diff --git a/best_blur_model.pth b/best_blur_model.pth
new file mode 100644
index 0000000..90314c8
Binary files /dev/null and b/best_blur_model.pth differ
diff --git a/new_nn_train.ipynb b/new_nn_train.ipynb
index 6a87cdb..9a8b1bf 100644
--- a/new_nn_train.ipynb
+++ b/new_nn_train.ipynb
@@ -1,18 +1,13 @@
 {
  "cells": [
   {
-   "cell_type": "code",
-   "id": "3cb1cc2f-2059-492d-8218-ccc66e976d92",
    "metadata": {
-    "collapsed": true,
-    "jupyter": {
-     "outputs_hidden": true
-    },
     "ExecuteTime": {
-     "end_time": "2025-05-16T05:48:49.033600Z",
-     "start_time": "2025-05-16T05:48:49.017234Z"
+     "end_time": "2025-05-18T20:33:21.545777Z",
+     "start_time": "2025-05-18T20:33:21.533147Z"
     }
    },
+   "cell_type": "code",
    "source": [
     "from data_gen import generate_random_shapes_image, blur_image\n",
     "import torch\n",
@@ -38,10 +33,13 @@
     "print(device)\n",
     "\n",
     "transform = transforms.Compose([\n",
-    "            transforms.ToTensor(),\n",
-    "            transforms.Normalize(mean=[0.5], std=[0.5])\n",
-    "        ])\n",
-    "sigma_range=(0, 10)\n",
+    "    transforms.Resize((128, 128)),\n",
+    "    transforms.ToTensor(),\n",
+    "    transforms.Normalize(mean=[0.5], std=[0.5])\n",
+    "])\n",
+    "\n",
+    "sigma_range=(0,\n",
+    "             30)\n",
     "sigma_mean = (sigma_range[1] - sigma_range[0]) / 2\n",
     "sigma_std = (sigma_range[1] - sigma_range[0]) / 2\n",
     "# Custom Dataset\n",
@@ -213,6 +211,7 @@
     "    print(\"Training completed!\")\n",
     "\n"
    ],
+   "id": "c52e12ffcfdcb63a",
    "outputs": [
     {
      "name": "stdout",
@@ -222,1741 +221,267 @@
      ]
     }
    ],
-   "execution_count": 15
+   "execution_count": 13
   },
   {
+   "metadata": {},
    "cell_type": "code",
-   "id": "16f2abed2c2f1da0",
+   "outputs": [],
+   "execution_count": null,
+   "source": "main()",
+   "id": "51351ec23357b507"
+  },
+  {
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2025-05-16T06:21:28.196339Z",
-     "start_time": "2025-05-16T05:48:52.961022Z"
+     "end_time": "2025-05-18T20:33:25.673483Z",
+     "start_time": "2025-05-18T20:33:25.667838Z"
     }
    },
-   "source": "main()",
+   "cell_type": "code",
+   "source": [
+    "def evaluate_model(model, test_loader, criterion):\n",
+    "    model.eval()\n",
+    "    test_loss = 0\n",
+    "    predictions = []\n",
+    "    ground_truths = []\n",
+    "\n",
+    "    with torch.no_grad():\n",
+    "        for images, sigmas in tqdm(test_loader, desc=\"Evaluating\"):\n",
+    "            images, sigmas = images.to(device), sigmas.to(device)\n",
+    "            outputs = model(images)\n",
+    "            loss = criterion(outputs, sigmas)\n",
+    "            test_loss += loss.item() * images.size(0)\n",
+    "\n",
+    "            # The normalization in BlurDataset is: sigma_normalized = (sigma - sigma_min) / (sigma_max - sigma_min)\n",
+    "            # So, to denormalize: sigma = sigma_normalized * (sigma_max - sigma_min) + sigma_min\n",
+    "            outputs_denorm = outputs.cpu().numpy() * sigma_std + sigma_mean\n",
+    "            sigmas_denorm = sigmas.cpu().numpy() * sigma_std + sigma_mean\n",
+    "\n",
+    "            predictions.extend(outputs_denorm)\n",
+    "            ground_truths.extend(sigmas_denorm)\n",
+    "\n",
+    "    test_loss /= len(test_loader.dataset)\n",
+    "    predictions = np.array(predictions)\n",
+    "    ground_truths = np.array(ground_truths)\n",
+    "\n",
+    "    return test_loss, predictions, ground_truths\n",
+    "\n",
+    "\n",
+    "def plot_results(test_loss, predictions, ground_truths, save_dir=\"plots\"):\n",
+    "    # Créer le répertoire pour sauvegarder les plots\n",
+    "    Path(save_dir).mkdir(exist_ok=True)\n",
+    "\n",
+    "    # 1. Scatter plot : Prédictions vs Valeurs réelles\n",
+    "    plt.figure(figsize=(10, 6))\n",
+    "    plt.scatter(ground_truths, predictions, alpha=0.5, s=10)\n",
+    "    plt.plot([0, 10], [0, 10], 'r--', label='Ligne idéale')\n",
+    "    plt.xlabel('Sigma réel')\n",
+    "    plt.ylabel('Sigma prédit')\n",
+    "    plt.title(f'Prédictions vs Valeurs réelles (Test Loss: {test_loss:.4f})')\n",
+    "    plt.legend()\n",
+    "    plt.grid(True)\n",
+    "    plt.show()\n",
+    "\n",
+    "\n",
+    "    # 2. Histogramme des erreurs\n",
+    "    errors = predictions - ground_truths\n",
+    "    plt.figure(figsize=(10, 6))\n",
+    "    sns.histplot(errors, bins=50, kde=True)\n",
+    "    plt.xlabel('Erreur (Prédit - Réel)')\n",
+    "    plt.ylabel('Fréquence')\n",
+    "    plt.title('Distribution des erreurs de prédiction')\n",
+    "    plt.grid(True)\n",
+    "    plt.show()\n",
+    "\n",
+    "\n",
+    "    # 3. Erreur absolue en fonction de sigma réel\n",
+    "    abs_errors = np.abs(errors)\n",
+    "    plt.figure(figsize=(10, 6))\n",
+    "    plt.scatter(ground_truths, abs_errors, alpha=0.5, s=10)\n",
+    "    plt.xlabel('Sigma réel')\n",
+    "    plt.ylabel('Erreur absolue')\n",
+    "    plt.title('Erreur absolue en fonction de Sigma réel')\n",
+    "    plt.grid(True)\n",
+    "    plt.show()\n",
+    "\n",
+    "    # 4. Boxplot des erreurs par plage de sigma\n",
+    "    sigma_bins = np.linspace(0, 10, 11)\n",
+    "    digitized = np.digitize(ground_truths, sigma_bins)\n",
+    "    binned_errors = [abs_errors[digitized == i] for i in range(1, len(sigma_bins))]\n",
+    "\n",
+    "    plt.figure(figsize=(10, 6))\n",
+    "    plt.boxplot(binned_errors, labels=[f'{sigma_bins[i]:.1f}-{sigma_bins[i+1]:.1f}'\n",
+    "                                     for i in range(len(sigma_bins)-1)])\n",
+    "    plt.xlabel('Plage de Sigma réel')\n",
+    "    plt.ylabel('Erreur absolue')\n",
+    "    plt.title('Distribution des erreurs par plage de Sigma')\n",
+    "    plt.xticks(rotation=45)\n",
+    "    plt.grid(True)\n",
+    "    plt.tight_layout()\n",
+    "    plt.show()"
+   ],
+   "id": "9f71e6d008face0f",
+   "outputs": [],
+   "execution_count": 14
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2025-05-18T20:33:29.283554Z",
+     "start_time": "2025-05-18T20:33:26.906718Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "# Hyperparamètres\n",
+    "batch_size = 128\n",
+    "sigma_range = (0, 10)\n",
+    "test_size = 2000\n",
+    "\n",
+    "# Charger le modèle entraîné\n",
+    "# model = BlurRegressionCNN().to(device)\n",
+    "model = BlurRegressionCNN().to(device)\n",
+    "model.load_state_dict(torch.load(\"best_blur_model.pth\"))\n",
+    "# model_path = \"best_blur_model.pth\"\n",
+    "# if not Path(model_path).exists():\n",
+    "#     raise FileNotFoundError(f\"Modèle non trouvé à {model_path}\")\n",
+    "# model.load_state_dict(torch.load(model_path))\n",
+    "# print(f\"Modèle chargé depuis {model_path}\")\n",
+    "\n",
+    "# Créer le dataset de test\n",
+    "test_dataset = BlurDataset(length=test_size)\n",
+    "test_loader = DataLoader(\n",
+    "    test_dataset,\n",
+    "    batch_size=batch_size,\n",
+    "    shuffle=False,\n",
+    "    num_workers=0,\n",
+    "    pin_memory=True,\n",
+    ")\n",
+    "\n",
+    "# Critère de perte\n",
+    "criterion = nn.MSELoss()\n",
+    "\n",
+    "# Évaluer le modèle\n",
+    "test_loss, predictions, ground_truths = evaluate_model(model, test_loader, criterion)\n",
+    "\n",
+    "# Afficher les résultats\n",
+    "print(f\"Test Loss: {test_loss:.4f}\")\n",
+    "print(f\"MAE: {np.mean(np.abs(predictions - ground_truths)):.4f}\")\n",
+    "print(f\"RMSE: {np.sqrt(np.mean((predictions - ground_truths)**2)):.4f}\")\n",
+    "\n",
+    "# Visualiser les résultats\n",
+    "plot_results(test_loss, predictions, ground_truths)\n",
+    "\n"
+   ],
+   "id": "73a7e2a2700bf97e",
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Epoch 1/100: 100%|██████████| 80/80 [00:15<00:00,  5.05it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 1/100, Train Loss: 0.2484, Val Loss: 0.1575\n",
-      "Saved best model with Val Loss: 0.1575\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 2/100: 100%|██████████| 80/80 [00:15<00:00,  5.23it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 2/100, Train Loss: 0.1384, Val Loss: 0.1116\n",
-      "Saved best model with Val Loss: 0.1116\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 3/100: 100%|██████████| 80/80 [00:14<00:00,  5.46it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 3/100, Train Loss: 0.1109, Val Loss: 0.1029\n",
-      "Saved best model with Val Loss: 0.1029\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 4/100: 100%|██████████| 80/80 [00:15<00:00,  5.05it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 4/100, Train Loss: 0.0954, Val Loss: 0.1138\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 5/100: 100%|██████████| 80/80 [00:16<00:00,  4.95it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 5/100, Train Loss: 0.0853, Val Loss: 0.0755\n",
-      "Saved best model with Val Loss: 0.0755\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 6/100: 100%|██████████| 80/80 [00:15<00:00,  5.14it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 6/100, Train Loss: 0.0821, Val Loss: 0.0793\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 7/100: 100%|██████████| 80/80 [00:15<00:00,  5.23it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 7/100, Train Loss: 0.0830, Val Loss: 0.0755\n",
-      "Saved best model with Val Loss: 0.0755\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 8/100: 100%|██████████| 80/80 [00:15<00:00,  5.01it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 8/100, Train Loss: 0.0685, Val Loss: 0.0781\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 9/100: 100%|██████████| 80/80 [00:15<00:00,  5.14it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 9/100, Train Loss: 0.0647, Val Loss: 0.0648\n",
-      "Saved best model with Val Loss: 0.0648\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 10/100: 100%|██████████| 80/80 [00:16<00:00,  4.86it/s]\n"
+      "Evaluating: 100%|██████████| 16/16 [00:01<00:00,  8.01it/s]\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Epoch 10/100, Train Loss: 0.0658, Val Loss: 0.0562\n",
-      "Saved best model with Val Loss: 0.0562\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 11/100: 100%|██████████| 80/80 [00:14<00:00,  5.36it/s]\n"
+      "Test Loss: 0.0250\n",
+      "MAE: 1.3159\n",
+      "RMSE: 2.3728\n"
      ]
     },
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 11/100, Train Loss: 0.0640, Val Loss: 0.0815\n"
-     ]
+     "data": {
+      "text/plain": [
+       "<Figure size 1000x600 with 1 Axes>"
+      ],
+      "image/png": "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"
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 12/100: 100%|██████████| 80/80 [00:15<00:00,  5.09it/s]\n"
-     ]
+     "data": {
+      "text/plain": [
+       "<Figure size 1000x600 with 1 Axes>"
+      ],
+      "image/png": "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"
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 12/100, Train Loss: 0.0587, Val Loss: 0.0810\n"
-     ]
+     "data": {
+      "text/plain": [
+       "<Figure size 1000x600 with 1 Axes>"
+      ],
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1MAAAIhCAYAAACrJeDiAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjEsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvc2/+5QAAAAlwSFlzAAAPYQAAD2EBqD+naQABAABJREFUeJzsnQd4pFd5799R16rsrqTVNpfdde82NtgUh2oDxoQawIQWLu0SSEIJhJLkhnKdQBIuLYBDj4FQA5gSsIGAwbjgXne93t5XW7Tqde7zez8d+dNoZjQzKlP0/+2jRzujKec73ylvP4lkMpk0IYQQQgghhBB5UZXfy4UQQgghhBBCgJQpIYQQQgghhCgAKVNCCCGEEEIIUQBSpoQQQgghhBCiAKRMCSGEEEIIIUQBSJkSQgghhBBCiAKQMiWEEEIIIYQQBSBlSgghhBBCCCEKQMqUEEIIMU8kk8liN0EIIcQ8ImVKCFGW/M3f/I2ddtppGX+e+MQnWiXxyU9+0q9rIXja057m/Vvu7Nu3z/70T//UzjnnHHv84x9vAwMDC/rdb3jDG2z37t0l069zOYa4rve973325Cc/2c4++2y75JJL7E1vepPdeuutU17H9XLd5c5f/dVf2UUXXWSPPPJI3u995Stf6T9CiMqkptgNEEKIQlmxYoV96lOfSvu32traBW+PKC2+8pWv2F133WUf/ehHbeXKldbY2Lhg333TTTfZr3/96ynPMVabm5ut3Dl48KC99KUv9T59+9vfbqtXr7bDhw/bt7/9bXv1q19tH//4x+3yyy/31775zW+2V73qVVbObNq0yW644Qb7zGc+YyeddFKxmyOEKDGkTAkhypa6ujo7//zzi90MUaIcPXrUOjs77YorrrBS4Mwzz7RK4Fvf+pYdO3bM/vu//3uKcnjZZZfZn/zJn0xRpk444QQrdxhDP/vZz2zt2rXFbooQogRRmJ8QouIhxOad73yn/cVf/IUrX3/2Z39mu3bt8pCnL33pS/asZz3LzjvvPPvud787aYl+4xvfaI95zGP858///M9t586dk5/3ve99z9/LZ8RJDePiNXgjXvjCF9q5556b0Ys2NjZm11xzjV155ZX+Otr4spe9zG6++eZpr8VC/sxnPtND1xBcf//730/zxnA9/P3SSy+1//N//o/19vZO/n1oaMg+/elPT74GoZfvHh8fT9u2W265xa+D3zOFLuGZeM5znuNhX095ylM8rIxrywbfy/cjiPM+ru0//uM/pn0XIWW8js+l3fTPPffck/FzuRfcpz179nj7aQscOHDA3vOe93h4Gn394he/2H7xi19MeS+v/9rXvubf+bjHPc4uuOAC+8u//Evr6uqa8rrvf//79oIXvMDHDu36l3/5FxseHvbv5Tvg6U9/+uSYSB0fPT09dvXVV9sznvEMvybu/3e+851p1/GJT3zC/umf/sme8IQneJv/1//6X7Zt27as/cp95rMJd6X9tIfnUvnDH/5gr3jFK/wauNZ3v/vd7mXKBv2QSCSm3dvq6mp7xzve4V6rTGF+IyMj9s///M/2R3/0R5PXQj/G5xPv4flvfvOb3je8jvu9detW+9WvfmXPfe5zvb2M/wcffHDaGGS+MYd43/Oe9zz76U9/mvf6EPrwIx/5iH8G45Lv/clPfjLt/YWMeyFE5SDPlBCirBkdHU37PIIdAl8AgeqP//iPPVQnrjgg+CA0Y2FHQENgQ3DbsGGDC7B8Pu+56qqr7Ac/+IG1t7fn1b7PfvazLmCuX78+o2Ub4fIb3/iGvw6hcv/+/a7wIMD/z//8z5TwNNqK0MdnffnLX7bXv/71/l6E8R/96Ece0oZAzOds2bLFr4FcIX5TDIG8FkLf3vKWt9jpp5/uStL/+3//z5XFD37wg1Yon/vc5+xjH/uYC+YI7gi59O3evXvt//7f/5vxfSh7KB8orwj9t912m78ezwdKbADPACFW73//+/06uJ63vvWt9stf/tLvdSoorlzXAw884P9ftWqVKwEoT/X19fa2t73Nli9f7t/N9yA0Mz4CXAsK3r/+679636CY8D08BpStD3zgAy7QE+rGa/iM7u5uz6/53//7f/u44bvT5SkNDg7ay1/+cjt06NDk/URR5v7STu5T4Ktf/apdeOGF3gY+/8Mf/rDfY5SNTPz1X/+13XjjjX6dJ554or/2uuuum/Ia+hrFgXwn+orPxqtEWB5KXUNDQ9rPRmH4+te/bi95yUv8h3w0rpH+QXnLlq/4d3/3dz5OuXdnnHGG//9v//Zvp73uzjvvdMUXxQqlhnFCDhpzmv5iTvz93/+9K0E//vGPJ+/Jhz70If9s+ovr+fd//3d/DWOLMZCJ1PWBMca4uOOOO/z7GHvXX3+99ycK8/Of//xZjXshRAWRFEKIMuTd73538tRTT8348/nPf37yta94xSuS5513XnJoaGjyuZ07d/rr3vve90753Le//e3JJzzhCcmenp7J544cOZK88MILk//4j//oj7/73e/6e/mMOE996lO9XQFe8+pXv3rGa+E7v/zlL0957mc/+5m//8477/THn/jEJ/zxT3/608nXDA4OJp/4xCcm3/rWt/rjv/3bv00+85nPTI6NjU2+5gc/+EHyq1/9qv//f/7nf/wzfvSjH035rk9/+tP+/KZNm6Zdx8033+x/43cc+pQfOHbsWPLcc89N/t3f/d2U13zrW9+a8rmpbNmyJXnaaaclP/e5z015/mMf+1jynHPOSR4+fHjK/Yvfk//6r//yz7733nsz9ivXwLUEPvKRjyTPOuus5K5du6a8jntEP4Z+43OvuuqqKa/5m7/5m+T555/v/+d1j3/845NvfvObp7yGMfeCF7wgOTw8nHaMxPv1a1/7mv/9jjvumPIZjEeunTEX3sPP6Ojo5Gs++clP+ntD/6RCf/P3r3/965PP0eYrrrjCnw+89KUvTV555ZVTPpt7csYZZySvvfbaZDb4+2Me85jJ+cb///zP/zz529/+NuM92L59u9/vL37xi1Ne89rXvnZKX4W5vXnz5snXMLZ47qabbpp87gtf+II/193d7Y+vvvrq5Ec/+tEpn33fffelHfNx0q0PXAfv+/GPfzzlte985zt9rIyMjOQ87uNzRQhReSjMTwhR1gUosKCn+yE0Jw6eJnKsUsE6HofQOsKdsMrjleIHrxWVvCgqkC+pn58OwsNI3Ce8irArwg1/+MMf+t+wgseLaoRcFMDDQrgUHgbAw4BnjTAnPCL33nuvhyaFcDwqrdXU1HiIX5zgkUmtxJYreBHwtBDOFfqMnxDe9bvf/S7t++hrPADp3oc34vbbb5987cknnzwlP4fiB5BPhT6uDw9FqoeQ66eoAp68QGouHl6N8F30MR4lPFdxCE3D05VL8RPaQjtoT2pbuPa777578jm8jnHvW/CwZLp2xhDEw+uqqqo8VC3Ae/kOwh25B6Hvjz/+ePfCZLpnAaok/va3v/Vxxv8pQoHn5rWvfa394z/+Y9r34AXlu1LHH+GNqSxdunRKsYeOjg7/jfc4sGzZMv+NFxPwYuGF4jHeVzzJeKtS51E6UtcHwmfxgtE/qWOTsfLwww8XPO6FEJWFwvyEEGULwg+CZi40NTWlfX7JkiXTihaQF5EuN6KtrS3vNqZ+fjpQev7hH/7BfxO+hOKwZs2aaecUEZaGUByHsMMgTFJogRAlQrD+7d/+zcONENgRMPkbYU98RmpYHEppyOEpBPoMCMNKB+Fa2d5Hvkk6CHcMpFbiC/2QKdcrHVw/ykIqQVAP/Zjp+8K9CO3ON+QztS2h3wtpS7Zr57OBex0n/n18Pu8nDI6fVFDUZ4J2oVAGpXL79u323ve+1/MQUehPPfXUKa8PuVip/ZauHzNVPcw2n3bs2OFhhChCKLQoSISy5nLeV+r6wD3mPeRMZhrToZ/zHfdCiMpCypQQQsRoaWnxRP+QhB4Hrw6EXKxUYbavry/v76M4xOte9zrPOSH3AwEQYZmy2uQJxUHZQcCL54KRXxNX8rDy88Nr8RwgKJM/Qw4J1v4jR454cnxcoQpCX6rwPdO1BgG0tbV1Mvdr3bp1GRWEVML7KJqRTtkNCuVcwfXjVUglPJfu+rO1O7VQA31LjlaqtylTW1A+ZtuWdIT3MjbifRiUQKC/ubevec1r0iqzmcrIM3ZQnsgZIpcoDrlZ5LTxt82bN09TpoI3MbVdMxW8yAXGJ0oNShSeaTzCzFfagYeqkHUAxY18tXRwreRTFTLuhRCVhcL8hBAiBiF+CGAIY3i9+KFKF8UeCGOKW805mDXAYZ5xYTVXCC3jfST945EKXoff/OY305QYQrPiFf5QaChQcfHFF/tjCh+Eog0Ig89+9rP9nB9Cj1CYuDb+T0nrOCGkEIUrlXTXikU+fngpoVcIsXiSQp/xgzBLwYbUqocBQieDEhJ/H8I1hRAK6c9sPPaxj/XQrPhBuuH68dogIOcCCi8KC5Xl4iC0I9BTsS7Vg5iuLbSD9qS2hb6kEl2hEO4Jqfc53l7uK6XaGX/xvj/llFPco5lavTGAEk6pcEJRuW+pEAIJqYpUGF+8P8yjwM9//nObLbSF76bASBh7meZRLjBX+vv73XgR7x8qfVIchnlU6LgXQlQW8kwJIcoW8iDIjcgE3p58D2pF+aCaH9XlqOBHuBOV0Ki0RolqQHkhp4rcECruodTwt5DDkQ9U+UOwpeofQhg/eKRCiex4XgyCG2FUVI/jPZQKJ2eDNgchmgpnVLojl4pQLnJasJoT7sRn03a8BwiAPEfuDt4rSnyjzKXrQ/JhECD5TrwZVDCL9yuKBd41FCA8bXwHn89jXh9CrdJ9NjlCVHNDsUBpRSCmOtpxxx2X1to/G/A2oqzgjaGaIfeLstwoqFRem0kBCqAQUDGOan6EqJEjQ7sZA+QP4XUK3isUB+5F6mGvhMERjonyi4eH66UyIUoKbQvvLwSUQsqT048I/RgGUPQ2btw45XWMI5Q/qkhyH/A6ffGLX/RcqjCm0sH4IQ+Pa8AIwOejrJC7h9GB+ZNuLBFi+aIXvcgVDRROxgX9E5S8XPs/HdwHQlrJkSKnjP6jmmHwLOWTWwfkSqHw0g/8cP8oxc895siB4A0uZNwLISoLKVNCiLKFkKj4mTapICjnUgAiDgIQAhmC6Lve9S63TGNlR5ngzCBAUMN6T+EIhGGEOARgvi9f8CCR30RZbRQzwq9o87XXXutlzykmEBLaEeAQfBFGuXYs47wOTwkgxCKk/ud//qcL6ih8lK0mzC8URUARQiBE6MUDhBCPUJ0urDEoDrweZYPXEbpEsQw8GsELEbxieHf43s9//vOuUPDdvIdrzATlvmkTbcb7hVBMfhefl67k+WygfZSR575RQjsI9PR/uLe5gtJEGNgXvvAFV7YR4Llf/ACCNeGifBc5PCi+cVBGOU+LvwdhnPtI2XO8K7MFpZp7xfjAk4gCQLl1SqAHnvSkJ3n7UbhR6BgjZ511luc8ZTsMG6WXsc594/MZi9wrFCiU/WztR3Gm31DauGbGCGXkmV+55Bdmg/tI/1GIgnxK2kOpc8Yu8yj1XLRsoNhxz7g3XCcFRwhTZJ7ES/YXOu6FEJVDgpJ+xW6EEEIIISobwjYJu0Oxi+eE4UmlCmKm0EIhhChl5JkSQgghxLyDNw7PEZ5XvJt4ogjTxbtFWK0QQpQj8kwJIYQQYkF48MEHPdQQJYo8phNOOMHDUwmbjFepFEKIckHKlBBCCCGEEEIUgEqjCyGEEEIIIUS5KVOUEKWCEOc5kJBKVaehoSH/286dO718LRWFqOzE4ZPZ+NGPfmTPeMYzvLoVlXbm4hBAIYQQQgghhCg5ZYroQhQpYqZDGWLOmiCWmr+hEFHWlTM3nve853nZ4T179qT9LM5+eN/73uevoUQtZ6u85z3vWfBrEkIIIYQQQiweilbNjzNKSED93e9+50oToFyFwybxTHHuCNV+OCyPczpQrDgoMRUqAT372c+25z//+f6Y81qe+tSn+mdwSKAQQgghhBBCVIxnikPuOOAuKFIBDvHj9PUzzzxzygF+F154oStf6eD1F1100eTj1atX25o1a/x5IYQQQgghhKgoZaq1tdXzpALj4+PuYbrkkkv8NPXOzs4pr29vb7d9+/al/awDBw7k9XohhBBCCCGEqJhDez/60Y/aAw88YN/5znfsy1/+stXV1U35O4+Hh4fTvndwcDCv12fi0KEeK1aheI7XaG9vKWobRPmhcSPyRWNGFILGjcgXjRlRzmMmtKVslCkUqa985StehOLUU0+1+vp6O3r06JTXoBg1NDSkfT+vT1WceMxp6/nAjSv2zSuFNojyQ+NG5IvGjCgEjRuRLxozotLHTNHPmfrgBz9oX/rSl1yheuYzn+nPrVy50rq6uqa8jsepoXyBTK8nL0sIIYQQQgghKk6Z+tSnPuUV+/71X//VnvOc50w+z1lR999/v4fvBW6//XZ/Ph08z98De/fu9Z9MrxdCCCGEEEKIslWmHnnkEfu3f/s3e/3rX++V+ig6EX44xJeKfJwV9fDDD9s111zjZ0m9+MUvngzh43VjY2P++KqrrrIf/OAH9u1vf9seeughe9e73mVPecpTVBZdCCGEEEIIUXnK1C9+8QtXhj7zmc/Yk570pCk/1dXVrmihML3whS+0H/7wh/bpT3/ay53DnXfe6a/D+wQXXHCBfeADH/DXoFgtXbrUrr766mJdmhBCCCGEEGIRkEgmyynFa37p6ipuNb+OjpaitkGUHxo3Il80ZkQhaNyIfNGYEeU8ZkJbyqIAhRBCCCGEEEKUI1KmhBBCCCGEEKIApEwJIYQQQgghRAFImRJCCCGEEEKIApAyJYQQQgghhBAFIGVKCCGEEEIIIQpAypQQQgghhBBCFICUKSGEEEIIIYQoAClTQgghhBBCCFEANYW8SQjY3zNk3QMjtrSx1la21Be7OUIIIYQQQiwoUqZEQdyy7bD9fONB6xsas6b6arv8tBV28bq2YjdLCCGEEEKIBUNhfqIgjxSKVDJptq6t0X/zmOeFEEIIIYRYLEiZEnlDaB8eqc7mOquqSvhvHvO8EEIIIYQQiwUpUyJvyJEitO9A77CNjyf9N495XgghhBBCiMWClCmRNxSbIEcqkTDbdnjAf/NYRSiEEEIIIcRiQgUoREFQbGJde5Oq+QkhhBBCiEWLlClRMChQUqKEEEIIIcRiRWF+QgghhBBCCFEAUqaEEEIIIYQQogCkTAkhhBBCCCFEAUiZEkIIIYQQQogCkDIlhBBCCCGEEAUgZUoIIYQQQgghCkDKlBBCCCGEEEIUgJQpIYQQQgghhCgAKVNCCCGEEEIIUQBSpoQQQgghhBCiAKRMCSGEEEIIIUQBSJkSQgghhBBCiAKQMiWEEEIIIYQQBVBTyJuEKFX29wxZ98CILW2stZUt9cVujhBCCCGEqGCkTImK4ZZth+3nGw9a39CYNdVX2+WnrbCL17UVu1lCCCGEEKJCUZifqBiPFIpUMmm2rq3Rf/OY54UQQgghhJgPpEyJioDQPjxSnc11VlWV8N885nkhhBBCCCHmAylToiIgR4rQvgO9wzY+nvTfPOZ5IYQQQggh5gMpU6IioNgEOVKJhNm2wwP+m8cqQiGEEEIIIeYLFaAQFQPFJta1N6manxBCCCGEWBCkTImKAgVKSpQQQgghhFgIFOYnhBBCCCGEEAUgZUoIIYQQQgghCkBhfqIocP6TcpuEEEIIIUQ5I2VKLDi3bDvsB+pyDhTly6m6R/EIIYQQQgghyomSCPMbHh62K6+80m655RZ//Dd/8zd22mmnTft51atelfb93d3d01578cUXL/BViFw9UihSyaTZurZG/81jnhdCCCGEEKKcKLpnamhoyN7xjnfYww8/PPnc+973Pn8usHv3bnvlK1+ZUZnavHmzLVu2zH70ox9NPldVVRJ6okiB0D48UihSVVUJ62yu83OheF7hfkIIIYQQopwoqjKFEoTSlMQ9EaOlpcV/AniqnvWsZ9kznvGMtJ+zZcsWW79+va1YsWLe2yxmBzlShPYd6B12RYrfPOZ5IYQQQgghyomium9uvfVWD8f75je/mfE1v//97+22226zt7/97VmVsnXr1s1TK8VcgveJHKlEwtwjxW8eyyslhBBCCCHKjaJ6pl7+8pfP+JprrrnGXvCCF9jq1aszvuaRRx6x0dFRe/GLX2z79++3iy66yN7znvdYZ2dnXu1BsC8W4buL2YaF4pL1bba+o0nV/OaAxTRuxNygMSMKQeNG5IvGjCjnMZNPGxLJ1Bi7IkHRiK9+9atTCkfs3LnTLr/8crvuuuvs5JNPzvjepz3tadbW1uYKFJfzsY99zAYGBuzb3/62VVdXL9AVCCGEEEIIIRYTRS9AkY