resnet5_keras_baseline_COVID_LFLImages_21May.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.6.7"
    },
    "colab": {
      "name": "resnet5_keras_baseline_COVID_Finetuning_24April.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "machine_shape": "hm",
      "include_colab_link": true
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/VeereshShringari/COVID-testing/blob/master/resnet5_keras_baseline_COVID_LFLImages_21May.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kApyA4pB-TqH",
        "colab_type": "text"
      },
      "source": [
        "### ResNet50 Keras baseline model\n",
        "\n",
        "This notebook takes you through some important steps in building a deep convnet in Keras for multilabel classification of brain CT scans. \n",
        "\n",
        "\n",
        "*Update (1):*\n",
        "* *training for 4 epochs instead of 3.*\n",
        "* *batch size lowered to 16 from 32.*\n",
        "* *training without learning rate decay.*\n",
        "* *Weighted BCE instead of \"plain\" BCE*\n",
        "* *training data lowered to 80% from 90%.*\n",
        "\n",
        "\n",
        "*Update (2):*\n",
        "* *adding competition metric for training*\n",
        "* *using custom Callback for validation and test sets instead of the `run()` function and 'global epochs'*\n",
        "* *training with \"plain\" BCE again*\n",
        "* *merging TestDataGenerator and TrainDataGenerator into one*\n",
        "* *adding undersampling (see inside `on_epoch_end`), will now run 6 epochs*\n",
        "\n",
        "*Update (3):*\n",
        "* *skipping/removing windowing (value clipping), but the transformation to Hounsfield Units is kept*\n",
        "* *removing initial layer (doing np.stack((img,)&ast;3, axis=-1)) instead*\n",
        "* *reducing learning rate to 5e-4 and add decay*\n",
        "* *increasing batch size to 32 from 16*\n",
        "* *Increasing training set to 90% of the data (10% for validation)*\n",
        "* *slight increase in undersampling*\n",
        "* *fixed some hardcoding for input dims/sizes*"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "60ahCsN4EKAE",
        "colab_type": "code",
        "outputId": "9318fab5-84b7-4fd6-9939-1a9d543bfe36",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "!unzip data.zip"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "unzip:  cannot find or open data.zip, data.zip.zip or data.zip.ZIP.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
        "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
