Covid19_classification_focal_loss.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "MSCS19008_05_Part_02Final.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "machine_shape": "hm"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "2b325b49a2f0407fa00458683167ff51": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_4bde53f852ae424a803ab83bb0e5f421",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_0e2fa1fd52af4a60ad9ad9bfc218de99",
              "IPY_MODEL_586de0652cc046eab899762fe7b38a26"
            ]
          }
        },
        "4bde53f852ae424a803ab83bb0e5f421": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "0e2fa1fd52af4a60ad9ad9bfc218de99": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "IntProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_cd947ad8db60450580ec5703d5972a92",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "IntProgressModel",
            "bar_style": "success",
            "max": 46827520,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 46827520,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_e86463d34cd24bd7810a4563e7b79c3f"
          }
        },
        "586de0652cc046eab899762fe7b38a26": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_513872c4b6264b38b53a85addf9dac6b",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 44.7M/44.7M [02:38&lt;00:00, 295kB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_3bf0dfa11eac40adba4490fe1b4e62f2"
          }
        },
        "cd947ad8db60450580ec5703d5972a92": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "e86463d34cd24bd7810a4563e7b79c3f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "513872c4b6264b38b53a85addf9dac6b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "3bf0dfa11eac40adba4490fe1b4e62f2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "5Ineye0rjIOb",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#!mv \"./Assignment 5 Dataset\" \"./drive/My Drive/\"\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VpnLApJ-BQE2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "! unzip './drive/My Drive/A_05_Part_02_Dataset.zip'"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3AbPeEZtmpvC",
        "colab_type": "code",
        "outputId": "bbfe71d5-b6ad-4203-d3fb-6a9539b46063",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 71
        }
      },
      "source": [
        "import glob\n",
        "import pandas as pd\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.optim as optim\n",
        "from torch import cuda\n",
        "from torch.optim import lr_scheduler\n",
        "from torch.autograd import Variable\n",
        "import numpy as np\n",
        "from PIL import Image\n",
        "import torchvision\n",
        "import natsort\n",
        "from torch.utils.data import Dataset\n",
        "from torchvision import datasets, models, transforms\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.metrics import multilabel_confusion_matrix, f1_score, precision_score, recall_score\n",
        "import seaborn as sns \n",
        "import time\n",
        "import os\n",
        "import copy"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
            "  import pandas.util.testing as tm\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fGwL0t3qqxRv",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "data_dir = 'A_05_Part_02_Dataset'\n",
        "batch_size = 120"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uaGO-EjkTyvB",
        "colab_type": "code",
        "outputId": "904f1063-e66c-4036-be7c-29d783557c85",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "source": [
        "#Define transforms for the training data and validation data\n",
        "train_transforms = transforms.Compose([transforms.RandomRotation(10),\n",
        "                                       transforms.Resize((224,224)),\n",
        "                                       transforms.RandomHorizontalFlip(),\n",
        "                                       transforms.ToTensor()])\n",
        "\n",
        "valid_transforms = transforms.Compose([transforms.Resize((224,224)),                             \n",
        "                                      transforms.ToTensor()])\n",
        "\n",
        "\n",
        "#pass transform here-in\n",
        "train_data = datasets.ImageFolder(data_dir + '/Train', transform = train_transforms)\n",
        "valid_data = datasets.ImageFolder(data_dir + '/Validation', transform = valid_transforms)\n",
        "\n",
        "#data loaders\n",
        "trainloader = torch.utils.data.DataLoader(train_data, batch_size=batch_size, shuffle=True)\n",
        "validloader = torch.utils.data.DataLoader(valid_data, batch_size=batch_size, shuffle=True)\n",
        "\n",
        "print(\"Classes: \")\n",
        "class_names = train_data.classes\n",
        "print('Train',class_names)\n",
        "\n",
        "valid_class_names = valid_data.classes\n",
        "print('Validation',valid_class_names)"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Classes: \n",
            "Train ['covid-19', 'normal', 'pneumonia']\n",
            "Validation ['covid', 'normal', 'pneumonia']\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "40uomI5Sj6Gu",
        "colab_type": "code",
        "outputId": "5fe2d20f-154a-4bcf-ac33-e0ed2a881aad",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "len(trainloader.dataset), len(validloader.dataset)"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(6200, 628)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "92llWT844ZAX",
        "colab_type": "code",
        "outputId": "6e178e47-582e-40b2-ec03-eaece00bff90",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 152
        }
      },
      "source": [
        "def imshow(inp, title=None):\n",
        "    inp = inp.numpy().transpose((1, 2, 0))\n",
        "    plt.figure(figsize=(15,15))\n",
        "    plt.axis('off')\n",
        "    plt.imshow(inp)\n",
        "    if title is not None:\n",
        "        plt.title(title)\n",
        "    plt.pause(0.001)\n",
        "\n",
        "def show_databatch(inputs, classes):\n",
        "    out = torchvision.utils.make_grid(inputs[:8])\n",
        "    imshow(out, title=[class_names[x] for x in classes[:8]])\n",
        "\n",
        "# Get a batch of training data\n",
        "inputs, classes = next(iter(trainloader))\n",
        "show_databatch(inputs, classes)"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1MAAACHCAYAAADz/wM4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOy9eZBsWX7X9z25VWVm5VLLq6q39vR0t5hW0yOkiaAxEtIgRxgYYwtjGyzMMjIQlsGEsVnFYk8ICUcQdkhhAwMhwBPCZpDwArJsI1lLS2hAA0gejTSa7unp7fXbul7tlZW1Zdb1H5mfU98871a9ektPd4/uLyKjsjJv3nvu7/yW728554Ysy1RQQQUVVFBBBRVUUEEFFVTQg1Hp3R5AQQUVVFBBBRVUUEEFFVTQ+5GKYKqgggoqqKCCCiqooIIKKughqAimCiqooIIKKqigggoqqKCCHoKKYKqgggoqqKCCCiqooIIKKughqAimCiqooIIKKqigggoqqKCCHoKKYKqgggoqqKCCCiqooIIKKughqAimCnrfUgghCyHshhC+990ey4PSeOxPv9vjeK9TCOFvhRD+8rs9jneLQggvhhD+yLs9jneKQggfDyH83Ls9jvcDhRB6IYQPvtvjeDcohPCBsc2svNtjeafo17KuhxA+FULYCyHc+EqPq6CCHgcVwVRB73f6uizL/qIUHe4b7/J4HpjGjuTj5zz2xRDCR9/ZEb13KMuy78yy7K+c59gH4U0I4Y0QwgceYWhfcRqDkU+d89hPhBA+8c6O6PFTCOFcDz58v+r6o1CWZTNZlr12v+MehDchhI+GEF581LF9panQ9Ylj3/e6nmXZxyX9jndvNAUV9GhUBFMFFXQf+mrOhhb0+CiM6Kvapha6UNB56KtdTgpdL6iggpy+qo1BQb+2aZyR/NMhhM+HELZCCD8UQpi27/9oCOHLIYT1EMKPhBAu2XdZCOGPhxBekfTKOIN7I4TwZ0MIKyGE2yGE3xVC+FgI4Uvjc/wF+/1vDCH8ixDC5vjYvx5CqD3m+/tECOF/Hd/XTgjhF0MIX/cA9/87QwifG4/xn4cQPpzc/9P2/6dCCN8zfv+gvJgKIXx/COHW+PX9IYSp5Fx/ys71HadcdzaE8KMhhLshhI3x+yuPgY+fCiH8jRDC/zXm42dDCE/Z9785hPCvxjz8VyGE32zfvRhC+N4Qwmck9SV9cMy7PxZCeGV8vr8SQnhqzOPtEMIPIwvv1D0l93c/HndCCD84HsObIYS/BFAMowz5Z0II3xdCWJP0iTG//mYI4f8Jo9azz4QQlsfzuhFCeCmE8PV2/j8fQnh1zItfDSH8e4/z/sbXeCOE8F3j82+EEP4nZP0c9z8VQvjvQgjXQwhvh1Frad3u/+eSa0XdeAhePDuWmc0QwhdCCP+ufXc/OfTr/tshhP9vLE9vhcdUmRhf4zvHsrs5Hk8Yf1cay8abYz7+YAihM/6ONrw/HEK4LumnEtnZDCG8Fka69PHxmFdCCH/Irv2O3FNyf4Wuv891vaCC3pOUZVnxKl7vy5ekTNLTZ3z/hqR/KemSpDlJX5T0nePvvlXSqqRvkDQl6X+U9LPJuf/f8e/qkj4qaSDpv5ZUlfRHJd2V9A8ktSQ9J2lP0pPj339E0m+SVJH0gfG1/+R5x37O+/+EpCNJ/8F4TH9a0uuSque4/6+XtCLpBUllSX9ofPxU3vgkfUrS94zfPygvvlvSz0talHRB0j+X9FeSc333+Fwf0wiozOZcd17Svy+pMb7OP5L0jx+DHH1K0pqk3zier/9F0j8cfzcnaUPSHxh/9+3j/+fH378o6fr4nivje8gk/RNJ7fHnB5J+UtIHJXUk/aqkP3Seexqf/4884v3dj8c/OB5vSyNZ/ZKkPzz+7uPj3/6J8f3Vx/xa1UjGpyX9lEZy9wfHsvQ9kn7arv8faiSDJUm/V9KupIt2/p97DHP4hqRfkXR1PGef0b3yetr9f5+kHxn/riXp/5T03542PpluPAgvxtf+sqS/IKmmkQ3akfTr7ieHOdf9qKTnxzz9sKS3Jf2ux2RTf1RSV9I1jfT6t4+/+0/G4/+gpBlJ/7ukvz/+7gPj3/6gpOZYTpCd7zBeXJf0NzSyuf/W+P5n7ndPdv5Koetfnbo+HvuNR5Xh4lW83o3Xuz6A4lW8Hval8wVTv9/+/2uS/tb4/d+V9NfsuxmNApMP2Lm/1b7/qEYBQnn8f2t8zAt2zC/oFEAj6U9K+j/OO/Zz3v8nJP28/V+SdFvSbznH/X9S44DGvn9Z0rfkjU/3BlPn5oWkVyV9zL77bZLeSM5Vse9XJP2m9Lo59/8bJG08Bjn6lKS/Y/9/TNJL4/d/QNK/TI7/F5I+Pn7/oqTvzpHLb0x48efs//9e0vef5570+ABWLo81AkSHkr7WvvtPJb04fv9xSddz+PUD9v+fkPRF+/95SZtnjOdzkr7Nzv+4gqnvTObw1XPcf9AI8D1l3/0bkl4/bXy6N5g6Fy8k/RZJdySV7PtPS/rE/eQwvW7O/X+/pO97DHzMJH2T/f/Dkv78+P1PSvpj9t2v08hmkjDKJH3Qvv+4pFcSXmSSluyzNUm/4X73pMcbTBW6fvLde0bXVQRTxet9/Cra/Ar6aqc79r6vUdAkjbJnb/JFlmU9jRz7ZTv+reRca1mWDcfv98Z/37bv9zh/COFrxm0cd0II25L+qqSFR7mRUyiOMcuyY0k3NLo36LT7f0LSnxq332yGEDY1yur7b8+ic/NCCa/H7/06a1mWDU4ZZ6QQQiOE8LfH7Snbkn5WUjeEUD7nmM+ic8nJmN7U2XIi3cuL0+Tknbwnp9N4vKBRBjudn8dyf5IUQviD4aSddFPSr9c7rAs6v4xd0KhS8As2vn86/vy8dF5eXJL01lhPfZzO69PkcIJCCC+EEH563K61Jek79fh4el5deFOjQGrJPktlJeWFsiw7TRfeyXtyKnT9hN6vul5QQe8pKoKpgn6t0i2NAgpJUgihqVEbxk07JnuE839S0kuSnsmyrK1Ra094hPOdRld5M+59v6LRvd2P3pL0vVmWde3VyLLs0+Pv+xqBTGj5EcY4wWuN2ofOM8aU/pRG2fAXxjz95vHn7wRfoXTs0mj8j0tO3o17clrVqLqQzs9jub8QwhOSfkDSf65Ru1RXo3a8d1QXdH4ZW9UIED5netDJsgyAuCvTgxDCo+rB1TC5cUHK6/PSP9CoNfFqlmUdSX9L77zM5OnxQJPg+lF04d24J6dC198/ul5QQe8pKoKpgn6t0qclfUcI4TeE0WYIf1XSZ7Mse+Mxnb8laVtSL4TwIUn/2Xl+FE4Wcn/gnNf5SAjhd4fRzkt/UqOe/Z8/x+9+QNJ3jrPBIYTQDKMF4K3x95+T9PtCCOUQwm+X9C3nHE8efVrSXwohXAghLGi01up/fojztDQCvpshhDlJ/81pB44XYj8K8IH+b0lfE0L4fSGESgjh90r6Wo3WlTwOOvc9pRRGmy58/FEuPq4u/rCk7w0htMaA6L/Sw81PHjU1Amh3JWm8GP7Xn+eHYbTByosPcK0/HkK4MubjX5T0Q/f7wbhK9AOSvi+EsDi+7uUQwm8bH/JLkp4b24lpjVprH5Y+q1GS4s+GEKphtLX3vyPpHz7EuVqS1rMs2w8h/EZJv++0A8ebCHzqIa6R0qcl/ZchhCdDCDMa2cwfSqogj0LnvienQtfPR+9lXS+ooPc7FcFUQb8mKcuyn5D0lyX9bxqtM3pK0n/0GC/xpzUCAzsagbX7ArsxXdWo9eK82ep/otFCXxZO/+4sy47u96Msy/61RhtH/PXxb7+sUU879F9oBPQ2Jf3Hkv7xOceTR98j6V9L+rykX5b0i+PPHpS+X6NF0asaBYz/9Ixjr2q00cUjUZZla5J+p0ZZ5TVJf1bS78yybPVRzz2mB7mnSGG0Q9i8zhc434/+hEYVmNck/ZxGFYK/9xjOqyzLflWjdSP/QqMKxvMabQ5xHrr6AMdKo3H/uEb38arOL2N/TiP5//lx+9VPaFRBUJZlX9JoMf9PSHpFI/48FGVZdqiRTv0Ojeb7b0r6g1mWvfQQp/tjkr47hLCjUXLih8849kH5eBr9PUl/X6P2tNcl7WskO4+LHuSenApdPz+9V3W9oILe1xSy7HEkdAoq6CtPIYR9jSox/0OWZX/53R7P46AQwl+SdDfLsr99jmM/odGC9N//jg/sfUYhhL8j6R9lWfZj7/ZY3gkKIXyTpD+eZdm3v9tjeacohPA5Sf/mGOTe79g3NFq8/xPv+MDeRzQG4r8k6cPnSbK8H6nQ9fc/hRD+rkY7Aa5kWfb0/Y4vqKD3GhXBVEEFvU+pCKYKKmhERTBVUEEFFVTQu0VFm19BBRVUUEEFFVRQQQUVVNBDUFGZKqigggoqqKCCCiqooIIKeggqKlMFFVRQQQUVVFBBBRVUUEEPQZWzvnxM240+EoUQ4ov/S6XSPd/5MaVSKb6yLNPx8bGyLIvH8BmfU51Ln2jMMQU9XmIO4fuD8pg5TN/nfZdex+WoXC4rhKDhcKjhcKh3g0qlksrlchwb4zuLHtcxeb9x/sBLP1f6PoSg4+NjHR0daTAYvGt8fC/TWfxj/vOOOe2vH3faPOfZMbeF6Fxh3/IptSvpd/zNmyP8U6lUmvBJrlN+rjw9S31Rno9iLjn30dHRxDHvJ8qT7dO+Sz9ze58ew3z4vPixeTqEb0AvsyzT4eGhBoNBtHOF3hT01Up5Ns1fbstOs39u+1J/A0bIsmziPcek9u69RFmWnQqszgymJKlcPt/DuR8UvDmTnHFMEMasVqup0Wio0WhoenpatVpN09PTqlarqlarmpqa0vT0tFqtlqamplQul9VsNtVqtdRqtbS3t6f9/X1tb29renpapVJJ/X5f29vb2tra0v7+vobDoQaDgQ4ODnR4eKiDgwPt7u5qc3NTOzs7GgwGEwb7NHDzlSac53uVSqWSKpWKpqam1Gg01G631W63NTMzo4ODA62trWkwGGhjYyPOgyuUdAJq4HW1WlWpVNLBwYEkaWpqKs5fCEH1el37+/s6Pj5Wp9NRpVLR/v6+6vW6Njc3VSqVVK1WNTs7q1arpe3tbUnS6uqqDg8Pcx3rWaDqUR3r8vKyLly4oN3dXVWr1Sj3brAg1w/467z2Y1ODdxpQ88REuVzW1NRU1K1qtarj42NVKpU4JjeUzO1gMNDW1pY2Nze1ubmptbU1bW1taXd3V/v7+zo6OvqqBB9nBT2pHQOYVSqVyL9yuaxut6uFhQW12+2JzyuVyoQ8+BzgsCqVimq1WrTRfE7whF4MBoMIBvf29jQcDrW3t6ejoyPt7u5OfI4eokOHh4cx2cD5PAn1XnZ8Z5E7+vTzCxcuqFKpaGVlJX6GL8L+4HumpqZUq9VUq9VUqVRUqVRUr9dVrVZVr9fVaDRUr9dVr9fjeSTp8PBQWZbF446OjqJuHR8fa39/X4eHh9rb29Pe3l6cx6Ojo+ijtre3tb+/r6WlJW1tben27dtxrlNbms7Zafw4z+f+2f0SjmmyIAVdDtBS2faE12n2zO/Fr1Uul1WtVtVoNDQzM6NWq6Vms6lqtRr1UFLkP7awXC5renpa3W5Xc3NzqtfrOjg40J07d9Tv93V4eKi3335bt2/f1srKijY2NrS9va1+v6+Dg4MJ/SiooEehNBn8uClN6qCD6A72rFqtanp6Ws1mU81mM2Lx1PaVy+Vo8/BpWZbFJCt4aX9/XyEEHRwcqN/vx0TQ7u5uPL7f7yvLsvi5JB0cHGhnZ0fHx8fq9/u6efOm+v3+e0bf7htMOWA7i87KkJ5FDpSZSCYVsOZOycEewVWz2YxAfWpqKgLgfr8vSXHiQwgTzK/VapKkwWAwAUikUaCCMEk614R9JYOr94LwQKVSKQa97XZbs7OzWlpa0uzsrNrtdnRi0oiPBwcH2tjYULPZ1NHRkVqtljY3N2Mwe3R0NFEtct6Xy+UYSEmTGQ1pBFKq1aoODw81NTWlnZ0d7e/vq1araX5+Xru7uzHwfv3113V4eBhliKA5L3t8Gj3qnHc6HU1PT0vSPcGUZ7jPojRYyssgSZoAVH6cv6rV6gQwlBQDgBTQ8P/U1JQuXryoS5cuKYRRpQ/w3uv1YoC1vr6ura0t9Xq9GGS9l+T4fpTet3SvQ/Jj0iCIzxzwVSoVHRwcKMuyCQCZ99vUPvp7B6n8lv+lSR1CBgiOAN5etfJADBDf6/W0sbERwTr6gq6it+lfAGaaecwLwu5XDXocNDU1peXlZYUQdOfOHe3v70sa6dnFixc1Pz+vV199Nco94IIkHnoyPT0dfQs+CfDB3EmKQS/XACCFEDQzM6Nqtar9/f3IpzQBUalUIp+RLa6Pz9vc3IxjkxTn9H6VyUfl8/2ObzabqtVqubaD3+d9lifX6fHISqoPzBe4gOBoZmZmIgHm9o7/CZqZh1KppM3NTQ2HQ9VqNS0tLemZZ57R4eFhnKft7W3t7Oxoa2srJpK2t7e1sbGh3d1d7e3tRd/2XgF+Bb336XEl7c9TYfekHzYMO0bBAgyO3SOQIqFXqVSijRwOh2o0GtFuYbPcDuTpQmqX/P+9vb2IK65fv64f/dEf1fXr1+9J8D0Izx5EF+937H2DKen+A3uQtoI8B+rgwwFHmp3174mKmXAqVjgKKhXHx8exYgWIceGpVqvxN35dj8odYKevNON/Hn69Xwm+N5tNdbtdLS4uanFxUfPz85qdndXs7KwajYbK5bKGw6HK5fIEGDs4OIggYWZmRhcuXNDh4WFU0sPDw5h1dcAGWKPytL6+ruFwGLOJLlOuuGR3syzTwcGBnnrqKd24cUP1el1vvfWWjo6OFEKIAVW1Wj212pfONfQoc10qldRutyfkmUCGZAIAA0CUVu7SceRVojA0njBIM8QOTNJz5AVdzheOYb7hIRnhxcVFSYpBVr/fV6/X087OjjY3N2N2F+DhleCvNHH/efz1zDkAEflGPtx+YUvSjLlXqaj+7e3taWpqauJ6Tsx7+rkDUP5665eUH7x40HRa1SKVQUlRRpHFVM48CDs6Oppoi3LbnCfLPi4/fm9vL/fe8+g8Ca8PfehDev755zUzM6Of+qmf0quvvqpSqaSrV6+q1Wrptddei8k2giLARVqdzQP+fi0SCthDdB3epe1jBKHOo3RuuG6r1VK329UXv/jFOFeedHJe5r2woSn/HsSmncXvcrmsCxcuRFCVjv802fVW4RRwuQ66nkmKVUEAX6PRiImyubk5VSqVeO/wt1QqaXp6OvIeX1GpVDQcDrW1taV+v6/BYCBJ2t7ejnPGeJvNpur1uubm5rS8vKy9vT1lWRaDKDpjbt++rbW1NfV6vVjFwrelYLCggh6FziNHaRDl2NcrU9h8TxLlddDwf61Wi4ULihiMKW9cjqVPGzdJdpLxt27d0urq6Fna+NP7VcnvZ9f4nu6APNt4Fp0rmDqLzpNVPCsyTjO2HjClk+rRb71eV61Wi1lBgNrU1FQ0dvV6fQIwSLoH3OCEPJhKBQsDn4KQ0yLh93NQ5WBxZmZG3W5X8/PzWlxc1PLysubn52PLBJk+Mm4Evt4uBGj0ci/nb7Va0XHRuoejQ3kAZsPhMFYop6en43y2Wi3dunUrBmo4p06no52dnQgsW62WdnZ2NDc3py9/+cv3BE0HBweampqKn+fNXd68PsocVyoVzc7OTpTIqVL1er3c9hjAKoYlBc5+LPKZZnjTipUbsfS49JVWZDzLjaGjjB9CuCfrS+tNrVZTp9PRlStX4pwdHBxob29P29vbsWWw1+tFUJIH/h4nuazyvxtYjCz3Cv+np6cnwOtpVam0MgU/q9Wqer1eTFT4PDhxfk/swBM/nwc6Pl/+nb/crjlw53/aAA8PDyMvUnmDN9z3YDCIgSLA1YFjen14miarqBCcNe8PIhMhBL366qu6c+eOyuWy9vb21Gq19MEPflCDwUBvvPGGQhhVW93npBVjDwTSxGBetpVginvGZkonrcpZNqpOum7nAYQsy1Sv19XtdvXGG2/E3zkvfY48ePKXt4T6NdL7OY3f97N9lUpF8/PzsfpOAsLnHfvufppAA/vvNuwsTNBoNDQ1NRUTR2TVSawxp/yO42jto/WVezs+Ptbu7m4EhbSnl0qliC3wO5444PeOT/CTBG+eLDw+Pp74fZqAKKig+9GDyEke5vYCgtu7hYUFLS4uqtVq6fbt2xMBlF879U2082VZFuXfE0ruS9PChCfp8HFUgYfDoba3t2MyHH/tNu60hHMe+X3Mzs7qmWeeUQhBu7u7eumll6I+n+dcjxRMpRnF9PP0/zSTlwYveX3TPrHeVpFGojgosmD0mkuKRvbo6ChmjRzokXH0CfGInLH7tRCOdOJSIXuvB1TwdWpqSrOzs7p8+bIuXbqkbrerTqejmZmZiSqggwmqdmkw6qAIvjkvUN56vR7bkVA6B8ye5c6yTO12O65VOD4+jgFZs9nU/v6+9vf3tbe3pxCCGo1GBH/NZlOlUkk3b95Uq9WaqCRIJxlkMip5inPWPJ6VUTmLpqam1O12JwJMeF2pVLS3txflC70gs+38QQ5T8Az5e1rJ+Nz54EAM/fNrpwGY6yuBM0GPgwfnUcp3DwQAO1NTU+p0OlpaWoqBFEEaer2/vx+BzOPK6J4WaHog5Rl1twcEww5qT2vJc2fEd7TRsQ4nb2zMUepsCOA8sOW7vGAKB5EGUqlzkxSDXHSJKgBglCoCdgSZocLogBGg6Imo05yfJw/W19cfOFHn5yiVSpqZmdHy8rLm5uY0OzurUqmkL33pSwoh6Mknn9TOzo7eeuutiUCYVzqHnNP13gNASdFvpQE5/KtWq1F+XQ8IGOCX6x3XZD3wW2+9pb29vWiX0Tv3VWll29e+ue1I5yKPv6kNvJ/OtVotPfXUU/Hc2HxsnAeXWZZF/0zl5ujoKLY/evbb52ZqakqtViu2kXv1m/WyOzs7MUnFulja0qmekxRBz+v1ugaDgXZ2diJo6/V6UfdoZ9rd3Y1JQHjHHKNX2HO6Z5hzeDIcDqPv82SZJwiLoKqg0+i8QUOKux1bp1ibxOcLL7ygJ598UoPBQJubmxN2Ki+owsYgz/1+P2JHabSkxosUbj+xVWllGl/P+VZWVrSysjLhi7k2yXy3az6+03BcpVLRk08+qZdeeknSqKvm2rVrev3118+N4x86mEoz06nRTY1yagjdaZ0GMnzx28zMTAQrabbeQRrX7vf7secZoJNmu/x3ZLsw3D5WB0385Vp5bTJ5GX8f77tJXjFoNBpaXl7WxYsXNRgMtLS0pOXlZS0vL8cgybN/DrK9ZSwFoAArz7BynDsazgcIqNVq8TcOVjjHzMxMlAnpJEhi3mgT7Ha7unv3rq5cuaJms6nPf/7zWl1djRslAGQg5ijNQrzT89VsNmNWGp6yOQAOG+MEXwFoaQDB/efpojQJ/pgrB8Pcd1pV8QACvUAPXM7RuzQR4sYu1VMfW3p9Hz/nJLPNphcp2OJ+vGXqQQItEgQ+Ps+iEfSWSqUJgJfKMYDRAbFXMwi0XL4ITgCBvs7EeePZOwIo/+y0RADH5TmG0+YGYEq23nUX++Cta9wj3/m4GS9rVPzaZ4FFDxzTMefdp88Jn01NTenbvu3b9PTTT0c+Hx0dqV6va3V1Vaurq7pz5446nU5sC8bOII8AArdJrlNuM/KSgn4cwNsBM1V9D4K4NvNSLo82LBkMBrpx40ZcvE0Ahr65/3UflCYLvcLtWeW0Yug8fxCbSPubt7VKikFTrVaLutJoNCLAAjyxQJ1NN5gbT67OzMzE9mLahNk8BR80PT0dN5TyAL9arWppaUlXr16Nm4CABXwdIC+fY5K3+CyCHsbmcg9PaTvnOj4WACjzw/hoe3dfWlBBDxpAnRZE5f3PsVNTU7px44bW19e1vb0dCw9pe59fw5Nn2BM6VdAL37Qrxch+b1mWRTswHA6jDt+4cSN27nAOSRN2zH22JxnTa/HbVqsVN0S7cuWKXn31VX3t137tA83JQwVTaSDlgVO73dYzzzyj1dVVXb9+PRo/nKgvVktBmjSZBWcRaavVipkgn3jO5yX+vb097e7uxonFeHnGEGPrzJdG6wGYbLJeGHA3jAAAj4DPAm8OiPz/d4rSYDPveqVSSQsLC3r66afV6XS0trYW2yfZGELSRDbWgyK/H69GQcyTr13DwaFgXKNer8f58oXtjJPr0SrC/bTb7YnMLpmJTqejubk5zczMaGdnR8vLy7HNr9/vx9bANPjw1pcH5ffDOLmZmZmJ/nsHx8h3XkbZgR7fsysVu1J6uwm6h8N3efD79YAXnue1N8Gz1FB7xc+BJ7rNtT2DLuUbOM+OcS1fi+PzRfVE0kQbISAO4EOFxXXVCZ7DF+wUNqFUOllf4WN14AxIKpVKcf0Gc+cgPF0PR+bcAzCcj8uYA1oH5x4seVDkx3mw4TY877wOJF0mHeynNpzzON/9WB+nV208IZUG6Km8pvKSBlBp4IpMtFqtieTYcDjU3NychsOhpqamtLS0pMFgoEajoZWVlVhRAOjX6/V7dOY0O++AxH0kMsDaTFq9fC6QZ+cHfq5UKsXNDby901tT4TXgO7WN6V8HOGmliDH7vT5IQDU9PZ1bPScZxqYN2K56vR43B0If0Al2NkQ3wANZlml7e1urq6vq9/tRXrxrgiQF3QsenLA72OXLl6O8TU9PRxkFQ9y9e1eVSiViEbeR3pnhvg77zU6NXp0NIcSkD7bN7xebziu1XQUVlEepT06To2nAlLYwI/dgjNXV1dyNdfz8/sJ3oIPHxyet/2nXhQdVeUlX/A+JiN3d3bi+OvUhqV1iLOiw2zW32243QwjqdDpqNBqn2viz6IGDqbMCKWm0Oxm7ueGUfOcPIlvPSjMx7hBDCLEEj7H1nZQAWgAgBIJsn7eWANzL5bIajcZEltQnjcwZToTME86G+0773/nfF5TCn3SiHcg8jqDKwcT9zpmCkLm5OS0tLUWHnWVZ3EDCnb//1jP0kiZAswea7mBcweAZveS0XTiIgIdcz0E+7ZoAW87Ffc/Pz0+A+1qtphdeeAGiLn8AACAASURBVEErKyt67bXX1Ov1tLy8rKmpKa2trUUn69mUh5mDh6HZ2dl4j8gfwSb3UK1WI9CBvwRMnqRA3nC+Dl5oJXGgmwLnFITmVadcxk7LuCMngARvC0sDKQxennHzQErSBD+wJRhn1juwvqHX6038hirs7Oxs3IxGGrX8rK6uamtra+Ka2BgCVgIKdhZl3UQ696ktI7jkmlQPuf+0SouzorLuwDMNdtLAyfno8gPPPVFwVvDvVRMPjn2s6fjTIM0rCOi82wd3vp5V9ASAg9m8oPu0AMr9gVcQSZwAyNnienNzUxcvXtSFCxfUbDZ1+/btOB4qCXfv3o02B51gYwOAt+9whX9JdS29Z3jE3LuOuy0l8GL7ba6fru/x9+6fvK0zBSEuq96x4b4t3SjhvNRqteLaIhI5jLFer6vdbscgaXd3N+4W2Ww21el01G63JzYB8WpnlmXa3NzU+vq67ty5M1F1oioMf1mLyf8k5Ei03r17VyEELSwsqFQqaWdnRyGMWsVDONnCmet6lwzJLXjnvGXs3tpEcMdjOhyQMmbuk+MdzFKhe7/thFrQo9FZc+3Jx8XFRb3wwgsTm2uRTKAC3uv11Gw2tb6+fo8txuekXUl5bYBpIMVY0uRaWrXywM0xQ7qLM3bo4OBAvV4v7prpyRKu6djN/Ru22IsnebHL1taWrl27puvXr2ttbU3Xrl3L7V46i84VTOWBnXQwIQQtLS3pqaee0sbGhiqViprNZmxZcaPA7/OiYhwLfeH8xoMSBxju4HziAJ0pSEzbyBBEIma21vYyoQMjHJw7Fgcf3ouergfIi6Cdfw9D3F96jnTy/X+2CadKB3BkjYs743TsCGKajeDatCd41p05ZYzMN7wlk5oGU2mGFrAL4PX2kXK5HINvHPLu7q7K5dGOUp1OR1NTU+r1eqpUKmo0GpJG65bIWJIpTnmWzo1/7mDpQahUKsVnC3nVFaDhhqhWq0VDJCkGRwT8OG6qKgRV8IdMLOd34CudbMnOuNIAwe/VQVpekCUpjo3qrgdNeedOAZoHaz73Lh/YE9+edWpqSv1+Xzs7OxE8o5s8q8JtDIkVCFvliRrGR/BPxYl78jG7U/MMHvzArnmg6zwl4ME5MLeMA9lLddF5yTGezOH7FKingaxXPz2RwrkBf3lzD5/4ndtk9xnOq/TlNjYN1pxSAMC4sOOsgXEA8Pbbb0d7NxyOtu195pln1Gg0YuD/hS98Qa+99pqGw9FmN71eLzp/t1/QYDDQ6upq1D82KXDn7u2BkuJxbvscGGTZZGuLy7EDAxIo3qbnwQo8hEfefuNBlWeZvRLOeNwW+5q39LiUQghxd1fmlsqK6xqb7vAIC8DTxsaGDg4O1Gq11Ol0YsaYpMTa2ppWV1d169ate3bcQ48IPNLd97LsZKdG9JPnDc7NzalUKml/fz8+U0dSfHRGu92O1TNkwPEEwI8AiiAbOz47OzsBaumiwb6hP1454FWr1WKFnQpbajvZLOqll14qgq2vMOGz6JaiK+HWrVsPfc6z9AsdxsZVq1UtLy/rhRde0NbWlq5cuaLt7W3dunVL165dkzTqhvn85z+vhYUF/diP/Vg8VypzZ1Ww0mAqTRi5Xkkn67SHw2HEm+ACt1WpTWHjIwIpkqTEAH6s2zr/PA2e8v4HI9DaR0vhl770pQeaq3NXptzYp0aU8vdHPvIR/Z7f83v06quv6md/9md15cqVCUCQMh8j4YJBuR/DWalU4i47ntkpl8vRSLFrSLPZjE7h4OAgAu+ZmZn4mQOFNEIGZHn7IIY+3crUwS2Om8/yXg5OUhDiwPZBgHkKYtO5yvu/0Wjo8uXLWlxcjAGHt84RjftaMc+k+jV9lxZpcp0bGcIQwkTLIDwHLALyaAlJ+estf9LJQxbhITKF0z08PNTs7Gx0nlx3YWFBWZap1+vFuW82mzG7uba2FjOY6+vr556Dh6VKpRIfKozcwcujo6OJZzR45cKzq8wXIAQn7ZU6glsH7HkBEPPnAUQa1DhAd0PI/XAtb//yoCPNjPO5Vy8htwn8BbgAPLwdwbNplcposTjZZCdPhvB8IYiNL5rNpkI4yVxLiu19BL+szUjlMC9j5zwmqIKnDoDdIcE3kjRu+1J++vx5QMNY0kA2bXngN1To4ZMHccyX65Wfj/890eXEsW7vaXeDr6whxJ4SMPzMz/xMrh9h7j3YAcCmgeGXv/xl3b17V88//7yeeOKJqEs8oH1lZUW7u7u6cOFCHN/CwoI2Nzejf3GZpPrhMugJPeYGfnqVlyrT4eHhxCY86DTzx8sBO9djPvA7bBDiwTW/966PNAnqlRp+nwZNDj4IinwXu7xqValUihUcgBU2YjAYxPVO3Ddyhd/wqg5rI7Fx8I/Aud/vx3WmHjChXwSnzIW3AqPP2JfNzU3Nzs7G4ImHWvPoDrc1bI4Df1l7RxsmclmpVDQzM6NSqRSfxejJpu3t7bjZkAdUDnA9sPIxkBz4xm/8Rh0fH+tjH/uY+v2+fuRHfkS3b9+O20cX9PgIG8B6fn94t/s2qtvo3YNSXuLd7V4qG9VqVdvb2/r0pz8dkwLYnJs3b6rb7cbOl7t378akbXruPB+WfuZJML6TTnzL4eFhxDbYnjRJzmdpos91vt/va3d3d2L/AxITHOt8gtepn0xtqts0xrSzs6PPfvaz9wRk58Xk565MnZbNpLR4+fJlPfnkk2o2m9rb29Pi4qLq9bo2NjYmHJFPCJmpvb29WMny/nAWlG5sbExkegCMc3Nz6na7scfRy+2+gJvsIO1+OFLvX2Zi/bdMHmMk440wcC1vw/LWCF+8SnbMKy1MPvSwQZXP02n/l8tltVotXb16VZcuXYprzLzVsVKpRHDearVy23jyqhueuXbyTLtnZ1EeMqEOKD349PeemWX9wvHxcWwXabVacc7SBZDHx8cxaBwOh+r1elEZyaDW6/X4rCNa/7wK4HPic5Nn8M5D7EQIPz3QANiyOxr8gQ+e2cGYu1y6g/fAB7l2g+Zz6ffhAY6Dw7yst7eAMa8O9njv85iW5T1Y9PP798y5A0tAS1oNwtmRJT4PtdttXbt2TbOzszHLTDtNtVqNARTB+/b29kRwkVZLUp56YMU84tC82pv+zpMxvkGGJ7gcPBKc8B3kx6RZunQRfarLjBn58QSLy5q/HPxja9xOlkqlmKSanZ3V/Pz8PffrbcCMq1IZ7QoHQHUZJPNJtRu+l8tlLS8vx0CLXaHu3r0bd3115885L168GAM+/AegFh31jDR8QW8I8vjfAxrmAv2HV/Dek1TwjHFyXtcHb293IOXzwBwwZ5yfa/Ed53abQpsh77GXefpFspOgiYDPAxpaFT3o8I6IEELcxKLf78eHrLNpRaVSiTu8djodbW1txTZOl1lkCt4QmNH2C6bwBIbPOdiBYM+TFVTD79y5ozt37mhjYyMGOMw3tmV6enpiDQqBNA+Xp7rmiQDkAhnzDh/eDwYDPf300/qO7/gO3blzRy+++KIWFhZUq9WKYOoRyJOJXm3ydT9p0J5iMK8mP8z1+eu66zLgz1D1RDXdYXx+eHio9fX1CVuNDZEmK0qpb0/tuRchHBOmPEEP8ZGOx5xv3q2GzlGVovqKTeL+83bZPi1hxO/5LC+g8qDKMU8a75xF9w2m/IR+0hBGPcXdbldPP/20vuVbvkXD4VD/7J/9Mw0GA126dCluauBZLm6mXC5rfn5eOzs7mpmZiTu1kS3c3NyM25JSgq/VahMgl2fQXL58OWb56Q31ipO32QCK+v1+dAgEWAhEvV5Xp9OJ2TXPQmxvb8cWDI/CERgHKBhw37UpXUSKM3XgCI8eBKSnoAki+Gi321paWtK1a9fUbrejM8KBHB4eTrSFEaByDh8f9wmf8zLwvh6KLK4DaMaIgnK/vt4HWcCpOKBEWS9cuBAXK6OYgBwHljjITqcTr0HLFQrb7XaVZaMKJ87/vPPg4PY8BEjgnjzR4FUYf6AkPGadBi1ADp49s+wZ/DTASVsFSQygo/z1Bwn7fXKst8RKJ0EY79Pf8H/aAsbnGGbPOCFvHlimoD0vi4bhZ/vi+81Pu93WhQsXVC6Xtb+/PxGgYe98bSVZ8TR7n8qFg9Y04EgrDam8uROCJ97C6ckNdwySYpInLwmA/jqo92t70Ozj4nxeAeE37uR9rUueo2WuAAMkSFzfkQNkCV7V6/WJ57PxTLxGo6HV1dXIEyrfZIx5QHQatJNkSMcKoafYBebCg3YPbrxyzD17cAu/sY++i5zrBbbZ5wI7CJDHx6AH+BOvfvo4XK5cH9N5dzCf2ieCF5I+BKMpzy5evBjnWDp5nhY2ynk7MzMT5RXZnZmZiTxfXV2ND8yl8uPrGknEHh0daWtrKwY/yCVzg/0YDAYTjyJAZpgD+Oj+2WUSnuzu7k7seAYmkRTnoFQqTVTOPCDjuY2NRkNbW1sR82DvHStwPvwc+nZ0dKQf//EfV7/f1/Xr13X16lXNzc3pV37lV/SRj3xEL730UmxzLuhe8kDB28axYak9TneE5vP0nJIm/PmDEjYPOadFGFs5OzsbE67YMaqV6QZiaWUpz+/mJc9T2+CBW7oEh/vFfjhhMx3r8hmbq6R+1NeEuv5514J0kmTmftOALf0fYvyndZPldZWdRfcNplKmSIpl6/n5ec3OzmpxcVELCwuqVCp6+eWXozGq1WpaXFyMjPDnlXS73ZhNT3uhydI4k9KMnyT1+32tra1Fg+0L8T1K99aKnZ2d2KvM/15qBPwfHR1pZmYmZtcwtNVqVVtbW/cAWe9HdwHiRe+2PzMHEO8VABfg8wL5vInmPjqdjlqtlmZnZ3X16lV1u92JB1KmVQeEcTAYRH4CyMmy+LFppsIDIebUwQHkLUTcA7/za5XL5Ri88qLS6M8q4X7IXEonW50zd2QMsyyLWW0WPx8eHsaqKg7cnX7KW8acgs7zUrfbnaiswBMH14wVECJNlqy9AupjcrlxYwKPCCAxuvCa42nlIYvLYnuAS5plSgM5z25xDPLigIbfSCdtgsgJc+lVkFRe0mAqDbAAMuVyObYCcRx2CCqVSup0OhHcMR7pJOHB/HgbJUE753Awmma7fE7csTkfvPLEsZ6E4Fjnuycj0CFfI+LO0LNw3qbM+M9KnvHXgyoAXV7w5E41dcql0kl7FdfhPtOgmOtRRZibm4st3R5QVSqjh8Ry36yFJAFHtcSrQFyPe3FQ4QkgQCzj8GASuSZg9/lymeA+ee+6iw9MwZnbZJ9zvmMHT+TL7VAekEjnFD3xgMkDNmyBB9sEit5SnNq+ZrMZn+eVZaONjdi4xdsaK5VKTLqmCQYC7bW1tbj4HLvkCU2v8pdKpfjsQY7jnkiSYGeGw9HaiBCC5ufn4wYi/PVt1vGn2ADOvbW1FZ+Ltbe3F21HCjZdDtC7/f19bWxs6MKFC3EL+XK5HAMqeMUc4BuRP+98GQwG+uxnP6uDgwNtb2/riSee0Dd90zfp27/92/Vd3/VdevXVV1XQ5C6tBNJgR7fdXpF13ZDyH0FxGlYDe/hmBuehEE6qsp1OR91uN2Ltbrer27dvR1zpCdLp6em4pIGCAbYvrUh70jMNGhmD35f7LK7n58HWUG3Gz5G8chzgyfk00PHfgaulST+MDSJ5ni4HcZvrRRz8Dtj+8PAw2gRvWwYTUk0/rZ3Z6YF280MA2+22Zmdn9Q3f8A26evWq+v2+JGl1dVUHBwcRCHpkXy6X1W63YztfqVTS7du3o9EbDAZaX1+PbTk4MAQByrIsLlrl5mg7YK0DYAnnc3R0FDNGtA6yvqPZbMZsKIrEzj9MMr3fZOLL5XLcASzLsokHVLpT82wnBpG/7F7U6/Um+t1dadPs4XkIZzczM6NutxurfouLizGT4btdeQAJz8lwesXBQS7X4VgAoAdljMOdnjS5BWYafLny4nxwYO7Qae9AAZAzBN9bZ6B6vT6xNS6tpVyLlqtaraZOpxNlwR/U6GN8kMApj2ZnZ+/hl4NI5hsZdsPgGRwPpHyeIIxeqXSyIJug0gOzFPzROigp6g8tTYyvXC5Hp+8VC1pyHcC5PjFO/ysptmG6rLgec+9ppo/vPRBJ+Qm4efbZZ1Wv1/WTP/mT98wJgTvJGV97I52sZYE8aEz574EAY/RqUDr37sTzKjpp0AhP4ZcHif6Zy6xnJE9LBKR/fQyMiXORKfWEjP8OGfHzeHUDm+M203dmlRTtLJnimZkZtdvtKNveWsdifq4DSDo6Ooqtu4eHh3GjCB87GVLki/F5wiOdb58L9ND5g7ynIAzw4nbe79kTIl5dwrZ5cgrfxPnJpqbJBsabRw44vCLN7wGVrPPKsky7u7tx2/g8arVacS7x2bS0Ije9Xk8rKyva3NyMYJPMO/zf2dnR22+/rbW1NW1sbEz4ZunkIb2VymjddrvdnkjmdjqdicQCiTru9fDwMO4QtrS0FGWCzgwAKeu80E14sL29He0pwNkTpD4HZNoBiCSXj46OdPny5bh7IbLM+kESWsx5mqzwNWzT09N66qmn9Gf+zJ9Ru93WJz/5Sd24cSN3jr5aCb1BPtkIDZzqHTmSov76vKXnc5v7oL7f2+nOS6VSSUtLS/q6r/s6dTqdKENsHjY/P68sy+JW4fv7+9EesnxmYWFBh4eHMdh3+fEEGrrOffv9pfYc2+0dEW7fPGnmfsgTqeALxuP+Mi8R5Ulx9z2OOX2sqe3jb6PRiIk42gnhmbf3SdLGxobq9bo2Nzfjevr7VRfPFUyRAWu322q323GXm1arpY9+9KOan5/XT//0T+uXfumXohB7qRTBbTabWl5eVqVS0cbGhi5duhSNAb3EBDdkAh3oYYTYFQtmArqYJB7S50DKM26cM3VuGCUy8iGcbIuKUrLDYKk0WtPVbDbjU9Y9y+6LaL11gZY/qmIbGxtaXV2Nv/cIPgVb9wuoaIFptVqam5uLi+mbzWbs2fa1ZAgs2VZvH/DsQVp9SsFiGgChWDjOwWCgfr9/jzNH+F3JvCWDc4UQIrA9Pj6eAEEYRp5D4P253F9aucBRbm1t6fj4WHNzc9GJEWR7gOgbGZxmSM8b7DImAjYfn/PZeQVvMAApcPUA1e/TA2YCSdpyvJriv/E1QG6UPDj1Si5tJl5ZQae8ddbbVVxWHBySZXYQSmDvusu8k5Hythl006/DWKanp9XtdjUcDuOCcucD9+J8AVxz/zgF+ON64vOQ6ivZPOTMg0uXf9cf9MEXNvsxfhz3684eOXbeOaWVEA/MU5kio+fyCqUBpgfXqcPlumlmk88AM9jhtGWKbOVp2VE+T+1Tq9WK8sL9esse7xknxwL8GYNXAX2sJB6w/fiv1M4yTlqTHaB4Jd/lL82a+jyzcYFX6ZCLPFvlVckU+DgvHLjgy1qtVgw69/b21O12J/yTk29r7utMyuWyer2erl+/rvX19YnHUvhGOQRet27d0vXr12PrXtrpkGVZ9Nu9Xk/r6+tx5z9P0lH5kpT7vKb9/X1tb29rfn4+toEfHh5OPNuQJDI7wJLo4Xi3O7RAIpuenOHa/B7s8vTTT6vb7apWq+nu3btRhr0yBbmdQAZZY3L37l198pOf1HPPPacXX3xRzWbznt9/tRDyxbMll5aWtLi4qJmZmTgvu7u7Wltb040bN7S6unpPV4IniVIepfb2LB66zXeMiY16ECqVSlpcXNTS0pI2Njaiz+KeSDx1u92YMPBEJLaRpTgkRMEP3Nv09HSUD2xtmgRNAy9fW8X3nhzCZ6GfaVKPZTUkPRw/pDiS33iCFT8gneA1H6fjIBLvCwsLsXqNvcBW7+3txXvnmYJLS0uamZnR2tqatra2tLq6qpdeeunMOTtzlglO5ufn9cQTT0SwTcS/srKiH/qhH9JTTz2ll19+eSKLhdFlo4iLFy/Gh8H2+/1YkgwhTLQbUYrHWDnghlHSqMWP7CKOiQwOQZk/Nd0zzIyRUl6lUomVCtrwyEp5gDIYDLSwsBDXHCEMV65cidUpb3uRTqJz1ks5WMYBcvzm5mZc8O4BigtVHmhHsViUPT8/r263G5/zwStt0wJwe7+4LxRHcBHKVMgBB2kA6IoDUKJ1kv9xKukuiVzbFcedNfwi8yhp4jlhngVygERmEZkgW+Xgv9frxYf94pArlYpWVlYmAiqfA+YIYHMeKpVGrSi8h9dpMJUaZoBTHqh2XnvWaDgcxrZZMtuAMwcbHkT7OFxvPMBEj8jaI+MecDsglJQL9JAFr+qm9w9/Hei5DqXP26ENwHk6NzenD3/4w6rVavrVX/3V+JDWzc1N9Xq9KP8Ogj2Ic0fBPfkYGWdesJPaLo5LKxGeGXQ9Su1AOu8+Fm/X8IDCA2IHkR4YMRenBeVpNSVvjH5ffs9edU+DLZc1KtkeZPhfdJ22U/Td5YngmWvxXbPZlKRod9wBuw6mdi7lb7oG2PUUOQZ0pIGmz186tx4kwDPIwYNXa0mU4AMI5NAXdCZtg8lLxLh/SRMXWZZNbIjDbngrKyt6+eWX72mBoZ0cP4lv7vV6unHjRqxGIYvuC9le/ubNm7pz507czdeTJu5fSKZyn1yr0+lMBEcHBweanp6OG8x4Nwr83t7eji2XYIHUJgM8sbckIjyw86APuwKlOky74Orqqi5evKjFxUVdvHgxVuuwrR4A80IeCOj978/8zM/EdnI2y3m/EkE9c7q8vKxLly7p0qVLunLliubm5iJ+YyttOp2kUcVzdnZWh4eH2tramkjsSacvlUgTXHnB1mnkGC4v4XAWcfze3p6Wlpbi/aN/nlgBS6fbjYN1SEDX6/WJoIouMVpJU7vN/TJusCH35O3HYADHcnyPHWI87gPoFMrzN9JJcsp9AnbCj2Xs/n+1WtXc3FxcguTr0KmMDQajHTuxERR82LyKjdg6nc59W2XPDKYqlYqeffZZLS0taX5+PjohZ8bBwYFefvlllUolzc3NTbQAMVntdltZNnpQHYJCvzwg1hfAu4FFWJkkghIylyiRAzqEgAw8PZxkOlk3RbBEGbRcPtmBiHNwPqpJ1WpVV69ejYB7d3c3Bi4hhInoP20voOVkMBjtVJJO/NTUlDY3N+9Z1J4qZArgyC7AaypSZNUAHXlg3XdR8nnjM+9vd2VjTDh1f/k8ch9klD3T4P3ezCvVCcaC0/ZrejsUiu8VDa4BAMHx0ao5HA7jDk5UKGjVWFtbU6/Xi4CIdjwe3JkCW8jv7X40NTUVW1IdwLleeRACWMB5p9d2UJHuKLa1tRWTBgB7z47yW29v8/ZcDyQ88HKQynySpWXMDhppr0EfuTYZWgIigCE6geF0x8HvObcDGh8XhhfZePnllyUpBs3IXa/Xi3rENb06xr2hxz4HHmCngYvPj5/D+ZaXbXZeQ+5MUgDuvPTgKi+QS6+VJke88pIC/3SsHjz596cBkzRI9mBZUswYezLHA0NetOmRaeRcrv/eSse8IdNkM3HuzKcf42P2INPXb3KvyJlXbrFP6JnbegcqPt9encTR+/x5NcPJ9WowGMT2Sc9Sp/w5K5jiWuigV2bRQ9rpfKMin+vZ2dkY0MKPw8PRA5DZ9Il5Qa9p48QP3rx5M2IG3wTHeZdlo6pPupNhCEEbGxs6PDzUwsJCfJ7YYDDQ9vZ23AXY9ZcgCV/kiQVkD5tI+xRtQNhYf7gvfAEEu54RnDI/JPp6vZ6Wlpbi//68MYI7T2BC6AUdD/1+P3YS7e7u6stf/vL7IpgicUJSeGlpSVevXtXVq1e1vLw8sXmW2yBJMZB0/IiPlyYfzbG+vj6xZtRtbZrs4Ld5di1NsKbn4p7Qg/MS415fX9fU1JQWFhYkTdpI6aTbSTppcwZL8kKewUkLCwuxyuuJF7c9jJv3vjzE8QU2Ie0mgWfMiy9z8U3GvEpIACgp4lb34WB0EuDenui6X62OHn1w4cIFNRqNqMcecIL5qFB1Op14TcZcrVbjbqE7Ozv3nbMzg6laraarV6/GSZMUs/VuQAEirImC2USyGCnAO4EU5XKyMjzPCaZjwDyg8Kw7E+rRLw4xbW05OjrSzs6OdnZ2ImiUTowo5z04OIg7AwEkMeTVajVujHHp0qWYFT86OorPPvKACEDjY0BofP0Jjp+F/isrK9rZ2ZnIbsHnFNSQcWm326rX67EF0wMpb7v0CpNf20Ga7/TnbU1kIOAb83laxhVBZx4IJpk3vmfLXQ+YPTPjG4F4NYyF5QcHB6rVarG1w50gATHXZ70eikzbJcEXzwDh+TIc45nM9D5TUHw/ItD1+ffMd14VgkCBzyGuTaDlYJgkgVdYHGDze+bBAy0PsLm+ywsZsrTahIy4ffBMGcbfs2sOgPyekQeyaswBO4R64oTEDveR6p4bfZwC7QboFiV+ZMYBaBoocP5UFtL78cAEQOVJhpQc1HJed2qp00qDGQ+sUvKEiYMAB2YOCnz+Tgu+uB8Plpx8TlNQ6veR7gzqtp77nJqaiomrRqMxAfQhQIt/5+1j3paV8tNtmd8LY6Wrwm2iBwNp4ov3XANbWCqVoh54sMeYPOHAb9PKEtdGTzgeEACQ8kyu2wDmOJVT+JA+tgJekSTkt3lyViqV1O12J4IvstA8282DS+ao0Wio0Wjo8PBQt27dit0iACiSuT4e+FwulyNm8AA3hBDXy/mGRbu7u3GjilJp1CYO2CJ4RlYIegBvtAV5pZ8KHs/xQV4rlUrcqdj57LqBTkxPT+vg4CBWqMjIHx0dxTZ2Ns9ye+vJsRBCxGLwlc1a3kvE/DQajfig4UuXLuny5ctaXl7W/Px8HDdy7EkMSROy6DsloyfogXRit9CFUqkUg21PdkiT7e387/Y/DZbuR/jMBwmmSqVSTCyAg5AhcB2y7BVe6cRv7u/vR+w3MzMzkeCHL2xY4sG/8wu59GQAwT28w1Z6Yt39lW+Ewbw7ZvEkFN092DGqH3iI1gAAIABJREFU1Myb79BKcjztWGk2m5qfn48857fl8mhzl+3t7aivbDjDmN0voGPgxvvN+32bOWnFOz4+js8mSltdMDg8r4cABefj/eQYB2kE0O7cuaNbt25N7J6RGmFvCeGGAW0IlwdxrNNJs8qSYrub91TDKIAIWQEmCQEol0db+LITIM9yYOwI0vz8fNymlw0mfNyMpVwux8CNYAGDvrOzo9u3b0fn41UKn1SejbOwsDDxew+eTqtM4aS8JQal8es4cEyBlVeH0ow7xtx5iyxJii0ZvnEEwIFreradsTg/kC1a4NgVCflAmaiEMV7fihPjACiBh2xukZfh4T6d0kD3NCLpkM4H5MaIe02dMORZUA/EqKR6mw9BpTti/jK/nqVC9o+Pj+MOUvv7+zGji4yk4NSDQ5wg/PXruB7wW88Eptk+EgXHx8cReLgOe7UNx8s1/cV3R0dHunv3rqSTPm/vSQfkO/jk3F71SF/S5DOCPChzWUnn0ec/7zg/xu0JNjgNEJx3aWCQ6lNKeYkRd7CSojyl509bc91+p5Xx1AmngN5BEJvp4IfyKu2ekGB+/DsP+NF1xu7BqVeaSASwyJ/jqAIxvtS2Yoecx8ybbxRBIM+8pYGU20aASKpbPhe+lpANInxsXpH2YB3/1mg0osxubW1Fn4hs+TnyqFqtamFhIQZA8Ih2fvcbjItqY6Uy2t2P5zsxT/Ch1+tN8AOZA1SlcgTPKpVK9NFgFFrA8P/s8oWtJyhnLaVXMPmcZRCpHfVgynf/dN6nQTJVs/39fa2urury5cuSFINBOlbwa2Cf1Ac4LkKPTtPzd5oYU71eV7fbjc8kvXLlii5evKjl5eUImJ1HBDgEBSS5UruHn+L5f57ko4LhdhZ5q1RGa2nAaD4nPhZPPqY2+0GIFsTzUgghJpHhSal08lxJ5Iz1iL4xi48Rm8Uc+HMNCSh4xIJ3e7gNc2yNHPpOu3RzIGPomT/2olarxa4T/nd8l9o3bCdzj976Mhh00JPcPBRbOqlaVyqjZTwrKysaDoeRZ9iHdH7dvqcFjTPn+H4CsLCwEB9ymrc9MhEiiwDZnW9jYyNuquCgnExDlmW6ceOGXn75Ze3u7saJ8T5znAlZNYyZ72xD5MwOfxhCb1Hz5/LAZMAHgZ1nLZkAQCMMxUixYQZ92QhzuqC03W6rVCrFLZm5tgPL4XAYAx9Kk7VaTRsbG2o2m3rllVcmeqfhI3Tt2jU9//zz9yy4xYGQ1cehIbReDXMgwBhdsMh4egDr7VAuiLx3xwIo9Wwm4MRlQzpZ+O0BUblcjjxFNnxLaknxGL7zTJYDGwIFV0Bf1M54qJrSkuc9+W50nM5raP0ZOW6oUpDrGZv0/M4LD4TJAON407nx6lQKsFNnlQZ8jANA5OvbkDmcWLrVLCA1HYMbTr93tzHSyVoRlzucgDsGr1xKkxWzvICKFhA3orznHtJ1jMgb13Lg6/OWJn/S+cuTn7zsoGfM8mTMg8e8a/lxKRhhzOk184KoPDlNP+MvMpAX7KROC1ki4QHv/XsSRd5unAbmnoHkGG+/BeDyGfdAOx7z6m2lg8Egrjd0Ow6w4Ty0oLidZYyeHEl1zbsi8hJSnMOr9enOgdyD22vppGWWKgVz7aCL35PQI9DIskzr6+t67bXXdHh4qGvXrk1kez0QTImNXlxG4B/z5PJeqVRiooROBZd5dtyjW8Oray7rDnyQGXw4voRr47OPjo60vr4e8UO5XI6YxgMxErDMr6Q45+CR1K5B2D/XE08ceiVye3s7JhMODg50+fLlqAd0y3i1YGZmJp7Xg0pfVM/37yRh38GBi4uLunbtmj7wgQ/o0qVLunDhgrrdbgxwvFvE18fDG4IkAioCHscobs9JkqetjKm9cV6CjXZ2drS5uRkrfp50Se3+eSjVdemkinxeomWT5DB6zaN8XKcJDJGnVEckTQQmzAHH+rpGAi8/N/O1t7en9fX1iAW5L4Ijjw08uYFf9gQjfOK9V4M9ac05SHgQX/AbbHCWnewYikzAt16vp62trdh55JUx5Davcsu9nGbnUjpzhmGiVwowUIAn9sInsKhUKhMP2XUQRYBSqYy2qbx582YEqezE5tlBSvfcGBOTZtV8YmZmZmKmDsY1Go24fopMF2NKQbpXUA4PDyfW6xBIINgsVkUIeD4EQlCtjnYdbLVa8b6OjkYPIaZqgNHg/IDCZrOpq1evajAY6PXXX48BVZp9wFDQmgWfPXvmAYJXaVKg4+DcHRTrzDy74cGbt9qkf5lTPoN3nlnCEVE9AaQCnD244rkg3C9yyfyjhL5zkgM3eOgKg7IzVoy8txXSVsauSZwnnYf7UQhBc3NzkTduYOCTB1EpCM0jwKeX/RmvV/68SuUZVA8A0jnBuDLHvpOijz0NwHmRYEiDIsBDqVTKNf7OL5Im6KqvafR5SOUYO+EBFJ876JQmn5njvCYj7+0M8AHZ8Hl0UOVrXlLZc4fr85jOMffgOpR+nwJ0r6j4XPD7NJDxsbhNzQumPGGTyqR/xv37JhEpoJFOFhhjj2mPTscJ+PFW5dTB+f0SbHvVhTn07gpfK5DqHdXdtL2Ie0kTf94uTObV7ZwDB3wjwJng0SsVzCe6EkKItseTFe5jU7uKvpH9h0e+xoDxuO/d29vTL//yL+uVV15RtVrV22+/rW/91m+NC7K9dTAFIiRdORcJS+Y7DWSxtbVaLe7aRwJve3t7Ygtxn2uI+YLADNid9fV1lcvlGFhh3/FhJELZ1RUfSbACj32OvDpEOx7BMb8FQ3jXjds+ZB/bjT9mbdSdO3dUrVZ14cKFKMfILetomIPNzc04LvwYVUlk/HERiQR2Sbt48aKuXr2qixcv6umnn447ofkGUbSi0UECRkyrKp40gt/4MeYtDcbz8AvVRHQfHaPLhM1Ims1m3NRic3MzbloBjnC+nRWQ5n3ndvM8YNypUqnowoULmpubi0HKYDCIm0nt7e1FXZQmn9eIDKS2HZklgeN4g4AFu41s+9ytr69He+i8oVIFj/2ZcfAmrSTn8cl5hSwMBgNtbGzEFlpPyGKDWE7kyX30j0ci0TLqyV94As88GYhvhz/Or1Pn7KwvsyybKM17pn52dlatVmuipYzolmh3ZmYmTgRCjQG7e/fuxBoHJjpdRCidLK7kGtwk2Zm9vb2JljYyzp7lwLGyfShZDJ6mzoTD1L29PVWr1bjriXQCvD1wCiHErSmZDG9L2NnZUbvd1tzcXDSU0qTB96yFt2f0+30tLi4qyzJdv349Gh2fH6/oeTCF0fCMpQdVtEK6c3OAgmNIs90e0LrjJsBJlQSHzjk94+NOnX542jto2yKgZZ0MjmJnZyduV+8ygfEFJDjg4N7gmfPFA0sUxxcIe6XPd1biPv3vWYSTAzSiyFBaNeK8qcHhWJwO99rv9+NaKQ+gfB6RG3dCvDywJwvNy0Gvz2deUJmXjfedhNBv9N71Bz4xJ15lYM0KYJX59KDN+YVupxUpDwgYh+9UxDjc+PIeZ53KOnRW9TJ1umliJJ3fVN+5fho0puSOzOc1DZS4pzQYzpNpdzR5gVYaSOfpXt5xDqTyAilk3Kvo6XlCOFm/6UCXCoPPq2dW3XZgBxm375rn+u4VCH9kA+cGqHmLTlqBJgGC7joQcJnHP6Tj8XZD5jetcMFz7wrg+iQOvFPEA398NOsIb926pZ2dHc3Pz8fxURFIgymSHQBgbCeVFffLVBL5u7u7q4ODg1gpkDThf51HjhFSGwQPvPJzcHCgVqulCxcuRHsmKWIDNuGp1+txcyrm3tvhsR9bW1vx8Rp+XXwZ6259cwyvdiC7VPIYD/fZ7/f19ttvx23mfb1Iu93WxYsX471NTU3FwJM13OiKr7d9UAJbdDodXbt2TVeuXNHS0pLa7ba+5mu+RteuXYtBY7fbjTpGgEnwSDcO+rGzsxNBv1dT4IsnqlwnkB3mxrsaOA/HIFOchzWyAHACAPwMdoa1aQQJKa5Jk3jOK8cRfjy+5ayAzAmes8ELGLZer8fdH2lTdj6iE9hKT3iCB+CPd1dIky2QBLBspuY4nfvwBAAJej8O3noyCn6hV45PwHvgQD/P9vZ2tJedTie2dWJn3R4wRjqfPCHi/gK7y/8+Rrel/oyuM+fsfhN66dKlWP4k8qxWq+p2u3HbTbIACB4TQssAzMIwr66uxsV/ZPT29/djhsjXKkmaCG68Z5wH8B4fH8cgDUa4oPB7otzBYKBOpxMF3YVQUgR7nAOmr6+vx4CNZ06RAaXitrm5OWEsPTM3Ozsb1+EQNePouE+E7+joKO4kBy+uX7+u3d3deH6UBqOVro0C/OPMvILElvIYev8OAaJtjs+5hgOLtGyPMUUOXLl83G5YWDslnewag8xgBDDYku7ZSpNFxK4wKTCDUGLG4AoGKCDb51ltDBnjYh7g13kCKeaLjKEDUn+lgDXlF9fESHMswYpnP914eZAg3dt+4oYSuXNHBEhyfjFe5t6DQ3duvh6A8zKWNNBz0JSCJ0CRA2f4k663TPnHeN0pY2PcqTp/pcldRVNAnGbb0kpUOl9577lGerxX604z5D5fjInPUsfh1/P78Gu6vOSNHduSBkj8xuXC5YnvXd+YT68eu876y7OSDlY8KPPWWQ/GPegGrHsVNA3gHJi5/2K80r3Vfz7nfr2im8cj5iwv0ITnroMOFHxXK37r67nS+YHXKaCjywCgjt3hRaJqaWlJa2tr0c/lBb5OjUYjAleuBR7wVnuoXC7HNco8E8g7VZgrEqXwArDIMdgqfCjzwFgI8I6OjtTtdiVpYiG7649X61hDjQ8DbJNQBcOQtaebAr+M3CBLPjfM7fb2dryOj2F/f1/r6+u6evWqGo1GbN0EnywtLcVNKcBEHnwSJJ7HP6FD/uL+n3/+eX3zN3+zZmdnNRgMdOfOHX3xi19UCCG28FUqlVj18C4IfAC2mzVx+EEqdsgEiSp0gODHAwH3fch3WmVO/Qz3As8dpzHX09PTarVasQOF9WqenHQbyjl9fRrvQwhxzARA51035et60sSUFxFIUvgGM2kwQCUr7ZRALvxY6eRRM6VSKQbF4Gxe2AhP5HpHhFcZ3bf7fPkuythcAjh/RAbBXaPRiM8PczyLTWA8XNM36QCns2OvJ83SZJonskgG+brH0+i+jZy+/mFqakqtVituNMFzoyAMCoRyenbj4OBAGxsb0Whi3FIhdUDglRsMp7caIExcCwfngReTxeI9rkUljev5QwTdAbL1OcDSJ77T6WhmZiYaGtaKoaS0RQ6Hw5iBKpVOKhQItRt/nD7PIyIzeePGjQmF7Pf72tzcjELEk+dxtCjycDiMFR4Mipcv4ROBFMGNGxwUgF5zsh/eMuJ88woQ8+qgnTmRFNslPbDFILkC48gwfoAHnBDXx+lLJ5WCFHg6peArlV2ABHNPsOngx+X2NGLDEZyll56dR7xP9cHnxBMGIYSYSWP8fMd8p1kz+OxBjRtv1zd2u0orNZ79SoGyZ/BTnuYZW4j7w5AxXg+6AXs4s+Pj44kgnLFgO3xLaWQA+ZcmNxDw+Uv54ADVs8yQ34ffS969+TXS+5dO1g+6gfeMX17QkQJC7E4qU6l8+b34OeAT5IkKDyjdMfn/DrrTAM3HeHx8rLW1tbh758LCwkRw66156T277vq9e6Y7hBAz93kBZZqIAcgPBgNtbW1N7M6KXcIfeIAEoEd2IfxkGtQyDpdJByOsVcAvYHNclzzYA3iEcLIxk4/N+UNyU1Jct0zLzMzMjJ599ll99rOf1Ztvvqnnnnsubs/MuVLZgHgmI2P0YMjHgm0loGObcddV7pVOF/jrAIxz+cZWvJgDn7fd3V01Go24dpXf0wbmIIr58ioVwQ/jIoHlAa6fF776kgJknpYoB93gDTpICEB9EwbWbXc6HS0sLEzsOkuLHWPgOXquw9huf7kv8mTWYDDQK6+8okqloueee07SCHd0Oh3dvn07tsr5vHB/BErIAcGhdwLQio6MpG2J6ELqU1I7gF9FJtz3uH6nbZUEgQTG09PTGgwGsbOKINCxifOPLiCCHHhXqYyWs3Ddt956SysrK/foSx55kAQvHLeAI8CCqc0n8eAt/u4jUhsJUZGCh3zHngngdXQTX0sV3pN42H/XUfexboeRaQ9cPdlJ1YrzpwEP5IFz2mqNbaMC7AF3qVTKtRmOfR8pmMLYEZzU63XNzc2p3W7H0pr3xaIQLrwYISo3b7/9tlZXV7W+vj6xIwigCGDEedwBEzg5Q+v1utrt9sRzI2CiP+QUoec5BQQuvqgT4UgBJYYGw+agn370drsdA02yMG6gMfZZNlrPMz8/r3q9HityrrC7u7txXRWROxmTbrer1dXViUoaQjoYDCYW+aLYfm/w1Z2gA1wH5A4OvFze7/cnFv4xN56JzAOIHsgQuLoxAPg7SIIvVEXI/vFMDn4bQpgIqIbDYXS+ZJlSIJsG78ggRtWDHV7cJ+09afn6fsEUmZU8MAx//Fx55AaD3zqwcD67sXEH4wbaAyz4wDl9Hn171jxAy5hcLt1wOshmvB5Q+1x4YMbxGDUezFipVOLCVOwGSZTh8GTXHulkF6KUh2RBfWOD04y936d/z/FkYVP5ypM7bJiPJz0HcuxA0GUrDSxcZrAJnDuvipCCdw9MPJDxe3YQ4+f2+z0toES//FjO9eabb+ozn/mM3nzzTT3xxBP6rb/1t+rpp5+Ox3iCKJVTBxEuR+4oAQh+zbz5dbmURjJP26yvA6JFGgAM0JU0MX/wAz/n68ckTfCZ+XCA436Brg1vd0znMQ3SXM898HO7Qet7CCHa40ajoQ996ENaXFyMnRwscvdsb15iqt1ux/H6OkQAaAqAm82msiybaO2jSlCr1WKWnHnB9tJGyP1hH7xNmzmCp8gFz41hAym2j8aekJwBnLs9wq7gv/jOAavbTZcvt/0eDHMPBFMkPx2ATk9Px+4dSdrZ2dHNmzcj/mF9FnrGZgOAX3bCTGXO9TG1VdDOzo5+4Rd+Qa+++qquXLmiD37wg3G77ddee0137tyJz8dKW7+kk01eNjc3o1ysr69HGeDa3tbnY+Se6PRhTtPj3P9hS/0+fTkE3SbgLN9NGoyHbCL33iaI3Hhy0jt6yuVy7ICq1Wq6cOHCuYMplzfkxxNYzuM0qeqJ77yENsT54Q3BjAf9zsM08evBqM+Dty97IOwbVMBP8DyBPzGFb7uOLFOdBCemOJ179jGhY+mOmugV/gzesW7M/Ueejcujc1WmqOhgbBAgd/JkP3ySmUTA5/b2tm7evBmf9ZRmzFNn4s7PS7wIEwZpYWHhnofUMnYUkZ2K2OBgOBzGbCNRcVrJcoBdqVQmttF1gMdT3RcWFuJzn3q9Xsx8I4CHh4fa3NyMuwKhlGy37g9Sa7VaMdja3NyMD/klO0D1yzMjXqHDqfjDkeFHr9eLxzk48cAToaNf3LP+DvJw0PAVwWT+3BDAtzwB9YoC58FZ+bbc3ibiwSO/Y10PBhelxrljRNJMPDxANjwI9ZI2GUbaUgDoGASA22nU6XTuCaS49mmBRwr6eA/RDuIgLA2o/FoOslPwRQYnnQtJsSLnMoLhxrChm26IJMXWKndULgNcw41jlk0+L8n5AZii0ooMcT30wVsFkCPm1sG9Z81PA9t8J00mjLLsZKOe0wJ2f58GI4ydzBif4djzFvq7Hvk85ckKx/u8u531bDI8QE7SQDuVTyfOi2N3HuDomAvn89HRkV566SV97nOf0+XLl/WLv/iLunTpkj7wgQ9EecBepsF2muxI+YuuO0hLee9BsleAfMwu527zHJAA3AGOg8EgZmwdWJ9GeXLm7fMOQACg2DUHVMg1AIJqHOtOeRbSzs5O/MxlimoY4Bs+e7LU+ehUKpVi+7u3DnJ+r7Qwr7VaTf1+Xzs7O/EY3xCDbDHyCj+8rddlzZMBZOjhLTLERlC0/OFDq9XqRMDBnDNO9BEQznwwT57AxUZ5pSWtAnGP+DIfK+uCXT48EBoOR89DZB0qDydGBkMYPbiYTUscA3gQdR6giKzx0GUwYbVajZ0vN27cUKfTiZtuIWfwcGtrSzdv3pyoMsCvubm5iS4c/jJOB8iOQdyepLaU3xHUIXdgWeZ0ampKu7u7ETO4bSJ4IoHhMu++zJNefOdzPj09ratXr+oLX/jCuXjtmzi4P3JbReCRVhM9QAfX5SUIsVdgLdZhQXl4D/sonWza5FVGbIUnaUmEg8/QScaS2mhfW8m9M8fYOrCd4w0PHB3fcX7sIZ+hT5JiKzDXcbviPvAsum8wxYSRuSHb45kXjLgbPGcsmx2wiDWEEA0Wg6Z9zoE3QgqzKP3hGFiI6dkQnyDKty5UfIfxJtPD996r6ZlQjDzX9x7ZSmX0DIG7d+/q2rVr0QCySA7B5D691Mh9ECzhJAG0riwp4Or1ehMgn9ZLPvO2AwQLB4RCkP0COCBAGAn44IEXPHSgS/UKBU0DAgf2/rtSqTThgHB0tDl62ZxtQd1I4fzhBZVUHJRnPJhj5CmPcHK+mHJ7ezuW8ckUo5y+TpCWmbOCqW63mwtEodRw+ufpZ8i4GyIHrumaJb5zo+rZLs8a+VhIqHBs2srh+u7Am78E/+mugnntZ/7XAYcHVh5IoE/YGmkyU4+NQZZ9LQAGXzppKfUxOO+5rl/fr5VmeH2+8gLg9B68pc+vy2e+lsJ1Kg3u0jGeNh53Mi4HHhBie/N44+fPk+c0UZCeM5Xrra0tlUqj3VgHg4FWVlYmbAL2lnE7EEXvnT/cE0GPd0C4w/UAEvDrjxVwAAYRuBIwoIN+bsbgCRnau53SREc6fsaF7OKLAaQksfgMP4xNRweQF/wNGeFU5jiOdh9/po3bUGwP8gFVKpW4cyD+yufbExwArRBCtA8kR0gOEWTQeu6Zf0/IuB2ik8OfqcNYfY3YcHiyFqRcLqvb7Ua+4NuxG74TLokk2o53d3fjLmvSydouEqR0qXAe5sHtLTrGBkeSoh9kUX673Z5IbiF/R0ej53XyHDbGeOvWrbilOOdLg6fURrhMuo46HR0d6ZVXXtHq6qpmZ2fjfLM7Hpio3W7rypUrqtfrWltb00svvaTXX389BqP4lFarNbFBiOuA60na9sXn6LnbPjCLV5zY1AA5YhO1ZrMZl6xgS5zPnhjEd2GzPYHnAR/3gN43Gg1duXLlTOzh5BjbgzjHo5Imqj/8T1DIcega8+o21PEWa8TY9CstLjj/fV6QKxKWbifQB+bIOyywTY5B8hKxfO4+0+OGvE180mQBdprgj2sSI2Bf3Q9ybQ+6zpyzs75kEuhr9p37PDOVXtADIozl3bt31ev1Iujk5gleeNHG5YJbrVbj55T6dnd3o3Fx4cT55H1GhWJ/f1+3bt3S+vr6RPDnuy/xG4xfqTRqVcSJeVWJzBUG3zdEgD+eTXDj6vxKDalnxdrtdjScHENJE8PhVTmfC4yW70yTBja+6yJZHBQZYfT58PPCHwf1fm3eS/e2ADFXnu2g7E42lOwXG3bgdDwjTZXIM3/ee4/zdOeRGkxPAJRKo4W7n//853X37l0dHh7q67/+62PgCh+QD6qv7qhScrCYBv+pPKSfp47OjSH6who+gLkHUqlDcMMqaWJMvsGDy6fzC/55NsjBAX/7/b62trbimgyfgzQoSOfGM2KANu9nToMxZBDDSxuvV1h8frFtkiayVqkjJ3GQBi5QGsicFki5g2F8Pld+jRR8wQeXGae8QO004OTj98CTsbnzdoDulY+8ykQeCMobH3Pk39F+3Wq1NDMzo0uXLk1kTn0xsc+ln8PlLwUR6TEus/Cm3+/HZ0pJJwkw70pw30CW1CsP2IbUNqbglHthnM5bBxQuj1yLZE6pVIpgQjrxLYAhKmecA9DjOyO6TDr/AEm+7TdjhDygglhvxXlp0caPVKvV6LcIAgk8vEpwfHwcn5XEZj8OXPFbzifXOdoV6dDwjhrum53lJGlra0vb29uan5/PrYJ6RYhk7MrKStyVGB2Fz9hiTwgjb/j2fr8ffQg8wo9QuRkMBtHvwWvkynWWgJRErgfh8Ai7n2bcOQbd5Pd5+uvJh7t37+rtt9+O98AGAdxLpVLRl770Jc3Ozqrf7+v27dtaW1tTrVbTwsKCut2uOp3OxNo+6STg5oWck1xmnD7mNMEyHI6eTeWVO/d7YM8QgrrdrmZnZyd45TrgCS+v1kiKQTv4D7137Ixtu3btWgy070eOp/NsXmrDkXMvCDCPbn/QY2Tp+Pg4rkOnMkWQjs11vmKH/B78Wp7YoriS14nj+ozODIfDe+IB5IHjuX90LIQQA+JUlp3AR+Bktxduez3YZB5SzHoa3bcyRYDhW226EXUG+0TzGUJLIEWpnCgRg0YgAtO4kVqtplarNdHHibMi6wCAdCZIJxGtZ98wcmS+uJZnAQFYMJxrbW9vq9fraTgcTjzckHs+OjqKCzMJvnzhcTrZvosSk+wLYDF8ZHLgA7xgEay3DeAEfI0B95GWTt3xUE3xygv89DYQqj1kcQDhKRj243nvlALv4+OTSg+Biq+X8u20vaXKAyoyy2QfAawOvD2r4crtSsXnw+GodeGpp57SG2+8oddee03PPvtszFj6FsgOhB3AOfnOUaeBYsbiPOSvz4k7VXrlU2CROlDnQwgnVY8UYOK4SIQwJncQ3oPsht955z3YebKRZpU9+eDz40Eg5ODagYFnubAF5XI5Zj8BdKnRTcFTqifuBNKg97TgJu+epRNbgS57xd0dsf8Om+c2FzuVF0gx/rSqkKeHnniBn962iw4BkDguDa6dzgrmeHGPU1NT+vCHP6z19XW99dZb+vCHP6znnntOtVot2qU0AErnxgNkPz9yn/KVABGeUJEisCUY4bxUprxqzYvspgMO9JIEJJQCP+Yg/W0KFtNAOE10OYhxew8UBC/6AAAgAElEQVQf3LcB+AA7zk+fK7eTaQCcyh1EZcntAC1hbj+Oj09asL0NDeC8tbU1UQWAH7wvl8v37IzmAMnHBkArl8uxIyaEELPVriesWSJhCIDnvSf/eIbN/0/amexIlqRl+3OPIWP0GDIyKzOrmkYMUreQWIF6zwUgseNyuBoWrJBYIiG2SEjsEKPohqKrszLmOTJ8+Behx/zxN80j8hdHckW4+/FzzL7x/Qazw5hop7Sd9WJ9d20wPq5F+yV4i3Y9AniWDbBWy/OEpg4gSUZmwGQbMxgM2mYvl5eX9enTp4XgzHz2mH1f5Jxngq2srLTqpyukrFteWVlpOxNubW01eSGIdPBvH8qYndxw67jt/WAwaM/GpBplOpAEp3IJ3QgC8V0kkB2AIqds5kVnFH6T5Rfwm23tq6rev3/flmm8dGxvby/g0Ex45AEGouiQ+pB849qPj48LD8S27TGOpzLLfXpJIuP/XkIox+H73N7etm4k+0MOaJh+2wERhQ3G4AMsxJpMkkQc9g1cz3Z3GVbz8WwwNRw+PcgLgwLxyKBYWSjN22nzG6oJDGplZaU5LwstE3ZbxsbGRu3s7LStUrkXAQPKlVkyGApAdnWN+YxGo0ZQBwQ4cYiJYNBKxrlXV1dtZyAAtR/caAW0MsM093MDVNzTbKf36tXT872ur68Xns6eiyAteMy5qpoxNjjnLzzAKOLsDawRegIZ5kOWk/tA+wws8r4AFINwetBpAXh8fKydnZ0WtAFwANUYPRwfdPPOc8gS98jMiD+z7MILXt9++22dn583Xq2vr7egeXV1tS4uLprxXnZ4J0XG4P8N9JJeHAaP0NqtqT4wNPwFqCTYBGhYfjjHz3sAELnt0wYaGsIby4GDvzT2BoAOoOywE+ilnXAwbAeCzAKQ/LmvydoQ6G3D77/8/xx/koeeD3LnQMrrpHrXtw1ijgYRCaxybD1HlmM3yHfiwffz+kmcTQLXl450Toz5pz/9af3Zn/1ZPTw81M7OTu3v73/Bc3QvbSl+yICravExCFWLNt4yi82wA8b2DYfzjRAAjgTCgDDkEZn0eiZ28+I+lu/0Bw4OXekYDp8qIc8F8k6MoMsEWNgrbFpVte3VOT/tj+UCmTDQRi8zwCLJiF+GjvwGfrk9162Ad3d37dlNq6urrQ2PsZGABFN4XPYzrNNwMJib1iAjW1tbNRqNant7u6oWq13YDVfl8C+vX7+u09PThWSt/ZsTDswPW+RWwcFg0Pwa7XxV8w0SwCNsJ44swAv4zPvpdPrFRlSMbTgc1nfffVfffvtt/eu//mv96Z/+ab1+/bp++ctf1l//9V9/AWJTHtKmuDsBmb25uWmgFb5UPdnY169ft9ZAV7D4H3mYzWYtMMD34D+Zr1vXnSQkQEUeSX46aQ12ImnL2nT8OoEteu3uAeyxH05NYoNNr8A33lzj8PCwDg8P6+TkpF46GKNlOulv2kIP7F3aHAeC6CVJWNbqWeYzwYKcOmljW+rNeRzYpj1ygsvj9W6KDuK4B0EqgazvAzaHt/ZZmXgjcINn9g8+7Eu/1se92OZHWx1EBJwQwcNsr0Pgt9mGhGHlYXbe6YVqC0KEolCpoc2NrAtGEGbxO8ZlQ8hBRcnPrCL4cfkYpjDmV69e1du3b+vDhw+tlev4+Lg+fvxYw+GwPUOAqgObJgD+LaDQxuAH2qL0OAoqTRgrt1oiWDiHVBoMttcZOCtmh4gRp+rmliqAH8LtXVbsSG20oSPKmff1b/wU+ul02nbPIZvptQBVtbCAkfF7UwQMKTtsMZZlhwEL4AmguL29XX/wB39Q//Ef/1Fra2v1u7/7u40Pe3t7C9WzqqdWEe8wlQeBYYJs/q9arF7YQSf45HyMEDx2IGQ9Ng+cEcIJmE9kCTG4nO+WHetKGl/GTfsHiRbPk7l7rPDTOmGg7PvMZrPmBLe3txd6te0YOMbj8UJWGr3FmDqrn7rksTF209RHzskHdPTLlQbuk4EUfPNW/Fwv5aYXVOF4Mkjnvm6nyfv32nSn0+nCxh/+vjfvpEcvoFpdXa03b94sAL+0Fw6eMqCyPWPO/l0voGSuGYigS1npddYW3mGv7Xfgo6vXBp29oN/0tQ1jjRB6iD/i+uigrw0dHEhVVbPtw+FTyzctYRkYZzBFgs9BYFYDOBxMmcZu94HH/IVu9/f3X6zxyXUpVbUAUFnPiT3wOiruYbDPLn67u7stMbq/v1/v3r1riVa32CNvYB9kbXNzs96/f1+fP3+uH374oQVJruiyJICxeMMMbxriwB3gyNpaP9/w+vq6tYxZzm0bkS2Av1t2CQL+/M//vN68eVN/9Vd/VZeXl41nJC2fO5CBDKyRi7QvjHFnZ6cODw9boOLNsvAR+Al2z6TbAnzHsz3Rr2WJGfygdcF403S3v7bvYCMNb4ZgG+BA1UlK72RnMF71FOB/88039e///u/P0pj59ZIb6T85l0qtecB5/g3zBnN5N2WwQyY8q+bJUVp/vU6Z+do+u/JunJu410Gqx4yNQq8s72wSQxEBPEjbbNU8uDQNbN/ADczRMmx69Xz8suPFNj9XdiAOi0INViGid1HBSHvnFgIphI/MHwEVLzPSvY78j7Dh0BAUO0OAIteyYXYGxIYaptKPOZvN6rd+67fqD//wD+v+/r7W19fr9evXtb293fqZGQML+DDoGAi3UkBTR+4oJYEruwzCfAzi5uZmM0QYRs7JVhK35PQia4QHgXV52GVWABxC6myvK4Lcl8OGLoGDnS2GfDAYLCy2hU53d3fN2dBiCKDAUUJ7Zz1Z9+PxLMswJMiB9q9evaqf//zn9e233y7IKzsBkdWpmm+xbaOQB47Ah8fTA5wJbtJ5ef2dA6KcHzrCNZHLzKhyj8lkslA59Jj8uwxyci7oJODUQbaBsIFagrS8VlagWbjeC9BMA8BFzmEwGLTETFYnmFNWsl46eoELICfbKcjI5U5QnM9fQK0rrfzNDJzp53lYD1JWnOGHt7w3reCFkyd5pANbpnvO3qes+0jA6Iw/c3cyJOdq+Uyn6Xn597Y5WeV24OT/nQBgTHyfFRqPzdVAVy1pk8XHIKPwz7QB4JBwvL29bWDcfiNtiOme4CuDrNRL+7Kqp4RRzg9aUtm2nQVQPTw81Pn5+YJcQBNXI5BZBxRUClmPhf9Ln0USEpBOFYLHq3g9rH2v7RfLDqALa02ur68X1s5hU5grHTXMq7cxlgPqi4uLOjw8bGOihWs4HDac4IAV38f1e3pHoPZP//RP9Ytf/KI2Nzfr7/7u72p1dbUFFfbT1o3Uy5S99P38bnV1tT3SZW9vr3XxwE+WMLA7IrsFZmLHrYtUjfg+gfxsNmuJ7clk0jYZA7fwSBX0lLG4kkRLIo8yQfd9L9tA7II3DHMigXF++PChvuZgDRD0TJ9mO5C2PXmQsuzgEuxIUsjJMyeRuBa0SBtqnhLwMHcSDNyb60J/kuHQlhbaq6urxm/jG+zy9vZ262IaDAaNx07AmgaJNR4fHxd2FjR/k35Zge8dX7UBRUaZABMYAVNdrYKw7Ix0fX3dsk5Z7rNC2FlBuB4jnF13RsqESwcMqPz8+XNryYKQ7r9lHOPxuHZ3d+vDhw91dXVVFxcXtbu7W3d3d/Xtt9+2xae0vuA8r66uajQatc/JGDAut3oxNgwtxpOIPAMwO0U7LQyJwR/CUlVtPZKVAKPuZ0s5G8v3s9mstVsS1fOsrgRZBmQGt5aplC9fy8/RcGYPXvG8MretwDd2/3PFiOyFjb0zqxgOZyJRnslk0p4fNh6P6/j4uG0GgtLScgndMkPtg+e09MC+HVkaRI87DYIBulsdfM0MYAyibWQs/+ij5dqtB8w/EwUOIMxnb7TgZIOr0+h6ykrShHnY8PGynUr6Auh6gMMA2Pc1b/Plw/czbTOQwnZCD6/pq1qsHlgfDbgZk++dwYoD8R4tfB1evcOyw/vMCuc9eo48vzf9nAXN4DyDDgdPtqX39/dfbJpiP5BVxV7wSeWdFjj8GLbFLbLw0DqGDEMj2+mkQfLO4NHy7c9sW7Fr3Ic1KVXzTUNctXAQYnlNHqV9zv/T9tCixcGmUE6iYD/4fSat6FZxu6uBqvXD9EgeYv+5B4k6r7eFN/hH22z8a9V8BzK2TOccNofgmrSrsUkGfMK+YVcZKz4ZgOiWtJwz7dvIOfRz14jlmk4Y3xubCv8eHh7q7//+7+sf//Ef23bqrP3KAOalo2fr/Z4NZajy8JBoKhsrKyttJ72Hh4f6/vvv6+bmps2FOYOjSE7jo6hKck/jRHgMvoHO4FFv8mEab21tLVSkHh4eWueDkykcVGasOz4HGeO7t2/fNmzx3JFrkjnsa7P627Oz/h2+lXnZ3nKgF4kHuZ4TFGw6Ala2nbH9Yg5O3Hg86+vrtbW11QI8vvOz++ARto2xkeCwXXaQbRolPe3L0CMnXI29rUPLjhcrUxAOI0W7DoNzNpXKFAaqat43ToWByB1QhbHJYAZDgHGgAjQYDJpC3N/fN6CbkbSrUCgVBpdtTL04EUDvKJdFoSyM47kUrOGCFisr8+0v2VSDB9ui8I5wAXQY/xQyFN1Ai3N4EJ8DUGdLDDrZmcoG2MLkVhIrgRXJLUC0JjgQhMc2Dr1qB0c6bmdInZlYWVmpy8vLdg7VKu8IWTVfNMq9AEMADDt0AywrmuUGY828OIds2MrK/JlODtB5dhdG18EbtKKlsRfk9EDvc7QDXCfdCXS8SYc3XLC8GKC40mNe2LCQdbKMJa89L65n8IHBA6gyL2e7ssIArxi/Zdhj9ecJBA3gHXRCS1owHWR5rP59T679We88Z9QcWLpN03oMfw0+kcWcH/fyGO1Yc7zICTRFXpkzdE0ZtR7xO9/na+Q3x+Yxeay9ayZgcKsO+ti71zKAaBqT+DOANc1NJ+uqgbIDAbdOAQK9657HYJvklrLZbNZ2/2RH06paADHoF/4XH2ZawWdaBlmoDU/T9vRAmYFSrsGxLrKTnwNfBy9OWnAdWrvgxc7OTqvKpa2E5u4UYPMPbL59JtiFJBz+GB9CAMza59ls1vx4VS2cQ9DjpEZV1e3t7UIw6Lk5WeSuHcZF8JbrRaDTzc1NW5vj4CJ9OfP3Vu8sY0iw/Pnz5/r5z39ev/jFL+qP//iP62//9m/rL//yLxtvs9XPupNyz/8pK2zOwO6cBFH4Ue/+9/nz5/rNb37THjhMpRCbSDCbOCarQMZQ6KwT7sgTlT7smIG4N/GAl4x7OJxvxsU9kUUvE2Fdn9vu8LXffPNNC4SfO+zr8zOSPoyB8eMnezxDJ51EwDZwHVfCvcmOx2G7ge6lzGCHXGTht5YReGAbBdYCt0A7J9wyAUiC27Jom5+Y0/7dfDQ93FH2tcezwZSdQ1W1B/O6t9JOGEbzm9ls1ioNXv/ibMrnz5/bOezkkZlzG1QUiFI9TPfW14wdRqJYbGOJIGTJFgNK2XE2m7W1RN99910b2+7ubl1dXdXZ2dlCJhVFp/cXAQegc46BMM6PzCf09WYLdmIY9wQ3y/gH4HG7ir/D+aSSQjPTnBZLlIdMPyDUgXAP/Bkoum3UbYUY2bOzs7q+vq69vb1GI5SHHnIysbRo8WJeKKS3veawcXHmGjpVzR9YjZF6/fp1kw2yXff393V6elqz2fw5ZC6Xc5CNBPTZ6C0LOnufpUEl+KB11tvBV81b/JxY4Bq56xY0oeUBOXWbqYOsZYCesWIgfU++y0yPgZ2NrgMvA1kym/DcgUPPAKZc9oAg869azIZbdpbpWo93Bh7YJ9aBeBfTDO4Mgg1K/awjxpzZeevYsoAkeZW0JjvIPTLhgz6mDe3J8rJAqyc7ySPrfI6VcXFOZkV9/QyMPG7Px/yuqgVdRhfs52gx4TfYd2wOGXCAGHxMv+N78lv4wTkOmEjY4GMs74C/qlqwscgEc3WbnPUteWSdtDynD6p68vkEJOiPW0EJ5BxksggeGrMDLv4Q3fGOfFXVdtFzEsG6jC89Ojpq14C2+MHt7e3a2tpqyU+Sswmusi337u5u4flS7IQGDRkL5zEm6IFMuDWf9kI/YJmk8+XlZe3s7Cy0L9lWgRnwSw4u80DGj4+P60/+5E/qu+++q/F4XH/zN39T5+fnbezLfJJ1PH2YA1eCqO3t7RY8bW5utja/nZ2dOjg4qKpqG44Q8BofQVcSCrYzPR/Jkbo2GAyabNr3TCbzDc2sL3Qv8T3dML3AHtuO3jnYdzJqMBjUu3fvmsw9d7BJiu1n0t0+z3LBd+kTjKP9OxL7FD3cQcH4qQz6nugFPAJDgLc8pmU4h7FalvmN6WibAc5hjOyhQJAKz6rmft3XTX/Otfmfapex8tcEVS9WphgY1RmiRoMlOyYzlLUsGWRBDKo7rkDlJL0zHUachWNra2v14cOHevPmTQP1BhMGdmQGR6NR6yGvqlbS3dnZqaqqy8vLurm5aTt+DIfD+u///u+azWb127/92zWbPa0j+rd/+7f2MFdnuZxdYrxkpdxLyvwcRDrDQJbNrSNVi5UYzxHhRqAIoAjCcGpkrKwszup5vRI7IhHsVVXbEhT+stmGQYWVxODfypJGGcC8sbHR2g4sBylzOB2+42HFyCzBDsFtT6E9FpSJ6pvXFBmgVNUXz75yRYugnQwhh3dAyzEsO3pjteLPZk/tTRhAeGDH36usMCfWBQIC3Z6D3MAXV/gcyHj8Dp49ZgdklkuDSTsE3meLC47AWW5ABzLrdQumoQ10ygH6Zn6lnC6Tn95hMOYsGrpuoOiMXF4bWjnDZ7rx17zwy9fpASRo7LYOHCu/AXgCJgxYbf/zXj48ll4CqBdsm98ZaCGb2E/fL7OJCSiYk69Frz26ZRAHDw2O+K1bsrIlj3G40gStnPhJejnRxzjxj8ipKzZ+OREE/Qi+rNfwO9sQ+U1PjgzgCCqtv1XzBCrPq+Ec727odmFXA7GxTqyxK5zBE/wjKLCvtR1nnLR/OokJwHd7FuOBXraV2FCuR4UPfOPHUiAfzAOe5DbsXkfC/Qgm4Tn3Zw0d5ztRB+bBB1M5mU6ndXp62mTTujadTuuHH36o//zP/6zHx8f613/914XWs0xcwPcMnqzHq6urbWnD1tZWC1TBWgRYBFPgrqurq9ZK6WdBMQ/uiw/wZlhOBto3GhAT5MATJ01cdbGuWj+Hw2ELiFkrja45MIFn4DUHe/5/f3+/9vf36+zsrJYdxhumMfaFlsy0o70EWv51UnU6nTb8gP233GI/kG0vMTFetdxnccW2OGliWaqqhUDOu3zDG2MV3w/dur29bXsZoO8935I+MenLfDIJ99LxYjDlnTIMfNJp2AhzHsqBgqyuzreQZscW9/SiICsrKw10zGbz3cUg5Gz2tNvKmzdv6mc/+1m9fv26fQ8BMTqOeNfW1mp3d7dGo1FdXV21HWIoQQ+Hwzo4OKibm5vW0och/+Uvf1nHx8dNUWklm81mLaq/vb1dWPjoCNdjsVHCWZiBZKFZcAnTp9Np6ze20FoACEr5Cz2gP0LD/a2UOOnc0h4nAnjhwBk5cwitrTgpyMyJ8XntE3/JIFKF8too5Auack0DFJQS2cksncdk54DRJRPDnAiE6O/m9zgNBw63t7f1D//wD3VxcdHuR1tiAlqPxWPwe3jmeWLwvIkKRj4DZQ4HC1mdQY/tFHhINfxA/xLsebxZITCfGI/ll7EkoIMu/l0Gb1WL7RgJunuyB134rQMVV1j5jQP0rzl6CY7cBt08cjXdc3fAZBmAXtjJdA7PGX2DjqpaAIQOulNOem0pzh4nYPNckg/PjW2ZHvh76EIyw2N1sJ1Buudvp2kAx7y4F2DM1yXoYv621wBa6OLAhkDYIMMybb77RRKQ8bFFtvXPQVrVPKhx26A3f6iaP6vPNtk8TpnLhAKbvqROHh8fN7qRZDQwoSpmHwPghXZUfdjMKQG9gbZ1Av4QoOBDvKkEtprfsvkBVfeUpfQn1u3Pnz/X8fFxeyaWE6okH9xa5nZ0EqiAc9blwMfHx8eml/DdwatlGBmhokzym3tnkvrx8enZQn/xF39R6+vrdXZ2trCe2jQwLfifz0mssokEO9zSvkdlir9gF7fL0y2EzsJ3dCbtoOdjfjkIwp7bHzlwcLLZNpT5sG7dnTy0gpIccFBrHbGOW4/wqZubm/Xu3bv6r//6r3ruSHvJGMGWVfOOsQxe3GXiYILfO3hBDpkLtsbdVMb32BT7cDCcbVZ2LKQ9t0217vKAZT+SBV5m1RJMb7q4YtWrzJqWmVQzD0l2/P8cL25AgbJVzQXWRMnKFL+rWty2eTx+WmTqjR8YPEbAgNMBAq1fdpCj0ah+53d+p221ifA7e85YYATPamJ3GYN0Di88daXJwolxNMAlAGQnGkAUW3070INOua7LQScKjWGBhuvr6629xAvvTC8HGRzOcFvhqDIR1JlmGDz62V19oSrFk+UJrBPUMQ6MIs7fDpygbHV1ta6vr2tra6tVQm1AHbi7JM+cqZ7xG7JKBL4Z5GVGwtfjmR48Zw2eEshPp9N6+/ZtTafT9sDD8/Pzury8rMvLywWZqqrWYvA1R4LrBNTQ2muzMCgZgBgQJGCiYmqHw+8dqGXlA17aYDIGn4Oc9eaAgTRPc7wcaawtSyQzsB3MMYNlrmN6wWsCRz4HjCZN4McyPlquAFvYQJJKDoKdhTcQcNBjUMd4l42hF1AtC7J6gQsy5Ptk0J38tT3zdU2H3nhTB/1/BsB+T3UG+vbm2bsX9EOOkBvbxazMbmxsND7ZPuPkuT/gwYuxndzIpIYD7t74s3pF6xm+xUESmVl+Z3BguXIQCohgXq6qZRbW44ZH0AIw6vUiv/71rxvAoYsFmQKAEgAAaAmkeMaUg2RoyO8NJnOc0NzfUxXitySFh8OnxKnXZxhDOIh20sivu7u75hu5N0sZAJXYWABe0tJtvgRV4C4O2+3Uwel0vhkJ8wD4O3CzvHG/s7OzL/QlgaX/ImPQajQatQ04RqNRHR0dteSiN4zwboxUJb2bHCDYHQvwzt0HyCJ0RaZto+1LsG0kOL0VuIMFziewWF1dbbtS2v6Ah5hL1fz5S8h/yrs/B7O9f/++njusO8ZU3C8D/eQP+gDv+c7JAHAR10HeBoPBwnb+Tjj68UXMyzbLxZC074mflx20xrp10N0Gd3d3TZ8caDvpQbcUONHJXWIDV64SExLT2Fd8zfFVu/nlznJWOCaLA/J5lA7Z9tR9om4fcVSZTo0sE0SAcEdHRy0T4l2cGLPbqrxuaH19vQ4PD1srGd/DJJQYRjniZ2xeEMn7qnnb2+Xl5cJOM46uLWDZ8131JKCMxa1XfDceP+0weHNzsxAw4uR5b1440MLZI7AAVcZpOk6n09ZqgRHkc1e+7MQzkPL3CZLML+hFoOrdbNjAwE4fxYWeZBKgA8ETawsyGE+AadCB4Xt4eKi9vb2WFcSoz2az+vjxY/3www8t23VyctK2/Sez44MKm+fdOxJQ+n9/BsiznDJu5uGsroEBsuegwfLMvZiLgdoyI75M7jB66JaDa2fGkW++T4CLnJIUcUaKl52feexx+nMbUt/D1ak0/ssCiHQe2MWsQhl4mXYOXBibjbllk/OXAXKfaxrw3kDbNoGDVhUH2vDSQBdbaR+wTK5TfjnsyNIm8ILu8IiqlGlo2erd1zaJz9AP64/lL6s4rmDZ3sJvwCK+zAEcQDsTN71589eBCg9BHQ6Hbd0vv4P2VV8mzciIcz3TEjvujTE8JycQLEecQ3XKLdCnp6c1GDxVgA4ODhqIsz1wcElVii3G6YDA32f7nO1Q6q7thVvMAIf81uvgWDJAK5q7ZJh7AjESnSTbWNfL79Ep+GxdSnlkXN4EimQd8kTw0EtqmE5c8/Lysr3PpB6ymzacI/EI12cTBqp4h4eH9e7du9rb22tjdMLQANcB+mAwD0KxjQBog2RjH+YJZnRCAPxo2kIncNerV6/q5ORkoZJMEtlJBHAj3U/YGK/1Izh2dcbzs322flp3Pnz48IWu+RgOh21No+UeuwIP3U3CPY25nABiHFR+XVVLm2l/7J0m3SprTGtbhQ6Yrvzesp9+yjbRiQN+T2stmJ2AlwPZMHZz0G15BsuCS6BxYqX0Ly8dLwZTuVOQD26KYbLhQVGurq7q5OSkBUp+4KCDHDJsMIaMzWAw370PRaa9AIHC8RkoOAhCKB8fH1vGZn19vRlvtpW1QmZ07aCJ78iaUOrmXpTRMZAwzg4A55zgyKCFazM2ql3s6Fc1d4xejGxFToeIsBLYTqfT5sB4fgbBFiV31ggNBoM2V5xSz8gbMCRfekFCAlTu8/j4WBcXF21xsJU2W1xcRbQyIksEW8uCFK4FrTim0/kDSqmuslbq6uqqKTwLbGmx8H0w0i9lZbhflpd7vLSRMzBDTrhOBuwGwgQzXlhaVQ30Vc3X6FnWeJ+BU8pZZmsNXA1YLbeWE5wp+kMwBb+caMh79/jrc8xvwAJyaDpnUOLrOjjxZ2mMPR47Ec73X66DsyNYyaCG/9Pg98a57MCGmVfw2c4NZ+PkkBMxBObIlgOCniNKHiWflv2G++GXer/n6PHM7xmrE0T+LUDKXRTYGIN565SflwKfkrdOCObYoLltGbygNY4dwpzcSZ3mfg6UWU9gYJG+IuneA9wZlFtfSASen5/Xzc3NgmxZhjxebMrV1dVCpQw770w8+MBg07SF7u50oBPl+Ph4wR8BmMEVPG8qfRIgDZ/kjoC9vb16//593d3d1Ww2ax0jJP8AbrbB0JjgyvhkMpm05QU8nJdKFeNNWQbU4wf9mJPUCQ7LlW2I9Zfrbm5u1vb2do1Go1aJIog8ODholb/hcLiwmYgrWASfAHHWM7PJA1jEPt1ro2iHzE4iKtS2ESm38On+/r7Ozs7a3EnYVlXjn4Np5s9B0IecMTc2UUn9YBAlCUoAACAASURBVDxZRZ1MJvXhw4dGt97h4MK+wlUi3kMj+yH0wHRBV1wph2dcF0yaHWXgdGT07u6uyYBtge1aBlM9e855zAH/zq7HBJBem06SArvHWj2uyed0hdiv92xd2rz0Azne545ng6npdLrQZuYLm+kuYTtDxtojO9409CgOWzLiyNz3SWDGOiICKpiIw8iSL4edBsbQWWIDS/a2d4UBQUGZMyK3YQOITiaThe3cmQf0s1PJA/peX1/X7u5u226dagRrvaqegKAzOhlIItA9gM6uSS7/4qQQRubMvLkPhswCbGOcoNqHM7bISm5AQnaBIJx5ovDQAbDvuU+n823nkTNKv84IGSgwD2QEJzqdTmt/f7/Oz8/r48ePDcxwbx6kuLa2Vm/evKn7+/u6urpa2PTCGZgeyM73zhzyvYGpaW0HbVCIXnDPNOoex2w2fzAmsoK+enE08p2y5Ot6TPwOR+idH3Gy7tu2o3DVLDNHGYg4E5Z8zYMxpby6jfAlo2s6+762bb1g57kx+Vzzrpcs8m+WHRnw5RgS3GU7CfOw3Tbo40UFPcfTcz7Lxrss8DLdHVh7+3L/1nKXR2Zg00byO2wslfDkn8eIfckMqNubDIacZc75OYDFLjo4Go1GNRwOF4Cnn6FHssiP1LCtINDBtkETJxBTxvM7B4gen+UFu01m2HbGY7H/pLrv6/AbJ7KcmCSI555pc2jtIwBCxv2MI2TZW6EzLydu+L3XQ5E8Ozg4aA9198YUBFIEWawL8rN08vrQFt3/5ptvajicr+uy72KcDijAa5xju2mdsF+wHXM1D/vMkgjWOsEPeMtacesP9CGIglasIauqlpgk6KY6iq/20g9wBzyazWbNJln/+etKMz7n3bt3bW07ugZPCAaRB4IodJtqJIEdj2ehQ8nVYPMmecBnBwcHtbm5uTSYcuJ+mW/39dPG+DzrhavClglabB2Iu1XZQQzXY7MI+GOfnJgK+Ur6GKPDewfyr169ajtmVlVLUHCQ9L+7u1vYGITqIR1JLnTwO5LbaSd7/tW+9Lnjxa3RnV1Lp8/nGHM7FB6WZhBgRjPZh4eHtuCMoItIOA0ZOxARFbvcy7UNrCACuwRlVjt3aPO8MO4wvdf3jEJZsFF6gj2v1+EcGGoamnGMj5KsgwBovbe314yG6WWB50gnyf9kGarmu80hvI+Pj7W7u1u7u7st48rzE+hzZwt3Mno24jZwppfHVFULATBOyG2NfE5gx3UIwjkHBXSpl52cMJg2lB6HQblbHl+9elU7Ozv16dOnhXshp6PRaKFyC093d3eb0Qa8cN1eVSHHw9wdqBt0YehMD36DfFhXM4OLseReXA/nx6JhB7hJX4PfBJnI9GAwaEbO67um06d1Zl5PMBgsLib1NXwf5g1tEoTDowTVCURMf2S5ByjzZT5lUG6ja3tnXjGWDK495xxjzvEl3e45YB8OiEhO3N7etjmQUEIXkcVs1+U6XAPQ81KwvWysy16mGTaVa2SVp3f4GulX0FEy5gCKbIFOXWKeVMmpEgHgc42fAz63J3sO6BnvCY4sywRQ6Dx056/BnW2jdZBEUE9uHNQlTXt+xUlB2xyDsqxGkxzlvXePTJvi3YMTi7DbbLacExhQSRmPx3V4eNiAOy9AsoMabFCCQ+ibOg0AZCmDsRC+h/mtrq62oAJZSrvG/O/v7+v29rZtrsVutbYr2Ezo4i3azcu0AegsfKbKNBqNWmDplmqeHYYcEbBzGEzzO1oC0QuuWfWEtU5OTur+/v6LTbwIMNAF5kxr72Ty9KBkaJAJAY60QazxOj09bfiFJBD4AjpRkSPAA+M5AKY6QzABBrNupN9lXjs7O7W/v7+wQZUPkrX87+AnE+Y9vUQfsiPKtgv60YFEkG/bYbwONnLS/vr6uiW1bS+es/WZXLF+gZUY683NTV1fXy9UX3kZCzEH9M1LPogZkjfoQOogxzKs8Nzx4m5+PUDmGycYZSBkK6rmW6KSHYN5BFJ8Px6PF4g2nU7b7nu0loxGo3r//n17yCbj4UjhQylRZoIxnBQtZAQuVdWAhQMYC7IzCnxGvzC0QQAI0jLCtyD13qMgbPLgBbt3d3d1eHhYGxsbC8/fsOJiiOxgEsQxFhsksmvwwNkXeEKvPKCAPlYCLt87s5tWbIOyqvlOMmx1yXUYixWPOVZVqyLu7u4uVIQI2JEpHCLfG6xDO3rnV1dX6+DgoC4uLur09LRl2KAbrY8HBwctIGXTjL29vRYs4KChu4Op5xQ1x8Y5gL8EJ16bw2ecbyPX01s+h+4OKqk8kLjIAI+xJsBzpZpkhgNxsny94CfXLbi0n6DLz11DVnpHBjwYZO7Xu3bv5WukM0ueZlCFDBrM8LtMAPl/7Jbpj2O3nj0XjPj/pAvXIBucD/r0/9lWivOieuC2zLzvcwFPb1x5wF8CHQcAHEnTnL/HhcPFTrilLEEQ13LigXUiAAzsYa7pgCZph7O6lnM3DdBrEhrYIMvRdDptu9LxwHJ0zQe/I6Ay8F7Gq/RLtkuWe+ZPYLq7u9uCkJwLeMDJN9NqY2Pji64Fg3jvkOc1ma7mrK2t1d7eXo1Gozo5OWm4ZGtrq/b395tv7dlk66IrTwQ6fjwHgJzvZ7N55YOgcGdnZyFpDC8Asm5Dn81mbTdDEpZOCEEry7Bxif1s2kTO397ersPDwzo6OqrDw8OFHfds6+wz19bWGt3AZOgk16ZVC79H0DqbPXX8nJyc1HQ6bYm229vburm5WWjts78EDDMu8I5l1nKcbW/oL9uSf/r0qcmMNwmBZoPBfAkDrYvT6fSL51m6Wmo9MO71NY0x3rx5U7/61a9q2ZF20/7fMpr6yf8ZnKdMMy9sgVv8nPTkeqYxOort+fz5c2vN8xjSBzEX+0Pk9uLioo6Pj9syHF4O3gjSExORsCBZgp4gK4lP/HliUI+zN5fnjhcrU84Ep3LmYJg0TOH84fCp7H52drbwlHkDqeFw2LIZjipxWGxZfnR01IwziuNME4Jt8M34Nzc3m/Em48X3R0dHzaD/z//8TyspOhL2OhXG59Ywgj0qNVzbbXTOpOX3yxSC6pRBLyXWbLuo+nIdgIEtY7ChwigT5LoaCOglwJxMJm0x9MHBwQLY5nsrnQU4ASZjRI5YvL22ttYyilXVKpyWJxwYgRVOigWxVU+GhxIwThk6poLwnsoMxpIWgJubmxqNRgvBCnPHkbOmgcwW2TRnIuGHMz/O3DIW/7VRdUYO2aASlb9JUMhvkBEHKegRCQDW+93d3TUZwUghXwarGCz+N4+RJbcFkIm0jKRNgCZ5Hx/M5znA7t/2AIbXAabh7wUmpifzhpauXjMHZIXMn3fx9HUAy96MBZl0ttiJhgToOVaunfMhGHOg5mo559mx9jLAPi+zmcto2Pt9npv3Ya6uoKWNgcd5/bw248MvkTyyvNmm2W5WzXfwchsZVSpf29Vo093j9meMDX3BrqyszJ97Zx+