkjv+sH+Y2GiPyuH+Q3GiLyu344bX4bcGpz58414uLiDA8PD6Nnz57Gjz/+aHZIjc6aNWsMSbVeSUlJZoeGJo78Pn3kNxoq8vv0kd9oqMjv0+es+c1zzgEAAAAAMBnXnAMAAAAAYDKKcwAAAAAATEZxDgAAAACAySjOAQAAAAAwGcU5AAAAAAAmozgHAAAAAMBkFOcAAAAAAJiM4hwAAAAAAJNRnAMAAAAAYDKKcwAAAAAATEZxDgAAAACAySjOAQAAAAAwGcU5AAAAAAAmozgHAAAAAMBkFOcAAAAAAJiM4hwAAAAAAJNRnAMAAAAAYDKKcwAAAAAATEZxDgAAAACAySjOAQAAAAAwGcU5AAAAAAAmozgHAAAAAMBkFOcAAAAAAJiM4hwAACcUHx+vO++8s962d+eddyo+Pr7etgcAABxRnAMAcAYsXrxYFoul5uXl5aV27dppzJgxyszMNDu8OqWnp+vJJ5/Uli1bzA4FAIAmx83sAAAAcGbTpk1Ty5YtVV5eru+++07//Oc/tWLFCm3btk0+Pj5mh+cgPT1dTz31lOLj49WtWzeH91577TXZbDZzAgMAoAmgOAcA4Ay68sor1aNHD0nSPffco9DQUL344ov65JNPNHTo0Frrl5aWytfX92yHeVLu7u5mhwAAgFNjWjsAAGfRJZdcIklKTk7WnXfeKT8/P+3fv1+DBg2Sv7+/brvtNkmSzWbTnDlz1LlzZ3l5eSkyMlIjRoxQfn6+w/YMw9CMGTMUExMjHx8f9e/fX9u3b6/zuwsKCvTAAw8oPj5enp6eiomJ0R133KGcnBx98803Ou+88yRJw4YNq5mOv3jxYkl1X3NeWlqqCRMmKDY2Vp6enmrfvr2ef/55GYbhsJ7FYtGYMWO0fPlynXPOOfL09FTnzp31xRdfnO7hBADAaTByDgDAWbR//35JUmhoqCSpqqpKAwcO1IUXXqjnn3++Zqr7iBEjtHjxYg0bNkzjxo1TcnKy5s2bp82bN+v777+vGcmeOnWqZsyYoUGDBmnQoEHatGmTLr/8clmtVofvLSkp0UUXXaSdO3fqrrvu0rnnnqucnBx9+umnOnz4sDp27Khp06Zp6tSpuvfee3XRRRdJkvr06VPnfhiGoWuuuUZr1qzR3XffrW7duunLL7/Uww8/rLS0NM2ePdth/e+++04ff/yxRo0aJX9/f7388su64YYblJqaWnMsAABo0gwAAFDvFi1aZEgyVq1aZWRnZxuHDh0y3n//fSM0NNTw9vY2Dh8+bCQlJRmSjIkTJzp8dt26dYYk491333VY/sUXXzgsz8rKMjw8PIyrrrrKsNlsNetNnjzZkGQkJSXVLJs6daohyfj4449rxXrssz///LMhyVi0aFGtdZKSkowWLVrU/Hv58uWGJGPGjBkO6w0ZMsSwWCzGvn37apZJMjw8PByWbd261ZBkzJ079wRHEACApoVp7QAAnEEDBgxQeHi4YmNjdcstt8jPz0/Lli1TdHR0zTojR450+MzSpUsVGBioyy67TDk5OTWvxMRE+fn5ac2aNZKkVatWyWq1auzYsbJYLDWfHz9+fK04PvroIyUkJOi6666r9d7/fvZUrVixQq6urho3bpzD8gkTJsgwDK1cubLWcWjdunXNv7t27aqAgAAdOHDgT383AADOiGntAACcQfPnz1e7du3k5uamyMhItW/fXi4ux/vG3dzcFBMT4/CZvXv3qrCwUBEREXVuMysrS5KUkpIiSWrbtq3D++Hh4QoODnZYtn//ft1www2nvT/HpKSkKCoqSv7+/g7LO3bs6BDbMXFxcbW2ERwcXOsaegAAmiqKcwAAzqCePXvW3K29Lp6eng7FumS/GVxERITefffdOj8THh5erzGeDa6urnUuN3538zgAAJoqinMAABqY1q1ba9WqVbrgggvk7e19wvVatGghyT7S3qpVq5rl2dnZtUakW7durW3btv3h9/6Z6e0tWrTQqlWrVFxc7DB6vmvXLofYAADAqeGacwAAGpibbrpJ1dXVmj59eq33qqqqVFBQIMl+Hbe7u7vmzp3rMAI9Z86cWp+74YYbtHXrVi1btqzWe8c+e+z56se2/0cGDRqk6upqzZs3z2H57NmzZbFYdOWVV550GwAA4DhGzgEAaGD69eunESNGaObMmdqyZYsuv/xyubu7a+/evVq6dKleeuklDRkyROHh4XrooYc0c+ZMXX311Ro0aJA2b96slStXKiwszGGbDz/8sD788EPdeOONuuuuu5SYmKi8vDx9+umnWrBggRISEtS6dWsFBQVpwYIF8vf3l6+vr3r16qWWLVvWinHw4MHq37+/pkyZooMHDyohIUFfffWVPvnkE40fP97h5m8AAODkKM4BAGiAFixYoMTERC1cuFCTJ0+Wm5ub4uPjdfvtt+uCCy6oWW/GjBny8vLSggULtGbNGvXq1UtfffWVrrrqKoft+fn5ad26dXriiSe0bNkyvfXWW4qIiNCll15ac0M6d3d3vfXWW5o0aZLuu+8+VVVVadGiRXUW5y4uLvr00081depULVmyRIsWLVJ8fLyee+45TZgw4cweHAAAnJDF4E4sAAAAAACYimvOAQAAAAAwGcU5AAAAAAAmozgHAAAAAMBkFOcAAAAAAJiM4hwAAAAAAJNRnAMAAAAAYDKKcwAAAAAATEZxDgAAAACAySjOAQAAAAAwGcU5AAAAAAAmozgHAAAAAMBkFOcAAAAAAJiM4hwAAAAAAJP9PyojVR7OAWsoAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1040x520 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ncols = 4\n",