2f/exndsYZZ2RVpODHP/6xJRIJa2ho8Mef+MQn7ElPepLdfffd9pjHPCbn7zt0qMeryxVLA25vbylqG0T5oXEj8kVjRhSCxo3IF40ZUc5jJrSl7JWpG2+80Z7+9KfP+LrGxsYpj9vb2726HyF/+cCNK/bNK4U2iPJD40bki8aMKASNG5EvGjOi0sdMydYPJ1zv3nvvndGz1Nvba4997GPt5ptvnnwOJerIkSO2YcOGBWipEEIIIYQQYjFSssoUZ0v19fWlDfEbHBy0gwcP+v+bm5vtwgsvtKuvvtruueceu//+++1tb3ubXXrppZ6HJYQQQgghhBCLSpk6dOiQ/166dOm0v/3kJz/xnKjAP/3TP9mZZ55pb3jDG/xw37Vr19o///M/L2h7hRBCCCGEEIuLkqnmVwp0dRW3AEVHR0tR2yDKD40bkS8aM6IQNG5EvmjMiHIeM6EtZe2ZEkIIIYQQQohSRsqUEEIIIYQQQhSAlCkhhBBCCCGEKAApU0IIIYQQQghRAFKmhBBCCCGEEKIApEwJIYQQQgghRAFImRJCCCGEEEKIApAyJYQQQgghhBAFIGVKCCGEEEIIIQpAypQQQgghhBBCFICUKSGEEEIIIYQoAClTQgghhBBCCFEAUqaEEEIIIYQQogCkTAkhhBBCCCFEAUiZEkIIIYQQQogCkDIlhBBCCCGEEAUgZUoIIYQQQgghCkDKlBBCCCGEEEIUgJQpIYQQQgghhCgAKVNCCCGEEEIIUQBSpoQQQgghhBCiAKRMCSGEEEIIIUQBSJkSQgghhBBCiAKQMiWEEEIIIYQQBSBlSgghhBBCCCEKQMqUEEIIIYQQQhSAlCkhhBBCCCGEKAApU0IIIYQQQghRADWFvEmISmR/z5B1D4zY0sZaW9lSX+zmCCGEEEKIEkfKlBBmdsu2w/bzjQetb2jMmuqr7fLTVtjF69qK3SwhhBBCCFHCKMxPLHrwSKFIJZNm69oa/TePeV4IIYQQQohMSJkSJQPKy6YDvQuuxBDah0eqs7nOqqoS/pvHPC+EEEIIIUQmFOYnbLGH2ZEjxXce6B12RYrfPOZ5IYQQQgghMiHPlLDFHmZHsQmUt0TCbNvhAf/NYxWhEEIIIYQQ2ZBnShSdEGaHIhXC7FBqeH6hFBq8YOvam1TNTwghhBBC5IyUKVF0SiXMDgVKSpQQQgghhMgVhfmJoqMwOyGEEEIIUY7IMyVKAoXZCSGEEEKIckPKlCgZFGYnhBBCCCHKCYX5CSGEEEIIIUQBSJkSQgghhBBCiAKQMiWEEEIIIYQQBSBlSgghhBBCCCHKVZkaHh62K6+80m655ZbJ5z70oQ/ZaaedNuXn2muvzfgZX/7yl+3SSy+1Cy64wN773vfawMDAArVeCCGEEEIIsRgpejW/oaEhe8c73mEPP/zwlOcfeeQRf/4FL3jB5HPNzc1pP+NnP/uZfepTn7KPfvSj1t7ebu95z3v8/3/3d3837+0XQgghhBBCLE6K6pnavHmzveQlL7EdO3ZM+xvK1JlnnmkrVqyY/GlsbEz7OV/96lft1a9+tT31qU+1c8891/7hH/7Bvvvd78o7JWx/z5BtOtDrv4UQQgghhKgYZerWW2+1iy++2L75zW9Oeb63t9f2799v69atm/EzxsbG7N5777WLLrpo8rnzzz/fRkZG7KGHHpqXdovy4JZth+2am7bZF2/e4b95LIQQQgghREWE+b385S9P+zxeqUQiYZ/97GftN7/5jS1btsz+7M/+bErIX+DYsWMeKtjZ2Tn5XE1Njb9n3759ebUnkbCiEb67mG2oJPBEXb/xoCWTZuvaGu1A77A/Xt/RVFEHA2vciHzRmBGFoHEj8kVjRpTzmMmnDUXPmUrHli1bXJnasGGDveIVr7DbbrvN/vZv/9Zzpi677LIprx0cHPTfdXV1U57nMYUt8qG9vcWKTSm0oRI4MJy0EUvYhs5mq65KWOOSettysNcS9XXW0VF5faxxI/JFY0YUgsaNyBeNGVHpY6YklannP//5nv+EdwlOP/1027Ztm33jG9+YpkzV10dehlTFiceZcqwycehQj3syiqUBM3iK2YZKIjk0bLWWtO0Heqyzuc49U7WJ6Pmurh6rFDRuRL5ozIhC0LgR+aIxI8p5zIS2lK0yhVcqKFIBvFQ333zztNfyOhSqrq4uO+mkk/y50dFRO3r0qBetyAduXLFvXim0oRLobK63y05bYT/feNC2HR6wpvpqf8zzldi/GjciXzRmRCFo3Ih80ZgRlT5mSlKZ+vjHP2533nmnnx0VoJgEClUqVVVVds4559jtt9/uxSzgrrvu8rwpPFpi8XLxujZb195k3QMjtrSxtqJypYQQQgghRPEpiUN7UyHEjzypL3zhC142/etf/7p9//vft9e+9rWTeVIHDx6cUsiC195www12zz332P/5P//HS67nG+YnKg8UqFM7m6VICSGEEEKIxeGZ4qwovFOf+MQn/PfatWvtX/7lX+yCCy7wv//kJz/xg3k3btzoj5/znOfY7t27/ZBecqUuv/xy++u//usiX4UQQgghhBCikkkkk+UUlTi/UJigmAUoqDJXzDaI8kPjRuSLxkzlHgcxnyHNGjciXzRmRDmPmdCWsvVMCSGEECI3OJCcYjt9Q2NebOfy01Z4zqgQQohFmjMlhBBCiNw8Uj+PHVDObx7zvBBCiPlHypQQQghRphDah0eK8/SqqhL+m8c8L4QQYv6RMiWEEEKUKeRIEdrHweTj40n/zWOeF0IIMf9ImRIOISGbDvQqNEQIIcoIik2QI0WyNAeU85vHOg5CCCEWBhWgEEpeFkKIMkYHlAshRPGQZ2qRo+RlIYQof3RAuRBCFAcpU4scJS8LIYQQQghRGFKmFjlKXhZCCCGEEKIwpEwtcpS8LIQQQgghRGGoAIVQ8rIQQgghhBAFIGVKOChQUqKEEEIIIYTIHYX5CSGEEEIIIUQByDMlhBBlDkcZKExXCCGEWHikTAkhRBmjQ7eFEEKI4qEwPyGEKFN06LYQQghRXKRMCSFEmaJDt4UQQojiImVKCCHKFB26LYQQQhQXKVNCCFGm6NBtIYQQorioAIUQQpQxOnRbCCGEKB5SpoQQZc9iLw2uQ7eFEEKI4iBlSghR1qg0uBBCCCGKhXKmhBBli0qDlzbch00HenU/hBBCVCzyTAkhyr40OIpUKA1OIQaeV9hbcZHHUJQjiz1kWAiRP1KmhBAVURocRUqlwUvTY8h94TGFMiSgilJFBgAhRCEozE8IUbaoNHhposOERbmhkGEhRKHIMyWEKGtUGrz0kMdQlBsKGRZCFIo8U0KIsgdh59TOZgk9JYI8hqKcDQDj40kZAIQQOSPPlBBCiDlHHkNRjgYAQvswAIScKY1bIcRMSJkSQggxL+gwYVFOyAAghCgEKVNCCCGEEDIACCEKQDlTQgghhBBCCFEAUqaEEEIIIYQQogCkTAkhhBBCCCFEAUiZEkIIIYQQQogCkDIlhBBCCCGEEAUgZUoIIYQQQgghCkDKlBBCCCGEEEIUgJQpIYQQQgghhCgAKVNCCCGEEEIIUQBSpoQQQgghhBCiAKRMCSGEEEIIIUQBSJkSQgghhBBCiHJVpoaHh+3KK6+0W265ZfK5u+66y172spfZBRdcYM985jPt29/+dtbPuOiii+y0006b8tPX17cArRdCCCGEEEIsRmqK3YChoSF7xzveYQ8//PDkcwcPHrTXv/71dtVVV9k//uM/2v3332/vec97bMWKFfaUpzxl2mfs37/fenp67IYbbrCGhobJ55csWbJg1yGEEEIIIYRYXBRVmdq8ebMrUslkcsrzKEUdHR329re/3R+vW7fOvVbXXXddWmXqkUcecUXr+OOPX7C2CyGEEKJ02d8zZN0DI7a0sdZWttQXuzlCiAqlqMrUrbfeahdffLG97W1vs/PPP3/y+UsvvdTOOOOMaa/v7e3NqJStX79+XtsqhBBCiPLglm2H7ecbD1rf0Jg11Vfb5aetsIvXtRW7WUKICqSoytTLX/7ytM8fd9xx/hM4dOiQ/fjHP7a3vvWtaV+PZ2pgYMBe+cpX2tatW10Re+9735u3gpVIWNEI313MNojyQ+NG5IvGjKj0cYNH6vqNB42gl3VtjXagd9gfr+9okodqASmnMSNKg0QJjZl82lD0nKmZGBwcdCWKsL+XvvSlaV+zZcsW6+7u9rDA5uZm+/d//3d7zWte4woYj3Olvb3Fik0ptEGUHxo3Il80ZkSljpsDw0kbsYRt6Gy26qqENS6pty0Hey1RX2cdHaXf/kqjHMaMKC3ay2zMlLQyRTW+N7/5zbZt2zb7+te/bo2NjWlf94UvfMFGRkasqanJH//zP/+zPfnJT7Zf/epX9tznPjfn7zt0qMctWcXSgBk8xWyDKD80bkS+aMyISh83yaFhq7WkbT/QY53Nde6Zqk1Ez3d19RS7eYuGchozojRIlNCYCW0pa2WK/KjXve51tmPHDvvKV77iRSgyUVdX5z+B+vp6DxOkyl8+cOOKffNKoQ2i/NC4EfmiMSMqddx0NtfbZaet8JypbYcHPGeKxzxf6m2vRMphzIjSIllmY6Yklanx8XF7y1veYrt27bL/+I//sJNOOinja6kEeNlll7kH64UvfKE/19/fb9u3b7cNGzYsYKuFEEIIUQpQbGJde5Oq+QkhFqcy9Z3vfMdLoX/mM5+x1tZWP3cKamtrbdmyZX7ILzlSbW1tVl1d7eXSP/nJT9ratWv9uY9//OO2atUqD/UTQgghxOIrZc5nSIkSQixKZepnP/uZe6fe+MY3Tnn+cY97nHuq7rzzTnvVq15lv/jFLzyc76//+q+tpqbGz6wiPPCSSy6xa665xhUtIYQQQpQ2KmUuhChXEsnUE3MXMSSmFrMABVWGitkGUX5o3Ih80ZgRpTZu8Ehdc9M2/9xQMILve8MT1smzVMZorXkUHSBdfmMmtKVsPVNCCCGEWBwgZOKR4kyoqqqEK1QUjuB5CZ6i3JHXtfKpKnYDhBBCZLZmbjrQ67+FqNSxhLUeIROP1Ph40n/zmOeFKGeYbz+PHSDNbx6X4jwUhSPPlBBClCCyZorFMpbwPl2eUsqcx/JKiXJHXtfFgZQpIYQocWsmlnoeU+pZG7CoxLGkUuaiEol7XUM+oLyulYfC/IQQokStmWy+wZrJY54XolLHEgrUqZ3NUqRExRC8rhQzwCPFb3ldKw95poQQosSQNVPMFRpLQhQXeV0rH3mmhBCixJA1U8wVGktCFB95XSsbeaaEEKIEkTVTzBUaS0IIMX9ImRJCiBIFoVeCr5gLKnks6UBUIUQxkTIlhBBCiLKk1Mu+CyEqH+VMCSGEEKLs0IGoQohSQMqUEEIIIcqOcir7LoSoXKRMCSGEEKKsy76PjydV9l0IURSkTAkhhBCi7FjsZd8JZ9x0oFdhjUKUYwGKnp4e++EPf2hbt261N7/5zXb33XfbSSedZCeccMLct1AIIYQQIg2Ltey7Cm8IUcaeqU2bNtnll19u3/3ud+0///M/ra+vz37+85/b8573PLv11lvnp5VCCCGEEGlYbAeiqvCGEGWuTH3oQx+yq666yr73ve9ZbW0Ul3z11Vfby1/+cvvIRz4yH20UeSLXvxBCCFGZqPCGEGWuTN177732/Oc/f9rzL3vZy2zz5s1z1S4xC9f/NTdtsy/evMN/81gIIYQQlYEKbwhR5spUW1ub50qlcscdd1h7e/tctUsUgFz/QgghRGWz2AtvCFH2BShe//rX2/vf/35705veZMlk0m6++Wb7r//6L/vKV75ib3vb2+anlSIv1z+KVHD9s9DyvBZZIYQQojJYrIU3hKgIZYpwvs7OTvvCF75gDQ0Nnie1fv16++AHP2hXXHHF/LRS5O36R5GS618IIYSoTFCgpEQJUaal0Z/2tKf5jyhN1z+hfXikQrlULbZCCCGEEEKUgDL1qU99Kuvf3/KWt8ymPSJPyIeKu/nl+hdzOZ6EEEKIXND+IRYreStTt9xyy5THY2NjtmvXLjt27Jg985nPnMu2iQIP7ZPrXxSCDoEUQojSoZyUk3T7xyXrtX+IxUHeytR//Md/pH3+//7f/2sJSsqIolTuIz+Kx3ilSn3RFaWHxpMQQpQO5WTcyrR/rO9oso6OlmI3r6IpJ4W7ksm7NHomXvnKV/pBvmJh0KF9Yi7ReBJCiMIF2k0HeufsGJJyO+ZE+0dx0LmiFahM/frXv7b6emnFC4UO7RNzicaTEEKUhkBbbsqJ9o+Fp9wU7kon7zA/qvilhvP19fVZd3e3vfvd757LtoksqHKfmEs0noqDQjSEKF/mKzy63I450f6x8Ou/zhUtc2WKan1xZYr/19bW2tlnn20nnnjiXLdPZEGV+8RcovG0sJRTToQQYuEE2nJUTrR/LOz6X24Kd6WTtzL1whe+cH5aIgpClfvEXKLxtDCo4IcQ5c98CrTlqJxo/1i49b8cFW5b7MoUxSVyrdT31a9+dbZtEkKIig7hUIiGEOXPfAu0Uk4qk7la/8tR4V7UytTFF188/y0RQohFEsKhEA0hKgMJtCJf5nL9l8JdRsoUeVLp6O3t9UN7ly5dOtftEkKIig3hUIiGEJWDBFqRD1r/K4+8c6bgK1/5in3+85+3rq4uf9zW1mZXXXVVRqVLCCEWC7mGcMiiLYQQixOt/4tcmfr0pz9t1157rf3lX/6lXXDBBTY+Pm533HGHfepTn7K6ujp7wxveMD8tFbNCJZiFKL0QDlm0hRBicaL1fxErU9/61rfswx/+sJ83FTjjjDNs5cqV/ryUqdJDJZiFWDgUwiGEEEIsHvJWpsiTWrdu3bTn169fb4cPz/7kbzG3qASzEAuPQjiEEEKIxUFVvm8gtO+LX/yih/cFKELBc+eee+5ct0/MUf4G4UYhf4PHPC+EmD9QoE7tbJYiJYRYVAbcTQd6/bcQi4W8PVPvec977E//9E/tpptusrPOOsufu//++214eNiLUojSQiWYhRBCCLHgKQWnr7ArO1qK3SwhSk+ZOumkk+ynP/2pXXfddbZlyxarr6+3Jz7xifbc5z7Xmpqa5qeVomCUvyGEEEKIhU4puP6hg3bhKZ0m062odAoqjb58+XJ7xSteYVVVVXbgwAG7/fbb/Td5U6L0qvQpf0MIIYQQC3okxJEBO9I3Yp11iWI3T4jSUqZQnP7qr/7KPvrRj9qGDRvshS98oQ0NDdnAwIA/9+xnP3t+WipmVaVPJTiFEEIIsVApBc111ba8qdZsZLTYzROitApQXH311XbFFVfYeeed52XSCfP73e9+Zx/84AftE5/4xPy0UszoUuc3j5X0KSoZJTcLIUTpphQkEuYpBfy+7PQVtnppY7GbJkTpKVObNm2yV7/61dbY2Gi//OUv7fLLL/fDeh/3uMfZnj17CmoExSuuvPJKu+WWWyaf27lzp73mNa+x888/35W33/72t1k/40c/+pE94xnPcCXvz//8zyu+THupVOmTcCsW0hN7zU3b7Is37/DfPBZCCFEaEBnzhiess9decoL/vkTnWYpFQt7KVEdHh23evNl/HnjgAXvqU5/qz1Pdb/Xq1Xk3gBDBt7/97fbwww9PPpdMJl0h4ru++93v2vOe9zx7y1veklFZu+eee+x973ufv+ab3/ymHTt2zKsOLhaX+vh4sihV+iTcioVCnlghhCh9dCSEWIzkrUzhLULRedGLXmTnnHOOe6Q++9nP2j/8wz/48/mAQvaSl7zEduzYMeX5m2++2T1TH/jAB7x64Bvf+Eb3UKFYpePaa6/1XK3nP//5dvrpp9tHPvIR+/Wvf+2fsZhc6gtZpU/CrViMnlghhBBCiFkVoHjVq15lF110kXuJnvSkJ/lzl1xyiT3lKU9xRSYfbr31Vrv44ovtbW97mytLgbvvvtvOPPNMW7JkyeRzF154od11111pP4fXv/71r598jIdszZo1/vzxxx9vlUoxq/SlrdxzeMCfl0VKzDU6L00s5oqsQgghKqw0OooOOVPkMdXW1npVv0KUlpe//OVpnz948KB1dnZOea69vd327duX9vWUZc/n9ZnAu1Mswnfn24ZVrfX+s9AsW1JrzamVe+qr/fli9uNio9BxU24wxjkAknNLKLdLlSiSm4sx9sudxTJmyoGbtx32Md07PDY5pks1z0TjRpT7mJHhovRJlNCYyacNeStTe/futXe/+9122223WWtrq+c39fT02NOe9jT78Ic/bMuWLbPZQpl1ilrE4TGFKtIxODiY1+sz0d5e/JO6S6ENudDR0WIvHk3adXfvsb19I9bcWGvPPW+NnbW+o9hNW5SUy7iZDVd2tPgBkJxbQrldVYmaHYthzJQye7sH7Dfbjlpdfa2dsaLF9h8b9MeM8VIe2xo3ohzHzI0PH3R5pXdw1JobalxeufSUFcVulijhMZMPeStT73//+/2w3htuuMHWrl3rz23bts3e+9732t/93d/NSXl0yq0fPXp0ynMoRg0NDRlfn6o48RjvWT4cOtTjuT/F0oAZPMVsQ76csbzB2i5cO8XS09XVU+xmLSrKcdzMBoL6/ADIkVGNtQJZbGOmVNl6oNcOHRu0dcsbbWhw2FprEu513br7qNWW4Lk8GjeiXMcMHqnv3LLdxpNmqyciaXjcVpOQh6rESJTImIm3ZV6UKTxS3/ve9yYVKVi3bp0rUi972ctsLli5cqUXp4jT1dU1LZQv/nr+nvr6FSvyszpw44p980qhDfnQ2VzvP1BO7a40ym3ciOKjMVNcWhtqrakuJQ+wrtqfL+X7onEjym3MHO0fsd40Od48H+QXUVoky2ydybuaH9X1OGsqFSrnxRWs2cBZUffff7+H7wVuv/12fz7T6/l7PBSRn0yvF0IIIRZzRVYhFstZkaVwlIyobHLyTH3/+9+f/D+V+zjTiTOmKI1eXV1tGzdutC9/+cv2Z3/2Z3PSKMqtU5GPs6Le/OY3269+9Ss/S+rqq6+eDOHr7u62trY2//6rrrrKXvnKV3pFQNpE7hbVBSu5kp8QQojyppgVWYXgbEiONKEyL8oFyjxjslINF1wrhotwrZpvYq5IJKkgMQMUl8jpwxIJ+8UvflFQQ0477TT76le/6qXSYfv27a60Ud78xBNP9JysJzzhCf63W265xUu0813HHXecP0foIflaKFlPfOIT7YMf/KAtX748rzaQg1HMnCmKOhSzDaL80LgR+aIxIwpB46aywBN1zU3b/F6GMFPu8RuesG7OlIxSGzOq5lf6JEpozIS2zJkytViQMiXKDY0bkS8aM6IQNG4qC0L7vnjzjsk8IsLf8Nq89pIT7NTO5jn5Do0ZsViUqYLOmTp8+LBt3brVxsfH/TH6GKF3hP694Q1vKOQjhRBCCCHEAqCD0IWYO/JWpr71rW/ZBz7wARsdHfWwvuDY4v/nnnuulCkhhBBCiBJGeUSVh8IYy0iZ+uxnP2tvetObXGkil+rb3/629fX12bve9S677LLL5qeVQgghhBBizlABlMphsRQTqZjS6AcOHLDnP//5VldXZ2eddZbddddddvLJJ3uBCBQrIYQQQghR+qBAkSMlRaq8PVIoUgSKkQPHbx5Xcrn7slemKEdOzhRs2LDBHnzwwcmDc/fv3z/3LRRCCCGEEEJMA88iHily38KhxDzmeVGiytSzn/1se/e732133HGHXXrppV6S/Gc/+5l9+tOf9hLmQgghhBBCiPlHhxKXYc7UO9/5TmtpabEjR47Y05/+dHvRi15kf//3f2/Lli2bPFRXCCGEEEIIMb+omEjx0TlTMXTOlCg3NG5EvmjMiELQuBH5ojGzsFRCNb/EYjpnSgghhBBCCFEaoECVqxK16HKmhBBCCCGEEEJImRJCCCGEEEKIhVGmPvShD9mOHTsK+zYhhBBCCCGEWKzK1A9/+ENLkJUlyjI5cdOBXh3kJoSYhtYHIYQQIn/yLkDxmte8xj7wgQ/47zVr1lh9/dRkN54Tpcct2w572UwOcgtlMy9e11bsZgkhSgCtD0IIIcQCKVOf+MQn/PeNN97ov4OXigrr/P/BBx8ssClivsDSjKBEmcl1bY1+oBuP17U3qfKLEIscrQ9CCCHEAipTv/jFL2bxdaIYcO4AFmcEpaqqhHU21/nBbjwvYUmIxY3WByGEEGIBlam1a9fO4utEMeAAN0J3sDgjKPGbxzwvhFjcaH0QQgghFlCZOv3007MWoFCYX+mBdZkcCEJ3sDiHnAhZnYUQWh+EEEKIBVSmvvrVr055PDY25qXSv/SlL9lf/dVfzaIpYj4hmZwcCEJ3sDhLUBJCBLQ+CCGEEAukTD3ucY+b9tzjH/94W7dunV199dX2rGc9q8CmiPkGAUlCkhAiHVofhBCi8gsOyWhWAspUJtra2mzLli1z9XFCCCGEWIRI4Ft8FOOeL7ZxpiMwSkiZ+v73vz/tub6+PvvOd75j559//ly1SwghhBCLDAl8i49i3PPFNs50BEaJnjMVoBhFbW2tnXPOOcqZEkIIIURBSOBbfBTjni/GcaYjMEpMmfrlL385Py0RQgghxKJFAt/ioxj3fDGOMx2BMb9UFfKmnp4e+9rXvmYf/vCH7fDhw/arX/3KK/oJIYQQQuTrKdh0oNdGxsYnBb7x8aQEvkVAXMhfqHtejO8slSMwONkIxZHfOgKjiJ6pTZs22atf/WpbvXq1//9Vr3qV/fznP7f//u//ts997nNpq/0JIYQQQsyUu3L8sgbbeXRQZ54tEopxzt1iPVtPR2CUkDL1oQ99yK666ir7i7/4C7vgggv8OUqiU83vIx/5iBeiEEIIIYTIN3cFReqF56622uoqCXyLhGII+YtVsdARGCUS5nfvvffa85///GnPv+xlL7PNmzfPVbuEEEIIUcGE3BVyOELuCo9RpE7tbJbQt4jgXi/0PS/Gd4rKJG9lCg/U1q1bpz1/xx13WHt7+1y1SwghhBBllvfE73xzV7YfGbCDvUP+u9JzV4QQlUfeYX6vf/3r7f3vf7+96U1vsmQyaTfffLP913/9l33lK1+xt73tbfPTSiGEEEKUJIWe2YNHgBypH9633wZHxq2htsr++OyV8hQIISpbmSKcr7Oz077whS9YQ0OD50mtX7/ePvjBD9oVV1wxP60UQgghRMkxmzN7eC85Uqd1NltzXbX1Do/5Y56XQiWEqFhl6vOf/7xdeeWVXhpdCCGEEIuX2ZzZk/re9vFkxZ/3I4SoPPLOmfrsZz9rIyMj89MaIYQQQpQNszmzZzGe9yOEqDzyVqbwSn3mM5+xbdu22fDw8Py0SgghhBAVfRioDhIVQizKML/f/OY3tmfPHi86kY4HH3xwLtolhBBCiDJgNmf2LNbzfoQQi1iZ4oDeBOYjIYQQQohZHgaqg0SFEItKmfqnf/onV6hOO+20+WmREEIIUSFQmU5eFyGEqFzyVqYOHDhg1dXV89MaIYQQYpGfvySEEKKClannP//59rrXvc7++I//2NauXWv19fXT/i6EEEIsZmZz/pIQQogKVqZ+8pOfWFVVlf3oRz+a9jdyqaRMCSGEWOzM5vwlUXwUnll56J6KklGmfvnLX85PS4QQQogKIX6GEoqUzlAqHxSeWXnonoqinzN122232ejoaNbXDAwM+IG+QgghxGJHZyhVRngmv3nM86I80T0VJaFMvepVr7Lu7u4pzz33uc+1vXv3Tj7u6+uzj3/843PfQiGEEKIMwfL9hiess9decoL/liW8fMIz8SaG8Ewe87yo/HuKgrXpQK8ULTH3YX5J1PgUdu3aNaO3ajZ873vfs/e85z1p87Ieeuihac9TEGPjxo1Tnrvuuuvs1FNPnbc2CiGEENnQGUrlhcIzF+89VSigWLCcqYXiiiuusEsvvXTyMYrbq1/9anvKU54y7bVjY2O2bds2u/baa23dunWTzy9fvnzB2lsuKAFTCCGEyB6eiVBNeGYQqrVfVvY9VfVNUZHKVENDg/8EPve5z7mH7J3vfGdaL9nIyIide+6500q1i0eR1UUIIYTIDvsiQrQMj4vnnqr6ppj3nKlic/ToUfv3f/93e8c73mF1dXXT/r5582ZbvXq1FKks8b5KwBRCCCFyAwH61M5mCdKL5J7GQwHHx5MK7xTz45n66U9/as3NzZOPx8fH7frrr7e2tsiz0dPTY/PFN77xDevs7LRnPetZaf/+yCOPWG1trb3xjW+0++67z9avX2/vete73FOVD1RbKhbhu2fThpu3HbbrHzpovcNj1lxXbZedvsIumfA8HRscsb7hMVu3PGZ1OTLgz69q1WZRrszFuBGLC40ZUQgaN6KSxwxy0OWnr3AZCtkoyFCSjxbvmEnk0YZEMl11iRSe9rSnFe0cKpr39Kc/3V73utfZy1/+8rSvoVDFr371K/vQhz7kHqpvfetb9sMf/tAPGObxYmBv94B97PpN7nFa2dpg+48N+kB422Wn2uqljTP+XQghhBBiMYOsdKRvxJY31Uo2EnOrTBWTe+65x6666iq76aabbOnSpWlfQ3GKwcHBSc8Zl0R1v+c85zn2pje9KefvOnSox5WNYoBi097eUnAbCO37ws07Jj1PuKmxrvyvS05wt/ZMnitRnsx23IjFh8aMKASNG5EvGjOinMdMaEtZF6AI3HjjjXbRRRdlVKSgpqZmSggi5dM3bNhg+/fvz+u7uHHFvnmFtqG1odaa6lJKf9ZV+/Ph8y4+sc3WtU1NwCz29Yq5oRTGrigvNGZEIWjciHzRmBGVPmZKvgAFnqnHPOYxWV/zyle+0j71qU9NyefizCkUqsVW+hNNmgo0/E5XzlVJtUIIIYQQQswNJe+Zevjhhz1kL/VcqcOHD7u3iup+5HR9+tOftjPOOMOLT3z1q1/1ghgveMELbDGhcq5CCCGEmC06k1KIClKmurq6rLW1dcpze/fu9aIUKE0XX3yxveY1r7GhoSEvQMHrzzvvPPvSl740JfRvscCip4VPCCGEEIWgMymFqLACFAtJV1dxC1B0dLQUtQ2i/NC4EfmiMSMKQeNm8Xikrrlpm9/jkH/NvX/DE9blbajVmBH5UkpjJrSlInKmhBBCCCHE/ENoHx4pFKlwJiWPeV6Uv6JM5Wd+i0UW5ieEEEIIIeYfcqQI7ZtSGbi+2p8X5YtCN+cXeaaEEEIIIUTOlYFF+YAnCkWKsLl1bY3+m8fyUM0d8kwJIYQQQoiyrQys6oMzh26iSIXQTRRlnldfzQ1SpoQQQgghRFlWBlYIW3YUujn/KMxPCCGEEELMebGDvd0D8/49CmHLjkI35x95poQQQgghxNx6iobHrL31gP3RumV28Ynz4ylSCFvlhm6WE/JMCSGEEEKIufUULY88Rdc/NH+eongI2/h4UiFsWUCBOrWzWYrUPCBlSgghhBBCzPk5VStbG6x3eP7OqVIImygFFOYnhBBCCCHmvNjB/mOD1lw3O0/RTJX60oWwqbqfWEikTAkhhBBCiDnzFBHqt+3IgLW3NthlpxfuKcq1Ul+8+qCq+4mFRsqUEEIIIYRIS75enuApOjY4YuvXLrPakVHPnZptpT68XTzmszO1o5D3CDFbpEwJIYQQomxRSNf89WGhXh4+Y1VrvXUsbbSurp4Fq9Sn6n6iGEiZEkIIIURZopCu+evDYnt5CjlsVgfUimKgan5CCCGEKDt0YOv89mFqZT5+83i+KvPNRaU+VfcTxUCeKSGEEEKUHQrpmt8+LAUvTyGHzS7GA2oV6lpcpEwJIYQQouwoBWG/kvtwSmW+wwOTIYALLazHK/XN53vKFYW6Fh8pU0IIIYQoO0pF2K/kPlyMXp5yoth5bSJCypQQQgghyhIJ+/Pfh4vJy1NuKNS1NJAyJYQQQoiyRcL+7FEflicKdS0NVM1PCCGEEEKIMkPVC0sDeaaEEEIIIYQoQxTqWnykTAkhRAWhErlCCLG4UJhmcZEyJYQQFYJK5AohhBALi3KmhBCiAkvk8pvHPC+EEEKI+UHKlBBCVFCJXCo6hRK5POZ5IYQQQswPUqaEEKLCSuSOjydVIlcIIYRYAKRMCSFEBaASuUIIIcTCowIUZYoqdgkhUlGJXCGEEGJhkTJVhqhilxAiEyqRK4QQC4MM2wKkTJV5xS7yIniMNVoTWVQi2qyEEEJUmmFbe1vlIGWqTCt2oUiFil3kR/C8JmN5o4V1OvLCCiGEqDTDtva2ykIFKMoMVeyqTFhYr7lpm33x5h3+m8eLHZ2bJIQQAlj3Nx3oLZn1fzZHUWhvqzykTJUZqthVeWhhTY/OTRKi8oVSIcrR2Dgbw7b2tspDYX5liCp2VRYK3Zx5s6JP5IUVYnYotEiUG6WaJx4M27SF/TrMp1zapL2t8pAyVaaoYlfloIV17jcrIUR5CKVClKuxsVDD9mz2NuVWlyZSpoQoMlIaMiMvrBCVL5QKUa7GxkIN24XsbfIsly5SpoQoAaQ0ZEZeWCEqXygVYrEZG/PZ2+RZLm2kTAlRIkhpEELMF5UslIrKZjEbG0NY36H+YXmWSxgpU0IIIcQiYDELpaK8WYzGxnhYX1WVWf/wqDzLJYqUKSGEEGKRsBiFUiHKjXRhfT02aoOjY/IslyBSpsqYbFVdVPFFCCGEEKIyCsb0D4/Zc89aZe1NdZLtSgwpU2VKtqouqvgihBBCCFFZBWM2dKjgRClSZSXM9ddfb6eddtqUn7/4i79I+9qbbrrJrrzySjvvvPPsVa96le3cudMWi/uX3zzm+Wx/S/c5mw70pv2bEEIIIYQoXsGYRMI8rI/fCusrXUraM7V582Z76lOfah/84Acnn6uvnz6Q9uzZY3/+539ub33rW+3SSy+1T3/60/bmN7/ZfvjDH1qCEbiIzguBXCq+yHslhBBCCFGaqGBM+VDSytQjjzxip556qq1YsSLr67797W/b2Wefba997Wv98dVXX21PfOIT7dZbb7WLL77YKtX9u/3IgDXXV1vvhEIUqrrMdJaIzisQQgghBCjHunRRwZjyoKrUlal169bN+Lq7777bLrroosnHjY2NdtZZZ9ldd91llQgT6/hlDbbxQK/9bssR/83jMOlmcg0HzxbKVvBe8Th4toQQQghR+RClcs1N2+yLN+/w3zzOh7lKF1DagShnStYzlUwmbevWrfbb3/7WPve5z9nY2Jg961nP8pypurq6Ka89ePCgdXZ2Tnmuvb3d9u3bl9d3FjMiMHx3Lm1gsdl1dNBO62y25rpq6x0e88cHeodcabpkfZut78jsGl62pNY9WnHvFY95fqH7QBaxhRs3QoDGjCgEjZvKg/33+pQoFR4jP+SyH9+87bBd/9BBl0GQRS47fYVdEksXyHXMzPQ5YvGQKKF1Jp82lKwyRR7UwMCAK07/7//9P9u1a5d96EMfssHBQXv/+98/5bXhdXF4PDw8nNd3tre3WLHJpQ0HhpM2Ygk7fe0yq65K2Nh40rYc7LVEfZ11dETvD7/Twd9ePJq06+7eY3v7Rqy5sdaee94aO2t9hy0kNz580NvQOzhqzQ013oZLT8ke0ilKd+yK8kJjRhSCxk3lEGSJDZ3NLks0LqmfJktkYm/3gP1m21Grq6+1M1a02P5jg/74wlM6bfXSxpzHTD6fIxYP7WW2zpSsMrV27Vq75ZZbbOnSpV5E4owzzrDx8XH767/+a3vPe95j1dXVU4pSpCpOPG5tbc3rOw8d6nELTbE0YAZPLm1IDg1brSVt+4GeSc9SbSJ6vqurJ6fvO2N5g7VduHaKVyjX986VRew7t2y38aTZ6olr4HFbTUIeqnkaN0KAxowoBI2bymM2ssTWA7126NigrVveaEODw9Zak7BtRwZs6+6jVjsymvOYyeVzxOKJ6EmU0DoT2lLWyhQsW7ZsyuOTTjrJhoaGrLu729raHnUBr1y50rq6uqa8lscoYPnAjSv2zculDZ3N9XbZaSu8aEQ4CZvHPJ9P+3k9P+F7F5Kj/SNeOCO16iDPhzaJ8hq7orzQmBGFoHFTOcxGlmhtqLWmupRiV3XV/nzqe7ONmXw+RyyeSszJMltnSlaZuvHGG+2d73yn/c///I8XlIAHH3zQFay4IgWcLXX77bdPCft74IEH7C1veYtVKuVeMjPTgXTxqoNCCCGEKD1ZIhS7iitihZyDNFefsxgotUrMlewhqxhl6oILLvDwPfKjOEOKQ3g/8pGP2Ote9zovRnH48GEPASQ36kUvepF94QtfsGuuucbPpeKcqeOOO64iy6JXSslMLaBCCCFE+coSc2XULXfjcCmcMbrQfbZYPGRlXxq9ubnZFSSUJpSl973vffbSl77Ulam9e/fak570JLvzzjv9tShOn/zkJ+273/2uvfjFL7ajR4+6QlWJB/ZWEky8Nzxhnb32khP892KeiEIIIUS5gRB/amez/55NefP454iZI3rGx5NFi+hJ9ZAlk+aPF3NZ+5L1TMEpp5xiX/rSl6Y9j/K0cePGKc89+clP9h9RXpSzd00IIYSYS8o1dCqdp4JjWkTlRfSUkoesVChpZUoIIYQQlSmAi8oIncqUy8N5VTOVWBflFxKpnPfpSJkSQgghypRyFcBFaRcXmCtPhai8iJ5S8ZCVElKmRE7I8imEEKW1Jm851GfX3bfPGmqq50QAr4R1vlyvoZxDp+SpWHyUgoeslJAyJWZElk8hhCi9NXnfsSHbeWTAzl3TMmsBvBLW+XK+hnJWSBbaU1GuCnOlUWwPWSkhZUpUbOiBEEJU8pq8vn2J7T02aA/s77XljbXWNzJekABeCet8rtdQqoJ4uYdOzYWnIpd7898P7rcbNnbZ6Pi4tTfVlZXCLCoXKVOiYkMPhBCi0tfkM1c12z17emzL4QFb1RoJ5PmuzZWwzudyDaXuuSr30KnZeCpyuTf//cB+u+b32210bNya62tsbDw5b0p/qSrdojSRMiUqNvRACCEqfU2uq662s1e32HPPWmUbOgoTKithnZ/pGsrF+5aPQlIsgX+uvzeXe8Nrbth00EbHkv5c/8i4HeobtuqqxJwr/aWudIvSo2QP7RWlFXrA+cdY+fhdTqEHQghR6Wvy885eZY9f31bwulwJ6/xM1xA8VyhawXPF43KtOIfAf81N2+yLN+/w3zwu1+/N5d7wfxSplvpq6x8ZsyW1VdY7NGo1VVVzqvTrQFpRCPJMiYoPPRBCiEpiPtbkSljns11DJXjfAsXyss3X9+Zyb/h/e3OdjY4n7Uj/sO3vGbaa6ip7xmkdc3rNsw15LaXwwFJqS6UjZUrkhKq2CCFEZa/JlbDOZ7qGci/wUAo5bvl+L8L8geGkJYdQkjK3K5d7E39NTVXCVi9N2DNOXWHPOmPlnF7jbJTuUgoPLKW2LAakTAkhhBCi4qkE79tsBf7ZeCvy+V6E+es3HrQRS1itJe2yGYT5XO7NQty/QpXuUsrJK/fKluWIlCkhhBBCLAoqxftWiMA/W29Frt8bF+Y3dDbb9gM9OSkWudybhbh/hShtpVQRsxIqW5YbUqaEWOTIOiWEEOVFvgL/XHlOcvneuDBfXaal9vNV2kopJ69SKluWE6rmJ8QiplgVoYQQQswOBN9TO5vz8pzMRTXDmb43LsxzFlQ5F/sox4qYi62yZSkgz5QQi9RzJOuUECIf9nYP2NYDvdbaIC92ubGQnpMgzJMzteVgr9UmzHOmKn3MlFJO3mKpbFkqSJkSFUmlhK7NZ1xzKcV4CyFKm5u3HbbfbDtqh44NWlOdcizKjYWuZsjYWN/RZIn6uhmr+VUSpZSTtxgqW5YKUqZExVEpiZXz7TmSdUoIketadP1DB62uvtbWLZcXu1xZaM8Jn9/R0WJdXT2+j4nSoZS8aJWAcqZERVFJp5fPd1xzKcV4CyFKF9ac3uExW9naoByLRZRnJSoT5KFNB3r9/xoLc4M8U6KiqKTQtYXwHMk6JYSYCdaG5rpq239s0FprEvJilymVEv4uCqdSIndKDXmmysySUI4elmIpIONlXkVooTxHslROR/NNiEdhbbjs9Im16Ej5e7EX4/wupHLrYuynSqaSIndKDXmmKtySsNgsUZWWWCnP0cIjy50Q07lkXZtdeEqnbd19tKyr+WWb35W6XxaSf6t1cH4pxlirpMidUkPKVAUXIVisi2GlKSClVB1osc+3hd4AK1W4E+XJ6qWNVjsyWrbFBLLN722H+ip2v8xXiNaxGfNLsWSzuU4d0P70KFKmSpxCLQnFXgyLPcmkgFTW/SyF+bbQwtZiNYaI8qXU14lM83tLV5/dsKlylYd8hejUflpSV21bD/XblkN9FdEfxaSYstlcRu5of5qKlKkS35RGxsYLsiTMhTu30I1Rk6yyWEz3M5PQwTxcyA2w2MYQISpxncg0vy1hFR3+lK8QHe+n4bExe2Bfr42OJe26+/Z5on2p3ddSVuxT21HsULu5iNzR/jQdKVNlsCkdv6zBdh4dzMuSMFt3bqEboyZZZTEX97NUNrXZCB211VULugEWe8MVIh/KZd3PNL9pZzmcuTebtTQfITr00w/u22f37e2xmqqEnbumxepqqkvyvqb2D57G23YeLbpin06OKoWxNtvIHe1P05EyVQabEorUC89d7QJdrovobNy5s9kYNckqi9nez3KwVucidDAnFnID1IHKopwop3U/k1JRCoWLsilLc7GW5iNE89njSe7tqG1oa7SWxlqvkBu/r6ViKKMdv32ky27f1W39w2O248iArWltsLNXtxTNAJhJjnrDE5pKYqzNBu1P05EyVSabEooU5asXwp07m41Rk6yymM39LBdrdS5Cx0JWiQwb+WNPWGa37ThathuuWDyU27qfTqkoduGimaoMFmMt3dDRZKta661vhHSDqUeNlIqhjHa4B21Pj9VUJ+y4ZQ02OjZuR/qHrX9krGgGwGxyVLHH2mzJtB8CpfTL8Zpmi5SpCt+UCnHnzqYNlVaafLEzm/tZTtbqXFiIDTB1I3/s8ctcoFmMm5MoHypl3S9W4aK4stTRXGf7jg26ghCUpWKtpdmE5lIwlIV+Gxodt/qaKqutTvhzDbVV1jM0ZgPDY9Y7sZYutAFwJjmq3Itkpe6H2w71+fllxVaui4WUqQrYlGbjkk733tlujOVudRFzcz/LzVqdC/O5AabbyIn7v+D4ZSUVbiREOrTuF05QlmqqzTbu6rXh0TEbGkt66NqLzl9b1LU03X3F+1AKhrLQb2ta6m3fsSFLJpMeitjSUGuj4yN2sHfY2pvrimIArBQDQy774f4clNBK31OkTJX5pjQbl3S29852YywHq0ulT+65pJD7uRg2k7mkVDx5pRK+s9BoPZg95bDulyKMuaoq86p5De5hqbKx5LjnAD3ppI6c1tL5HL+p97VUDGWhHYQhrm9rtAf299roeNJOaa23l1ywxja0P+rVz6V/5vq6imlgWMj1rHuGvSuXPSW0d9mSWuvoaLFyQ8pUGW9Ks3FJ5/LeSt4YF6vAuNDQp831Nbane9DWLG2ws1a3FrtJJUspCCjlnOc2G7QeiGLC3LrwuKWe9zOSSFpdTZWdubLZRscjQZW/ZxPMF3r8loqhLN4O+ur0lc3ej0EBzbd/5uO65lKOylVBWujxsDTL3jVTCGtqe5vrq+3Fo0k7Y3mDlRNSphapJXs27y13C+5iFRiLQeqi3js0KiG1hAWUhfaOlcJaUonrQSn0q8gPFAA8UeT/ELaGt6U+EQmq2QTzYo3fUgnrnKkd+