L3vte1nGDYrcWAoDG43FLAhqA9PgEbXsBP4v5q6rZkNnsabe4169fLySQ/FsqosyRTR0A3egT/jIBEr/FHpEM8NoX1k2srq7W3t5es/+j0ah1XmCrnWw0rdBNbK35i/5Q3WbMVfNNgvD5/A45MS7B3nq9Kve0zvvwbr7gGl7YDwc6+dvvvvuu3rx5UwcHB21XPmiBLLrSNRjMHy9DoIvsMmfm4OQB79nMCv1ibbm3wOflzS542C/0c5LANpSxpM/jL3NeW1trrXpu+USWfG1jRDYG41E65qGDC9MjD3Tx6Oio+33VfLMH090Jfa5jnfSR/jPPI4DvtSfCQ1/TyRi34LGO3fYobUgvSVQ1T7iOx0/boX/69KlOT09bFZcqlfFRVkDxCejnw8NDW2JhP2R/YX32fHOMnNfzecuOr3porx2NB2RBtdNlQwcmdHl5WWdnZ60s6EpVVTXHBWHu7u5qZWVl4fkEbLc5Go0aURFi2s28g58DCoPwzBKSNedztp/FUDEmV0jIoJDNnc3mi15R9IODg5pMFne/QoicAfXLnxlw39zcfFHm964mZrQjcwtF74XxYFzOZkBXwDXXRQYAFKncOHxn0Py9hdNBAHNzLyv8IVvioBQFRuZWVlZaFcoOq6pa76yNXU85MK4YZnYcms3mVS6ciNtYNjY26vz8vPWGMx8eQEyQaOOcgTN/E/zZMdjos94tgwO3JSH/adC4bq/tlRYl73RJEOprGRSm0cyqpCtNzMFtSdynd600zEkTO67kazqTzLjxOTzP+/WAuAOoNMIGf/T9uxKN7jMG05NrQjv0rlc1cxDFmDKoSMBrued823BAps/NdqwcYwZv3DOdcZ7T079lzorP3WKUR2+Ovp7nDd/42wv8DLK2trYWnk0D/bF/rMX07mCMyS8AkrPm5hf39+97so+u83LCZ3V1tbXMcy58Rs4N0jwnxp5AI+lsuwUtHMi5gg1QZhOG1G0SKlXVOk1I3GF3GI+7JUwHZAM7QncJAB68wYF9I8ByJwI8Me25rqvK3h3QmyxwP7ePuYKIb3Q1fTAYNOzCufANEOs1WLZJbmF3wMmcTacE3BsbG/Xtt9/W/v5+e6CuK1Pu+rG/M4CF77Y9+Mu0EWAhEptXV1ftMTnIjm2Tgxw/A4t701kEXay/HMZ7/mxtba2t5cef0o4IH6sWK8y+NjaeAMJ+zUm1TD54DN98882CL/OBfHIudGWOltPndDODF7C5lwG4g8UVZCcwwb7Qk+tT+HCyNfGF59wbK3ifBwdnxQh9t291WyF6gE0ej5/2aqDqadk3TXqBpnUrk2vL/NMC3146wZkN+hPNbBYm2mhTPmRnHx7Qa2PoB5/xOc4A58DmC6urq3V2dlbr6+utKgJhMMIwE6E2mEvhQEHZSYZMCIAXwcJxY8QA6DDXvakGuGyPvr6+3pQOA5stCw567Cj93psw2Ll4txLmblCRQpvBm7N3GEGCBWcYuTd0pGfZztRGHHCKMedwBYbvPH/oboDn3m7umVkaDAsvwIRbYixnCQJtqJ0h+/HHHxuoJKi1s+G6q6tPvfDX19cNYPGZZdIg08Y2lRv59t8EU8hOVjWZC44WY2C5RQd8X3rjqYqwHT6Ljm28kSWDS8Zqw2XAytjRKQwnv095NfjtBQ3pqB2s98A04/bcOVwN5z6mpemU3zOvBC4OngCy/NbBH/PxJjKuvCPHzsT1gpmenUvnkVlD20MyrKxDgUaMn2vbHxh8p3zn/Z8LlvK7HCN0o2puG29b99yRY8HWmw+uTnJdMs7wx4k4ACI7rvl6AJVskTXPLPfwaJm8+eW5AxicZOGzyWS+0Yuzyx4POtijO+Pgr/XH8seRujmZTFqSiXMBWIPBoK2RWV192uaYijjjB2g7KHTlFxCND8Knst4XH07Ahp6zHgY9cxv2Mh2rmq/zxicjD15CgB6QRCUJhcy5okggBl3oaMBGug0OG42+ITvQyUlYd2xAszxms1kdHBzU1tZWbW1t1d7eXm1vbzf/m/YFn2P9QF793CHbauw1FWW6TgaDQXs4K79BRpExsBR6ibxCa2jkIDhtX9U8oLJ8EwBcX1+35DkVObAs9DNvvE191XxDBILhxFfWG9NsMBjUhw8f2n1eslvWb88teeHf2I4wHq91M294T6DizomqaphgOBw2LAA/qGgaty2zxSkX9kckRJivE7DYfs8Fm2t6gFHRCdb+Ju1sZ6FRjhN5Np1fOl5s83MvrgE+2X52snGbF8+M+PHHH9uuH1W1oDBWercvYDSYACVJCIhQw9yM0JNAdg4uURIM0mrkrasNyP1sEIyvhZUDI0kwcnl52e7pneX8MjCxY+y9Li8vW78ogYtBJ8Kd2VaEIjOmBoXO3qVQAw4TyE6n0zaeyWTS6OSgNo2baWY6WiYy4CBLiezhYG0EEnQS+BAcIEfMh898GAAxLldPGfPNzU3rB3eGant7u5XOkRt6rN0y0JNRy6rHk2AK3tnYOViB5jYggPrBYPG5OvAQB+Gqmp/pBeiwkzXYzPlkQGSZ9ligK59nltzXI2GBjLtqamfSo6N5y33T+QD2PKcemLfeuAKbugIvrPPQHjuD80KWCWKceWe+BLTMmWsb7GTQaUe7TKZ8jsEk87Lz4je2PdZjrzvh79c4oZeCIA4Aao/fzzm95Cf39HoHr9OdzWZt98kMktELQDltZQ6MuIeTSlmZdaLPSUjTAn9jG4E8wSe+Z/xsemOgzTq2qsVEAz4P+rl1N3nX062UIX/mJAe7hEFn8wu/ThKV75Ahrgl4Reb98FA6F3h4Ljvd0opVVe25j58/f66Dg4MGyq33Th7Ac8s8rV/eiMjBNZtE+Pck2nZ3d6uqWgIU/Yc+thdUy6rmwaKDK2hEJYvkszEFfLf9TT2j+2c0Gi20xltOEzs5eHVC0wkHEs8Onh1cEiQRODI+88EtY7TdO5jje1f3jC1S5znPIHp1dbX5bLdLXl1dLdDClVx45C4qAhTkw9Uu+xrrz2w2q6Ojo5YszgOZYvxZrUkfyd+ct3nu6iK6g7x4e3SPk6AdHbUtwL7RIm/5TTtsOaqaB3LYNPSVpR2TyeSLJSzQlXiEjSXQLcYN3ZAzZKxHN2hkW+Pvenrz3PFiMOXsatVTxuDm5qZOT09rb2+vPZgPA0Df4m9+85u2KxxCw0J9CIIBoFTK4J1JQ2ANPsxAE9LAILOOVfOF3IDf0WjUHiCX0TbjtVCj6AgkWSkyEza4jGt1dXXhSesGl6atQaWNCudSBiUrg5OkCuGe3eRhj68YIBbVQh+MIPfGWUNL7mWgncpetdjOlwAHZwJtOY+/GM7eMxAYF+MloEOxuQYZSa9fM72tOIzXFQPz3Mbw7u6u3csZwOHwqZJ6fn7esmm0ZjJnz7MHoHwkWIGeBG82Hr6+M2JJf5fxeU9yArmiD5prARwZg9tKPT4+M738G+4FzeA/+mhj7qAAwGLnYNBmMGr6mo4cGbxyP5xnjwf5nnFYXvPevo/HZsA+m81qe3u7tZ8SGJ6fny+sIcC2YO8SkPfAvGXHTsNjQnYSAMPX5AW2kWtgq0g6sL419cWBwzIn+9zB997kJvWT+yy7ZiYB8DsGgoAB5BA9c7u3F8BDP9ZyVNUX1WDrVmbqU7dThsnOmgZuJefFM+0eHh6+WHfCOYBuVwBcqTJoSX3msE4vOyflCJuLj8mAk8rZzc1N829+VAJ+EdvkQB2AnL57a2urdnZ2FrbRpqK0vb1dd3d3tbq62jYO6VWkGWPqB4G05S2z93SmwENAH/iAVkBX5ZE3+EulznRC/ly9hl7WOQdVtjt5kER2VdX+3xUdf8ZhGcouizyH4Pfz5891fn5e4/G40QLfDt0ILrxeEcANj524tc2yDFpuoYnHh34SXLMT43A4bOu6HLzCF3Sc+WH/sJW2Q/zWcouM7e7u1t7eXuu08kE1G3vlTVesPz3skMUFB6eWOeSI8dr/Ut3FDyB/8JIKac8X+cV4l73PgJuxgevcYTSbzdqafmN89JndravmwShY3YG0aWO+8N6YD9px/5eOrwqmAAEo7+XlZd3c3NTKykr9+te/rp/85Cftga93d3d1fHxc19fXbWJ2DI+Pj+2J14AqDBEGaDx+2sqUIMpCRo8za6gyivf5EI8MIcZ7MpksZDuur6/r7OysBTyMG8PJ/Ily3SfLd8PhfAMNhIM2x/v7+9rZ2fkiaIK29LkbzJv+vE5OTur169cLwo/DRhiXMT0BH/RyRQo6whcyEz0AzTUwOBgUQJXPNW8YJ99z73TWKCl8oE2UTR74vTcpwcASeCOvBFwZiHBAN9MfZQcE+DfX19c1Go1aIMBvt7e3WzsDDrQHHrnfsoDKBjNlmgyfjVFe04bL10deOEgGAAjRN+7tdhiDL/MzAZTHi6G0/EE/xsPv3cKCTDrzj45ZNriPnaplqBc0J90JGJNPOeYeD5PGaah9wBcCDwMIdJo5GaBh5xKAewwGO/l9jtMBCZ9jB+3w7dzyfICdbQM6QMbYYC5lIm3QsvEakNBe7jHwnXnd40ECZeTOINp2w0DemWM749ls1h5oyu/xM87+Ju8MAlN3epls+y9nirF7tOVcX1+3alkCCO4Jf2hlsj5Y7tOmeP49fXDQl3aBuTroNUhfXV1tmMIBE4eBOvzzXLALYIOdnZ3mI1i/BP6ApyQnp9Np41/aDNsAV/zdzue2qOn0aTdZruV1TMjUYDBoGyoQcHEPQOLe3l5LPtHyZyBPZ8V0Ol1IuiA/YAsvIehVP1jXnnpiO8JnXNc+mSQE1we/cT3bDq5DctTJA+ygfSY0r3qqxu3u7rZuEHiUnRLIUy+g5z3zsE3g+gRyaUsYC2OlykoLvGnk+1o/+C00JEA4PDysX//611/wxuMDI5qOvesv+7yqFjaqSdq4OsXaUGQReYBvbOyCTLKDH/KZQWzauAxe4BnBE11sYGzfv6paoQP5ofrENUj0c28nx3r+uGpuc3t+3OP9PwdTFgTKybe3t3VxcdHA6fn5ecv0DIfDury8rNPT0wVlsdLd3t62XVGqqjHSbXs8pHZ7e7vtnrK/v9/25ydoccY2hcmKY4Bpw/v4+Nie7E3pHANtYmKY6V1GGaxsCTzcw319fV37+/vNOWNcAcZsv+1nlTBeAoCHh4c6OztrFQ9nQyy8PUCBosPTBIsIvatbGaBaUVBu998SrOaC31T2ntJzDei6ubnZnAFb2KL0W1tbdX193ZwhRvfi4qJWV1drf39/YUEvgJVA2UbNY3BbD2N69epVnZ+ffwFI2YaTLfZdZd3f319oQ7D82Yn0XqaHAQ20cVUuee7f4BRIUtiBWM9oDaICy/gIDOmFzpYy8zD/T/mDt2RSDUYwlB6fd+JhTM6aMm8ba4P13pHBE5/xv0F6ymdPVn1OAj1/jsxmJZpAH1rv7u4uZMgTOFqHzAfG4mAgZacnK/4tLU/5IFx0h6BgGY+5Ri+AN197r6Rtj/Z85jZQwI0zxL258r+vlXSAdthqDrdbOpmCPpCE8LMKMzPqtR6Z9OvprWnLva3PyCk8oyozGAwWtvimwozfAmig3wRinP9cG1HyIX1Lygf05J5ePwvv3FI4Ho/r9PR04XEknOuKgv0pvDFPtra2FuiXyUXkmS4ZkmQkb1ZWVho4dnsmOgrdsgph38FnOzs7bVzcly2f2ZmXgLBqnhAgyCMhy5gA7g4gSEgzV+bEQ0wdmPbalabTaR0fH9fv//7vN5lCh7OLAl6n7+Q888RywHtkEn+Oz2QtvW08/gkfbl1CDwD0JLAti6lbthVpq+EzQeqPP/64kDx1V4CTfCRd7Hfy/r4P772h0Gz2tGatd3AO/jGvk3xJffT3jDsD5Kr59ug3NzcLSXtXuFkjji4im4zFeAKZ7gVNiRN8DjTEZqELzAmsAi7NThdk8PLysumMEzmcY/o9Fxz5O7oTnsMFHF+9NToBFRtKuLJDq9xkMqnT09OFrBGBwOXlZV1dXS30bSaxUSieK0XEuru7W0dHR1/sOJMLszP75+tW1UIWCYZRhud+9Frv7OzU7e1ta1XkfEf7Bqswbnt7uzkXAgOExNE09+aZFdCP8jcKB714PtfJyckXz9gysGH+jq7zMJjGMduY9vqErYBUOMwzZ3+4v7M1KYxWNsbC+dBtMpm0kjttGbQnMEbohKM6Ozurw8PDhfUAd3d3rTKYmQiPx1Uzr39Lus5ms7ZLn3uPx+Nxe5Af6+TggY1ulv6tAz3Ax70JvhP8ZFBmY24+8Vt47CCb73AeOLIEfFmpSNnjc8sA4ycBsba2tqAPBkVed8bvHIh6rKxJcKLEWe2erFlfeG+HnHNxRcPn+F5c30GTgydn1x10cLBLpO2EqyLom/WMw/RxQNm7D87IY3F21puCuM2GLDggJyuDpqkDu2XBwnOvDNrttPkeYG1H7vv0DuuFbQCVA/MmASgZWbfJUDHkGq4GeI1br3MiM+h83suSGtQ7sQGgvLy8XHjWIMEN1ydAtA4wZ4II66llpqfjOb60X4wN/+wF+/bVyBXjt8yyvqlqMUDl2rZbw+GwBZJVTw9MZ1xOdDkJNRwO23bbXM98ykDAOAi/jc++urqq09PTNiZaCtl5mN0MAeyskbKdTKxydXVVOzs7LVjHpzA2qge2K2CVh4eHOj09rcfHx5bF71WmZrNZnZycNHlxUtrzTf5DEwcd1lProYNVAniCS6ogbrMlOIIeuSMq4D6D7mXAmDEwB3TCVVLeb2xs1NHRURuX/Qg+yziT3+VmDWn3GDf05LvJZFJv3rzpjntjY6MF/bS6MY7n7FvvyBZP7Obl5WVdXFw0HcEGggdJrGWLOd1atK3apkBj2y3Tw7TgL//TceYkg/H57e3tAo7n995gB7zItTy2HN9LAZXp18MUvePFYOr6+rrtzvfw8NB2LLMwe5EhT7bmmQxse0g1ygLh3kbaBOl3ps8ao8dzIfhNgnE7rMzwJniwEcA4oeSAPhz2xcXFF1F31byagQElomaNmJ9nhOPCsCDEPHOLLNPV1VWrwCEsZMNOT09bJuH+/r61FJgGpq2zZik0BiCAE4yDgyocOLTF+AGyvUMZRttAmGtULT6HjPv68HXILhA8oVysL8GQGwRhJHDQ+/v7bZzQHaVy5o3xJJ38fChAm9sA0Q2qZt5db3d3tz1AzrxIUJ/BlPmTgLKqFuaQtGMOXhOGDPNbVx5wAl73xnWo+mbAnrKVdsB85N6MF4fJgYPieRWcw+8tp8ieaejg0JWJDEQN+JLeCRJMzx5A8HfJx2wnwJETlKBbzAOdW1t7euaJNw8ADBOspzxYT3rBVNKAz9ATZxbJ8FU9AVfW1jD+wWDQEj0OHnFitNC6srmMzvl/jrH3Od9lpdcJhQT2L/GT3wFWsP8GAVQ6cPQZdGAHeY8NMN9TZnpz6wVUyL5lzLxzIIgfSRlhzJt6SgAAIABJREFUFzmSYclT/DVjtU30vU3DBHPZfmS57CU7HVwOh8OGERJscZ/czGBlZaUFTlzTCR/A3urqagss2GSiqtoidzZdICnrdW0JAOGJ7a6DbmSd5/FBO6pDrLmgm4QH+Xp9kOkGXa+ururNmzetM4POH4JEV6Xwk7/61a/q5uamrq6uajJ5enYma6J6fubi4qLRzfprXbVOmxe8N/99MCfwDz4GjEBViCUTnIseci4yPJvNFtYgG8+ljllmGYt/Z15yHT/64+bmpo2Z37qNzPKBTQeTeN7cizk7Cfj4+Fjv3r1bCB6SfiRJrPOMGfnzkedwf96DmbyTIn7YCQr/Fhz0+vXrJr+umqJ/6HzPN6X82Y/wP4UT7ILXsZJoBItdXV013WXMdLwMh/NnBxL49WQifYZpyNisX71z8/jqDSjG43HbmYeIlBeKUlVtJ7/Hx8c6Ozv7wllxTRiPAeSaOG8WKaLAZHvctlS1GCGnM02HxPUJnNge/eLior7//vumrGR0YCbtSfwWQtsIkAGj99TbLRM0MY+Li4u6urpqAabL8+fn5/XNN980Z04G7Pj4uCne+fl5HR4eLoBIG5JeBsY8tUDZeRgUGdC6aoNhAwjCz6pqDtOZNjvhZYCX60Ivjq2trdbeyK5+lhl64zHuKAGbpNA/v7o6f8K5M0RWHF7pPLa2ttoT0A1GquYPv6OF0NvVb21ttbZQ5mw5zPccPdDlLBjnI4dkxAF4VBkZJy2lzBPeIY/8Hl6zRSzX8JH8XHYkIPP5Bl2AHfjnVi6fm3Q3DR2c92htuV9mPHsOuQfoDS6gJ0FqBlLYD7Le/s7tVV7bgINCh7KywWGHZV5bz3vgIsES8u7k1Pb2dlVVs1mAPrdCovMJoD0GO3KPI+1V0jl5AG29Y5ZBfE8WfOS9GROJvsx8E4gwDhIQa2trC1VUV/EIfL1WFH5msJvjSZ45KWbQnoko2wT+ms/IkX/DeLDhyN+yNs7n6Goe+3sSKdnm5+CKjg62pSYTjm8fDuedCD4cvFi2mJfXFQFCHVxubW3V69evazQatey/d8lkfP4d93CgDQBFJ/b392ttba0uLi5qPB63LbdJwtGx4I6Tzc3NBVzgFj5jis3NzRb4OZBCL2xHTk5O6ubmpn338ePHevv27cJDlX2cn58vrP+Gp/DROoYN8GcG4Ekr5Bf9Ach7vtARWXSw5eDNwYyBP2DeurPMlqAvqVdOyiA7dDCxL4D12skAAPxgMGjPuHTyy3gXnIcNYl+AV69eta37Obgf1RgXB0zjnh/ms17QSGWUsVTNA2kqYdPpdCGJDEb0I1IIcJ0syQAqfZZ9hXENLz9L1jjwxx9/XFgHCs6mc4iNJ8A/JCC9JGJZ0jrjBT7LlxM+zx0vPmcKxhDREqkmeGCCPIjvhx9+qLu7u8YMDrJLKJrbYlgPg/ExYQDsGH4iV37P5wmwsryfGVSMyWw2a8D8+vq6dnZ2FtYyuP3FDo+xVFW7Fko5GAwaoGeNDXO2EYRm7Gr08PBQe3t7NRwO6+TkpK6vr1tfeFU1GnkzAfNrMJhvg22lg2cJghwUQycOZ6SZL8YNuXDga+UyALACZXCFceSayAyZ752dnbZYlvMJYpwZYuyM6/b2dsHpAZyYk+/fc9KTyaRGo1FdXV21aye4vr6+bg7ZiYfRaNQyf5zLPV4CKj2ngBHxmjvaQmktTX7aSSHrrFvEKHKes245Pju1BOn5Pn/LmLkXdCJJYQeHATcwd0XAjtcB+3OGMoNWO6JlRvY5h2zZzWy+nVjSxpU37AP65d+R/cPBZDbU4CId2LJgyvYwga/BEHqCs8yAmmtmwiUzrv6Naew59IK95IEP7Bk0sl4/F0j55SCY7KeTJNhTyyQA0gGxbU3vXm6zs93LlvQMrrB7mXG2v3L/PxUPuhuwWVyLwCDvWVUL5zlI9DyW6YfH6+SmAymv00JekdnV1dXWjUEgyrpNwCk2zzKFX8N+cW1XbSaTxUdfePwOqkgAm7bWW7cPIw8O9sAmdFAAptFnqkUEilVz7IJM4Te8kYZpSIsT1Ti6g0hwGEOwjOL6+rrZKOj75s2bLh9J6PZAuQ/7CegDbQ30bSdtn1mfTpt6VS3wk7l47THX4fd7e3uNzyQsrWfcq+c7bR+dDIZ+1jdjD9Yl0nptnOoKDTQCh6b9yyARH7e3t1e7u7tfBFNsdMZOdZlM6fln88e+n7n5EUPIaAZNdP1kMExSYGNjo9k+Akd03LrtBF8PL9iPusLsZ5yOx+OGmbEl3slvZWWlnbO7u9uKFx4DW6NXVcOoqQeJE5K+yOWyHX99vBhMwRRa/JzFd9Q7GAzq5uam9eyenJy0ia+vr7egoqpalsItOmtra21dENWh0WjUMuWOgl0KTuDyXMTuwA5gibKybSrGkcwZGa5s8bPQQPBc6LayslKfPn1qtCGyZzw2UoDd6+vrti5qOp3W+fl5bW9v197eXpsLrYJ25gYsVhgHDRyMn9K2gwvmZCWDR9DabZDck/dkIZ3d4/5p4HqANAMCsoe0R5A5gmYAKpTai66n06f1UgShzuZxz6rFNS/cu6raYuGqamX8lDV4tre313qPyZrs7u62Ofre6ZzyMJ24J4bO2+LzUOusIPVaoLgnmejb29tmnOAh8gQPe7xL0O4j55NAmnNWV1dbhpK1ZWR/MrufcmFAzvcJmL5mbBlMLZNRA1wnI3rG17/vBc1cExlBLrgGANI84L7O/iUvki8ZSDlYM+9wYs4iIjsErpzHePjf17DNMHjgGr3A1Ed+7xcADNvqSmpm1HuArycDBgyu4CJ/fEeSZjabLXQacH0cusEGvHAAA40z0WQ59vieoxe0JTDAd7nC5sSEW3kzcWb7l5Wtnpzne8aTfHRAYBo4sGKTKvhIMi6DM64L39l4APqBB+hSwFdQIZrNZgtr2Lze2jYlg03u6+DJGGA2e0oMbW1tNdALcGO9NZt2eddXdJn2QxJi2ATLCvOj88GBFNejwsozPf28rfX19bq6umrV5jzu7+9bBc32z7bKOMBV2pRXJ+H8Hv4iA/hsgqLpdL7Bl3XSyaTXr1+3NWPYx0yee/x52D727ITxEvenBX13d7cF/lXz54lmsikTI4nzkN3xeNw2V9vY2KiDg4P6+PFjlzeubKKzvq751fNHrlyCHajegOmhu9en4ZedAHU3FzbT88/EjeWIF7pkmpk/4EY/qgc87lZr437mdnt7u1CpRIagtTGR+WJ65Qv6ufvkuePFYArjfXt7W5eXl22wHCbM7e1tnZ2dtY0mBoNBExqyNwYUEIkAAKE9Ojqqg4ODpkwYQhhkY20hNpDx2AkUDJohEBWf7e3tptwEcKxjMlhgzigzET6tC+zc4yzveDyuo6Ojtu1pZmbTQbHDGk5+e3v7iwDx+vq6Pn/+vPCcEwdSFrzMLpuG3onHzg5w4Tasx8fHBaDgMWEkEwD05sr/dtoZZPE9Bu3i4qLW1tZasE22y9l+xuygFiNAVsKOKAMNgAHX4JlSVBahW46T6hS7EDIunhXiRao9UG/eWI6dgbHhc2YsaQsQNl8wYlR3yHbOZrO2mw/6inw7I58ONg1ljj/fG/hMp/OHPQ8Gg4VHAiCT7pVnPJYdZ7+ZY96X+Zouln8bS/8mHTPXeQ7gJg0MxHk5m+w1lAatBhJua3EyKbOsfp/BVDqPPAxKzSfe24Eh+3lP5CmDBPQpndiyeyX9OLiGF9yTTLA94khgkdc3z7EZTiRl5wDz8LoPfFYvUeHn3nljo+Sd5+8xvgQOoa0DOObLVtlJD1eH0Z8MpB20wx9++1JQld85eOLFvfCbPF9nOBw2W4TPs510B8RstviwYvhhu8WY19fXv9hanHE7G+459GyZgTiJNb/3piPo9ubmZuuoQG4dhACUWQ9ONp7rAXjxWVSvnKzDh/HsJnZkc9UVW3pxcdEFgpPJpO3oB11S/qxPDrqzKpzJUdsB3oOr8NleG2M7C39Ye84zpkjQOKBOG2aepW/MYMrv0X3kFPphe1+9elV3d3cLQRldRCxTSdzje2FDsC2bm5u1u7tbb968qX/5l39Z4MvKyvzh0NZhB/O+R+qg8QwywnzdzQTdCfK8EYX18P7+vk5PT9tGMX4QbrbTJj5KbGc/ZTtlzIH/RCZIGKALrqhVzTvCrq6uWkDFfPmegBCZdELbgRX0cyBKwP9/CqYwYqzlub29bRtFcBiE3Nzc1KdPn1p/OwODCVWLzxRi5zoiSMrSh4eHNRqNmkNifQwZIhyAM18w1uVjPkNZ+JzxQ9BXr17VwcFByyT52lW14LhwttyL/xECP1UdYzEej+s3v/lNvXnzZiFyTyWAnhgarsMcbNioiPgBasyPcdvRpLMw7/gLvTkPXgHGmRtCD41x1pYFv+/dLwOqzPpy3mTytE33wcFBc1RUM1ZWVlr7n6tUdpwEhs4QoiAGDNCWcdNfzPq3T58+LdAw+8cvLi7q6Oio1tfXF54xMpvNFoB08jF5YhmAdgDyqmq7WSYgsyylUTcAq6oWbNlo4sT9QOwERelgDdo5stpqp8UBqKItwq226B7zMHj2fA3sew4lx5rgz8YyjX7Kp/Unv/d1nb3DcTp493WSntwLe8eOls6AZiCTdHhufP7roAt+8D9jns1mrW3JNgs+GSTzeW/HK2Si52h7iTD/PxgsVqbwRwZuCTDM755uofeAM+sBsma7xj0eHx8bT2j7MUjHdroK5GxtVqVMg568erymiRNBVbWwDfBwOGz20LbfQDXpm/YcsNWjXep8z76j2w74TQfa3pCrXNfFnJGxDACrasGOYbPwH3yGP8B2kICjIgXPLLOmh+nkpI43xUA/qbQzNjYgwu77nswT/8C8CGRdGR0Oh63KlMkZ/AlrqkkkgheoIIxGo/ag3J4uHB8fN95Cb94jR04mZUBl+29aMU90CprCYwJn25isXFhOSSianmn7Upcslw6CegGj8QAygG0h0GVbd3cUMH8HKR4bdsGtjgTE9/f3dXh4+AVf/GgYxumqM7KYtt28wHbhiyx/rhyB773tv5Ob8J+WVbAD17ZdMxZP/nlsyJkxHrtaDwbz5DPfea248ZfnQZzBI3KQK6rUtjm9IIr/rUPGitmK2TueDaam02nd3NzU7e1tW6zorB0MQ9gvLi7aQ81WVlYWHCHKwqAxeAQNBB57e3utvQ+DBYB0hQoDC9izQqCMVhAEwJk3iM5fsv1V1RyjFYDx+AFnBFl2VFwDcDWdTuvTp091e3vbHj5nGqcDRYhc5s3zYPjx8XGbH2P0dROkWHBQLJTe2X4HcgmGrNzcByPnh68m6M4MEgfKZRpY4VZXn55wTRuDlcpBGO+rqvEUPrBgHNm1YjkT4SomSkSFyQGiDX5VtTbYnZ2dBTCNkltfEvz5MK8wZvQ6A7LtHMxfrmdwyHfmYQY3PHzPwMOG0fzL3+b4e7LsgM56yCJwG05XFflNOu0M7jz/BHvMne+sP+PxuK2Hy98vc8ies3kEv02DlH2+c3WEA6BJOzR2pLfINwG6v+vNgbE46QQdfZ4TQ4DMXB/L9bMrgPvwmceUTrX3ynHzst4yD1cke4FTykXOj6qWx00QZfvm5Bz6z7oNfs85ZNFpKXMQYT5lBa9HQ8uX2+hms3nSKe20gSKJEXyHv4OmVV8+AiOrUwbQeaQccw3ajnlvOgAsSVhmoiPnnskOB8B+YKoDLSdxOfBVbJmMfrHuxvKerayWPftUB1YkVVkWMBgMFnY0drCAPiNfgFTuB60S1HoLa/zJeDxuG1cljyaTSV1cXDQwmXaa4/z8vPHRdprDFU3GQkBFEGGM9Zyvh4ZuZYbmBJHMG589m80WnkXFWmmP5zkAbx9mTJM20d9n0gCcgGy4wwq6cQ14yLiY7/n5eX38+LFOTk5aS+jKykq9fft2IYhFdlO/ej7luQP58OYJ1mfkMzER55k+Hk8+UiB1toc3l/HGnzuwcmLA+rO2ttY2Zquqdp4fUfHw8NC6yVhmAT1MT8/J9OVzV3+n02nbIfO548Vg6urqqs7Pz9tiet/QGaTHx8e2PghnhZPqRdBbW1uNUKurq60n9fDwsLa2tloWaXt7uwVWDoTyxX0hnIWKuSD4tOVNp9O2uwkgzou+p9NpvXv3ru0mR3bESunFuzh6nEhVLQSNFxcXC2tcrDBuL8nI2RFyglXWpnE9BNLXTgE236rm7Tx28A62+N8GAqfr+2RmwnPMzxPYJkC108Jx7e3ttc03kl78FmeEksFnHA/OpadU8BPnyGcEVKenpwvzz0Dh6uqqBf88n6yqFjaMSP6ZHvz1dQnkqeLyXS+YsAHEuPA3nZt1BPpmYsJjS4fF7/Le6VRNYw4CQtqiSKTQWuuKE7R1iZ77WeYTBHieCbaZHwu3lxn3XkLC59jYLwsg+d+ZRQf5VfP1UbRvYn9yl7Fe4sOOO0FwyoaBL3yx/CKvgAbm5F567AutNoBl5CGDfHQ4P8vzln1PMgodZmwE/pa5lPHkOTaecWMH/LJdgbYG2AB5ggJnih00ZCUq5+ZKiMdo2TLv+Au447BtplV9MpksyBg8wx6Qte0F+9jx1GHLN+OwX8LeWhYcTNGu7+Qg2XeCK9sq7oH9dqWA8VHZhk/upPB1vIaXJI7p62qi+ZJ+2L6Yz9BH+Ebij3FXVZ2dnTWacA3m6vl74xkqdgQObsOiCuYNKXq+hZ1wM+Hog04iVzLNewdTPZuc9iJtDzR3hp/KLsk0+2L8nWXI9ooqj/Ur79f735jGdEoc4O+sZ/yOe9oe8nsKA2A+ghl4d35+XicnJ7W/v99knx39vMEWY3WAk7TNJETPvzkY4FzkhUCf8RlLYQtIxCDL4HH0xbbDyese/f257TUJeJKqxtPIgYP58Xjczqe7xT6ILiR2c0bHEg/b5nM/Yz6vq2XJx/8pmJpMJnVyclKnp6dtZ410PhzsRpcRNcbCSg4RCKBoxdrc3GzBFMSCcTbeBg9WxmTksmieMVGyvb29rePj47q5uanRaNTGc319vbBDDwAQQ9IrARqY8hv6aXmgbAIcM9HGmnMMxjOgYkce7xDG95yfSp+OOrPdGAvmnc47QQg8XQZs7Nx6zrmquqCDc1FWr5+Cd4yb6g10dLsEfdppOMwDB1PsjnV3d9cCaLIc8CgNedV8Y5DXr1/XxsbGQuuH+fc1B/JLcIHDsgFLPibdM2PIdTFSOH9vBY3cYnTSQTtg8zj5a7r2wJjlBVqy/TzXBTS7KuJxONNrmqRc9YIr85sNb5KmCTIZFzLcc8BcGwcBfQzGrO9ZnaAlhxYpnFNWnjLbb/CX+gN/GGvPudlGeO5V8wDBaxSqqlUIDORtC5bp8td8lnTxQmgSWeixx5zzsg5w2HEC3GwzzS9+66DWsoOdp6Jl8OTnDXk85rXn6CMTD6nrPf23rPGMQsuYQQm+tGoOTBzE0JKVcu3/kRvmi9wDqFNOoZ934vVzKNFnr8Nxy2EGntCfLg9oD7+ofDBf1h/5USvmQy+wzWRZgnT8CHLiZ9pMJpMGDtF35um5YjNpySM4BGCycZcxgDPm+CgC6J6dPjs7+0LGfIAfRqNRm691gnk6WIIO0IkkBfzz9+CJqnklwUk7r/3iBT9MR2yhd2Pk9/CdI3lmPVrGT3iR/DeWQteRX7AFSQIXB9InjcdPzwo9OztrG8bMZrN6+/Zt2+SCg3Xw2BvrZ/oV/vf4HUyBvSeTpwfa0hLq9XdOrJDksJ+jmssW/e5KgAbZJdFL9poH8Akfj/5YF6bTadvVGlt9e3vb9HtnZ2cBp4K1wat+zpzxsLGKsYWTJcZpX4vbXgymjo+P2wNjWWTpiI6BXV1dNQfFjW1MHNUPh/MH9mHM1tfXGwiF0BAGQMt53vwBAuDsrPRpRPI953L9x8fH9pwGFoPaqAO2YbSzflyvqtruPq6SVFXro676EsSjpD3GOSuV0TSbfrDYFdo4U5VAFtr5f4O1nvHNaperQ4AHX68HbJInCYDtkPkM/qN4m5ubtb+/Xz/++GP7LbwnoIInDhBQJldMOUynx8fHurq6qk+fPjVg4cwhDxF2SdyAnk1Utre3W5ANEEyH1zts3OG3M6dJz2W/h9dpZDnQQ+aYgW4a7l5g5WtaNnrJjQyGLG9sjQz44eC8vF8GzdC/B/r8uf+fTqdt440e/Xx+DxgbbDmwQ84MtJinP4NG6Nv29nZrYeg5KAO/zMouC148F8/BcuPtmtkRDfuSQAXgw/icbOE+tl+9cfGZX8sCqar5znsGXT2ZSh5z/wQ1BkbIQcqyx55tTFwTx44cmD+WTyeiEpxbx3yOM6Ape4zT1RnkaDabP2CYDYqwm5YDfscYkAHrr+diWie9XdWwTGfQT4WV5COt8m5NZ5zIEGMkUTcYzNv7uB6yiM2lKgBPNjY2ajQa1e7u7he7fWWCKg/oY39LsMF4LPvr6+u1u7vbZPbNmzdtHcjHjx9bks6yBa0tk+jiaDSq0WjU5N4+mK2uWaPpJGz6/OeAIK3z3333XaNn+s4MPhMo2xZlFZogmq3jkWdjCObi5zDBdwJLNothPIB+eOIkAePKoMn6kzSxX+E9cmh9tmw+Pj62JQAONBmT7zuZPD0f9PLysgUxyKPxjH/DGOAjes/7nFPOxZVfHtTrzSgSy1kv4DMBvfXHOpa4JG2cP09c4XvYZ/hcCjhOmJOEwad6TSJJFsZIG63jkAygln3GMZnMl1o8d7zY5nd8fFwXFxcLwuYbO1BA0FwBqJpXLWCUo0mMHaAchqK8XKcHxHvOMB2PGcp3LglOJvMtIgny6MuteopuvWDf2T4ypTDSY+Ga7HxCNiB3abFzd0Dlw8Ferp96fHys8/Pzev/+fVPSfKgyY2cRNTThOjaOTTBktKy4CVadhUepEuils0rDk0aQ+2J4UQzWte3v79f9/X3r9x4Mnp7nBd2pejBfV6pwSssCm9nsKTFwdnbWNvhg7dyHDx++CKaqvtwW9eLiou3ux+e9tR95cK6BZBq3pF8PIOf3CaR7AIJrGygkX/M6CVLzvrzsQA3UE3SPx+O2rsHPxHBSAn67VcZgNw1k8tbjYI2D55NO2GNPmgIWXa1wBpFzfH0bdeaPwwJcwXMnYjKQso3sBSY9nXOADa0AIQ4ccMKmsxcBm2e9AMk86I2pF2D1Ph8MBk3fAN5c2+Cpx+/USc4BXJhP0CXbxWyHPH/0Ans8Ho9rf3+/7eKW66S4ptsF/bL+phyn/HKuHb7n7XUEPDcRoA9ocaeIW7C4LnNOOmZQRfKRti3LKS+CHuZOEOVnzHF9jw9++HtAEpUAxr2ystIy2ARV0GA6ndb29nbt7++3Ta3AGq7ApWzDp0xA9mQZHcfHQLujo6MGYkejUcML7uyANxzYhrW1tTo4OGitSt4cAPtJm1+vVWuZj8ljPB7X8fFxN/hw8OBAxjgsA5nUO18PnWD8JG6QDSoyBGDYI7onoLPbbznXY+thxeRZj0a20TmPxCzMhyqVK6+eN/cYj58etUOSwzv/jkajBZ547Jmwd4IxEzHc1/6VgMJreqFxrqWz/cXGOpj17q5eI+/qo2mcn9nfV83bXqEl9HRywd1SPNLFiXA/iw06TqfTVtGkCEEXgbF3Jn+ND0xLsOT/KZiazWZ1eXlZd3d3DaBawWw4aNOpqoWMXYI9KjMEUru7uzUcDls1BwUZj58eyIWxpPoBMyGa2ywsXBgAiJCO08z2Di0ozfX19ROBVlfb8ys4x1klz5N7whSMBJ8TTBnMOOPhuVgZcTB+FgDzmEwm9Zvf/KZ+7/d+rwV4CKeFAmNIIGn+GQA5K+Wgy2CE/xFyQHCWfzNAs6BmZgWe8LkNnoEkZf7Dw8P2sGN+w8JIroUyDQaLD1vOYMighHuTzeE5XxsbG/X+/ft69epV203QRtvGdjx+2tjA26U6gPP9Uj9seJD9pKXlx59B15xL0tqOwobZgCqDpnS0CbCSz/nX4A3HD/ADdNkBYAwT5OQ9Mou0zElyrl9U03vO1tfKc5hHtt/1wBnyZj1nzPDWa00YL/LL9xm0pNNaFpCks4QOBkGmJQv4SS6trq629a2mnR1iD1w6wdQDoV/7ckXK66cM7M2rZdeB5k5IIXv8TRBiIANogkb2QzlffETSxK/eGHvzsBwZDHjsuaEDdgPbzOd8BrCwnNMSYxuS+mMdMxbwhgJOImRldWVlpbWVeT2GK55ugbMtd7ve2tpae3Bv1bytDnkxZhgMnhJte3t77dEkDkjM/9Qn6LC2ttbmaLtoHjvhi+999epVvX79uj1Mdzwet42UvAlIrq+jtfvg4KDW1ta+qBY7mAUP9BKtL4E/aPfDDz/U7e1te5Yj3QpgFtMUH4qvz04SDvt5dyKgx1dXVwsPzuYc8xF6kxgAFFNtqJrvOIxdTQwBr+yv/Jn1LfWu5z9tO/mcTdf4vmd7wLOAfzqK9vb26ptvvlmgHa2dVbUQXNhnJHZIH0jVjmqUdWw4fOqEWFlZaWuo0p/aVjigwlaQtCeYcjDbC6htB/2552B/BN1J5jmgBvOlPgyHw5ZUN74Hv7swY7uaARa4BLp/TUdR1VcEU7njhwG/hZDAB2VhQjakZMQc0SJk7t0ka0FFgSOjyF5Wj/cGLhgCDDfMwdgCFBib99tHyd2eB0NevXq10C6S7QmMg2CUbBKGwL9zP7QdJgcZPZ5LYRB5cnLSFGIymdT19XVrDyAwgB/mGXRxkIQQ9xy4hQ+DQpbP68S4byqU/0+D1ftNOi5no3Z3d2s8ftpyHl67SuDACSOM00KGU0G4Dw+MPj4+br3u4/G4Tk5O6vXr1/XDDz8stIch35bLq6urhetnxnBZAJJGO4Mp84/zMxhMOmeWDlnwuPwZDiD51guwfM6yIMYOK+mNrqBL3gShp/uW4aRtjicP6xuZwhxPb06Wzx5Q7iV5HAimOIKcAAAgAElEQVSiK1XVeuPtrAjW7ZABesv60TOAyiArdWxZ0OmxL0tm8N5tQG5ttEylzec6fN8LIvJz2wH8Cu2z8JAscE+PenzENyFfHg9jdIABTQgWkBnmD6iYTqft0R2WI9PYr5yjkxmp06mrJIQMHBkrf71WdH396aGtgFT4RcDCGAGubL6zLJgyrbGrrmp5TtnmB1C6vLxsttNVMWSHhOpgMGgbRQDUWP/gR02gj/CHjX8A5XTFMKessls+Uqct064mJ/A2xmCe0HF3d7fev39fs9msvv/+++aLGB8+FB3HT21ubrb127e3tws2nr9O0CTe+ZpAioMNMnz9tBnwezgctsQsG2YYF+T9+b1bOPG9yAuyYt/mqgJdQoyH7hPruPmV/pFxJF7MI/1XyoTB/nQ63yzDAagT8r6358K6KR618/bt26YLOT78AHbLSdxMklqWfU9kA1tNsAC9uL4T5caIxAFgYjbyss0zH5a1gyamSFuPnFXNNzWxvyVBAZ3B4dCYvQMIRhkLssT/y5Kexovw262oLwVULwZTTA6BAfA40iWAoofXggtRZ7NZ2zoWonz+/Lm12hDcYIQslH5uhYF8RsPcz8RCWd3raYYjQCwKZN5WmM3NzcY49+l6fplxRlgQarJx9/f3rQ8Uw+LFpdl2ZD74uUeOlk9PT5sADAaDFnRZGLy7UI/POC0bM/PANKmat2RAbytFBg4G86ZLGrU0gD6fl9tH6E+/vLxc4HcPEFKWtnz1QDgA9+3bt3V5eVm3t7d1e3tbKysrC88KA+S5ImcwXVWtPbZHk6R90sCy1QMzGdj06NX7PkFmBks90Gf98r1TNpKm5gWHZRD5dDuj13IYnPk6adQMZnoyZWDAuchN0qVHp5SRBP7IVK57dBaO9zyglPHwUEpacw0M3BrG31zD4EpHL1iBBgkeLZeMzaCWubLQ2iDUyYOstCxb68X1MhhcFmDxwt6y3iRtYupET945F/CTMo/MeR7IG86V+2AjnW03T3D2pk1WP/L+Bn+WvXT0pqHH4CDc7czYMnYZI0hx8OxkWibpUi+cIDDgdSKpB6hWVp42Dri+vl6oBuM3bauHw+ECSHeASKKQYIqxkagkULMsst7Da18N4jMbbroydnjpYMrXSZvt3f3gBY/OuLy8bL+jDTyfYcP2+qyhdBdF8gS+Lat+f83x6dOnhU07rAO2VXS9wAfsRfraxG8OlLA1fvwDARrfZSt+b5c82u+zyps2oPe3Nz/mZR+BTEMXXyf9HOMF36W/ZG7D4bB+/PHHOjo6aliZ51M60cbhII0KrGltufW98HHwCLxmjMp1sf3QwfxirmBT63YvkdLDFT3a+73HTQcc+seaQMsTNvbx8bG2t7cX+Eby23hxMBi0AJzg1PQzNsPeY+e97uz/FEylMGUFperJuRBE0RrnYILz6LlEGKbT+ZO6AfLumXYWjmoOysc1YIQBPQJoInGedx5yNguQw8I2K35VtQwkTgsB57AhM9BkPLSM0S8L46DjS8EU1/v8+XN7VpXXXrFDi9eVMC8E0Qpj4M+1E8Q4kMZYABYzePBvku4oInzq8SWFNK9rEIYRwKAeHh7WdDptbaYEjSgBzhL+IJuZQUpwvLW1VR8+fKjj4+M6Ozurh4eHOj09rY8fP9be3l57WCJOO53rYDDoro/z/71Xb/4OcJ7T05SZXgDEdXvj4jv/nyAw78FhB5QOh/+Zi/lPGy3gD+BkQJhA0IkMB3FugVhGI+Tg9vZ2YQelnL958By9DcAcKFoXTOvBYDFb7kXwyKuBXq6/4fOvDaRyTj1+2Gnk3HFEBr3mA59xPhXFbF3ieC546n3Hc+XY7tbB67IERTp0A4vka+oBdsFBF7Ty5kgAFAclGeiad6ZHT5+s58mr1G0Hxtm2AiDhM1q1AChV89YoB1PYetqQevYmZZr/De7Mg6TFx48fF9ruAcdcB5DEuDnPQTBYg3lQvcKPwie2cGbHP4Negjv0x+AtA/8MPE1XeOHAzvIFMGQ5w5s3b75YS849nbRAf8A88MJ+wLzgmlmd+trj4uKiJQcta1zHvt9tZzzjcjQaLfCT60BrJ82gNY8XYMxc8/r6euE99GS9Ds85yo2HzLNloB5+9bCK5RcZTPzlYJXP3KU0HA5bC6e7j7jW1tZWnZyctEIClaO9vb3a3t5um5NsbW21sYC5c51d6pvlAnsH9iaIAP+mb7BdchBru+0ktv2u9SQrUsmHxFtpJ8ADjGNzc3OhTZF7JH6w/8VXsHbSdtJJsZ5ts6wzrl6CadnxYjBlI2GFct85WV4yF66CAIS8+LaqWv85hPOGD0SdEJ3gCIHNjBJG1xkyZ1jSWbs64V1BBoNB7e7utpJmCi8ZEjIpKcBWbBSd6z48PNT5+XldX1/XwcFBK63i7LJlIo0BwsuToglEq6pt8by3t9fOZRehNLQum9qJPweW7dQt7FWL66kSPHnsznyk8+0FFL1xYbAxwlyfnQyROWhL0O5MBNeB/66qpfMfjUb1+vXrVvm6u7ur77//vv7oj/6ofvazn9U///M/19nZ2cJDF/mbxo77PWcM01n6WpaHdNzctwei0wjgJHufWx7y8x4gttP0/FJ24D90cJsHcsT1WSPghAjOOAOqDKaSPr0xcP9lO/nl/9ACmcnvmTef2bFYlh3EQ7fhcL5jqYNGB02uTvVe6YyW6bDpwJjSsRhcMGbG4E16qmrBljs4eK4qlWCnZ5vzZRnjYYys7VhmN3w/y4f1yADZ93CSBYC/srLSbC0BMDS37zEfElykftrm9ACO/+/x0rYNO8aYHTgBItgNN3XJgBXA4jUyvTHZTiUwNZB30Dkej1ui0pnjqlpoyeNYWVlpa2IJ8MbjcaviIJf2meghvpSgyXRAn+x7bFMS7zjYg18OApAj/3WGP1sVHx8fW0Uc0MYzNdkwic1oeq2j6V8M8P1KWX/uuLu7q/Pz8/beQZSDs+l0/jgBQCs2C7rat2O7TRfOd/IZrEWlBrxE4nZtba3t1MhzR3t+xkcPQ3FYF+23XQToHcgEeuR18A5GnPznYOtvglHk+uHhoXZ2dmp/f7+Oj4+/uCfXM638mAZ03nrogM+2y/INb9AJzvVmJn62pSvuYG/7pcR9+Up5NP3tL9Bb7BP6RzDI9+gbdoAWZbZLp6PMPHVl3TxNeXeQ6Bbol/ToxWCKgWQwZQf18PCwkNkw2KYUy3mubiWwt+EiQmSHJO90hWGGEWxKwNodhM4tAjh6C9VgMGjtYig0JXq2sfRW78zdmb8kvg0uzOKc8Xhc5+fn9e7du/a5n6/gfn7o7v/JONDu52zixcVF7e3ttQCKJ20nrxAcGw/e94Q9A5m1tbW6ublpv3OQhAOBNnxuoTXwMhheNmcHFx6HWzaREXrLnUVE+WzEucdkMlkARnnt9fX1Ojw8bBlVrv2Tn/ykvvnmm/Z+c3OzPn36tAAkega5BwANYDOI4TfLjDvn9ZyGPzMNe79LGj93bt7bcuWsK/RNp5eZPRtUX9NAxf87MHZQt4zu/p4xjsfjlgHt0SvplnQwjQzSe+PD3nnODtr9kHCAR68C1QtSuKfBeo6TIwPv/Cz5ZcfowDUX/Kb+LAv6egHfsiDQ9L26uqqqJz24vb1tHQ69QCr5zzW8HheepExU1UKCDDqkY3VFw37Lcv6cPj83156uJh+d+GMunIPdYkw8WJi1GZZnV3PhLWN3y1Xqk8/D19uOATAzmeod/Di3qloiyllgdtyjCjUcDlu2mcAxfa/HCv3AA7yw7waejCN55c+9xpYg1kEZmXMH18wdrEDrIu3+DtQInACs4B0CfcuPQasTWMYYaXNfOh4fH9vzqHoyCX2x0QQ9BLesoQHPZRLa+u7dnOkEYi0kWMtYbGtrq+3UBh+hje1uL0DyYV1dWVlpXVBpKxNo52fuSOI6/JalLvDE4+C5k2ASP+9pNpvVwcFBO9f22ZWSDPAdwHO+zwErp266YogNsU0kYd1rt3XHD0tJsvKets3ya/7YZ+D/CFb5fnd3t8k6j2hCh80L6w52nNa+HtbiMC2tR67SrazM93t47vj/qkzBEJgFoa+urhqQQsAJojDwVnYqPHZEflYC93V0SiXGrTlsZAGDYZTf28E4qML40g6AIVhdXa23b9+2gIXdDDHozppTguQvNPHcmDNCe3l52aooOHrvOsjce6AQ5WL9AMHXZDKps7Oz+ulPf9qEiOdHWKidtbRBTuCeQAqlGQ6HDdwgE8sO5MUG5TnQ81zA4HMMWnE2ZK6QES8Ehc8YYcaB8+Z6eW3+7u7u1t7eXp2fn9ds9rTRysnJSf3kJz+pDx8+tAc7b25utn54jp5TM5jo/YXPBko9WiefenT1Z+nUzBNnx5NHBunJB2TK3xukOng3j7PygiPm+tgYb6qCseczALWBWB4GtDaWZIeTPwZZz9EywW6PL7meE3oABB0c9QLqDEwcaNkJJZDwmBJc+pWA0XNz4OFAYWtra+HZQLb3bFedMrMM7DwnU5zjezvwceDeA26eM/zGFrnjIMEnNh4acB5dE75mjteJRm+cwnU9rt7/KYPL7CPANhNV3ANw58BkY2OjPfKjar61tNeKcU8H/wQgOc4EJvyPXwS0OakFeISmJBCMCbgWgM+VUHbwYqzOjpsv2HweTUGQgg5VzbeXTjBtPeklHJgb/twAEnmEXuYJvx2NRnVwcFAXFxd1cXGxIOtsPOHkSk9WUnd6gZRxxNcEVuCH9Cf5Wye9ptNpw0i7u7utIoCuuwqAbNHeRsvf/f19Cy6ZB/dhLene3l7b0h6aA5zhKZ1PtotJN97nHNO/ZmBiPYOvyJ6DXfht+vA77wXAcXd314Kvx8fHOjo6qqpaqLzbXxm7+PqmL/9nUr5q/ngdVxYdbFleoS1BPbJpW8uaNebWS5RxmMZ5jvXMQRJVssHgCaMTA1xcXDT9Y0MPglpkgAQNuNr2mPmax7Zl6JGT8b3gvHd8dWUKBroMiNIAsBkUAk9pnhIdYAhD7Sx1BkRcm4jQiubMBBGkhcLgizHZAfH+1atXNRqNmtAS1fPcl8lk0gIrr20icEJAMQ4ADfb191gArKx7gqau1PWcaIJcxujKVNXTjn5k81Day8vLlt0xyHV07uvDV1cJyEjwmUFWGqo8DNIAIOYj901Ql4Fe0sQglRfr8VgQTZDNHCi726ghM3ZSnINh3NjYqP39/bYV73g8rv/93/+tg4ODur6+bu2VOzs7C8+9sswmTWwoezzhvJSHNNjWTR8ZFPHZMj49d34P+Po3/s5j6xlWX8vBRs6fDBg6XjUHDdgPZ4KdHU/6+X8DDycF8kBXE5QwlwxMkSk7Bs9pMBi0LDYOg/NcyXVACW0M2DJA6wUqyaNlvH5OPiwP0MzzBfjaL9BWkY7S43Q1p/fKeUwmk1YFt223H+oFopYBr9PwtbHfyJFtmsHd/2vvzXobS5Lz7yClKkkUqV2q6q7u6vGMDfvG3x8wDF/ad74ae8bALJiZri7tK0lRK98L4Un9+CjykKrufj3jPwMgSJ4lT2Zk7BGZp9VqvShvpN7S7/F4HJ1OZyLb7fyTjZn4F+2R7rL7RIPqj2iNGXjRLLMeKmtVdkcGhwwYymQPTnC+3TEX/vV+vm63W8rxxMvSV9wpS89RVYaCqnymtkGXXaF1UNyESeu2de/KykqRyTIIKVua1oQxEOq8wPHLBlheXi74k8HJ+zRXsnk6nU50Op04OzsrlRIyAFdWVoohzblwntJ/6g7JRMqQ18DFxcULHiVeqKtUNnp3dxej0Sj6/X6sr69PvDKA8lP9UWBIjhTXJbZaT9k7ZTy63W6srKxMvMeI9C0nSvJRPKdr+Fznoww4t/rm/Lk+1nOpU8bj8URpmdrV61QE7Xa7LM2QU+U74wkkW4VXzgNlQcaPohdWP1E3CYeUj8IvX9fBZ9/d3RX+4/u9HN81m4H2gOtT2Wlc08ms59raWslwPzw8TLzTTwFtOkIRT8uJGJhwu8BpXDwk+SdZmTmoDjM5Uxq4R/OkaKRcNfmMpEpgKtqlzIHXIHJdlAah86zNJaEK6VSmImxGTV0oRTylBTc2NiZSl1KkEc8G1d3dXdlq0ZWCojMSrFIKymidnJwUwUjG0257GifLUGoMz3HIqKSwOjs7K2WJmh+9/4A7iIk4Il4qyMxB0vOEd+24xvnhnLBNfTPyqWsy44fXkek4fs0No/qKBq+urpaXR7PmXJELtaPnUvG4AaioqZzRxcXFWF9fj/fv38eHDx9if3+/tNXv9+Pdu3dxcnISFxcXE7tPsf8+dgo2OmGcV85JpiicRvg7w3ntnhrNZddlwpGOFHez4vNrSlpzKBlAnqBBKPmgIAYDOsKV41fP8QiujPRszJnxRP7zLAsdSMkPyiPd70EDRsvVfwYvSOfMQmWfTIFxLE4TmTPlvzVWORDMJsoYogMpme4ZM+9j9qk57cSncCTZS3nPOdO9kskuU9R31uCLZtlXLZLXedGpZK+CK6Qr0gfnh32rGRsZ/p2/NG6W+0mf0umVvJPBSpoSneq31k1khvos8oF0PRwO4/DwMJaWluKbb76Jjx8/TlR0yOhVBpBBJTrICmRFRKyvr8fOzk6sr68XA0pZNertra2tuLu7K9uKK8DWarXKtsmkZ0b7qW80z5K5bmBz3Ip8u4wmDuVkKeugJQlqT+VJMgKbNlfgHJH+man9ElApu9Om9KA/V1kObcSkjbHkIFL20eCXrOYSBAVD19bWotvtTmSeRN+Ug+JHd7I8Q+J0SvtB4IY1cU56ELizrP+aZ/ZjeXm5BNi1E6nakg5Sxpi2AoMlcty1CzZ1nOt4H4d0owIWTDKIb7jDneS77pWM01zpHjn+nIMM7457tpv9ZiCIOuHh4SE6nU5sbGzE3d1dnJ+fl2sdD3SCJRe4RtTn0+0F4Yv6R+85mwYzl/kJyLhSoNyOVIQvIlBal+sGpMCUJlxZWYmNjY3Y29uLr7/+urx8VcgmM4uR9JsCTISsWk89T8glsd/f38fq6mrxzCVAafhofCynU4RP9/C/cHN9fV0U8f7+fvGO6Vj65NUyPS5QxcDujGobbzl0Wse2tLQU7969q2alakaMzomgZDAy7an+ZBFnCiIyju5j1J7AqALHx+c7Hei8jBtlHegY0ahhaZUzs+hDSkTRREV3tRWnaEUlDioDVI2/hIAb5hoXgwDEhWiEyjTDZ5MxJsjuyYSvG96ztK1zjCxlRqPPox9zw9PXF0VE4Q/fgEblt6In4pbj8wiuIqPZeGpOBtt1J4Y4ID3SMKUzICWVGclqz42HJkeKPFz7blLAxHU2b1SqCsYsLCyUwBLbE93LWM2MwExWOP1ERDFCvC80Dji3TmtSpHyWdo7kO6seHx+LjO71erG6ulqUtWds7u6etuPl7oKkU+GChh3nOSuJ8bmg8akxuYGl6zQ3CnJJp0gXr66uRsRz0I61/zTgtPuVwLNdGb1oXGqHEe3b29v4/vvv4+PHj7G8vFyeq8yEnDn1g87A8vJybG1tRa/Xi16vVwxJOpEeMI2IYjcoQ6KSOV2n3yxTJf2Rl11G+//aXJLW1b7w9fj4WNb/aKcylcfS8fB3+GT0ozaFO8pN5+NZ4PT0NAaDQXS73QlnQ3zPTIZKSWVzXF5eluwU+ZE01Go9B94jnl+tsri4GN1ud+Jde7T16CjpnAI6mR50nmoy7v14ZmT7M2iXkg+ZlVQAcGVlZcJhZnXP7e1tnJ+fx+7ubgkiREzuUMnsOWWdOwAEXUdHgk6fxqogEfFGXSTHWustuaOe6IDyuubU1nBP3aX+6V6NST6C+n5/fx+bm5tFfmsN2ng8LmulnJ+FB+dh3UeZSpxTd3Oum2DmMj8KBU6sJkxEpYFIeGodS6vVKqV/MkrpTMn50ESr5pnRKBGhUqM0rH2iWCdJBddqtcpaKwlY9Z0CmmMjQUtRyMkS8bEMUDW8iort7+8XYlAWSzj0tWZNcyDQfQTVH4t4Tk9P4+joKL755ptUKQgvVNScYzLYw8ND2dyBcy2CdKXiRjWNOBEys1l6Jsepb7XFCH5ETDAdaW97e7tEWxippbOqtqgYafiwVOHh4aE43YPBIE5PT4uCj4jY2tqKnZ2d+K//+q9CD3KmPdpEBs4EIoUK8ce+ZcZwRjvEY+0c7yX+adRTKLox533XvLiC82e7YxIRxXCVEeFKJIv6S+jTGMkUovOzssnTwPGi/jKDxGdROYl2Iia3iBctttvtsiZF41KQSQrLlVPTh33MDAYqEJ9jzp/GmClrpws6vzQwWRbCSGOm6Gr06TgUSMY6fWTzTxmpKgIpYTnoojllMTqdzkRpERW+jCJVT0ixK4DCTJY7wjUnioargM415bPwr/GSL9VPGlI63u12y5xcXV0VHcRKBTmUfI+N5BnpyQ0zyiwZWXqfZLvdjk+fPsXGxkapLpFDx7kWzyuour6+Huvr6xO7w8mGUDBPjr3oQ7zHgOnCwkJZn63yK+o2jd83IyDNMCDG/3LU9OxM7qlPkhcK9Ha73dja2oqLi4sydlbtyH5xI47ymPqJSy/07NdCv9+P8/Pzss05g4ySR7KXhMvb29s4PT2N9fX1uLy8jK2trVKq5xmUhYWFks3S8dXV1ej1eqU94StzHl0mNjlRTfqO1+s3++kyRLSeBayII94v3nv79m30er3Y2tqK0WgU//Zv/xYRz7K23+8XWzYrqxVvcZ8CzyxR/4o/dJ8CThHPmUA5KLpHPCG8yuGVXc422+12CSy7vZfNQZNcd70iPDOITH2vipSIp0q14+PjiaQEdRrllHjb7VviTMDABG1f2r1NMHOZH0GT6gpNb1kXg0sYaxG/GC0iyroiOTfD4TBOTk7i/v7p5XsSdjs7O/Hx48fY2NgoEQwpMdWcSomJ4TUJLAMUQ0ZEUZ5SOopySIEwy+ROld6BdX9/P/GeLJXe0QigIDg4OJgQeMSdl5s5ZEYdU5ERTwR2dXVVoosqRREhZFmpGlG54c5PTVC78Z0JLDpQZGQeq7VNxnXDNeLZKBfTeRRIzrcilhISrkQ5Dgnw5eXl6Ha7cXR0FKenp7GxsRHffvttrK6uxsbGRrx9+zY+fPgQp6ensbKyEufn5xM7WrGfHCvH64KRhussimLacbbneM7mlBFQn4fsmPDF8zQCszZqdCelLf7VtSz9k7Ji+6RXz1SRn4fDYaNwzPDBvjqNuHMoQ0/3SYaQ/plJ1f/MWeEny3ZkvJaNgU4T+0oDnffUnCk6k+qTDCTfwl3zwTFMc6j4Xxl2gnApPcMssfpIehEdqLRP98jJUGBE2QwZfiwp96wh54uOje6TQ6IIuuMkCzwRx8IDjSbNl88TaZ6lU6J56StuzSyDgW0yi6dAkOPVPxo76V7OqYKmw+Ewjo+PY2dnJ1ZWVoosjnjSw6urq2UdkUqH9E3cMzvC6gjNRcTz+jXhbTAYlPFIlvtaF9erxK+uUx8YbJXdwIXvvE5tUkaoZEh2jF4Gz6CLdBRtGRqIpBNm2pt2Ap4FRqNRfP78Ob777rsJY9QdC2VdZJArwLi5uRnD4TDW19cn6MP7LPwpUK5somgok3nOP5m8yHSVA/GYOVOur0n7XurI6/RsOjvD4bAkDRYXF8vyA/ZtNBqVV3S4Dcjsq2xpHpf8dWdCOM8cNOKOc6nxeZnt4+NjCRTpoyAJ+YZzkOmk7Pm1j/SFeF59Ex8/Pj7G2tpajMfPG99xAznNoWw8yWU/n9ljEc8O1cLC8ysihM9p8CpnihPGHejUsTdv3pTInzI0u7u7sbW1Vba2XFhYKAv15dTwvQPaPU+bJkhQrK2txd///d/H9vb2i+2EqdhE2G7I6VoiTsclvHQ/y1RohOmYFI+2nuVuUYz4KerXaj295fzs7KwQHSOsZGLiuob/iJc77Tw8PG23qc0ShsNhrK6uTpRV+DaRwgUFNhnDhXhNQLsBSMWkudB1UoYyyAQieBrBAncE9E0nWcfkTOlZrdbTy/IkuGT0UYByvByTGFHr6y4vL2MwGMSnT5+KAyVa3N3djZWVlbi+vp5YcOpCJxsbcZvNhSuLmsHsfc+Emb5rc8rn81o3tH0sWaYx62eGAzcU/Fr9poGlY+JB0lzN6NPn6uqq6rhnfeI4KFekcBioiYiJ98bRgBCeNJcs7xVde2ZqWnbKaTebcypbHiN+qCBdtrgDrnGzJEdlcx5R5vMzA8iPsf+UBwI3IrWeIOMjbgWsTYQiInq9XjHgfbG1eJ79I++5U0X80HjOPi5XfbycJ+KIMs9lIR0J9Z2VC61Wq2wnvLCw8KJsR9fJ2OD6jFq2lzqONEyHStUZj4+PcXh4GKPRKPb29spcjMdPmYm9vb3Y3NycKN1nVobOCCsNXI7o+/r6umTfIqKU/Lfb7YlgLnHpQaCI53WfkuMMeuo/N9MQrn0OnWdarVZ0u91YW1uLk5OTMhYabk5vmaymXMn4hDiZBR4eHkoFDZ9LftO8y4nU+fPz87i6uop+vz9BPy6H2+3nd46q4kj0S5nhzhTlTET+4m8/71Az7Amy37KAowdN2CZpiPJ8MBgUR/vw8LDwmmhDFT8K9EZE4R3xMHe29Ix1pudkM6tNBo+Wl5cn6Ez9lpxU6a3mVvfy9QW6R+dZ2eV6MpsDt4fUrp/nnAhnqjjiJi1yMlnOKLtbDlUtYJV9j8eTyyxqvJXBqzagEGhylf0gsWsy3759Gzs7O7G7uxubm5vR6XSK49RqPb8xWvXUErDcmlbELedKNb1EppwuRgzlEOkYSzZECJoYjU1t6BoRnY6J0NS2HEotTuO6KoEE+uPjY2xtbUW/35/YRUkE3JTB4HEKaI/6RDxFid6+fVucO+2GwywVoxcRL4W1GxGuFFygu0OTCSkaAxwLFY8LhYz2MsaQ48QME6ML6pMUgDblYKSHz3HjLuKJobnqBhUAACAASURBVJWFur6+jk+fPpX0/DfffFNofmtrK/b396Pb7b6IDGXz2uRo1Rwih5rDkimK7PmZ4o94Lk3J+q72aw65K383DrIxuHPF/vE+8SD/swR0miP1+PgY/X7/RR+yvmRz4tkBGtA0Mt2BIS+RxqlQZEgyUMRz5MuaUe7zyTnJ5swdKMpxnuP1mSFDQzBzqJr663SlD+Uqx8ZorWcrdR2junKk2u12rK+vR6/XS7No0/rpTo7wxf7REMzGRbqext8Z35GOXBYrYq0gYMSzw67IuHY7VQWDgp+cOxluLJUmH4h+WV2R0TF54ocffij6ezwex97eXrx//77YBlz/SpoVDSmK7vpC9HF///TuuLOzsxiNRuX5Mlg9aKrnSH9l88rMG3mGdCjcccOEiJfBpVarVSLoCs5RVxP/Gb/y+e7MzmpDTIOjo6MJvGROtOZEyyREQycnJ/Hu3buy9k59VL/EfwpksFSM8qLVak3wpcu6TJ40BSgcJ7VzxDn7rbFnQPnpDoHWgnU6nTg/P4+Tk5OJe3V9v98vNERg1l1OFXfAFq+zf3Sm6GhTP7VarYnNauiMk6/lEGvepGOlm4g3t/2yOaDz5lk44YJ8QN5jQOnNmzelZFjB8oWFhRIgVd/kUC0sPG+bznnOaJs8xao2HZsGU50pV8wUIoPBoOx2d3p6Wiaw3W7H3t5ebG9vx8bGRnnxWq/XKwKk2+1OrKsS0uQZk6FUZkWl3mo973rDCeC+8jR6KIBrEQY3VrgQTROk62m8C/RsEo6Ujd6XoPs9es1+zAqZM6XUrDZOUH0/ceDOlAxnNwbc0PE+cpy8z/tFI0DC2M/rmwRPXPNaMnHE5HtDIqKUIJDBtQ5C9KeoIhWrj5GCXuUZvV4vLi4u4uzsLPb39+PPf/5z7O7uxmg0irW1tTg7OysKM0uzO2TChwrVcV2bCx/DtOdm7VDQ1Oa71n9eQ2PP5zrimS99XjMDwp/ljktETESb1T6Vg4ShjOvaTn4Zbvx/5kzomVJKOs6IPZVIxLOxJZ5RRFIBEPKUGxU141/ghrw7qh7EYH81lkzuZ0EOPosbEXFsNDSyTyZrWq3WRFk0gYa8r5vSHKmUTVtmt1qt2NjYmKiQYJYwwyXHRifG9YQbVPrtEXUfH//z2/GvZ2T8ovOiLxkRdKjU5nA4LJu2yBDRdtyDwSAGg0FZEyzcyLAQXvXtAQqOi7pb+m00GsXZ2Vn0er341a9+VTL5nrVzPDNISmNXjpT6NBgM4vz8PO7v78vabNoIisB7UI8GvPOzcC59SqNPeOCOkKyUYRs0BkXX3W43tre34/j4uOgv3e/GLYEGL7cod2iSozXY39+fMCR9zklv2lxBzzk9PS0bUmxsbLwILOnaXq83YZO4fKs5UrrWnSnn2ZoO1Dm1kwWjne95jc8haYi/1cbNzU1cXFzEeDyOv/zlLxObJVHuiCcZ2Nez6Uyp3I604U6B6zruVSCZKPtL/+WY0FZRVocvIZb9rjI/Bg9cZrodR/3htqA7fAzMUCdRjzLQGBHF+WMlmOZYjruX+gkHom3Ht3iq1WoVf2cavCozRYOCjKzIksrWtre3Y3d3NzY2Noqw7vV65S3PvV4v+v1+mcTHx8eJN5TzPQJLS0sT0QwKRxKTJowKlVEnR5yIzIWVkO0Gte71Z2py6d2KYBcXF0u6dWVlJba3t4tA1yJ4Za9qAsDnoAm0oFEKheWQHtVyIckxuQE0jSayaCzvdeaiQyVirV1bG7sLPl0rIyKL1ImelFVyZ8qfLyBdbm5uxvHxcZyfn8fFxUX89re/LbSkd+0MBoNYXV2deMG0tzttbJyfbA5qBvQsdFS7zg1oCsPaPRyXG1VNc8YAyLQ+uuLLjFEaWe5IMcpX28mvhpuassz6J3kg+eG07pFEKgZlpNzQz5ypWlR2ljHw+ZQJ0xx4XcPzNEoU8ac8oGHtSjaLZPo4WP7Cfitoli2615hk5Op9R7u7u8WR8qyfGwI+15nBkAWceK2iqH5tbdwZ7xHPbjRxXvg8/ZczIXrUdsaXl5dxd3dXSmBUHaEd5KSTtHGDSgTZHxpBXJdG3LBCROuLh8Phiy2KNZfsKwMMnHN9e2ZJmy/xvUPqs28Gpb7TadMzqMfH4+d3QnL9Cw1tL9NlFJxt0VbhnO/s7MT19XV5h6Hw5pkml2nCmTZU8QqITMfMAicnJ3F9fV3W16m9rF2+/iLiqbzy6OgoPnz4UJxc2hpcS+dGOHnEN37JriMfuc3h/c5g1qAd/5NuiA/JQH1Ec+PxOK6vr6Pf75dX5PhY5dBo/4CI58wP9RX3FqjZcBHP76SisyV7WY6V41bOlGhP/CJHajweF9nB7ftZhu72QSY7Ha9ZAMblL/lGuGfWu91+Xi4j/0FBEzlWzjfMTLNtXiP+Ek832SiEV29AIcKSM3Vzc1PeDaAytg8fPsTa2lpZXKodbLrdbkQ8bcMZ8bxZgBSBJke/9Y4fZbY2NjbKhHrUQJMiYS0hRuJX+0IOHS9nEhpQzB7pehKD2tTOIVIi+lYfe71eRERRAFJkxC2/XwuDwaC8fI/OFIkyM0w5LjccmvrihlHtGrVNAcL7RANNAk5tuXPB+/TNdXRss91ul1S5vtU39tUNHQof0eHx8XGcnZ3F5uZmXF5elp1vdnZ2ioGhHf8cDxkO+SzCNGdG42o678/z9rNrsrYy/GfjcQPfhT7vz9r0/lGZZf2JeDY4GS33Uon7+3wnP++X96eGD66jkIJhhE38zyCPyyNGs6XIsixxzZFyueX49fG4g0TcudPocoLX8Fq1yfp84YGK0pUln+WfiJjYxIX9ZvQ1K/GLiBLNvbm5ie3t7fJSUW7rmzlPjmM3BLJrHG/ZvGVjdoNtGr1l88vnyZnSRxlOZqxarVYMBoNYWFgoMuvq6qqU9tGJoaPDuWZmkIvn2R9mg2QQrq6uxi9+8Ytot9txfHxc5kJ8IIOS60ZoKEpviJ/k+KkSgC/jFe0pQLGxsTFhNzCgwzaFO0ao9WLaVqtVNq7i9eILbmTFF606PUQ8v/9L752SUev8JXBDbzgclvcUzVKCNAtcXV3F5eVlbG9vv6A3ygwZrx7IODo6ivPz8xfvnHp8fCyyTdc6f4j2KPv47CY56PzXpLuyY349g2OUjaQbxw8z4wyke1sRLzeZ6ff7LzJTLFPmsg3fTIz0woyKHAHSN9foi5cUXJKz12o9rX+V46FXRfCFzJIRvq5X86rnu/zMZJ47qcI/gyr+X9VqEc86ggGRVuv5RessWRRe3YmibSyc67+yerM4VK/egEKg1JfSl91uNw4PD2N9fT12d3dLiZlK9FS+IsZhOZyUnBC+uLhYXqgrh2xtbS22trYKEp2BuJOQCMcnlEKeRMiJovAmA/r1TiQS1BqXcESGFA5OTk6KovfMyI8BRRSFU0a5nfGycflcTzPOXKjxXrZBBeu44fOJKwefS7VPZoiYfGm0G2IU2G/fvi0ZisyJzCLIihip3v3y8jLOz8/j6OgoPn78GJubm6VWWtdRSGTj4nkqW13v+HUFUIvMZTCNxnQ+c7aFY3++KzMfQwZ+PRVOdo3As43eFyoZd6aYqZhWfpm17SBDigYAceVGPpUYo158l5G/10htuxHhRj3HkBkYmSMa8XILdM6xR+Scz3SMyp9Z4cyhcJnB/vonIibkI8EzkI7rx8fHsrtpp9OJra2tCUeKOGUfMp53eZBdQ3zWxtP0DM5fLeDg/RDN1YIJMtTccaT+HI+fItdbW1txf//0fr6Tk5MYj8clw6IMBA0e8ZZo2XUYjR4Z32/evIm/+7u/i/fv35e25aCoL9yRl0ah5prOBCP9tYX1Cp7R9qBeF578w/VmapfbgCvrFxEvNuHSt5x9lq35sgLhsNvtFoNavEP+0rhl6I1Go7i4uIjhcFiyBxmtZDzfBKPRKA4PD+NXv/pVqm8FcoydN7Wzn945tbS0VHjTd2dkX7PsNenI5Z47Upnt4ZCNx/WJwHUS+Uxt1Z7FOXfIeFrOlGdb3aHX3DOAxD7SWZJuEWhDCtGq6J4lcQygeFaX1WLUWV5pUZOnukZ4Ja/xXtG+7lGfdB35TPSiccvmU1WWnim8qa8eQKdeY5Au4omHpdNmeZXKF2emlFJUWdnGxkaMx+Pi8CgysbKyEqurqxP10TI6Gb1R1EL3KgMlZbi9vV3aIOGJcLJSBG1CoMlyphAR+ST6BPpuPSRkrqtie4w0UFho3Iqq1Rj6S0AZL6VkGT0gkxJHGdCoosFG3NSMg5ox4E6riJzzpqg276ew9LZrgpT90lio3CUQ/D0Fjn8fm/AoGh0MBnFwcBAbGxvx8ePHGI/HMRwOy3slWL5Yg8yg0ly6cOJ13t/sObPQU6Yo3LjI5oBAHGd85v/9+TTqvV2/1g0mXkeDyx0pfWtHx+w5Wb/VPy83k8zhfY5HGoL8LYVHo8HXCtCAyIwK/lcfsu9a/7J5YKbK8eHOFnFOWmRWUIZ8k7zw/+yPIqQOHIcbFa3WkzGsQIm25KZBQF7OFH9m1PGamgObycvMcczmL5sP/haPuCzmvGbydzweT6wvoANKPhJ+Wq1WfP/993F1dVXK9HUNDS3J72zzA/V3YWGhbP/97bffloCTyr1okA0Gg+j3+9HpdCbaIJ0RnzLUecxlg+Q8d/GT/Cf/0eGks8XxtdvtuLm5ifPz87LsICLKRlakL31Go9FEuSNxrrlURo7vw9G4aVQL3woSXF1dlXJ1x/2XwsPDQ3z+/PlFG67bPcsueHx8jIODg7i6uiol75mj7bzVFHCp/W4aM/+7jUD8NunMzObgvNEpJNDoZ1Y4u15tspxZ5Wq0EWWncLMdfSh3uemVni07Xcsa1EcF2iVj9dJbBmD4Yl/Nr9Nwpi8z+yHDE3kiw7kCjJ6Ni5jcVK7dbhcnSvTBrKg2sBOe2G/qZV8Hqpdq1+ba4YveMyXEahe+0WhUFveura3Fmzdvyrs7JHj4grtvvvmmKLx2u13K0bRVOoWdMiya4GzSqFTo3Hh6kNdSAfA8J7PVak20x/M+GbqehpHG0G63Sx2yBKKYRfBjhKBAxjwX3dFAyzIAHC/HQweT486YpckgoHCioOHcqaaXBitx4sZIk4NFIc25oREgZmPUlWPxProg18s9I54XPh8eHk4o7+Xl5djf359qDAo82uY4dVxmSqTmFPpzM+XjdJzNnd9LXFOwZwrM58ufn+GG/arRIWmahjxL+7jzW7b5RDYm4oHHW63nl4y6HPBj/N90XjhkxiT7UL44bXrfa5DxDQ1VGpOOf58fx0tElN2WdO20bE2GZ/3mBgOEzMDhMcnYra2tsumRB5bEX5mhlvV5FoOuaWyZrsr4MPvvcpjK3R32LPsR8Sz/dNzX+6g/6+vr0Wq14ve//33s7++XzA9LV/UMGnWOC9HSYDCI9+/fT7zWhP1qtVql5FBzrr55SSYj6cQVg4UaB9dci185B/wvZ7Ldbk+U+DDooTGr+kU7ExPfkv/qS6fTmZDjinDLsF1cXIxOp1Nkk4xB6mwaerK3Tk5O4urqauLdRTUeeS0cHx830qeMYK2tclrX5kx7e3vR6/WK8UoZ44GhGt3PovMy+eyyyufbr6/pJA8c8dl09F1OaLxae+Q48L7L4HccU2+pbJkb7rA9rl+kravlJHLG6OipPdnf2nhGG1DIRhffcVc8rnOk7sp0EAM5tBM0NwwiCAcKQOgZGivpaGHhaVfJwWBQaIlli8ILs/U+3wwueSZQuyz+LM6UE7OYezAYxM7OTnz33XdFeC0uLpbNJ+gIacAqL2Akh5FDPm8WQ4S/dT/T8V4/6h42xygkZ0qX/cgiBDR49EwpEEUItLiP2atZDKFZoN/vx+7u7kTJECN37L8LzNrv2nlnnky41ISWFB4jrnR+eZ33NWsrE/5kTpXoCVTqR6Ve6zfHG/G8202r9WTAX1xcxOXlZSwtLcX79++LU39wcJAqu0zwM2rk4+BYPSqetVtzNms05g6LX1tTOHwWI1g1g7v2TL+mptwoZHUd8eHrpbiAVwa2NiTwvmTPc+EuniZNutNRy/xShmX0xfHQ0OAz2Q/ihO3UcM0+cox6Zs3R9mud1zxKyxcoZoaQ47XW/1br6d18WTCiqZ8Rz7s7bW5ulo0VvNyNfeH3tHN+nY8vGy9xlT2HY3J952MV7l2WZ8d0Lx0tPYPrdZ02e71e/PKXv4z9/f3yHhxmjBYXn15kroyP0ywzLisrK/HVV1+VDXr0TMl5ReIfHx9jc3Mzer1eyR67fmD/PdjnekKOlOafBp3jO5sHGVbt9tNrMeg4qS3pEN2jsQvfzFzrXu06JkPt5uYmlpaWSpXK2tpaLC0tRbfbfeFM3d/fx9nZWXl5vAeGmmT8rPD58+epdgkrYFxX39zcxOnpadlUgTSpbDVxJWiSi028wmu9r9l/v5bPzeRj7TdtFp6TQ86qqNq6TuKTu8f5nEt3SZfRoVKfdYzjUbvCvRwKZVtUGcayONGc5k90zN2QPTGhfrucpJx3m