    "nrows = 2\n",
    "\n",
    "fig, axes = plt.subplots(nrows=nrows, ncols=ncols, figsize=(ncols*2.6, nrows*2.6))\n",
    "axes = axes.flatten()\n",
    "\n",
    "fig.suptitle('Model comparison - Optimized Threshold (ROC_AUC)')\n",
    "fig.supylabel('True label')\n",
    "fig.supxlabel('Predction')\n",
    "\n",
    "for i, (model_name, model) in enumerate(zip(results['Model'], [lr_clf, dt_clf, voting_clf, bagging_clf, rf_clf, adaboost_clf, gb_clf, stacking_clf])):\n",
    "\n",
    "    disp = ConfusionMatrixDisplay.from_predictions(\n",
    "        test_df['target'],\n",
    "        preds[model],\n",
    "        #normalize='true',\n",
    "        ax=axes[i],\n",
    "        colorbar=False\n",
    "    )\n",
    "\n",
    "    axes[i].set_title(model_name)\n",
    "    axes[i].set_xlabel('')\n",
    "    axes[i].set_ylabel('')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "783e5fef",
   "metadata": {},
   "source": [
    "### 5.2. F1 optimized thresholds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "646ffb2a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LogisticRegression new threshold: 0.485\n",
      "DecisionTreeClassifier new threshold: 0.000\n",
      "VotingClassifier new threshold: 0.226\n",
      "BaggingClassifier new threshold: 0.209\n",
      "RandomForestClassifier new threshold: 0.206\n",
      "AdaBoostClassifier new threshold: 0.449\n",
      "GradientBoostingClassifier new threshold: 0.126\n",
      "StackingClassifier new threshold: 0.532\n"
     ]
    }
   ],
   "source": [
    "model_thres_results = {}\n",
    "\n",
    "for model in [lr_clf, dt_clf, voting_clf, bagging_clf, rf_clf, adaboost_clf, gb_clf, stacking_clf]:\n",
    "    TunedThreshCV = TunedThresholdClassifierCV(estimator=model, scoring='f1') \n",
    "    TunedThreshCV.fit(X=train_df.drop('target', axis=1)[:num_samples], y=train_df['target'][:num_samples])\n",
    "    \n",
    "    model_thres_results[model] = TunedThreshCV.best_threshold_\n",
    "    print(f'{type(model).__name__} new threshold: {TunedThreshCV.best_threshold_:.3f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "e7704432",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f1 optimized -  Logistic Regression 0.485\n",
      "f1 optimized -  Decision Tree 0.0\n",
      "f1 optimized -  Voting 0.226\n",
      "f1 optimized -  Bagging 0.209\n",
      "f1 optimized -  Random Forest 0.206\n",
      "f1 optimized -  AdaBoost 0.449\n",
