encoder for sentiment analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "train_dataset = pd.read_csv('./Corona_NLP_train.csv', encoding = \"ISO-8859-1\")\n",
    "test_dataset = pd.read_csv('./Corona_NLP_test.csv', encoding = \"ISO-8859-1\")\n",
    "\n",
    "to_drop = ['UserName', 'ScreenName', 'Location', 'TweetAt']\n",
    "\n",
    "train_dataset = train_dataset.drop(columns=to_drop).sample(frac=1, random_state=42) # shuffle rows\n",
    "test_dataset = test_dataset.drop(columns=to_drop)\n",
    "\n",
    "def clean_text(text):\n",
    "    text = text.lower()\n",
    "    hashtags = \"#[\\S]+\"\n",
    "    mentions = \"@[\\S]+\"\n",
    "    url = \"https?://[A-z0-9_%/\\-\\.]+[A-z0-9_\\.\\-\\?&=%]+\"\n",
    "    text = re.sub(url,\"\",text)\n",
    "    text = re.sub(hashtags,\"\",text)\n",
    "    text = re.sub(mentions,\"\",text)\n",
    "    puntuations = \"[\\.\\?!,;:]+\"\n",
    "    text = re.sub(puntuations,\"\",text)\n",
    "    return text\n",
    "\n",
    "map = {\n",
    "    'Extremely Negative': 0,\n",
    "    'Negative': 1,\n",
    "    'Neutral': 2,\n",
    "    'Positive': 3,\n",
    "    'Extremely Positive': 4\n",
    "}\n",
    "\n",
    "revmap = {\n",
    "    0: 'Extremely Negative',\n",
    "    1: 'Negative',\n",
    "    2: 'Neutral',\n",
    "    3: 'Positive',\n",
    "    4: 'Extremely Positive'\n",
    "}\n",
    "\n",
    "train_dataset['OriginalTweet'] = train_dataset['OriginalTweet'].apply(lambda x: clean_text(x))\n",
    "test_dataset['OriginalTweet'] = test_dataset['OriginalTweet'].apply(lambda x: clean_text(x))\n",
    "\n",
    "train_dataset['Sentiment'] = train_dataset['Sentiment'].apply(lambda x: map[x])\n",
    "test_dataset['Sentiment'] = test_dataset['Sentiment'].apply(lambda x: map[x])\n",
    "\n",
    "train_dataset, val_dataset = train_test_split(train_dataset, test_size=0.15, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "from torch import nn\n",
    "from prettytable import PrettyTable\n",
    "from transformers import AutoModel, AutoTokenizer\n",
    "from modules import Encoder, EncoderForSequenceClassification\n",
    "\n",
    "def count_parameters(model):\n",
    "    table = PrettyTable([\"Modules\", \"Parameters\"])\n",
    "    total_params = 0\n",
    "    for name, parameter in model.named_parameters():\n",
    "        if not parameter.requires_grad: continue\n",
    "        params = parameter.numel()\n",
    "        table.add_row([name, params])\n",
    "        total_params+=params\n",
    "    print(table)\n",
    "    print(f\"Total Trainable Params: {total_params}\")\n",
    "    return total_params\n",
    "\n",
    "model_name = 'prajjwal1/bert-tiny'\n",
    "tokenizer = AutoTokenizer.from_pretrained(model_name)\n",
    "device = 'cuda' if torch.cuda.is_available() else 'cpu'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch.utils.data import DataLoader, Dataset\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "ohe = OneHotEncoder(sparse_output=False)\n",
    "\n",
    "class TwitterDataset(Dataset):\n",
    "    def __init__(self, set) -> None:\n",
    "        super(TwitterDataset, self).__init__()\n",
    "        self.input = set['input_ids']\n",
    "        self.labels = set['labels']\n",
    "    \n",
    "    def __len__(self):\n",
    "        return len(self.labels)\n",
    "    \n",
    "    def __getitem__(self, index):\n",
    "        return {\n",
    "        'input_ids': self.input[index],\n",
    "        'labels': self.labels[index]\n",
    "    }\n",
    "\n",
    "def dset(set):\n",
    "    tweets = set['OriginalTweet'].to_list()\n",
    "    tokenized = tokenizer(tweets, return_tensors='pt', padding='max_length', truncation=True, max_length=128)\n",
    "\n",
    "    return {\n",
    "        'input_ids': tokenized['input_ids'],\n",
    "        'labels': ohe.fit_transform(set['Sentiment'].to_numpy().reshape(-1, 1))\n",
    "    }\n",
    "\n",
    "train_loader = DataLoader(TwitterDataset(dset(train_dataset)), batch_size=128)\n",
    "val_loader = DataLoader(TwitterDataset(dset(val_dataset)), batch_size=256)\n",
    "test_loader = DataLoader(TwitterDataset(dset(test_dataset)), batch_size=256)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## Тренировка модели"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import numpy as np\n",
    "from tqdm.notebook import tqdm_notebook as tqdm\n",
    "import matplotlib.pyplot as plt\n",
    "from torch.optim import Adam, AdamW, lr_scheduler\n",
    "from sklearn.metrics import classification_report, f1_score\n",
    "from IPython.display import display, clear_output\n",
    "\n",
    "def train(model, train_loader, epochs, lr, val_loader = None):\n",
    "    optimizer = Adam(model.parameters(), lr=lr)\n",
    "    criterion = nn.CrossEntropyLoss()\n",
    "    scheduler = lr_scheduler.CosineAnnealingLR(\n",
    "        optimizer, 25, \n",
    "    )\n",
    "\n",
    "    # pbar, step = tqdm(range(len(train_loader) * epochs)), 0\n",
    "    \n",
    "    batches = len(train_loader)\n",
    "    total_steps = batches * epochs\n",
    "        \n",
    "    validation_f1, val_loss = evaluate(model, val_loader, return_f1=True, show_progress=False)\n",
    "    losses = []\n",
    "    valid_f1 = [validation_f1]\n",
    "    valid_loss = [val_loss]\n",
    "    \n",
    "    fig, axs = plt.subplots(1, 2, figsize=(15, 5), width_ratios=[3, 2])\n",
    "    axs[1].plot([0], valid_f1, label = 'Val F1', marker = 'o')\n",
    "    axs[1].plot([0], valid_loss, label = 'Val Loss', marker = 'o')\n",
    "    axs[1].grid()\n",
    "    axs[1].legend()\n",
    "    \n",
    "    for epoch in range(epochs):\n",
    "        model.train()\n",
    "        for batch in train_loader:\n",
    "            inputs = batch['input_ids'].to(device)\n",
    "            labels = batch['labels'].to(device)\n",
    "            preds = model(inputs)\n",
    "            loss = criterion(preds, labels)\n",
    "            \n",
    "            loss.backward()\n",
    "            optimizer.step()\n",
    "            optimizer.zero_grad()\n",
    "            \n",
    "            losses.append(loss.item())\n",
    "            # step += 1\n",
    "            \n",
    "            axs[0].clear()\n",
    "            axs[0].plot(losses, label = 'Loss')\n",
    "            axs[0].set_xlim(0, total_steps)\n",
    "            axs[0].grid()\n",
    "            axs[0].legend()\n",
    "            \n",
    "            display(fig)\n",
    "            clear_output(wait=True)\n",
    "            \n",
    "            # pbar.set_description(f'Epoch: {epoch+1}, training loss: {loss.item():0.5f}')#, lr: {scheduler.get_last_lr()[0]:0.5f}')\n",
    "            # pbar.update(1)\n",
    "            \n",
    "        scheduler.step()\n",
    "        \n",
    "        validation_f1, val_loss = evaluate(model, val_loader, return_f1=True, show_progress=False)\n",
    "        valid_f1.append(validation_f1)\n",
    "        valid_loss.append(val_loss)\n",
    "        \n",
    "        axs[1].clear()\n",
    "        axs[1].plot(valid_f1, label = 'Val F1', marker = 'o')\n",
    "        axs[1].plot(valid_loss, label = 'Val Loss', marker = 'o')\n",
    "        axs[1].grid()\n",
    "        axs[1].legend()\n",
    "        \n",
    "        display(fig)\n",
    "        clear_output(wait=True)\n",
    "\n",
    "    return losses, valid_f1, valid_loss\n",
    "\n",
    "def evaluate(model, loader, return_f1 = False, show_progress = True):\n",
    "    pred, true = [], []\n",
    "\n",
    "    model.eval()\n",
    "\n",
    "    criterion = nn.CrossEntropyLoss()\n",
    "    running_loss = 0\n",
    "    \n",
    "    if show_progress:\n",
    "        pbar = tqdm(range(len(loader)), total=len(loader))\n",
    "    \n",
    "    with torch.no_grad():\n",
    "        for batch in loader:\n",
    "            inputs = batch['input_ids'].to(device)\n",
    "            labels = batch['labels'].to(device)\n",
    "            preds = model(inputs)\n",
    "            running_loss += criterion(preds, labels).item()\n",
    "            pred.extend(torch.argmax(preds.cpu(), dim=-1))\n",
    "            true.extend(torch.argmax(labels.cpu(), dim=-1))\n",
    "            if show_progress:\n",
    "                pbar.update(1)\n",
    "    \n",
    "    pred = np.array(pred)\n",
    "    true = np.array(true)\n",
    "    \n",
    "    if return_f1:\n",
    "        return f1_score(true, pred, average='weighted'), running_loss / len(loader)\n",
    "\n",
    "    return true, pred, running_loss / len(loader)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+---------------------------------------+------------+\n",
      "|                Modules                | Parameters |\n",
      "+---------------------------------------+------------+\n",
      "|           encoder.te.weight           |  3906816   |\n",
      "|        encoder.layernorm.weight       |    128     |\n",