0Qf0njG08KpHCr9Fh7ka7BtBiI0bmVKeZ2+PQCrNYnEbbZjpsMXZaaEHC3oHg6HcXt7Gzs7OxNCUBPA+zJDgZEhKmESII1NP8dUMifRhbGEGiMFnFiBfrs3y0mVca4aWU2I+iEClEARYd/c3JRaZ1/A6b+/BCRgPVLgmQ43BLNzjptM+dSMbB+HG+tkIB3ztV1si33IDEI+U3QgxuKuLlTEHq2dZhSphrfdbk+886Tf78fl5WWsrq7G4eFh2Ta25kj5x4WTR6IdpxROmYFLoKHcBJnCyvBN4FzpObW5yY41jY/P8POMWoknKRS9xI/O1Kw7+ZFO9V+yLKNRHq+ddxxkil3/aZSSj319VmZk1J6XPYvZCQ8KcUxNdCCg7BUdO/3VeMz7GxFVBUbZrTHQSF5YWCgVEgxq1Zyi2rma46XfPh7Oq/NwjeezeWP7LEtzOaxjnllkFolGHx0N4Ym40/w/Pj4tvH737l0Mh8OULyUDXW+Tbh8fn96vqHcxPjw8b9uu+7nhipy29fX1F2U9meOk4+o76U6lhLJBpK+p84UDD5QKB9LhdKRoeOla8qRnDZxmZB/JGVWplZzbhYWnd4DpfVz8XFxcxK9//ev4wx/+EOfn56lj+GPh6OioUUaK3phd8xLos7OzGAwGxUEUzh0vzgM+low3eY2PexY5yG+XbWrLHSSXmS67XS+JPnx3yEzme9ta06hAoHSWfvN1EOz79fV12XRBx+gc6PkPD08vrZdjpAob2Unj8bg8R8/QRnBcOqIsfybnPJCWzW0mzzwDRN2hsbLkUfbd6upqXF5eFueQ63Y1Zr2ni1krPtMDsP1+v2TvZnGkIl65NToJTp0cjUYxGAxiOBzG7u5uEZbcBtM9VLaTOVCu/DlZTpQ6pvuENCHaI5sUhFn7HK/a5nVUGiRAGXdUpho3Xy6prVa5cPSnEIICldi4gsvG6ATN8fp1hCbBPc1QcpDyZ5SHAi2bp8zx9efrw6iFniFgJDW7V9eon+oXI6rigU+fPsXt7W15F8NXX31Vork1HLlhljmFTfib5sjWBBr5h7yYnfc2m57LPs/qUGTHPdvhz3XjUP+ZlfJIk5TSLIKxSTnU+iRwY83brf33cww4efSXNJMZGR54cLnNc1RYpMNptEhF7lnJVqtVDGcq94y/srYF/X5/qjxyOmi3n0uq3ZnSczzT53qI38KLX5+NJfufHXeZk80/ZY3PGeeI/6mDxBP+LZzR0RXfUPbqvIw7rv+gscXxZSU/mgv1VQa45CerUcSnvV6vZG50T22OyJecL2WN9Dw5pFw7FRETpZ8cw9u3b2N1dfVF2aSMS8kZ0p7wRBrlem7hu9VqlXHv7u6WKLvsEuG53W6XLdD/9Kc/xb/+67/G73//+4kNC0gvTjtfAqqy8BI+gWwt9VmvGiEt6rUhw+Ew1tbWJjLVNGQzmne9WJMXfj3759fU8FQDzpXarMl99tWDGx6s5z3kP7YtHHEXTFUxeVZKIBuEDhwzWcK/eEGZz263O7FnQafTievr67IPQqv1lMXSbqjcrZYBKi+fdp1JkBxyB5N63INyTGooiH19fV1wJXki2cTAB/We6NUz+LQZhDM5U4PBIDqdzkz89EUbUFCwyJnSglUhlbsmCYncOtRBx5zZKKjcKKCRoWcIwUQS783Gw2/+9mfoNzet8EicC1muB3p4eKhuZzoLk88CrhiptMW8Tswaq0MmRGpGZW0MmeNDuqAB4JFIOlgCd4j82VSKVGxOT1TSvI/XUHlrviVkuLh+PB7H1dVVvHnzppS8jsdPm7EQX25oudHm/eSYndabYBotzXo+41UaXDz2GjoRZPe44cjn+byTNiQsdR2zUvp9fX1dXWPg8+9z5fOY8Y0cC+ermtxx2ePPqxndWf8Is9KKxpEZ5k0BC3+Oj5Wyn31tcqjYJ0GT4+syjLwqpa8IqvNyk3PE62rna7TiY6rN2bQ5zAxCGh78pt6kvMycfBqIXI9A/lJpm4wTymRdGxHl3VFce0VdKPpgn3SO26Fna6G5KxnH57KdBh1BfNhqtUpQRfOm+6Sj6WgRH1wYTzoXjoQDORK1SD3HKDxzHG/fvo3d3d24v78v7yOSMfz4+BiXl5fx/fffx3/+53/G6elpwWOWlaKMapK7TXB9fR37+/vxy1/+MqVpta2xr6yslNJp0Z12G+z3+xPGfK1E2vkqGxPBecz1hvrp92ZykvYm78v66cB7mRFR254J5Zy5zCUwCKglIXRgmTXSR7JSz5RDpHV8WuetZTjKiqr/cqbUtl45oDXgKumk0yQeFu1ngT/ShWRQJk/Fs9Sf3KCC/BgRZb2hjmsskkvuhxBndKY8I6WPslweOGqCLy7zEwKUEru6uop+vz+x/bkmmY5URuB+nAIyi5wKEfzNyaWx7PcyOsT7nUFdIHv0QH3URPE8jbqI5x1E5HRye+aawPhSUH2o+qhn8NsdKRfC2ScTqrw3AzeMIybx74qBuGZkW9foPlcmPjdkSs9EcfyMDuleXqNjxI+OsTxFCuTy8rJETra2tmJra+uFos8UhhulHBd5zZ0un9tpdOTnHYfe5iwOUoabaffMck7nM6fE6ZHBExeQdKpk+GTP8Q9x3eScuBPix2rj8ufqvxuWs1zb9Az+dhzqXMbrPiaO13GVjZ/R/mljyfAyHj8FKGpAeZC1UytF8f+Zk8Xv1zhTHFs2ptqnpn8yWcTz0+bZ8cL5pdHCcTldSCZzlzBGomXgqU1+C7iDovCvEj/Su7I11KduvKkdzi0zow6SBzRyddzpQM8T/XKJgrcpB01Ohd7rSAfP7R7nWRlob968iY2Njbi/f3qBqzZ4aLVacXBwEL/73e/i7Oxswr4hZLTzpfDw8BA//PBDcaa8XeFRW+lzh2U99+HhoWyRrvJO2ktsqyYnavpsmgyp4SW7X9Ak4yImnS6/lrxJWqYDIPrz4F/Wprb5107Zo9GoOAdMDHAMWkriz1peXp54Ubk2TfEXJwsURHHe9Uy+816G58wJJT6ZrZXTQiedzgvHJntNuBW++v1+adt3JWy1nhMslGV+jNk8zZUculngi3fzI0MMBoOytale1isG02JgZgGEaBc8RLYTcsYEFIzywGXMMlXOCWQdK+uenUBFPLUMjitaGiNqm17z3d1dXF5elsxUkzL8MSDD3tOuZBz2lc+tGYZsm9+ZIeDX0jBwgePtS3lxIwF3wvQ7YtIZ0HFFhDS3vE/PdUMpM8x0DZUwo4quGCUA2+12XF5exh//+Mf4x3/8x9Q49ja8Hxlkwr9JWdT+extNkNEH260Z3K99Fvkne5b/zoxFd6SydVPcRripL5msmYXG/bvpGREv38vh/FHrh/PzLAZF1r9sPjOHqtZmTa5Q7jThLBsXoSl732pNbumdZcKkqKljXN/M6jjVfmfzkc1J0/WZU+zz5fPB+aI89ECTByCpe2kkSP5Qp3JnuvH4OfotXMlYUR+Eb63JotOk9qinGRlne/66EbbbZMy5PGBGjO2xqqTVet6xV+NWu3Ks2DZ1GZ206+vr8jwFlDP+4Dw8PDwUnCoDIPtADtUf//jHOD09ndB9GZ05vX2pQzUej+Pw8DCt/qDMeHx82s1Zm3xorbb6cXFxMRE4dl5055b3cr4yyPDq1zLj6mNgG7IX+Hxex2v1e5pDpTZZdq4MnT8jsyuls0ajUVmzI0dDdg0z0JST4sfl5eWyozblo/47TjQ/mc3MOeIxylgGvnkt55Rj5aZzkg3KAskO9CQF+YbntS5yOBxOvOfV17hrHrQMKFsOoGtI5012GeHVu/llhsxoNCqMw/frKGoVESXqRHDBQIYTwlutVlkHVFNYNEKoWG5vb0ufuZtTRLzw9DMv2g0Ov8b7T2eKW7w+Pj6WtVJamMnnNBlDXwLKTNUUvY/Lj3McmTJWv3l82hiytrjDDJWi05g/l5ES73PN2M7azSJ8TlNunLAtXadn6uWU7XY7rq6u4i9/+cuE0nCl7vRDps2uqc1ZNtam+cjaeQ0NUijOch9xVJvXrG+iQSqrmkPHiBwFJD/TnKlMadaMlaY2fBzZvTX+rBlhTUZ4E14y5e99yuanxuuODyrJGq78fh9vds943JyZovHhjpHao07InCXyWO1c9rvpU8OXj712fQ3n05S595uOhAwKGmCsuojIZY8HJtkP6WplY2iQZUEC7tabOb7MVLE9/ef6a8r/TL4SV9pVmP8z41m4V3tcC83zcupkxMpgXlhYKBF1GbMR8eLVKtStjJKrz7Kl2u127O/vx+np6YvMZDb3TedfC58+fZpYp5LJa/2+ubmJ5eXlGAwGE88fDofR7/djMBiUXddq1SSZ45LJkExOZjook+PZedJy1o9Mz4mPeEx0IvrgNaIR2YJquwkUjOeap1ar9WIz+opyiAAAIABJREFUCdIRZVmr1XqxAQtL8jIcupPUak0uldDYyX90pHhMMsfxrGfqWXou+U58LpzJn1AbWksmWSY8vX37NkajUVxfX5cdNmUH6xUFCqpqzsiD2mVb7bP/nLsm+OKX9vK30mzD4bC8SI6d5SdiclGmE6EmT0JL+9lnzOFKiNFpnVeGTCDPu0bYJCZGEjKCcMVIg44LBlVipLStL9zN+vFjQNs6ZsKwhr+Il3X1HHcm4GrgRjMFkeOTCoaRKzKarteHwopCpWawE2hg8FiTQHdDw58nEGNq4fPFxUWZa4/MufBzXOh8Ng8Or6WhzHjLnDIXhlReNSXW9Dynq5pC9XtpNPl1NBBdQDLyxCjwNKjRw6xjI9QcTpcdmaHg86RjbqxmfaZCmwaZ4zcN77zWHaqm8TpOate6IevAenq2mdGk9414zObAnTP/rTFn8+Zjyvgpm/Osvz5/xF92joFEpwvPOtFIohz153IHS42ZgUKOg3qdcry2toIykRlG9ZnGGg029cfxyvGoPzLOOH7aGeoj1zlrXOyL8BrxvLEJt0/XdTc3NxOGYTavWdCHa8/39/djYWEhrq6uJuwZzlVGX7zmx8DBwUFZM6L2sgi9aEG7vOmcbJ5+v1/sQW1ooXZq5WE1HcTfpOOM57IgKa/PguDZtX7O+0RdyKUDGqfoipUy5Jes7eXl5bJmRzv00a5ltsZpQev26UiJ/rl2L5OVnon1IJMf0/EsMMI5cNz7/HFOyJPkfe2ArQSJXrEwHj+/yHg4HJbXDtze3sbKyspEoEjj5sY2+sg24AYf6rvzXxP8qDI/EoQiEfLwbm9vJ7JARDKzVIrsKDVOoaeBuWLIiFCETMHGd1g40xGZJASlQBmZIfFKEGgsJCZvV2lLLmrTZLIvPzWw5nQWo9AVqePLDf5M4BEyA8Ojv67cqZAjJktWMny12+0JY8v74AaeM4QLzczoUv/caKPwZHvj8fMmCJmgc0PChbrj2MeVzcEs85GNmbRPpUThVlMk4jW21UTLGY6bjmX3sm98LgWlO1JM3+v3tD56nzJlnt3TdMyhNs/ZfNZ4K+NlV9DeNufbr3fc6jyVr48tkwv+bG+vCZ+89/7+Pn3BcnYto6U6Jl6bxieUbzXnyY2FLPPi81UblztltWunjblmVDfRD/lestWzWAxOeHu+JlnrOSRbKd/8Pul8X68hOciouetXRsjdgeMzXJbRQaPM5nh5H++lQyXcCT/6yJbJAnvSwT7flFMy4q6vr+P8/Dw+ffoUnz59ipubm+h2u8U24qdm//yUoBK9jY2NcszLqBkUfnx8LGVaGqtwoA0U5FTLIajxgeshnuM3oUnfTeOtGi55nweUvJ9sS3RBZ5uBHx+f2zbKyjBT4rRL20PPa7VapZRN79XTWilmhDM5RV5wh3MWWVj77/isOb/kSe+j2tNaStEQHSoGSzudTnlHlpIxGrtol3zKpQAs9WPffM+EGvwoZ4og71EEMBqNYnV1dSIKw/pgCh/fvprePYkvUx4CKgZOgHvai4uLE3vUs11GnPgMCgFOhK5xZ4rZqYiYcKqydQCvVabTQM6s48mNAXcQqFQyI7mpn7MYFGpTClfKRscENFYYbXXDT/f6x8vCpAhrCjMzHqg4M4c+i5qTsem8SdCxxMWdS+IvU8A1IfQa2nH8sH+zKB03wn0OpgFxnTlQ3k7m9GXn9PGMFB0qlQbMMjYdy4xmH08TvmrX1hTatGPZcVdSNTlJWdrUZ8dzZjgIN+RPb4P3sSxmVnkSESmPERi9rK2Ncv7y485rmYEw62fauPzaWeiJ+st5htdT92XP1XnexzapNwWcN0aL+ZG+5zb0Pt52uz1h2Ol4Vn6k692hIh44R9lcKeCpgK1omplpLxmUsUb8Erc6num0VqtVdJln/GSc+WYEMtxub2/j6uoqrq6u4vLysgSWvaqHumjWBfFfCqPRKI6Pj2NzczMi8vWoKoWSztM7NQlypmT4yo5yHe184fou++3g8k9z7kHBzMbxc96m66wMqIMiotiROuaBX/8oiK/r+YJezYHPP20b2Rcql+WaJA9guKxilqlJ/nnfaxkq4i5i8h2UPrduR/O3xsxjvV6vbO1+dXUV5+fn0e/3i3+htWKe/aVsoBz1ddXuTMlx+9mcKVfgeigXzCntxg6LuN6+fVvK9zzlX1M2zEQ4QYig2u2nxXusvWaZlfrJ+m8fgxwtZ2y2R2OZROCGnYSmhAnfCO7j+ylBO6Kw/4QmpZ0ZuTVBx3NU0DVjgc8Szhx3Ps80wrxPNBAIzFRk/7PxZd8RzzW67IfSzZlRI6ZmFETlJr5NsxtyNeeqBrNeNw2ycbij7deTHnRvZow39Z33s+2a0chMpeaTv72sz7eYrfXLn0llQf5WXzI8NdF6Rlc1mpvlu4mfa7znBmGTccA59PZ4nJ/sOP97RsH76HLl9va2yLCsb85HNASyKGzmRGVR1ddEYWtjqf2elpVyHsqyRE0gOedzRfplJlf/JaN0D+nCF4FHPJfgLy8vl1I2OrZ0utRvykDNHx1htwM8e0UHvjZXjgsH4pJ6XePPdHON3ul4ZeuLKDuEY2bIudidNkqr9WRYq8Ih28jAnWLno1nkbxM8PDzE0dFR/MM//MOEPcNAFdefKpMm+0jP51bTdLQ805bNXybjascyHiLvZAEe11XkPYEHgDiXWR9l73HdFMv92D7tJK073N/fj6WlpVL2qe3JfbyUpcKb1gUyQCE+9eUF7Lva8Gv8fIbzpk8NXGdQFjEz5rjXsYWFhSJ3Op1O9Hq9koGS49Pr9cpumOI3yhPOHwMbLMfU/PuGFE3wxZkpIkYPomcnZdhut4shI0UnoKDo9Xrl3RZUCiLGLALOY1SsOheRb3HtKT8JCXn4GfHQMGZ6m8wqJqZXrAWY+rgg/Dng8fHxxe46rhz4m0KHx5uMaN3XxEyZ0ZAZF96XGlPyWmdEv86NBr83MxYzx9MFWsTzuwwoIMkHUh6idb3ngYYDDQZ/foarDA81o6zpnoiYyNhmiiHDvQs5Hhdk+Oa9/pvnnS5dYWS0yPILBi98ndS0nfzomOl5mdLjuF4zBzU61u9aW7W2mxRWht/s2TQ2WPb0mnE09ceVsRsi09qU/GyCVqs1UTpGY5zGbOYEZesXeb7JqZrFiKjRDGXFLPKxllGkvOExltzpt88vr4mY5GHpQR13uem4lyHIUj7h13EseedOlG+/nOHF59N51G0BnfN3XglHxIdAOPP5UbuU1/pNfBGvWUYkk4GMvnNslGk0+Dz7k9HGj3WmHh8f4+TkpLSVOT90PJ1m9K1AJDdS4LtG3dFx3MxiC/gx6h9mNx3o6Gc2WU3f+di978KH/ssOVukZ25X92263YzQaFZrUC+Yz28NxRJk3Ho/LRh/iR65DzLLGeiZ5NcO9yyWXpS4r2VeC8wV1r/5n9oD7D8Jbu92ODx8+RLfbjePj4zg7O5vI3inwzbWc7JeXYjJITmdqFvhRzhRBTgkdKqX/hQym3bQFYvYGcoEG7pkK/+1RSkaLmGplOYEU1c3NTVxeXsbDw0N5cZkWSwrUL0WjRDBOCGQqOlQURoQmo+fHwHic74TlysEdqNpH5zODrsZ0mXEQMVm2QUeZis7vd8UWMVmil42f56jk3AFqAtGH2qRiGwwGVWNPZWVifEVS/N037kxRGDGgME2gZfPThBNX8oRsDvWbxl3NGZuGU+dh9s3lhAt3zr2vY2MEkI7VrM5UzVDz+cnw5OOr/c7mqMYvTXw0je5rAQZdE5HvYNXE7zVoohf+Jt830Yr6rzWm2Rj1rXnOMhl8Ri2Lkcms137cAW3CgfNvJlNfy0+Om9oz3NnInOiIl2uS2Aca07xf8kxrNdg+s/GSl8xeue7OApnu8Loxl+HBcUiZq6oYBl7phEkPcawuj2kD8DhpMXNA9TwGYvXhq11kTzHb7iVIPyd8/vx5YjdiZjApm6kP9F9jlSPFMVDv1fQAj+m3z7PbBD4Xmu+IyYA65QdpkIE5v5agvmhutPvi2dlZnJ6exsnJSRweHsbp6Wn88MMP5ffl5WVZO1ZzdkV3LI1054vGf8bvEc9rFz2oRDyS9vnJghr+Pws8+X3sV81ByuYuu1bXaYMNztHj41NZ7Pr6eqyursbm5macnp6WRAadJr7MXUC+Er2qbdHOz1rmVwN5gZxwCloJW3WMaUg6QO12u+xEwnIPNwDITCojEGPw+VK4MmYFd3d3MRgMymJRCVlF3EQUZNIsG1FjwCbi+Lmh3++/yJyxT1nJW9bfJmXeZFTwvNp25uRxKhgK6ZrCjJgsWald42On01ZzCHRMZat0iuVIaaclxyHvlePe6XRerJnyLYIzQ74J59PGWhu/t1WjSzcCHEdUVP5Mf17WX2+TdKk+Zw4zr6GRw1IYOlVypJo2n/DsRmZITjuWjdP/1wxo/s6eMUvbrwUqCxqwNXnlc+B9e82YsnscL9PWhrCfNGQzIz0zCtwgqH2arqs53RnemoyTGn4j8khuE/97e3699KHLV47J2/EAj3iQm+0wEk7nhO+1kR7O8EJ9ShnjjnDtU8MjsybMVHjwlbghTRGvrhdbrdaEPeEZAMoqXkO5xbZJt3Sg+J2VIvrvnwKOjo6K7uMmWu4giS7u7u5eLNOQgaoX9zJI5nMnOTSNT0gf/t5Q6lG16eD6y2lOeke27PX19YSzdHR0FPv7+/HDDz/E8fFxnJycxPn5eVxdXZVsUpOuaQL1XzapB5M0TgbmnU6dth2H02QdnazaPDTxn/pRA7ed9d3kTGlOst/UAXd3d7G5uRmrq6uxsrISf/jDHwqtir/Es4434lYg2p+Vt6Y6U7M6AeyYOj8eP79oLCLKC974IjHdKwOaEQWPGHg0hAhhVExM60KZztx4PC6bZnBxurzUbKId+TTaGCnX8ylA9Pv/L6eq3++XbbpdYPkYMoM7M6Yy4d2k0GpGjIDMXjPWecz74yWgTeBGxbSoldb90WAQc11cXJRohfdf+ByNRsWhPz8/L1Gs9fX12NjYKOsJKcSyiE8TjjMDzA2BGu6a/ntbGU24IVB7VobbGmTnmJWmE8XrqWDIb1zkPY0+phlp04IF0wybjCeyazOlVWuraSwReZTV50Lyko5ekwzQcf/tyinDTRNPO/609rYGVPYcG49nhkXNMJ/VucquZf+nyUF91+Z/Gi81QSYP/JiPg3qVOpPX18Yl3lLVB9eFjsfjsvhdO95RtwtqGSk3zHku+9TG6zJD0X71kZtKEW86nxnkDCARj1w7xWu9b7SR3ICTs+WOlIx0byujk5/Cxri8vIzhcBhLS0sTdszj49MufaenpxPPl5Mkm07jUkBSdqB2nKPz2BQg8ON6hmirxm/S4VwfT7xoLXu/3y+O0uHhYRwdHcXBwUF8/vw5jo6O4uzsLK6urqLf70+UK/5cINthNBq9kH+iEzr8Ec+bXbBstuZUZb+ZlcrkoON2FtnnOiAbZ8bbtPd5f41OOBZubvPu3bsYjUbxpz/9qfBQp9MpskV4pG1HG159EF5ngZ+0zE9ES+ZrtVoTxs39/X0sLy+XDvIaDVSeo4xXIU3fmZEppDAqkBnaEoSPj48TizmJUDILBbKn5vlMCTo36rheypmwSYj8WJAQW11dfXHOvX83oiKaDePs4/dNY0JX7Jor/Wb0tAY1BnMjwJ1GtU/B4evgqARY3nd7exuXl5cvtkB3nEZEEYaDwSAGg0EcHBwUpl9ZWYm1tbXY2tqKvb292NnZifX19eh2u9VtTGvg85YJnBo0KWOnCT8/jV5oPNT6TT7K5lG4585cvDdicqejjP9mVX4Z3WTyJvud4eA1/5v4xsHvq+GXY+AYvR2Pkuu6mjPFdlwGZM6V9zczgt1Bm2a0MKsgvUEDgc9yw8KzvzUZ5Z9acCOjidq4/Xp9uxHvOCaOnD+bZLcf5/y4PKSMzOS740J6VLLUy/e4PsbvZRlSVu6X4b5J92Ryj06a+iacMJjm+kH6IOMB4jFzbDLcM2Cn41lwlQ6WgglypDJ+mKYXfgzIYfr6668nbKObm5uizzJ9Lj6UfmS25u7urmyU0OTw1Xgv21qfeNCcyN4aDodxcXERx8fHcXh4GPv7+3FwcBCHh4dxeHhYskp6pY87rP8bINmX7T7rpe2EVmtyd0yXh5SDfjzL5Dfxl5+jE5bxDPnO7TqXZZR/kisM9rmOJt40Fp37+uuv4+rqKg4ODsrc6p1oETGxIYx4LMP5rPbDzM5UzahyBaxvpnQ1OF8kLiMpM1bYfobA2iTr2RHPL81TuR/PMU1NYBtSFjzHfmWRcHeo3JlyZvgxjNt0r95f4WPz/jcZTJlx0MRUDrzOI8auaDztrn7OojC8VGlaf9g+x682lJU6OzuLk5OTuLq6iouLizg/P4/Ly8u4uLgotbWvASlRZanOzs7iT3/6UzEo3r59G91uNzY3N2N3dzd2d3dja2srNjc3o9vtTkR+s7bdkJw2fh7PDAO23WSgNbXt7bMNp0HRifOc+IgOFa9RW8wA0/nNossERubUBx33Mc7yOxt/7X/2u3a+6TqCOzo81nSfP2OawePftblnXzIZkvG5SoMcFhYWYmtrKx4fH0tZd7bZQdY/NyRmicLWxlHDQW2up33PAn5t5lz58ewY28tkPPmSGSt9hHPy46dPn2IwGMTa2lqsr6/H9vZ27O3txfv372NjYyNWV1dLoEhGDzesIP/5mlLXH1nfMzmupQUuG13nSGZQP2XOpZ7DT63sLpN3/O04lhMgOeZldU3Bhabnfinc39/H8fFxvH//fsKGUdlepm+IK41JAUkGtkg3GX/qwyUiS0tLE7i5vr6OwWBQnKWDg4M4ODiIo6OjUoJ3enoaV1dXZfMHBkD/GkE4UCC3RkOZvKEO00fH1W6WpWrarCezJ/g89iGzAzLZUpORmQMmPc/gB21zPYP8Qxui0+nEx48f4/r6urwAeW1t7cXGO/RXHN8/aZkfG/b/bqR4ulC/Vdbnuy5l16ttta91U2pf5/mbhKFtEiOipJTFmKyxHY1GhVHFpDJo2R8KVK6/kgHPiIkMOAlxlTj4rip0JGoEluH8NecfHh6i3++nyjObu2mGkB+fZjjUmIvX8pmehao9u+ZguTL1cdNxEpMq0qZUv6JXimCdnp6Wcr0s4vJTgYx/reE7ODiI3/72t4W+lpaWotfrxdbWVuzs7MTu7m7s7e3FxsZGdDqdQmdNc+I4ctzwP+kkMyay895eJiD9ea4c+Hy/xjeh4TMkcBlNZlZ42ryxNMIVVI2ma4rCx5ONr3aO/91ImeUe4Svjj6w9yoHsGdP4rCYXXntd9pyaE9xqteKf/umfYjAYlHUaNLpJn7PIkGxOs7nOfjcd8z43tVWDLBqbQXbej3kWXwaH+sFMjes+9Zf6WuuK5Qzd3NwUg5bXLy4uxvLycnQ6ndjY2Ijt7e3Y2dmJvb29+Oqrr2Jvby+2trZibW1tYvOnWkTd5yrDs67l5gIs7dFv0Y3sBcmYjD5qx9SmyzI32r26xe9ncFcZKW6j3rRe6ueA8XgcZ2dnL47p2W6cMzBNHvQtpzUGvluUOONYb25u4vz8PE5OTuLz58+xv78fh4eHcXx8HEdHR3F1dRWDwaBkHbxk8m8NZJdcXl6mwVp3wjNdRV3Gdn3OfP78t+7LEh1NMtLHk+lJ9o28kAUqlATRNe4naMzOUxFP/L2+vh5ff/11/PnPf46IKAEcyr6abSe6zoLXGcy0ZirzzpxoFxcXo9PpxOrqaolCaQ94/V5dXZ1wLvimZv32us3l5eUU+WRc98T5AkH9p7Ol6Mjy8nJBupyp1dXVFxkACj3Wo+p5ehEwF+ByopRy7na7JUrCdn8OEFM2CZfMocoMXELNwMoMiWlGTFYfy3mlcHQjMOuD/yb+tZBUtdGfPn0qkayzs7OJrUv/WiJXzLAMBoPY39+PiCgCZHl5uWSydnZ24t27d7G7u1syWcvLy2Xxt+7zOW5yogmZw6XjNWPO2+Sza32QsM1ok/XvGZ14/fOsJX4KqmRCv8no9uO89zXHHE/+/zWGU4bPGr84f9WM09pzph3jXPNcTQ7wntFolMouRRy//fbb+M1vfvPCgJqm8JuO8dOUvco+/uwMN7Xr+TsLRDju/HiNX7x9OUrMNLlu5XHRiH7T4OJLQWvyQg7Czc1NXFxcxOfPnyf6I92skueNjY3Y3d2N9+/fx1dffRXv3r2Ld+/exfb2dqyvr0en0ykyzQ0cNywZpY94LpGSA05jUuezDBxxns0D583nx41DX/OkOZHDpHNcY8SX/vocZ/z2U+mu8XhcHGMPVFHOsi9erq21PFy/OhwOYzgcRsQTj19cXJRd8E5OTuL4+Lhs6KBNHf7aM0o/JWhb9ExvsZTdHfMs6x4x6YBkyYssE8VSuUxeZr8FTbJL/fRgqP92ucNMEv+znzwuB0iB6O3t7VJNJB9A+oX2Qmb/ivZngZkzU5lRSwT1er0SMd/Z2YmdnZ1YW1uLt2/fxtraWnS73VhZWSnRIL17Z2VlpThSTPkzu7CyslIQLo+Sk+I1okICFbWIREhaXl4u2ykqla4smvrHyZfCEuLVHg03tbOyslI+Ev7j8Tiur69jNBrF6enpC6H6GoNJUDOUBHKmWAZHx1DfNUfEn1VTmrMYCZxXn79a+4yWsm8ZjMdPJQUqnzs7OysLST9//hzHx8dlUe2sL2H7awUp5n6/H/1+vzhZEVGcrF6vV8oFZZBsbW1Ft9st74YRuGGSCTqCz9VraTdT+jxG/qfx51EnX+9AQayd/HxRaQb//d//XV4AqICKvhV04QJ7Ly9rMqqbDO0m52Uab2eGlBt2fj0NPo+QN/Ul6/80pUn6YR81p9OgZlCMx+M4OTmJd+/evaiAYAAtkzuZ4aBr/HjNuZrmSE3DXXZPzTDntTVerD3Ts/O8xnVazWjKxhnxvEaC+H4t0NliybOPQa8rcWfrw4cP8eHDh9jZ2Sml0L1eb2L3QLXhRqiOawykZc43daZwT5xFTL6vLaNt8ljNAeFGCVy/QSeEvPCl9sJr4fDwcOIF9HJqfLlC5mSx8uPXv/51/M///E88PDzE9fV19Pv9slud2v9b1sc/FWj+tYuigLYsNzvjx50ptef03JSZcl7PjtXsVZd/Nfpkn3VtRksCL4XNnkufQPwkPnzz5k30er149+5dHBwcTGSfvd887v2dBaY6U6pvzdYo6CGrq6vx8ePH+OUvfxkfP36M9+/fx/b2djHctFWhnAp5h+12O5aXl8uGFF7u5wqbCKCyIGG40uFvEYfSh3pJsCZAbdMB4WLaiOcdjBjJ5oYb2op5OByWLbRlxCqKJ6/YFT/BhXmWnRsOhy/WRRH0LiQanQI3prJ0bg0ctxmua0zI5/C9Ae7sOS7oUN3e3r4ozfv06VPs7+9PvNPhf3sx6f8GPDw8lAXC+/v78Zvf/CYiojgFeheDAh/v3r1LM1ku7DKh02TcCTKnYFrbFIwRURyqjP+puF2Zz5pp/Pd///f4j//4j0KjDPi0Wq3iPElOraysxOLiYtmZVGVMKysr0el0YmlpqfB7r9eLxcXF6Ha7E46aAktybJk9zwx0x08Nt01z4A6OG9wa/6zOVe0ab9/1heaIypr3icdrcHZ2Fp8+fYqIl+9UyXayajIa3Mma5jjN6lD5/5pjkl2X4ctx5Ndn5zPayIIPdGyY4ZU+U8kl39mmwKGXn/2UoFKvm5ub1NkS36ysrJQy6O3t7fjqq6/iw4cP8e2338Y333wTGxsbsba2VipjpHvchmCAyZ3YzMF1A69JLxLPvIfOSavVmti5jyV+s6yX0rN+Kri4uCgvk1Xftf5EaxrFy3wnlge5Dg8Pf7I+/V+Gh4eH+N3vfhdXV1fp+p2bm5s4Pj6OiKcM1vn5eayvrxe9vbKyUvSM1pmJ1qVn/DUg4gOWv0bkAaDMnnNa5/2Zw8LkBsdGhzEiXti7DP6Q76hLFhYWJgITEc92z9bWVgwGg8bNS2q276zybaoztba2FktLSxPrRjQAMc8vfvGL+Od//uf47rvv4sOHD7G9vR1ra2sloqt3RnmKMSJeOBSM3tQUiCsMRtd88C4EMwXH+nwKA06aMhqs/9X9ul5pa89U3d3dxeLiYvR6vdjb24uTk5OJbFFtnIyAkXgFKl+rgRw61ZTSUFOpo9pkfbk+bqj4Iu+ISP9nv2mkat6FO7XhES8tNNWi0k+fPsX3338fBwcHcXJyUnbh+X/RaXotMJOlNVkStHKyNjY2ioO1u7sbOzs70ev1qk69l2nqdwZuSGdGXyZbdK2MNjpajNzKAOFmMK8pp6XBJEXW9KLfaeDGOh20drtdFJoctE6nU0ql5YwtLS1Fp9Mpx1ZXV8tcvX37tmTSut1utNvtsvUr333j48t2ESOOphn608bs7fnzeV0W+dfv8/PzqhL7y1/+Eufn5/Htt99OGHPccER4V2m2FGzmXGXOVs35mtWZyhQzdZXg8fExhsPhC35QQM75RjuJErTuk/w4Hj+9uJ1OqWQq35GndrX1M4+rvJjGsRwc8Zt+/2+AZ7a0LkKgMkSVQ29sbMTOzk589dVX8d1338XHjx/jw4cPsbe3VypJSIMRk7v9Cdw5YpkedTRLjb1s7/Hxaac7OSeiWzkq+siB9cwU+/JzwWAwiH/5l3+J6+vrsuudduebZ5N+enh8fCzOUgailc+fP8evf/3rCfkV8bzkJOJ5HbD0gTZjk9MV8bRBy+rqatFLvV4vWq1WWaPdarWKnvHfvF7BQtl4nU4nIqJcJ6AMcXvd5RF/cz0cy7r9GHmM/Hl7exutVqvwd/bKAedz2jSzQquJIVqt1lgGNB9CGI+f3yfhLyJ1RyBpv/bcVw2iCaYhRQj3Yw6zROCY4fL23FCc1cD7MdBqTa45czz48dp3du/Q6MR7AAABOklEQVS04z/mXERMKKS/lm1L/18BCUHxtEdvBD/1XDQFTfifBktm9DBb9X+BXmi00/CnsmSwYpq8+1uA4XAY/X6/et5L+6aVWzr8mGNfck1EPVOU0anod9ZrM92UHftbmf+fG8Q3Wm6g4IbDNHzV7KLa+SwQ7M4bM+3/l+TYHP52ILPdpWP0m7sy8vcs1RVNUJOT0441+SgRL5cqkc/4PaVv1QE1OlNzmMMc5jCHOcxhDnOYwxzmMIccXr9ydA5zmMMc5jCHOcxhDnOYwxzmMHem5jCHOcxhDnOYwxzmMIc5zOFLYO5MzWEOc5jDHOYwhznMYQ5zmMMXwNyZmsMc5jCHOcxhDnOYwxzmMIcvgLkzNYc5zGEOc5jDHOYwhznMYQ5fAHNnag5zmMMc5jCHOcxhDnOYwxy+AP4/7FjcoG7ItK8AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 1080x1080 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3tldU6HAmoOm",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def one_hot_label(inp, num_classes):\n",
        "    # 0: Covid class, 1: Normal class, 2: Pneumunia class\n",
        "    vector = np.zeros((len(inp), num_classes))\n",
        "    for i, label in enumerate(inp):\n",
        "      if label == 0:\n",
        "        vector[i][0] = 1\n",
        "        vector[i][2] = 1\n",
        "      elif label == 1:\n",
        "        vector[i][1] = 1\n",
        "      elif label == 2:\n",
        "        vector[i][2] = 1\n",
        "    return torch.FloatTensor(vector)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FLhylykhzkZB",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def get_model(name, pretrain):\n",
        "    if name == 'vgg16':\n",
        "        model = models.vgg16(pretrained=pretrain)\n",
        "        \n",
        "        num_features = model.classifier[6].in_features\n",
        "        features = list(model.classifier.children())[:-1]\n",
        "        num_features = model.classifier[5]\n",
        "        features = list(model.classifier.children())[:-2]\n",
        "        num_features = model.classifier[4]\n",
        "        features = list(model.classifier.children())[:-3]\n",
        "        num_features = model.classifier[3].in_features\n",
        "        features = list(model.classifier.children())[:-4]\n",
        "        num_features = model.classifier[2]\n",
        "        features = list(model.classifier.children())[:-5] \n",
        "        num_features = model.classifier[1]\n",
        "        features = list(model.classifier.children())[:-6] \n",
        "        num_features = model.classifier[0]\n",
        "        features = list(model.classifier.children())[:-7]\n",
        "\n",
        "        for param in model.features.parameters():\n",
        "            param.requires_grad = True \n",
        "\n",
        "        n_inputs = 25088\n",
        "        model.classifier = nn.Sequential(\n",