      "f1 optimized -  Gradient Boosting 0.126\n",
      "f1 optimized -  Stacking 0.532\n"
     ]
    }
   ],
   "source": [
    "preds = {}\n",
    "for model_name, model in zip(results['Model'], [lr_clf, dt_clf, voting_clf, bagging_clf, rf_clf, adaboost_clf, gb_clf, stacking_clf]):\n",
    "    print(\"f1 optimized - \", model_name, round(model_thres_results[model], 3))\n",
    "\n",
    "    preds[model] = model.predict_proba(test_df.drop(['target'][:num_samples], axis=1))[:, 1] > model_thres_results[model]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "34dbecc7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1040x520 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ncols = 4\n",
    "nrows = 2\n",
    "\n",
    "fig, axes = plt.subplots(nrows=nrows, ncols=ncols, figsize=(ncols*2.6, nrows*2.6))\n",
    "axes = axes.flatten()\n",
    "\n",
    "fig.suptitle('Model comparison - Optimized Threshold (F1) Normalized')\n",
    "fig.supylabel('True label')\n",
    "fig.supxlabel('Predction')\n",
    "\n",
    "for i, (model_name, model) in enumerate(zip(results['Model'], [lr_clf, dt_clf, voting_clf, bagging_clf, rf_clf, adaboost_clf, gb_clf, stacking_clf])):\n",
    "\n",
    "    disp = ConfusionMatrixDisplay.from_predictions(\n",
    "        test_df['target'],\n",
    "        preds[model],\n",
    "        normalize='true',\n",
    "        ax=axes[i],\n",
    "        colorbar=False\n",
    "    )\n",
    "\n",
    "    axes[i].set_title(model_name)\n",
    "    axes[i].set_xlabel('')\n",
    "    axes[i].set_ylabel('')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "562c405a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1040x520 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ncols = 4\n",
    "nrows = 2\n",
    "\n",
    "fig, axes = plt.subplots(nrows=nrows, ncols=ncols, figsize=(ncols*2.6, nrows*2.6))\n",
    "axes = axes.flatten()\n",
    "\n",
    "fig.suptitle('Model comparison - Optimized Threshold (F1)')\n",
    "fig.supylabel('True label')\n",
    "fig.supxlabel('Predction')\n",
    "\n",
    "for i, (model_name, model) in enumerate(zip(results['Model'], [lr_clf, dt_clf, voting_clf, bagging_clf, rf_clf, adaboost_clf, gb_clf, stacking_clf])):\n",
    "\n",
    "    disp = ConfusionMatrixDisplay.from_predictions(\n",
    "        test_df['target'],\n",
    "        preds[model],\n",
    "        #normalize='true',\n",
    "        ax=axes[i],\n",
    "        colorbar=False\n",
    "    )\n",
    "\n",
    "    axes[i].set_title(model_name)\n",
    "    axes[i].set_xlabel('')\n",
    "    axes[i].set_ylabel('')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}