fZPLp9XjGvOVUEqxnhYmWXvCiGhmUJY07X3urv3WNuFa62zuXzWLilTi9SSXeh7K8GCWyrhVKXCfG0OlSikzjfFFlAW0jtWKmtJpa0HpdKv+bLYFUCu+Xlnr/J7d7BvJGdjSjHHb6lEr2RrRyH9k+nzijW38tlLizUeLs6wd8VDWKsT5uX2AcPB6StbPGrlUN+wnd7ZPNnevX0j/jlSpsSCMJvylIUWuqgE4bgUwqlKhfncHCpNSF0oFkpAYT4f7Z/b4jP5fHeprCWVtB6UUr8uBgWwFIwphYzfxaS4ztX8LubcymcvXaj1LFPxspUp7QkhrH/YftSGx5OWoI0NNfbIwV770i07vYz9jqMDNjaetLNWReeCNTfWlt36K2WqzJlNecr4e0fGxj3plQkylxaeUqQUwqlKgfneHCpJSK00bnz4oH3nlu2TZYPnqvhMrkJaKa0llbQelFK/VroCWCrGlHzH72JTXOdqfhdzbuWzly7EepbvGDp9ZYs11lVbXTJp7Y21dmRw1Pb1DNuapY2e50bREA5aHhwZt86WOnvJeSd6e1kTygUpUyVOLsJJWHwL2ZR4HgUsl4lRScIxffKMU80wk2xYpJv2fG8OlSSkVhKsE8Skj89x8Zl8NthSW0uKHV45V5Rav1aqAlhq5Dp+F6viOhfzu5hzK9+9dD7Xs1wKSqSCob6TPMChUT//K5kct+qqhL+fOb9sSY1tPZx0papckTJVwuSr/ReyKeWzuFaKcDy9X6NrKwUWMvxiITaH+RZSF1O4ylxBf/UOjtrqiY1sLoTXfIW0UlxLSiX/YzaUYr/ONHeJiig3BbAUyWX8LmbFdbbzu9hzK9+9dL7Ws5nOREsH7T1heaPtPDJgQ6PDNjJmHta3/fCA1ddUez4Vhy6fvarZC6+oAIWYMwqxIGUSjtmsMuVQ5bu4lrsFt5Qtc5Va3na+FvVSD1cpVUWP9jQ31Myp8FqIkFbua0mpUg79mjp3j1/WYDuPDpa8Alju5GNAK9X1azHPrVIw+ORyJloqPPfYE5bZnbu6bXQsae1Lai2RqLOuvmFLHuj1585d02ItPj6TKkAh5o5CK9CkCsdsUt+7Z29GgbMQ70QpTOhCKVXL3GIvb1tJSnGpK3r0z3PPW+M5U3MlvBbq5VzotWQ+BMRcPnOhBNP495za2WzlMndRpF547moXzMppHVoo5mr85GpAW4j1q1yVtXKWf+aCXM5ESwfpFCcsa7QVzXWeP7Wkttoe2t9rT9rQZnfs7ra6mmobH0+qAIWYWwoVTlKLSqBIzXQwb7mEhVRyTkGpKnmlSqn1V1wwgFJW9ODSU1ZYW01iWjW/QimHdWQ+BMRcPnOhFOtSVuBzmbsoUqWqABaTub6vc30uUyWP1cVILkpu6ploXf0jNjo+7jJnJvi89uY6G0uaK1KMKx5fenKHHb+8cXLvaK6vdmOfClCIOSEf4SR18IefcFjaTAJnuXonKknoK5aSV66bWikpxal9iJWulBS9TNCWuQyjKOV1ZD4ExFw+c6E8qKXuqZ3N3C1XD0Yp39ds3pX5NlRxTRQsGB4dt1WtDdY/PFayY3WxkY88gHcKheqhA712qH/E2pvq3HjfOzSa9j3ZZC9+wt6xbEmtnbW+w7q6eqyckDJVwuQinGQb/PmW0wS+K/64EilFoW+hlTw2tC2H+uy6+/ZZQ031rDfqhRZ4SkUpTifssMEQU14Kil6phsAs9HiZDwExl89cKA9qqXlq52rulquxp9Tuaz7zbb4NVRQquG9vj9VXJ2zvsSFb395oo2PZQ8RE8RX3MIa2dPXZbTuP+rgcHhs3nFGndTbbictnliGyVVEOe0eCg6jKEClTZSyczDT4tWmVV9zzQil54V7vOzbk1XVI/JzNRl2ssVMKSnEmYeexJyy1B/b3zruiV45W+2KMl/nwhuTymQvlQSW8hjCb7UcGJoWaYirwM/VfLnO3nLxt88VcjJ9859t8Gqq4pxibaqoSxr+BkVG7a1e3rW9vyhoiJoqruIfjcw71Dft5UGtaG+zs1S22uavPi0ictap5RhmilKsoV7wytX//fvvwhz9sN998s9XX19sVV1xhb3/72/3/qfzv//2/7Ze//OWU5z772c/aU5/6VKsUUjeoXKxW2rTKi/lW8uL3en37Ett7bNCF/uWNtV6SNN+NeiHDmNKN4WIrxZmEHWLK+ZlPRaccDSDFWmvmw7CUy2cuhAc1tLd7cNSFna7eITuhbUnRwpfz6b9s7Ssnb9t8KZ6zHT+Fzrf5MlTxeePjZp1NtfZwV78Njox5Dk1Vot++dMtOe8ZpHfasM1ZaMSlHA9VckK0adBhDFI/AM3Wkf9j6R8Y8THPTwT7b0zNka8z8NyXOU2WIxSBjlqwylUwm7S/+4i+stbXVvva1r1l3d7e9973vtaqqKnv3u9897fWPPPKIffSjH7XHP/7xk88tXbrUKoV0GxQDMRerlTat/KjkxTT1Xp+5qtnu2dNjWw4P2KrWaOPO55oXYuyUstIwk7Azn2Ga5bg5zcd4yXW+zodhKZfPXKgDNMlhwDM1Mpb0ynhnrW7N63Pmon1zOS5LKS9yvshlbZvN+JnNfJsPQ1Uoq32gb8RaG2r8rCG0q2ODo/bg/h4PPbek2bPOLI5CVcp7TaHkOrcz7WUUhgljqH9k3Jrra/zg3YHhSBFet7zRegZH7NcH+jx8j8d4suLftRhkzJJVprZs2WJ33XWX/e53v7OOjg5/DuXqn/7pn6YpU8PDw7Zr1y4755xzbMWKFVZpZNqg3vCEpjmxei6GTWsxL6bZ7nVddbW765971irb0FF6Ak85KA3FCDdcyM1pLo0Lcz1eCglhmmvDUi5C53x5UFPbS4hfqIxXSB8i6KKUZTovJoyHUAUytCFTtERTbZUbarBm53v9pZIXOV/ks7YVOn6KvbdzjQeGk5Yc4vvrHy2rvbfHEskkepPVTYzVZY011jM4ZjdsOmgXHL+sJPJfS22vme/1Md1eRr/ExxCFJgZHB+1g77Atqa+2C45bavfuPWandtZmLChS7HG4qJUplKLPf/7zk4pUoLe3N63ilUgk7Pjjj7dKJNsGPxeCXKVvWvOxmJar9yrdvX7e2asKVhjne+yUi0VrocMNF2pzmmvjwlyOl/kQfkpp05+rvK1snxXvw5rq6CBOzo8hryXdunDjwwf9fLLeoTHrHx51YbipriZttMTw6JiHEI+OJ+26+/dZVSIS1sotL3K+KGRty3ffKebeztpx/caDNmIJq7WkXTaxdoSy2j2Do2ZHBuxI/4g11CTco9pSX+0HuBZjfU93P6hUd8/ubjt37dIpRRjKYSxmWx8hW2hptlDlzpZ6e85ZK61vaNTv4x92dk/mXuNtbK6rnjaOF4OMWbLKFOF9l1566eTj8fFxu/baa+2SSy5Jq0w1Nzfbu971Lrv11ltt1apV9ta3vtWe/OQn5/WdxawiEr47XRsoFdmcsmHymOd5fepPIVyyvs3Wd1TmppUrxwZHrG94zN3Uk5vbkQF/nhC4wM1sEg8dtN7hMV84Ljt9hV1SJO9VtnGzUPd6PsfOTGN/seIhmaev8HHIGA3jMD5OZztm2IyvT9mMecy9ns09nqvxkut8Xah+nUtyXWNyaW+2zwp9SC7Epp291lBT5Qdxcn5M6r0+0Dtk1929xxWoFS119rstkWHzieuWe74lr3/DE5u8PT+8d5/du7fHCw0gZOEBL3TscC0L3f8LQb5rW6H7TjH29sm1g8NaVzTbjoM9/riloca9pn90Urv9YcdRH3+0C2N4bVXCli+ps47musk+WEjlJfV+3L+vx/Z0D9qPH9hvt2w/Yscta7BdRwdLYt+fzfr4uy1dbjQJ1/HYE5Z5VEq2Pk4dQ3DN77Z5JeD2jrppudfpxnGu47AQmWa+yKcNJatMpUI+1AMPPGDf+c530ipTg4OD9qQnPcne8IY32PXXX+8FKb75zW966F+utLe3WLFJ14aOjhZ78WjSN7K9fSN+OjSHmlGLH0shz/cOjlpzQ40/z4GchcD3LGZGamusvfWAHRtN2srWett/bNDaWxts/dpl1rG00V+zt3vAfrPtqNXV19oZK1r8NTy+8JROWz3xmmKQ79id63s9X2Mn29ivZBhnR/pGbHlTbcZxdWVHi4+7mV5X6JghPAer8obOZquuSljjknrbcrDXEvV1s77fczFecpmvhTDbfp0t+a4x2do702eFPjzUN2KjyaTV1lRbba3Zhs5WO9AzOOVeMx7YZzasaLHDfcNWV1Pl0kZVXa2duLxucmxceVGHtbY22jW/2WKndDZba2Od58bM1dipFPJZ22a77yx0n0+uHSuiteOEFS12y5Yu+9qde13BRlZ52tmr7I2drXbbti773eZD7pla0VI/L7JNLutq/H5sOTpo+3uH7cSOJjvv+OW27VCv/eTBg3bmmlY7Y+2yktn3s11PuvWxsaHG7jvQbw0T4+ieXUfsC7fstHUdTZN9n6mP42PogT3HpuwNj1mXtNu3HbFdPcO2ZnljxnGczzgsBXk8H2rKRZH6yle+Yh/72Mfs1FMpUj+VN7/5zfbKV75ysuDE6aefbvfff79961vfykuZOnSop2gnLqMBM3gyteGM5Q3WduFa1+qprlI9Mmq/vne3fe/uvUYOZ2t9te090m//8dst1laTWJSepdmCveWP1i1z69+DuwcnrU+1I6OTB8ht5YC6Y4Nu7RkaHLbWmoRbe7buPuqvK7VxM1cUM7whPvbD95fSgX5z3Tf5WKAZs511CbPYGJ2rMUOeA+E52w/0TFrOaxPR8wvZ/5n6N5f5WiiF9OtcUcgaE9q7/3Cfvy701UyfFfoQTxK5DjWj437o9KFjA9Pv9fCwC7V4GZbUVfuhqzA+PGLb+4amvL69tsraG2qsq3vQqsbGizZ2Sp1c17ZS23dm4uChHusfGLaNe47aaWuW+e/93YO2rKHGVrdEXu5f3rfPTn5igz3n1BV20erWKX1w/9YuDydFtlk9sfbweDayTS7rargfhPb9aHjUTm1f4v1dk0z6YbS1yWTJ9P9M15NufTx7VbPdtqPbOhtr7PCxAdt3ZMAGh8dsaW2VDQyM5NzHqXvD+Ei0bjz37Edzr7mHheyLCyXT5NOWilCmPvjBD9o3vvENV6ie+cxnpn0NFf5SK/dt2LDBNm/enNd3ceOKffOytYEEzq1dUb1/Yns5V4RyuCtb6mzjzl6PUafaynfu3G0vOn9t1gFcTrG/C8nFJ7bZurapruj4/WhtqLWmupQchbpqf76YY2c+x24pFOVg7PMDxZ6j89k3Huf+0ERoXTgE8aGDPibnep7ONGbo78tS4tx5zPMLdQ9m6t+Z5muhFHN9LHSNyVjxdYbPCn3IYarkQKAj1Sds2r3m/1icEbgO9gzbcUsbPIzrYN/ItLFRCmOnVEkdW7msbdwvcs4e7uqzNS310TEWJbDvZBuHRwdH3YPJ76aaKmtrqrMTlk3ND6OQSbj+eB/wfG+afLLw+vlcV/n8c9Ystd9vOzI5b2gLIbAoLu3jyaLv+7leT+r6CIT48XrqfvQMjVpLfY011lXbktrqnPs43fz+41iO5c1bZ78vloI8ng8lrUx96lOfsv/8z/+0f/3Xf7VnPetZGV/3N3/zNx5ze/XVV08+99BDD6X1YpUzqQmFlME9QLWcnkFPAsZSiPXkxw8csF3dgxkLC5SCcFzKZCsmsBgSKecyyb+Slfb5KIBQagU34gUA8IiT78B1L1TeRS79m2vxj1zHIuvjD+7bZ90Do7a0sWZWBVoKoZA1ZrYVX3mMAW6ms9EIAcJynama31wUj7h/7zHPVVmztCGv8u4LwWzXs0L3XkpNE+6GN+ThA312YlujveKi40puTaV/mDvIIqd1Ntvh/mFLVFfbs09rt189fCjnQimsNXNZCCbfdTXdHPzjs1fazqOD87bv5zO28rme1PUxXBfV+GqqqzxPDUUq3z7ONL/3V2BVxLJWpjg36t/+7d88B+rCCy+0gwcPTqn0x+OWlhZraGiwpz3taX6Y78UXX2wXXHCBXXfddXb77bfbBz7wAask0pXB5TRqFqyqkXG3mlBelFKjLGbpBvBiHehzSSVXmMp10eY8kJmuv9KV9vlQfEqpmlyAa0GYW+h7OZf9m+tYZH382h92uTGqOmG295hZ7+CueV8fUwWpfNeYuar4GgQv2rPpQG/Gal9xy3Uun5crX75lu/3wvv02ODJuDbVVLsC+5uITrRSY7XpW6N4b3tfRXO+HMe87Nuj5aqEqWymBd5Oy5/XVCdt7bMjWtzd6Tk1NVZWHgeH5zKSMpPbv8csa5kx5KWRdzVQmfD72/XzH1mz2ifh1cWTBbTuPFtzH4bXdAyOTj0vNIGiLXZn6xS9+YWNjY/aZz3zGf+Js3LjRi03giXrhC19ol19+uf393/+9v27Pnj12yimneFn14447ziqJdBMICxVhflRtIcmXnGDijDmF+lDf8LQBXKoDvdw8GAtdCruUxhwlka+7b5+fZJ9p4V8MSvt8KD6l6Pks1r2cq/6Nt59KYQijWM/TtR8jQaiKx+GUePp5zPPzda2ZBKl81piZ+iqfzyqWEQSPFIoUrF1ab4f6R/zxY09YviAeqmx70FzMgUL33tT3Ndc1Fbxnz+c+y2ejLFFgAg92Mpn0kDIKErBfjI2TkmH22BOmn2OWrn9RpDh8ms/Kp73prjHXdTX1vemMC6d2Ns95v6VeO+sTclymsx9nu0+E6+JaONOrkDFBu0No8HhMFogfj1AqBsFFrUzhkeInEyhUcf7kT/7EfyqZdBOIEBT44s077OjAiI0lx60qkfBYWIQBLA/xyV+Klu9K92BU0phjMySMmZKoYfzEhYqwGR3qHy5JpX0umS/Fp9Q8n2nPX9nfazdu7rK1yxoLOux5Ifs3tJ9zlDbuinJLh8aSLggQ2jYFBvdknH4yzXOlqajOVV9la898lycntA+PFIoU46x9Sa3t7h7y5wtVpvIJ7cy2B83FYcSF7r1ztWfP9z5LHyFU44HaenjAC5r0DY/6784ldZNtp4Q2ylTqe9PtFyhS+Sgv2a5xpnU103vnst/SjcfUax8eG3PvHmHGfvRBhu+bzT6RTmnMhxAKTTtRns9c2ez3+dt37bE/u/iEkjMILmplSqQf+Jlcz+3NdXbCSKPtPjrgFiAOSuxoqnUXbvw08VKzfC8GD0a5Ex9zeDt/cO8+3+xSlaR4KBhKFx6sUlHa47H4+Vo6F0rxme0GN1+kCnNsoHhqONCSakeE4/yvS06cFwPIXPQv72M8YiUniZz7j9EJi2qqhRzFEG//7u5BL9U8lkz6Y56fDyF+LiMF5qKvsrVnrpSpTH1DjhShfXikUKT4zWOeL4RsQnC8DTDTHhSfA7keRpxuPhey987Fnr0Q+2zoI75jdSsH3vb58wd7hmxNy/T9Iv69c6Ew5nKNmdbVTO/FID1X/ZZpPMavnSqZrFMoKBvaGr3QSLbvK1QRmotwVVJJCOdkPb1vX4/VVSWsd3jcvnTLTvuT81fbG56wbsa1qNwikrIhZarEyRYCkmoROqMzsg4sqa3yqn4nti2xnsGxaQsX72eRKIUk31INOxRTiYc7pNv0UFJSN50eG7XB0bGiK+1hDpFfiDKIoEbuQba8mXwW+LlQfErZOxsX5vBI7Tw64BvpeDJpFMfeeKDPPePzZQCZbf/y3guPW2r37enxA2nJN8GSStW61Nw/fkjsTy1AkUkASzdOMt3LdK9n3lCVlWJC5MDO1ugw276KC3Z4X/b0DHnI+FwZQbKNc/YhcqQI7cMjFXKmCtmfsgnWqfl/jIWZ9qAwB9wav6fHaqofPYw4nbCb6ToLVXhnqygvxD4b76NNB/s9d/v0FU22vXvIFYRljXUun6Qb33OhMM7mGjO9Fxkp389MN89nUvTCtW891G+jY0kfWy0+F5M5X0Mu+9Zchquuam3wvDiMC90DVPSs8Zz92urEZPGbuFcxtX2lvOcVgpSpEiYMfBaglvpqLzCRbuAHy+veniGrSZgrUl7Gcyj9wpU6iMkLKNYgnq1Faj4sG5VkLZlvwRqh4hmnrnDrVOqmw7h97lmrrL2prmh9GZ9DfUPReSD8n598hKDF7p0Nwhznr1z7h10+PxCy2ThZl3YcTZ9XVCpzCQ8Unqih0fHJ0tLHeofS5v7lIrhmU5jS3csj/SMeJRB/PfA3jrdAye/qHZpU8ovVV2F+c4//cHjALBGVXkYBma1nKpdxTrEJcqRma+jLWDinq89u2DS1DYwL9s+Z9iDuL8YDlOz17UustaHGxsenC7szXWc6hTeXeVKIojxf1fEy4X2UjPoIzwoKwdLmBrtt22FXFELYWrrrmK3COBtZItN7GYP5fGamdSGMR3I2jwyMuAeqq/fRnPZw7ayhrEl1NdU+tnK9hlz3rblQqkNfsYdSYOSu3cfci99YU2Unr2iyNa0N0z4ztX2PPX6Zr4fFCCeeL6RMlTAMxh2H+33QjownrbYq4ZMwdeCHsqlY3odGCLFKuADLaxm08UorpSa4zcYiNR+Cb7lbS1LDV+ZDkKU/2BBu2NjlVnUWRYSudJvOfOXT5ErYPFoaqj0sB68UwisFBnqGpnptizU3iuGd3ds94AeBck5Krt/B685du9RaJ5KjSTAfGTPP0axiACTTz6VDXoI3UrqfdebKebmemYRRnvujk9p9zO7sHvS1MVvuXzbBNds4SZtfdqDXBXgMDsEohvXeJr4frxmeKQQSEu6LXQ6c62hpqLFTOxFsGiYND+s7mqyjI7cDLNOBIrPv2JAL2dnGOdc/2z5IDZ0KFfAYpunmGkURCN2baQ/aMCHs0SesIemE3Xzn83ycVeeV2g712W07js55dbyZ5hprPn3kZ2HVJ62+ttrOWd1iV3Kg6wxr6Ww8q7ORJdK997EnLPM5y2/6cabPzLYu0FeEvd+0tderhI4lzc9pi4+bcO1VE0aWQo9FYC359l17PfoodR7NRThlvK9Gx8xOXdHsRyWsbK13RSr1M9P1C+sh+/Hpnc3zFk680EiZKmGwJhE3DiGGfGB03J/PVjYVQedPzl9j3YMjvgj8z+ZDk5OSMwXiCz0bDRaj+axWVUgeWC7vn2vBt9QUzXyJb8ok/ibMbEldzZwrhfQT4wrLbFiQeRysTcUO60u3eeClJQ6dOYQFDYF2tkLQXLdxofLLbt522H6z7agdOsb5dPmNDfrhOWettEe6+rxPE4TNVSds7bL6KXlFYS7tPzZkR/qHXXHdcqjfhdlnnTG3ClUuwiivYYyOjo27YndKR5OHJ6bL/Ss0JCisX6n3krLQB3sHPXwnGMVINlvWWGsnrH70mIuQcF9sQtj4yR1NE9XjosM8g1GuEOh/8osIEd17bNBD67C+z9c4DwIfZe6pxkil25XN9W6cTDfX8FzOdMZW/HOzCbv5zOd8BOFciBsw8Bavxbu3qiWv6njZlKVc5tqUPjoyYO2tDfbH56zyA2Tnm9l4t6aVDEd2ejiSndjbWN+yfeZM68KjtiZ25mTGmjazORZhz7FB2364344OjNqXbtnhcmD8/sxFOGW6Nm6bCJ1N95np+oXDnDnjKt0cYfwdGE5acoi/lb7cFZAyVcKw6OFhIjwJazqhe8SlxjfcTGVTma8sBqmKAYvpZCLt2Jjdu+eYDY8l7dt37naLyEJ4YTJZrHNVonKtFjdTqduZquqUU/5WfFNe0VRrmw72+vNPWN+WMaStUDL1E5tNoWVWs11XarGVfPOZwubB3BkcjSzV/MxGCJpL5mqDywX67/qHDlpdfa2HbxViMHBlKGl27e277MCxIautqXIPCxtqfC4xv1GkWIs4voHvxjN0wXGPFsQJbSp0zORiAIm/5vSVzf6ah7v6cgrtSke2cZLewr3Uvnln3xSjmAv3LXU5H2K6kGGScz0PQv8zRsgFIX/mnj09dvaaloz5aHNx/YyB5oYa62iqs57BEevqG7Jv3rnHnnZyux3hANwMcy0eyVGIsJttPqdeTz6CcF57QHOdGy8IHyXsP9fqeDMV7cjV2Bj6aOuhPlu6tNHaaxfOSDBb7xakhqDx+PjljVnHRrZ54zlFdTX2xHXL7ejgqHv1MUZlki3yuYaQ5kERiAM9g54Lutxzl6qm3J8w9nhMPtNs9tXUNq5sqc94Jle6EFMKpqXz+LGHXL/xoI1YwlA/LyujyCApUyUMg/KE5Y1RSMGEdR0hML6pZZrACDuZSo2GJNHbdxy1wdHxScsjVjwsYnNV7SxbImYhFuv4Qj9Ttbhsm0IuVXVme67NQgtA8U2ZELxqyqwZYVjjc64UziRM5vMd+VhBQ5hKvuEwceEnXTW/eBuKVelyIcqhc53kPHX1DdsFK1psaHC44LGB0vybLYdsVUv9lFCwsHFzDRhKmN8oDQh0rC2EhWaLpc/Xg5qLASTTa3IN7cpX+U29l/y+YVPXNKPYE9e3Zf3+YoUcz7Vyn9r/FCIgGuK5Z6/Kej1zMTYYl4MjY+4FwxO4v2fYvSWUb05dB/L5vpnWuXTzOd3nhzN58EihSLEfpxOE8+3n/pEoz5ojUgZGRq13yKbtZekMVdmUpXyNjQjGN6QRjEsljzIT6a7z9p1HvUod0Q25eOQI7cUj/YzTHq0WyvsoGR/kHtZHIoJme24V/dwzOOrRAvzGeHDSRO4Sec2s+RhvUnM2Z5KH8mVlbE6kfiZjmrGCRwpFKnwHhrV4WsI1N23z8behs9m2H+gpq8ggKVMlTHxyUpUv3aaWuvGhZBCDTwhWpqprhPqtb1tit2w7Yi111dbRUu/VuVgAPvu7bW5Bme3mnS0RM1eL9UzhEBTOwFUcKgTFrX8oi1xTqqAHuVTVyVeImIxTnzhRfKEFoNQqXJR0BjblufawzJWwlS4pNYRSQOr9psrXaZ3NBYVgZhJ+0o3TXEq6lppFNR9vMKFW1TVH7JS2JQWPjWyhYOE68DhjKGFuoEjhZQ/FSOYqrDZYZfE0hcISqdeTSfnPJbQrk+CXi3ci/lw6o1j8+4OSz/fxPg6wJdyLAh/FCDmey4qvqf1PP5AXQf5MJubi+l2hr6IQ06gtX1LrRqaGmoSHW6Z6aMKeES9Qku37eD2CsM1wsGo672j8evAQsOZwrXikEDqDIFxoYYDQz+zzpAUcxBPQFAmw2RTH1BSAVGUp0zxi7IYDbdNdb1wwTleIJd2B78VUtlKvk70HbzLzgT5i/mbzyHGNnhM0Nu6el+UTqQzsb9f8frs/H9ZD/p5N7pmJ0M8rmuv959YdR7xIytL6Gj/CAm/n9+7Z669jTJ29Ogr5zFUemk2bkhOfSTvu3NXtn9dUV+MeqVSlGlD62J+IHqgus8ggkDJV4uRisQ6vCadR37aj2y2eqQmnPGZiETeOZRDrSB9VzkbGPHeCAxNnO6l8k+nq8/h4wjrSJWLmYrFOJTUcgms4MjA6YeFdPuW8GPqBCcwZCJTupOIMiZLBRZ9twyjEQzApqPYNexGQdItWPgpZIZtIXME52DfisfL4pqgYNB8eltl6UlLDEu/b1+sby4b2Jb7JpJYrbp4YnwijcxWCmU3AQdCKn3pfDot5JlLD3Ej83Xl4wMZGx6cJWdk+I36vc/HierGJhE0WKkn9rrkIqw3Fd/A2PHygz8+EorR5NoNT6nzI9F0zWWxzVX5nMoqllupmnWb9ispJR6WGcxWucwltzuWstdQoAAx0l57cUVABinyNL3w3h38Wcv2p34tn4P59x2zroQEjOr6httoNTanGA98z9vR4hUqKZKxva/RwqXTfR/tCLhbKVBhz8bGRLpwPQZHwO7xG8bEeFFdC+7gvqUn8ue4Lqf1MQYArz1o5Ldcn07oXTwFIN6fT3ccgU2SqXsfnV6cUYqHoTSYZo9gFoEJfx/N/WbvIC+VvnD8Xeadq0o4NzyfeST5x7bTCNtwHjCqMgcbaGj/CJp6LWMheikLv43WiuuT5a1s9hJYzvkiFQA7AcMH5owd6aX+tLWuocRkhF3koWx9lamt3iocUwzkK5HHLGvwMVPZ55k1QqkN+N99Pnh/702NPqi/62ZT5ImWqREl34N9Mwh0KVLw6VTzhlA2URQ+rIKEmQ2PjPoD5GRiJilqw0axrX1KQYEPbgjLHIX27OLV+VYtP5PhnIaTOZLFORxDesBKhSGFxY6Pl/fET1flM2sCCx3UTm8ymjBAZPn8mITAXISkumMTj1B8+0OtFQFKve6bPm4tNJFXBgfm08M3GkxIWXApsIewc6BnyUITxtkbvy9RyxRSM4NwZrILtS3IvGZtLG9JtJKkCbi73o9gW1VyvE0V/y9FBu/yUdjtnzdKCx2YuAjKhu/FQjvjfZxtWm1p8Z9vhfn+eNWG2yn8+XrNc7num70/ncccDS0EKvBSsc5sP9nmlv5n6JpfQZtZOPxS3qc4Fu3RnYEFoE/OT9RXl7o5d3fbK0aSdsTz/Q3TTrU3p9rLQH6zd+V5/Ohh7JyxrtN3HBq02kbBEVcL3vDiTe4YfQJqI9oz9U/eMVA8W+xsGHkDA9kiIsXE3+BDKtKmrb8p9wCsWHR/Q796xpvpaW700MkoAnj9ypFKPnMhnHaJteE6yFZkIob7BAzBZcXJ/r3shgxKRLkwt9T4GmSJT9bowtxuXRIIxn4dQnanoS7ELQE2LlDhhmXtP9/cM2sf+Z4uHzYecR3Jv44XActlTeH3kKWUPe3S9KzSaxYu63LfPdh6ZKOqyqtnPPWN9f8zapfbbrYe9Wh4Ga6L+UbqGRsatmoJBsUqC+Zwrl4ucsjR276lciOGcNTkokHGlOjW/GxmAcXj/nm5rra3y0NBS2kuzIWWqBEk3YCHbIM40iUM4AxuXl4iur/ZQBg7Uq0qMe1ljLAFMoBOXL8la8jVbe/0ww709vmCwxfBdboEw80kT/6yZLNZxUvNZsFjikUKROm5Zo3sqCGMIC3IIP8KzQXwyoX6EdWBZDZ8/2xC1/35g/4QbP+mbHsopn/9IV6Tk8Ri39urWBrcOztSH+YaY5KPgzNVCNNeKQgjPQmiBSHAjzGnYTlwWhUSl5rRwgOdclPaNtyFT2Eoum3q8TxB6uIcHe4Zd6XvOmSvnrQx4rmRLAF4xUeZ8pmpJ2cKtclVQMindsw0XTfVWI8QSKkWo8lNObp/irc7Wjpk+O5txabZ5NqkeC9YzBHLyHihP/sjBPl/v1owls/ZNNkEU+H84Xw3I3wrhzwj68RLawSuMoLPxAAa6yBjH+nTd3Xus7cK1BVXZCtefKXcotbAQYw4hnxw/9ojnn5O9WEU6+EwUjFM6myc9cvHzfcJrfM9Y1eweLA4hZU8Me0Z8nvObM5QQEim1jnJAnz+wr8eLOaH0DYyM2fHLGu1JG9r8PoQy+AiwKLP7eobNeobd2xEv2pJ65AQ5ieTAEL4V7imfNZ4mtDBdn6bm4kyJoJjwACB4s28T+vutO/fYipY634O4ztQwtdT7GGSKdHOE76YNFBPYcrDXRkfH7KzVzfbwwb6MxpO58FRnYyavber8CSF4gBIVDNFUgyVM99jQ6DSDQKY9JZSp7x4YdmWMMY0xIxRhyFeBnFbUZf9EUZfVUVEX3v/ggd7JowE86siSftB6tT1qUMh2rlwufZSurSvjUTITRcb8qJ7aqmlKdWp+NwZ4PJkvesxxtr61TtX8ROHEBywHvOHloAIRAlq6sLnUSYxlE2UISz7CKpsTnxn+zvMoTkcHkj6Ily2pc4vaqSua7PLTO3M6TyFde1FaUKJ6J6wgK5prfSLdv6/HLjhu6WTVprCgsUhlslhn2yBIHCYR9NjgiO06OmAP7Y/yprDusICH66T/HrO2ddLSEjxXM1mJZxIM//vB/RNxz0lXTPmuQ30jLnQcmPguvpzFCwGPMIuZ+jCfEJNiMB+hF1wXwgobOeOGBRRBkoMKuWcki6fLaZlLpS6TMI/AlSkkJ104CvNs15EBO9g37IIUghnJwPGiKgvttcpUvIProJ+fe94ab8dEal3B4Vaz8U7ONlw0vuaF5P26KvNQ26/ctsu9Daw7hYzVXLxmmYwg+RTxQcia9FjUVvnaiWCBsEV+IML56rFx+7OLj5+Wt5Qq5GcSRCEY0kbiZ61hiOoZduGdEKH4IbZUen24a9gFyMbaKi/gsHppgx0bHPXPLFTISSeQIcihODJv4oWFEPy4p+xTjL1cPz9dOCrXET9vKl1OHW1ir8LDybedvrIlbV7n0sYa23pozJUSfBNjY+NGnAf92t5UazuPjLqngHvIfbh3b49/D+drcV0dE5Ei3GvGT7zaGvsv14oCSIj6jsMDLtxyT7knrJcoc+Hw22wHRWesaNnZ7B4APGr83n6k38t0Hx0YtgO9g3bXrqSdt3apndyRXbifaY7QNs4mu3Nfn/3moX22cX+f39seixT5VBljrgpAFbKHZZs/Xgis7VFDMzIUxoh0B36n21Pc2zehMF14/LKJM+XGJz2IlF/PV4FMbS8K75bDA/bcsx5d70I77trV7ftYfU3CqqurPJQehQaZic9hvUp3rlzqPc9V2d2f4iENimToD7ydPM6U340h6cJ1y612hKqHVjZImSoxJsOfqs027up1KxmTt7W+2i3J8TKnYTKEjQOBiRARBHk2ZAblD+7ZNznZw+Ra2lDjSbkspCgFFKN46QVrp1VXyUWwCe1lErIJsQESYpdIJPz9CMVhgucilKcLn5uaz7LOJ+M1NwWFJkoKxU1OhbEpVpG+Ef/+dEphqhCYa9vcajg27u9lgQpCCqGSWJBZ1I5b3eobLspk/OyddOQTYlIM5iLcKTWEKPwfRckFNzRHFyz7XTBFqcyU0zJb4T0XYR6FOQi43FvGV9zDmNonVHpCeUKpx4qfrIrK3v74/v0+n4LXCiGIcVGogJ9Ktv5OvWfxkN9lS2rtrPUd1tXVk/Wz04VbYSmvroqEwbki7iGIP57pOsNcD8n7zXVVNkw+Q23kXWRczcbDO5PXLJ0RZH/vsOe+IKzkEpaFUIHHYueRfjcO4XVgDGGcYY1GiLv8tFXTFKl0Qn42QTQY0hgUe3uGorPWJiqK4QWJh14xJ9kbqLhKNTiUrsccv9SVh0R11Le5kMvxEwhTWMQR5ChmQrsRto8OjNhm5lRVws49fqkrc+kUhPjnhwiJ1HkWP28q5DjFLe/xe02/c50oRl+7fZfn4xFGGi+Rfd6aVhdQWba4BuZIVIUPb3u0BrBPH8PjlURJiMQsjEQYi1C+EFjxwu/pjsLjX3T+Wr8WDAEI6niNaOzQ6JhtPTJgSwdH7O5dx1wBC0aQ0B+hTzsmLP0ojem8b/F+Dx4AjKisdeTRIFQf6hty4yC660zCfa6e5Xt2H7X6WPrB4OiYywRx79pkrlIeB+TOtNdkCqXNZoye6bgDQtZQ+LMd+J2umidnfYa+j58pV6gCmfo+DDle1CUmb4RcvKtv2OSPGa8YoTFwIyuS144y6PnqE9E16YoJZfpOfscN9iszeJ3ThXozV4KMhpzKeoNBkkp/l52+wlYvbcy6P5UiUqZKjMnwp31ReAUTbjw5anuOsUhwgFyUY8QGHiZDiO9FYMKayZmQ9+095psnoRrBYooiEs4XYEPETZ1ajShfYTVu+VvXtsQVCIQYLHsoWJ0t0QQPCxqva2mIKlplSz5lM+GgS7xNcQWSthPHTBy8J3JyZlDt1MmfrhR2mOyzURg89GE8KphBm/heNkQEgeeds8p+cO9+V4hYLP0shRnywFJDTDYd6LOhkTEbxYIVC0ssJrMNdwoCDnl0CBz0GcpJeA0CT3gfCiTXnRqeNd/Ex3xcwEWo8tLCo+NTPIyp4WUk9g6zKSWjRGVCZ6sSkXcSq9zX/7DLrcBY2N1i3bvdLdqM40KvM5vyP1PIb+T9yE78MxjThJvt742Ea8qhkyuBsD8XSmEmIXim64R48j7jCwUAEL5Zf9igyRHBEDWT0JWP1yydEQQPBEIWwgFrwEwHsIY+PrFtiecJeI6nmbU11drwaNLGkubhiumMOpNr6YSShJCfTRANoTwITZ4jWxdVuSMclfdOqVw24RU7a2Wzn11zuH/Edh0Z9Hu/cmmDfe/uva5UF1LWPFUgQ8Fgs+BehbHKdV10/FI3rODNaWmsdSUkvu6kUyZ/8sB+z2VCEdh7zKx3cJffv3De1ClUycsQRh0fR+H+USESBQyFNj6P2pbU2akrml3JxGjIXvW7LYft0MCIrW6pd+t/grHYM2w1NVU+noExzrX1Dkdh6nj8GLOMI9Y89ir6fsp5ZMmkbTvU74Wi8HpjMEDhixdVok/x+Ny0tdevnXFzXCwnJpMgzP60rq1pIswKkm4oYf3iLKTjxmfOTZ2p6iPtQyFdnZInxXen2/f5LhQ8+jhTFclse02659Mp8FsmjNHxaIds8ye+FnCPfnDvvqwHfqfKUbkoavlEBPF32otx93D/o9UaIR56iCf52MCYtS2p9XnPmaLcD/4flMEw5/kdZJd0VRpT2xoKRwSD/WMn8u0yyVHxa0otmsa8wADFNV1SJudKpSJlqsQI4U937z7m2roXUphImkWA7x0ecuGNBZ8whDBog8cEgRTrFDlR7FLD48lpscxzKajGJxiTge8fGU1aa2Ok9IWFgUkZt7qlVsRJVWhIdqbNWNiYzHwWillY1LFgsGkgnGcqIpEpeTfbwYnZFIbQDu4Liw+fgzeCROFLT+rwPs93UQybXBACuV98Dn1TKIWGlKV7X67hTumUUTZZBDgszCyWCEds9o/3g4TNQ8gI28Rjkk8J5vkMmQtjAcst840zWlI9jKnhZVx3Q7UZDmEslxhBUPLJPzjSN2LbJ0Lr+KHtD+3vsy/+foed0PZoAYB8mEn5n4twmfhnUGYXazfCPt4DBO34d3b1DhVcQptruTZF2ewZjM67OzYRSpMtvDmevI+wSljUWDLhxgmEUJTdHz+w327efmRGoSsdmYxL6fJsCJdGCEYgyeUA1tDHhJ5hrUdwZq3nc1F0+DyENtaD+LrlOS+H+31OEbbHe7g/RBdkiiygzwjlOWdNq4f0IWQhnPM9CK+bDvZNVC5LTl4Da+HFJy63u/ccc2Vx9dJmO33tshnPf5lpfMYFMoyChLGl5uqesarFcz6iUvdThfp0n889xpNM1AVhfBzS+8ihfjdmkOxPn54yUcafz0u3vjO+2cPCtaN4USGS+9Nc1zTZBsY5+w/fH+YX7UBR3d0dreOez9tQM014/O5du+3HDxzwe4oixvgJShHfH8/NYY+ssoSHV0aHsUbhhIy11KJKj0ZDRfknqdFRmYR22nTig40+/wgpZQxyLYyRXCsvxudSqoGF9tEPeL+oGIwRNb4Wpdv3b9p62Dqb6m3t8oZpXvxse02mMYdSQN+xXrP/EPnBOP/2Xbvtd1sO2eGJuRwUgtTqh6lkOnYm0/o6k8JUaAVhFD9kPgw6tDuc0zQl136iylhTXY0tb6yyY0PR/sYeFZRBX6+Ocv7ViK8JyDiZqjSGtsYrNoc+uIE88vGkh5HmGrIYiqa1N9a6YeU3jxyyxxy/rKCqocVGylQJ0lQX3RasTYBVCq8/Ma/NtdXWMxFTbtYTJbsubbDD/dGkip5PujUd4aSuKjGnsceFVpFLZ3XrGx5yQTs15t+VxqFR31wIRcBijQXvOTHvwEzWnEyLbmqydfzgxJkWx/iiSLvWLG30kMOQF1PIougWptj5E7i/Z3P+RKH5TZneF64ZQRXLe/Ac5BJL/eC+HtvS1R9VEEqYe0RQgPGaUpSEjfVff/WIC4JNE2ebzeTtKOT6MoUZpuvbVEWEwy5TPYzhnn3/3n0eEtPRVGtryUniUEISaqsS7jmlnzwgasLwi7cBiz60NlZ7O+I5E7kyk/KfbfOmLwj1GKmtsWyrQWpi8giCXDLpAlH8O//j1u1207aj7oUj54ciIa+5+MScr4VNOa5sMiY27u+1j/96i68Z3IcL1rbOuDmnWjp7B8dcocHDyOaei9BViEc+5NkgcHuIXm1Vzgewhj7+4s077HDfiHukRyxp/cMYaRLW2Vg7+V4EHQoSsBayviP8Un01rKWs9/TXTMpfOBdsT/dRF8hRFhB2n3ZKuytjoULblLWwrsbXJoSuXM5/mWl8pq6TwegVH6soyKwF6cZwusIHVDRjfGIt7+rDAj/u1/vQvh674qxVOa3v6UKnCAlEOZupbX9+6XofW6x5v9rcZY211WnzTwjnw3sWPwMRu1loSzw3h3BM9n5CMvEYrWmlKt6IryMoSyF6gf7gHj1x3fLIGFeVmCx9nc4bkLr+UdY91TOcbR+bKRQ/ngPGWsP6efPmLveGMK7/aEP6XCUKOD24v8evj+cO9A35ehmfN5nGFsac1OepGofySn+jnJLTPEDfNtTYmtY6/74/7Oj2M8jOWd3i1xHSBYB+DZEtIe8nXQ5qLobTfM+lS9ff8bzhyfy3ldHaxtoAPBe81bwGQyXeUq9mOZF+QQQRxoAwzsnDwxPJBo2X/dSOJq9Gmc3DFNaTuHfucB/FJqpyVjInU1qqzL2tzAkMrqzhhKGXG1KmSgwmClY2NgZcsmzIrkhVJ3zQdQ+N+mKJUEqCK8oBZ0ZRGeZZp3d6qAOhCSOj4zZWlbCHDvZNlsDNJ/Y4W/sy5S+kPo6/h8WOSU6eEwsbYRssbOSVkHMSj/nHqca1YQ1lgjfURnG8UZnS6PuZ2CFkMV3b0y26VIbCg8eCEQ81DAcn5uJVmimEMN8wSQhhlgiTSxtq3etYSBWjQkvLzrYkbSZPCN41aggxlhEuwlhGQGDRba6vsq6+IbO+hG9mmRJfZ9POuPKFcQIhpCnLodS5hF24VXDnUVcO62oSVluTsOOWNkbFJ4ziK/W+WQFjFss7Sd5HRqM5zX3eeWTQx/fQWHIyZyL1WjON7Vw8T+k278m+GB6z9tYD9kfrltnFJ2ZWRIM3gzBWrLF37T7mlkQUfgRNEuJv3NLjAvbapfUu1JOz+dgTlufuoQrnM0w8YE5x7h3VxRqqo1A0rNXhO2eyANOPhEzd+EiX/ezBgx5CN0XB398zeS7LbCqGZQuzyecA1mDIQRAfHMEbZT6GCNVGeeEHofBbd+52gQcD2cDo2OSZa6ylWJmZa6xFmYiPGZSxB/f3+vilPDfe1F8+fMieekp6RSGeMB7KXGe7D/l6RjMJmpmeT/f5eIFHx8dsy6EBvy5C1RrrqzxUD3JZ39PNfRSNdGFstOsZp9m0MPmH9lMZbzDtGYdhbwihzenOAIzn5uCBYo3Aw8bagheP0HYiNJiPoahS6A/mR6Zclvg1ZjJEpPZzuvmQqZJtpqMl9nYP2u5jQ7a8sdoNGxgafrPlsBudgtJG2xGmI4GaqBXzcGmMcBjjQjhetrHFvYk/T14QkTAYC+prq9wDSP+R24pnnb+xbkWhjQmv/EvBKu5dUMCISMAAjNzC/aD9RCyk5qDmYzidC+No6uHKhC0iA/YPj7hnNe6tZjxy3eyZ7L8oga+95AT/bD4XmQhFkX3Ri4NMzH3WfbyrmXLw0oaNNtelDZXMdAzCZErL/pDSglybsNt3dtve7oGsxr5SRMpUiREstQgPrfU1HvaC65QJP8JGO27WQPncibwkT2hNJv3/xy9vnBIb3tU/4hvjRccvmyyRG7cqMdGw4qZWpMlEfCFl4syUXM33BItOlH8y5lZ8KkHds7fH24pQ058S87/r6KBbKJjEK5vrXEjrGR6zO3cd9XCUXLwSYaLi+QoWQBZ/QpKwWqaGGubjVcoWQljoPae/dhwhjrkqcrMvb8zbk5jq3UO2ipeNz+V96SqBcZ244k9YnV6ByRZCgocGYZBwuZqJzQuFijKp7U31vrlxr/F6oTBzzwu1eGeznJKcTU4BYMHNVn5+plyZ8JkkojO/CPPafTQaVxguHr9u+WQ/vfDcGjtheYOHILHJIehF541EWsRY8tGcifA9uRwWm6twGJ6boogub7Rjo0m7/qGDnjORaWykejPOHBv38rvkG5DszPOELAbrJF6SHUcG3ThC0n4ugkNQNhFeRsaG3bMAjAWEBhRxCrswhqJiDDOHzjI3UcC4ZgTKcP5KlFdyePJcFqpaQWp1t1xIF2YT7lH8AFZCJBHoM53dQh8PjSa9UE7T0no38OzrGXTjA88xjrwYRA/hzhQJqDbrT9oxPx6izj0nvId5k+0a4mMGAZXPRAHGu0XIGeFpKAwoC+nGP/vRD+/bZ3ftOGJLahL2xyne6XTHWGQan9m84OnaHfopPE43/hHMKeDxxZt3+txCgCZ8MVRFzXV9T+c1i4c7Md8h1ZMTPAYznXGY7jvSeY9CGBVrBHsnihV7GHMtXH98/Y0r9tzbeC5LLntTLoJ+aiXbJXU1Pr/S5dpwb/AE0Sd4LI4NMC9GvGBUvDgMhkwE/Fu2H3Hva2RIwIAa9TXGuPjBYJnWvrgRgLUCzy15QswNhHTW6NM6m1yJoo+8+nAiCu3kN49DfiDz1ddnyohXV/m9Z4/iPsTzt0MO6kznf+ZDpvPeMh2u7CGf+3snjz3gefLlgsz0u61HJs/hC5Usw57Hb4xOXG8oQEIf7uuJ9jJXODPk4GW6DxenFDFj/qSGHoaxGK/oy5ggz9XDXsfNw+M763Kr4FkqSJkqNWKWWvIugtB1Unuj7e2hFGsUDY3QzQaKknXumlbrHx73EAMvm9pc5+7+kdExV8zYDH6xqcvWtTW4YMIkYdJS7ABlggHMRMyWLP3NO3Z5yA+CMBMPZSSTMDp5nkVvdJ5FsOiEcqzJZHSBfC8WtlA9BsGKiUjCOJXy8K4Fyyvv+NXDh9wln+4clVQ3ONZ+FKd9x4ZdAUNge+KGNi8zD6mH7+XjoZutJyd1wUSRJHyKClY8j1fu3NUteS/MwVqEVY6Fn4UxXjY+E1FMeXLappgafplNgckkIGAF++ItO3zMsUlSMr86EYVv7j826EIhNxdrJKFKIVypUIt3ahETzvWgHG10noWlzSPMJMTNpNAxRjFmMGdQmFAkeBzKqaNUcOYR/erKyMpm38h5HEpT81y8DH6uYyuf8v7pDuqkOuGDuwezKtqp4yIcCBkqcTG/btjUFR0Cu6TWFSJyf36/7Yi/J1cDw5NPbvdNH4EFZWpodMBPQkFQQJhhnXvm6Z126ckzFybJdP4Kih8rJwrIuWuqXdm6dftRF+42dCxJe65KpvkazxcIYTbxaqIhh4v18n7KYmc5u4XPRCBHucNribDT0lDrCg6GEIxWVNKjPyMiQZ1wQhS0nsHpJaYzEYoF3Lr9iLeZHD/GK/ePz8O6P6NwnUG+SacgUfAo18pqmcrJZ1K80o3/dT1D9ruth93zehwGNM4YZB+cWE9y9QyE16VrK0oUVnrW6HiOX2jLTGcczti/E3/js7j/5Bjz/eyTjLdXXHT8tGIqccUewyrfu3ZZY9pKc4WSrpItxomWOpSnKAdtytESfZFXB0NDkxc7GXPZhL08FIehHzG24uUhn43+pF+5Z4QHEpKGMQ65IB9PJmsdES8oaXfsjvL9WEvIE2Qe0iYKqiSqIu8vHl9kIbx/LcFQnUjYkf5RO2F57WQeLPspRkFCv8O+M5PhK12IXiYjXQhRDsbtcN5bukJCIfSeaqJcD0YVzvDiGnw+11T5e1hTg7GLvD+vAn3o0SrQa5c2TilAEv8VkbCx8SjtgvU+lz1o5QzzJ158KV7RlzHBmKHa5cjYWNmpJ+XV2kXAVEsteQjRKeutDXW2ammjl0rFGswAZtKfsRIhOeFCyE3bjtjDBzmtPqqaxqZSNSGokmB4244jXgbd48k9jHDcWhMczNnj1pveofEpydJh4t+586hde/sutxzhUegbSeATdkUvPjGBxyFpnJAELPLBohPKsT5pfZvdsbvbhTM289TqNmwWJ28/EsWOk+h/OLIkd/WN2LHB2inn3bAAITDFY5lRnrB20L5TVyyxZCLhggfWmXDGCt/p1uKGGvfQPZAm2TITuSoY6UhdfFkwfQEZj0rZE97AIk6oZrYKhOkIFspsZeMztSfdYYLBu5NLAQruewh5gWCpm7S0xv6Oh5E2ojgRIsdqe7ifwykRLOsyhivN5JFJXby9SlH/iP/2wiG+SSRnlUeY7nR3xhCbEsICjwlf9RCR/hFb01rtIT8uYPSPeNluxvwghoxaznsbnVJ2PZ+xlSqQZTvsO/WgThRZjBiZrj91XDD/OprqPeSLAiLh+8mRIrQPRRJFirGDxy7fEEwX5lvrPc8JDwm/2eNRqjwMrqEmq8KYyTt75irz4i5PWLfc/rCzO6oWR7hgXbX3xUkdS3xNohhKOkNSujBRDEqEIaKsxfMz4lUD4yGS2c5uCWFfCOSscygrCJZXnLXS94Jwf+7c2T3pvWMckwf2xieuyyvMKH4trIOci7a7+9Fct0yhmWFeUd761DUt0wpQZFKQ8DrEjTiTSn3f8JQk9d9vO2yf/M1Wb0cIl8z1/KT4Z7MXcG3kGnFMB8YbLOqFVp5MNxcJQcIDhlGPvQklgfuGgs3aRjQE+zKhZN2DI9POOMyVKTnGTY/mxaHwZvKU+97X1efjhEOK892bZppX6SrZIsQ/7dQO748gIHthpqpI+eAxuYAI9RgLuIbTVkbGW+Y1EwojD33J/B/nugciIyCKDsa4fIxdQX6g4AzRBxxXEApOYKhAtsGw1j2wxz0wyBSrWutcVrlrd4+Hz1L4hkqLZKbR73iXaSuvdQNHLKQ32/jMdNZfpoq3QWZBJuAzUTIYT/GzQ8OexbxCKWHPoa18LjIdcgThxZ0tDZNVW+N7N2tY/Hws9nmun5BU7hfr2/KmqHjY+WsabCdr8dEBP2ia8z3/5PzVaQ9wTocXy4md14hCi/KXek5aCHvduL/H28i6+fkbt9ozT23PGoZeakiZKjEYmH+akhD6Rye1+0IVkoOxrKI8sYhTMh3LLfOGhZv/s3CxkAcDP4sHAl9UhW7IaqmesqTWJzUL3JH+YfeCsWhMJjxPFGpgQUcwxuKBtYONvH9o1K1RKHphYoYymfydMBoEjbamuol448iic7B3zNuwpL7G/mhDuwv56YTiuNBMyB+CFeGBVAkM593w/Sw0LDih2heLDoIdnrAgwOLNIxaahFxWCtrMWTCJiRjiZYnElM/IRQgstFpaOOQznnwcznQJGw+LDmVzPbcsxw3wwHDSkkO0pT5r2XiIb5b37z3m3kiUU7w3Gyfu8+rmer/3HCY4U7ItG0H8DBfaTh9HQt6jZa7j1xHayAaK4gsoF2wAWGC5v4SPpCsdni1EJlX4YTNGOWTeYAVFsGKzYCzkas1PJdPp7ijC/B4cHZzy/Lr2JT6XsI4y93aQFJNI+MZH3gDj+dKTov7xsLS+KN/Bx1Rt1eSh06nKay4Wf8YauAcldlAnbVvb3uTneaS7/vhnMS7w4jAHG2uqfVwQ8hU2VIpNkCOFFw6PFIpULkJcOsUXYxBCKGvHLdujErts7ChxVHkKZ3ZlswRn8s5y7aECI5ZWP3A1aa4EeVhUwqYJC/E2xsNEL1jT6spldJgxCdzHPA+LZPlQNdBDFGMhkvGzWwDhmwWTMR4P7QrPpfZZ6p7AvMo1L82NHV19rjBiXGPtYZ/oaB5xJZPKedk+Kz6v4gUogiGNuZrOAMDfwzV194/4eh+iFehzjGuhamtzXY1fF8+H9Td1PtNmckP43Pg4YO/BMIhAiDB6wrIGX4+wuDPus4X0ZiPdOo/CR1siElOiSWgTYyAoxRhFX3J+dDh2vrB+tkwI0Aj0FKEIFTWzJfGzp9CGB/b12ONOWJa2zHV4X3z9mMnDkq6SLWHL7GW/2RztE5ef9qhwjdGF+4TCtYy1cQV7WhRtg7FwaaP5GGZt3HgwMoZGlxblSB+/rNGrGBZCfI3G6x8/cgN+ftM2z6Vk3IfCMRTHYJ1FxmFtxluFdxiFnH4/fvkSe+G5Uyv9pSuEkhoeH1/fkE3Yk1hLQr52KIhDPwaZhT2RENX+oTE7bln9tAI/4Z5FxhaMwVHOETIexjlCdgnBxGjNWo+MQbvSnY/1k/v3e5+z/9IHKGarWxvc4EeaCEeaoAyHQ9tT51E2BXxL7EBy2sG4XFJXNZnaET4reMwpRuVnq42M2b27j1rXsYGsYeilhpSpEiSTwJg6eU9c1uj5C09Y1+ZWVzZsFgBi4RF+ExadVcJaz+LARCXvimKBblF3RSeK2U8kmEhJt7az4eHSR7BlYWEShKRGNquh8aS1TlTaY2Ii9G06GAka56+dEDT293puCkIFAiMuaGKoURJYWNho4tbXVGsb72NB2X100N8Tyq3bwKiHKq4ZS9rj15FjFVmbXWDBRT8y7pXiiH2O3PvjLpCSf8D10BdcQ4gFZnNgYiPc52rJm8lDkgk/5HNvz5TkZAQuwngQctl4wiGxS+qrJzetbJb96zcetBFCfyxpl03kKaUrG5+uGhFtiQRCFmHi2kftQN+wXxNha485blnWZNugHGIJZexh/SMHispTWFP3HktMhsDEr8E35onSwtxjxgrSLBs+ivhn92ybFEYQItMViUjXJ+nCFfGUxNsO+RRcSUfc20a1MGLOEcqx3L/0gjUu9AfDB2OL+0yoGRsoHLcUpXnc59Wgjft9QYkNVkuEw10DA74xpoaI5XqWCuOYPDQg1y0c1En/XHj8Unv6OWtsbWNN2hPmU707KCQE2h63rMHHVeqGiiBOXD79nu1Ax0zfEeYufYWi01hLEZZ66xsacWGGfkIZ/skD+9y4ks1Tkck7i8Uei+2du7rdeNQ7PG7Ibh4Chuea5OfxqUJ3vI3xMFHODsJ6zD39zeYuN84gpUQFJCILVjyvIW5wQbmgwFBc4A5jPHUNjI/TQiqFQhgvWH3xHlJ4J6w97BVY8bOFAKcqFaEARdzCzX3mcaoF/Es3b/cwa0LPWItZ6/GKMgZZ7/iNko4Qu2ZiTrDmRRVco+uczA0Zi8qBc0+/fefu6LD0hhpf69l76Pdacm04QLlnaGKvqsoppDcdqWcPMb/5vCdtaPM9JO4l5B6S38x8p/gMjzlol6mFkJjP94WiRnjvyXkZmhBwEY45SiH17J/UJH5eS5/Rj7Q5tcx1MHryefHc1lw8gIx51inuGwYz7mmmcEKvcJs0H+tjlrDlDdWey42iEFWkjIy8GEnCnozni/uIgkCYezolONdw/HSFotKuO3VNk+tkCM8kDJH6QU85uc3OXrPUZYd4kZFcjKrTv6fawwx3HOmPDFoT+dqhCiGyCePF38feMFHNmGIzGJiCF5Q9IqynqTlHzfVR8SoM7Ad7h+ybd+1xoznzBKMOZ3fduOWwK4fMtaGxMS9ugnLDNdbXjtuSiZxt1nx+kLWQD05OU0gnmwK+P3ZeI9d8oAejZtLqa2o9hDP1s/AGMm+pWN1UV2PjGCcmDDJSpsSsSCcwpivdirs0nMmBFYtNhIp/WOWaEZL7o8WEk9nx7rDVM7mY2EuX1LrVoapq3Cc7ccIsLAig7JJMuv37hqJDKSeq3owkkz7xLz2p3QUbFCk2rCjuNukLALlQd+46Zvft6/VqUeev6bSbth9xwaijORJo2YyCtTldqFKwYLI4kEjK4jJIvHZzvVus/uzi412AYxN5tIx1lGzKZF3dWu8WETZfFn0WSq6ZTeQJ69smF1gO9USIu3N3t098JJxcvEz5CjfZkpOvOHOVrW5pmCzuERRYzpjJFHYYt5pv6GyeDL1JV5nQD9PbcTStlQwBk007CCC44rHmbe7q9+s6qZ0iA72+6KfG6nPtbIwImQgNhIQhXBCYShgVFr8QApMa2pRqOSS06d69x9yrw8bB/UFJ+cLNO6aFX6WGFcYX9tRwxZCcnDq3ciHbxo1yc8PERrK/hxC3cVvmgl2fl57lfoWEaCyy9LWHjiBYGpWqouMOgseY+8GmGe4POWYIT3xWsOLRD6mezSBspNvYEUogPIcXAEPHTVvN9vWPpq3mF4o2BO8YMgjXgVISPJ2Mh9SDcGc60DGTBym0jXGCoEbIsW/25E6NJW2Zn/0TVXm6Y2d35G2YyPuKh9cR/hQU5nTeWdrMWOU+kEuK9575wPirTiStoaHGLcbkIAXLcrwfaFM8THSkBmt1g3t3GYdYXmkv4yLk/0ybhycs8zUPYYoxHtbB1PL42Qo0ZBq7mbyVfDbrHsITS/TAMCFa088pyvZZ4d56AYqdR6zKx/BUC3ePcXzF2KQFnPlAWBTXSWj6wV4Onh/0cDBCr7inGJFsq/l6jUc+Ch0b8iMnwnfTZ+S/sJajdOG1xPKO0QaDDXsS84F1P6rOFhUTGCJEvsYKCulN7X/WSAyBKPecuUW7CTuNewm53/GDvI94pd3RtKFRmb5v88FeNwKyv+IlYA0NuTqMxfPWtqQ9+yc1iR9jCUcDcCAz84L3MNZXNEWKJ7AHhvXjGada1tDieH8c7h/yiBf6es/RIV8XuMfpFFbuNdfST1XikTGfv14MYaKQEQZb1uumuhrrbI6ibfDwMq6CEhxfa5hb6Q73zjT2vQBWV597ROOHO2daJ+k7olgwvrL/HhkctV8Hr1tieoViV7YnDqpNZ1SNfw9znnkwOMKRKnWT+dp4m1jrNh0c9N8Y0FhzO5sJI8UgOWSnrIhyy9uXTD8fLZ5zFCquYiRhTwFCeTH24KnHGEDOG3snxqPB0ciQzLhh7WKsDVJtti468oGIqF8+HFU/xmhBMa9QrGemENzu2IHknItK2ltkHExMRhbF5yRGyWMDUZ4c6xVpLU4aY1+pImWqjMilkg2bPkoVGw9bC+WnB0aS7rYl7A6LKQIoEwgBGcsRCx2eoyhRl5CvpT6osYAxwOtrI0WLz8YqdcmJy1y5IW/Aq2K1Nbr1EdhUWOzYLPnBQrave8jjzDmrhoWXiYlVj9fFF4b4BA35VlwXmyzu+KH+pAtBLz5v9aSAHO8PFiMWlGD1RaE6pWOJHR4YdQ9WVVW3bwZYwIIwSkEAvpc+Q5AjvOCvnrIhJ4E7m3CTykzJyc86c6XnNcVzzkKYFxtIaoJ2ptCbdJWr+P0/mw9N8QLgwUNRRbF+YG8UKz5WTW5erQv6LLRU/YoKYoxMHnyKcBO8iY8mz0cCN+9hFER5elWRlX5K6etHSdfGP0wIyyiTjJu+4XH3lv3Lrx6xNz7hxMn48nhYITH7CA3h5HXC0lBGGMuMdcZJIYf8zmR1C+OUscQ45f9jrZHBIWwqXj751Oj6g+UaoYx5h3DLHF3fscSVsI2c91MfneHEZsZmw/iNF2jh2AM8K2x6qWWX8S6EuYAAgnDPYdIINDzHvSNEqr66yuf03qMDdv1DI1PCKFLzgxCQuR7C5NxbWlvtQlu6g3Dj9zTdgY7ZKkCiEKH8MvdQNvGaIwsgFPA4kYgMNBhLWFPi5Y8RUsjnRPjBCkteJAJOqneW/uBesa5sP4zledzHK2OVNY3qeXi/8KrGq4aGfmiqq5kWJsp8QnhB+KGd2L6J4sSbEPIaUsf4dfftjwxdrkwlpq2D6UKB88k9i49V94Tv6fHph+W31c+fiXJXGK94CuJV8mYa906Saxz3PYP9IJROpn8oTIJVm7lL8Q3vkwSh1TU2Pj7kYx8hLyql3zA5V6uPkSfKOhydU0OIWBDiUT75G8WIzuhs8j2MfYmGsCdxbexFWM9Pamvy8DYec7+5xnRVILMJ36lnJ9G+32897AIl4xGlpndwdFq+Gu8PYaQcqEw0RqbQqDjhfmM8YY9kn0a5ZByi/BKmSP9iuAzFB+LnCPG5qUn8jJn25mpXQKZ7VyOjJ9cZ9gw/1DWDh2XKnhxTxvAk9g6P+vhnbyZEO577Gd6HgaSpoc4OdJN3M+T3IoRsIj9wD71oxoRHl78xHwmJZn4zr1lrOLuLqAXywOOFP+LnRqLEs5/yfn9uGEPXkHtB8IiGQjGp5btRysBDt/uiiBbGaabCCamhxvE9cYqH/PhlbiB9pCtSkjEmY2Sg7RyBQh+wTwZ9ob6aMM4xa6mrsXPXLPXQzl8/cnjaodGpVfVCzhGKKDLFuvZG/39LXZV7o5n3yE219VVuMD2Gh9Ki0GWGw8OEh45ExiX68N693XbW6qV24QnL3GDBfsu9CPst159anr0zRZkOxjIPoyTEuo6ooIQrY0cnIoviZx+yl4ScL19DRwddVgo52OWAlKkyY6ZKNgjjlJV85FCvXb+xyyfr8iVVLpAxeeKFCLASsUixkOP9CWFvTHgWFIQy/s+GeMLSBjuhfYkPeKwXk9Wy9vXapoP97iFDyGNDQMBBMCJfhFLEIV+Bg4Vx9zKxCRvEpRtfGFIrpbFpeKWdUQpR1Njy6ir3NuDqpqQnCgCWN6pGBSWEawpKCNeAIhWEOhYENqWEDbmwhPCFEOdnktRElppd3QOevJytIk8hhMWFBStYvxCmT1/Z4nHL/SOjdsbKFmufyLXwZM29x1www2JHEj0hJPHQDP6P9659zOxQT7QxhP4Mil6qpyF48LCIsSF3LKm15U11LmzRD6ta6t2aNZYcdQsRmxzeETZ1+o/wKaz+oTQ+C3nvYKTcjE2UzHUr7fDoZAhMajWmdMoolYIYZyhlngOBoFtlXqCCexQUynhYITDWIsvroyev430Igkcm4TDbPc3V6oZwcc+eft+Q3HsyHuUf8v/Uwih876PeuEh4QFnC80euA9dKSCsbFtfr98cLBUQFWvhcPHdU9xoaieL6H9g3NsW7wHUhXHjVrfFxFxz4TpRKEvzpM+YLcwvFgWTvbFUE8TQgIPMaPss9lGkOwk1VkqJcmlFra6+blusS7+vU6ltcC22jgAfz1AViC9676PDkYAkO5Y/xtofwyVAemL8TPorARFlmxiqKJc+HsT8wQpiyn4Lm4557iCBMnso37+ybcqYMIbAXHb/UIwBYX8K4uWPnUfuve/ZFHqtkdD4UXHB865T+mBriGlXuC23mvTwX7mG6UOD4OUWp6xGkG6uMLwwoCMoUtUkMo1CNuXCE5wflBAPaxv19M4Z7hbyOUIBi456j9sjBPrd445VmLV/b2jDpJfZ8mrFxFyDxzliSKnN4AKPqgShTiPT0ZZgTCNRUMXzqKe0eIsZnIMSFQkyEO927r8fDy1inydtjfvN5GAgIC2Ve8TnsQ7QZhZxxE4wqUyIfqDRXFSlufFa6CqCMXdrMOriypc4FSNYn1jrWy8enOa6AvLQQGsU5PSjN2Y6n4H7fvftYFEpPsZ+6Kr8+lFX2ScaVn6FUV+OFAPC4MdxCDlUYBygRKFTpzq4KihL7YtzoGYRz1mfyndKFrcdTC4IyhneD8E2ukWJOKK+jyUG78qxH83loF/sX7YlSDSLlhAJZeLNC1AhrIB44D/1LRHOc+fb7rUcmjSGsdV09w24kje4DJbzH3NjHusG+SM4Y6y2VKtk7iALAe9g7OGJHaqp8XeH6SYtAkUlXaTIc+s28Ye6hCKIMM8ZC4QTmKso0YzfMEdYZImwghGCicLhHbKIYEUYx2oTMEpTwJRMVfLm3pEQg67AX8B5y3g+NJV1m4/qzpRPEZT9kIPq5piqK0MC8SQEu/iHvoMi0NkTjmvWNfeW2nd2+92IMYh3ce2zYXnDOo2t6OG5nWX2NHR0a9T34Ty88Lmve+MqJ+cBr8XpxBAgKLfeUOUTRklCJmXbT31wr6zf7DlyyPv1xCaWKlKkyJC4ox+On46FHbArLvZpU87RCBPGN3kME9vRY9US8NSEYh/pGXKlYtqTWsJswKfdRlaUVAaPZFZGQY0QCNoLSKx57nC/KccHIE+8n3M0oW6FyHxu9V9OaCIsJ7UkN/WHzYtPCQ8FCwAGYFMm4ccshtxZGyapRUuYT1rdPOZH7lPoaD1kcGh31sIIQj88GihKI5ZDqhHgAWOgpwby0JjqEkBLsTzklWrTyLX2ajtSQAKxfxCgTy/6PNzzsmwRbJN6LK87sdMH61h1HfTNlIaWNu472uzAXLNWE83F/2TzGksd8c33qye1TLM2cCxIEa/oeC3uookSfUZkqlG5GmUZoCHlb3o8uZEa5Ngj4eFRoE9b9kDvDpvS+Z542GXbHJoA1Od0ZLJn6KfTxgR5Oqh/xks3QUIVyR9IzORSj/v3xsEK+kM/cfrjfX4/lk1A5vGvx807iVmaEnW2H+j2HJtM9TZd/FA83CeMUhRhhLaSEo6wgSLU1JSeLmrBBYJ1jU/nLJ580edB0CD+hcAPzizGIsorll/aftgLvcdKVhijGPToqAMUC7wLjtKmuxucvsA5wvWzsbNy0mY2Z6pxPPrnD5wntDpW4DvaP2MqmyLuY6ZpZLxBuqN6HEcbn9gP7p1RiS11TuJds6OEsp7DRMtd5nr5KrQaFsIiQxzkv4fUoUoxJQhPpB/qAMRs8uGGd4bBWhNFQpTPy4uFRHY3OkquucssxAnrw0pGrgBDj58zUED5T64pCJwdn11V78YtQgRSjBgoHSfNUIA2hRYBhh3YRGsgqF+WiVNlJ7dHZM6njPF3lPowN+ZxTlK4a6LSKc7u6PbyMEGbOWyJPlvUBj9+SWvKMalx4T/UcxsO9GG/0D4JgyOvw54cjowtjzX8myquPjw+4R49+xjCCYQFjEWt9JEyZtVDOvTby6LEOs2ZgPDq1I5oTjBH2Fq6RJYB8jmSS74lCtY/2j9rde7rdU3nC8iXW2VRnu44Nerjl/7rkRD8ahENalzbXuAKJ8kifsq6FdZh5GBULIVd43Ocb959xHkozhwqgoWw214unaErp6DThRyGRnr6nD/BIDI4k/R6kO54i3O+o6lpkPGKvA7ydeBEQvjEanNy+xHZ2R9549umBKvJpxr3SbljLeP9jT4gKLaTzALPvYHxgzcLoGarSBXkiXnU1CLrxPTkoY4T6s9my/nAf1y5r8HsU9yIQ+kw/A7nOh/uivRYlPBxcfuaqFv87c/2pJ3d4njD72SNdzHcMU1H4J/ObPcn3Me/3KNSBPTNURWWtC+dFMdY8nJrIgWQUXsaaj2LOmoDXPD6P46CQueG2scb7lzFFe2kjuUWsaSjTwcMW1sC48Yz7EBSuqEry2IRxuMaVFvY4lGIKPSDLMKf66qJDjVm76DeObaD9eACRnUIBjUx7abiHjNKwxqGU4U2lt/BOMf8x0nBXMGjwfkLwWG/Is6e6slcw7BtxD3DYF5CrWD8e4t4RQjs6bg8d1zPpeQsGq8tTFL2g5IWS79gGIkPpuK/dv91yKPIKTswn2ry8sc6ODCAjmq1qKawASbGQMlUmpE6g1E01COpxwZFcGLwPLJqZ4sZDiAAThIN+ycNKWpQsePyyel8weA+bP+EH5EIxAUM1JoRzJnuoQBXKkvJ9vJ6KZbSJScymwgZLeygZzHfE8wWmhP6QUFtb5ZMNgTJYIX+5+ZAL9nweVi02c/JNTupoypozEs/9YKFAoPQ8Ikt6fDpCKosnbmg2t3QVeWYqfZout2nafZoICUAQ+cLN290ryGLm3zk4atfdv982tC2ZFAB4ngUSgYQ2RQsuIV/7vC9Y6LAcJZJJ+/mmLo/pX4GA2lpvv3z4kFsR6Ufir1HgnndOdEYQeNWgWOlmNljuD9ZHP818Ykwh5O/pplT7cOSmHxxxj0oQpBES4oto/NC+XBJV+ZsfqowXlXCowWjMsOCz8bOBcy8RuuNhhfQh4zucno4AzSbEcQGU1E3NYyD8BkGJjffE5UsmQz9SvSupSn260LZgdWMeRNULE248QHh/zHFLPWyPpPmNu3pdIOf+sam86Py1/j30GQczI3ixeRMChfXQw3gaamzjwT7PD6QdDe6liATaniGqdiZcwOPQWTagcChi3KqO0LnpAOeORGGIbUsQoqMQLLeG1tfY02On089UAj91bqd7TeoZTxgzGDue87aqxZ9P7WvGBiFhCC19I2OT952K+WMTgiOWZAwr8eTr0JYQ+hcqnrE28H8EmZM7mqdYjoOw++D+HvvVw10e2sfYp2/x0mCsQNgJHvGth6MCI+6dJxd1cMS6ere7QIdCgpJB21jH+EEiQIhjnNy+62haL2i2yn3ZQoGBEuIhBDisR5yrxHcH4T/Kt+uPCmKMMe7G/XqY84zRP7lgrf/N8zFrq3zM8turnU6EezHesVgjfCFEEgodnj+GpZ0cQS+MUG0t9Y9WncNwA/zG0sx6w9/293L9o9bs52dxnEa/rWiq82th3rMGxkOxGCOsx+wfXBvjIRgsUFyPW4aFHYNEvwvUGPZYD9Yub7Tq6oR7bw9NCH1RxcYeNwgELxACHWscigUhzAiD5BbTDvYy5h/zgfWN+0xfc30o1KxFlJKmr9IpzHjAOM+RY0iCQokSn+54inC/+fz79o7a2ESYNMIvfcuc4fUYvDjkPhzGG3lMCNuPPhfjSxgPCPSppdhThdr+iTkTP2gaUMyisP9xN1IEhSOujLH/MC4pP8/YwSu47dCAjxMUsaAwRh6Y6L6xBjEv8B6RQ0N4H9eCERZliHHG3ofwzDqIAkDgAffFHWmEfxGCluDcJO5D5O0KERIYtWgTP7wc2QBvKWtH31CU7/1IV7/LH6xLKA3pQi+5H3uODvp8CeGqXCcfetKKJT4mmO88jofepVYVxihETh9zJioTH3mtuS7GJL/JEcbwF883RM7hO+tqojzAySp7w5zdd9SjWGaqZppaeINjX2gbRhHagsK6bElUiIv+Zxwwxvk6lvFD/aO+x4cz51ILnHBPyV/92UMHfI1FYeOaNnQs8e9Nl2/JnseYZM370i3bfc6FUM2u3h32gnNX+7xj7T424enEAYCyhUEt3yMNioWUqTIgnUCeWlDg+/fuc8E7lCaOl4UOScFhU44Tj7llseR1LOJUhAmHcWJtR1DDUtRYSzn18YmqbVH5ZxbluEckCJoI63igsLChgPWPRJvFQPegTyQEZzxCQciMV/G7zb8zWrTZZPksqmdxoBuCFgIsCzDfd7B3xBfv1HyyEAtNWw72DLuXJZ77wWbgMcRjkfcNKzObRSS01/rkx0q0IYfSp+mE8nShU9w3QiARxNg4oHrCIoliyHlLA8M9Ua4Bq3Ai4YspIXQIBmHxJpkXRclj10fHbffhAf8eFkn/nu1HJqoSIWCOeHglfYb1GMEqbOTx0s0INihb3NNJq1fCPLSPio+e5J1IuJUPxSceV5+6gKYmwWfKAQkeETYwNkD6nvuERw6vJMIylYQopY+iwW+sflgH8dTQpqAsti8xr0YYTnnHOsy4RXmir1GWGY+MFTZkX9SrIut7qsc2jCXu1+5uhKypoW145Ah1+PXSLk/M5d6EsCEE94e7HnGvQrQBYdEd900tbjXmeTYzrHFYn93qyPlUvUM2PlE9DyHu99uj0rhYXpk/hAFx/BxzPXUdYBzfs+eYK1IhXBfhkQ2qqW7E20ifPevsVS7kffzXj0x6EZl3mUrgBxD0Q6nd1NekerfoV6pKIVAhXHOf4iFrYWxwfxD6UZw95458BXKcGPM1WOwTbr09lHKfQltQDEnmDoeYM35Z/1LnbBBEmNM7jwy6YOUVPpP17tFAQArzAc/NHTuOGo6CWgoqjODVTXqp4I/+crOvgQhr4fDWRDVGGc6gIk8osor7ERMTJcHTFZPINRSY6+M+7e2O8o4uOC5a41HSNx/s92tg/BAGhgWcPEMEx5oEXqnxKPHbzF56wdpJ5Y4xgVeGtqMgsO6h2D32+BG75vdUQozOFKIv3ejS0WR3cYwDBYzqq12gZ34uSZJAP+brB5+JIsX4IlyNtnhYWjI6JJ3nUY48zNsLRJBrFVU4jObjsF8T94b28/eR0dHoTMSJ9EvayhgnrxMvF/OLsGRCAin131CTsL3dkfGsKkEoXJUryjdOhG/xWTSIfedA74jPSz6X4kYUbmJsMC+uOKPT53EI2QqFD6J2V2UsrELfbjschSHzOYwTPxYhTaife4qH8bhTdCEqGsB+hCJ13NJG99xERtHo/nkuJZ5wPKqeO1XrcyREY8xUtTASiKtdCQn5Qxh0QvXAr9y609dbPovIFOZAKF0dF9AxrKIssvbQn2GcBKNF8Daj5DGOGutqbHAij9uLWI1FhQ8YQ8B6z/2kmibeRO4H++FElKMboloTkfeE3Cwv3DMR9RD2d9pGKKhNvJ61knHP3CF0lLM58Sx5heLx6YUcfD9/aL89sD9SWmkr/cxY5fE9uye8no01biRgjwkVHonWwXhG/2N04ww/+u7OnVE+peexc/7gSDTuGRcoUuGe0ReEU7PPMFZR2JBtiApijxoe6/Vr5gB48IqRORzjgrLzotWtvucgY/1u4mB05CrGZBiDj1+33I+1YG1BYeWzUXhDOGJqgRPuAwWqvNIrBvP+US8qQUoHx7PEK0WGueHh3y5PRQWmUCi9r3qGvE0YT5Y2kGM15t/5uA1tNl7gkQbFQspUiZNOIA8nkRP+wYAkzAnXLBYNhLMQ6x/KQgfrBEUYWFAzWTOC5YpqScsaaiYPdQxV+jwZ3sOrhn2yn7um1RdNwrrwqMQr7DD52XxQRMhbwsrgoQwTh15yPWwShAm6QFRXM+UcknjOAgLepRsIYSPel1CA6LR3FiMWALwBCGrxhP9g8f3vB/b7wsZ3uxs8YZO5H1463cP8ojZxoGpI5GcDJRwgXmQjHP6brvRpuo0s3WuiEJwdkXI6oaT6gj0RNkL/8FxoE5XGqiY2WCyuDbV1vrhR4jQcoIgSjbJDH6CgIuR4gQPCufpGXPmkOhLCzo/vP+DhDigl6TwRqSVgQ2U0Ns2th6OCIAjGbOQvecyaaae2p46tkANChhMhDORfsQGlJqqGSpQISowz/kY/PPuMla48+rlXE5WVKKnP+8gN/OmDUaUv7u/AaNLaUNATj+Ys4alBIUcZ4TrJY8ADh+CGAMEGwfihxDXziD5c2dLgY2l5wxH7XTceWQwIAy48n7dm6bSwDpTUeBgGhPDZkUR0WCSCP8JBfHykXjt5QuTO8bqGCQ8gOXNB8aUfuDY2aCz5WHnZ1ML4QqDBUICRgI0boSGBVXQ8CnehP89f0+w5j5v299itj3RNHj7s3r5YYn2odhms78yjUG0S62ZqOBGvQ6n08++ODHguHv3OZyDQpAtZ47oI+8EaGa9RMhEZ7OMawQshHI/E6qWRlzxuXML7d8HaVjtnDXlKDZMl6VPHdTzkM4TrcP+YL3jjSawOAlKwDg8MTyTrj5vfIzcWE07aPRh5qqLzpl1hILeP9iIskvR9YluTjzP6LO45yCYcxA1R9KefMbih3UNiUJb4LpL+mW94BGg/HoHzj1vq3sXtR7D0T4RCJZNWy5k5JH1XV9kVZ66cMi8f7e/klMfMf/YNr4RI4Qjm/aF+L76BR/8ElM3xcQ+zDsJhbVVkCAnlrV0Jmjg7z0M0CR9vb/EQ8uBFYByHA5RZx1C8j2w97EIn38l6Rxg0Hjq8EVA7IZiyhiNsE2WAYMn33LvnmI8/BFPWuqbqKvfQU4AIAwXKFvfF87YssvxzO2sSCWtqqHaPW1QBLapWiwIZQr0nQ+ByKKzi4YrjUQESVnGuA0MWfZcaFYKxh7bzvaynhJoS9vjM01d46B7rDJ9F+DeGkeExnxDeZ1juURJo50znHU4eltw7PKUSJiHGeMb5DsYP3r5Q1Zc9Cc9fvBLr77Yc8upuhOzhA2ItOGMVFUib/B7H9z88dITlUnDFC4JU4QmkaAH/j+Y2vc1nUN2O7/Tzl8iZGRlzgx2vqa9Cea+xRFXCTlrR7Aas1GMunn1Gp+cUd/VFBRj4DMY8nvvnnLXSFTy890HpT9dPVOUkksOrdo6N+2uHxjGQTcgrKHgUaRrnzM7aSUPW6ETV2bDusYewl7P+jSRtIoc2MmBxDA3rPH3Lnhg8sWHfBeYF45s1vn8oCpGNDn+O8quQgQgNzhZmneq5YnyyVqAEcn0oPfDEddEZbHjF6Cv2B+Qu5mI8NxrjPW2NvGVRURevEDo06nOIaxodj/KXGZtUAUxbOGeiGBUeZzzBQ2E9PTZorQ2Uf6+145ZV2+mrmm3l8ibr7R3M+0iDYiJlqkQJ1v50ByISdsEmGiYvCy0hC0weBnSwzjKRvHz4PXunVYdLt6EHyxXfET/UkY3xG7fv9o26e6DfczbYAG7edsQXPhavRwWy6FwhFgdct0xWShEzUXkNFkc2QhYnD3FYThz+mIcXuFUay9JolN/ExAwLBgsBnxeqUD1IcQy+t4pSo7X27Tv3+PubJq0i5sJyZGWNCiPglfJkyo4mtwLt78UzEYV7MMGxlGEhCcnbIVyJ3KJbd3R7+MdJ7Utc0QoFILJtZKnhYljgQ3hAOET13j3khESvR+lZUpewvuEJAWccRana2/fyC9dOWkrDveL9hEKQ6M3GwzUiEFYPR4tyKKvLojWaTFobYTY1Vd63v9lyyBe9eFWjdJ4IT5Stj0pLs6nxmXh48DH+/KEDnk+CwM71JJNTwydCToCH47mQl/SFOi5Qx71Ann8yHFlGERZQzPksNsK4MQHlHUse1Ya2dPW6J4a5wrWxGdDeEHYRDiNG0OK72DQiL+mYb+BYzPn/f/xh16SVnnHG5xCi5oc21kZhI1gf+UzalpoTFbxOmcJnOQARoSWqmDldeGZTRlBm06yuihSw7Yf6XTFyGQ8hEM9jImnLGupsA+d+kJie5hwekrRR+j10Emt+pBP4Zrj50IBvfoS9kE9IaA3CbDyxPsTdT5aGbqjxkNLJeTReNyWcCEXrW3fu9jBQrpE5xVyDM1Y2Wc/Q+LTqlUBfIGjRNhRFFD6YmA4+Duh/LMxcz2NTql0i6Gw6OOAKJQaCYMgJlU3jVQ3jIZ98LtUB8XSTkM79fsVFx/lYu3FzV2SBbqi18eSIjSLcT6xVAYSkarxmE/JBckLQZx57EZKBYXtg7zFroyjCRNW0XM+wSwWlD08A1lx+xpN4VUc9DJQ1lfWJPiLfkOvz6ocT3stEYszGxxN2fHujPeb4KEEe+P7UIyIoFBC8y9HhrHjJh3xMIShiCOHeouCvdqNUdD1edW7iPEPCWxEsGRso9Ywz1gpC+sibCbmioThICE1k/GG0spHI4+LrwFB0WC1CHMIbQiTKIcovStyjVULJm6FAxbg11FW7Uoxxhfm6oZ0qs+brXTgGgtdRiiLKq6EcdpWXTx8dwEsV9Q9CKyHBqd7qYNQI9zNdYZX44bZbu/q8vwjfZfzH84MRejkwfV/PsEcPrFnWYMcvjQ5H96q7a1onPUbfvHOPj+lgQMSz8YoLj7OXPua4SeE50xoeL7iBkZIxQ58GLzbyAXP1gX1R5ANdQCgmBgLfpRPRZ3zxlh1e9ZV+5e8oeP3uper3/LV0+x8Hej/uxOW2s3fUNu054t5y1h3ajwGKdcYNRnieaqLcIhT5i45fZt+7e6+NToz1ML5YO1ILiUThzVG/XHpSx0QuX9L3npdfeJyHd8JM50ISZowxiaNc8Gge6B10z6WfRxmb59xz+u23Ww67Ih/2/l6LzmVirWKcnbS6xc89+9XmQxNGoajQTWtjrXtvMOaiYKbmGjF+keF8zEwYVVHIOdQaowL7EzIcOeGpVRfja2PceHPKikgpxohEO8J5eXx+GMOsk/QvMhfrJms1BpUofPegr7237DjichfvR05ib0KJGq2KJo4XsPAc70fltngUgk2EYG4/RmGMR2H/JceuvQnjao3fU+ZPvkcaFBspUyVIfKFIdyBiCK27YVPXo5N34lA1JgKlosMZMKEaTzghPSQQhtC6QDovCsJYSLhHWNo9cVo4NE+cmcOESa10hMWLcJmQc8PG6KEwblmPfvgYNg2s1liVdhzt9wkUrKpewWmEBSryeKCccdZDyMlAKGbxO255ox8c+Ki1ZbkrcN+8c7f3DZb96EDIKNYa6xALC14AhDcUmIlqsR5iQC4BB+FiaXvcCcu9HSxmbLznH9fq5UsRfhHms50xESBUi5C8jQd63HtIGAveNBSzcIgqZ5J4zkctXgqESwR9hOfI0otCy8YZJ2wOCK6rli+xPUcQjqPwNa+i1xadtI7g4Qeh1kQhZSiNPCY/AKXSKyFO5G+EpOPUAyR5XdzNj0cEiyifjSeF+/aHHUfdO9BUVzO5gPKb+Px6Eqm9lLF5uCIHMMeJe0ZDKEKmcYnSQFtCaBpex1NXNE8cxDrmghCWNEJngtBDSAVjE8EPBccLkUx4Gfk7m4KXbI4KBLrCx7gh4b2xlq2cik5YhiNvKwd34uXNlBMFXD/KKl5kvhOlgXmL14S2pMaBL6mvsaahMRc0MThQ2Y9xEp1fVePCDf2Nkb5twgMcjkRgPNIn3F+Ea+6/l1/Hm+K5BtFCj4BMW2s9HC7y+kTK3aOJ9YSlMX/joYMkhTMvycML84jv4t4g7P3b77a5pZHnULY896au2tcFPM/BiEN/xxVOxhaChSv8KWOCsUKxGf6OwEflRKpOseYxPk9dscT2ErpaE3nRWNeCIp+uqiHEDRvkTSBUMo/j3tiQdxOFUpHDhufp0TXC2xZT+MLh3xhAEEgQJiwZeVifeUZnTp6DQDznLJzFQwGOUIHNE+EnhPinndJhmw/1+XUc6h/yUBu+v7aKanCRQEnGCpXP/uT8Nf7+UKwo5EGgqHme7ERRmhCmGxR8xhTX2sAB7dVVVjsWeTcRiGnHCRMVWzEmoJRccdaqybykKLyr2g9xZ8w+9+xV7nUJh4cT2cA8xDvD92K8YHwHz8jo+JAbtRh3zAmKaYyFktUe8uaimwvbfSMUiTDfB3k9fY1iT8geAhoGE9YA1lw8oShfJ3cssaec3O5CI4aUoEiRq8McD0U24sSNY8cGh70MO0sVhwjTmuC9Ys3HyEPYIG2kkBOtRnlCcWQ8R3mi0ZlCXvGtZ8gfh2MouMesuyin7OvMuagYhrnij9eM+8l3hqI2qcV98LbwnYzjYLwLB8YyVkI4LAI0lfmi4y3IgYnWaiq7Mq+v33jA32eJR48SYD1iT2FMIoxjkEi3/7Hu3bP7qB06Fnmp+UxKqHPdURg0nsboAFuuOXhSfvrAftvVPRwJ54ShtjZMrh2p0TquGLKHTijNk/nSsby2bOdCIm/dhGEYhRfjSkvdRA7k1FojzAUMBPQjnhmuORh8WWuffHK7/WLTQb9O7hlFpOgrQk1HxiOjLXsQbeRan7S+zdebsO/SJu4tc4+QXr4HDzlrKcYdroX+Yx2Nyxzx8GU8eid1LJk8+4tqzXiRaS9GBsZ9OC+P8vNbDg34Xsd4wHjINYXccvZS9hEiaNgXmI/uRW3EqEp15ijdgQ6qxhg2kd8crxQZjyTg/nhRspR1D3kAWe2E5UQWddgfdh61LQd7rTZhdlmaMVWqSJkqMUIeQbwgRPxAxHgoVZTAvmNy8rqi1Vw35TDN1ARCRj4WJ4TWeJhOeJ0v3hPfGyYDGx7t8TAONmkPlSARPirRHCYn4C1iIWya8BB5wYULmuy/7t5jv3i4KzonogpBNmErJ845YKPxRGM