        "            nn.Linear(n_inputs, 2048), nn.ReLU(), nn.Dropout(0.55),\n",
        "            nn.Linear(2048, len(class_names)))\n",
        "\n",
        "    elif name == 'resnet18':\n",
        "        model = models.resnet18(pretrained=pretrain)\n",
        "        \n",
        "        features_num = model.fc\n",
        "        resnet_feature = list(model.fc.children())[:0]\n",
        "\n",
        "        for param in model.layer1.parameters():\n",
        "            param.requires_grad = True\n",
        "        for param in model.layer2.parameters():\n",
        "            param.requires_grad = True\n",
        "        for param in model.layer3.parameters():\n",
        "            param.requires_grad = True\n",
        "        for param in model.layer4.parameters():\n",
        "            param.requires_grad = True\n",
        "        for param in model.fc.parameters():\n",
        "            param.requires_grad = True\n",
        "        \n",
        "        n_inputs = 512\n",
        "        model.fc = nn.Sequential(\n",
        "            nn.Linear(n_inputs, 2048), nn.ReLU(), nn.Dropout(0.50),\n",
        "            nn.Linear(2048, len(class_names)))\n",
        "\n",
        "    # get gpu\n",
        "    if gpu: model = model.to('cuda')\n",
        "    else: model = model.to('cpu')\n",
        "    \n",
        "    return model"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BvK4u05Mq1zQ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def plotCurves(stats):\n",
        "    plt.figure(figsize=(8, 6))\n",
        "    for c in ['train_loss', 'valid_loss']:\n",
        "        plt.plot(stats[c], label=c)\n",
        "    plt.legend()\n",
        "    plt.xlabel('Epochs')\n",
        "    plt.ylabel('Error')\n",
        "    plt.title('Training and Validation Loss Curve')\n",
        "\n",
        "    plt.figure(figsize=(8, 6))\n",
        "    for c in ['train_acc', 'valid_acc']:\n",
        "        plt.plot(100 * stats[c], label=c)\n",
        "    plt.legend()\n",
        "    plt.xlabel('Epochs')\n",
        "    plt.ylabel('Accuracy')\n",
        "    plt.title('Training and Validation Accuracy Curve')\n",
        "    plt.show()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Di5VNium6WO1",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "class Focal_Loss(nn.Module):\n",
        "    def __init__(self, alpha=0.25, gamma=2, logits=False, sampling=None):\n",
        "        super(Focal_Loss, self).__init__()\n",
        "        self.alpha = alpha\n",
        "        self.gamma = gamma\n",
        "        self.logits = logits\n",
        "        self.sampling = sampling\n",
        "\n",
        "    def forward(self, y_pred, y_true):\n",
        "        alpha = self.alpha\n",
        "        alpha_ = (1-self.alpha)\n",
        "        if self.logits: y_pred = torch.sigmoid(y_pred)\n",
        "        \n",
        "        pt_positive = torch.where(y_true==1, y_pred, torch.ones_like(y_pred))\n",
        "        pt_negative = torch.where(y_true==0, y_pred, torch.zeros_like(y_pred)) \n",
        "        pos_ = (1-pt_positive) ** self.gamma\n",
        "        neg_ = (pt_negative) ** self.gamma\n",
        "        \n",
        "        pos_loss = -alpha * pos_ * torch.log(pt_positive) \n",
        "        neg_loss = -alpha_ * neg_ * torch.log(1 - pt_negative) \n",
        "        loss = pos_loss + neg_loss \n",
        "        \n",
        "        if self.sampling == \"mean\": return loss.mean()\n",
        "        elif self.sampling == \"sum\": return loss.sum()\n",
        "        elif self.sampling == None: return loss "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "V9SU5JOly-w6",
        "colab_type": "code",
        "outputId": "85d888cb-7482-4303-ed03-7051ea2b6113",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "# Check if gpu is available\n",
        "gpu = cuda.is_available()\n",
        "print('Gpu is available:',gpu)"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Gpu is available: True\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wbpizTHi0Wn8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 103,
          "referenced_widgets": [
            "2b325b49a2f0407fa00458683167ff51",
            "4bde53f852ae424a803ab83bb0e5f421",
            "0e2fa1fd52af4a60ad9ad9bfc218de99",
            "586de0652cc046eab899762fe7b38a26",
            "cd947ad8db60450580ec5703d5972a92",
            "e86463d34cd24bd7810a4563e7b79c3f",
            "513872c4b6264b38b53a85addf9dac6b",
            "3bf0dfa11eac40adba4490fe1b4e62f2"
          ]
        },
        "outputId": "d0309660-8d0b-4a4f-8928-b2cdc012b4fc"
      },
      "source": [
        "model = get_model('resnet18', True)"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading: \"https://download.pytorch.org/models/resnet18-5c106cde.pth\" to /root/.cache/torch/checkpoints/resnet18-5c106cde.pth\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "2b325b49a2f0407fa00458683167ff51",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(IntProgress(value=0, max=46827520), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PwOsCD1A6Zzq",
        "colab_type": "code",
        "outputId": "b4bdd3cf-ff50-46aa-c975-41b88e01cd09",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "from torchsummary import summary\n",
        "summary(model, (3, 224, 224), batch_size, device='cuda')"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "----------------------------------------------------------------\n",
            "        Layer (type)               Output Shape         Param #\n",
            "================================================================\n",
            "            Conv2d-1        [120, 64, 112, 112]           9,408\n",
            "       BatchNorm2d-2        [120, 64, 112, 112]             128\n",
            "              ReLU-3        [120, 64, 112, 112]               0\n",
            "         MaxPool2d-4          [120, 64, 56, 56]               0\n",
            "            Conv2d-5          [120, 64, 56, 56]          36,864\n",
            "       BatchNorm2d-6          [120, 64, 56, 56]             128\n",
            "              ReLU-7          [120, 64, 56, 56]               0\n",
            "            Conv2d-8          [120, 64, 56, 56]          36,864\n",
            "       BatchNorm2d-9          [120, 64, 56, 56]             128\n",
            "             ReLU-10          [120, 64, 56, 56]               0\n",
            "       BasicBlock-11          [120, 64, 56, 56]               0\n",
            "           Conv2d-12          [120, 64, 56, 56]          36,864\n",
            "      BatchNorm2d-13          [120, 64, 56, 56]             128\n",
            "             ReLU-14          [120, 64, 56, 56]               0\n",
            "           Conv2d-15          [120, 64, 56, 56]          36,864\n",
            "      BatchNorm2d-16          [120, 64, 56, 56]             128\n",
            "             ReLU-17          [120, 64, 56, 56]               0\n",
            "       BasicBlock-18          [120, 64, 56, 56]               0\n",
            "           Conv2d-19         [120, 128, 28, 28]          73,728\n",
            "      BatchNorm2d-20         [120, 128, 28, 28]             256\n",
            "             ReLU-21         [120, 128, 28, 28]               0\n",
            "           Conv2d-22         [120, 128, 28, 28]         147,456\n",
            "      BatchNorm2d-23         [120, 128, 28, 28]             256\n",
            "           Conv2d-24         [120, 128, 28, 28]           8,192\n",
            "      BatchNorm2d-25         [120, 128, 28, 28]             256\n",
            "             ReLU-26         [120, 128, 28, 28]               0\n",
            "       BasicBlock-27         [120, 128, 28, 28]               0\n",
            "           Conv2d-28         [120, 128, 28, 28]         147,456\n",
            "      BatchNorm2d-29         [120, 128, 28, 28]             256\n",
            "             ReLU-30         [120, 128, 28, 28]               0\n",
            "           Conv2d-31         [120, 128, 28, 28]         147,456\n",
            "      BatchNorm2d-32         [120, 128, 28, 28]             256\n",
            "             ReLU-33         [120, 128, 28, 28]               0\n",
            "       BasicBlock-34         [120, 128, 28, 28]               0\n",
            "           Conv2d-35         [120, 256, 14, 14]         294,912\n",
            "      BatchNorm2d-36         [120, 256, 14, 14]             512\n",
            "             ReLU-37         [120, 256, 14, 14]               0\n",
            "           Conv2d-38         [120, 256, 14, 14]         589,824\n",
            "      BatchNorm2d-39         [120, 256, 14, 14]             512\n",
            "           Conv2d-40         [120, 256, 14, 14]          32,768\n",
            "      BatchNorm2d-41         [120, 256, 14, 14]             512\n",
            "             ReLU-42         [120, 256, 14, 14]               0\n",
            "       BasicBlock-43         [120, 256, 14, 14]               0\n",
            "           Conv2d-44         [120, 256, 14, 14]         589,824\n",
            "      BatchNorm2d-45         [120, 256, 14, 14]             512\n",
            "             ReLU-46         [120, 256, 14, 14]               0\n",
            "           Conv2d-47         [120, 256, 14, 14]         589,824\n",
            "      BatchNorm2d-48         [120, 256, 14, 14]             512\n",
            "             ReLU-49         [120, 256, 14, 14]               0\n",
            "       BasicBlock-50         [120, 256, 14, 14]               0\n",
            "           Conv2d-51           [120, 512, 7, 7]       1,179,648\n",
            "      BatchNorm2d-52           [120, 512, 7, 7]           1,024\n",
            "             ReLU-53           [120, 512, 7, 7]               0\n",
            "           Conv2d-54           [120, 512, 7, 7]       2,359,296\n",
            "      BatchNorm2d-55           [120, 512, 7, 7]           1,024\n",
            "           Conv2d-56           [120, 512, 7, 7]         131,072\n",
            "      BatchNorm2d-57           [120, 512, 7, 7]           1,024\n",
            "             ReLU-58           [120, 512, 7, 7]               0\n",
            "       BasicBlock-59           [120, 512, 7, 7]               0\n",
            "           Conv2d-60           [120, 512, 7, 7]       2,359,296\n",
            "      BatchNorm2d-61           [120, 512, 7, 7]           1,024\n",
            "             ReLU-62           [120, 512, 7, 7]               0\n",
            "           Conv2d-63           [120, 512, 7, 7]       2,359,296\n",
            "      BatchNorm2d-64           [120, 512, 7, 7]           1,024\n",
            "             ReLU-65           [120, 512, 7, 7]               0\n",
            "       BasicBlock-66           [120, 512, 7, 7]               0\n",
            "AdaptiveAvgPool2d-67           [120, 512, 1, 1]               0\n",
            "           Linear-68                [120, 2048]       1,050,624\n",
            "             ReLU-69                [120, 2048]               0\n",
            "          Dropout-70                [120, 2048]               0\n",
            "           Linear-71                   [120, 3]           6,147\n",
            "================================================================\n",
            "Total params: 12,233,283\n",
            "Trainable params: 12,233,283\n",
            "Non-trainable params: 0\n",
            "----------------------------------------------------------------\n",
            "Input size (MB): 68.91\n",
            "Forward/backward pass size (MB): 7539.85\n",
            "Params size (MB): 46.67\n",
            "Estimated Total Size (MB): 7655.42\n",
            "----------------------------------------------------------------\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1eKex2OA9Gd2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# hyper-parameters initialization\n",
        "Epochs = 300\n",
        "#criterion = nn.BCEWithLogitsLoss()\n",
        "FL = Focal_Loss(alpha=0.25, gamma=2, logits=True, sampling='mean')\n",
        "optimizer = optim.SGD(model.parameters(), lr=0.001, momentum=0.9)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xzQnukEA9OxO",
        "colab_type": "code",
        "outputId": "59fbf55e-2817-46c7-ef3f-221a599f309e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "device = torch.device(\"cuda\")\n",
        "\n",
        "model.to(device)\n",
        "model.train()\n",
        "stats = []\n",
        "\n",
        "print('Training Started.......')\n",
        "for epoch in range(Epochs): \n",
        "\n",
        "    train_acc = 0.0\n",
        "    valid_acc = 0.0\n",
        "    train_loss = 0.0\n",
        "    valid_loss = 0.0   \n",
        "    train_pred, train_orig, valid_pred, valid_orig = [], [], [], []\n",
        "\n",
        "    # Training dataset pass\n",
        "    for i, data in enumerate(trainloader):\n",
        "        # get the inputs\n",
        "        inputs, labels = data\n",
        "        labels = one_hot_label(labels, len(class_names))\n",
        "        inputs, labels = inputs.to(device), labels.to(device)\n",
        "\n",
        "        # zero the parameter gradients\n",
        "        optimizer.zero_grad()\n",
        "\n",
        "        # forward + backward + optimize\n",
        "        outputs = model(inputs)\n",
        "        out = torch.sigmoid(outputs)\n",
        "        #loss = criterion(outputs, labels) \n",
        "        loss = FL.forward(outputs, labels)\n",
        "        loss.backward()                    \n",
        "        optimizer.step()                 \n",
        "\n",
        "        # print statistics\n",
        "        train_loss += loss.item() * inputs.size(0)\n",
        "\n",
        "        # Calculate train accuracy and thresholding\n",
        "        thresh = Variable(torch.Tensor([0.50])).to(device)\n",
        "        one = Variable(torch.Tensor([1])).to(device)\n",
        "        zero = Variable(torch.Tensor([0])).to(device)\n",
        "        predicted = torch.where(out > thresh, one, zero)\n",
        "        \n",
        "        correct_tensor = predicted.eq(labels.data.view_as(predicted))\n",
        "        accuracy = torch.mean(correct_tensor.type(torch.FloatTensor))\n",
        "        train_acc += accuracy.item() * inputs.size(0)\n",
        "\n",
        "        train_pred.extend(predicted.cpu().numpy())\n",
        "        train_orig.extend(labels.cpu().numpy())\n",
        "\n",
        "    train_loss = train_loss / len(trainloader.dataset)\n",
        "    train_acc = train_acc / len(trainloader.dataset)\n",
        "\n",
        "    # We don't need to save grads because no backprop is required for validation set\n",
        "    with torch.no_grad():\n",
        "        model.eval()\n",
        "        # Validation dataset pass\n",
        "        for i, data in enumerate(validloader):\n",
        "            # get the inputs\n",
        "            inputs, labels = data\n",
        "            labels = one_hot_label(labels, len(class_names))\n",
        "            inputs, labels = inputs.to(device), labels.to(device)\n",
        "\n",
        "            # Validation loss\n",
        "            outputs = model(inputs) \n",
        "            out = torch.sigmoid(outputs)\n",
        "            #loss = criterion(outputs, labels)\n",
        "            loss = FL.forward(outputs, labels)\n",
        "            valid_loss += loss.item() * inputs.size(0)\n",
        "            \n",
        "            # Calculate validation accuracy and thresholding\n",
        "            thresh = Variable(torch.Tensor([0.50])).to(device)\n",
        "            one = Variable(torch.Tensor([1])).to(device)\n",
        "            zero = Variable(torch.Tensor([0])).to(device)\n",
        "            predicted = torch.where(out > thresh, one, zero)\n",
        "\n",
        "            correct_tensor = predicted.eq(labels.data.view_as(predicted))\n",
        "            accuracy = torch.mean(correct_tensor.type(torch.FloatTensor))\n",
        "            valid_acc += accuracy.item() * inputs.size(0)\n",
        "\n",
        "            valid_pred.extend(predicted.cpu().numpy())\n",
        "            valid_orig.extend(labels.cpu().numpy())\n",
        "\n",
        "        valid_loss = valid_loss / len(validloader.dataset)\n",
        "        valid_acc = valid_acc / len(validloader.dataset)\n",
        "\n",
        "        model.optimizer = optimizer\n",
        "        stats.append([train_loss, valid_loss, train_acc, valid_acc])\n",
        "    \n",
        "    print('\\nEpoch',epoch,':',f'Training Loss: {train_loss:.6f},',f'Training Accuracy: {100 * train_acc:.3f}%')\n",
        "    print(f'\\t  Validation Loss: {valid_loss:.6f},',f'Validation Accuracy: {100 * valid_acc:.3f}%')\n",
        "    f1 = f1_score(np.array(valid_orig), np.array(valid_pred), average='micro')\n",
        "    print(f'\\t  F1: {f1:.6f}')\n",
        "\n",
        "    # Plot Confusion matrices after some epochs\n",
        "    if epoch % 7 == 0:\n",
        "        train_cm = multilabel_confusion_matrix(np.array(train_orig), np.array(train_pred))\n",
        "        valid_cm = multilabel_confusion_matrix(np.array(valid_orig), np.array(valid_pred))\n",
        "\n",
        "        fig, ax =plt.subplots(1,3)\n",
        "        plt.figure(figsize=(4,5))\n",
        "        fig.tight_layout(pad=3.0)\n",
        "        fig.subplots_adjust(left = 0.025, right = 1, wspace=.7)\n",
        "        sns.heatmap(train_cm[0], annot=True, cmap='Blues', fmt=\"d\", ax=ax[0])\n",