        "id": "D4RjvLfX-TqJ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "38a939d7-77ff-493e-a9d0-0636bbfe2b2c"
      },
      "source": [
        "import collections\n",
        "from datetime import datetime\n",
        "from math import ceil, floor\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import os\n",
        "import pandas as pd\n",
        "from tqdm import tqdm_notebook as tqdm\n",
        "import sys\n",
        "\n",
        "\n",
        "\n",
        "import cv2\n",
        "import tensorflow as tf\n",
        "import keras\n",
        "\n",
        "\n",
        "from keras_applications.resnet import ResNet50\n",
        "from keras.preprocessing.image import load_img\n",
        "from keras.preprocessing.image import img_to_array\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "from sklearn.model_selection import ShuffleSplit\n",
        "\n",
        "from tensorflow.keras.applications import imagenet_utils\n",
        "preprocess_input = imagenet_utils.preprocess_input\n",
        "\n",
        "WEIGHTS_PATH = 'drive/My Drive/resnet50_weights_tf_dim_ordering_tf_kernels.h5'\n",
        "WEIGHTS_PATH_NO_TOP = 'resnet50_weights_tf_dim_ordering_tf_kernels_notop.h5'\n",
        "\n",
        "\n",
        "TEST_IMAGES_DIR = 'drive/My Drive/data/LFA_Iamges/Test/'\n",
        "TRAIN_IMAGES_DIR = 'drive/My Drive/data/LFA_Iamges/Train/'\n",
        "GENERATE_IMAGES_DIR = 'drive/My Drive/data/LFA_Iamges/General/'\n",
        "\n",
        "\n",
        "\n"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Vu4oeW23yNu3",
        "colab_type": "text"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yl7PogfSyN7a",
        "colab_type": "code",
        "outputId": "b38c72c7-ecc7-4dff-c8ed-b55427b85f75",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 123
        }
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive',force_remount=True)"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Go to this URL in a browser: https://accounts.google.com/o/oauth2/auth?client_id=947318989803-6bn6qk8qdgf4n4g3pfee6491hc0brc4i.apps.googleusercontent.com&redirect_uri=urn%3aietf%3awg%3aoauth%3a2.0%3aoob&response_type=code&scope=email%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdocs.test%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive.photos.readonly%20https%3a%2f%2fwww.googleapis.com%2fauth%2fpeopleapi.readonly\n",
            "\n",
            "Enter your authorization code:\n",
            "··········\n",
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jje61omNyOb1",