8VzVYkEctOTLuCg/W93jFoSAksWn2TlScSrW2BIGb51H4DvYm3MpCHhi5D5GgEYEM5xbdifORaBNCOaGOWJVQAvlOhJLGGqxBgy5MsOGy8KQ7YwLi5yFROQpBgXh+Ng6KctjEwXVcH3lKhPIQd4z1ib7BguXWn/4Rt+jGk51DbhkhnAgx69cus9sfPuCKYLyK3qMKymH3qrAZswG6lXPcPM+OvuZwwvgBlcT8EzLBAowgQrgMxT5YYMkdQnjyMBn6eiLJm0UUwQ8lNwgh9AmLNBsOY4dNGkGVPIV01vkQiuCHuE5U3qNvggDDBsQGzGcRdkkYGVb7J5/UZvt6a2xrV78r7tx7D5VMRIfleshrWyQ80Ud4BrEac19Z4BEIGtABElVuJMA7g5cy5CcxhjwkqMqssS4KW/M5lSUnivuP0Ljn2IBb3b2y2ShHA0SCQNhAg/CMYtQ9eNg9LBgECA3BestmRhIuYSt4PPDqPvWUSOAAfj9m7VLPCTy9s8kFuK0H+3xOkrcCTfV4Xof9vDlCRlCIWDvIxWOsM2+4f1hFPewtpryGM6/Y0EMlQARVQu64z8xFPM9s+swfBBL3SlRX2RPXt/k9DR6JVGHLD4Ve0eRSCkIobeC6mVcoUS+5YI2H7pEY/8279kwWU+G8OipLoTgQekk76V/uZchnjFc1dG/mlkOTwmk8pzLkEAVFg7mMt5z7gMKNYI/XyUMFJ5TmIGOzhhBKi2CPFZuJUFtLTkqDHzeAJ5mw1mART5e3Gs83TGfQIgoBAQrFB4GNuYagxLqBN4P+Jy/N+3PiYGXCyvBg8d2E4nJPQ5ESroPwuZB3iBeO+44QFxU7OObz/lzGI4a6JTXuUWAtQgE65/hltu/ogF103FLvIzxaS2P3lnU/muvj1pCkIMqoV0mkX9lHsHIzNuiH6Dy4pD3vnJX20wcP+Blqe45FubJY5hEKKUnPPfO50zvsnttTOhtdwUJRjT4/8vqzZnuYtIcJVtlLLljrno4Q9km4JQY8ft70xHU+57jvhGKGHB6cXsmJnBLmHopPai4lhYNYYxgHrG9Y6z9941afrwi0jGPWceYsihX9hve0tmrQ92tyVMg/I9SYMc09Zc0J6yvrDEbRO3d2+xpEkaAzVrW6h4Y9kXFODl2m0Oqw/7B+skaF864w3rH3xg/zDYo+/crehLBMzh/7AO35kR+Y3Of9wToVnUcUeQOXLom8jBQ1iRdSiY/t6x86aHX1tT6noqqeUVg6CjbjHKNrCNGNn9kVBG5CsFGo2OcCqfPE88N7KcefzHpgeDyXN95G5ifrE95qFAq8g4yVKGIlShEIEk7vSFRFlXUcBTfkizHG8f6yVsfPzuP7OTdqeJQwxGjPY81l76Q6KGcrxtuYelYf+Y/xM9j+5II1kwWeWH8ZN6Ec+ug4xTxGorO/KDZRW+XjkblLnhNe0IcPRsoW44EjbZir5GCizKAwhwqoeK4Z/3hyOasUecANAw01kSd4OMod9NBYD0usspoqoiIappTnD3tcODuUMRZKsANvj/bWhOcbUw30MScss0R9nSWHGJvloUiBlKkSIxy0yKKHC50NsSl2IGJcaGczwOqYzX3N/0MCIYIUE4FJQNhTvPAD/yemGgE1dYMl3poZEwmmQ5EQPZZ04RxrX3S+Dnk6kb2WiREvuMDER6DyQzmp9ra21Z42EXpDGOO1t+1yAQFFzZNwa8hbqXdlIa4Y0k4+j40KrxULKtZqcreCQseCSvGF/uHIrc5kxQLj5XVRJJMJF2raGmrcU4QyFcJ1sH63TJxfQr6YV6yqi6r7oYDcseuoLxQ85lwqFoh40m1wtZPL81/37rGdh/HkRR4F2kefIeAixO4fiw5Rfv45kUDHZ/xy00H74X37XIFCgGITZoNmcwjKx7SY6NNX2BOWNtolLMJt0y1voZLOd+/a7VUPCctkbK1uJcl1zH70wAGvfoflFmXoho29LgSgdNYSSpWIElm9whLV0qg+5HH+eG7GJi30wPURHvO123dNCqruvfMiCdH5L1i4UXhSrb70247DA74pPXKoz6sq+Rk1tt/OXt3sGxwWUMYI3hfuEzlsCFRbjwy4d49PRGHlPnCdwfIdDu1EAUVYX91S6/lffmbKRAI6ac91xoGxUbK3e6omQutCS2k/myzX4966lJwoNknmCoJKqFCI1c9P1UkmfX6M9kRhuEHoTxd6xkbMPXzmaStcEQ5n7TCGCbEhrBEvMUIx/RKVcx62jQeT/nlsgGzqzfWN3hcICvR/dSIKE/FwvwN97gGjD2g/8z0o4HEPDteEEsvawxzw6lQcrHrqCt/IWVMQ4IcnTrf3wyE9r7DGxx0/6azBYa6Qi4NgEpKWN7Q1+WYfebwjAQvPUhS+GFmC66vxiI7bUJLcBorRVHnODO1CCMHTxFrG5k2fp3oz4/le6XIM/nTCOuxn3xhhlTVe2ARBthbDDQUqiBrAm9vW4IIOBpCg9BKqunZ59ZSDe0NuYTxvFcL3eH7USdPzGL00/brl9vXbd1tjXZQ/goCB94K1F+G0dUJYCooez69bUmfveNpJPnfjYbIoMKwDzHe+L3iVaOOXb9k+sUZElmqum/HGXOd+8d3MC54/fVWLh3ClHlzrRQ2WNU5WA+T7HnP80slKcSg9CKCcJdS2pM4VJ+4zuWHcCw/hmzBcoNygSCHE41mprx3wuUqODO957AnR9zM3OGLCq8NNxKrVJCJjSvDexg/6phofecT0DW33PYz1icPpLfK6IFwSapZ6MDX9RA5lcJ5jTEQxxWDGOVkorswB5gXKOH1Abgnjt60xMjKgxHgOHJ6wqqjaHHMbyz9rP/t+OEePtf+uXcdsVeugKzwI94Tr1U+sOaHQSbrCR9wrlEquH0Md38F4CnvqZDjsxNlGrD+MLXLMmJMYjXg9x3IcHYiKBDE2CHNlD8VbFIqapIN5z/pxxooWGxqMSqvvT5o9aUO7DUzk/pHHhheXPg6VAHkfMg9GJYyj9BHyQ2qebZgnjDGeQylnfZ8pry21jWENZizyep5D1jlnTYvPh/HB6MxEz5saRpGMvPBELRD2SN+RmxXPrwtHNyDXkPfH+sa94/6HfLFMbQw/rC88z3dhvGLNRdmAuDdqc1evjzXWxGhMjvt3HRuMPEaEVTItWCO6ekdcmSaahyqqGCXZS5cPREcFPP2UDp+fGBgYoyji7Mm7CcMlB5CCVv2RHMlnUxafNfxQPwbnWq/MF87/DMYK+pcoGfqVPbSHnLnYvtrSUG2vftzxk9fGtXd0tFhXV09UzbFMkDJVQuztJkSOevzjUSLuRMgRG0lqYYBANvd1AMEBTxTVm9jwmdBYkEMlLHJQQpgQliYIGywwMRBYeA+eDT8lnHjn/hHf6Hn+kYNReWWsgiGMDW8ZeVW3bD/i1ggmKT/bjwy6xZUNkdhrBAk2iiW1UcgMmwClzuOKFHgOWR9WmijccEVzrVfyY6PAAkmbsayy8IckV4RUBAw+h8M0//vBg76Qd/c/GiOP0IzQgPKAohgd+Bs9x6aFpZmNmOt14aqR5PphPyciLPIsbmHD9vC2iYUtRAclJsOsqKYTucKxDuGWZ+Ej9wkFl+9ECCcngj7jmng/QiL3F+GL12Nt+v/tnQd4pWWZ/t9T0vsk0/swAwwD0osUO6go2JBFXeuuYFm9dNf/2nZta1vLVnWxi4KiILoqqGCjCAzSyzgwzRmm98lk0nPO//rdz/eefDmTTIlAEve5uXIxSU7Oeb+3PO9T74fX4fk7edGUMFJWcfR802sIDypsQaRfYajJE4sBtKvTDFsZnZYWSGpTz8BAiYWvLjHoeRA8Yeo1sbljv8/jcoGgItblQQdMrjbpeHaBWGSHz4tAifrmnWsVPfvj1r1S8FHgrTFnRgYDUSWehXTOymSdeH48nrwvhheGKivE5StmxuYaXVixaSd/H/tDYeAgxBszeVGPr99Dk0b2r9W/oITHGr5obLPnUG7YGyiTx27fV4ogs+5qIPzAJilHXCiLptTpPUTL30uaIBEDFOLB/k7pHi58PpFAM6AHm89S54cc6B0gOlkpI8EINPZIKUY+GMU0qbLmAVwY0363dqiWBbYm2N4wgELM/89nVX+IV5HC7liPUF5fQCF5NNpo+BobnYoooR3GxEJJ+WWmULJi7yRQLpPSUVsGMqkOY6BalzaF2yjQnDv2IkoJc4Eix5qoR1aHRTpJi2VeeBbqcjCWmTMUWRRO0lqQW6LarcBQq1HEHWMGmuE0oUB5PzC+ogHE+shgUtpfXkrDDLFsWl829jYRTaJYPD8GTro2LF2PGlMeSUOODT3TBCAQRZSnDvNcNDEVKURlTr37bl61UxEL6g8ZH+CsEP2Nkem0MpOONLI3UfJEKFRpLQ4GGxAX5VQiWg2hBePjTCvduTIro6Gvb2CI1x8lGLlLeii1VUSnGQPzSbTo9jW7pIwhV9inGI9qygy5QHVe5C8YEuzDKQ35UJvPaRzIKNLfmMeYok6a8jfvfFx7JX4+Zwq5wt1U7DcjF9lOHa2ig612htjrmC6kNl1x13qtLfNEBJX7L6YVxnpTokqxn15UdJknIs1Rx1PNSCjK6YQpxryyDUzeY2j1ac/xftObq7XWsf8SaUwxhVN3hVILTSaTcs7hUERYtOykWfUqogzzHfITJR3Wx7ShEe9JnFpy5sjxiPNhQGc37Wwdrml2JDsizTjuESJHf1i7W3NBxgQkP885qi2cf8zgGR8OyDj2G+x2W3d3ag2ZX6KT7Cd0AWQwuk56jqOxxLwdqCEsfxP79GH0UeeMgZmuaztYnWL8LLUUSPoV8jocnehF+7qNZCHe5VF48u/2nkKozYfQn8OJZXI1/n1s3ZBm5sTB9+xFber1xnuXj9GanQ/qceUsipEQKB3t4T54aCPOWiPqQS5C80UUkno0HJTWwN5YFLkXY/PxWNO+q6tT8gm9JPbSe+Ppc0ROwt8xt5QFcH5Dr7E3TqrJhPpqy1LY22PMzuz3H9y3sZSVhLHO+ElfvX1Nh/YiqYOMmzOKPMV4v+j4GfuVMUxEuDE1jkBEgwvWLsWMGNB0kcVCpREwXPi6/PdnzW+R0YEYQGjLk94Te4PQRd563KAIlwvoyMzDZYvSiveVyxlllfeiLxKeYUbJgUPRA0QxCDcDvOhczFyw3Db0/lAKWULzGdNOeJPoJS9/JgQKhhr1MHiCSdbj0kbwkVuLp/1rt6/Va5kyFONMkTz8vMLjhOG5NDEquxMJyRxjmESPPUoj0QTynMl1B4THo3FlfIRmFCEgMawYF0oYEQQuy3TNTzrfWj1CElprIo4IurvX7dG4eT88uvNaoU2HQS7xDlfmpBSjYJHqc0/SzJd5h9kQqlj2zeqNu8KNy4d62EHa604jWBR+0g1QfFFYCc2rmeJ2WNWszw/Dj0ZEbBxLfUmkVOeS5gJYU4E3bOjexIaE7YzIAMoohjyGsZiUJGAt0hEvRpQwPGQIZS5uPNbMEcJ25vQGeQhRekjPw+uIh5e0pth8mCgCwh3aWOZHzHCqy8uEI5I6sWi4xUv4G3euk2EA8C72im2JHiADUshiypVSWlLryN7Ga8x5IK0lthR4fA9KY7cUfTEVdfVb7VtSH0hKBHNJ5MBofwfbCGAs/VjnyshQiCDFJqUoNBgEf3XiDD3PD+7fqPfe3dUuY5LPYk4pZo7MaewL9QmJab91lSLhIOLFM7BuOAJY++pK6jcqxZSXNm6Hc9DESBJvwjrhXY8ELVzEyKujJ5Gy0xBOmd1ScsSMlMbMWYned/ZdW11OxnxsIk30gzXluWUEVyaOjnZLOZ43pU61e7ev3inlTwpCnTVXpq0CxhlrTF4/e2XJ1Hp5YOuqrKEy6VYomSjG1IIRQUjXvrFveS7OgzGQZUJN0tiXdSL9lNQ0Nbput2bWzC+eXNJi07Vh0QlEFOORvdB/W5orRjcOmTQBCO+BBzrNSMrzk74t73ihKKY3I0qpHlLful9bg2FIcNSXTdEKzn9/qeH1XWt3aozsi3WkWRcxjomEFnVX8Mw4ynBkNFUOpk+qfcDaXSIoQVFiHBhwSr8WG1pG7GhGuW29guKpQs5Ql/jNux4vEaZkQ15Reb0e3r1MZkiKOmu2aW+3mAEZI7KDvcvdBsmN0r3Vl6qguyc2BkYPZK5xQKjOLmMsqowfZ8spc5plFPPnyELkenmzexxmX79jrSLw5RAVey4j5wJnmfRgk4dWL0zK1c6OXkWELWXeagapweHOR2ZhkNk+LKi+GCeP2GZJNU1SiVGeSW22yH6mdAel6/D4HQYG4JlJ12K/PHOhpZGlkVb803sEHQRdACcndXmQ/RBRwokykmM3nu+03ICu/sp7NoT7H4dG3ZrH4jRGvyCVkGdjj+Bgi/pG2lgaKeMm3SAZwyCmc3O22WcYR/G+OVCdYvwsUiLRbTi/OKIYE8/PRZhTGnuSjpZYVDCiAoyRbH8hXHP/hv2a6saaap41Mlhy31tz5qF18BgcOCnLUzd5v3RvKf5WkaB95mTnbiYSpXolMhcq6PuWVXSVvcvziKAo0RlwOsXU1VJt5Ma90jXSPbiopX7lCdP1b/Q4PpN9zf7mPXEmkla8uX17aK23mkecoeydWLduZyuf3J1WZ6Ym9yrnIEMoq7Egz3CwTZR+UiPBjalxhJY6ixpxSbB5xZJHU7WUoBmu+/WhYDh2MbzUbHIIGlDzDiSgYeZBccfbR6QBrwbKDYcZr6MOK7Ul0D8T9YCRr9NS5QDKE5eqFVGbMYJ3EiWaQt5zjmgNLzxmqj53OGHNYePgY0ypHmbAPLEYYTzDjx7aJO9dZLYztkDr9xLzmRFK6gheVOKVio2nN9ZIOOBBffnx0yVE8QK31pEK1iUhh0cJwYTRhmKFQksqLwoAAok1w9uKoiOeu4TYIp0ixudBcFBk3FnS02gGarkDpKbxDCh2zB0pAd39Vmciw7W1VjTLj0N9nAhl5tD6x/SERzbuDtffv2lIeiVzBVB48TSzpqTeQKdL/wuUf3LCYdkiL5vLF4UAocpYYhaepTaYNxmygU17TMFnLUcKwWN0UhsVa4ioK7PaDlPu48UYGa0e3dIh447Pzueo0bEIR3N1Ram4GEU5NiVF8SRFiQs+vg/7ibPR20/Odwj1KnbvHMIcCFDyUUyPGKjVXHNRkORHOsxAwZpK2/Ob2RwfUYxu0GD3D2Wke95RIWzY1RV+3bfdmowW6CMC85IV2ENhzeLjhaa+BqWKfkVERX6+bHO47sHNljoaijJscHBiFLL/UJyUsrN8mxRgPNpWGF6p9BeMJwwulNhY2IvzgHQnazRZp7nG+COahGLZ0QOVNfTqnIGiPns42uZyBw1n51cp9iz+jsgbjHVnzasQKx71LZxzNYRNKQQyxBKSE8bFXkLptTYLKMgDUqQxsom2YKyR9x8vZkhOkBWcB6WA1FWEk2Y1y9hl3ChMsb8ae4jxqaF4XyFs2NMZHtq4V8oP8iDuJ/UBqsmHtTuLitgSsS8kjitSK2c01xhxT872PzICRZ59iMIrCvJCCMdOazA5uGVvkmZXDGctGNpHimdnL+7a16M9hAEkZw8KBnTUkHaIxXQg3JCqN4WRNK1cYjgs3W6sejLMoFeeZvTGw7U1KFfq0pFGSHHw6sfmqpxl1b8mUSiUM6sJyyrt67TZTWH1ri7JwrmttSUjjibhd6219DuR+SR1hlbDasozKdMofMVCwiUfG9PmQ+jsN4dNSw2pxQNKHa9st1RSGoGSIogTj9RomjpbdCuEezbsCdPqaeTco9cxTowpNXeF0RCNLWEPiAal1Xqa40AOB3oU7uvT99ExojNbYcyvyPw0tb5SMruTovsyIKs425xH7jhlFeRYaxw63InUuw1GrmKGAntUNX85UqgwUi1yzvOqjjgUdd5xMPV09YVlmzp0x/B7ZHtV3s44iHV/RBF4fPYl55E0Q+TxzSt3qM6QfZLWH9KKNY22kb84GHH28V5E/3FaHUzZLSfOQk7i5CAFOzYcZn2Ru9wvfI/uwd3DWU+nfh9Kxk3s01feU66tLh9u/5PtbeQm+/1AulLaMOMs4Ii6b8Me62dFSmoSJYxgbuO3nFxk6YbdPXJ0IrfLe4JCHkEEnK0fDTw7J1YHz1yNlPaXjkJx7h/d3KGzG50POB9Vr5Qj0mPZDey/x2VMQlxUDMdNb5CxalHOwXpltbCRs6d/2B5c6TVAhsmw2ter9+R1D2/u0BhwQqzo7VTEFN2FuymuBYY88mzh5Nrwh7VQ7+NMgdCpGPppZVJXMWyq6kSEG1PjCNNJoTmyTZ4xvCSRzhvlD4GS7vUSKTUPJuDSzGwINxRcGUBJHQbCnPQyaoCwexA+UUCXb2yEFsowyg+HgUNvHgfoTRPqT6ILiYGDEdg3YFGQ7k4rPOXAR4UdYcIYuHzpRwHj3VvPmj9ikSiGIPNRncfzb+kcNKwVUUBCdS2jJSHIQMiB5xxpXmLeg0jYGXNz4dY1O/VaIlAIc5779Lkt8pTcv8GKg0kdJB0R4URdEZ4qa4QbwouOmaL0B7wt1JExB7wPSoP40RCclZZTjYJcXWlRlZoK+kZl1XA2BhxJ6VCqv9jVjPUImGFdlBJAVKbIZVyZk4dYl5MK5EO47t4NYeXWDnnz42XApQiyNcbiSBohxjO9Q4gokj89lV5FiYEyvakQjp5cH36ybIsuv1zenoEhonTyeV+4ZY1SYZTWlfTDiPGMeA3GiNZx0+pVZBz7MZH6xLqj2EUFm/VQg8+qnMYWmY/i/kARqKvMK4LJOInAkIrB2aDfCRdXXFfSeLg82gsw8wUpqUQsytOtIr0xa80ZoP7IiqItKoTinM0MSPmWMqnaODMuI0kJF1O54oAijLEoRSnxJlNjAmPTTx5BQc6qpgJjnL378KZ2RTZQiiKJy8b2XhldfHBlzmrmeD/GSbSWecK4J1KKoZWtzmufogAyNs6biFZoAlqRLaU3AuWjk2Lz2LawYhsNq80DXgxd4fiZjQe8xMrZs0hFpuYMJRmly6Ih+VJ9S6zNQfmHCfChTRjvGBIohlVac54XoxgPLU4Bi8iT5mWU0JFQhqif6hgwVAeKYWZzlZST6NVlzUnjw1gh/Y79hafTvLEDYc0OU/RRLvayJyrzMiJxynA2GRNKF89ojZcz8tQeN32vxoYyyWdjsEIRz9qxDpGWnudUf56qfGjJZSUziG7HdBwMWMgvSD0VdboOC7IRxacowpBMu1HKsz68T/Syo9xylohQWZ2R9ffJJNFOIuw8B8ZErG+N61Xe/40IKDVKjBsZx74XLbhqGKmBKCYMj5bqyn7CCCBNjXV6ZEtOJDRVVYNzz//ZE1aLUyG5HxFlG4ZnoWCpUkTiorwwR4VppUoZTwxs/g6FjGeKNXykirMOal4uL1lRKddK8RTBEC0hcqE2b5FF1pC901I3qN5wn4liXD2ITMFMkyzF2raoOJanWkaHGUTw5eBZFk2hj5w50lgn9hbGFXcbPDCs2THT6+W827mvR5Fi9isyivNiRoxRvE+pqwz9tYlDRzYhxpYZ4US8bQSwBOaVnsv6RIIRshqIyImshPEkUXp2PbVPtAphj0anVtQfcMagFBu9d59eQy+lyMI5UqR5OBmBYs/zQaCBgwTnkByLxcH7QWQ1fVDuozcg77KSFekavINl3JRHXWNDe/pakv7MXcl6x/N4oPey+nOLxLDu6qlEPRxyiQKfhLaTuylucJ5FhEoJCzGNjFmjdBsP6muJNLNXkf3sEa1oBuKHJqUbAxxQw6UmAnNgDYT71rdbWqhSxW1ezRFh96DVYxm7M+tFyvTvSe9u7wlnzrfskDSVOu/PnSFWwhF6cMU54zzgzDp5dpNITRgLG9Fqyq2ukrnCCYKsi+8T5eSWdmOixKFA1DXepZyn0bSMGI9wY2qc4QXHTNVhROmxhqDWX6W8ZxKH8mDWfFT4uIhQuFBKyVHHMyJ6WZoB9iU58QW8Znmx2aXrJ8rfiyaoGA0ocaREoUSi3CJ85C1RPrd5HrgYEKbIHw4aFwqfISKMJA3Cym8tDAxLHilY5c8U02Tw6iBsyNvG88jlipedw4iHkeoiDCillhTss+e11Cq3eygDUJWMpGVb9slA4kI7Z4EZUkQLYPTp6DYlmfA1vaUgHqC/CmQHiFMUuH//3Wq9rzHamUCNGSD8D6Uar9GC1ipdTniSY6THntowJGtE0Q+ifNbUkbGh9BCJIc+aaF9tonQQpYqkHdRs8D39MojOYQQADBnGx5rxMVy0GLfML+sxqDAYYQQPd8Oyrcq3xkDVXBZNKZTymRgVgPVvrEI4wupoihhRKYvyWJFy7MdEmicXM4oldVTp9eDyRrjzNxj0GAR4BDEGIusUSit1eZ39BRkm7DH2I02aeZ9m7WXGm1XxOSlZeJjjZTXcBczlw7pxGWDEsS4xhSca5cxvFPxWR1GjddBScaklOf+8pqvfPI6RaIP1OG5mY3hAxkR32Lu7y/rh5LOqYbL88cIQJUNGadGiU3ZvWsR4D0a9jB8z7kjN46zlaXLZb6yWk+tr9TOiKMx5OnUvyhbOXiQBoGYK58RIykap192+wV53vD8GHZ50ZBMGH2eBaMzju7pD66QKGb1/3LJXCgQ1JrGxI3O4eseAjH3STTCWUeBYjyXT6lXLQrrwwrZ6Ge2sArKLiDNOGDWk7kYZzumZI829Xd5GJYwhxV6DmCSmr5w6t1mRLSJ7LbWkIdNnqaDov5hDk5Qzq7Upimo7rjHyBI8ua8zacWZitA/Fk8jqfu0ptnaI8IXnQwGnLlIKo+SlOVuooaM2D2XCHEGpdDFaAPQPaPz8PcY+5x5Zpb/JmBEk8gZYNiGcSJTdGMGPzw6pAAYX6W2xCSdpe0RYY4oZ89nbb3TgvBdz36uoH3WZFkFB3pNuOnd6U7h26dqS3CDS/d17umXAR/KIKP94JIybkIpaIRPyRCWSyElEfB1gn8U7iD3Is7IOzAvOBqLPap+RyCIMZtgklXWQEMIw39SmxCgp68M5wtkX0ztxrCmlMhXVwajiK8qodJrrjn3dYXeqNiaC+5i9xDkiC0JOQpqa9lr4JZPNhbo891JWBlZHPheOndEgunhqd5GJaiwvQpMaOZDifojRDPYQWRwYrfyee5c7FLkYWxkgw/ge8DqcLsjrfM4MTOqLlalAymjByE9YAhxUkS1T9dRJ3x/S3EkTJApdXs9TjijPia7fu2p3knZsbHOcLQxK2idGQwqGOMbPeqgfYu+A2GWh1U87ig+UiTNcOiDOM0heYo1gVXf/fr3ARnrPEpNe0q8JQwU5zJlFvr5gyRTVPEHQBckSz2up1ZYlQ+9KHAGk36WNIuR8TCvn/FKfi6zCuMfJQRP6ckKN2JMQnYQzs3yzpfYi6zgqfCb1RnwuMhzDKd0QO0brOOsYtmn6+nTaYGT3ZO/GOUTfjIYc81POnki7C9LzuQd6+qydRsBRgjO9KqcshaZh2negH6ruPpNVnSFptKzBROsnNRLcmBqHQOlBsMY+KQhbFXoPkDO9f6+X4YRb9BQpFKzoioV340WCrO6Nl05lVkIWlj0EX7kSlvY6UX8RU624rGfV1+ryQghy+MlP54CTGgJbHoo1v5/emC+xHJl3xqAC/UQYISCscHxQ8AE8hShViiAkdWTynOayVneR0BdH4SyjppCVUnbBkql6bi4H5pKxI3jqqitU30HtEgJz+77+8K+/ekweW7Gm5UkXogFnnxhqUM5QkpRKlM3oeVEmUcZRwNPpfNBs44FBqSX6g3BCGeH1UFtzeeyJRVspWKTMIk/MNUocdREoFghVFL/IuMgEcDFzWYnem0hfT7+YxFDuKAKn4FqU6J29pXk2j6AZV5ecPEvUtPFSgc3rhj9uDXup96GoWoITQoSs6rJiz41S8TUGJBS3TVW6QDDAIgsfc/j0ec3h1lU7lYr5iFggzVCMNSkIUGrYGL+UpUxRhivPzDxEAhKiON+/d6MMQbyeRSKe/QNhX6Kx8T7sEQyfWF9I82Mupli0m748uVAiAULssWFpV+YlZV6JPsWG1Dww+xNDigiRSEOSJppEC5kfLhIMb2ONHCTaQJFFySPVhrXnaNVVkk5hlyUKZZrkgs8hohXTwHAKcPH+aVd3aKiytJaevqIMZgy5+dTeKXrSb8ZMXYUuVy7gB9UQulBSSklD+eXybfrMOW11YW+n5eqzp8rlSHnkLeb4o+Tz2czl9KZq1W8Q9fzV8m0ySklHZcxEJIhsqECfKSxSuJzTfOPIeN1ps3XufvPYdimvsb/Jss37FMniOaEcJ9rB3zaRdlVF2wRT+KjfgP1S3lQIY/pJbzGCAxSDcxa0ltJX+D0KI+eU+WWt1Bg86Y2ltOCkVwpKJmOGSMEo6xvkxWcemV/IPDBEaf/AeaQOhQ0S21PgiEHGoHCResY5xBPLWFA0tib1lDhLqEdhv5DGO6epOty3sV1eX4wbnCDqWxUslZpItuwNGdikeFmfJfVWIrU6UX5iBB/FhWeFcIb5i004eQ+MUJwtnBXmBEUXBkZFhijer6tU+4aaCtINOQ82TkZ+zqLJYVI+o8gQc0J6I0Nh3g+EGOVGXhBV5XzzMxS4ckkowiLq0o6dVkqvPW56Y1ixrUPjpbYOqBcY0TTtMbkZtI9YH5wUrBv7g78jZZufE72nCTd7GMUXhkFSKyOb5Eg02ji2MOQxCpOe6qU0L5RcUp6IppHWdcMjxlqodhgVWck00hXTCmupIfmjfxpSe4hDkT0DOxzOA9YTxwmkNEQBcK5xX3PvYFBJriRGzJod+3S22XcyQBOnG1lq/QkBEHcS60pDZu4wzgVzwfwh2yIxEUa2sTraHUsvrTS5SXlGTInIQZkDRP+NPZcdrDgh8j2514hmU1dIGqEyHJI0RyOVsj3Kc6OkExFBPsXm29KNUihPBwTMY2xkjiHB2Ym9wOIZSdcWp58lGg+cA/oJoiuRqcGzX3LSLMkqmgMv27wn3LLa+tfxfLObq2S8I6vJ9EEPIWtFpEXQ3ieNqIkS8RrODGdw9c7OcOPyreG8o6boWdOEGjgRGR8G4gNEgiC3YY7qrHH1nJZKZTLFuizuavYf0bBSP72U4U6EG6TZPXkd+xLHJTJMkfRVO8JPH9kypMVK2tjjrHPHSH4qGl+hf2PI4eguR1yjGx7ZLCcin43Tg32fplGfyFEp4MbUOESkAI99UkifQZHEI1/e62Ukaz56E+hNYz0KQtjbbdEEeQgTylspGESpamFQMsW0/D2H6+tAWgICBoEFLXr03kox1SVfoX4nohAmvNxULWWrjrStOnpX9Urpjylx5oO3ehUU9/RcYExy+XZ0m7LMAT73qDZ5amMfhlDqCZMJ3WIKspSoHz24SRcD3nMuNgk4XTg5RZpQRGMnc5rscXHhacRrQg4wz42nB2ED/S4KGykxVnBuSkr6guV/Mkah3R4wRQQFC4OBueCiUk3OCPVGPFuDIjtFKcHMDQrLMp4juQDOmTxJiv1tq3eYtzygEGXDVqWoWWRoWyr1C+UVKAKitBEiWgVdxNHAwNMEDSzjo/bN1rYYKui/khRQSOlPlCKNlRSSXFZrz/PAzsgFTA3OG0+frYuIyyLd0JXnK9VSzW5Ooq1G5Y3ChsBG6aEBYjTwYl1RNJSM6Q4muzpdEKfO7lPTYBSMmGYiryH1L6t36PzQPDheyHidI7tXrB9kD7HmvA7Djjlcs8P2NYoJ646hYJcGUSHzAFLv1tVnDgMovVGQS3nxGVNceR4imz0Dvaqj4CnYh8yHau2SVCHON0a+GmkmDYgZi1jjOiyaq6a4GPC9lreOsUfUc9WOTvXbIZWLKAb7bfmWfdrzNE4lSsQlz+eybtvau8XKxeVO+mG6XqE8rY816QjULXWqNoD5Z/e3E8Gl1gGSpwHmzepurK7PahUhZwDsD+pGpNCwfmt3KpUSo5S02gXdterVhDf8adOt8TDKYnYPUQ/kWJ8UPYyERTV1ki94+ltqcor6EXXn98w9ZxjFAsWaeVi1HaYqc04oFSZAopNTA0sMfwyH2P8EunVmAjmFrCN9EDmxYXevol8UqtPTi+hDjOQwv8wf9UPMHWcOeWNNzTtlZLV3dagvXozYqI4uSRkiGyCfz4ptkN4vagFQKKoxMco3cjo6MQh2aD77LApM/yLWP92nij1NFI7PJ3rPOWZ+IqECxhkecfY88hO/DrULsd/QmfNpSr1L8pQ6L/YW8ujy2/4UOgaK4eLjpqlPW2zsy5qj4CuaqPNntRvMZzSx0qaWUv+KQYof80W6XRooZZv39ChLAXlHiiZnd0p9RdjUbqmFysLi3CRMfij8yHT2P2Ndt7tH90g2Z+0cGNu0ButRhpMPRY57COMV8MycuZGyPYi0siZEytU2IyGVqUjqXTlzfzV7htJpYRCMLTKog04TiaSdOuVMi+ypRzZ3SIZy/6NgQgQQ0/QhZLH0fO4PM36oSUL5hS1N7RMwYmAlrbTIXZTTJQKFLGntPeHutbt1TpVlkaQTq81BhWW9IHfIqrBGv11yamDwGOvk+v3mKDqpvnDrmhKTbbwR2eaRfZZ1w/ATU2ubMcVh/D20ob1078Y9SnYOcxFp9jmLDDj2WRopHTDWf3EncE/jnOIc0R8Mp2RMRR6JNj0ahuxtzjxnmTuANeV3OO52dPKzQWcwsq2h2qKWpN+ShsscIn8x/mmtgHOOuVVrJqU5W6o1GSswSqJjlBNqRIKU42c1ql4Z2RtLHGjxgHFZHkWNRjhkPPHOe3piMJbvuRg9iy0GqImX7sAds72gz/z0hUtKdc/oEqwZhGbMC4YiZ70mYV9l/+CgLZ9bomFE6tHj2IPU8Jav40SHG1PjEOXGC5cjns76qookBD/Y66V8E6IUU/THIcb7hELFRUK+PgdwH/n61aYo4AlWHVPCYKa87qnW6+Fgucl4VRAGeH64yFAWJSQSWuv2LvrNZNTdmssP7xdCtDJn/XJQALr69klRR/Raur0J99vW7NBFh0cDb8bSP+2ymgKoZyuyoaWuKjz/6KkSsoN9GKpDXUVOng8idygRGKA8M0qv2O8Kef07l8+oWB7FkoaUFNgyH8o9FgtVUQ0nLfUnE2qr8srnxgtP9ET1NSoYT7L+y9Lo+WkjntdMRgrr0nW7NHZSm6it2kyDz2FgUSMzuLgA8G5SbB9T5fBmPbatS9EmFEMEG2snEoWk7gbPo3n5B1QMTfSComMiITGdCSD48D5FKlTmwvLsYZLKhqqcGVPUc6G/x0dM24Cap8pMePZCqPd3DmFjRAHCY9aWPDO1ZAxz8dQaGRBcTBRCsw9E+VxBk8AuRQZZVxSE9bu6dClas9ZMGOgxBSFHRChgxBSMMh6qakWF6OllFw2GB8bPtQ9skrJHmhj7kwv5Aqj1C0G9olh3IkjsF85XT7X1p8IBwZ5rylcqAifGLYzTGlKwLNKCAqxi+6xR+MYmtSjznBfOEiknKAlKUSJ62ocB3iNvMWeINBoMEd4TJib2IvTYeMxRnDCwSAlTClqS0kndT1c/4yuGHR09YR+NY6vyShfhrF197wZ5slGgiRDf9Og2nU+aR0q5ThRKU9IggOiXFxR5wJljbVDGieqUmseK6S2nn6GQEp3b1W1zwthiFCqmQ3ImFXkr2+OSNxUY1O3hmGkNMhiWb90XHtiwRz5sPPQY44wTI7yu0mrhTBHDmxnUk4i9I/rdJNVXdY9qSG1zz+/PywSl4bUn0TXR/NZCWEG/F9bRojUd3bBjWW0n9gBGyuwWawdgEel9UiiZe+YBIxBDobYaqvD+cNtqaMrrdc5wRCh6R0NWoo5JA1vmkIiv+pdB3U9WAOyUhQH1d2NuRS7SWBle+rQZcpTwd8iYJONRsm9fVOIzxqpoRCJD5TTGCWx37EVkAmthDdaDHDvsK84pNbpENkk7krc4eR/LMiiGjp6+sGp7l8ZFjSVn75q714fptfnw2xU7LLU7h7FP3avVzKLkZ9JdjYdxGvUPDIT2bkg7OuSh3u/3qh+E/KdgLTKSqA01hTCeKrW8czA9MxQLFONqvq0OKWuMrXjKiYYmdwxpb0Ty+B45ifnMfWD9CQsj1m4QpaUlA/uDL/YZyXTsAer3iEjx7Ow5EJX7aDxFlCv96XuVO5M0V5RRmm9HNrRLz6wr9V+iLm5IX7IFrXJGYIQB7gicCTo7Sf0azh4MAcDPcTgs/ZOl4TVV5SSfRaFOamc/6cNkXmRVR4lMpVnr0rW7QzP3Y631mkPOl2ePABRn9n45WCfO9NFTSe+tkLKOE5JzHYk40GfSpB+MAUWdu4N9BKkPEWYyCrgbDtSwmO85z6S2Tq2vUm0ZcoEIDJkE1P1wLpnz8hTAuE44+v7rltXGKkmqZmUI300+G7p+IojoG/yS/YQc4D7CyUlEjUQ2dBuMFObd7nUjbVKNNyntSdZDV7LWkN4MR6iRJkjBIYYRg86F82beJGPBK09XxPkUM5u4hzkX/Hw4XS5mdqgP3LaOEnEX8mJ3Z0f4p58tU6/GWPd8ZFudxorsZP+w5zbBsrqrawjDaLruKzrnZjcl6f6rdsiYinv7LwFuTI1DlG94BCsKKl5QBF/s9RKbnEV84ZZV4RfLt8mAQWFBiQQoB/LqywgoKhS7eGqDvC5cJRefOEOCqrypZBRS5bnJscO4pRD2yXOuSI368Fjndy48DjzKDxczF6jViJhCiLLBWFAqUM64qBH5PGNkd6F+4poHNpa8/MiudhgOlUZWGBLiJ40Pr05nP8x7KCnUbJmxk6+wNAKKvTFS+EwEFAomn00d1LFVeUUX6APDJbyjE5KFEOa31km4kdYDMKQiDTJIgmqllAqRFvBNQnWLIMNQYU4GCtSL7Z9zH0HePZ45LhD85xjB6hHSMxBaaypkXCmFSgXxkH+QUgi9aGXYvrdXSqXVrSUpZFDe56GNrlf6BWujdJ2k3op0JHneuvu1B1Aw9/RSa0YthSlE1q0cL6UZ3epNFQ2yghl/UvRSkR72K4KZ381vwXgygpLevgFdxDwbOenUgeHl4jlQ3FCoUR4ZJ+kORJW44DEoxP6UKGeY3NTHkSYGyQFGCZ/BrzHOFPEs2jyT4sHeV2oCNSA9BbEhKQVq+77SeMnTjykSKKjWD83y4evy2dDeZR53lDmMNC5ISwu0vkJE6Mqb1ALOlNIyE8WPvd2d6Bs8K/OP0vKWs+aUvNh48Tg/KBO7uqxtAYoP9WIYDzOb8qGpmoa8vRo7exCjjOaenDEUHlKFVF8Im1qHES1gUGgcSrsbCNl8TtE9mPGiJ5FICsbBqm371GcIxa6xulKX6O4Oet3EPmlWA6XUURlOZsxzVksEE8zRgKW9WvqwOSGIWO1UGwhSVCEMMVYz9u8jWzpEcy3GJxgO6yrV4BuvOZc8eyOy1kWZhFKE4oJRzL/5uaVl9YdZjVVh+dZsKMIC11dQJJq1pf6Hy5xaVFKoxWYJCyWGd4U5KLYUexWVOn1eS/jhA92hTnUecBDb6UeWScbSK6u2QgoiUU7enz2k9F/R3JtMJwKOgylGjVB4mDcK8WmoXFuFvKxWKjVGH72oYj2VtWUwA5wDt3haXZhSb02NGSt7DsUkzgnDhMkSZw4GIZ/52xXb9bnssVjDuKMjRvP6RZmP8jmtwQw7yfK89Q2rTqjpN3f0SgHF2YfixHyyLy1dS403ZBBk5SAYPvwOYydPIy93bVXoH4ABb5DExlKgjK1sX09fmFTXIFZHajFIecJRM7k+o/WVsakWB7nQXGuMn8xrjA5jVMR9ykKx3zizrXVVobna+lkB7oM0c1+soQLMp3r0kJ4Nk2NSG8kY2W8o9pMbhzJiDtcMergaoLhe5Q3JofAvN+yGsKvt2CdZxbpubu+VU6KV5urZTFi5o1NGMgprlAFEYDm3yAXOMuQQ6mqRsbrP/s6iDEP2PjJehCo1FXLIIa+NyymVfjHM0qKQ7+w03SBtR0PaSesS5AxOG+7dqMgzl7F2jLsSA4ElwdkZibJoX7G53coDkE8YHBAIRZnFWNPEFXZPx6GWYsFad5zMMQWQqCRzzl2EaEwbZTjoRCVeX1EiDhLzHnehasK5V3Jhen2FHGqcn2jM8ncYj9RpMvfcR3LS5KEtxwnaJ4cBYgQac3QejBGMsOEMnVjLx3N+/jcrpUtQRoAzYm8qSpgmG0tnNpVHicrrzLj31FMLDx8se0nUN66hUiUHCuGMedaPjfo67llL5bQ7rZG6zSSjopzIIgYHONexX1xM93/FCTPDXwrcmBqHGG7D07VawjrVfySN79+7Xl5YNjaeOQwcFAwuU2Ntol9LrxQUcpCXdu9SLjeUpxwyGt6VU3OmhVSpOLOs0SU9f2AYQnnAM6sksiSqQ0E+uegVFMz3B/VjQLgbXSwes4xqfNI03BxaBNbDG9vDrRzC3qRpHkp0YhwpVz7VO+i+x3erePVPO61upj9r3t9BMB4rjMBzxqEm+kCUAoHA5UxKn1J96KcShUk2qOeEsYtZXycROSS3hFUt4GknLahCY1WNCGPtNSWeyxbhgtJL3v7+lVIGhCp58kSdeCY8UFCtk57F56P4kLaI0GaMuWIm9GcscpTJ5/S8eFsR1Hyp5qoqp1otFAmEHalPlqIyoMhWNAxh+MMT+Pyj6bexSZ7PEjsfXsGaCj0/+0gGOGlcWYrCMYwzqsMhtSk2faQJIO/JXuMyhEYdRZa6LjyceP3I5Ubg829qFrg4qUlAWVRzTi6gxDgVfXCKqMPqJEyBx6j60076S5F+Z2yIXE5MAIo8vVlQuIzx0dj2UESmNVi6DhFfHBV4fdnjMfWAtB0uLva3NZLGaCBaZ6mxvJF2YGJMoKQq2tY66En+xR+3SFEX6YdIR2DLzIR9MKclZAOsAcX7vA+pGDHFblZzrRRTIgmgua4yHFlfqbSuKfXV6vsVo3gYsrOba3V2H91qERwiUqSPsT/EblY0VsDe0hlCncUbbzWYOBGIvt62aoc5NzAEeuk/tCccPbUunDanJfzv1s36W/YnSr7q3BLymFgGadGCYsiHbJjfahEfPot1Ze9yPqIhxcWKAsfZY4ycFaWOJJ/RUlul/YvMYS+LxTRF4z5cU0t+B4kMKatc7CiG7F9SUFFqONMYUjSJhB2VInCYH2NUin2/j32YyyptjD3PXhEjWiayDA72YLOGsRk9h/qbKS2zzxw5yfczm6ul7POMKJJzWmrl3CC9kLkzI9OIc5irwWiGUWmLPVXsoMhxFOO+sH5Xj0hy8OZDQjHSnKC84i2nxxzOA+q+GO+9Gyggtyjs9KYqi9h1G+PYlo4erStOAhRpGvhOzRtFupRJRSGJbvYqCh4p+3k2HAOs4fpdRAH217gjsQuGL+eTjAIUT5xhyC3OFu9TXYEybxHmyOLHZ/F63qSxqkLpUOx/UsFecfwMPV+M8uBEig14GQdGMDIPmUyjec4ddSbUOTKmWLuBkQm5QIz+YLyR6UHD6xqM9aq87lGiU20o0n3UwXSHFy8xxyZGGPNeniYbHRU4K5m7NJtemvQgNiTH0cUZ4R4opzEHKL/ISerPcCphaIosQtHcnO4S0n45Bxg4nEwMfhwRrKHVSvUbc2vOUpQhK0GpZ744m6R4IUtnNFXK2CUaGNMLuU/SBifjh/yIfSCWtkQTV6Pf6io5CCzatH+KGXXWyL6V2zvCHWuobSzomRgPi8OYeL0a2vcVZXQgF5BZ1BmS/liXqnkDt67eIT3EejgN6Cyj7xD1p2wAo4Dzpt5K+dz+6X7J3rE+W5nQXiQ7uWC99pLaJSJLZA1wxql7RB9SqUPWZB0/x4jFkcNYMHRxluaoGx8wUg7SUpEx/B1y6kD9tXAeQJ6DTIjOArJlkI9pUgn2F05b9u5wjYvL68wwapiPyIKcPrXcU/wMvYfIHj3e0BkiiYmM4SXTtK/iGIYbu2p4N+0tERFh0Md0/3JiEFK929qGEkdNBIxrY6qnpyd89KMfDTfeeGOorq4Ob3rTm/Q1HJYtWxY+/OEPh8ceeywsXLhQf3fssceGiYrhvFDp/g3lm5C0Cy5mckuL2dAAADLbSURBVJNF0arsB+ugzsXB4ULwi8UruVQ4CPEzyoUckQEiPdZQzw4Gr8XrjXBCkHM5ECkhOMEFUFFpRYUxpxfvCT8nLQbBgtcQAw+vO96VeHnaa+0Qkxu9dK2lVCC0lKKX+JcsncioR2ONBwordTdKA0q8wFZHNAhe2kvls7yiRoVMkXtdZV5MX/yeaAPzqBqvJKpEmt5da3eLwYfIRXt3VUJ7a4ZUc00uaXRrXmPmGS8aTUAh0iBNhd9gUCL4RjKk8Ny/+uRZYp6jFo1UTdbGyndJz7BmgmLvE9udFXyyvvzXRR8RaEorzCNONAsFBOGrOhHS0fpQnQvqh7Ijoe7mOVGY+rIZPdd5R0+VEfOrx7Zrr5C+iPBjzXmGjGjejX2Ofkb5QPqaRRoRjvJGVWQV2cTDSQ8RevywJhhy9InB6KI3EdFDhC4eQIwB6Nq7+mhACHOYkV0ovSJGdBLjFS8qe0x57FmjziY1JFeDV5UaJArLMSKtXg3Fj3ojoEtehnRfmNtSq8jY5hRtb6SzBo9saJchKyMquV5Q7Mjtj5TSXES8P8/EmK/8w/oSYxIXfqwHI7KGIsPP5OjY1W2e08RjjpFJ/RpIn0UuTBRnRViaqsV41lZbqdHcunqn1gVlB+UOWcDZhRIYRYQzCB05nli8wqS8bO8cZFbDIIWamzpB5onzpv4viQEr9rkk55/9hMeYMyYSgb09+pvJdaYcoIgyFuaWImx6bGFIs+dh6aMOiJ3M/sWbj1OCaBvGBBEN9vm0BvPwoqShdB09tVGpZdc/Qu+b9hJt+XOPHJR75alTrB/sgdTY8VpShR/fZWyDEEkwZtYUg5nXEj1FGef50jWMYhctFsOJMxrlBCFiiAKpxszqAUf9nu1L1m92S60iaKTrEQlVH7yEwRBDiT1G/SNpWXjfSV+lllX1lmIfpd4AYo8BKcTINtKZqDGIkUCxZRaIsgyESbV5Rdh5LmRdWpFP15GU10csnFwvhWj5tn2hNo9xSj1VQUQAyD+J3GDNi2P01BRYIzFCCceo/+H9m3XeMSYxzMgYwErlOeSY6kcWG/NhNYZX4oxgTpBHePr53ZyWain93G2c2Tkt7Jl+GVdEjIliM+fqT5cwvHEH4QFnbpgnnDpHTakNzz5yslJQY5NRUqphZdy+ty/k88ZqiNLM/rvsrPmls2aRvZ5SavwP7tuoz+TekZGyvUNnm9TdpGVcibGTqJpS2QeImA7Skw+nzN6xdpecFfwtdzAyM63A6/wmUTDuXjmhMKp2dg1x9oCo/HKGeCben9pHeiPheEqTF5w6pyXcvHK7nDIY0reu2q6IYn11hfYvvSPN11IMt67epf1oqf4ZKc6c78n11brvicxjYFJvx/vzrDj4uDvUDkUZKBZhjqyo7P+XnzRD+x+jO53WyLlAlrIGjAvDL/Y94z1JN33mEa1JZLAQuvT+Fs2/Y/WOgMQkxXRvbb+o7rln0v0VcdRxn7LHIZrCiRJT3TCWqVEkChh7K2GUxD2BcRDPvFKhE4Oe+wnnKPKB9UG2s4ak4QHGKvIiGe/WqDYS3HCGracgZ970nu2dtJcw+RIj7iP214qbr/SN/Y+5ujthdIxtKXB08P9Ye1UecU2nomLUsO7safTH7TK+Y0103N9Qs+/WZyETNH+tVtPJfpuavB97m7OEUyUyQvPzSTV5ZTGpOXU2ExZNrlVEN7JEpnVc7r2L+othcUt1mEgY18bUZz7zmfDwww+HK664ImzcuDG8973vDTNmzAgveMELhryus7MzXHrppeGCCy4In/70p8P3vve9cNlll4Wbbrop1NbWhomKEsf/OjsoRFRQVBGkHIDIqqMUpyQVhEOsnO4e0rXyyotHACBcuOTY+PK09AyEdQjXldtVxFpOzYmyni6EJJ1KaYY9A2H51r0yOhCa0ROFoMHr3dnbJ6WKQ2n9B4pSPJUilpAl4OGyuhZTzKOiHOnECwUMFGtyW878R54y+b/kit+xZme4/pEtuoD5AzNo7OJmrupI78uboghBBBc/Sg8XEMKGwnEuFgQeRfDGCGViiiw33pEIxZ1/2qmL6fLt+0rea15D0Wnse4KSgCDACCLiFal+94pauE9sRogGE8KD74EyRhoTqWArt28WIxPeLJvXoowTFLpzj5oiRisicFzszGsk0MpkzBuGBw8htbfHCkK54FC6UHyoD8KI4N/MB6mC8e/zuYKIOvBIKiUtE5RnjhA1o5uIoj1bXAdlOxUGpBjjeY+KIh5SK56vKzWcRqHge4SsFMA++plY5EwRx0IIrTX58Ji8f4P7n6WnH496x0DFnaeeoF7jRznA+9+LUVldlHJwwswmKYUwrhF1YC/xeSHZX+xBxsTFgMBWT7SurNIY2N+shQqKSZXZvs+UJqiMewaUSoQBgqIXXXc899S6SnlHGR/KHAYHFzqeSZjryPVHwRno7dfncdmQ/oGCSeNponawdBFdKE/xZe3rKq2/EGMURS/U0tR8JZTJNPTkM0lZgbKWf3OJPWthqxQT1gzCBKJY1jjY6KMpRbD+LsZshsKRpvXn3ygvVUXY9+xip0eUIskDKFEoA7B0VoUz57eoETQOBCi2yWTl7KA44G3lNewlzmxFxox96wVkPYVIacI4YX3IzlSUtTon7zhyij1kZ570zz1iS4sNm8vTqpRqtK9XhpgpLSYTxYYH6UDSsw+FFKY71iYlYkrPjyNkyz4zUok2putVaCzMWk9vqNZ5jkyf7D3mttSsm5YFSlvOSLGOLK0UeaepxCEMiECxv2vdLhkSKMtqhh71KBlUZqSSfoyBqP41KUU+nVJWni7O3mANSd+EYVNskYnxrHRr1SIOpqHG2WCNYi1kD58vEpNMyOdplEw9pDnTSFdkPjkzTdVZ9fvhPJMqSFSNCBlKNHPLufirE2cOiQYz3ujdJmrEfXPBrKnh6GkNJSYyXkd04cZHt+oeie8TlTkU0dhjq2ePpTpzDtbutF5tF58wY0iUJ52Op2jplg5FGOTI6OwT4QoKZUg5dbROnRCi7JExB/nGL5cX9XzlyizMsaSRk0aPMYqjjTmKDKblaXzIByItGBQ4oAAKPXsv0t9H5Tf2IWR9iajx/rOTPmV8NlkGLzluqu5yIrPIGlgHSQtU+mZTdThzQYvOAfuM5+U9ifDhqENmzmquHoxAHDtNZxFDivuZebp1FanF/dpfJO73dFk+NvuOu/aiE6aHt5+3OGzfvnc/JtW4p4gk37Jqp96He5t6T4xZ7mUiuqqtGigogs16ir0SIyVpW8AXWR8nzTTWUkCLBubwlFlNSk195QlmSMXfYxRanzGbP9aejJt0LVY889RI8fnRsUe5AFElambPnt8aXpHsP5AuO/jfh7boPNPHDbmBHIqEFTW0NiBNuGD7MN0nbqT+WguGOBJ6dXb5Xo7KMoIw6o9ZT4xAZB77IxrAaYd8ZC7EgDbGywplGnHVM7/oeMhw7gAcrOyFnlAMyza3W9uIFGHZ0rLU1liDy5wjCzir7FvODHIDuUg0Fp2Su4P9wthxBP70gY1h0skzlTkxUTBujSkMpGuuuSZ89atfDUuWLNHXihUrwlVXXbWfMXXDDTeEqqqq8I//+I9S4D/4wQ+GW265JfziF78IL3/5y8NERvRUU7iJclGdXHoIv+jZYjPjCWDzbthtNMlsdATpK0+cpfegEfD//J4Ltz/s6C/o/+RMX/vAxrBka+OQXgMxDSEWQqLU3b2zS0rELPolbW6XMi/lh1oMBqrokl3OBIGM6Sqni4xLG0OFix+hhKDEQ88lzPcI0Gh4cbDRszB+8HRFIMgqiVRkMrqYvnXnOnnKeC6Uaww2BD3Prn9XUwMFha/l1UeWPUsFM0Y/2AY5rHjIGC8XKNGTtJca4XHzyp0aFwIFT1p/UqcSX8drUIys6aUZgOmccZR+lFYEJMbHiu3G1sV4n7OoTZEJLifW+MFE4MJSBahdohEfAvOFx0yVd/3Ke9YnCc0WKVAvk4wV/BMFgPCAVBl6ECnqp5TLQpjRXK386G8tXVfqO8W4AAIP5Y15AHiZ6QUC2QGXycwmapF6jfiCyBGGRwEFwpprSima3Sxl9lt3rZNByd5AgUaBQolTxKy/oH3A58CmprqjbAgrtnRa/U2KJtzYy8wYP3FWo/XcKRSliGIk8mxcuBiNNM1EWRQ7FJG4AYuiseYYQZwH0t5Ux0NfNZT7bqi4OyTgWRsMaoAXm2cTG5OcEIVQ7Le+GBCCQM/NPuB3XNS8H+kiMpYqc/KGbu/olgKydpfV+hQS5YP1JhpC6l+J1bEY1MCX52Qeucgx7jhzKOuR7ZJIS7yUmT9eD7sZHkKIVNK9RHgfmBAtjc3mjaWVNz+TVfojBuziaY3hoY171FR7OJBWk81YOiDRNfL0N3b2aj/GFBmUIAyqFX8gxcnOr6jdB4p6TurMmF/kE/ss189cEJ3MSYnBUBLZR3U+TKvIad1RpjnXohxvsqJmxr6h3Rj17llvbGfpBtDME0YOe09RuXxGtXox8rx4uqXRRIWUqHq5IVU69xApJMxgaUV9OI8xn49nnYiypQoXQyaJXuOswZhEcREz6aodSguL9QZpGOtaMdyxZpeRjZBynAgZVgcHCRTw7F1qWeoqLcsgEtQMx1A2XNrQd+/doM/h8yB1AAlhZ2lM0QGO7IgkO6zT43t6lBKIQhczFDgnyJyKXC7UhYwi4UacYk10mWMibjB0YniGALMq4aywn/JYbhCR9sxnzW7eq/spRkM4Wzx7+n0icDiyFxgbing83+l+XHHd0ul4D2/ea4a3sh4G2xPwH4o5z5KeI5YG2WKMd91KpYzKLEo0Kb4wtVnabCHUV+SsN1wmE7bsM/rrcuZc631nTWwZu5FjFHXuY/pnWvkVgcNAMSyZ1iDnaGxuThop55Izxz7HCYVCq6az1LhSZ9g3oPpR7mPu9Zjix53N+xqZR76URUA6nu6kRHEnah4zOZUR0Q3pVDbUV+TDzJZqPd+zFtl5i2D+y5lUiUiJFCuJ4LCvuNt1Jyd1P8wvRmh03lYlrH/IesB4SYXHSGLuaVbLc7HPMRqJ+pQbzjCU8llW6xf2K3OASZHsBRy2pJHHOz+Ho0mcJ8XwyJa9YWrS6iPNLIgTljrPzj72el7pfCu2d8gJhWAwQiGjCI+sngcD7/uaMhKS4ajLcbbj7MYJwjziyOTcRaZb7jz60HG+cILyt/wOY5Q5XARL7qTacPfje0qRJBwHoQAPqpVwYAzh+MVIHe4sbU3JoWiwHTejUT3hiHxzp4rAKp+Tg3zZ5r1yGsb+pRis/J0bU08Ali9fHvr7+8OJJ55Y+tnJJ58cLr/8chX/UXcT8cADD+h3KJSA/5900knh/vvvn/DGVPQuxoZsHDwUGC7p8sJjYD02gpopUhcQD6H6wjRQjM1FY0W7eB1U/J30Zom9BmKufTycfDbShs/UpVCVl5KCQOa98DRyp9WRzpawnqF4ERXAuEHB6U2UaJSthW14H2igWFR6QjSk9NlFI6NIIyrYRIuyNCjNomwRWWoK63Z1JkQbVsSOkovCi+KOhzwWGAMuRuZLBfT5jKi7eQ7yf1FIe5Ixpj+e4Amv/+3KHVL28Z6s2210z2nMbqoKHX35sHpbZ+lneK+iN5PPwDCgnxX56dSgPHNhm55ZjYqTy0l9b/jj5NKwGm5qLeyoYrCQMsKe4KI21jC8XUWlGDJ3eK7ZI0SpEHqkXnDpxDQZLgcMQwgZSK1C6DP/+OwwODhfXIodPVCWE3mqUfoYqWKMV+k+tdbD5S1nzRsSztdzJ/VsGGQYXzwC+4voCylJ9I26ZfVOKeakFM5vrUmatFoaX2wmqpqzilw4dnqjjMOmmspwyuzm8LNHthjTXCajOjmYlVjvppqg1A4iZFygeH8tRdE8nUQJUb4gFGDPpYu+OVd2pzH/lnrSl/QU009JWUqM8Vg3QP8vDDI8mtSyyEBPmC6ZV/6WVDoppFlrpAkzHIYpkUKUN+aI6Uo7R/hC8cN5QWT13vW7lc7CGlq9oNHuWzTYwgjpXiKR6p50N842kS9FM4kCk5KbsVScWS21eu1DCR3lcORrKHKcCwxNjExeT9Tt9NnNYWpTdSlFZu7cWilHkBig3+B9ZM6s71zWUnz7rFeVGuP2Giua0b3nw4VL2rS3o3IZvbs0xsbwR1mHjY11JD0nktQMl6a8sLU2PLylIxT6ilpvZNiM5lrJN/YqqS54Q40h0+ia07IGhwd1Tem9fSCPMT9DccFpIydEJdEji/Qxzyg8Me2OKKGajecyId8fz7d9rogb1JQ56Kywf3GXqNFpUmOn+k2l3g6UerlRh8MzHkp9BP+Hcpp9QZRRRzVx5DDPyOWo2kXlFkTKbMZNmhXPQLTPvPXFhDo7cUT14uQwh9mkupyiItwZKHTM+ZzpI9NSpw2iqNwSZUHhRgbhkIAGHJw1r77EehffJ+4FkSoUjBKeO0hR1Z6BIZGg8n1DKtgK+kOxITrpLYSibQ4mY6s0xwYGFEYJP4dMhQbHtOOQIkpvOdIMVVdIm4gKye77N7RbXR5GUUJCxPsTiStn+ENJFgV5ElXnvPCzeDbKlV/kAndJbG6OIYVRiHxnDdbu6lF0jEi5OSyt9YXOY4U5O9gPyOFIyoD8xxjCkCpvqppO1YtED8r+SOqdIVKB8Q2HSJrNMM55mkmVcTJPi6dWy6mJkcAdhpEBiQZEE2QJkHKJs086Hk6h/mLIVqK4m/OEMTMPGAfU4IZgjWSjXMPAKlf2Y7QN/O+Dm/c7QzH1LJIsQIWO/hIDt6TbYoBe8Yf1QzKFMNqIuODAYX9Nb6wMe9Q2wIiT0FF4ZgbWVMkaDVOzNQLKz3N8fXSayNnV3q37DwOb51Sd3kBBBvOj6xOHfMbkOndlJF6Kvaqik+qCpIaJ1Gll7uBA5+7BOV60dHnSzTmrx0ytH7YeLo4z7tkTiRSWGiHb6zCg2OOMUWnwEGhJVxypMGJ8YtwaU9u2bQstLS2hstK8IaCtrU11VLt37w6TJk0a8lrqpNJobW1VJOtwEAXDWCB+dvkY5Gk+enL4yUObzctatPxvsaZUwWBk+ednzJ8U5reNkGubvM/zF0+Wl6+YKDOkCOFrE5uTFPBsqXieUO9Ny7cpHM/h45DwmRwKBFo9Xg91VbeifpRrEUoQAie1DsMPiuQkxYzCXfot3Le+PdRCQd3fq4gNBp6eIZgXmehbuUYncgt5cKxpIUYTz8DFzgVLXZOUmOq8Lk+U7q17zYu5W/VaVkiJgMYzRMQKZiHejwNMiJ1xX3v/Jr0Hnk+8gzDvoNDLK4mXE2ppKywYrD1JPJYdMC62QB9qxcPWXDN5XZamfjUqUqWoF6F/UZIagEenPpXWZRE0jDgYCc0gI6LE2tg8V8igY3z0c1iKIKJmiVSaGlsDXfJ4/qvo6UXx72CNHQod80lNAW/YXGEdzNUrJkcPkj7lcKPMIexe8rRp4Xcrdgy5bFGYSFe58Lhp4dgZg8omwINHzjjsW4oqJXVNGFE04eVS++oda81bXG+N/1Zu6zR2sXZqcaz2ojI/EBpr8mogal7KEM49mrqIOtUUMB4EcVtdlZTxC44zQ4L9/7tVO3SRsAdZS/boCxZP0WfGXG7mXUXfD1nRN/NrQYCiUsxQJERp3VClf/M+KG44C0iteEnq2fHrxLOCgXnWgkk6Tzv39Wiv1uShPc+FRW31cgJozyiyizff1ovCesYpKuAkxZH1xZONYcVejkQEKuquymndUbfYT6wFbEsRKB+kZfD+KPTqAYXRTY1XdT60NVSFyQ12kc1oJmo0yDBV2t8ojZkQjppar7GRaohHmS/qCaPxyP4l5XHhlPrQPbDXetQkqb/sI+SHUpGaakT/TK1OZKVrqzJKZFKLnp0qWEZe8blEnn/68JbEa45Hvm7QiNvVpfniTMQzxJ5oqbM+NhijKPLIBF6bPkPWJNzaTuzo6pNCzC8p6Ieyfbi9fSDgPCK9zSIZGaVTsq40/EY2Az63Wg1XLZqRtl5jA/C+YlHEAecc0RpuXbVDsoReXcbQOdgPijP6xjNmK8L3x6RJafSqp++FOJd8xftl7qRaS3VVFLVdtSzm3GIeTbnlfMkLn6SvnjGvRXcHlM6QFihekxABIaNILe5K2m6IdZWeYpmM7pMjWuusL+FAIcyZllK2kvWLY0vLEMYXW1YwTpRunssiq3oSyZfy94l7YV+KMh9PN+e2fF7K9w3PzJ5Tr0KUdYgyOrrVJBtHl8467He0UsCQTM4A6VBzG2u0z6MMQOaixMZnOGagoLuPOUCuRZZM2hZwb6f3PbKlo2e9lFEQ5U2cJ/QB5M02OcQqJBfZq9zZ3GEYti3V+XAE0Z/G6pKDk89WQ3oM+aw182WtT5vTLEOX+2bx1HqNva4yLwU7Pg+fUf75yA72UlMnhD1mJNVUWSQEWRnnO+674eacsaFAcwxOmdMc7ic9s78oltHXnjZbz3XK1iZFW6i1i9T03Mvsfe6GOOaYtsa9hwzGQcWZR3YhD9N7Ku4b7lL0pfSY4hliyPyNaPRJXazIhe4MKcnGXkkLkXSmEGuJrOb/avg9rUFzSPYA8gFdqDZjzdyJziID0DfSsr/8LAyH9HmOiPrfgxv2yNlIJLLU1qLT7hyMGJxvOrP5rGQfe4j2AyeV9apKfw5/z+up60RGKTqYrGkkorh3PSmvYVg5pAh5as+UNxfmPHNnkPannlWkCddVleT1WOJwPn/cGlNdXV1DDCkQv+/t7T2k15a/7mBobR17BpHhxvDitoZw8qIp4dfLtoala4xoooZ88uNnhCXz20qvOxgDyoWnzg0Pbd0nGlKUelJr5HGEKraxOsyf2RzacO+nPjM2Hly5tUN5rHjiF0ypTxpCFkJPIYRZLdasFYVPXujWunDcrKawbW93WL1tX+iDKrSxJkBevXiG0Z+LDaujV0odQoBneWj9nrDq14/pvWPRJl+q7chbr55WqKxh4cOLV18VchX5cDxKu+phsvIInrWwNfx+5Q4p7lkalRKNSrxmA/lCOG1+SzhhziTN4+ptHWFya0N45ozGMHtqY/jib1eGre3dYc32TnkbuXimN9eoeBIlh9SNTEKQYIc9EyZVZcNbn7UonH7EpHDXmp3hk9cvk4cN4w3jUwJ/ZlNorKkMbU21+sxMVaXWiy+KLePczm2rC8XtkcwiF6Y0Voa5rfWltUm/fiCTCYunN4qZSVTv0C83VIVpTTXhNafPUdoNa0fdXEQfKRitdSHHJYPnqbtfJBBEWmKBLPM6fZKtyTmLJoe2SfX2ed2ZcMKclnD6/Nbw3GOmDHnf9Pu3Nm41g6ulWoogn/es42bo9T97YKOUoJkt9NTIhikVubBhV1c4c9GU8PDGPWEVbFiFQlg4tT68+RlHSEGPezB+Xnz+TTAb1VSEi4+fq3GCTXu69Pnt/cUwlT2nz68Izzt+5pDxMo+cneam2v32NX87Z3J9eNXT54ZFUxr12WDFlr3yLi6aVj/kvcrPCli9G9rpnpBj/w8UrDFuJqO5OGthW7hp2ebQTxPKYgjHz2oKA5ls6QxKDsRnaKoNs1qsvoY9j5LOxTMHL2ZjVThjQVt47uL91+KYuVBeG7ujpeCacTSlkTqJNqW5XHLqHJ0/xvzt29eE6+7dqDMD2HtcnBiSJ86bpJSUlVv3hr99xhGKsqXnP+6TbHVl+Ootq3TmMcyOmFIf/vacBXq/OMeLEg8m6cikER01rUERsvSZSOP/vfjYcN7xMyUbfvvoVpGXVFVzDrtL88Wzp/dEY11VWDIzo/Wf2mh7MC3f+IyXndoT/uOmx+RNR5lpqK7Us77uzLlh0dSGYff2gXBKRT4c9+AmPXs0khZMrgvnnjCzJFP53Fed0R3+dMNyEToAohyZVEPt+W314bVnL9B8smfZUygxt63YIdkvR0hDVWnOhzsP5fdCGmn5Qe3TyfNbw5IZDaG1rjrMnlSjNfr+H9aFLXtIfQ7hmUdNDS87yc4OZ4vxbNrTGb526+qwfPPeUgsHWCqpoXvGkZMVmVi62kiEprXU2T5urpEBZufSmjmX3znlMiS+lnuCvcw9Bb28UcVlQmNddWjv7d9vbePz7SZ6R+pXXWWY1Fg97H1ZPndvO2NuSeYw71ctXaeUeFJjV23bK4WXPbh5j9WGpvd5WlaZDHis9AyNdYVw9HQiOP3h2Bl2F8T7p3zfR3lyqPIm/o6fxzuM+4C9xDwfOb1R98GNj2wJVy1da06YYhCVOsrys4+bGV5eV7Hf+z03We/yO2Q4ecdYIef546Y9ZlQl+skx89qG6DXlc866XDJ/kvrWsR9PW9C23/0S9RYia9SFkm6OHHzzOQv2u+Oi/Nf901pb2mfHzJ0U7tvcMez+K5cf8QyxnreuxXHXr88gvV39AjPm1OVep26qoiIXFkxtkG6wF5bZkNFdwlmtqsqH++m32FAVKrtz0gdwcnHORJ4yvXGI7C8/C4cD5nbSpLr9njPeOb9atqV05zxtRmMYyOVCa2OuNAcHkm2L790QHlrfPhjtI+pbXxlam2qUCrp6W0c4c2FbeHD97mHl0HB7du6KbZpz9BCc8VMaq8L81vrQ0QuJVD7Mm/HnzcdTjUxxOP7ScYCf//zn4eMf/3j4/e9/X/rZqlWrwvnnnx+WLl0amptN6QCQTxx55JHhPe95T+lnn/3sZ/V60gIPFTt2GPPYWEApMa0NBx1DeZ+KwwVkCngIKEJGOFBQTwoOnqe0Z/tgnw1Ek/7wZnnYiaqQcoBAJBUDYYZHiwJWCu7lGUw8XPOHoTOOUZoPXf9HeaMhD+D3agJcYR58ohwolFxis1tq5NEuf8/4XuXPiQHHF6kp9OmK3hPu5UvPmleay/h3K7Z1SBFFwSJCEJ8Dz0os7owGz8ueNi288Yy5pXmC/ADvUGwaKS9aXdWIn1k+t4TVfw2Dkf6ucti1SdOIrtjdHX5093rN10ivH24P8CzM//OOtBSr4dZkNPsuvn96zeN4SD/75+uXSyQTMcCjiWr0Ly86Wp5GUc+iiKa8tQfbi+WvO9DnH+y9wBPRlX1w/3UpFZOIApHLOBY+U3Uhj+8pnZX0OMufAa+tvLxJT4/h1qgc37xzrdjJxNKHF7QqF86Y2yLWw+rqivCGU4YW+LI2v1u5PSwjitBf0F6ln0hMF0nv25HmXxE/qBTLWjikX4/B9s2k5jES3Ax3Jkaa05HWNf0Z5PMfbA9wTkkzPdRzczAwPjIIiAw0V+f3ixZG0LMOEg3Gq3SsWnrZGA33+UssJfBw9/zh3gujea/0HcX4v3r72oTwiPrYCqV/EtmZ00xd7V4RziCn49yCQz2X5WudvkdQbjkHRCBGep90z52DnZXDkSWcQzz5vC8K8UitSob72+iNx3A+0F3w52Kkc8Jz/ufvVkkesD+RO6RhP5GfXz6XI+k15a872P49HJk+0msPR36U6wPqyUgWSVIXy8NYSnPCFJnLaS1ffvz0cJ0axQ+ucU/spdXVp/Xn3uWsiGQkpcP8ObLnUJ4/1nbeO8Kdc7D3pByBvnLRwc05KL8bwGjkEHcGcxMjyBedNjccM6l6zPTxiLh/J7Qxde+994a//uu/Dg8++GDIQxvEgt55p1j67rvvviE1U//8z/8c+vr6xOQXAfMfpBQf+9jHDvkzYZwZS2MKz8JTMYbDuWgOhuGaE5bn9B7ORc/7GYNOr1Jfzp4/qdSTQIWp2awaFqcZoA52SaafM92PYaQO6iPNT7lCWE4BOtx7HOpnHujvDzRncd88sma7omdPhDL1ROBA7w8xQqTBZY0vPHZqeMPpc5+yz3+qcCjn7IlUkIcDBhL9X6A5JlUOr7Euq9Pninp2OFlTuuBSdLWHum8PFSM1NT0YDmdODuW1T/Q+OdT3eyJl8FjdUbSl+NWj20uERZHEKK4pxgOGRvrZ/pz1ezKcHoeCP2ePlP/taPf9EzXmp+rznwy95ok4+6NZy+HOKogkKWkGwFgzNdIcP1V7+Mm4V2JqPI4M6qyf6LthS8pBTERrLPXx8v07oY0pUvdOP/308I1vfCOccsop+tkXv/jFcMcdd4Qrr7xyyGuvvfZasf7B3ke0gEc677zzwlve8pbwile84pA/8/+KMfVE46lQSJ7IzxgLRfvJ+syJum9Q8g9kjDqeeIzmsnoyz8p4MHgdf56sOdwIg2Ps9/1T+fkT9X46VDyRRttEw/8FnSbzl2BMgQ996EOKUH3yk58MW7duVbTpU5/6lAwlSCcaGhrUzLejoyOce+654UUvelG45JJLwtVXXy3Dima/h9Nnyo0px0SD7xvH4cL3jGM08H3jOFz4nnH8XzGmBnPlxiHe//73q7/U61//+vDRj340vOMd75AhBc4++2z1lwL19fXhy1/+crjnnntEhQ5V+le+8pUJ3bDX4XA4HA6Hw+FwjG+M68jUUw2PTDkmGnzfOA4Xvmcco4HvG8fhwveMYyLvmb+YyJTD4XA4HA6Hw+FwjFe4MeVwOBwOh8PhcDgco4AbUw6Hw+FwOBwOh8MxCrgx5XA4HA6Hw+FwOByjgBtTDofD4XA4HA6HwzEKuDHlcDgcDofD4XA4HKOAG1MOh8PhcDgcDofDMQq4MeVwOBwOh8PhcDgco4AbUw6Hw+FwOBwOh8MxCrgx5XA4HA6Hw+FwOByjgBtTDofD4XA4HA6HwzEKuDHlcDgcDofD4XA4HKOAG1MOh8PhcDgcDofDMQrkR/NHf6nIZMb+s8dyDI6JB983jsOF7xnHaOD7xnG48D3jmMh75nDGkCkWi8UnczAOh8PhcDgcDofD8ZcIT/NzOBwOh8PhcDgcjlHAjSmHw+FwOBwOh8PhGAXcmHI4HA6Hw+FwOByOUcCNKYfD4XA4HA6Hw+EYBdyYcjgcDofD4XA4HI5RwI0ph8PhcDgcDofD4RgF3JhyOBwOh8PhcDgcjlHAjSmHw+FwOBwOh8PhGAXcmHI4HA6Hw+FwOByOUcCNqXGAnp6e8IEPfCCccsop4eyzzw7f+MY3xnpIjnGOLVu2hHe+853htNNOC+ecc0741Kc+pX3kcBwKLr300vC+971vrIfhmADo7e0NH/3oR8Opp54azjzzzPBv//ZvoVgsjvWwHOMcmzZtCpdddlk46aSTwnOe85zwrW99a6yH5BjHMubFL35xWLp0aelnjz/+eHjDG94QTjjhhHD++eeH2267LYxn5Md6AI4QPvOZz4SHH344XHHFFWHjxo3hve99b5gxY0Z4wQteMNZDc4xDoMhgSDU2Noarrroq7NmzR8Z4NpvV3nE4DoTrr78+3HzzzeFlL3vZWA/FMQHw8Y9/XErO17/+9bBv377w7ne/W/fTJZdcMtZDc4xjvOtd79I+ue6668LKlSvDe97znjBz5sxw7rnnjvXQHOMIPT094R/+4R/CihUrhug4b3/728ORRx4ZfvjDH4Zf/epX4e/+7u/CDTfcoD01HuGRqTFGZ2dnuOaaa8IHP/jBsGTJEgmav/3bv5WS7HAMh9WrV4f7779f0ahFixYpoolx9bOf/Wysh+YY59i9e7ecN8cdd9xYD8UxQfYLysy//Mu/hKc97Wnh6U9/enjTm94UHnjggbEemmMcAwcfd9Rb3/rWMG/evPC85z1PGRR33HHHWA/NMY6wcuXKcPHFF4d169YN+fmdd96pyNTHPvaxcMQRRyjCSYQKWTRe4cbUGGP58uWhv78/nHjiiaWfnXzyybqsCoXCmI7NMT4xefLk8LWvfS20tbUN+XlHR8eYjckxMfCv//qv4SUveUlYuHDhWA/FMQFwzz33hPr6eqUTp1NEceQ4HCOhuro61NTUKCrV19cnB+C9994bFi9ePNZDc4wj3HXXXeH0008P3//+94f8HP33mGOOCbW1tUP0Ygz08Qo3psYY27ZtCy0tLaGysrL0M5RkQp94BR2OcpDeh5cvAqP7yiuvDGecccaYjssxvoFX+O677w5ve9vbxnoojgkCvMOkZv34xz9W2vlzn/vc8MUvftEdfY4DoqqqKnzoQx+Sknz88ceHF77wheEZz3hGeOUrXznWQ3OMI7z61a9WiQKGd7lePGXKlCE/a21tDZs3bw7jFV4zNcbo6uoaYkiB+D1FeQ7HwfDZz342LFu2LFx77bVjPRTHOAXOmQ9/+MNScPAaOxyHmoa+du3acPXVVysahZLDHkL5Id3P4RgJq1atCs9+9rPDG9/4RtXDkCpKmuiFF1441kNzTFC9uHcc68RuTI0DD075Bonfu9LjOBRDCuKSf//3f1expsMxHL7whS+EY489dkhE0+E4GPL5vNKHP//5zytCBSBJ+t73vufGlOOAUXCcexDdoMdQowkD7f/8z/+4MeU4JL24PDMLvXg868RuTI0xpk6dGnbt2qW6KS4ugPePTUM6l8MxEvD0odRgUD3/+c8f6+E4xjmD3/bt20u1mdFh88tf/jLcd999Yzw6x3iuz0SxiYYUmD9/vmivHY6RADvx3Llzhyi/1MBcfvnlYzoux8TRi1euXDnkZ9xf5al/4wluTI0xKMjEiKKwDla2WPSLJweqa4djpEgDqTf0fHEKfcfB8J3vfEcOm4jPfe5z+j90xQ7HSKDehRTRNWvWyIgCkAmkjSuHoxwovaSH4rSJ6Vrsm1mzZo310BwTRO585StfCd3d3SWDHL0YEorxCtfWxxjknr/0pS8NH/nIR8KDDz4oPn2a9r7uda8b66E5xnEu+pe+9KXw5je/WcKFSGb8cjiGA8ovnuL4VVdXpy/+7XCMhAULFoRnPetZ4f3vf7+YZ2+99VYpOa961avGemiOcQya9FZUVIR/+qd/kiH+m9/8RlGp1772tWM9NMcEwGmnnRamT58uuUO9HTIH/fiiiy4K4xWZorcyHxfFdhhTN954o2ho/+Zv/kadnx2O4YBgoYZhODz66KNP+XgcEw/ve9/79P9Pf/rTYz0UxzjH3r17lVJ80003yfkHAxcNNTOZzFgPzTGOQZrWJz7xCSnBkyZNCq95zWvC61//et83jmFx1FFHhW9/+9uiSgdENum/Ck06Tj9Y/84888wwXuHGlMPhcDgcDofD4XCMAp7m53A4HA6Hw+FwOByjgBtTDofD4XA4HA6HwzEKuDHlcDgcDofD4XA4HKOAG1MOh8PhcDgcDofDMQq4MeVwOBwOh8PhcDgco4AbUw6Hw+FwOBwOh8MxCrgx5XA4HA6Hw+FwOByjgBtTDofD4XA4HA6HwzEKuDHlcDgcjnGFvr6+8N///d/huc99bjj22GPDs571rPCpT30qdHR0lF7znOc8J1x33XVhPOMXv/hFOP7448Njjz12yH/z2te+Vs/ucDgcjomB/FgPwOFwOByOND73uc+F22+/PXz84x8Ps2fPDo8//nj4xCc+EdauXRsuv/xyvebaa68NtbW1YTzjW9/6Vnjf+94XjjzyyLEeisPhcDieJLgx5XA4HI5xhR/96Efhk5/8ZHj605+u72fNmhU+8pGPhNe85jVh69atYcqUKWHSpElhvOOrX/1qaGhoGOthOBwOh+NJhKf5ORwOh2NcIZPJhDvvvDMUCoXSz0488cRw/fXXh5aWlv3S/Hgd0azTTz9dX1/60pfCueeeG5YuXarfH3XUUeHnP/95eOELX6i0u7//+79XtOt1r3udvn/1q18dtmzZotcWi0VFv3h/UgzPPvvs8IUvfGHEsfK6z372s3rdS1/6Uv09aX2k65111lnh+c9/frjqqquG/M1NN90Uzj//fH32RRddFO66664nZR4dDofD8eTDjSmHw+FwjCtg5HznO9+RofLhD384/PKXvwzd3d1h4cKFoaKiYr/Xf/nLXw4//vGPw+c///nwzW9+M/zud7+TsZTGf/3Xf4VPf/rTeu2NN94YXvWqV+nr6quvDtu2bVMUCfA+V1xxhdIKqXl6+9vfrhqmRx55ZMTx/vSnPw1f//rX9f49PT3hzW9+czj55JPDT37yk/De975Xxh3vC5YvX66fvfWtb9XvL7zwQr2eFEaHw+FwTDy4MeVwOByOcQUMGKI906ZNCz/4wQ/CO9/5znDOOeeEH/7wh8O+/rvf/W5417vepejQMcccI6OGCFEab3jDGxQJOuOMM8LixYvDmWeeqUgV/z7vvPPCmjVr9Lrp06eL7IIUQ9ILMbgmT54cVqxYMeJ4MYiIfh199NEyrFpbWzWeefPmySB8y1veEr797W/rtRhdF198cbjgggvC3LlzZTg+4xnPCN/73vee0Dl0OBwOx1MDr5lyOBwOx7gDBgpfu3btCrfddlu48sorwwc/+EEZLaTfRezcuVN1VMcdd1zpZwsWLAhNTU1D3g8ii4jq6uowc+bMId/39vbq3xhbDzzwgKJcq1atCn/84x8VuUqnHJYj/V6rV69W9Im0xIiBgYGQy+X0b96TlMPvf//7Q9gLMQQdDofDMfHgxpTD4XA4xg0wREiJgwUPUCNFFIfaIyJI1FKljal83q6x8khU+ffRmInIZodPzLjmmmtEfvHKV75Sn0dKHtGjA6Gqqqr07/7+fkW1PvShDw37Wgwr0vqor0oDg87hcDgcEw+e5udwOByOcQOMDeqeli1bNuTnlZWVMjjKWfwaGxvF7peuaaJeqr29fVSfT7odaYYf+MAHZPBgzO3YsWM/42wkzJ8/XymDpAiSxsfX/fffrxqw+Pv169eXfscXUapbbrllVON1OBwOx9jCjSmHw+FwjBssWbJETXrf9ra3qf4IwwNjBCIKUvGIFpUD5jwIJu644w5Ftt7//veXWAEPFxhPvA8G0cMPPxze/e53Kw0vpgEeDKQmQpZBZIqUvptvvllkFtRRxdqtG264QTVU69atUy8qvqivcjgcDsfEg6f5ORwOh2Nc4T/+4z9ETw4l+caNG9Wcl5oi6qbq6+v3e/2b3vQm1U294x3vUDrfpZdeGu6+++5hmf8OBiJSfL3kJS+RAQRJRU1NjWqnDgWMD2ZAUgWJbDU3N6s/1mWXXabfn3DCCeEzn/mMGAL5/5w5c1Sfdeqppx72WB0Oh8Mx9sgUDzV3weFwOByOcQhS5KijiimAkFJQt/TrX/9a6XYOh8PhcDxZcGPK4XA4HBMa1DhRa/We97xHqX3/+Z//qYjWtddeO9ZDczgcDsdfOLxmyuFwOBwTGtQnwc53ySWXqIcTNOZf/OIXx3pYDofD4fg/AI9MORwOh8PhcDgcDsco4JEph8PhcDgcDofD4RgF3JhyOBwOh8PhcDgcjlHAjSmHw+FwOBwOh8PhGAXcmHI4HA6Hw+FwOByOUcCNKYfD4XA4HA6Hw+EYBdyYcjgcDofD4XA4HI5RwI0ph8PhcDgcDofD4RgF3JhyOBwOh8PhcDgcjnD4+P8kldi/E+T0RAAAAABJRU5ErkJggg=="
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Epoch 13/100: 100%|██████████| 80/80 [00:15<00:00,  5.23it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 13/100, Train Loss: 0.0575, Val Loss: 0.0514\n",
-      "Saved best model with Val Loss: 0.0514\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 14/100: 100%|██████████| 80/80 [00:14<00:00,  5.42it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 14/100, Train Loss: 0.0581, Val Loss: 0.0537\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 15/100: 100%|██████████| 80/80 [00:15<00:00,  5.17it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 15/100, Train Loss: 0.0600, Val Loss: 0.0504\n",
-      "Saved best model with Val Loss: 0.0504\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 16/100: 100%|██████████| 80/80 [00:16<00:00,  4.97it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 16/100, Train Loss: 0.0601, Val Loss: 0.0569\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 17/100: 100%|██████████| 80/80 [00:16<00:00,  4.86it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 17/100, Train Loss: 0.0619, Val Loss: 0.0505\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 18/100: 100%|██████████| 80/80 [00:15<00:00,  5.16it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 18/100, Train Loss: 0.0627, Val Loss: 0.0679\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 19/100: 100%|██████████| 80/80 [00:15<00:00,  5.11it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 19/100, Train Loss: 0.0531, Val Loss: 0.0447\n",