        "        sns.heatmap(train_cm[1], annot=True, cmap='Blues', fmt=\"d\", ax=ax[1])\n",
        "        sns.heatmap(train_cm[2], annot=True, cmap='Blues', fmt=\"d\", ax=ax[2])\n",
        "        ax[0].set_title('Covid (Train)')\n",
        "        ax[1].set_title('Normal (Train)')\n",
        "        ax[2].set_title('Pneumunia (Train)')\n",
        "        plt.show()\n",
        "\n",
        "        fig1, ax1 =plt.subplots(1,3)\n",
        "        plt.figure(figsize=(4,5))\n",
        "        fig1.tight_layout(pad=3.0)\n",
        "        fig1.subplots_adjust(left = 0.025, right = 1, wspace=.7)\n",
        "        sns.heatmap(valid_cm[0], annot=True, cmap='Blues', fmt=\"d\", ax=ax1[0])\n",
        "        sns.heatmap(valid_cm[1], annot=True, cmap='Blues', fmt=\"d\", ax=ax1[1])\n",
        "        sns.heatmap(valid_cm[2], annot=True, cmap='Blues', fmt=\"d\", ax=ax1[2])\n",
        "        ax1[0].set_title('Covid (Val)')\n",
        "        ax1[1].set_title('Normal (Val)')\n",
        "        ax1[2].set_title('Pneumunia (Val)')\n",
        "        plt.show()\n",
        "\n",
        "    stat = pd.DataFrame(stats, columns=['train_loss', 'valid_loss', 'train_acc','valid_acc'])\n",
        "    torch.save(model, './drive/My Drive/FINALwith_focalloss_resnet18.pth')\n",
        "\n",
        "\n",
        "print('Finished Training')\n",
        "plotCurves(stat)"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Training Started.......\n",
            "\n",
            "Epoch 0 : Training Loss: 0.055653, Training Accuracy: 68.677%\n",
            "\t  Validation Loss: 0.051028, Validation Accuracy: 66.773%\n",
            "\t  F1: 0.100575\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 6 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 288x360 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 6 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 288x360 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "error",
          "ename": "KeyboardInterrupt",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-16-65ba30c3d521>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     15\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     16\u001b[0m     \u001b[0;31m# Training dataset pass\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 17\u001b[0;31m     \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m \u001b[0;32min\u001b[0m \u001b[0menumerate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrainloader\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     18\u001b[0m         \u001b[0;31m# get the inputs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     19\u001b[0m         \u001b[0minputs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlabels\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torch/utils/data/dataloader.py\u001b[0m in \u001b[0;36m__next__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    343\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    344\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__next__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 345\u001b[0;31m         \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_next_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    346\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_num_yielded\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    347\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dataset_kind\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0m_DatasetKind\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mIterable\u001b[0m \u001b[0;32mand\u001b[0m\u001b[0;31m \u001b[0m\u001b[0;31m\\\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torch/utils/data/dataloader.py\u001b[0m in \u001b[0;36m_next_data\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    383\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_next_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    384\u001b[0m         \u001b[0mindex\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_next_index\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m  \u001b[0;31m# may raise StopIteration\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 385\u001b[0;31m         \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dataset_fetcher\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfetch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mindex\u001b[0m\u001b[0;34m)\u001b[0m  \u001b[0;31m# may raise StopIteration\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    386\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_pin_memory\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    387\u001b[0m             \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_utils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpin_memory\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpin_memory\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torch/utils/data/_utils/fetch.py\u001b[0m in \u001b[0;36mfetch\u001b[0;34m(self, possibly_batched_index)\u001b[0m\n\u001b[1;32m     42\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mfetch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     43\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mauto_collation\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 44\u001b[0;31m             \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0midx\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0midx\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     45\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     46\u001b[0m             \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torch/utils/data/_utils/fetch.py\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m     42\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mfetch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     43\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mauto_collation\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 44\u001b[0;31m             \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0midx\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0midx\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     45\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     46\u001b[0m             \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchvision/datasets/folder.py\u001b[0m in \u001b[0;36m__getitem__\u001b[0;34m(self, index)\u001b[0m\n\u001b[1;32m    135\u001b[0m         \u001b[0msample\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloader\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpath\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    136\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtransform\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 137\u001b[0;31m             \u001b[0msample\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtransform\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msample\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    138\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtarget_transform\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    139\u001b[0m             \u001b[0mtarget\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtarget_transform\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtarget\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchvision/transforms/transforms.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, img)\u001b[0m\n\u001b[1;32m     59\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__call__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mimg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     60\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mt\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtransforms\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 61\u001b[0;31m             \u001b[0mimg\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mt\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mimg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     62\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mimg\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     63\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchvision/transforms/transforms.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, img)\u001b[0m\n\u001b[1;32m    996\u001b[0m         \u001b[0mangle\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_params\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdegrees\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    997\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 998\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mF\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrotate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mimg\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mangle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresample\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexpand\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcenter\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfill\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    999\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1000\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__repr__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchvision/transforms/functional.py\u001b[0m in \u001b[0;36mrotate\u001b[0;34m(img, angle, resample, expand, center, fill)\u001b[0m\n\u001b[1;32m    763\u001b[0m     \u001b[0mopts\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_parse_fill\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfill\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mimg\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'5.2.0'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    764\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 765\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mimg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrotate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mangle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mresample\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mexpand\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcenter\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mopts\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    766\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    767\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/PIL/Image.py\u001b[0m in \u001b[0;36mrotate\u001b[0;34m(self, angle, resample, expand, center, translate, fillcolor)\u001b[0m\n\u001b[1;32m   2021\u001b[0m             \u001b[0mw\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mh\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnw\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnh\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2022\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2023\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtransform\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mw\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mh\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mAFFINE\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmatrix\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mresample\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfillcolor\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfillcolor\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2024\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2025\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0msave\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfp\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mformat\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mparams\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/PIL/Image.py\u001b[0m in \u001b[0;36mtransform\u001b[0;34m(self, size, method, data, resample, fill, fillcolor)\u001b[0m\n\u001b[1;32m   2343\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2344\u001b[0m             im.__transformer(\n\u001b[0;32m-> 2345\u001b[0;31m                 \u001b[0;34m(\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0msize\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmethod\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mresample\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfillcolor\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2346\u001b[0m             )\n\u001b[1;32m   2347\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/PIL/Image.py\u001b[0m in \u001b[0;36m__transformer\u001b[0;34m(self, box, image, method, data, resample, fill)\u001b[0m\n\u001b[1;32m   2419\u001b[0m             \u001b[0mresample\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mNEAREST\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2420\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2421\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mim\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtransform2\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbox\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mimage\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mim\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmethod\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mresample\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfill\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2422\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2423\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mtranspose\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmethod\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Vg696pR5W3q4",
        "colab_type": "code",
        "outputId": "ecfc66b1-4b5c-4a01-f5f0-3b0549dfd87d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 791
        }
      },
      "source": [
        "plotCurves(stat)"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_37WpZuxxysj",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "stat.to_csv('./drive/My Drive/finalPLOT2.csv')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3k32za79o6zc",
        "colab_type": "text"
      },
      "source": [
        "Test data predictions"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PUbE7GAeQzl-",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Load the trained model\n",
        "#model = torch.load('./drive/My Drive/Covid_focal_loss_weights/res18_focal_loss.pth')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VHfLI_GzUxh2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "class TestDataSet(Dataset):\n",
        "    def __init__(self, main_dir, transform):\n",
        "        self.main_dir = main_dir\n",
        "        self.transform = transform\n",
        "        all_imgs = os.listdir(main_dir)\n",
        "        self.total_imgs = natsort.natsorted(all_imgs)\n",
        "\n",
        "    def __len__(self):\n",
        "        return len(self.total_imgs)\n",
        "\n",
        "    def __getitem__(self, idx):\n",
        "        img_loc = os.path.join(self.main_dir, self.total_imgs[idx])\n",
        "        image = Image.open(img_loc).convert(\"RGB\")\n",
        "        tensor_image = self.transform(image)\n",
        "        return img_loc, tensor_image"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "483T8G_iU5XF",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "test_transforms = transforms.Compose([transforms.Resize((224,244), 2),\n",
        "                                      transforms.ToTensor()])\n",
        "\n",
        "test_dataset = TestDataSet(data_dir + \"/Test/\", transform=test_transforms)\n",
        "testloader = torch.utils.data.DataLoader(test_dataset, batch_size=1, shuffle=False)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S7t3dBjFKyUX",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Prediction on Test dataset\n",
        "device = torch.device(\"cuda\")\n",
        "model = model.to(device)\n",
        "pred, names = [], []\n",
        "with torch.no_grad():\n",
        "    model.eval()\n",
        "    # Test dataset pass\n",
        "    for i, data in enumerate(testloader):\n",
        "        # get the inputs\n",
        "        name, inputs = data\n",
        "        inputs = inputs.to(device)\n",
        "\n",
        "        outputs = model(inputs) \n",
        "        outs = nn.Sigmoid()\n",
        "        out = outs(outputs)\n",
        "\n",
        "        #Predicting labels\n",
        "        thresh = Variable(torch.Tensor([0.50])).to(device)\n",
        "        one = Variable(torch.Tensor([1])).to(device)\n",
        "        zero = Variable(torch.Tensor([0])).to(device)\n",
        "        predicted = torch.where(out > thresh, one, zero)\n",
        "        pred.extend(predicted.data.cpu().numpy().astype(int))\n",
        "        names.extend(name)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "K-Y23I63p7h8",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "img = pd.DataFrame(names, columns=['image name'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wOymVR9Zv7G4",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "label = pd.DataFrame(pred, columns=['Covid Class', 'Normal Class', 'Pneumunia Class'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PDwyU0My4i9b",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "csv = pd.concat([img, label], axis=1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lMipXe0YuTsJ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "columns_titles = [\"image name\", \"Covid Class\", \"Pneumunia Class\", \"Normal Class\"]\n",
        "csv = csv.reindex(columns=columns_titles)\n",
        "csv.head(10)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lZBVEL0fuY1I",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "csv.to_csv('./drive/My Drive/FINAL.csv')"
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}