        "colab_type": "code",
        "outputId": "798f82ae-7550-4adc-b5cf-63cdeee032f2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "WEIGHTS_PATH"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'drive/My Drive/resnet50_weights_tf_dim_ordering_tf_kernels.h5'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 65
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "thLEukWZ6czH",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from google.colab import auth\n",
        "auth.authenticate_user()\n",
        "import gspread\n",
        "from oauth2client.client import GoogleCredentials\n",
        "gc = gspread.authorize(GoogleCredentials.get_application_default())"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Kc5Chlk1-64e",
        "colab_type": "code",
        "outputId": "cc5f8539-e22d-4e53-fc8c-171c2b86535f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 139
        }
      },
      "source": [
        "wb = gc.open_by_url('https://docs.google.com/spreadsheets/d/16isnCyge5lJ3LZ6viAv8U6zOTpQ3QU2B9ikxoosLTZA')\n",
        "sheet = wb.sheet1\n",
        "data = sheet.get_all_values()\n",
        "df_labels = pd.DataFrame(data)\n",
        "#df_labels.drop([\"reviewer\", \"Control_intensity\",\"IgG_intensity\", \"IgM_intensity\"],axis=1 )\n",
        "cols = [1, 2,3, 4]\n",
        "df_labels.drop(df_labels.columns[cols],axis=1,inplace=True)\n",
        "\n",
        "df_labels.columns = df_labels.iloc[0]\n",
        "df_labels = df_labels.iloc[1:]\n",
        "df_labels['Assay_no.']=df_labels['Assay_no.'].astype(str).astype(int)\n",
        "df_labels['Assay_no.'].dtypes\n",
        "df_labels['Assay_no.'] = df_labels['Assay_no.'].apply(lambda x: '{0:0>4}'.format(x))\n",
        "df_labels=df_labels.rename({'Assay_no.':'Labels'}, axis='columns')\n",
        "df_labels.head(3)"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Labels</th>\n",
              "      <th>Result</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0001</td>\n",
              "      <td>Positive</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0002</td>\n",
              "      <td>Negative</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0003</td>\n",
              "      <td>Negative</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "0 Labels    Result\n",
              "1   0001  Positive\n",
              "2   0002  Negative\n",
              "3   0003  Negative"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rCGQTI3o7snP",