-      "Saved best model with Val Loss: 0.0447\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 20/100: 100%|██████████| 80/80 [00:16<00:00,  4.77it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 20/100, Train Loss: 0.0560, Val Loss: 0.0462\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 21/100: 100%|██████████| 80/80 [00:15<00:00,  5.22it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 21/100, Train Loss: 0.0513, Val Loss: 0.0509\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 22/100: 100%|██████████| 80/80 [00:15<00:00,  5.17it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 22/100, Train Loss: 0.0514, Val Loss: 0.0474\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 23/100: 100%|██████████| 80/80 [00:15<00:00,  5.26it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 23/100, Train Loss: 0.0530, Val Loss: 0.0546\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 24/100: 100%|██████████| 80/80 [00:14<00:00,  5.42it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 24/100, Train Loss: 0.0498, Val Loss: 0.0610\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 25/100: 100%|██████████| 80/80 [00:14<00:00,  5.34it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 25/100, Train Loss: 0.0506, Val Loss: 0.0485\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 26/100: 100%|██████████| 80/80 [00:15<00:00,  5.11it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 26/100, Train Loss: 0.0464, Val Loss: 0.0442\n",
-      "Saved best model with Val Loss: 0.0442\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 27/100: 100%|██████████| 80/80 [00:15<00:00,  5.10it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 27/100, Train Loss: 0.0520, Val Loss: 0.0524\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 28/100: 100%|██████████| 80/80 [00:15<00:00,  5.21it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 28/100, Train Loss: 0.0497, Val Loss: 0.0487\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 29/100: 100%|██████████| 80/80 [00:15<00:00,  5.20it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 29/100, Train Loss: 0.0483, Val Loss: 0.0395\n",
-      "Saved best model with Val Loss: 0.0395\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 30/100: 100%|██████████| 80/80 [00:15<00:00,  5.12it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 30/100, Train Loss: 0.0492, Val Loss: 0.0423\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 31/100: 100%|██████████| 80/80 [00:15<00:00,  5.09it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 31/100, Train Loss: 0.0481, Val Loss: 0.0473\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 32/100: 100%|██████████| 80/80 [00:15<00:00,  5.01it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 32/100, Train Loss: 0.0455, Val Loss: 0.0470\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 33/100: 100%|██████████| 80/80 [00:16<00:00,  4.81it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 33/100, Train Loss: 0.0445, Val Loss: 0.0444\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 34/100: 100%|██████████| 80/80 [00:16<00:00,  4.92it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 34/100, Train Loss: 0.0438, Val Loss: 0.0376\n",
-      "Saved best model with Val Loss: 0.0376\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 35/100: 100%|██████████| 80/80 [00:16<00:00,  4.89it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 35/100, Train Loss: 0.0469, Val Loss: 0.0425\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 36/100: 100%|██████████| 80/80 [00:16<00:00,  4.87it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 36/100, Train Loss: 0.0438, Val Loss: 0.0381\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 37/100: 100%|██████████| 80/80 [00:15<00:00,  5.07it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 37/100, Train Loss: 0.0427, Val Loss: 0.0382\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 38/100: 100%|██████████| 80/80 [00:15<00:00,  5.17it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 38/100, Train Loss: 0.0439, Val Loss: 0.0364\n",
-      "Saved best model with Val Loss: 0.0364\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 39/100: 100%|██████████| 80/80 [00:15<00:00,  5.20it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 39/100, Train Loss: 0.0448, Val Loss: 0.0426\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 40/100: 100%|██████████| 80/80 [00:15<00:00,  5.15it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 40/100, Train Loss: 0.0397, Val Loss: 0.0361\n",
-      "Saved best model with Val Loss: 0.0361\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 41/100: 100%|██████████| 80/80 [00:15<00:00,  5.20it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 41/100, Train Loss: 0.0405, Val Loss: 0.0496\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 42/100: 100%|██████████| 80/80 [00:15<00:00,  5.15it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 42/100, Train Loss: 0.0421, Val Loss: 0.0399\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 43/100: 100%|██████████| 80/80 [00:16<00:00,  4.78it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 43/100, Train Loss: 0.0412, Val Loss: 0.0446\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 44/100: 100%|██████████| 80/80 [00:16<00:00,  4.93it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 44/100, Train Loss: 0.0422, Val Loss: 0.0381\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 45/100: 100%|██████████| 80/80 [00:17<00:00,  4.69it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 45/100, Train Loss: 0.0392, Val Loss: 0.0344\n",
-      "Saved best model with Val Loss: 0.0344\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 46/100: 100%|██████████| 80/80 [00:15<00:00,  5.18it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 46/100, Train Loss: 0.0432, Val Loss: 0.0378\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 47/100: 100%|██████████| 80/80 [00:15<00:00,  5.01it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 47/100, Train Loss: 0.0414, Val Loss: 0.0382\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 48/100: 100%|██████████| 80/80 [00:14<00:00,  5.48it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 48/100, Train Loss: 0.0393, Val Loss: 0.0336\n",
-      "Saved best model with Val Loss: 0.0336\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 49/100: 100%|██████████| 80/80 [00:16<00:00,  5.00it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 49/100, Train Loss: 0.0386, Val Loss: 0.0377\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 50/100: 100%|██████████| 80/80 [00:15<00:00,  5.16it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 50/100, Train Loss: 0.0366, Val Loss: 0.0338\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 51/100: 100%|██████████| 80/80 [00:15<00:00,  5.16it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 51/100, Train Loss: 0.0378, Val Loss: 0.0393\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 52/100: 100%|██████████| 80/80 [00:14<00:00,  5.40it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 52/100, Train Loss: 0.0400, Val Loss: 0.0334\n",
-      "Saved best model with Val Loss: 0.0334\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 53/100: 100%|██████████| 80/80 [00:15<00:00,  5.12it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 53/100, Train Loss: 0.0409, Val Loss: 0.0378\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 54/100: 100%|██████████| 80/80 [00:16<00:00,  4.99it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 54/100, Train Loss: 0.0383, Val Loss: 0.0335\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 55/100: 100%|██████████| 80/80 [00:17<00:00,  4.62it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 55/100, Train Loss: 0.0359, Val Loss: 0.0334\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 56/100: 100%|██████████| 80/80 [00:16<00:00,  4.94it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 56/100, Train Loss: 0.0392, Val Loss: 0.0342\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 57/100: 100%|██████████| 80/80 [00:15<00:00,  5.25it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 57/100, Train Loss: 0.0377, Val Loss: 0.0333\n",
-      "Saved best model with Val Loss: 0.0333\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 58/100: 100%|██████████| 80/80 [00:16<00:00,  4.89it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 58/100, Train Loss: 0.0376, Val Loss: 0.0317\n",
-      "Saved best model with Val Loss: 0.0317\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 59/100: 100%|██████████| 80/80 [00:16<00:00,  4.96it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 59/100, Train Loss: 0.0346, Val Loss: 0.0284\n",
-      "Saved best model with Val Loss: 0.0284\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 60/100: 100%|██████████| 80/80 [00:15<00:00,  5.04it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 60/100, Train Loss: 0.0361, Val Loss: 0.0277\n",
-      "Saved best model with Val Loss: 0.0277\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 61/100: 100%|██████████| 80/80 [00:15<00:00,  5.08it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 61/100, Train Loss: 0.0346, Val Loss: 0.0521\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 62/100: 100%|██████████| 80/80 [00:15<00:00,  5.12it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 62/100, Train Loss: 0.0351, Val Loss: 0.0356\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 63/100: 100%|██████████| 80/80 [00:15<00:00,  5.12it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 63/100, Train Loss: 0.0335, Val Loss: 0.0339\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 64/100: 100%|██████████| 80/80 [00:15<00:00,  5.14it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 64/100, Train Loss: 0.0344, Val Loss: 0.0284\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 65/100: 100%|██████████| 80/80 [00:15<00:00,  5.20it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 65/100, Train Loss: 0.0347, Val Loss: 0.0303\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 66/100: 100%|██████████| 80/80 [00:15<00:00,  5.10it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 66/100, Train Loss: 0.0344, Val Loss: 0.0304\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 67/100: 100%|██████████| 80/80 [00:15<00:00,  5.10it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 67/100, Train Loss: 0.0337, Val Loss: 0.0275\n",
-      "Saved best model with Val Loss: 0.0275\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 68/100: 100%|██████████| 80/80 [00:15<00:00,  5.04it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 68/100, Train Loss: 0.0327, Val Loss: 0.0343\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 69/100: 100%|██████████| 80/80 [00:15<00:00,  5.09it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 69/100, Train Loss: 0.0334, Val Loss: 0.0264\n",
-      "Saved best model with Val Loss: 0.0264\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 70/100: 100%|██████████| 80/80 [00:15<00:00,  5.01it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 70/100, Train Loss: 0.0324, Val Loss: 0.0294\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 71/100: 100%|██████████| 80/80 [00:16<00:00,  4.86it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 71/100, Train Loss: 0.0322, Val Loss: 0.0336\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 72/100: 100%|██████████| 80/80 [00:16<00:00,  4.85it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 72/100, Train Loss: 0.0319, Val Loss: 0.0318\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 73/100: 100%|██████████| 80/80 [00:16<00:00,  4.95it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 73/100, Train Loss: 0.0320, Val Loss: 0.0276\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 74/100: 100%|██████████| 80/80 [00:15<00:00,  5.01it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 74/100, Train Loss: 0.0320, Val Loss: 0.0289\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 75/100: 100%|██████████| 80/80 [00:16<00:00,  4.84it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 75/100, Train Loss: 0.0296, Val Loss: 0.0291\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 76/100: 100%|██████████| 80/80 [00:15<00:00,  5.22it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 76/100, Train Loss: 0.0324, Val Loss: 0.0281\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 77/100: 100%|██████████| 80/80 [00:15<00:00,  5.33it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 77/100, Train Loss: 0.0310, Val Loss: 0.0312\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 78/100: 100%|██████████| 80/80 [00:15<00:00,  5.01it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 78/100, Train Loss: 0.0294, Val Loss: 0.0300\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 79/100: 100%|██████████| 80/80 [00:15<00:00,  5.20it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 79/100, Train Loss: 0.0324, Val Loss: 0.0245\n",
-      "Saved best model with Val Loss: 0.0245\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 80/100: 100%|██████████| 80/80 [00:15<00:00,  5.24it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 80/100, Train Loss: 0.0317, Val Loss: 0.0246\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 81/100: 100%|██████████| 80/80 [00:16<00:00,  5.00it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 81/100, Train Loss: 0.0310, Val Loss: 0.0253\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 82/100: 100%|██████████| 80/80 [00:15<00:00,  5.21it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 82/100, Train Loss: 0.0294, Val Loss: 0.0259\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 83/100: 100%|██████████| 80/80 [00:15<00:00,  5.00it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 83/100, Train Loss: 0.0290, Val Loss: 0.0280\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 84/100: 100%|██████████| 80/80 [00:16<00:00,  4.92it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 84/100, Train Loss: 0.0308, Val Loss: 0.0263\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 85/100: 100%|██████████| 80/80 [00:16<00:00,  4.74it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 85/100, Train Loss: 0.0292, Val Loss: 0.0281\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 86/100: 100%|██████████| 80/80 [00:16<00:00,  5.00it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 86/100, Train Loss: 0.0300, Val Loss: 0.0228\n",
-      "Saved best model with Val Loss: 0.0228\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 87/100: 100%|██████████| 80/80 [00:16<00:00,  4.76it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 87/100, Train Loss: 0.0293, Val Loss: 0.0260\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 88/100: 100%|██████████| 80/80 [00:16<00:00,  4.95it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 88/100, Train Loss: 0.0283, Val Loss: 0.0271\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 89/100: 100%|██████████| 80/80 [00:15<00:00,  5.20it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 89/100, Train Loss: 0.0290, Val Loss: 0.0255\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 90/100: 100%|██████████| 80/80 [00:15<00:00,  5.09it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 90/100, Train Loss: 0.0295, Val Loss: 0.0261\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 91/100: 100%|██████████| 80/80 [00:15<00:00,  5.13it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 91/100, Train Loss: 0.0275, Val Loss: 0.0233\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 92/100: 100%|██████████| 80/80 [00:16<00:00,  4.99it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 92/100, Train Loss: 0.0295, Val Loss: 0.0279\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 93/100: 100%|██████████| 80/80 [00:15<00:00,  5.02it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 93/100, Train Loss: 0.0277, Val Loss: 0.0247\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 94/100: 100%|██████████| 80/80 [00:16<00:00,  4.88it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 94/100, Train Loss: 0.0282, Val Loss: 0.0225\n",
-      "Saved best model with Val Loss: 0.0225\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 95/100: 100%|██████████| 80/80 [00:15<00:00,  5.12it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 95/100, Train Loss: 0.0284, Val Loss: 0.0265\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 96/100: 100%|██████████| 80/80 [00:17<00:00,  4.59it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 96/100, Train Loss: 0.0277, Val Loss: 0.0255\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 97/100: 100%|██████████| 80/80 [00:15<00:00,  5.04it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 97/100, Train Loss: 0.0288, Val Loss: 0.0257\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 98/100: 100%|██████████| 80/80 [00:15<00:00,  5.21it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 98/100, Train Loss: 0.0293, Val Loss: 0.0233\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 99/100: 100%|██████████| 80/80 [00:15<00:00,  5.05it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 99/100, Train Loss: 0.0291, Val Loss: 0.0239\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Epoch 100/100: 100%|██████████| 80/80 [00:15<00:00,  5.14it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Epoch 100/100, Train Loss: 0.0278, Val Loss: 0.0256\n"
+      "C:\\Users\\birli\\AppData\\Local\\Temp\\ipykernel_9948\\800325767.py:72: MatplotlibDeprecationWarning: The 'labels' parameter of boxplot() has been renamed 'tick_labels' since Matplotlib 3.9; support for the old name will be dropped in 3.11.\n",
+      "  plt.boxplot(binned_errors, labels=[f'{sigma_bins[i]:.1f}-{sigma_bins[i+1]:.1f}'\n"
      ]
     },
     {
      "data": {
       "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
+       "<Figure size 1000x600 with 1 Axes>"
       ],
-      "image/png": "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"
+      "image/png": "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"
      },
      "metadata": {},
      "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Training completed!\n"
-     ]
     }
    ],
-   "execution_count": 16
+   "execution_count": 15
   },
   {
+   "metadata": {},
    "cell_type": "code",
-   "id": "eadfcd14ab426dbf",
-   "metadata": {
-    "ExecuteTime": {
-     "end_time": "2025-05-16T06:21:28.293948Z",
-     "start_time": "2025-05-16T06:21:28.284718Z"
-    }
-   },
-   "source": [
-    "def evaluate_model(model, test_loader, criterion):\n",
-    "    model.eval()\n",
-    "    test_loss = 0\n",
-    "    predictions = []\n",
-    "    ground_truths = []\n",
-    "\n",
-    "    with torch.no_grad():\n",
-    "        for images, sigmas in tqdm(test_loader, desc=\"Evaluating\"):\n",
-    "            images, sigmas = images.to(device), sigmas.to(device)\n",
-    "            outputs = model(images)\n",
-    "            loss = criterion(outputs, sigmas)\n",
-    "            test_loss += loss.item() * images.size(0)\n",
-    "\n",
-    "            # The normalization in BlurDataset is: sigma_normalized = (sigma - sigma_min) / (sigma_max - sigma_min)\n",
-    "            # So, to denormalize: sigma = sigma_normalized * (sigma_max - sigma_min) + sigma_min\n",
-    "            outputs_denorm = outputs.cpu().numpy() * sigma_std + sigma_mean\n",
-    "            sigmas_denorm = sigmas.cpu().numpy() * sigma_std + sigma_mean\n",
-    "\n",
-    "            predictions.extend(outputs_denorm)\n",
-    "            ground_truths.extend(sigmas_denorm)\n",
-    "\n",
-    "    test_loss /= len(test_loader.dataset)\n",
-    "    predictions = np.array(predictions)\n",
-    "    ground_truths = np.array(ground_truths)\n",
-    "\n",
-    "    return test_loss, predictions, ground_truths\n",
-    "\n",
-    "\n",
-    "def plot_results(test_loss, predictions, ground_truths, save_dir=\"plots\"):\n",
-    "    # Créer le répertoire pour sauvegarder les plots\n",