      "|         encoder.layernorm.bias        |    128     |\n",
      "| encoder.layers.0.attention.wqs.weight |    512     |\n",
      "|  encoder.layers.0.attention.wqs.bias  |     4      |\n",
      "| encoder.layers.0.attention.wks.weight |    512     |\n",
      "|  encoder.layers.0.attention.wks.bias  |     4      |\n",
      "| encoder.layers.0.attention.wvs.weight |   16384    |\n",
      "|  encoder.layers.0.attention.wvs.bias  |    128     |\n",
      "|  encoder.layers.0.attention.wo.weight |   16384    |\n",
      "|   encoder.layers.0.attention.wo.bias  |    128     |\n",
      "|   encoder.layers.0.layernorm.weight   |    128     |\n",
      "|    encoder.layers.0.layernorm.bias    |    128     |\n",
      "|    encoder.layers.0.linear1.weight    |   16384    |\n",
      "|     encoder.layers.0.linear1.bias     |    128     |\n",
      "|    encoder.layers.0.linear2.weight    |   16384    |\n",
      "|     encoder.layers.0.linear2.bias     |    128     |\n",
      "| encoder.layers.1.attention.wqs.weight |    512     |\n",
      "|  encoder.layers.1.attention.wqs.bias  |     4      |\n",
      "| encoder.layers.1.attention.wks.weight |    512     |\n",
      "|  encoder.layers.1.attention.wks.bias  |     4      |\n",
      "| encoder.layers.1.attention.wvs.weight |   16384    |\n",
      "|  encoder.layers.1.attention.wvs.bias  |    128     |\n",
      "|  encoder.layers.1.attention.wo.weight |   16384    |\n",
      "|   encoder.layers.1.attention.wo.bias  |    128     |\n",
      "|   encoder.layers.1.layernorm.weight   |    128     |\n",
      "|    encoder.layers.1.layernorm.bias    |    128     |\n",
      "|    encoder.layers.1.linear1.weight    |   16384    |\n",
      "|     encoder.layers.1.linear1.bias     |    128     |\n",
      "|    encoder.layers.1.linear2.weight    |   16384    |\n",
      "|     encoder.layers.1.linear2.bias     |    128     |\n",
      "|             linear.weight             |    640     |\n",
      "|              linear.bias              |     5      |\n",
      "+---------------------------------------+------------+\n",
      "Total Trainable Params: 4042389\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "4042389"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classifier = EncoderForSequenceClassification(\n",
    "    num_classes = 5, \n",
    "    num_layers = 2, \n",
    "    hidden_dim = 128, \n",
    "    context=128, \n",
    "    vocab_size=tokenizer.vocab_size\n",
    ").to(device)\n",
    "count_parameters(classifier)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "baa351e1162643ffb4269ed457e183af",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/15 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.16      1.00      0.27       592\n",
      "           1       0.00      0.00      0.00      1041\n",
      "           2       0.00      0.00      0.00       619\n",
      "           3       0.00      0.00      0.00       947\n",
      "           4       0.00      0.00      0.00       599\n",
      "\n",
      "    accuracy                           0.16      3798\n",
      "   macro avg       0.03      0.20      0.05      3798\n",
      "weighted avg       0.02      0.16      0.04      3798\n",
      "\n"
     ]
    }
   ],
   "source": [
    "true, pred, _ = evaluate(classifier, test_loader)\n",
    "print(classification_report(true, pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "66acff52409d4bec98a38bfa40772c80",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/15 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.042039150206138944\n"
     ]
    }
   ],
   "source": [
    "print(evaluate(classifier, test_loader, return_f1=True)[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "losses, valid_f1, valid_loss = train(classifier, train_loader, epochs=15, lr=1e-3, val_loader=val_loader)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f7fa0fbf2d46424cb64811fd1362bf79",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/15 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.80      0.63      0.70       592\n",
      "           1       0.64      0.69      0.67      1041\n",
      "           2       0.85      0.77      0.81       619\n",
      "           3       0.62      0.70      0.66       947\n",
      "           4       0.74      0.70      0.72       599\n",
      "\n",
      "    accuracy                           0.70      3798\n",
      "   macro avg       0.73      0.70      0.71      3798\n",
      "weighted avg       0.71      0.70      0.70      3798\n",
      "\n"
     ]
    }
   ],
   "source": [
    "true, pred, _ = evaluate(classifier, test_loader)\n",
    "print(classification_report(true, pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "487d63f059664d06a93126f0eab3d688",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/15 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.703129343691713\n"
     ]
    }
   ],
   "source": [
    "print(evaluate(classifier, test_loader, return_f1=True)[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAApIAAAIsCAYAAABBSEljAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAACm6ElEQVR4nOzdd1hT1/8H8HdYYe8pMgUHKiBYEay4ULR1z1ZcVG1rta7aVttfq7Uq2mGtXbZ2qK1+tXXvAYot7oGK4gAXInuFEWaS3x/UaCRYiMEQeb+e5z6P3Jx77rnHk/Dhc869EchkMhmIiIiIiOpJR9MNICIiIiLtxECSiIiIiFTCQJKIiIiIVMJAkoiIiIhUwkCSiIiIiFTCQJKIiIiIVMJAkoiIiIhUwkCSiIiIiFTCQJKIiIiIVKKn6QYQERERPS/aB7yjtroSzn+ptroaCgNJei60GPapppuglW5u+Qgbzn+t6WZopdEBM9Bz5QpNN0NrHZ4+E+vPceypIiJwBn4+tVLTzdBak4Kma7oJzxVObRMRERGRSpiRJCIiIlIXgaYb8GwxkCQiIiJSF0HTiiQZSBIRERGpS9OKI7lGkoiIiIhUw4wkERERkbo0sYwkA0kiIiIitWlakSSntomIiIhIJcxIEhEREamJrGklJBlIEhEREalNEwskObVNRERERCphRpKIiIhIXZrYA8mZkSQiIiIilTCQJCIiIiKVcGqbiIiISF2a1sw2A0kiIiIitWliayQZSBIRERGpS9OKI7lGkoiIiIhUw4wkERERkZrINN2AZ4yBJBEREZG6NLE1kpzaJiIiIiKVMCNJREREpC5NKyHJQJKIiIhIfZpWJMmpbSIiIiJSCTOSREREROrStBKSDCSJiIiI1IaBJBERERGpQsbH/xARERER/TcGks9AbGwsBAIBCgoKNN2UBufu7o4VK1ZouhlERERNznfffQd3d3cYGhoiKCgIp0+frrVs9+7dIRAIamwvv/xyvc6p0antCRMmYO3atTX2h4eHY//+/XWqo3v37vD393/ug5cHfRUVFYW5c+fK92/fvh1DhgyBTPZsv5RpzZo1mDlzZo3g+MyZMzAxMXmmbXmWxvTtiMmDgmFnaYqrdzLxyS/7cSk5rdbyZsZCvDO6B8I7t4aFqRHSskVY9NtBxJ5PBgCMDg9ERHggnO0sAQBJ97Lx7V9/42j8zWdxOWp1+mACju+6gGKRGI6uNug3oSucvRxqLX/lZDKO/HUaBdlFsHG0QNirwfDu4CZ/XSaTIXbzGZw/nIiyknK4tHLCy6+FwsbJUl7mf5/vRcbdHJQUlsLIRAjPds0R9mowzKxrjsG8DBF+nPcnBDoCzP1lklqv/VkY5OuLUQEdYW1sjJs5Ofjm6BFcy8xUWrZrixYY3bETnC0toaujg/sFBfgr/hwOXbsmLzM+qDN6eLeEnZkZqiQS3MjKwi8njuNaZsazuiS1OHMwAcd3V487B1cb9Bv/5HGX+GDc5VSPu16vKB938Uf+HXctnfDSY+MOAG7E38HfW88iKyUXevp6cGvTDKPe6Sd//f7NTMRsPIn029kQQIBmLewRNjoYjm62au8DdTofnYAze+NRIhLD3sUGvcaGwqlF7f15/XQy4racgiinCFYOFug2Khiefu7y149tPY1rp5JQlFsMHT1dOLjboeuIIDRr4QgAEGUX4sSOs0hJTEWJSAwTKxP4hLRE8MCO0NXTbejLbRgamtretGkTZs+ejVWrViEoKAgrVqxAeHg4rl+/Dnt7+xrlt27dioqKCvnPubm58PPzw4gRI+p1Xo1nJPv27Yv09HSF7X//+59azyGTyVBVVaXWOjXB0NAQy5YtQ35+vqabUis7OzsYGxtruhkN4uUQH3wwoTdW/vk3Br67GtfuZmLNR6NhY678evX1dLBu/hg0t7fEtM83o/fb3+ODH3YjI7dIXiYjtxCf/3EYg9/7GYPf+xknL9/BqvdHwdvF7lldllpcPpGEg78fQ7dhHfHGkhFwcLPFH0t3o0QkVlr+3o10bPnmEDp0b4M3okagVUcPbPxyH7Lu5crLHNsVj1P7L+Hlid0w6dNhMBDq4Y+lu1FV8fC97N7WGSNm9MG0L0dj5Kxw5GWK8OeKmn+ESqok2PLNQbi2dlL/xT8D3b1bYkrXUKw7dRJvbNyAmznZWDZoCCyNjJSWLywrx/ozpzHtz42YvOEP7E+8gvfC+qCj68OA6V5+PlYePYJJ63/HjM1/IqOoEJ8NHgKLWupsjK6cSMLBP46h29COeH3xCDi62mL9f427b6vH3etLRqBVoAc2LVccd8d3xeP0gUt4+bVumPjpMOgb6mH9Y+Pu6umb2P59DPy7tcYbS0chcsEQtOviLX+9oqwSG5bthoWNGSYuHIYJC4ZAaKSP9Ut3QVIlabgOeUrXTiYhdkMcQga/gHELR8LO1RZ/fb4LJYXK+/N+Ujp2fX8Q7UPbYPzCkfAO8MS2FfuQnfqwP60cLdFrbCgmLHkFo/9vCCzszPDXZ7sgLiwFAOSl50Mmk6F3ZHdERr2KnqNfxMXDV/D3XyefxSU/V5YvX47JkycjMjISPj4+WLVqFYyNjfHrr78qLW9tbQ1HR0f5dujQIRgbG2tfICkUChUuxNHREVZWVgCqp4QNDAzwzz//yMt/9tlnsLe3R2ZmJiZMmICjR4/i66+/lqdk79y5I59K3rdvHwIDAyEUChEXFwepVIqoqCh4eHjAyMgIfn5+2Lx5s7zuB8cdOHAAHTp0gJGREXr27ImsrCzs27cPbdq0gbm5OUaPHg2x+OEb67/qfVRJSQnMzc1rvL59+3aYmJigqKhI6XEAEBYWBkdHR0RFRT2xT+Pi4tC1a1cYGRnBxcUF06dPR0lJifz19PR0vPzyyzAyMoKHhwc2bNhQY0p6+fLlaN++PUxMTODi4oK33noLxcXF8n6KjIyESCSS9/uCBQsAKE5tjx49GqNGjVJoW2VlJWxtbbFu3bp6952mvTagMzZFx2PLkYtITs3B//24B6XllRjey19p+eE9/WFhaog3l/2Jc9dTcT9bhNOJKbh292EW6fDZJMSeT8ad9DzcSc/DlxuOQFxWAf+Wzs/oqtTj5J6LCOjpgw7d28CuuTX6T+wGfQM9xMdeU1r+1L5L8PJzRZcBHWDnbI2eI4Pg5GGH0wcSAFT/8Xdq3yWEDglE644ecHCzxeC3eqEovwTXzt6W1xP8kh+aezvC0s4MLi2d8OLAAKQmZ9b4ZX34z9OwbWaFtp29Gq4TGtCIDgHYe/ky9l9NxN28PHx1OAblVVXo59NWafmL91MRd+smUvLzkSYSYevFC7iVk4P2zZrJyxy+cR3n791DemEh7uTl4Yd//oapUAhPm8adMXvUib0XEdDDB/7/jruXJ3aDvlAP8UdrGXf7q8ddyL/jrse/4+7MwUfG3f5L6Do4EK06esDB1RaDp/RCUcHDcSeVSLF/XRx6jw5Gx7B2sHGyhF1za4WxlZOWj9LicnQf8QJsm1nBvrk1Qoe+gBJRKUQ5xQ3fMSo6u/8CfLu3RfvQNrB1tkafCd2hL9TD5aNXlZY/d+ASPNq7otPLAbBxtsaLw4Pg4G6H+EMJ8jI+IS3h3s4FlvYWsG1ugx6jX0RFaQWy7+UAADx83dBvci94tHeFpb0FvAI88EI/fySdvfVMrrlBCNS3lZeXo7CwUGErLy+vccqKigqcO3cOYWFh8n06OjoICwvDiRMn6tTsX375Ba+88kq9ZxU1Hkg+Sffu3TFz5kyMHTsWIpEI8fHx+Oijj/Dzzz/DwcEBX3/9NYKDgzF58mR5NtPFxUV+/Ny5c7F06VJcvXoVvr6+iIqKwrp167Bq1SpcuXIFs2bNwpgxY3D06FGF8y5YsADffvstjh8/jnv37mHkyJFYsWIFNmzYgD179uDgwYP45ptv5OXrWi8AmJiY4JVXXsFvv/2msP+3337D8OHDYWZmVmt/6OrqYsmSJfjmm2+QmpqqtMzNmzfRt29fDBs2DJcuXcKmTZsQFxeHadOmycuMGzcOaWlpiI2NxZYtW/DTTz8hKytLoR4dHR2sXLkSV65cwdq1a3H48GG89957AICQkBCsWLEC5ubm8n6fM2dOjbZERERg165d8gAUAA4cOACxWIwhQ4bUu+80SV9PB+1aOOH4pYdBjEwGHL90Gx1aNld6TNgLLRF//T4+mdwPp36ZhX1fvYEpQ7tAR0f5tIeOjgD9u7SFkaE+4q8r//9tjCRVEqTdzoZnu4f9INARwLNdc6QmKZ8mvZeUqVAeAFr4uiA1qTrILsgqRHGBGJ7tHr6fDY2FaN7CAfdqqbO0uAwJx27ApaWjwpTY7cupSDx1Ey9Fhqp8jZqkp6ODlvb2OHfvnnyfDMC5eynwcapbhrVDcxc0t7LCpfv3az1H/7btUFxejps52epodoOTVEmQfjsbHo+NO48njLvUpEyF8kDdxp1zCwd5nem3s1GUVwKBQICf5v2J5W+twfpluxWymjZOljAyNUT8kauQVElQWVGFC7FXYetsBUu72j/jNUlSJUHGnWy4tVXsTzef5khLVt6fackZcGvrorDPvb1LreUlVRJcPHIFQmMD2LnW/gdLeWkFDE2EKlxF4yBT4xYVFQULCwuFTVkyKScnBxKJBA4OissQHBwckJHx38tVTp8+jcuXL2PSpPov+9H44392794NU1NThX0ffPABPvjgAwDAokWLcOjQIbz++uu4fPkyxo8fj4EDBwIALCwsYGBgAGNjYzg6Otaoe+HChejduzeA6qh+yZIliI6ORnBwMADA09MTcXFx+PHHH9GtWzf5cYsWLUKXLl0AABMnTsS8efNw8+ZNeHp6AgCGDx+OI0eO4P33369XvQ9MmjQJISEhSE9Ph5OTE7KysrB3715ER0f/Z38NGTIE/v7+mD9/Pn755Zcar0dFRSEiIgIzZ84EAHh7e2PlypXo1q0bfvjhB9y5cwfR0dE4c+YMOnbsCAD4+eef4e3trVDPg+OB6izjokWL8Oabb+L777+HgYEBLCwsIBAIlPb7A+Hh4TAxMcG2bdswduxYAMCGDRswcOBAmJmZqdR3mmJlZgw9XR3kFChmE3JEJfB0Vv6B6OJgheB2ltjxTwImLv4f3Byt8cnr/aCnq4tv/vpbXq6lqz02L4mE0EAP4rIKvPXZX0hOzWnQ61EncWEZZFIZTCwUp/hNLIyQk6Z8GUZxgbhGeVMLYxQXVGf6i/+dmjSxUJxmNbEwQkmB4jTboQ0ncOZgAirLq9Dc2wGvvvtwobi4qAzbVx3G0KlhEBobqHaBGmZhZARdHR3kixWvO18shquVda3HmRgY4M/XJkFfVxdSmQwrYg/j3L0UhTKd3T3wUd9+EOrrI6+kBO9u24rCsrIGuQ51ExepNu5Ma5T/73FnamEkfy0/qxAAcHTrGfQZ0wUWtmY4ufci1n66A9OWj4aRqSGERgYY/9EgbFq+D/9sOwcAsHa0QMTcAdDRbZz5m9J/+9P4saU6xhbGyEtX3p8lInHN96i5cY2lBTfj72DX9wdQWVEFU0sTjHhvIIzNlC+hyM8swPlDCej+SshTXM3zY968eZg9e7bCPqFQ/UH2L7/8gvbt26NTp071PlbjgWSPHj3www8/KOyztn744WhgYID169fD19cXbm5u+Oqrr+pc94NACQCSk5MhFovlgeUDFRUV6NChg8I+X19f+b8dHBxgbGwsDyIf7HtwJ1R96n2gU6dOaNu2LdauXYu5c+fijz/+gJubG0JD65YxWbZsGXr27Kk0C3jx4kVcunQJ69evl++TyWSQSqW4ffs2bty4AT09PQQEBMhf9/Lyki8neCA6OhpRUVG4du0aCgsLUVVVhbKyMojF4jqvgdTT08PIkSOxfv16jB07FiUlJdixYwc2btwIQLW+Ky8vr5HWb4g3lTroCATIFZXgw1V7IJXKcPlWBhxszDB5ULBCIHk7LQcD5vwEM2Mh+gb74LNpAzH643VaFUxqUpf+/gjo0QYF2UU4uvUMtn8fjVffexkCgQC7Vh9B+y7ecGvT7L8res6IKyow+X/rYaRvgAAXF7zVtRvSRYW4eP9htvtC6j1M/t96WBgZ4eW27fBxv5cw9c+NKCgt1WDLG7cHNza+OCgQbTq1AAAMfKMnVkxbi8RTNxHYqy0qK6qw86cjcGnphKHTekMmleHEngv43+d7MGnRcOgbaPxX7zPl4uOM8YtGobSoDJdiE7Hr2wOIWDAcJo8FrUV5xdj8+S606tQCfj2UL9vQCmq82UYoFNbpd5ytrS10dXWR+dgNeJmZmU9M+ADVS+42btyIhQsXqtRGjY9mExMTeHk9ed3S8ePHAQB5eXnIy8ur8/z9o+UeTK/u2bMHzs6K688e/0/S19eX/1sgECj8/GCfVCqtd72PmjRpEr777jvMnTsXv/32GyIjIyGo4+ALDQ1FeHg45s2bhwkTJii8VlxcjDfeeAPTp0+vcZyrqytu3Ljxn/XfuXMH/fv3x5QpU7B48WJYW1sjLi4OEydOREVFRb1upomIiEC3bt2QlZWFQ4cOwcjICH379pW3Fahf30VFReGTTz5R2Dd//nwADXt3X36RGFUSKWwtFbPnthYmyC5QvuYpK78YVRIJpNKHd9TfTM2BvZUZ9PV0UFlVPYYqq6S4m1H9F//lWxnw9XLChJc74f9+3NtAV6NexuaGEOgIamQhSkSlMLVUPlZMLWtmLYpFYnn5B1mjElEpzKwevo9LRKVwcLd57PxGMDY3ql6r5myFr6atQ2pSJlxaOuL2lfu4fu4Oju++UF1YVh0ILIz4AQMmdUeHHm2e5tKfCVFpKSRSKawee99ZGRsjT1xSy1HV02JpIhEA4GZONlytrTG64wsKgWRZVRXSRCKkiUS4mpGBdePGo1/bdvjf2TMNci3qZGym2rgrrlH+v8ddsagUjm42/9ZRvd/O+WHCQ09fF5b25hDlVK9xv3zsBkTZRZj4yTAI/l3KMnRab3w2+RdcP3sb7UIUZ4AaA6N/+1P82I01YlHN2YMHTCyMUSJS/KOjpLBmeQOhPgwcLGHlADTzcsTqd/9AwtGr6DwgUF6mOL8Em6K2o5m3E8Ije6jpqjREAzdtGxgYIDAwEDExMRg8eDCA6nsQYmJiFJa2KfPXX3+hvLwcY8aMUencjTPH/oibN29i1qxZWL16NYKCgjB+/Hh5EAdUd55E8t93wfn4+EAoFCIlJQVeXl4K26PrKutL1XrHjBmDu3fvYuXKlUhMTMT48ePrdd6lS5di165dNRbRBgQEIDExsUZbvLy8YGBggFatWqGqqgrx8fHyY5KTkxXuBD937hykUim+/PJLdO7cGS1btkRamuIjbura7yEhIXBxccGmTZuwfv16jBgxQh6Yq9J38+bNg0gkUtjmzZtX535TVWWVFJdvpiOkvbt8n0AABPt6IP6G8vWM567dg5ujtcIfpx7NbJCZVyQPIpXREQhgoK/xv/HqTFdPF8087HDr8sP1dzKpDLeupKK5t/K/hF28HXD7imK/3Uq4h+be1et7LO3NYWppjFuXH5YpF1cg9WYmXGqpE3iYLXpws83EhUPx5tKR8q37iBdgYKSPN5eOROsXPFS74GesSirFjawsBDzynhAACHBxQWJ6ep3r0YEA+rpP/oNLRyCAwX+UaSx09XTh5GGH21cUx93tJ4y75t4OuH35v8fdo2OzXFyB+zcz5XU287CDrr4uch+Z7pVUSSDKLoKFbfX6x8qKquoA8pH3/oNEwbN+VFtd6erpwtHdDncfuXaZVIa7ialo5qW8P5t5OSIlUbE/716uvby8XpkMksqHvz+K8oqxcck2OHjYo9/knvLgm+pn9uzZWL16NdauXYurV69iypQpKCkpQWRkJIDq+yOU/b785ZdfMHjwYNjY2NR4rS40/tuqvLy8xkJQPT092NraQiKRYMyYMQgPD0dkZCT69u2L9u3b48svv8S7774LoHr93qlTp3Dnzh2YmpoqTIs/yszMDHPmzMGsWbMglUrx4osvQiQS4dixYzA3N693IPe09VpZWWHo0KF499130adPHzRvrvyGjdq0b98eERERWLlypcL+999/H507d8a0adMwadIkmJiYIDExEYcOHcK3336L1q1bIywsDK+//jp++OEH6Ovr45133oGRkZH8g87LywuVlZX45ptvMGDAABw7dgyrVq1SOI+7uzuKi4sRExMDPz8/GBsb15qpHD16NFatWoUbN27gyJEjT9V3dU3zN4Rfd53E528PQsLNdFxMSkNk/04wFupj8+GLAIAv3h6EjLwifLH+MABgw4FzGNvvBXz8WjjW7j0DdydrTBnaBWv3Psz2zInoiaPxyUjLFsHESIiBXdshqK07Jny6XmkbGqvOL/th+w+H0czTDs5e9ji57xIqy6vg3601AGDb99EwszJB2KvVa2GD+vlizcIdOL77Alp2cMPlE0lIu5WNAZO7A6j+pRvUzxf/bD8HG0cLWNqb48hfp2FmZYLWHasDwNTkTKTdzIJrKycYmgiRnynCkb9Ow8rBXP5L/9GsEQCk3cqGQCCAvYtqH5ia8lf8eczt3QfXMzNxLTMDw/wDYKinj/2JiQCAub37IKekBD8fPwYAeLXjC7iRmYk0UQH0dXUR5O6B3q1bY0Vs9dg01NNDxAudcPz2LeSVlMDc0AiDff1ga2KKo0n/PWvRWAS/5Iftq6rHXbMW9ji17xIqyx6Ou+3fR8PM2gS9Xvl33PX1xdpPd+DEngvw9n847vpP6g7g33HX1xf/bDsHa0cLWNqZI/av0zCzfDjuhMYG6NirLWK3nIG5jSksbM1w4t+Mt09Q9VS3ZzsXHNpwAvt++xsvhLeHTAYc23keOro6cPdpvE9k6NjXH3tXx8DRwx5OnvY4e/AiKsur0C60OnO/58fq93HoyOr+DAz3xcYl23FmXzw8/dxx7WQSMm5noc9r3QEAFeWVOLnzLLw6eMDE0hilRWWIj05AcX4JWv27LKAorxgbo7bD3MYM3V8JkT8WCHiY/dU6GnqO5KhRo5CdnY2PP/4YGRkZ8Pf3x/79++U34KSkpEBHRzF/eP36dcTFxeHgwYMqn1fjgeT+/fvh9Nidh61atcK1a9ewePFi3L17F7t37wYAODk54aeffsKrr76KPn36wM/PD3PmzMH48ePh4+OD0tJS3L59W9lpAACffvop7OzsEBUVhVu3bsHS0hIBAQHyG3tUpWq9EydOxIYNG/Daa6+pdN6FCxdi06ZNCvt8fX1x9OhRfPjhh+jatStkMhlatGih8BiedevWYeLEiQgNDZU/TujKlSswNDQEAPj5+WH58uVYtmwZ5s2bh9DQUERFRWHcuHHyOkJCQvDmm29i1KhRyM3Nxfz58+WPAHpcREQEFi9eDDc3N/lNTA801P9JQ9hzPBHWFsaY+Uo32Fqa4urtTEQu2oBcUfX0opOtOaSPZBvScwsR+el6fBjZB3uXv4GMvEKs2XMaP24/Li9jY2GML94eBDsrUxSLy3HtbiYmfLoexy7VPo4bo3bB3hAXliF282kUF4jh6GaLiLn95VOGopxihaUb1WvHwnDkz9M4vOkkrB0t8co7/RQCvC4DOqCyvAq7fo5FmbgCrq2cMGZuf+j9u75M30APV0/fQuzm06gor4KZpTFa+Lli+JA+0NPXjqxaXcUm3YClkREiOwfDysQYN7Nz8P6O7cgvrZ6GtDczxyMrKGCkp4cZPXrAztQM5VVVuJefhyUHDyD23yBRIpPB1coa4W18YG5kiMLSMlzPysSMzX/hTl6eJi5RJW2DvVHyyLhzcLPF6Ln95VPUotxiheyWS0snDJ0ahiN/PRx3o2YrjruQAR1QUV6F3Q/GXUsnRDwy7gAgbHQwBLoCbP8+BpWVVXBu4YCx/zcIRqbVn6G2zlZ4Zc5L+HvLGfw6f2v1jYnutoh4v7/ClHlj07qzN8RFpTi29VT1A8ldbTH83f7yqeqi3CKF97GztxP6T+mNfzafwj9/nYSVgyWGzOwHu+bV/akjECAvrQA74vajtKgUhqaGcPKwx6sfDoHtv2XuXrmHgkwRCjJFWDVT8QtK3l039RlduXppMuc8bdq0WqeyY2Nja+xr1arVU2fJBbLGmmdvAn7//XfMmjULaWlpMDDQ3B2lqampcHFxQXR0NHr16qWxdjyNFsM+1XQTtNLNLR9hw/mvNd0MrTQ6YAZ6rlyh6WZorcPTZ2L9OY49VUQEzsDPp1b+d0FSalJQzXsI1Kl1+Edqq+vagcb/u03jGcmmSCwWIz09HUuXLsUbb7zxzIPIw4cPo7i4GO3bt0d6ejree+89uLu71/mucSIiIiJAC262eR599tlnaN26NRwdHZ/JjSKPq6ysxAcffIC2bdtiyJAhsLOzQ2xsbI2704mIiKieBAL1bVqAGUkNWLBgQa3rCZ+F8PBwhIeHa+z8RERE9HxgRpKIiIiIVMKMJBEREZGayLRkSlpdGEgSERERqUvTiiM5tU1EREREqmEgSUREREQq4dQ2ERERkZo0tTWSzEgSERERkUqYkSQiIiJSl6aVkGQgSURERKQ2DCSJiIiISDVNK5LkGkkiIiIiUgkzkkRERERqImtaCUkGkkRERERq08QCSU5tExEREZFKmJEkIiIiUpumlZJkIElERESkJk1tjSSntomIiIhIJcxIEhEREalLE8tIMpAkIiIiUpumFUlyapuIiIiIVMKMJBEREZGaNLWbbRhIEhEREakLA0kiIiIiUk3TiiS5RpKIiIiIVMKMJBEREZG6NK2EJANJIiIiInVpajfbcGqbiIiIiFTCjCQRERGRujSxjCQDSSIiIiK1aVqRpEAmk8k03QgiIiKi54HnyMVqq+vWnx+qra6GwowkPRc2nP9a003QSqMDZsCnZ+P/oGqMEg8vxk8nV2q6GVrr9c7T+b5V0eiAGRi+jn2nqs3jZjRo/U3tZhsGkkRERETq0sQCSd61TUREREQqYSBJRERE9Bz47rvv4O7uDkNDQwQFBeH06dNPLF9QUICpU6fCyckJQqEQLVu2xN69e+t1Tk5tExEREamLQDNz25s2bcLs2bOxatUqBAUFYcWKFQgPD8f169dhb29fo3xFRQV69+4Ne3t7bN68Gc7Ozrh79y4sLS3rdV4GkkRERERqoqmbbZYvX47JkycjMjISALBq1Srs2bMHv/76K+bOnVuj/K+//oq8vDwcP34c+vr6AAB3d/d6n5dT20RERESNUHl5OQoLCxW28vLyGuUqKipw7tw5hIWFyffp6OggLCwMJ06cUFr3zp07ERwcjKlTp8LBwQHt2rXDkiVLIJFI6tVGBpJEREREjVBUVBQsLCwUtqioqBrlcnJyIJFI4ODgoLDfwcEBGRkZSuu+desWNm/eDIlEgr179+Kjjz7Cl19+iUWLFtWrjZzaJiIiIlIXNU5tz5s3D7Nnz1bYJxQK1VK3VCqFvb09fvrpJ+jq6iIwMBD379/H559/jvnz59e5HgaSRERERI2QUCisU+Boa2sLXV1dZGZmKuzPzMyEo6Oj0mOcnJygr68PXV1d+b42bdogIyMDFRUVMDAwqFMbObVNREREpC4Cgfq2OjIwMEBgYCBiYmLk+6RSKWJiYhAcHKz0mC5duiA5ORlSqVS+78aNG3BycqpzEAkwkCQiIiJSG5kat/qYPXs2Vq9ejbVr1+Lq1auYMmUKSkpK5Hdxjxs3DvPmzZOXnzJlCvLy8jBjxgzcuHEDe/bswZIlSzB16tR6nZdT20RERERabtSoUcjOzsbHH3+MjIwM+Pv7Y//+/fIbcFJSUqCj8zB/6OLiggMHDmDWrFnw9fWFs7MzZsyYgffff79e52UgSURERKQuGvyu7WnTpmHatGlKX4uNja2xLzg4GCdPnnyqczKQJCIiIlIXDQaSmsA1kkRERESkEgaSRERERKQSTm0TERERqUs9HtvzPGAgSURERKQmsqYVR3Jqm4iIiIhUw0CSiIiIiFTCqW0iIiIideHUNhERERHRf2NGkoiIiEhdmlhGkoEkERERkdo0rUiSU9tEREREpBJmJImIiIjUpWklJBlIEhEREalNEwskObVNRERERCphIElq4+7ujhUrVmi6GURERBojU+OmDTi1rSUmTJiAtWvXIioqCnPnzpXv3759O4YMGQKZ7NkNuTVr1mDmzJkoKChQ2H/mzBmYmJg8s3Y8C6cPJuD4rgsoFonh6GqDfhO6wtnLodbyV04m48hfp1GQXQQbRwuEvRoM7w5u8tdlMhliN5/B+cOJKCsph0srJ7z8WihsnCzlZVa8/TtEOUUK9fZ6pTNeHBSg9uvTtFcHBeG1UV1ha22K6zczsPib3Ui4lqq07JrlE9HJ37PG/qMnr2PKB+sauqnPVHx0As7ui0eJSAw7Fxv0HBMKpxa1j7vrp5NxbOspFOYUwcrBAl1HBsPTz13++vFtp3HtVBKKcouhq6cLB3c7vDg8CE4tHAEA967ex59LtyutO2L+cDh61n7uxojvW/Xp28oXA9sGwtLIGHfzcvDL6Vgk52YqLRvm3RbdPNvAxdIGAHArLwsbzh9XKD81pDd6ePkoHBd//w4Wx+xouIt41prY1DYDSS1iaGiIZcuW4Y033oCVlZWmm1ODnZ2dppugVpdPJOHg78fw8sRuaO7lgJP7LuGPpbsx7ctXYWJhXKP8vRvp2PLNIfR6pTNaBrgh4VgSNn65D29EjYC9S/UH67Fd8Ti1/xIGT+kFKzszHPnrNP5YuhtTP38FegYP347dR3RCYM+HH7YGhvoNf8HPWN/u7fH+lJfwyYoduHT1HsYO64Kflk3Ay+O/Ql5BSY3yM+ZvgL6ervxnSwtjbF09DQeOJjzLZje4a6eScPR/cQgb3x1OLRxw7sBFbPliF15bNhrG5jXH3f2kdOz54SC6jugMT393XDuRhB1f78PYhSNh27x63Fk5WqLX2FBY2JmjqqIK5w5cxObPd2HiZ2NgbG6EZt6OePPrCQr1Htt6GimJqXDwsH8Wl602fN+qT4i7N8Z37IqfTh5BUk4GXm7jj/8LG4zpO9ahsKy0Rvm2Ds0Rd+cGrmeloUIiweB2HfFR7yGYteN35JU+fE/H37+D744dkv9cKZU8k+uhhsGpbS0SFhYGR0dHREVF1VomLi4OXbt2hZGREVxcXDB9+nSUlDx8A6enp+Pll1+GkZERPDw8sGHDhhpT0suXL0f79u1hYmICFxcXvPXWWyguLgYAxMbGIjIyEiKRCAKBAAKBAAsWLACgOLU9evRojBo1SqFtlZWVsLW1xbp11dkjqVSKqKgoeHh4wMjICH5+fti8ebMaeko9Tu65iICePujQvQ3smluj/8Ru0DfQQ3zsNaXlT+27BC8/V3QZ0AF2ztboOTIITh52OH2gOtCRyWQ4te8SQocEonVHDzi42WLwW71QlF+Ca2dvK9QlNNSHqaWxfNP2X0jKTBjRBX/tPYtt+8/j5t1sfPLVDpSVV2Jov0Cl5UVFpcjJL5ZvwYFeKCurxIGjl59xyxvWuf0X0L5bW7QLbQMbZ2v0ntAd+gZ6SPj7qtLy5w9egkd7V7zwUgBsmlmjy7AgOLjbIT76YYDdJrgl3Nq6wNLeArbNbdB99IuoKK1A9r0cAICuni5MLE3km6GpIZLP30bbrq0hEGhXeoXvW/UZ0CYA0UlXcORmIlJFefjp5GGUS6rQ06ut0vJfxx3AgeuXcCc/B2mF+Vh1IhoCAO2dXBTKVUokKCgTy7eSivJncDXUUBhIahFdXV0sWbIE33zzDVJTa07/3bx5E3379sWwYcNw6dIlbNq0CXFxcZg2bZq8zLhx45CWlobY2Fhs2bIFP/30E7KyshTq0dHRwcqVK3HlyhWsXbsWhw8fxnvvvQcACAkJwYoVK2Bubo709HSkp6djzpw5NdoSERGBXbt2yQNQADhw4ADEYjGGDBkCAIiKisK6deuwatUqXLlyBbNmzcKYMWNw9OhRtfTX05BUSZB2Oxue7ZrL9wl0BPBs1xypSRlKj7mXlKlQHgBa+LogNal6WqcgqxDFBWJ4tnv4oWpoLETzFg6491idcTvP47PJv+DHuX/i2K54SCVSdV1ao6Cvpwufls1w8lyyfJ9MJsOJc8nw93GtUx3D+gVi75EElJZVNlQznzlJlQSZd7Lh2lZx3Lm2bY70ZOXjLj05A65tFX9Ru7VzqbW8pEqCS0euQGhsADtXW6VlbsbfQVlxGdp1baPilWgG37fqo6ejA08be1xKT5HvkwFISE9BKzvHOtVhoKsHXR1dFJcrBoptHZvjlxGT8fWgcZgc1AOmQkN1Nl3zBAL1bVqAU9taZsiQIfD398f8+fPxyy+/KLwWFRWFiIgIzJw5EwDg7e2NlStXolu3bvjhhx9w584dREdH48yZM+jYsSMA4Oeff4a3t7dCPQ+OB6qzjIsWLcKbb76J77//HgYGBrCwsIBAIICjY+0fJuHh4TAxMcG2bdswduxYAMCGDRswcOBAmJmZoby8HEuWLEF0dDSCg4MBAJ6enoiLi8OPP/6Ibt26PW1XPRVxYRlkUlmNqTATCyPkpOUrPaa4QFyjvKmFMYoLxNWvi8TyOh6vs+TfMgAQ1Lc9nDzsYGQixL0bGYjZdArFBWKEj+3y1NfVWFhaGENPVxc5+cUK+3Pzi+Hp+t9LJNq3bo6Wno746IttDdVEjSgtUj7ujC2MkZeufNyViMQwNn98TBmjRCRW2Hfzwh3s+f4AKiuqYGphguHvDoSxmeJxD1z+OxHu7V1gZm36FFfz7PF9qz5mQiPo6uhAVKo4jgpKxXA2t65THWMCX0R+abFCMHoh7S5OpSQjq7gQDmYWGN0hBB/2GoQP9/0J6TNc69+gtCP+UxsGklpo2bJl6NmzZ41M4MWLF3Hp0iWsX79evk8mk0EqleL27du4ceMG9PT0EBDwcPG3l5dXjfWW0dHRiIqKwrVr11BYWIiqqiqUlZVBLBbD2LjmGiNl9PT0MHLkSKxfvx5jx45FSUkJduzYgY0bNwIAkpOTIRaL0bt3b4XjKioq0KFDh1rrLS8vR/ljf90KhcI6tUlbBL/sL/+3g5stdPV0sfuXo+j1Smfo6evWfmATMqxfIK7fzKj1xhyqybWNM8Z+OgqlRWVIOJqIXd8dQMT84TXWXRblFeNOwj30nxquoZZqJ75vFQ1u1xFd3FtiwYEtCmsgj925If93SkEu7ubn4PuhkWjr0BwJGfc00VR6Spza1kKhoaEIDw/HvHnzFPYXFxfjjTfewIULF+TbxYsXkZSUhBYtWtSp7jt37qB///7w9fXFli1bcO7cOXz33XcAqoO8+oiIiEBMTAyysrKwfft2GBkZoW/fvvK2AsCePXsU2puYmPjEdZJRUVGwsLBQ2J60ZlRVxuaGEOgIamR1SkSlMLVUHkybWtbMAhWLxPLypv9mPUpEiovUS0SlMKmlTgBw9nKAVCJFQXZhva+jsSoQiVElkcDWSjHjZWNlipy84lqOqmZkqI9+PXyxZd/ZhmyiRhiZKR93YlHNrNkDJhbGEBc+PqZqltcX6sPKwRLNvBwRPrEndHR1kHC05rrLy/9chaGpIVp0cH+6i9EAvm/Vp6i8FBKpFBZGitdoaWSMgrKaN8M9aqBPAIa064hF0dtwtyDniWWzigshKhPD0cziqdtMmsFAUkstXboUu3btwokTJ+T7AgICkJiYCC8vrxqbgYEBWrVqhaqqKsTHx8uPSU5ORn7+wymfc+fOQSqV4ssvv0Tnzp3RsmVLpKWlKZzbwMAAEsl/32UXEhICFxcXbNq0CevXr8eIESOgr1+9+NzHxwdCoRApKSk12uri4lJrnfPmzYNIJFLYHg+o1UFXTxfNPOxw6/J9+T6ZVIZbV1LR3Fv5lL6LtwNuX1HMkN1KuIfm3tWPHbG0N4eppTFuXX5YplxcgdSbmXCppU4AyLibA4FAABNz5dOQ2qiySoLEG2noHPDwDxyBQIDOAS1wITHlCUcC4d3awcBAF7uiLzRwK5+9B4/mSUl8OEZkUhlSElPh5KV8jDh5OSqUB4C7V2ov/2i9kirF97FMJsOVf67Bp0sr6OppXxaN71v1qZJKcSs3S+FGGQGA9o4uuJ6tfL0pAAxqG4hhvp2wKHo7buZm1VruAWtjU5gJjZBf+uTgVKsI1LhpAU5ta6n27dsjIiICK1eulO97//330blzZ0ybNg2TJk2CiYkJEhMTcejQIXz77bdo3bo1wsLC8Prrr+OHH36Avr4+3nnnHRgZGcnvzPTy8kJlZSW++eYbDBgwAMeOHcOqVasUzu3u7o7i4mLExMTAz88PxsbGtU55jx49GqtWrcKNGzdw5MgR+X4zMzPMmTMHs2bNglQqxYsvvgiRSIRjx47B3Nwc48ePV1qfUCh8ZlPZnV/2w/YfDqOZpx2cvexxct8lVJZXwb9bawDAtu+jYWZlgrBXq9d4BvXzxZqFO3B89wW07OCGyyeSkHYrGwMmdwdQHSgF9fPFP9vPwcbRApb25jjy12mYWZmgdUcPAMC9Gxm4n5wJ97bOEBrq415SJg78fgy+L7aEkenztSB9zV/HEDV3GC5fv4+Ea6kYNywERoYG2Lb/HAAgau5wZOUU4qufDyocN6xfR8TEXYWosObjR54HgX39sX91DBw97OHoaY/zBy6isrxKfuPLvh+jYWplgq4jq8ddQB9f/Bm1HWf3xcPDzx3XTyUh83YW+kR2BwBUllfi5M6zaNHBA6aWxigtKkN8TAKKC0rQ8gXFmYqUxFSIsgvRvpvic/60Cd+36rPr6nlM69IHN3OykJybgZfbdIBQTx9HkhMBAG936YNccTE2xB8HAAxuG4hR/p2x4p8DyC4uhKVh9e+FsqpKlFVVwlBPHyP8gnDybjIKSkvgaGaJMYFdkFFUgAtpT/4DkhovBpJabOHChdi0aZP8Z19fXxw9ehQffvghunbtCplMhhYtWig8hmfdunWYOHEiQkND5Y8SunLlCgwNqz/s/Pz8sHz5cixbtgzz5s1DaGgooqKiMG7cOHkdISEhePPNNzFq1Cjk5uZi/vz58kcAPS4iIgKLFy+Gm5sbunRRXHT+6aefws7ODlFRUbh16xYsLS0REBCADz74QI29pLp2wd4QF5YhdvNpFBeI4ehmi4i5/eVTXqKcYoVHo7i0dMLQaWE48udpHN50EtaOlnjlnX7yZ9EBQJcBHVBZXoVdP8eiTFwB11ZOGDO3v/xZdHr6urh8IhmxW85AUimBpb05OvfzVVh/9bzYH5sAa0sTvB3ZC7ZWZrh2Mx1vvL8GufnVmQknewtIpYqL791dbBHo646J7/6qiSY/E62DvFFaWIpjW09BLBLDztUWw+b0l09VF+YVQaDzcNw5ezvhpTd749iWU4jbfBKWDpYYNKOf/BmSAoEAeekFSIzbj9LiUhiaGsLRwx6vfDBEXuaBy39fRTMvR9g0a3zPqa0rvm/V5/idJJgLjfCKf2dYGhnjTl4OFsdsh6iseimArYmZwg0yfVr5Ql9XD+92f1mhnj8vnsSfF09BKpPCzcoW3T3bwNhAiPzSElxMu4uNF06i6nl6lqSWZBLVRSB7ll+JQo1OamoqXFxcEB0djV69emm6OSrbcP5rTTdBK40OmAGfnh9quhlaKfHwYvx0cuV/FySlXu88ne9bFY0OmIHh69h3qto8bkaD1u8y5TO11XXvh/fUVldDYUayiTl8+DCKi4vRvn17pKen47333oO7uztCQ0M13TQiIiLSMgwkm5jKykp88MEHuHXrFszMzBASEoL169fLb4IhIiKip9DEprYZSDYx4eHhCA/n8+GIiIgagrZ9rejT4uN/iIiIiEglDCSJiIiISCWc2iYiIiJSl6Y1s81AkoiIiEhtmlggyaltIiIiIlIJA0kiIiIiUgmntomIiIjUpIk9/YcZSSIiIqLnwXfffQd3d3cYGhoiKCgIp0+frrXsmjVrIBAIFDZDQ8N6n5OBJBEREZGW27RpE2bPno358+fj/Pnz8PPzQ3h4OLKysmo9xtzcHOnp6fLt7t279T4vA0kiIiIidRGocauH5cuXY/LkyYiMjISPjw9WrVoFY2Nj/Prrr7U3VSCAo6OjfHNwcKjfScFAkoiIiKhRKi8vR2FhocJWXl5eo1xFRQXOnTuHsLAw+T4dHR2EhYXhxIkTtdZfXFwMNzc3uLi4YNCgQbhy5Uq928hAkoiIiEhd1JiRjIqKgoWFhcIWFRVV45Q5OTmQSCQ1MooODg7IyMhQ2sxWrVrh119/xY4dO/DHH39AKpUiJCQEqamp9bpc3rVNREREpCbqvGl73rx5mD17tsI+oVColrqDg4MRHBws/zkkJARt2rTBjz/+iE8//bTO9TCQJCIiIlIXNT7/RygU1ilwtLW1ha6uLjIzMxX2Z2ZmwtHRsU7n0tfXR4cOHZCcnFyvNnJqm4iIiEiLGRgYIDAwEDExMfJ9UqkUMTExClnHJ5FIJEhISICTk1O9zs2MJBEREZGaaOqB5LNnz8b48ePRsWNHdOrUCStWrEBJSQkiIyMBAOPGjYOzs7N8jeXChQvRuXNneHl5oaCgAJ9//jnu3r2LSZMm1eu8DCSJiIiItNyoUaOQnZ2Njz/+GBkZGfD398f+/fvlN+CkpKRAR+fhRHR+fj4mT56MjIwMWFlZITAwEMePH4ePj0+9zstAkoiIiOg5MG3aNEybNk3pa7GxsQo/f/XVV/jqq6+e+pwMJImIiIjUpYl91zYDSSIiIiI1aWJxJO/aJiIiIiLVMCNJREREpC5NLCXJQJKIiIhITTT1+B9N4dQ2EREREamEgSQRERERqYRT20RERERqwqltIiIiIqI6YCBJRERERCoRyGQymaYbQURERPQ88H7/C7XVlbRsjtrqaihcI0nPheV/r9R0E7TS7NDp+PU0+04Vr3WajjZh/6fpZmitq9GL8OMJjj1VvBE8HVN3fK3pZmit7wbNaNgTcI0kEREREdF/Y0aSiIiISE0ETSwlyUCSiIiISF2aVhzJqW0iIiIiUk2dMpI7d+6sc4UDBw5UuTFERERE2qyJJSTrFkgOHjy4TpUJBAJIJJKnaQ8RERGR1mpq32xTp0BSKpU2dDuIiIiISMs81RrJsrIydbWDiIiISPsJ1LhpgXoHkhKJBJ9++imcnZ1hamqKW7duAQA++ugj/PLLL2pvIBEREZG2aGJxZP0DycWLF2PNmjX47LPPYGBgIN/frl07/Pzzz2ptHBEREZFWaWKRZL0DyXXr1uGnn35CREQEdHV15fv9/Pxw7do1tTaOiIiIiBqvej+Q/P79+/Dy8qqxXyqVorKyUi2NIiIiItJGWpJIVJt6ZyR9fHzwzz//1Ni/efNmdOjQQS2NIiIiItJGAoH6Nm1Q74zkxx9/jPHjx+P+/fuQSqXYunUrrl+/jnXr1mH37t0N0UYiIiIiaoTqnZEcNGgQdu3ahejoaJiYmODjjz/G1atXsWvXLvTu3bsh2khERESkHZrYzTb1zkgCQNeuXXHo0CF1t4WIiIhIq2lJ/Kc2KgWSAHD27FlcvXoVQPW6ycDAQLU1ioiIiIgav3oHkqmpqXj11Vdx7NgxWFpaAgAKCgoQEhKCjRs3onnz5upuIxEREZFW0JabZNSl3mskJ02ahMrKSly9ehV5eXnIy8vD1atXIZVKMWnSpIZoIxERERE1QvXOSB49ehTHjx9Hq1at5PtatWqFb775Bl27dlVr44iIiIio8ap3IOni4qL0weMSiQTNmjVTS6OIiIiItBGntv/D559/jrfffhtnz56V7zt79ixmzJiBL774Qq2NIyIiItIqfPxPTVZWVhA8EmKXlJQgKCgIenrVh1dVVUFPTw+vvfYaBg8e3CANJSIiImrsBNoSAapJnQLJFStWNHAziIiIiEjb1CmQHD9+fEO3g4iIiEjrcY1kPZSVlaGwsFBhIyIiIqJn77vvvoO7uzsMDQ0RFBSE06dP1+m4jRs3QiAQqLQ8sd6BZElJCaZNmwZ7e3uYmJjAyspKYSMiIiKiZ2vTpk2YPXs25s+fj/Pnz8PPzw/h4eHIysp64nF37tzBnDlzVH6EY70Dyffeew+HDx/GDz/8AKFQiJ9//hmffPIJmjVrhnXr1qnUCCIiIqLngUCgvq0+li9fjsmTJyMyMhI+Pj5YtWoVjI2N8euvv9Z6jEQiQUREBD755BN4enqqdL31DiR37dqF77//HsOGDYOenh66du2K//u//8OSJUuwfv16lRpBRERE9DzQxNN/KioqcO7cOYSFhcn36ejoICwsDCdOnKj1uIULF8Le3h4TJ06sx9kU1TuQzMvLk0et5ubmyMvLAwC8+OKL+Pvvv1VuCFFsbCwEAgEKCgo03RQiIiKNKy8vr3EvSnl5eY1yOTk5kEgkcHBwUNjv4OCAjIwMpXXHxcXhl19+werVq5+qjfX+ZhtPT0/cvn0brq6uaN26Nf7880906tQJu3btgqWl5VM1htRjwoQJWLt2LaKiojB37lz5/u3bt2PIkCGQyWRqOc+dO3fg4eGB+Ph4+Pv7q6XOxubykQRcPBCPUpEYNi426PJqKOw9HGotf/NsMs7uOIWinCJYOFggaFgwXNu7y1//cfJ3So8LGh4M//AAFOUU4tzus0i7lgpxoRgmlibwCmqJgJc7QldPV92X16DOH0rAqb3xKBGJYe9ig7BxoWjWova+u3YqGf9sOQVRThGsHCzQfVQwWvi7AwAkVRL8s/kUbl68C1FWIYTGBnBr64Juo4JhZmWiUM/NC3dwbNsZZN/Lha6+HlxbN8PQWS815KU+E6MHBuG1kS/C1toU125mYPG3u5Fw/X6t5c1MDDHztTD0frEtLMyMkJZVgKjv9+Lv0zcAAMZGBpgxIQxhL/rA2tIEV5PTseT7Pbj8hDq1xYXoBJzdVz327Fxt0GNMKJw8ax97N04n49jWUyjMKYKlowW6jgiGp5+70rLRa2JxKfYKur/6IgLC/eT7T+08i1uX7iI7JQe6ujqY+sNkdV+WRoR6+CLMKxDmQmPcL8zBn5dicbcgU2lZJzNrvNw6GK6W9rAxNsfmhKM4cuuCQpmFvSNhY2xe49ijty/iz0uxDXAFGqDGu7ajoqLwySefKOybP38+FixY8FT1FhUVYezYsVi9ejVsbW2fqq56B5KRkZG4ePEiunXrhrlz52LAgAH49ttvUVlZieXLlz9VY0h9DA0NsWzZMrzxxhsavwmqoqICBgYGGm2DKpLPJOHEn3HoOqY7HDwccCn6Ivas2IVXPh0NI3PjGuUzktMRs/ogOg3tDDdfdySfSsKB7/Zh2EcjYe1sAwAY+8UEhWNSLqfg6NrD8AxoAQDIz8iHTCZD17HdYWFvgbz7efh73RFUVVQheESXBr9mdbl6MgmHN8ShT2R3NGvhgLP7L+LPz3Zh8mejYWJRs+9Sb6Rj5/cH0W1kZ7Twd0fiiSRsXbEPEz4dCTsXG1RVVCHjTjZCBneEvastykrKEfP7P9j61R6MXzhSXs/1Mzex/5cjCB3RGW4+zSGVSpGdmvsMr7xh9OveDu+/2Q8Lvt6JS1fvYdywEKxeOgEvRa5AXkFJjfL6err45bMJyCsowYyF/0NmTiGcHSxRWFwmL7PonSHwdrfH+0s3Iyu3EAPC/PHrZ5Ho/9rXyMotepaXp1bXTyXh6MY49BrfHU6eDjh/8CK2frELkUtHw1jJ+zYtKR17Vh3Ei8M7w9PfHddOJGHnyn0Y88lI2Da3USibdO4W0m9mwMTSpEY9EokELV9ogWYtHHD576sNdn3PUkAzbwxt2xUbLx3BnfwM9PD0x7TgwfgkZh2KK0prlNfX1UduiQjxaUkY1i5UaZ2fHd0InUcW/zmZ22B6yFDE309qsOt41tT5+J958+Zh9uzZCvuEQmGNcra2ttDV1UVmpmKQn5mZCUdHxxrlb968iTt37mDAgAHyfVKpFACgp6eH69evo0WLFnVqY72ntmfNmoXp06cDAMLCwnDt2jVs2LAB8fHxmDFjRn2rowYSFhYGR0dHREVF1VomLi4OXbt2hZGREVxcXDB9+nSUlDz8pSQQCLB9+3aFYywtLbFmzRoAgIeHBwCgQ4cOEAgE6N69O4DqjOjgwYOxePFiNGvWDK1atQIA/P777+jYsSPMzMzg6OiI0aNH/+fdZJqUcOgC2nRti9Zd2sCqmTVCx3SHnoEerh1T/ksiIeYSXNq6wj88AFZO1nhhcBBsXe1w+XCCvIyxhYnCdvfCbTRr5QxzOwsAgGs7N/SI7AWXtq4wt7OAu78HfMP9cfv8rWdyzepyZt8F+HVvC9/QNrB1tkZ4ZHfoC/WQUMsv2HMHL8HT1xVBLwfA1tkaocOD4OBuh/PR1X0nNBbilbmD0CbIGzZOVnD2ckTv8aHIuJ2NwpzqoEcqkSL693/Q/ZUQdOjVDtZOlrB1tkabIO9nddkNZvywLvhr71lsO3AeN1OysWDFTpSVV2Jo30Cl5Yf2DYCFmTGmfbwe8VdSkJZZgDOX7uD6reopLqGBHnp39cEXqw/gbMIdpKTl4bt1h5FyPxevDgx6lpemducOXEC7bm3Rrmsb2DhbI2x89fu2tuDu/KFLcG/vihdeCoBNM2t0GRYEezc7XIhOUChXlF+MI3/8jX5v9oaubs1fnSFDghAY7l8j+NRmvbwCcPzuFZxMSURGUR42XjyMCkkVgt3aKi2fUpCJbYlxOHf/BqqkEqVliitKUVgulm/tHDyQXVyApFztz4Q3BKFQCHNzc4VNWSBpYGCAwMBAxMTEyPdJpVLExMQgODi4RvnWrVsjISEBFy5ckG8DBw5Ejx49cOHCBbi4uNS5jU/1HEkAcHNzw9ChQ+Hr6/u0VZEa6erqYsmSJfjmm2+Qmppa4/WbN2+ib9++GDZsGC5duoRNmzYhLi4O06ZNq/M5HjyfKjo6Gunp6di6dav8tZiYGFy/fh2HDh3C7t27AQCVlZX49NNPcfHiRWzfvh137tzBhAkTnu5CG4ikSoLsu9lwbtNcvk+gI0DzNs2ReVP5epOsWxlw9lF88zVv64LMW8rLiwvFSEm4i9Yv+jyxLRXiCghNan5wNFaSKgky7mTDra1i37m3bY77ycr74n5yBtzaKvadR3sX3E9SXh4AysUVgADyvsm4k43i/BIIdAT47f824dtpv+HPz3ch+552ZyT19XTRtmUznDh/U75PJpPhxPmb8PdR/mHfM7g1LiSm4KPpA/DPX3Oxc/XbeP3VbtDRqU6V6OrqQE9XF+UVVQrHlVVUIaCdW8NdTAOTVEmQeScbbj6KY8+tbXOk1/K+TU/OgNtj/eje3gVpj5SXSWXY/1M0OvbrAFvn5ydQfBJdgQ5cLOxxLTtFvk8G4Fp2Cjytama4VD1Hp+atcSIlUS31NRaa+qrt2bNnY/Xq1Vi7di2uXr2KKVOmoKSkBJGRkQCAcePGYd68eQCqZy3btWunsFlaWsLMzAzt2rWr1yxinaa2V65cWecKH2QrSfOGDBkCf39/zJ8/H7/88ovCa1FRUYiIiMDMmTMBAN7e3li5ciW6deuGH374AYaGhv9Zv52dHQDAxsamRurcxMQEP//8s8JgfO211+T/9vT0xMqVK/HCCy+guLgYpqamdbqm8vLyGguNlf119rTKissgk8pqTGEbmRujICNf6TFikRjGZkYK+4zNjVEqEistf+P4NegL9eERUPsjF0RZBbhyJAGdh4fU8wo0R1xU3XePT2EbmxsjN01535UUiGFiodh3JhbGKKml76oqqhC76QR8OntDaFQ9xgqyqr8Q4djW0+gZ8SIsbM1wet8F/G/Jdkz+PAJGpv89phsjSwtj6OnqIje/WGF/bn4xPFyUr21q7mSNoA6W2B1zCW98sA5uztb4ePpA6Onp4Pvfj0BcWoH4KymYMqYHbqZkIze/GC/38IV/GxekpGlv4F3679gzVjL28tJrGXsiMYwtar5vxY+MvTN7z0NHRwcdejedhImp0Ai6OjooKld8DxaVi+FoZq2Wc/g5tYCRvhAn7z1fgaSmvmp71KhRyM7Oxscff4yMjAz4+/tj//798htwUlJSoKPz1PnDGuoUSH711Vd1qkwgEDCQbGSWLVuGnj17Ys6cOQr7L168iEuXLik8skkmk0EqleL27dto06bNU523ffv2Nf6iOXfuHBYsWICLFy8iPz9fvh4jJSUFPj5Pzso9UNvCY/Oe6vlge5auH7sKr6CW0NNX/jYsyS/G3hW74BnYAm1ClU8lNUWSKgl2fHsAkMnQJ7L7wxf+vYkseGBHtHqhem3PS5N74fsZa3D9dDL8e7Z79o3VEB0dAXILSvDxV9shlcqQmJQGextzTBzZFd//fgQA8P7SzVg8Zwj+3vQ+qiQSJCalY8+RS2jr3UzDrW9cMu9k4fzBixjzySgImtp33zWwYLe2SMy6A1FZzXW+2kyTo2TatGm1zizGxsY+8dgHy9bqq06B5O3bt1WqnDQvNDQU4eHhmDdvnsI0cnFxMd544w2lgb+rqyuA6j8MHr/Du7Kysk7nNTFRXIxeUlKC8PBwhIeHY/369bCzs0NKSgrCw8NRUVFR5+upbeHxd6d+rHMddWFoagiBjgClhYp/jZcWipXeaAMAxhbGEBcpLkAXF4phpOTmkvQbaSjIKEDY6+FK6yopKMGuL7bDoYUTQsf2UPEqNMPYrLrvHs8mVt+FrrzvTCyNUSJS7LsSkbhGVvNBECnKKcKr8wbLs5EP6gAAW+eHN5fp6evC0s4chbmK2TxtUiASo0oigY2VYtbexsoUOfnKrys7twhVEimk0ofv31sp2bCzMYO+ni4qqyS4l56Hce/8AiNDfZgaC5GdV4zl/zcKqbVk3LWB0b9jT6xs7Cl5HwLVmW+xqOb79kFW8/71dIiLSrH6nbXy12VSGY5uPIbzBy9i0pfj1HwVjUNxeSkkUinMhIr9ZiY0RqEaAj9rIzO0tnPB6tN7nrou0qx637VN2mfp0qXw9/eX3/QCAAEBAUhMTISXl1etx9nZ2SE9PV3+c1JSEsTihx/QDzKOEonyRdWPunbtGnJzc7F06VL5It6zZ8/W+1qEQmGDTGU/TldPF3Zudrh/NRUeHaqnnmVSGe5fTUXbnu2VHmPv6Yj7V1PhG/bwkSD3r6bCwbPmeqJrcVdh62YHGyVTkyX5xdj1xXbYutmje2RPCHS0Kwuiq6cLR3c73E1MRcuOD/vuzpVUBPZW3nfOXo64eyUVL/R92Hd3LqfC2fth3z0IIvMzRHj1g8EwMlOcqnb0sIeuvi5y0wvQvFUz+TGinCKY25ip+zKfmcoqCa7cSEPnAE/EHK++YUQgEKBzB0+s33FK6THnr6Sgf09fhT8G3ZvbIiunEJVViu/X0rJKlJZVwtzUEF06euGL1Qca9oIakK6eLhzc7ZCSmAqvwIdjLyUxFf69lI89Jy9HpCSmKjzK5+6VVDRrUT322nRpBddH1vsCwJYvdsEnpBXadm3dQFeieRKZFPdEWWhl54JLGdU3+wkAtLJzwdHbl566/s6uPigqL8XlzOcvUdXUEtfqnyynRqd9+/aIiIhQWOv6/vvv4/jx45g2bRouXLiApKQk7NixQyEl3rNnT3z77beIj4/H2bNn8eabb0JfX1/+ur29PYyMjLB//35kZmZCJBLV2gZXV1cYGBjgm2++wa1bt7Bz5058+umnDXPBatK+tz+u/ZOI68evIT89D/+sj0VlRRVadame9j/8SzRObX34jQHte/ki9UoKLh6MR356Ps7uPI3sO1lo91jgWVFagVvnktFGyU02JfnF2PnFdpjamCF4RAjKikohFpVALNKuqZ8X+vnjYmwiEv65hpz7eTiwJhaV5VVoH1rdd7tXRePopod9F9jHF7cTUnB6bzxy0/IRt/U0Mm5nISCsuu8kVRJs/2Y/Mm5nY8CU3pBKpSguKEFxQQkk/wZGQiMD+Pdsi7itp3E7IQW56fk4uOYoAKB1UN0eY9FYrd1yDCNe6ohBvTvA09UO82cMhJGhAbbtPwcAWPr+MMya2FtefuOu07AwM8IHU1+Cu7MNugW1xOuju2HDzoeBZ5eOXnjxBW84O1ohJKAF1nwxEbfv5WDb/vPP/PrUKTDcHwlHE3El7hpy0/IQva567LXtWj329v0UjX/+ejj2Anr74s7lFJzdF4+8tHwc33Yambez4P/v2DMyNYRtcxuFTVdXByYWxrB2epj9LswtQtbdbBTmFUMqkyHrbjay7majoqzuMy6NTUzyeXRxa4cglzZwMLXCK349IdTVx8l/b44ZF9AHA9s8XL+tK9BBc3NbNDe3ha6ODiwNTdHc3BZ2JhYK9QoABLv64NS9q5Cq6bnGjYqm7rbREGYkm4iFCxdi06ZN8p99fX1x9OhRfPjhh+jatStkMhlatGiBUaNGyct8+eWXiIyMRNeuXdGsWTN8/fXXOHfunPx1PT09rFy5EgsXLsTHH3+Mrl271roGw87ODmvWrMEHH3yAlStXIiAgAF988QUGDhzYYNf8tLxe8EZZUSnO7jgFcaEYti62eGlGf/mz6IrzihTWTDl6OaHnpN44s/0UTm87CQt7S4RP7Sd/huQDyWeqn5fWolPNx9KkJt5DYZYIhVki/PHeWoXX3lg9Vd2X2GDadPaGuKgUcVtOVT+Q3NUWI9/tL59eLMxV7LvmLZ0wYEpv/LP5FP7+6ySsHCwxdGY/2LlU911xfgmSz98BAPz2f5sUzvXqB4Ph2sYZANDjlRDo6Ohg96poVFVUwamFA16ZNwiGJtp5o80D+2Ivw8rCBNMn9IKtlSmu3kzH6/PWIvffZ0g62VsqTGNnZIswee5azH3rJWxfPQ2ZOUX4fesJ/Lzp4bePmZkYYtbEPnC0NYeoqBQH/7mCFb8dQpVE+syvT51aBVWPvePbTkEsEsPO1RZD33k49ooeG3vNvJ3w0hu9cWzrKRzbchKWDpYYOL1fvR/jc3zraSQeuyb/+Y/5fwIARrw/GC7/jk9tcz4tCWZCI/Rv3Rlm/z6Q/LuT2+U34FgZmSksf7IwNMG8HhHyn8O8AxHmHYgbOan4+tgW+f5Wdq6wNjbHibtXnt3FUIMRyNT1NSdEGrT877o/WYAemh06Hb+eZt+p4rVO09Em7P803QytdTV6EX48wbGnijeCp2Pqjq813Qyt9d2ghn3mddBS9X05y6m5s/+7kIYxI0lERESkJlwjWQf//PMPxowZg+DgYNy/X/00+t9//x1xcXFqbRwRERERNV71DiS3bNmC8PBwGBkZIT4+Xv5waJFIhCVLlqi9gURERETao2ndbVPvQHLRokVYtWoVVq9erXAHb5cuXXD+vHbf7UdERET0NAQC9W3aoN6B5PXr1xEaGlpjv4WFBQoKCtTRJiIiIiLSAvUOJB0dHZGcnFxjf1xcHDw9a//OYCIiIqLnXtOa2a5/IDl58mTMmDEDp06dgkAgQFpaGtavX485c+ZgypQpDdFGIiIiIq3QxOLI+j/+Z+7cuZBKpejVqxfEYjFCQ0MhFAoxZ84cvP322w3RRiIiIiKtoC1rG9Wl3oGkQCDAhx9+iHfffRfJyckoLi6Gj48PTE1NG6J9RERERNRIqfxAcgMDA/j41PyuYCIiIiJqGuodSPbo0UPhe0ofd/jw4adqEBEREZG24tT2f/D391f4ubKyEhcuXMDly5cxfvx4dbWLiIiIiBq5egeSX331ldL9CxYsQHFx8VM3iIiIiEhbNbGEpGrfta3MmDFj8Ouvv6qrOiIiIiLt08Se/6O2QPLEiRMwNDRUV3VERERE1MjVe2p76NChCj/LZDKkp6fj7Nmz+Oijj9TWMCIiIiJtw5tt/oOFhYXCzzo6OmjVqhUWLlyIPn36qK1hRERERNqmicWR9QskJRIJIiMj0b59e1hZWTVUm4iIiIhIC9RrjaSuri769OmDgoKCBmoOERERkRbjzTZP1q5dO9y6dash2kJERESk1ZpYHFn/QHLRokWYM2cOdu/ejfT0dBQWFipsRERERE2VQKC+TRvUeY3kwoUL8c477+Cll14CAAwcOFDhqxJlMhkEAgEkEon6W0lEREREjU6dA8lPPvkEb775Jo4cOdKQ7SEiIiLSXtqSSlSTOgeSMpkMANCtW7cGawwRERGRNmtaYWQ910gKmliUTURERES1q9dzJFu2bPmfwWReXt5TNYiIiIhIazWxnFu9AslPPvmkxjfbEBEREVG1JhZH1i+QfOWVV2Bvb99QbSEiIiIiLVLnQJLrI4mIiIierKmFS/W+a5uIiIiIasFAUjmpVNqQ7SAiIiIiLSOQMdVIREREpBa9v1uhtroOTZ2ptroaSr1utiFqrDac/1rTTdBKowNmsO9UNDpgBj4+tFLTzdBaC3tPR/uAdzTdDK2UcP5LLIjm2FPVgrDpDVq/JtdIfvfdd/j888+RkZEBPz8/fPPNN+jUqZPSslu3bsWSJUuQnJyMyspKeHt745133sHYsWPrdc56PZCciIiIiBqfTZs2Yfbs2Zg/fz7Onz8PPz8/hIeHIysrS2l5a2trfPjhhzhx4gQuXbqEyMhIREZG4sCBA/U6LwNJIiIiIi23fPlyTJ48GZGRkfDx8cGqVatgbGyMX3/9VWn57t27Y8iQIWjTpg1atGiBGTNmwNfXF3FxcfU6LwNJIiIiIjURCNS31VVFRQXOnTuHsLAw+T4dHR2EhYXhxIkT/3m8TCZDTEwMrl+/jtDQ0HpdL9dIEhEREamJOtdIlpeXo7y8XGGfUCiEUChU2JeTkwOJRAIHBweF/Q4ODrh27Vqt9YtEIjg7O6O8vBy6urr4/vvv0bt373q1kRlJIiIiokYoKioKFhYWCltUVJTa6jczM8OFCxdw5swZLF68GLNnz0ZsbGy96mBGkoiIiKgRmjdvHmbPnq2w7/FsJADY2tpCV1cXmZmZCvszMzPh6OhYa/06Ojrw8vICAPj7++Pq1auIiopC9+7d69xGZiSJiIiI1ESdaySFQiHMzc0VNmWBpIGBAQIDAxETEyPfJ5VKERMTg+Dg4Dq3XSqV1phK/y/MSBIRERFpudmzZ2P8+PHo2LEjOnXqhBUrVqCkpASRkZEAgHHjxsHZ2Vk+NR4VFYWOHTuiRYsWKC8vx969e/H777/jhx9+qNd5GUgSERERqYmmnkc+atQoZGdn4+OPP0ZGRgb8/f2xf/9++Q04KSkp0NF5OBFdUlKCt956C6mpqTAyMkLr1q3xxx9/YNSoUfU6LwNJIiIiInXR4DfbTJs2DdOmTVP62uM30SxatAiLFi166nNyjSQRERERqYQZSSIiIiI10eR3bWsCA0kiIiIiNWlicSQDSSIiIiK1aWIpSa6RJCIiIiKVMCNJREREpCZNKx/JQJKIiIhIbZrYzDantomIiIhINcxIEhEREakJM5JERERERHXAQJKIiIiIVMKpbSIiIiI1aWpT2wwkiYiIiNSkicWRnNomIiIiItUwI0lERESkLk0sJclAkoiIiEhNmlgcyUCSiIiISF2a2s02XCNJRERERCphRpKIiIhITZiRJHpMbGwsBAIBCgoKnljO3d0dK1aseCZtIiIiIs1jRvI5MmHCBKxduxYAoK+vD1dXV4wbNw4ffPAB9PRU/68OCQlBeno6LCwsAABr1qzBzJkzawSWZ86cgYmJicrn0bTTBxNwfNcFFIvEcHS1Qb8JXeHs5VBr+Ssnk3Hkr9MoyC6CjaMFwl4NhncHN/nrMpkMsZvP4PzhRJSVlMOllRNefi0UNk6WNeqqqpTg5482I/NuLt6IGglHd1v5a8kXUxC7+QyyU/Ogp68LtzbN0GdMCCztzNV6/U9LE/33v8/3IuNuDkoKS2FkIoRnu+YIezUYZtbV4/BO4n2c3HsR929moby0AtaOFgjp3wG+L7ZssH5oKElHE3AtJh5lhWJYOtsgYEQobNyV968oPReXd59G3r1siPOK4D/sRbTq4adQJis5Ddej45GXkoWyQjG6TO6H5n6ez+JSnrlXRnbBhHHdYWtjhus30hD12TZcvnJPadlff5qCFzp61dj/9z+JmDrjFwBAwvkvlR775YpdWLMuVm3tbgxuHE3Ateh4lBaKYeVsg8CRTxh3abm4tOc08lOyUZJXhA7DXkTrno+Nu6Q0XI2OR/69LJSKxOj6+vM77poKZiSfM3379kV6ejqSkpLwzjvvYMGCBfj888+fqk4DAwM4OjpC8B/5ejs7OxgbGz/VuTTl8okkHPz9GLoN64g3loyAg5st/li6GyUisdLy926kY8s3h9Chexu8ETUCrTp6YOOX+5B1L1de5tiueJzafwkvT+yGSZ8Og4FQD38s3Y2qiqoa9R3acBxmVjWD8PysQmz8ch882jrjjaUjMWbeAIgLy7Bp+X71XbwaaKr/3Ns6Y8SMPpj25WiMnBWOvEwR/lyx/5HzZMDe1QYjZ4XjzWWj4N+tNbZ/H4Mb5+80WF80hJRzSbiwLQ5t+72APu+PhKWzLY5+twtlRcr7t6qiCia25vAbGAxDc+XvSUl5JSydbRA4qltDNl3jwvv4493ZA7Hqp4MYOfor3EhKw4/fvQ5rK1Ol5WfOWYPuvRfIt8HDP0NVlQQHoy/Jyzz6evfeC/DRgo2QSqWIjrmktE5tdfdcEuK3xqHdSy+g79yRsGxuiyPfPmHcVVbB1MYcfoNqH3dVFZWwam6DwJHP77gTCNS3aQMGks8ZoVAIR0dHuLm5YcqUKQgLC8POnTuRn5+PcePGwcrKCsbGxujXrx+SkpLkx929excDBgyAlZUVTExM0LZtW+zduxeA4tR2bGwsIiMjIRKJIBAIIBAIsGDBAgCKU9ujR4/GqFGjFNpWWVkJW1tbrFu3DgAglUoRFRUFDw8PGBkZwc/PD5s3b274TlLi5J6LCOjpgw7d28CuuTX6T+wGfQM9xMdeU1r+1L5L8PJzRZcBHWDnbI2eI4Pg5GGH0wcSAFRn007tu4TQIYFo3dEDDm62GPxWLxTll+Da