-    "    Path(save_dir).mkdir(exist_ok=True)\n",
-    "\n",
-    "    # 1. Scatter plot : Prédictions vs Valeurs réelles\n",
-    "    plt.figure(figsize=(10, 6))\n",
-    "    plt.scatter(ground_truths, predictions, alpha=0.5, s=10)\n",
-    "    plt.plot([0, 10], [0, 10], 'r--', label='Ligne idéale')\n",
-    "    plt.xlabel('Sigma réel')\n",
-    "    plt.ylabel('Sigma prédit')\n",
-    "    plt.title(f'Prédictions vs Valeurs réelles (Test Loss: {test_loss:.4f})')\n",
-    "    plt.legend()\n",
-    "    plt.grid(True)\n",
-    "    plt.savefig(Path(save_dir) / 'predictions_vs_ground_truth.png')\n",
-    "    plt.close()\n",
-    "\n",
-    "\n",
-    "    # 2. Histogramme des erreurs\n",
-    "    errors = predictions - ground_truths\n",
-    "    plt.figure(figsize=(10, 6))\n",
-    "    sns.histplot(errors, bins=50, kde=True)\n",
-    "    plt.xlabel('Erreur (Prédit - Réel)')\n",
-    "    plt.ylabel('Fréquence')\n",
-    "    plt.title('Distribution des erreurs de prédiction')\n",
-    "    plt.grid(True)\n",
-    "    plt.savefig(Path(save_dir) / 'error_distribution.png')\n",
-    "    plt.close()\n",
-    "\n",
-    "\n",
-    "    # 3. Erreur absolue en fonction de sigma réel\n",
-    "    abs_errors = np.abs(errors)\n",
-    "    plt.figure(figsize=(10, 6))\n",
-    "    plt.scatter(ground_truths, abs_errors, alpha=0.5, s=10)\n",
-    "    plt.xlabel('Sigma réel')\n",
-    "    plt.ylabel('Erreur absolue')\n",
-    "    plt.title('Erreur absolue en fonction de Sigma réel')\n",
-    "    plt.grid(True)\n",
-    "    plt.savefig(Path(save_dir) / 'absolute_error_vs_ground_truth.png')\n",
-    "    plt.close()\n",
-    "\n",
-    "    # 4. Boxplot des erreurs par plage de sigma\n",
-    "    sigma_bins = np.linspace(0, 10, 11)\n",
-    "    digitized = np.digitize(ground_truths, sigma_bins)\n",
-    "    binned_errors = [abs_errors[digitized == i] for i in range(1, len(sigma_bins))]\n",
-    "\n",
-    "    plt.figure(figsize=(10, 6))\n",
-    "    plt.boxplot(binned_errors, labels=[f'{sigma_bins[i]:.1f}-{sigma_bins[i+1]:.1f}'\n",
-    "                                     for i in range(len(sigma_bins)-1)])\n",
-    "    plt.xlabel('Plage de Sigma réel')\n",
-    "    plt.ylabel('Erreur absolue')\n",
-    "    plt.title('Distribution des erreurs par plage de Sigma')\n",
-    "    plt.xticks(rotation=45)\n",
-    "    plt.grid(True)\n",
-    "    plt.tight_layout()\n",
-    "    plt.savefig(Path(save_dir) / 'error_by_sigma_range.png')\n",
-    "    plt.close()"
-   ],
    "outputs": [],
-   "execution_count": 17
-  },
-  {
-   "cell_type": "code",
-   "id": "6985f26ea8078f2d",
-   "metadata": {
-    "ExecuteTime": {
-     "end_time": "2025-05-16T06:21:31.710039Z",
-     "start_time": "2025-05-16T06:21:28.335271Z"
-    }
-   },
-   "source": [
-    "    # Hyperparamètres\n",
-    "    batch_size = 128\n",
-    "    sigma_range = (0, 10)\n",
-    "    test_size = 2000\n",
-    "\n",
-    "    # Charger le modèle entraîné\n",
-    "    # model = BlurRegressionCNN().to(device)\n",
-    "    global model\n",
-    "    # model_path = \"best_blur_model.pth\"\n",
-    "    # if not Path(model_path).exists():\n",
-    "    #     raise FileNotFoundError(f\"Modèle non trouvé à {model_path}\")\n",
-    "    # model.load_state_dict(torch.load(model_path))\n",
-    "    # print(f\"Modèle chargé depuis {model_path}\")\n",
-    "\n",
-    "    # Créer le dataset de test\n",
-    "    test_dataset = BlurDataset(length=test_size)\n",
-    "    test_loader = DataLoader(\n",
-    "        test_dataset,\n",
-    "        batch_size=batch_size,\n",
-    "        shuffle=False,\n",
-    "        num_workers=8,\n",
-    "        pin_memory=True,\n",
-    "    )\n",
-    "\n",
-    "    # Critère de perte\n",
-    "    criterion = nn.MSELoss()\n",
-    "\n",
-    "    # Évaluer le modèle\n",
-    "    test_loss, predictions, ground_truths = evaluate_model(model, test_loader, criterion)\n",
-    "\n",
-    "    # Afficher les résultats\n",
-    "    print(f\"Test Loss: {test_loss:.4f}\")\n",
-    "    print(f\"MAE: {np.mean(np.abs(predictions - ground_truths)):.4f}\")\n",
-    "    print(f\"RMSE: {np.sqrt(np.mean((predictions - ground_truths)**2)):.4f}\")\n",
-    "\n",
-    "    # Visualiser les résultats\n",
-    "    plot_results(test_loss, predictions, ground_truths)"
-   ],
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Evaluating: 100%|██████████| 16/16 [00:02<00:00,  5.69it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Test Loss: 0.0241\n",
-      "MAE: 0.5091\n",
-      "RMSE: 0.7755\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/tmp/ipykernel_39455/2777009023.py:75: MatplotlibDeprecationWarning: The 'labels' parameter of boxplot() has been renamed 'tick_labels' since Matplotlib 3.9; support for the old name will be dropped in 3.11.\n",
-      "  plt.boxplot(binned_errors, labels=[f'{sigma_bins[i]:.1f}-{sigma_bins[i+1]:.1f}'\n"
-     ]
-    }
-   ],
-   "execution_count": 18
-  },
-  {
-   "metadata": {
-    "ExecuteTime": {
-     "end_time": "2025-05-16T06:21:31.860426Z",
-     "start_time": "2025-05-16T06:21:31.722967Z"
-    }
-   },
-   "cell_type": "code",
-   "source": [
-    "test_dataset = BlurDataset(length=1)\n",
-    "plt.imshow(test_dataset.__getitem__(0)[0][0], cmap='gray')"
-   ],
-   "id": "acf4f38dcc622089",
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "<matplotlib.image.AxesImage at 0x7b8bd431a810>"
-      ]
-     },
-     "execution_count": 19,
-     "metadata": {},
-     "output_type": "execute_result"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ],
-      "image/png": "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"
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "execution_count": 19
+   "execution_count": null,
+   "source": "evaluate_model(model, test_loader, criterion)",
+   "id": "eca2586c16e71db5"
   },
   {
-   "metadata": {
-    "ExecuteTime": {
-     "end_time": "2025-05-16T05:43:16.005858418Z",
-     "start_time": "2025-05-15T20:20:50.761564Z"
-    }
-   },
+   "metadata": {},
    "cell_type": "code",
-   "source": "",
-   "id": "552a62e57e926d7c",
    "outputs": [],
-   "execution_count": null
-  },
-  {
-   "metadata": {
-    "ExecuteTime": {
-     "end_time": "2025-05-16T05:43:16.007187298Z",
-     "start_time": "2025-05-15T20:20:50.812330Z"
-    }
-   },
-   "cell_type": "code",
+   "execution_count": null,
    "source": "",
-   "id": "7e5084dd15accf74",
-   "outputs": [],
-   "execution_count": null
+   "id": "21039bca62958b11"
   },
   {
-   "metadata": {
-    "ExecuteTime": {
-     "end_time": "2025-05-16T05:43:16.008832755Z",
-     "start_time": "2025-05-15T20:20:50.874065Z"
-    }
-   },
+   "metadata": {},
    "cell_type": "code",
-   "source": "",
-   "id": "11b32ce1db90eccd",
    "outputs": [],
-   "execution_count": null
+   "execution_count": null,
+   "source": "",
+   "id": "df34fb8fd7af5587"
   },
   {
-   "metadata": {
-    "ExecuteTime": {
-     "end_time": "2025-05-16T05:43:16.009372872Z",
-     "start_time": "2025-05-15T20:20:50.934736Z"
-    }
-   },
+   "metadata": {},
    "cell_type": "code",
-   "source": "",
-   "id": "833120b7fd616c3e",
    "outputs": [],
-   "execution_count": null
+   "execution_count": null,
+   "source": "",
+   "id": "f1fb23720efbb534"
   },
   {
-   "metadata": {
-    "ExecuteTime": {
-     "end_time": "2025-05-16T05:43:16.014226874Z",
-     "start_time": "2025-05-15T20:20:50.993258Z"
-    }
-   },
+   "metadata": {},
    "cell_type": "code",
-   "source": "",
-   "id": "6181b358984f94b8",
    "outputs": [],
-   "execution_count": null
+   "execution_count": null,
+   "source": "",
+   "id": "cd0640f07fcebf3e"
   },
   {
    "metadata": {},
@@ -1964,7 +489,7 @@
    "outputs": [],
    "execution_count": null,
    "source": "",
-   "id": "69d89c5167b76f90"
+   "id": "df85fdfb8f0088b1"
   }
  ],
  "metadata": {
diff --git a/trainee.py b/trainee.py
new file mode 100644
index 0000000..af9f62e
--- /dev/null
+++ b/trainee.py
@@ -0,0 +1,247 @@
+from data_gen import generate_random_shapes_image, blur_image
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import Dataset, DataLoader
+from torchvision import transforms
+from PIL import Image
+import numpy as np
+import cv2
+from pathlib import Path
+from tqdm import tqdm
+import torch.amp
+import seaborn as sns
+
+from matplotlib import pyplot as plt
+
+# Set random seed for reproducibility
+torch.manual_seed(42)
+np.random.seed(42)
+
+# Device configuration
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+print(device)
+
+transform = transforms.Compose(
+    [transforms.ToTensor(), transforms.Normalize(mean=[0.5], std=[0.5])]
+)
+sigma_range = (0, 30)
+sigma_mean = (sigma_range[1] - sigma_range[0]) / 2
+sigma_std = (sigma_range[1] - sigma_range[0]) / 2
+import os
+from pathlib import Path
+import h5py
+
+
+def precompute_dataset(length=12800, save_dir="blur_dataset"):
+    Path(save_dir).mkdir(exist_ok=True)
+    with h5py.File(f"{save_dir}/dataset.h5", "w") as f:
+        images = f.create_dataset("images", (length, 1, 128, 128), dtype=np.float32)
+        sigmas = f.create_dataset("sigmas", (length,), dtype=np.float32)
+
+        for i in tqdm(range(length), desc="Precomputing dataset"):
+            image = generate_random_shapes_image()
+            image_np = np.asarray(image)
+            sigma = np.random.uniform(sigma_range[0], sigma_range[1])
+            blurred = blur_image(image_np, sigma)
+            blurred_pil = Image.fromarray(blurred)
+            blurred_tensor = transform(blurred_pil)
+            images[i] = blurred_tensor.numpy()
+            sigmas[i] = (sigma - sigma_mean) / sigma_std
+
+    return save_dir
+
+
+class PrecomputedBlurDataset(Dataset):
+    def __init__(self, h5_path):
+        self.h5_file = h5py.File(h5_path, "r")
+        self.images = self.h5_file["images"]
+        self.sigmas = self.h5_file["sigmas"]
+
+    def __len__(self):
+        return len(self.images)
+
+    def __getitem__(self, idx):
+        image = torch.from_numpy(self.images[idx])
+        sigma = torch.tensor(self.sigmas[idx], dtype=torch.float32)
+        return image, sigma
+
+    def __del__(self):
+        self.h5_file.close()
+
+
+# In main():
+
+# Custom Dataset
+# class BlurDataset(Dataset):
+#     def __init__(self, length=10000):
+#         self.length = length
+#
+#     def __len__(self):
+#         return self.length
+#
+#     def __getitem__(self, idx):
+#         image = generate_random_shapes_image()
+#
+#         # Convert to numpy for OpenCV blur
+#         image_np = np.asarray(image)
+#
+#         # Generate random sigma and apply Gaussian blur
+#         sigma = np.random.uniform(sigma_range[0], sigma_range[1])
+#         blurred = blur_image(image_np, sigma)
+#         blurred_pil = Image.fromarray(blurred)  # Convert back to PIL for transforms
+#
+#         blurred_tensor = transform(blurred_pil)
+#
+#         if blurred_tensor.shape != (1, 128, 128):
+#             raise ValueError(f"Unexpected tensor shape: {blurred_tensor.shape}")
+#
+#         # Normalize sigma to be between 0 and 1
+#
+#         sigma_normalized = (sigma - sigma_mean) / sigma_std
+#         return blurred_tensor, torch.tensor(sigma_normalized, dtype=torch.float32)
+
+
+# CNN Model
+class BlurRegressionCNN(nn.Module):
+    def __init__(self):
+        super(BlurRegressionCNN, self).__init__()
+        self.features = nn.Sequential(
+            nn.Conv2d(1, 16, kernel_size=3, padding=1),
+            nn.ReLU(),
+            nn.BatchNorm2d(16),
+            nn.MaxPool2d(2),
+            nn.Conv2d(16, 32, kernel_size=3, padding=1),
+            nn.ReLU(),
+            nn.BatchNorm2d(32),
+            nn.MaxPool2d(32),
+            nn.Conv2d(32, 64, kernel_size=3, padding=1),
+            nn.ReLU(),
+            nn.BatchNorm2d(64),
+            nn.MaxPool2d(2),
+        )
+        self.regressor = nn.Sequential(
+            nn.Flatten(),
+            nn.Linear(64, 512),
+            nn.ReLU(),
+            nn.Dropout(0.3),
+            nn.Linear(512, 1),
+        )
+
+    def forward(self, x):
+        x = self.features(x)
+        x = self.regressor(x)
+        return x.squeeze(-1)
+
+
+# Training function
+def train_model(model, train_loader, val_loader, num_epochs=10, lr=1e-3):
+    criterion = nn.MSELoss()
+    optimizer = optim.AdamW(model.parameters(), lr=lr, weight_decay=1e-2)
+    scheduler = optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=num_epochs)
+    scaler = torch.amp.GradScaler("cuda")
+
+    best_val_loss = float("inf")
+    model_path = "best_blur_model.pth"
+
+    losses = []
+
+    for epoch in range(num_epochs):
+        model.train()
+        train_loss = 0
+        for images, sigmas in tqdm(
+            train_loader, desc=f"Epoch {epoch + 1}/{num_epochs}"
+        ):
+            images, sigmas = images.to(device), sigmas.to(device)
+
+            optimizer.zero_grad()
+            with torch.amp.autocast("cuda"):
+                outputs = model(images)
+                loss = criterion(outputs, sigmas)
+
+            scaler.scale(loss).backward()
+            scaler.step(optimizer)
+            scaler.update()
+
+            train_loss += loss.item() * images.size(0)
+
+        train_loss /= len(train_loader.dataset)
+
+        # Validation
+        model.eval()
+        val_loss = 0
+        with torch.no_grad():
+            for images, sigmas in val_loader:
+                images, sigmas = images.to(device), sigmas.to(device)
+                outputs = model(images)
+                loss = criterion(outputs, sigmas)
+                val_loss += loss.item() * images.size(0)
+
+        val_loss /= len(val_loader.dataset)
+        losses.append((train_loss, val_loss))
+        print(
+            f"Epoch {epoch + 1}/{num_epochs}, Train Loss: {train_loss:.4f}, Val Loss: {val_loss:.4f}"
+        )
+
+        # Save best model
+        if val_loss < best_val_loss:
+            best_val_loss = val_loss
+            torch.save(model.state_dict(), model_path)
+            print(f"Saved best model with Val Loss: {val_loss:.4f}")
+
+        scheduler.step()
+    plt.plot([l[0] for l in losses], label="Training Loss")
+    plt.plot([l[1] for l in losses], label="Validation Loss")
+    plt.legend()
+    plt.yscale("log")
+    plt.show()
+
+    return model
+
+
+def main():
+    # Hyperparameters
+    batch_size = 128
+    num_epochs = 200
+    learning_rate = 1e-3
+    val_split = 0.2
+
+    # Dataset
+    # dataset = BlurDataset(length=12800)
+    save_dir = precompute_dataset(length=12800)
+    dataset = PrecomputedBlurDataset(f"{save_dir}/dataset.h5")
+
+
+    # Split into train and validation
+    train_size = int((1 - val_split) * len(dataset))
+    val_size = len(dataset) - train_size
+    train_dataset, val_dataset = torch.utils.data.random_split(
+        dataset, [train_size, val_size]
+    )
+
+    # DataLoaders
+    train_loader = DataLoader(
+        train_dataset,
+        batch_size=batch_size,
+        shuffle=True,
+        num_workers=0,
+        pin_memory=True,
+    )
+    val_loader = DataLoader(
+        val_dataset,
+        batch_size=batch_size,
+        shuffle=False,
+        num_workers=0,
+        pin_memory=True,
+    )
+    global model
+    # Model
+    model = BlurRegressionCNN().to(device)
+
+    # Train
+    model = train_model(model, train_loader, val_loader, num_epochs, learning_rate)
+
+    print("Training completed!")
+
+if __name__ == "__main__":
+    main()
diff --git a/webcam.py b/webcam.py
new file mode 100644
index 0000000..70aa9a5
--- /dev/null
+++ b/webcam.py
@@ -0,0 +1,122 @@
+import sys
+
+import numpy as np
+from PySide6.QtWidgets import QApplication, QLabel, QWidget, QVBoxLayout
+from PySide6.QtGui import QImage, QPixmap, QFont
+from PySide6.QtCore import QTimer
+import torch
+import torch.nn as nn
+import cv2
+import torch.amp
+from torchvision import transforms
+from PIL import Image
+transform = transforms.Compose([
+    transforms.Resize((128, 128)),  # 🔥 CRITICAL
+    transforms.ToTensor(),
+    transforms.Normalize(mean=[0.5], std=[0.5])
+])
+
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+sigma_range=(0,
+             30)
+sigma_mean = (sigma_range[1] - sigma_range[0]) / 2
+sigma_std = (sigma_range[1] - sigma_range[0]) / 2
+class BlurRegressionCNN(nn.Module):
+    def __init__(self):
+        super(BlurRegressionCNN, self).__init__()
+        self.features = nn.Sequential(
+            nn.Conv2d(1, 16, kernel_size=3, padding=1),
+            nn.ReLU(),
+            nn.BatchNorm2d(16),
+            nn.MaxPool2d(2),
+            nn.Conv2d(16, 32, kernel_size=3, padding=1),
+            nn.ReLU(),
+            nn.BatchNorm2d(32),
+            nn.MaxPool2d(32),
+            nn.Conv2d(32, 64, kernel_size=3, padding=1),
+            nn.ReLU(),
+            nn.BatchNorm2d(64),
+            nn.MaxPool2d(2),
+        )
+        self.regressor = nn.Sequential(
+            nn.Flatten(),
+            nn.Linear(64, 512),
+            nn.ReLU(),
+            nn.Dropout(0.3),
+            nn.Linear(512, 1),
+        )
+
+    def forward(self, x):
+        x = self.features(x)
+        x = self.regressor(x)
+        return x.squeeze(-1)
+
+
+class WebcamViewer(QWidget):
+    def __init__(self):
+        super().__init__()
+        self.setWindowTitle("Live Webcam Feed with Sharpness Score")
+
+        # UI elements
+        self.image_label = QLabel()
+        self.sharpness_label = QLabel("Sharpness: ")
+        self.sharpness_label.setFont(QFont("Arial", 14))
+
+        layout = QVBoxLayout()
+        layout.addWidget(self.image_label)
+        layout.addWidget(self.sharpness_label)
+        self.setLayout(layout)
+
+        # Video capture
+        self.cap = cv2.VideoCapture(0)
+        self.model = BlurRegressionCNN().to(device)
+        self.model.load_state_dict(torch.load("best_blur_model.pth", map_location=device))
+        self.model.eval()
+
+        # Timer
+        self.timer = QTimer()
+        self.timer.timeout.connect(self.update_frame)
+        self.timer.start(30)
+
+    def preprocess_frame(self, frame):
+        # Convert to grayscale
+        gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
+        # Resize to expected input size (e.g., 240x320)
+        resized = cv2.resize(gray, (320, 240))
+        # Normalize to [0, 1]
+        normalized = resized.astype(np.float32) / 255.0
+        # Add batch and channel dimensions: (1, 1, H, W)
+        tensor = torch.from_numpy(normalized).unsqueeze(0).unsqueeze(0).to(device)
+        return tensor
+
+    def update_frame(self):
+        ret, frame = self.cap.read()
+        if ret:
+            # Convert BGR (OpenCV) to RGB (Qt)
+            rgb_image = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+            h, w, ch = rgb_image.shape
+            bytes_per_line = ch * w
+            qt_image = QImage(rgb_image.data, w, h, bytes_per_line, QImage.Format_RGB888)
+            self.image_label.setPixmap(QPixmap.fromImage(qt_image))
+
+            # Convert to PIL, grayscale
+            img_pil = Image.fromarray(frame).convert("L")
+
+            # Apply transform
+            input_tensor = transform(img_pil).unsqueeze(0).to(device)
+            with torch.no_grad():
+                sharpness_score = self.model(input_tensor).item()
+                denormalized = sharpness_score * sigma_std + sigma_mean
+
+            # Update label
+            self.sharpness_label.setText(f"Sharpness: {denormalized:.2f}")
+
+    def closeEvent(self, event):
+        self.cap.release()
+
+
+if __name__ == "__main__":
+    app = QApplication(sys.argv)
+    viewer = WebcamViewer()
+    viewer.show()
+    sys.exit(app.exec())