2dsKdSVduItbl+6hT0RIjfOk386GTCpDz5FBsHawgJOHHYL7+yPjbg4kVRL1d4SKNNV/wS/5obm3IyztzODS0gkvDgxAanKmvG+6Dg5Ez5FBcGnpBGsHC3Tu5wcvPxdcPX2r4TtFja4fvgDPkLbwDG4DCydrdHylO/QM9HD7xFWl5W3cHOA/pAtcO3pDR09XaRmntm5oP6Dzc58NGhcRii3bTmL7zjO4dTsTCxdvQWlZJYYM6qS0fGFhKXJzi+RbcOeWKCurxMFDF+VlHn09N7cIPbq1w+mzN5F6P+9ZXdYzcT3mAlo8Mu5e+Hfc3XrCuOswtAvcOnpDt5Zx16ytG3wHdIaL//M77gRq3LQBA8nnnJGRESoqKjBhwgScPXsWO3fuxIkTJyCTyfDSSy+hsrISADB16lSUl5fj77//RkJCApYtWwZT05p/sYeEhGDFihUwNzdHeno60tPTMWfOnBrlIiIisGvXLhQXF8v3HThwAGKxGEOGDAEAREVFYd26dVi1ahWuXLmCWbNmYcyYMTh69GgD9YZykioJ0m5nw7Ndc/k+gY4Anu2aIzUpQ+kx95IyFcoDQAtfF6QmZQIACrIKUVwghmc7F/nrhsZCNG/hgHuP1FlcIMau1bEY8lYY9IU1lx84edhBIADij16FVCpFmbgcl/65Ds92zWv9oH7WNNl/jyotLkPCsRtwaen4xL4pK62AkamwztenaZIqCfLvZcOhlWL/OrRqjpzbyvuCqunp6cKnTXOcPPXwj2aZTIaTp27Az9ftCUc+NHRQEPYfjEdpWYXS122sTdH1xTbYtv2UWtrcWEiqJMi7lw3H1o+Nu9bNkXOL444e4hrJ55RMJkNMTAwOHDiAfv36Yfv27Th27BhCQqqzXuvXr4eLiwu2b9+OESNGICUlBcOGDUP79u0BAJ6eyv9aNDAwgIWFBQQCARwdHWs9f3h4OExMTLBt2zaMHTsWALBhwwYMHDgQZmZmKC8vx5IlSxAdHY3g4GD5OePi4vDjjz+iWzfl0x7l5eUoLy9X2CcUPl1QIC4sg0wqg4mF4lSMiYURctLylR5TXCCuUd7UwhjFBdVTPsX/TumaWBjVqLPk3zIymQw7Vh1Gx15t0ayFPQqyC2ucx8reHGPmDcDmrw9i989HIZPK0NzbARHv91ftYhuApvrvgUMbTuDMwQRUllehubcDXn335VrbeuVEMtJuZqH/xO51urbGoKK4un8NzRT7y9DcGIWZyvuXqllZmkBPTxe5eUUK+3PziuHhbv+fx7dr6wJvbyd8vHBTrWUGDngBYnE5og8nPHV7G5Py2sadmTGKMjjunkRbpqTVhRnJ58zu3bthamoKQ0ND9OvXD6NGjcKECROgp6eHoKAgeTkbGxu0atUKV69WT1FMnz4dixYtQpcuXTB//nxcuvR0a3309PQwcuRIrF+/HgBQUlKCHTt2ICIiAgCQnJwMsViM3r17w9TUVL6tW7cON2/erLXeqKgoWFhYKGxRUVFP1VZNOX0gAeVlFXhxcECtZR5kLP1CW2HyouGY8PFg6Orp4s8VByCTyZ5haxuvLv398UZU9fpRgY4A27+PVto3t6/cx44fD2PA5O6wd7HWQEtJ2wwdHIQbSWm13pgDAEMGdsKefedRoWTtMzVNXCNJWq1Hjx64cOECkpKSUFpairVr1/7nTTIAMGnSJNy6dQtjx45FQkICOnbsiG+++eap2hIREYGYmBhkZWVh+/btMDIyQt++fQFAPuW9Z88eXLhwQb4lJiY+cZ3kvHnzIBKJFLZ58+Y9VTuNzQ0h0BHUuDGkRFQKU0vlC8ZNLY1rlC8WieXlTf/NtpWISmvUafJvmdtX7iP1RiYWjf0RCyN+wMqZ1UH3Tx/+he3fxwCovhPa0NgAvSNC4ORhB7c2zTB0ahhuX07F/eTMp7puddFU/z08vxFsnCzRwtcFw9/ug6QLKfIp8gfuJN7H/z7fg/CxXeAX2rr+F6lBBqbV/fv4DQ5lheJab2igavkFJaiqksDG2kxhv421KXJzi2o5qpqRoQH69vHH1u2nay0T0MEDHh722LLtpFra25gIaxt3RRx3pIiB5HPGxMQEXl5ecHV1lT/yp02bNqiqqsKpUw/X8OTm5uL69evw8fGR73NxccGbb76JrVu34p133sHq1auVnsPAwAASyX/f6BESEgIXFxds2rQJ69evx4gRI6Cvrw8A8PHxgVAoREpKCry8vBQ2FxeXWusUCoUwNzdX2J52altXTxfNPOxw6/J9+T6ZVIZbV1LR3Fv59L2LtwNuX0lV2Hcr4R6ae1c/FsPS3hymlsa4dflhmXJxBVJvZsLl3zr7jX8Rby4biTeXVm8R71dPyQ6f3gc9R1VnjyvLq2r8ISDQqf65sWQkNdV/yjzok0dvRLqTeB8bPtuDsNHBCOzVtv4XqGG6erqwcrFD5vWHfSGTypB5IxW2HrX3BQFVVRIkXk1FUCdv+T6BQIDOnbxx8dLdJx7bp7cfDAz0sHvvuVrLDB0UhCuJ93AjKV1tbW4sdPV0Ye1ih4zHx931VNh6ctzRQ1wj2QR4e3tj0KBBmDx5Mn788UeYmZlh7ty5cHZ2xqBBgwAAM2fORL9+/dCyZUvk5+fjyJEjaNOmjdL63N3dUVxcjJiYGPj5+cHY2LjWx/6MHj0aq1atwo0bN3DkyBH5fjMzM8yZMwezZs2CVCrFiy++CJFIhGPHjsHc3Bzjx49Xf0c8QeeX/bD9h8No5mkHZy97nNx3CZXlVfDvVp292vZ9NMysTBD2avV6zqB+vlizcAeO776Alh3ccPlEEtJuZWPA5O4Aqn9ZBfXzxT/bz8HG0QKW9uY48tdpmFmZoHVHDwCAha1ilsTAsDrItnawgLlN9Y1OLTu44eS+izi65QzahXijoqwSMRtPwsLWDI7uds+ia+pEE/2XmpyJtJtZcG3lBEMTIfIzRTjy12lYOZjLA9jbV6ozkUF9feHTqYV8Daaung6MTA2fcS+prlVPf5z6PQbWrvawcbfH9SMXUVVeBY/O1e/Rk+uiYWxhAt9B1f0rqZKgMKP6DmJplQSlBcXIT82GnlAfZnaWAIDK8goUZ4vk5yjJLUR+ajYMjA1h8lgGT5utW/83Fn/yCq4k3kPClRSMHR0KIyMDbN9ZnWlcvPBVZGWJ8PW3exWOGzK4Ew7HXoaolkdYmZgI0bu3L75YvqvBr0FTWvXyx8l1j4y7w4rj7sTaaBhZmsD/0XGX/u+4k/w77u79O+7sLQEAlWWK4644txD597JhYPL8jDttmZJWFwaSTcRvv/2GGTNmoH///qioqEBoaCj27t0rzxBKJBJMnToVqampMDc3R9++ffHVV18prSskJARvvvkmRo0ahdzcXMyfP1/+CKDHRUREYPHixXBzc0OXLl0UXvv0009hZ2eHqKgo3Lp1C5aWlggICMAHH3yg1muvi3bB3hAXliF282kUF4jh6GaLiLn95VOtopxihcygS0snDJ0WhiN/nsbhTSdh7WiJV97pB3sXG3mZLgM6oLK8Crt+jkWZuAKurZwwZm5/6BnU/W3n0a45hk3rjWO74nFsVzz0hfpw8XbAmLn9oV+PehqaJvpP30APV0/fQuzm06gor4KZpTFa+Lli+JA+0NOvvmv74t/XUFlehbgd5xG347y8brc2zTDh48HPoGfUwzXQG+XFpbi85xTKisSwdLZFt6n95VOM4rwihf4tE5Xg4NI/5T9fj7mA6zEXYOfVDD1nVj81If9uNo6s3C4vc2HrMQCAe1BrBI3t9Qyu6tk4cPACrK1MMHVKOGxtzHHt+n28OW01cvOql9c4OVpCJlXM7ru72SGwgyden/JjrfX2C+8AAQTYdyC+QduvSW6B3igvKkXC7upxZ+Vsi+5T+8PowbjLVxx3paIS7H9k3F2LuYBrMRdg790Mvf4dd3kp2Tj89XZ5mfgt1ePOI6g1Oo97PsZdE4sjIZA1lvkxoqew4fzXmm6CVhodMIN9p6LRATPw8aGVmm6G1lrYezraB7yj6WZopYTzX2JBNMeeqhaETW/Q+kdvUN9n6obRM9RWV0NpPCkNIiIiIi3HqW0iIiIiUkkTiyN51zYRERERqYYZSSIiIiJ1aWIpSQaSRERERGrS1NZIcmqbiIiIiFTCjCQRERGRmjSxhCQzkkRERETqIhCob6uv7777Du7u7jA0NERQUBBOn679u+JXr16Nrl27wsrKClZWVggLC3ti+dowkCQiIiJSE4Eat/rYtGkTZs+ejfnz5+P8+fPw8/NDeHg4srKylJaPjY3Fq6++iiNHjuDEiRNwcXFBnz59cP/+/Xqdl4EkERERkZZbvnw5Jk+ejMjISPj4+GDVqlUwNjbGr7/+qrT8+vXr8dZbb8Hf3x+tW7fGzz//DKlUipiYmHqdl2skiYiIiNREnXdtl5eXo7y8XGGfUCiEUChU2FdRUYFz585h3rx58n06OjoICwvDiRMn6nQusViMyspKWFtb16uNzEgSERERqYk6p7ajoqJgYWGhsEVFRdU4Z05ODiQSCRwcHBT2Ozg4ICMjo07tfv/999GsWTOEhYXV63qZkSQiIiJqhObNm4fZs2cr7Hs8G6kOS5cuxcaNGxEbGwtDQ8N6HctAkoiIiEhN1Dm1rWwaWxlbW1vo6uoiMzNTYX9mZiYcHR2feOwXX3yBpUuXIjo6Gr6+vvVuI6e2iYiIiNRFA7dtGxgYIDAwUOFGmQc3zgQHB9d63GeffYZPP/0U+/fvR8eOHet+wkcwI0lERESk5WbPno3x48ejY8eO6NSpE1asWIGSkhJERkYCAMaNGwdnZ2f5Gstly5bh448/xoYNG+Du7i5fS2lqagpTU9M6n5eBJBEREZGaaOqbbUaNGoXs7Gx8/PHHyMjIgL+/P/bv3y+/ASclJQU6Og8non/44QdUVFRg+PDhCvXMnz8fCxYsqPN5GUgSERERqYk610jW17Rp0zBt2jSlr8XGxir8fOfOHbWck2skiYiIiEglzEgSERERqYkGE5IawUCSiIiISE00ObWtCQwkiYiIiNSkicWRXCNJRERERKphRpKIiIhITTi1TUREREQqaWqBJKe2iYiIiEglzEgSERERqUkTS0gykCQiIiJSF05tExERERHVATOSRERERGrSxBKSDCSJiIiI1KWpTW0LZDKZTNONICIiInoeTN/1tdrqWjlghtrqaijMSNJz4fezKzXdBK00tuN0zDvAvlNFVPh0/N9B9p2qFvWZjp9Osv9U8Xrn6fDp+aGmm6G1Eg8vbtD6m1hCkoEkERERkbo0taltBpJEREREatLE4kg+/oeIiIiIVMOMJBEREZGacGqbiIiIiFTSxOJITm0TERERkWqYkSQiIiJSE05tExEREZFKmlogyaltIiIiIlIJM5JEREREatLEEpIMJImIiIjURdDE5rY5tU1EREREKmFGkoiIiEhNmlY+koEkERERkdo0sZltBpJERERE6tLE4kiukSQiIiIi1TAjSURERKQmOk0sJclAkoiIiEhNmlgcyaltIiIiIlINM5JEREREasK7tomIiIhIJU0sjuTUNhERERGphoEkERERkZoIBOrb6uu7776Du7s7DA0NERQUhNOnT9da9sqVKxg2bBjc3d0hEAiwYsUKla6XgSQRERGRmgjUuNXHpk2bMHv2bMyfPx/nz5+Hn58fwsPDkZWVpbS8WCyGp6cnli5dCkdHx/pephwDSSIiIiItt3z5ckyePBmRkZHw8fHBqlWrYGxsjF9//VVp+RdeeAGff/45XnnlFQiFQpXPy5ttiIiIiNREnXdtl5eXo7y8XGGfUCisEfhVVFTg3LlzmDdvnnyfjo4OwsLCcOLECfU1SAlmJImIiIjURJ1T21FRUbCwsFDYoqKiapwzJycHEokEDg4OCvsdHByQkZHRINf5ADOSRERERGqizq9IfG/ePMyePVth39NMQzcEBpJEREREjZCyaWxlbG1toauri8zMTIX9mZmZT3UjTV1waruRiI2NhUAgQEFBgaabolRd2+fu7q7yIwSIiIi0nSbu2jYwMEBgYCBiYmLk+6RSKWJiYhAcHPy0l/REjT4jOWHCBKxdu7bG/vDwcOzfv79OdXTv3h3+/v7PfYDzaF/p6+vD1dUV48aNwwcffAA9vaf7rw4JCUF6ejosLCwAAGvWrMHMmTNrBJZnzpyBiYnJU52rMTl7MAEn9sSjWCSGg6sNwseHwrmFQ63lE08l4+hfp1CQUwRrBwv0ejUYXv7u8tevnbmJc9FXkHEnC6XF5Zi0eCQc3e0U6li3aBtSrqYp7Avo2RYvTeyuzkt75jq7+CLUIwCmBsbIKMrBzmtHkSrKVFrW3sQavb07w9ncHlZG5th97W8cu3tBoUw3j45o59ACdiZWqJRU4W5BOvbfOIYccUHDX4wGJP+dgOsx8SgrFMPS2QYdhofC2l35WBSl5+LKntPIv5cNcV4R/Ia+iJY9/BTKXD14Dvcv3kJRZj509fVg4+EI30HBMHOwehaX06DioxNwdl88SkRi2LnYoOeYUDg94X17/XQyjm09hcKcIlg5WKDryGB4+rnLXz++7TSunUpCUW4xdPV04eBuhxeHB8GpRXWm597V+/hz6XaldUfMHw5Hz9rPrY1eHRSE10Z1ha21Ka7fzMDib3Yj4Vqq0rJrlk9EJ3/PGvuPnryOKR+sa+imaoSmviJx9uzZGD9+PDp27IhOnTphxYoVKCkpQWRkJABg3LhxcHZ2lq+xrKioQGJiovzf9+/fx4ULF2BqagovL686n7fRB5IA0LdvX/z2228K+9S9RkAmk0EikTx1wKVpD/qqvLwce/fuxdSpU6Gvr69wJ5cqDAwM6pQet7Oz+88y2uLKiSQcWh+Hfq91h3MLB5zefxH/W7oLU74YDRML4xrl791Ix7ZvD6LnqM7w7uCOy8eT8OfyfZi0eCTsXWwAABVlVXBp5QSfzl7Y8/ORWs/doYcPug3vJP9Z30Bf/Rf4DLV39MbLrbti+5XDuCfKRBc3f7wWOAhfxv2OkorSGuUNdPWQJxYhISMJL7cOVVqnp7UzTqRcQqooEzo6Ogj3DsZrHQfjq2N/oFJS1dCX9EzdO5eEi9viEDCqO2zcHHAj9iL+/n4X+n40GoZmNceipKIKJrbmaN7BCxe3ximtMzs5DV5d28HKzR4yiQwJu07i7+92IvzD0dATau94u3YqCUf/F4ew8d3h1MIB5w5cxJYvduG1ZaNhbF6zr+4npWPPDwfRdURnePq749qJJOz4eh/GLhwJ2+bV71srR0v0GhsKCztzVFVU4dyBi9j8+S5M/GwMjM2N0MzbEW9+PUGh3mNbTyMlMRUOHvbP4rKfmb7d2+P9KS/hkxU7cOnqPYwd1gU/LZuAl8d/hbyCkhrlZ8zfAH09XfnPlhbG2Lp6Gg4cTXiWzW4SRo0ahezsbHz88cfIyMiAv78/9u/fL78BJyUlBTo6Dyei09LS0KFDB/nPX3zxBb744gt069YNsbGxdT6vVkxtC4VCODo6KmxWVtV/NcfGxsLAwAD//POPvPxnn30Ge3t7ZGZmYsKECTh69Ci+/vprCAQCCAQC3LlzRz5Vu2/fPgQGBkIoFCIuLg5SqRRRUVHw8PCAkZER/Pz8sHnzZnndD447cOAAOnToACMjI/Ts2RNZWVnYt28f2rRpA3Nzc4wePRpisVh+3H/V+6iSkhKYm5vXeH379u0wMTFBUVHRf/aVm5sbpkyZgrCwMOzcuRMAkJ+fj3HjxsHKygrGxsbo168fkpKS5MfevXsXAwYMgJWVFUxMTNC2bVvs3btX4boLCgoQGxuLyMhIiEQieZ8uWLAAgOLU9ujRozFq1CiF9lVWVsLW1hbr1q2rd788a6f2XUCHHm3h360N7Jpb46XXukNfqIcLR68qLX9m/yW08HVFcP8A2Dpbo/uIIDi52+HswYcfmL5dWyF06AvwaNf8iefWF+rB1NJEvgmNDdR6bc9aV7cOOJN6GefSriKrJA/bEw+jQlKFjs4+SsunFmZh341juJSRBIlUorTMb+d24Py/9WUU5WBzQjSsjMzhbP58/eIGgBtHLsAjuC08OreBuZM1Akd1h66BHu6cUD4Wrd0c4De4C1wDvaHzyC/xR4W+NQDundvAwskGls1t0WlML4jzi5F/L7shL6XBndt/Ae27tUW70DawcbZG7wndoW+gh4S/lffV+YOX4NHeFS+8FACbZtboMiwIDu52iI9++L5tE9wSbm1dYGlvAdvmNug++kVUlFYg+14OAEBXTxcmlibyzdDUEMnnb6Nt19YQaCo91UAmjOiCv/aexbb953HzbjY++WoHysorMbRfoNLyoqJS5OQXy7fgQC+UlVXiwNHLz7jlz46mHkgOANOmTcPdu3dRXl6OU6dOISgoSP5abGws1qxZI//Z3d0dMpmsxlafIBLQkkDySbp3746ZM2di7NixEIlEiI+Px0cffYSff/4ZDg4O+PrrrxEcHIzJkycjPT0d6enpcHFxkR8/d+5cLF26FFevXoWvry+ioqKwbt06rFq1CleuXMGsWbMwZswYHD16VOG8CxYswLfffovjx4/j3r17GDlyJFasWIENGzZgz549OHjwIL755ht5+brWCwAmJiZ45ZVXamRhf/vtNwwfPhxmZmZ17h8jIyNUVFQAqJ76Pnv2LHbu3IkTJ05AJpPhpZdeQmVlJQBg6tSpKC8vx99//42EhAQsW7YMpqamNeoMCQnBihUrYG5uLu/TOXPm1CgXERGBXbt2obi4WL7vwIEDEIvFGDJkSL375VmSVEmQfjtbIeAT6Ajg3q457icpf5RCanIGPNq5KOzz9HVBanL9H71w+dgNfPnGL/jx/f/h8MYTqCyvrHcdjYWuQAfNzO2RnHtPvk8G4GbuPbhaOqntPIb61cF2aWWZ2upsDKRVEuTfy4ZDK8Wx6NCqOXLvqO+xHpVl1c+qMzBuXHeE1oekSoLMO9lwbavYV65tmyO9lvdhenIGXNsqvm/d2rnUWl5SJcGlI1cgNDaAnaut0jI34++grLgM7bq2UfFKGid9PV34tGyGk+eS5ftkMhlOnEuGv49rneoY1i8Qe48koLRMez/T/osmvyJRE7RiHnf37t01ApoPPvgAH3zwAQBg0aJFOHToEF5//XVcvnwZ48ePx8CBAwEAFhYWMDAwgLGxsdKp2YULF6J3794Aqh/8uWTJEkRHR8sXp3p6eiIuLg4//vgjunXrJj9u0aJF6NKlCwBg4sSJmDdvHm7evAlPz+q1IMOHD8eRI0fw/vvv16veByZNmiRfl+jk5ISsrCzs3bsX0dHRdeozmUyGmJgYHDhwAG+//TaSkpKwc+dOHDt2DCEhIQCA9evXw8XFBdu3b8eIESOQkpKCYcOGoX379vI2KmNgYAALCwsIBIInTneHh4fDxMQE27Ztw9ixYwEAGzZswMCBA2FmZqZSv9T2cFZ1ExeVQSaV1ZjCNjU3Rm5avtJjigvEMLEwUthnYmGMkgKx0vK1aRfSEha2ZjCzNEHmvRwc/t8J5KYXYMSsfvW7iEbC2MAIujo6KC5X7IeiCjHsTNSzHk8AoH+rUNzJT0NmcZ5a6mwsykuqx6LhY9OyhmbGKMpUPhbrSyaV4cKWONh4OsGimY1a6tSE0lret8YWxshLV95XJSIxjM2VvG9FiuP15oU72PP9AVRWVMHUwgTD3x0IYzPF4x64/Hci3Nu7wMy65h/i2szSwhh6urrIyS9W2J+bXwxP1/9e1tS+dXO09HTER19sa6gmkgZoRSDZo0cP/PDDDwr7rK2t5f82MDDA+vXr4evrCzc3N3z11Vd1rrtjx47yfycnJ0MsFssDywcqKioU1hEAgK+vr/zfDg4OMDY2Vgi8HBwc5F+WXp96H+jUqRPatm2LtWvXYu7cufjjjz/g5uaG0FDl68UeeBB0V1ZWQiqVYvTo0ViwYAFiYmKgp6enkOa2sbFBq1atcPVq9ZTP9OnTMWXKFBw8eBBhYWEYNmyYwnXWl56eHkaOHIn169dj7NixKCkpwY4dO7Bx40aV+yUqKgqffPKJwr758+ejRX9rpeW1UUDPtvJ/27vawNTSBOuX7EBepgjWDhYabFnjNbBNdziY2WDVqcaxLELbnP/rKETpeegxc6imm9JoubZxxthPR6G0qAwJRxOx67sDiJg/vMa6y6K8YtxJuIf+U8M11NLGa1i/QFy/mVHrjTnPCy1JJKqNVgSSJiYm/3kH0fHjxwEAeXl5yMvLq/Odw4+WezAFu2fPHjg7OyuUezzrpa//cDG6QCBQ+PnBPqlUWu96HzVp0iR89913mDt3Ln777TdERkb+53qbB0G3gYEBmjVrVq+bhyZNmoTw8HD51HxUVBS+/PJLvP3223Wu43ERERHo1q0bsrKycOjQIRgZGaFv374AVOuXebU8nPXPhB9VbqMyxmaGEOgIamQligvFMFVyow0AmFoao0SkeONIiUgME0vl5evqwV3i+VoaSIorSiGRSmEqVOwHMwNjFFXUL1urzMA23dDazgM/ndmCwvLi/z5AywhNqsdiWaFiX5UViWtkKVVx/s+/kX75LnrMGAJjK+3OoBnV8r4Vi8RKb5ADqrOP4kIl79vHyusL9WHlYAkrB6CZlyN+ee8PJBy9iqABimsDL/9zFYamhmjRwf3pL6iRKRCJUSWRwPaxcWJjZYqcvCe/94wM9dGvhy++WVO3WTVtpi1T0uqi9WskAeDmzZuYNWsWVq9ejaCgIIwfP14exAHVGUuJRPmC/Uf5+PhAKBQiJSUFXl5eCtuj6yrrS9V6x4wZg7t372LlypVITEzE+PHj//NcD4JuV1dXhSCyTZs2qKqqwqlTp+T7cnNzcf36dfj4PLzhwcXFBW+++Sa2bt2Kd955B6tXr1Z6nrr2aUhICFxcXLBp0yasX78eI0aMkAfdqvSLUCiEubm5wtYQU9u6erpw8rDD7SsP/3KWSWW4czkVzt7Kp/ObeznizhXFv7RvX05Fc6+nexhs5t3qBf2mTxmQaopEJkVaYRZaWD/8PxUAaGHjgpSC9Keqe2CbbvCxb4Gfz25FfmnhU7a0cdLR04WVix2ybiiOxawbqbBxV31syWQynP/zb9y/dAvd3h4EE1tzdTRXox48miclUbGvUhJT4VTL+9DJy1GhPADcvVJ7+UfrlVQpfgbKZDJc+ecafLq0gm4tNzlps8oqCRJvpKFzQAv5PoFAgM4BLXAhMeWJx4Z3awcDA13sir7QwK2kZ00rMpLl5eU1vitST08Ptra2kEgkGDNmDMLDwxEZGYm+ffuiffv2+PLLL/Huu+8CqL4z6dSpU7hz5w5MTU0VpsUfZWZmhjlz5mDWrFmQSqV48cUXIRKJcOzYMZibm9cpkFNnvVZWVhg6dCjeffdd9OnTB82bP/lO3yfx9vbGoEGDMHnyZPz4448wMzPD3Llz4ezsjEGDBgEAZs6ciX79+qFly5bIz8/HkSNH0KaN8sXi7u7uKC4uRkxMDPz8/GBsbAxjY+WBzujRo7Fq1SrcuHEDR448fORNQ/W3ugT188fOH2Pg5GEP5xb2OLX/IirLq+DXrbpPdvwQDTMrE/R8pXp95wt9ffH7ou04uSceXh3cceVEEtJuZSk8/7G0uAyinCIU//uYjNz0AgDVQaKppQnyMkW4cvwGvPzdYGRqiKyUXBz8Iw6urZvBoZaF/drgn7vxGNGuN+4XZsof/2Ogq4dz96ufYTaiXW8UlpfgQFL1zIKuQAf2ptbyf5sLTeBkZosKSSVyxSIAwKA23eHn1Aq/x+9GeVUlTA2qx19ZVTmqarnTW1u17OGP03/EwMrVHtZu9kiKvYiq8iq4d64ei6fXRcPI0gTtB1aPRWmVBIUZefJ/l4qKUZCaDT2hPkztLAEA8X/+jZRzN9Bl8kvQN9RHWWH1mNQ3FELXQCt+NSgV2Ncf+1fHwNHDHo6e9jh/oPp9++DGl30/RsPUygRdR1b3VUAfX/wZtR1n98XDw88d108lIfN2FvpEdgcAVJZX4uTOs2jRwQOmlsYoLSpDfEwCigtK0PKFFgrnTklMhSi7EO27KX8awfNgzV/HEDV3GC5fv4+Ea6kYNywERoYG2Lb/HAAgau5wZOUU4qufDyocN6xfR8TEXYWosObjvp43TS0jqRWfFvv374eTk+Ldna1atcK1a9ewePFi3L17F7t37wYAODk54aeffsKrr76KPn36wM/PD3PmzMH48ePh4+OD0tJS3L59u9Zzffrpp7Czs0NUVBRu3boFS0tLBAQEyG/sUZWq9U6cOBEbNmzAa6+99lTnB6rv+p4xYwb69++PiooKhIaGYu/evfIMoUQiwdSpU5Gamgpzc3P07du31vWmISEhePPNNzFq1Cjk5uZi/vz58kcAPS4iIgKLFy+Gm5ub/AalBxqqv9WhbbA3xEWlOLr5FEpEYji42eLV9/vLp7ZFuUUKSw1cWjph8NTeiP3rFI78eRLWjpYYObuf/BmSAHDj3G3s+umw/Odt31Z/2HYd+gK6DesEXT0d3L6citP7L6KivArm1qZo80ILvDj44VpebZSQkQRTAyOEeXWGmdAE6YXZ+O3cDhT/+wxJSyMzyCCTlzcTmmB6yGj5z6EegQj1CMStvFSsPrMVANDZtXr97uudhimc66+EQzifpvxRL9rKJdAb5cWluLLnFMqKxLB0tkXXt/rLp7bF+UUKv71KRSU4tOxP+c83Yi7gRswF2Hk1Q/cZ1U9MuBlX/fiV2JXbFc71QkRPeYCqjVoHeaO0sBTHtp6CWCSGnasths3pL5+qLswrguCRL0N29nbCS2/2xrEtpxC3+SQsHSwxaEY/+TMkBQIB8tILkBi3H6XFpTA0NYSjhz1e+WCIvMwDl/++imZejrBppv0Pda/N/tgEWFua4O3IXrC1MsO1m+l44/01yM2v/kPEyd4CUqlM4Rh3F1sE+rpj4ru/aqLJz9xzMdVbDwKZTCb772KkKb///jtmzZqFtLQ0GBho97MEG9LvZ1dquglaaWzH6Zh3gH2niqjw6fi/g+w7VS3qMx0/nWT/qeL1ztPh0/NDTTdDayUeXtyg9X/1j/rG9ayu09VWV0PRioxkUyQWi5Geno6lS5fijTfeYBBJREREjU5Ty8Bqjc8++wytW7eGo6PjU3+9IRERET0bmvxmG01gINlILViwAJWVlYiJiVH67TJERETU+DS1b7ZhIElEREREKuEaSSIiIiI10ZJEotowkCQiIiJSE22ZklYXTm0TERERkUqYkSQiIiJSkyaWkGQgSURERKQunNomIiIiIqoDZiSJiIiI1KSJJSQZSBIRERGpS1Ob2mYgSURERKQmTW3NYFO7XiIiIiJSE2YkiYiIiNSEU9tEREREpJImFkdyapuIiIiIVMOMJBEREZGacGqbiIiIiFTSxOJITm0TERERkWqYkSQiIiJSE05tExEREZFKmlogyaltIiIiIlIJM5JEREREatLEEpIMJImIiIjUpalNbTOQJCIiIlKTprZmsKldLxERERGpCTOSRERERGrCqW0iIiIiUokAMk034Zni1DYRERERqYQZSSIiIiI1aWpT2wKZTNa0crBEz0h5eTmioqIwb948CIVCTTdH67D/VMe+ezrsP9Wx75oeBpJEDaSwsBAWFhYQiUQwNzfXdHO0DvtPdey7p8P+Ux37runhGkkiIiIiUgkDSSIiIiJSCQNJIiIiIlIJA0miBiIUCjF//nwuOFcR+0917Lunw/5THfuu6eHNNkRERESkEmYkiYiIiEglDCSJiIiISCUMJImIiIhIJQwkiYiIiEglDCSJiIiISCUMJInU6J9//sGYMWMQHByM+/fvAwB+//13xMXFabhlREQNg597TRsDSSI12bJlC8LDw2FkZIT4+HiUl5cDAEQiEZYsWaLh1hERqR8/94iBJJGaLFq0CKtWrcLq1auhr68v39+lSxecP39egy3TLsxu1E9hYWGdN3oyjr364+ceMZAkUpPr168jNDS0xn4LCwsUFBQ8+wZpIWY36s/S0hJWVlZP3B6Uodpx7KmGn3ukp+kGED0vHB0dkZycDHd3d4X9cXFx8PT01EyjtMyD7Ma4ceOwceNG+f4uXbpg0aJFGmxZ43XkyBFNN+G5wLGnGn7uEQNJIjWZPHkyZsyYgV9//RUCgQBpaWk4ceIE5syZg48++kjTzdMKzG7UX7du3TTdhOcCx55q+LlHDCSJ1GTu3LmQSqXo1asXxGIxQkNDIRQKMWfOHLz99tuabp5WYHZDPcRiMVJSUlBRUaGw39fXV0Mtavw49lTDzz0SyGQymaYbQfQ8qaioQHJyMoqLi+Hj4wNTU1NNN0lrREVF4Y8//sCvv/6K3r17Y+/evbh79y5mzZqFjz76iL+Y/kN2djYiIyOxb98+pa9LJJJn3CLtwbH3dPi513QxI0mkJn/88QeGDh0KY2Nj+Pj4aLo5WonZjaczc+ZMFBQU4NSpU+jevTu2bduGzMxMLFq0CF9++aWmm9eoceyphp97xIwkkZrY2dmhtLQUAwcOxJgxYxAeHg5dXV1NN0srMbuhGicnJ+zYsQOdOnWCubk5zp49i5YtW2Lnzp347LPP+BibOuDYqx9+7hEf/0OkJv/f3r1HRVWvbwB/BuQ2XFRKRBCRkVJJNMTsmCZ5KRWPly7LLigXiVITORAm1iJTEi9lBWrqSQM0NU2UBCkpQw+C2SlunkQURFHBZIniwZHr7N8f/pjjiBeY2cwG5vms5VrO3tuZR9d2+/rOd7+7vLwc3377LWQyGWbMmIFevXrhnXfeQVZWltTROoxvvvkGSqUSpqamcHNzw/Dhw/kPeSvcvHkTdnZ2AIDu3bujoqICAODu7s6Zfg/Bc087vO4RC0kikXTp0gV///vfsX37dly5cgWff/45zp07hzFjxqBfv35Sx+sQQkNDYWdnhzfeeAOpqalc09dK/fv3R2FhIQBgyJAh2LRpEy5duoSNGzeiV69eEqdr33juaYfXPWIhSdQG5HI5JkyYgEmTJuGxxx7DuXPnpI7UIbC7oZuQkBCUl5cDAJYsWYIffvgBffr0QWxsLIdqPwTPPd3xumeYuEaSSERKpRL79u3D9u3bcejQITg5OeH111+Hj48PBgwYIHW8DqXpz3LHjh34+eef0bt3bxQXF0sdq0NRKpU4deoU+vTpg0cffVTqOB0Gz73W4XXPsPGubSKRvPbaa0hJSYFcLseMGTMQGRmJESNGSB2rw2rqbly7dg3nz59HQUGB1JHatfr6egwYMAApKSkYOHAggNt/hkOHDpU4WcfDc6/leN0jFpJEIjE2Nsbu3bt516KO7tfd2LNnj9TR2jUTExPU1NRIHaND47nXerzuEb/aJqJ24+7uho+PD7sbrRAdHY3Tp09j8+bN6NKFfYLW4LlHpB1eaYh0EBsbi7feegvm5uaIjY194LELFizQU6qOi90N3fz73//GoUOHkJaWBnd3d1haWmrs37t3r0TJ2j+eey3H6x7diR1JIh24uLjg999/xyOPPAIXF5f7HieTyXD27Fk9JiNDFBAQ8MD9cXFxekpCnRmve3QnFpJEJCl2N0gqPPeIdMdCkkgky5YtQ3h4OORyucb2W7du4ZNPPsGHH34oUbL2jd0N8YwdOxZ79+5Ft27dNLbfuHED06dPxy+//CJNsHaK557ueN0jFpJEIjE2NkZ5ebn6EXVNrl69Cjs7Oz4pg9qckZERLl++3OwcvHLlChwdHVFfXy9RMuqseN0jPtmGSCSCIEAmkzXbnpeXB1tbWwkSdTzLli2DUqlstv3WrVtYtmyZBIk6hvz8fOTn5wMATp48qX6dn5+PnJwcbNmyBY6OjhKnbN947mmH1z1iR5JIR927d4dMJkNVVRVsbGw0LqqNjY2orq7GnDlzsH79eglTdgzsbmjHyMhIfd7d65JuYWGBtWvXYvbs2fqO1mHw3GsdXveoCcf/EOnoiy++gCAImD17NpYuXYquXbuq95mamqJv376cR9dC7G5op6SkBIIgQKFQ4LfffkOPHj3U+0xNTWFnZ8eRNg/Bc691eN2jJiwkiXTk5+cH4PbC/WeeeQYmJiYSJ+p4mrobMpkMjz/++H27G3Rvzs7OAACVSiVxko6H5552eN2jJvxqm6gN1NTUoK6uTmObjY2NRGnav4SEBHV344svvmB3Q0tbt2594H5fX189Jek4eO613o0bN9TXsxs3bjzwWF73Oj8WkkQiUSqVeO+997B7925cvXq12X6usXq4I0eOsLuhg+7du2u8rq+vh1KphKmpKeRyOSorKyVK1v7x3Gu5O9eT3rk+905NSwV43ev8+NU2kUgWLlyI9PR0bNiwAbNmzcL69etx6dIlbNq0CStXrpQ6Xofg5eWl/jm7uq137dq1ZtvOnDmDuXPnYuHChRIkat/u7Kx5eHjg1q1buHXr1j2P5bn3P7/88ot63Wh6errEaUhq7EgSiaRPnz7YunUrnnvuOdjY2CA7Oxuurq7Ytm0bdu7cidTUVKkjtnvs6raN33//HTNnzsSpU6ekjtKusLNGpDt2JIlEUllZCYVCAeB296Lpa8RRo0Zh7ty5UkbrMNjVbRtdunRBWVmZ1DHaHXbWdPfjjz/CysoKo0aNAgCsX78eX331Fdzc3LB+/fpmyy2o82EhSSQShUKBkpIS9OnTBwMGDMDu3bsxfPhwJCcnN3tkHd1bcnKyuqsbEBCAZ599Fq6urnB2dsb27dvh4+MjdcR2bf/+/RqvBUFAeXk51q1bh5EjR0qUqv26cynFnT+nllu4cCFWrVoFADhx4gTCwsLw7rvvIj09HWFhYYiLi5M4IbU1FpJEIgkICEBeXh68vLwQERGBKVOmYN26daivr8dnn30mdbwOgV1d3UyfPl3jtUwmQ48ePTB27FisWbNGmlAdBDtr2ikpKYGbmxsAIDExEVOmTEF0dDSys7Ph7e0tcTrSBz4ikUgkoaGhWLBgAQBg/PjxOHXqFHbs2IGcnByEhIRInK5jaOrqAlB3dQGwq9tCKpVK40djYyMuX76MHTt2oFevXlLHa9cWLlyoHmXT1Fnz9vZGSUkJwsLCJE7XfpmamqofLfnzzz/jhRdeAADY2to+dDQQdQ7sSBK1EWdnZ/WgaGoZdnXFUVdXh5KSEvTr1w9duvAy3xLsrGln1KhRCAsLw8iRI/Hbb79h165dAIDTp0+jd+/eEqcjfeAVhkgksbGx99wuk8lgbm4OV1dXjB49mo+qe4DQ0FD1z5u6un/88QdcXV0xePBgCZN1DEqlEvPnz1cPJj99+jQUCgWCg4Ph6OiIiIgIiRO2X3d31pqGt7Oz9mDr1q3DvHnzsGfPHmzYsAGOjo4AgB9++AETJ06UOB3pA8f/EInExcUFFRUVUCqV6vVU165dg1wuh5WVFa5cuQKFQoH09HQ4OTlJnJY6o5CQEGRmZuKLL77AxIkTkZ+fD4VCge+//x4fffQRcnJypI7Ybk2dOhV1dXUYOXIkoqKiUFJSAkdHR6SlpWH+/Pk4ffq01BGJ2iUWkkQi2blzJ/75z39i8+bN6NevHwCgqKgIb7/9Nt566y2MHDkSr732Guzt7bFnzx6J07ZP7OrqxtnZGbt27cLf/vY3WFtbIy8vDwqFAkVFRRg6dCg7aw9QWlqKefPm4cKFC1iwYAECAwMB3O6SNzY23vfcpNvzXZOSklBQUAAAeOKJJzB16lT+PTUQLCSJRNKvXz8kJibiySef1Niek5ODl19+GWfPnkVWVhZefvlllJeXSxOynWNXVzdyuRz/+c9/oFAoNArJvLw8jB49GlVVVVJHpE6mqKgI3t7euHTpEvr37w8AKCwshJOTEw4cOKD+TzV1Xrxrm0gk5eXlaGhoaLa9oaEBly9fBgA4ODjgv//9r76jdRjR0dF46qmncObMGVy9ehVXr17F6dOn8fTTTyMmJgalpaWwt7fXWEtJ/zNs2DAcOHBA/brpSS2bN2/GiBEjpIrVYTQ2NiIxMREff/wxPv74Y+zbt49PtHmIBQsWoF+/frhw4QKys7ORnZ2N0tJSuLi4qKdYUCcnEJEovL29haFDhwrZ2dnqbdnZ2YKnp6cwefJkQRAEYf/+/cKgQYOkitjuKRQKIScnp9n27OxswcXFRRAEQcjMzBTs7e31nKxjyMjIEKysrIQ5c+YI5ubmQkhIiPD8888LlpaWwu+//y51vHbtzJkzwmOPPSbI5XLBw8ND8PDwEORyudC/f3+hqKhI6njtllwuF/Lz85ttz83NFSwtLSVIRPrGjiSRSLZs2QJbW1t4enrCzMwMZmZmGDZsGGxtbbFlyxYAgJWVFQdDPwC7uroZNWoUcnNz0dDQAHd3d6SlpcHOzg7Hjh2Dp6en1PHaNXbWtGNmZnbPv4/V1dUwNTWVIBHpG9dIEons1KlT6js8+/fvr143RA83efJkXL58GZs3b4aHhweA22tMg4KCYG9vj5SUFCQnJ+P999/HiRMnJE5LnYmlpSV+/fVXuLu7a2zPy8vDyJEjUV1dLVGy9s3X1xfZ2dnYsmULhg8fDgA4fvw4goKC4Onpifj4eGkDUpvjHEkikSkUCshkMg6D1sKWLVswa9YseHp6wsTEBMDtbuS4cePY1X0AIyMj9XrI+5HJZPfs9tJt7KxpJzY2Fn5+fhgxYoTG39mpU6ciJiZG4nSkD+xIEolEqVQiODgYCQkJADgMWhfs6rbO999/f999x44dQ2xsLFQqFWpqavSYqmNhZ003RUVF6vE/AwcOhKurq8SJSF9YSBKJhMOgxcNH/OmusLAQERERSE5Oho+PD5YtW8ZHdj7A9evX4e/vj+TkZPU519RZi4+PR9euXSVO2L6oVCp88skn2L9/P+rq6jBu3DgsWbIEFhYWUkcjPePNNkQiSUpKwrp16zBq1CiNrxmfeOIJFBcXS5is41AqlQgMDIRcLscTTzyB0tJSAEBwcDBWrlwpcbqOoaysDEFBQXB3d0dDQwNyc3ORkJDAIvI+VCoVVq1ahcmTJ+PSpUuYPn06vvvuO+zZsweFhYXYt28fi8h7WL58Od5//31YWVnB0dERMTExeOedd6SORRJgIUkkkoqKCtjZ2TXbfvPmzYeuX6PbFi9ejLy8PBw+fBjm5ubq7ePHj8euXbskTNb+VVVVYdGiRXB1dcWff/6JQ4cOITk5GYMGDZI6Wrt2d0GUmpqKpKQkTJkyhV/PPsDWrVvx5Zdf4uDBg0hKSkJycjK2b98OlUoldTTSMxaSRCLhMGjdsaurndWrV0OhUCAlJQU7d+5EVlYWnn32WaljdQgsiLRTWloKb29v9evx48dDJpOhrKxMwlQkBS4+IhJJdHQ0Jk2ahJMnT6KhoQExMTE4efIksrKycOTIEanjdQjs6monIiICFhYWcHV1RUJCgvqGr7vt3btXz8navwcVRL1795YwWfvW0NCg8a0BAJiYmKC+vl6iRCQVFpJEImkaBr1y5Ur1MOihQ4fi2LFjzWbT0b01dXWDg4MBsKvbUr6+viy0tcSCSDuCIMDf3x9mZmbqbTU1NZgzZw4sLS3V2/ifl86Pd20TUbtx9OhRTJo0CTNnzkR8fDzefvttja4un85CYjMyMsKkSZM0CqLk5GSMHTuWBdEDBAQEtOi4uLi4Nk5CUmMhSUTtSnFxMVauXIm8vDxUV1dj6NChWLRoEbu61CZYEBHphoUkkY74VBEiIjJUXCNJpKN9+/bdd9+dTxWh+2MxTkTUMbEjSdQG+FSR1uEj/oiIOiZ2JIlEVFZWhiVLliAhIQETJkxAbm4uB0K3wLRp05ptu1cxTkRE7QsHkhOJgE8VEQ8f8UfUcdy8eVPqCCQxFpJEOuJTRcTBYpyo4+nZsydmz56No0ePSh2FJMI1kkQ6MjIygoWFBcaPHw9jY+P7Hsc5dPe3evVqrFq1Cvb29oiOjr7nV91E1P4kJSUhPj4eqamp6Nu3L2bPng1fX184ODhIHY30hIUkkY78/f1b9FQRzqG7PxbjRB1bRUUFtm3bhvj4eBQUFGDChAmYPXs2pk6dii5deDtGZ8ZCkogkx2KcqPNYu3YtFi5ciLq6Ojz66KOYM2cOIiIiIJfLpY5GbYCFJBEREenkr7/+QkJCAuLj43H+/Hm8+OKLCAwMxMWLF7Fq1So4ODggLS1N6pjUBlhIEhERkVb27t2LuLg4HDx4EG5ubnjzzTcxc+ZMdOvWTX1McXExBg4ciLq6OumCUpvhwgUiIiLSSkBAAF577TVkZmbiqaeeuucxDg4O+OCDD/ScjPSFHUkiIiLSilKp5NpHA8dCkkgkN2/ehKWlpdQxiIja1I0bN1p8rI2NTRsmofaAhSSRSKysrDBjxgzMnj0bo0aNkjoOEVGbMDIyeuiUBUEQIJPJ0NjYqKdUJBWukSQSyTfffIP4+HiMHTuWg3mJqNNKT0+XOgK1I+xIEomMg3mJiMhQsJAkakMczEtEnV1GRgY2bdqEs2fP4rvvvoOjoyO2bdsGFxcXLvMxAEZSByDqbP766y+sXr0abm5uiIiIwCuvvIJDhw5hzZo12Lt3L6ZPny51RCIiUSQmJmLChAmwsLBAdnY2amtrAQBVVVWIjo6WOB3pAzuSRCLhYF4iMjQeHh4IDQ2Fr68vrK2tkZeXB4VCgZycHEyaNAmXL1+WOiK1MS7YIhIJB/MSkaEpLCzE6NGjm23v2rUrrl+/rv9ApHcsJIlEUl5e/tC1jxYWFliyZImeEhERtS17e3sUFRWhb9++GtuPHj0KhUIhTSjSKxaSRDq4ezDvgwb1cjAvEXU2QUFBCAkJwddffw2ZTIaysjIcO3YM4eHhiIyMlDoe6QHXSBLpgIN5iciQCYKA6OhorFixAkqlEgBgZmaG8PBwREVFSZyO9IGFJJEOjhw50uJjvby82jAJEZF06urqUFRUhOrqari5ucHKykrqSKQnLCSJiIiISCtcI0kkIg7mJSJDUlNTg7Vr1yI9PR1XrlyBSqXS2J+dnS1RMtIXFpJEIklMTMSsWbPg4+Nzz8G8qampEickIhJXYGAg0tLS8Morr2D48OEPXTNOnQ+/2iYSCQfzEpGh6dq1K1JTUzFy5Eipo5BE+IhEIpFwMC8RGRpHR0dYW1tLHYMkxEKSSCRNg3nvxsG8RNRZrVmzBosWLcL58+eljkIS4RpJIpFwMC8RGZphw4ahpqYGCoUCcrkcJiYmGvsrKyslSkb6wkKSSCQRERFQqVQYN24clEolRo8erR7MGxwcLHU8IiLRvf7667h06RKio6PRs2dP3mxjgHizDZHIOJiXiAyFXC7HsWPHMGTIEKmjkETYkSQSmampKdzc3KSOQUTU5gYMGIBbt25JHYMkxI4kkUg4mJeIDE1aWhqWLl2K5cuXw93dvdkaSRsbG4mSkb6wkCQSiY+Pj3ow773WCi1ZskSiZEREbcPI6Pbwl7uvd4IgQCaTobGxUYpYpEf8aptIJCkpKRzMS0QGJT09XeoIJDEWkkQi4WBeIjI0Xl5eUkcgiXEgOZFIOJiXiAxRRkYGZs6ciWeeeQaXLl0CAGzbtg1Hjx6VOBnpAwtJIpHcOZjX2toatra2Gj+IiDqbxMRETJgwARYWFsjOzkZtbS0AoKqqCtHR0RKnI33gzTZEIhk/fjxKS0sRGBh4z5tt/Pz8JEpGRNQ2PDw8EBoaCl9fX1hbWyMvLw8KhQI5OTmYNGkSLl++LHVEamNcI0kkkqysLA7mJSKDUlhYiNGjRzfb3rVrV1y/fl3/gUjv+NU2kUg4mJeIDI29vT2KioqabT969CgUCoUEiUjfWEgSiWTlypV49913cfjwYVy9ehU3btzQ+EFE1NkEBQUhJCQEx48fh0wmQ1lZGbZv347w8HDMnTtX6nikB1wjSSQSDuYlIkMjCAKio6OxYsUKKJVKAICZmRnCw8MRFRUlcTrSBxaSRCI5cuTIA/dz3hoRdSaNjY3IzMzE4MGDIZfLUVRUhOrqari5ucHKykrqeKQnLCSJiIhIK+bm5igoKICLi4vUUUgiXCNJJCIO5iUiQzJo0CCcPXtW6hgkIRaSRCLhYF4iMjQff/wxwsPDkZKSgvLyct5kaID41TaRSDiYl4gMTdNNhoDmjYa8ydBwcCA5kUg4mJeIDE16errUEUhiLCSJRNI0mLdv374a2zmYl4g6KxcXFzg5Od1z7NmFCxckSkX6xDWSRCLhYF4iMjQuLi6oqKhotr2yspJ3chsIdiSJRBIREQGVSoVx48ZBqVRi9OjR6sG8wcHBUscjIhJd01rIu1VXV8Pc3FyCRKRvvNmGSAQczEtEhiQsLAwAEBMTg6CgIMjlcvW+xsZGHD9+HMbGxsjMzJQqIukJO5JEIjA2NsYLL7yAgoICdOvWDW5ublJHIiJqMzk5OQBudyRPnDgBU1NT9T5TU1MMGTIE4eHhUsUjPWIhSSSSpsG8XBdERJ1d093aAQEBiImJgY2NjcSJSCr8aptIJD/++CMWL16MqKgoeHp6wtLSUmM/L7RE1NlUVFSgR48e99x34sQJuLu76zkR6RsLSSKRcDAvERkae3t7bNmyBZMnT9bY/umnnyIyMhK3bt2SKBnpC7/aJhIJB/MSkaEJCwvDyy+/jICAAHz22WeorKyEr68vTpw4gR07dkgdj/SAHUkikZSWlj5wMG+fPn0kSkZE1HZycnIwa9Ys1NbWorKyEk8//TS+/vpr2NvbSx2N9IADyYlEwsG8RGSIXF1dMWjQIJw7dw43btzAq6++yiLSgLCQJBIJB/MSkaFpmp975swZ5OfnY8OGDQgODsarr76Ka9euSR2P9IBfbRPpiIN5ichQmZmZITQ0FFFRUTAxMQEAFBcXY+bMmbhw4QIuXrwocUJqa7zZhkhHHMxLRIYqLS0NXl5eGtv69euHzMxMLF++XKJUpE/sSBKJhIN5iYjI0HCNJJFIVq9efd8i8sSJE3pOQ0TUdry9vVFVVaV+vXLlSly/fl39+urVq3xUrIFgIUkkEnd3dxw4cKDZ9k8//RTDhw+XIBERUds4ePAgamtr1a+jo6NRWVmpft3Q0IDCwkIpopGesZAkEknTYN65c+fi1q1buHTpEsaNG4fVq1dzMC8RdSp3r4rjKjnDxUKSSCTvvfcejh07hoyMDAwePBiDBw+GmZkZ8vPz8eKLL0odj4iISHQsJIlExMG8RGQIZDJZs7m595qjS50fx/8QiSQzMxMzZ86Era0t8vPzkZmZieDgYKSmpmLjxo3o3r271BGJiEQhCAL8/f1hZmYGAKipqcGcOXNgaWkJABrrJ6lz4/gfIpFwMC8RGYqAgIAWHRcXF9fGSUhqLCSJRHLkyJFmg3kBQKVSYfny5YiMjJQgFRERUdthIUlEREREWuHNNkQ64mBeIiIyVOxIEunI2NgY5eXlsLOzAwDY2NggNzcXCoUCAPDXX3/BwcEBjY2NUsYkIiISHTuSRDriYF4iIjJULCSJiIiISCssJIl0xMG8RERkqDiQnEhHHMxLRESGijfbEOmIg3mJiMhQsZAkIiIiIq1wjSQRERERaYWFJBERERFphYUkEREREWmFhSQRERERaYWFJBGRBPz9/TF9+nT16+eeew7/+Mc/9J7j8OHDkMlkGs+Hv5tMJkNSUlKL3/Ojjz7Ck08+qVOuc+fOQSaTITc3V6f3IaK2xUKSiOj/+fv7qwfMm5qawtXVFcuWLUNDQ0Obf/bevXsRFRXVomNbUvwREekDB5ITEd1h4sSJiIuLQ21tLVJTU/HOO+/AxMQEixcvbnZsXV0dTE1NRflcW1tbUd6HiEif2JEkIrqDmZkZ7O3t4ezsjLlz52L8+PHYv38/gP99Hb18+XI4ODigf//+AIALFy5gxowZ6NatG2xtbTFt2jScO3dO/Z6NjY0ICwtDt27d8Mgjj+C9997D3SN87/5qu7a2FosWLYKTkxPMzMzg6uqKLVu24Ny5cxgzZgwAoHv37pDJZPD39wcAqFQqrFixAi4uLrCwsMCQIUOwZ88ejc9JTU3F448/DgsLC4wZM0YjZ0stWrQIjz/+OORyORQKBSIjI1FfX9/suE2bNsHJyQlyuRwzZsxAVVWVxv7Nmzdj4MCBMDc3x4ABA/Dll1+2OgsRSYuFJBHRA1hYWKCurk79+tChQygsLMRPP/2ElJQU1NfXY8KECbC2tkZGRgYyMzNhZWWFiRMnqn/dmjVrEB8fj6+//hpHjx5FZWUl9u3b98DP9fX1xc6dOxEbG4uCggJs2rQJVlZWcHJyQmJiIgCgsLAQ5eXliImJAQCsWLECW7duxcaNG/Hnn38iNDQUM2fOxJEjRwDcLnhfeuklTJkyBbm5uXjzzTcRERHR6j8Ta2trxMfH4+TJk4iJicFXX32Fzz//XOOYoqIi7N69G8nJyfjxxx+Rk5ODefPmqfdv374dH374IZYvX46CggJER0cjMjISCQkJrc5DRBISiIhIEARB8PPzE6ZNmyYIgiCoVCrhp59+EszMzITw8HD1/p49ewq1tbXqX7Nt2zahf//+gkqlUm+rra0VLCwshIMHDwqCIAi9evUSVq9erd5fX18v9O7dW/1ZgiAIXl5eQkhIiCAIglBYWCgAEH766ad75kxPTxcACNeuXVNvq6mpEeRyuZCVlaVxbGBgoPD6668LgiAIixcvFtzc3DT2L1q0qNl73Q2AsG/fvvvu/+STTwRPT0/16yVLlgjGxsbCxYsX1dt++OEHwcjISCgvLxcEQRD69esn7NixQ+N9oqKihBEjRgiCIAglJSUCACEnJ+e+n0tE0uMaSSKiO6SkpMDKygr19fVQqVR444038NFHH6n3u7u7a6yLzMvLQ1FREaytrTXep6amBsXFxaiqqkJ5eTmefvpp9b4uXbpg2LBhzb7ebpKbmwtjY2N4eXm1OHdRURGUSiWef/55je11dXXw8PAAABQUFGjkAIARI0a0+DOa7Nq1C7GxsSguLkZ1dTUaGhpgY2OjcUyfPn3g6Oio8TkqlQqFhYWwtrZGcXExAgMDERQUpD6moaEBXbt2bXUeIpIOC0kiojuMGTMGGzZsgKmpKRwcHNCli+Zl0tLSUuN1dXU1PD09sX379mbv1aNHD60yWFhYtPrXVFdXAwAOHDigUcABt9d9iuXYsWPw8fHB0qVLMWHCBHTt2hXffvst1qxZ0+qsX331VbPC1tjYWLSsRNT2WEgSEd3B0tISrq6uLT5+6NCh2LVrF+zs7Jp15Zr06tULx48fx+jRowHc7rz98ccfGDp06D2Pd3d3h0qlwpEjRzB+/Phm+5s6oo2Njeptbm5uMDMzQ2lp6X07mQMHDlTfONTk119/ffhv8g5ZWVlwdnbGBx98oN52/vz5ZseVlpairKwMDg4O6s8xMjJC//790bNnTzg4OODs2bPw8fFp1ecTUfvCm22IiHTg4+ODRx99FNOmTUNGRgZKSkpw+PBhLFiwABcvXgQAhISEYOXKlUhKSsKpU6cwb968B86A7Nu3L/z8/DB79mwkJSWp33P37t0AAGdnZ8hkMqSkpKCiogLV1dWwtrZGeHg4QkNDkZCQgOLiYmRnZ2Pt2rXqG1jmzJmDM2fOYOHChSgsLMSOHTsQHx/fqt/vY489htLSUnz77bcoLi5GbGzsPW8cMjc3h5+fH/Ly8pCRkYEFCxZgxowZsLe3BwAsXboUK1asQGxsLE6fPo0TJ04gLi4On332WavyEJG0WEgSEelALpfjX//6F/r06YOXXnoJAwcORGBgIGpqatQdynfffRezZs2Cn58fRowYAWtra7z44osPfN8NGzbglVdewbx58zBgwAAEBQXh5s2bAABHR0csXboUERER6NmzJ+bPnw8AiIqKQmRkJFasWIGBAwdi4sSJOHDgAFxcXADcXreYmJiIpKQkDBkyBBs3bkR0dHSrfr9Tp05FaGgo5s+fjyeffBJZWVmIjIxsdpyrqyteeukleHt744UXXsDgwYM1xvu8+eab2Lx5M+Li4uDu7g4vLy/Ex8ersxJRxyAT7rfam4iIiIjoAdiRJCIiIiKtsJAkIiIiIq2wkCQiIiIirbCQJCIiIiKtsJAkIiIiIq2wkCQiIiIirbCQJCIiIiKtsJAkIiIiIq2wkCQiIiIirbCQJCIiIiKtsJAkIiIiIq2wkCQiIiIirfwfef0cw8ehJgEAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "import seaborn as sns\n",
    "conf_mat = confusion_matrix([revmap[x] for x in true], [revmap[x] for x in pred])\n",
    "conf_mat_normalized = conf_mat.astype('float') / conf_mat.sum(axis=1)[:, np.newaxis]\n",
    "sns.heatmap(conf_mat_normalized, annot=True, linewidth=.5, cmap=\"crest\", \n",
    "            xticklabels=revmap.values(), yticklabels=revmap.values())\n",
    "plt.ylabel('True label')\n",
    "plt.xlabel('Predicted label')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.10.9 64-bit",
   "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": "3.11.6"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "3e5df068846dd857fd8f6e2337ec975361106f93a8ad476be75c82b4dc19b61a"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}