        "colab_type": "text"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3O5LXUtjRqsw",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "\n",
        "def convert_jpeg(list_name):\n",
        "\n",
        "  for filename in list_name  :\n",
        "    dst = filename.split('.')[0]+\".jpg\"\n",
        "    os.rename(list_name, dst)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "I8opJ0HhTekL",
        "colab_type": "code",
        "outputId": "b5c0483d-ff2d-4167-aa44-2b6cb74e3c1e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "My_list  = [\"test/try.jpeg\", \"team/found.jpg\"]\n",
        "convert_jpeg(My_list)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            " test/try.jpg\n",
            " team/found.jpg\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sFVPWluZDA5d",
        "colab_type": "code",
        "outputId": "48ca076b-965f-4c17-905b-9163acf64dc5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        }
      },
      "source": [
        "import IPython.display as display\n",
        "from PIL import Image\n",
        "import pathlib\n",
        "\n",
        "\n",
        "data_dir = pathlib.Path(\".\")\n",
        "\n",
        "images = list(data_dir.glob('drive/My Drive/data/LFA_Iamges/Train/*.jpg'))\n",
        "#images_path+=list(data_dir.glob('drive/My Drive/data/LFA_Iamges/Images/*.jpeg'))\n",
        "\n",
        "#convert_jpeg(images_path)\n",
        "\n",
        "#https://medium.com/my-world-with-python/how-to-rename-multiple-files-or-data-sets-which-is-in-google-drive-using-python-af6233849ee7\n",
        "\n",
        "new_width = 3000\n",
        "new_height = 1000\n",
        "for image_path in images:\n",
        "    #print(image_path)\n",
        "    im = cv2.imread(str(image_path))\n",
        "    #display.display(im)\n",
        "    width, height,_ = im.shape   # Get dimensions\n",
        "    if width < height:\n",
        "      M = cv2.getRotationMatrix2D((height/2,width/2),90,1)\n",
        "      im = cv2.warpAffine(im,M,(height,width))\n",
        "    plt.imshow(im.astype('uint8'))\n",
        "plt.show()\n"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAANwAAAD8CAYAAAAc9sq3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOy9XaxtS3Ye9I1RNdfe59x7+8+YlgmRHKHmIXnACRY2gocglI7jl4YXy4lEmihS82BLIPEQw4tRokh5gCAiUIQRrcQSibEEUVrIwjQWUZ4c2kJREjtx3DK2sOXYgjZOu+/Ze81ZNXgYPzWq1trn3nvO5vZu3VPSOXvvteasWbOqxt83fopEBG/am/amfTiNv9kDeNPetI9Se0Nwb9qb9iG2NwT3pr1pH2J7Q3Bv2pv2IbY3BPemvWkfYntDcG/am/Yhtg+d4Ijo+4joF4noq0T0Ix/289+0N+2b2ejD9MMRUQHwjwH8EQC/BuArAP64iPzChzaIN+1N+ya2D1vC/SsAvioivywiZwA/AeBzH/IY3rQ37ZvW6of8vN8D4P9Kf/8agO/JFxDRFwB8AQCYy798+/y5fwPS79GlAwJI3GO/pM8ealmiExFAuffr18W18VPsmuvXro2ILj576B6icb2IpHckfex4/IN9XRu79+vvKVfHIBDRMbyX4uPDiDfzeZF1mOmDi1mevsK4/H0s5LUBeYewPvL47NmSL7JPxrsKuuhPgu4NStfc3737f4vIt3/AkU3twya492wi8mMAfgwA3v7Yx+UP/MF/FRBBKQXMDGZGF8F+HLohRUBEsXlEBK012zgEZgKg3/fe0XuPe7iwER2NDW2t1goioDW9h4hQSsFpqyjMaL2jtYbjaPHc3gWtNx2LQJ8LARPFBiiloJQS48nvUAqDC2PbNgDA/f399L2Pwf+G0NKHjtf79sbMAAGVGbVWAM60BKVWtOOA0CBAIgLbnGl/Y878+0w4RISt1nju0TvI+vJx+BqVUiaGEs8rjEJAF0FruoYAdO7sHePZiYEpU7J3FYq58PkAAZV033ApONqBtvQnImhHQ+sNe2u63tC9U22+RQS/+Pd/7lc/6H5e24dNcL8O4Pemv/95++x6E4CYAREws0k3XfitFJ04IAhCCUwn1xeslBLd9T4+FyTJI/59BxPpMwHUUlBLwf15RzcCa8wo9u8wogecuAERRpMGgNCabgQhQNBifADG+xgTcMJBa+hdUMrQ9kUAkfGO3o90AReCiHLhUgpobziOI54jItj3HaUU7F1wtNwPwNYvEsG5lPExHocYExnyIQugaszQmR2ghOMaifIeFZt5rfz9dUN3cC32roO4BQDXCuk9nq/jApRxNIiw9THGNAhTpZUTeuES0t3/AUBhgltYnTnG7n29p8h/n+3DJrivAPgMEf0+KKH9IIA/8dDFrm6QLSabZNG5piBG/XPmfE54zAxi0s3JCGIku7/3rhzUFlQAdOdyLnXs+b133N3fQfqG0+kUz8kSSBuFpPHNzDTGA2CSVL7orTV0EUgXSK0Idcf3Ogk6LYRHHPeKCOpWIOiQjiHNE/H5mAEdE4F0fsiu76LSD0BJ78VZqkRfZFK14SwqyY/W0U0iskkJNqLzufKWGY34e6f197kR03O7rT0RoVRlBq0Bx3HoPDNAi9QXERzUgdZQaglG25OU8/5VEyCwM0oYn+ivoOI+0D5UghORg4h+GMBPAygAvigiP//Sm5LK1zEkAzCrK4MjCrgY908bC8UlYQHzkBLH0bD3Hs9w9UWIcBwHfIqyRBIAhxFtKaq3MBN6N1WVJZiEv4JLkJaIeeKwrmoZ4RhvhSBzYoJ0gFjfxdXs4zhiLnycriH7HNj8A8BgODYGvcafg1CTO3Uw52uUAfbu0kR/tnbYPCvxxjhsXkAExpBcsz2MkIpHGlcer9/nUlOvFxDTpDq7IGJnpE5wvWvfO02aw7hvSFOSmYmCdE+1b1EJBxH5KQA/9X6vVzWAEzIyCIyYsJ/3IVlobDih2b6Lz2VszNO2obKqnK6a9sT9eu9GdNpKKai1hP3ifdZaAQKO/YC4atc6yKWG9FCV/N++79PGK6WoSpsYikuWvKl67+gCFKitV0rB+XyeiFmngsLeXW3c3FfrDCGVbP69awFH70AzG80J1KTaqpL5xq5UVZKTaiacbM88l26nunRyokuLrBrHZOMWs5sP7HtHqWWaKyIG8dgfWYq11kBCwWwfImyBxLyLgydMqGn/vU57cqBJbr7pb7YNkuyd1hpqrZPUAQFEjFKq4ko07AXAbayxmb3VWlC72g6+HbINmCWH9sfoXcCsG8A3vZBxaNH+b29v0LaO8/05JNC68f0fG5jhUi7bOW6/+rVdFJzpIijlmIhR+x0IYUY4V0QyMxW3c2d7lKf5M60r2rphJ0nj16u4AUylzCZAVqmzvRnvgUH86xoqcxHg6CiVwVTiZVfp6fPp2oCbCD4HsX/yOxFDRBkbu0rOHwGCA4wLi6AY12ki2PcDIsB2swGsahYAlYQ2QT55WX3KEw0AzVRJJbCOwio1srRwDuybIdStDh1XKeAEzPjCuv23nbbxHomL5811tAa5vw8bT8yWdDuTkdRAAL119A682++xXaB+TqCDUazSyK+NeWht2LuUVMK0GbsIIATCIEpvWT0FFrcAkc0PgUn0XUXw4sWLkOqrOhzPsz4zcZdS0AFsPk7iiWCvSXKdR1XF/T1dWmamogxPzREhBqErY7P3eIz2pAkuL2RPxKEq3FjgUgqkC46joaOh1Iq6cG2/zm2tQBgT1yRicC0gMHpX4tj3IwCJaeMSoR8Nd3KHLh1sGz9vfldNtm0DM+F83id7y8emaqJgs3GKGNDdewAewTTs99YEwAxvr9JuBZHyhrwm9fK487tMIIT3vazPSoBM7nbRfybiACJ1qRwHSmuTeyETnLteeu8RnRE2eSkxL8NUkNgnK3PRF1PwI2saWQrGvJj/LRATBjoE9K0ImrxKY5NaYtwdBHC9VEGOw7mVXWMbsLWmEq83VFRdfAypR8wowLQICnNXnPcdpZf4rrUWxBfPbR20HygJWMjSQoqAWMBccMMEPjP2fQ9pR0SgPvvOfCMcrYESBw7ktXcU7hAZxJg3mW+qAExsD61gwrrhVqmQf/p8h8oLhPQFmYsBmTkgJEmAD93Uenv35uNM/fvYj9ZUA9EvzAXCQ/21z31efH2ujX91kaxEuRLn+HsECnxLopQftBHZxCX7qx0znOuT7eqJaVVjUgG0o4UE2xKHd8DDN9G+H+bw7WiNbMMgnnWNc+rz3cYZAMHk/2vqhFU7rYYT3Pvbap1sqixhBmIJSO8g9yeZDeTv0lpLczADJVwIvcOCADCpUX6/j8ebLPO0fu4b3m0tYqByDZV6dbwTgCamAhIrkjs6vZQ29t4iI/gAR0MpDVx4PDuNqZuN6/uiZBXdzYQmOHpDOzq2U52fucwHBKq9LO6M12lPmuCyIeyb6Th2CGARE0jcy/0pMEo1/sQMdI8IOYPs3tiMzLi5PZldNBA+sfu5qBENGRz7qtqSfs0omX4gJln1QmaKMWQCy05wNoQv++6QCHEGOhShO47dnjNHphCpYzdLPr+PSQEiwXVV69KtoPoWITE8jLlxiSYiKqFsrkopiJgQIjCX5NsbjvXJhjRwqOcxO3OAMo6zSdswD0rxzRDz5/Os71InhpTV2IxSrkT40SA4t1sSp3KAwieKmVUqFQGQuD4zemu62QqrMxaqjx/HEdEY7hivtQQxUXLW6jDMrtC/0I1DtqOhi4I60juOu2bOecGpbnDFn5bQsoiUEEFriw1BZNKSsdUN77zzjr5zEIw7ztl6BjwypPemkoyGC4RY7aFV8jko05sxKVy6D5yp+VzF2GAMonWUSpPEdAR533e8uLvDcRw47/scIMAM8nlwNZkJ3ezbsYYURNN7V+YHTExBXRsyfH6JMLpI+P8yg42gBi7WZw4J87CuwbhWjeB12pMmOCWWMpyuBkBkqVBKUYKjoVr5tVxKIHzSBEym60MUxu8NtO9oveN02pIEKOo8X2yiMemCzuoHkkOUixuBtt7R7s8gGfGY7vTOfZA7acHgMgMTzMA7b38M73z843jr+XOQO7iVOj2cFhZpq4TQmzF/lchsRKjP6eOBmF0c/j4iZq+Qfi5d1UQIG+MYjnGVjFB0mMd79dbDUX/sB7ZS8bvf+AbePQ6cz+cJwSWXjEQan1oKGnP4FCe1ugu6XKq9w2bGBUFkqZXXLxNfax1ddE8RAb23hBJLClx+JAMOT5zgAAr0Lz5Jm9alm/vBVtChOsy7qA5+Te8dRzsMZWQ3TEAypCcw2zP+U8OAAOYTeOHMuf9Si8byVY9sOcIBTEQgs/34hnB3PqO3hme3z/Bt3/7teOvtt6M/IY0fXQEPxOY8ofdDfVL6GiAmUBc06ZMUAzSgt/UOBoVfEUCEig0VDIAFETjD8mBhQoG7H0QAKhJ2dy0Nm/kWW28425zVUswZbwwBZnsZofm6+k8igrCgoISLJhPPCmL52KddtKiITlRHa5Bm0rzUZIYootlSPx8JCQcg9PxJ1ViMeDeUJ0c3YJNI2LBNfTgBttZC/YqAXwF22QGCwfl8EfcYtpHZTmt4VV7c3tzoHjaP+4Ckr3C6Esnz58/x1ttvodZ68W6X6mey1/gUUfpEjq+Z3RUc32w1i6HUd4nJVEJ1GzExNzEiUb2cQppqTrHbkeNa5gLmgtPpwLZtOtfufDZ112Mu931HDd9luQrqZAawMl2PhW1pbbO9RwbajLGNNd9bw3kX9NZRq/oLJeZhgGGP1Z44wY1IhCzmJW1CVdKH49JDpkTUXuOXLKJLKW9zCJXMXB6XztRat7BnMtFl1WU3+wW7YDttE2pGTBf+HWbG6bShVI2uyc/LP/P103thJg4iCzwOYvD4TAx7yt/d+yGD8b1D+2z8bYADDcDGL8zSt5SCslUU85s6s9HrAGkSgIr7TVfJ5f3l6B9/71gzl4i9q+xNzNH0zcm2y+Mj0hhNZ4oUUo5Ukhu6+1hK5ZMmOJERpCvASHdJHD9Dtm7j+WKtMYh+jS+6E1TWSAa4oIG7k/qanqU/h4qSjfjskhARBQ18katJS+gWHX4lAXYdQzXHvaTNl1HJLEkumtt4Pj5zNk/oIwiFRwbAal86oBK/e3/WtwGhdg8HaENkVGrpQsqUNP/O16ORgjgTyms/PconR/dklNb7yIQY4zTmUVzK+Xr4fTE9M+NiZg0woOF6crT1aDva0VRBeCSKe9IEB2ikBxEDjMGB0kTnpraDws2eoJqJbuW+rqo4ptCySrqobswcgM206EyApbTkqAdvHJwbCHUFxgAE4M04dWImQdyLKrUSWFYp47uw7fR5oCFF8/us0j76A1KgBQ1CI82k8ABxSZLFpalmFjj9DFSVMAKJj11AhUGitiUTmy3d4h4NWDhinbzVWiMcyxlxtun8GRMjyXvAJL0jl3ku8t+lFAuMRgBBnrL0uu3JE1zrPXLglGOObGmPJBkcj7EVHpEoSI5akxThQE9oJoyIC2tCo9pyBI+B0HSUY7LfHPhQG9Lyx6rll62IKWBpK4PYugjargBK2IgCi9on15JjjCugMKTzQnD2FA8UJlLgJDbhwqiUNkhtv7D7MEK3/LqumRQu9zxeFTJcDP6TyBnHeFaO28Q+mBix2uAA8LvHN1S9I0VM1dUx+8KyjTYxOFOJL95PZjs55nT53veH/3OXiDyaMqntyROc+0PEfFO+4Y5DETnpGgIVque2Dd9NliYYKtAF6ll8kwAkw5fXREDHMO63KqhbvbDtJoIojI02lFpw7A2tWXIkxsastWr5BIzUfndAB5dmjs02AQQy7J6sdk3R+gvHp0JDHcQgjmwfZTVNcrSL9dkxKk4RzePy64d0lfjMSyYAIzUqM4ccsR+fM0dw/tE72vk85dM5MeSolp4Yxqpyh11tY/NxXkt9cpVW/X7DF+fPfN325AlOcXMX9cOGqqWiG1wdGb+A1hQxYpPlny6QompZHfNWLcSqHS02ZJPBZVvr4GLSzDb8aswDSnSeedy7JqlmTuljAQDpGtvXLTJdJecANTI3jynJtkuSfKt9l+uH+H2Zo19TMQsBsI3myKM+/7r6JSKQ1b41AiARlMo2l0tGgMyhXQxCp9l2ZKh20aSH+6SUkbC673usmXQLcmYN6SLSIGmIqCsiq5X2k9xMOLwshWWQmI9Q1WR1jmfV9nXakyY4FQxuX4wQHYCmhEndmG4/yMQt541lm1bMwC4UKmmOQNBFHGqKq2NE5ty1Z27bZk5TmhIp86bR9KHr/pzY6I4cutO4DSf2ZLfRiLzwlhnKNb/hKr0ekoZ5zr1okM5pD7WzQ1CwXj//zcwqDYOgPCv+koHEmAx0CT+fDJSVbQ90dBSzB7P6l/sppeBUihY0GgNSqbYwYR+rZ5pQYY15ZUYmLb32IyPhLMUDrgaapDnatKlubm4g6NNGcl/XpF71DrEFq7UECHLsh0b1w2Fyh73t3iQ1elehe8hADh1QcbvOx1FC+qWaIDQ2no/TubFAUKrm1zENlRgwQMc2DffrBvzqB4xsCRoI6ioZ8+e9d+wmzQCAvMYIaRQGkTqh/X53iciVfoP5QMEUB4Iywfv8ss05mYrde9diP04gbPVRUsiaVyzzZ3q/Wdvx/skYlaT39CYioX2s79DMdKAmOJ/PV+f8g7bXIjgi+hUAXwfQABwi8t1E9CkA/z2A7wTwKwB+QER+m/Rt/gsA3w/gXQD/roj8H+/Rv0aL8Ihtc2JbHcFclsVM4MjYXGMjnvcjUvQ9lEnBA+O2iwqX+3U1Vl5oGbnTaYvFWtW9WqraUMjSyBA2AJKlVe8odQNboK/3dTRVcT1VR2X8aM7hsw8wj329xj+f7LaJAB1MchU0zUOSEOYPCGbIMp4XzzfpcHt7exEBEjC8zWlvDVQ0K1uZmTqkucuULS5C9itbJsSoM9N7x935rBqIgVGeNVDM7RAaUPrdx7K6dfZ9xw6Aj8taKK/SHqOXf0NEvktEvtv+/hEAPyMinwHwM/Y3APwxAJ+xf18A8JffT+elFiunMHPpvJFUvbCJ6gMGVodqDb9KrVWj9Mlz5UY2c4du5DZtqOF/AzDp/zBC2FNYl6s1pQwnbrW/I1DabAdmT1MZOWNeM6XkZ+bn6gsrKJHsteDiPC/nCkZMSOZih61hcblFdMpy/2oDC5GGkQFDkqd3n2Ip8/p1DQJXcag2MBcLbK4F5bRNar33le0ql2ZHa7Eu5+NQBgH18Z22DTenE25OJ7XzMJfQqLXidDrpfnPpaI7zrG28Tvv/Q6X8HIA/bL//VQB/C8Cfsc9/XHTVfpaIPkFE3yEiv/FQR7ph9eWV4wyVQH8KWhcUGSjUzMkHclZKQd18Y7PFaALN9PasnnQgiAKu5zOZTaH9eYUpEaB10aI5NJynw65yh7X6pFqzKH4IwBKqr24g5eIbV1Nrc50QGmFbL/HLXZvD9feV2IAZAV2vu/bZarsBw/ndPS4KrOqazNdNPlDxgGiE6gciq+UJm4fL57orRVVIGzeG/3ANRngIXIoSfuR2/fDtuY24xsm+TntdghMA/wspTvxfi1ZN/nQion8C4NP2+7Uy578HwERwlEqd3z57rgiWqwYommEdcDlsE15ukHY0CLuTFlHNKVc+9lbKDFcTqeOcC6OKVidmZq1QbIBLbsd+YKtDgsYYrJyb90nklooSTbHZdzdDawcqK2Hl5SVi85MNonFiuVDhstp4ZcEyka3E6P1dEhd73HzqaZTWy2MK4Kb3qCHi6DLRSPjVONZhu7r/0d9RVT5/C8+7k7Dj8nsWLlZcqQXh1ISaHq3hxd0dTgZwTVFHq8RN7yEi4eB/KgT3r4vIrxPRPwvgy0T0j/KXIiJEHywoRlKp849/8lOx9Xwht2pcSbzEGiJaIXMmohG+pZ9jUmeilHYpuLkhnO/3sAVCl28d56PhZJu9H00zrjNhGzBwHIiUHsIcPpajIrgQelSLG+qRclElRi1yOqqHuY3kYSCregdc2mQ+vlWKPaQ6XttQk7qYGIdeqxrENamq4a1GkFDC83vz+8bm10nXz6AVrJnY5qKa6uclyGGo7+wuqUwQY5D+Wbbf931HN4nlqq2X6pveP+zsxCgfmJ9Xaa9FcCLy6/bzt4job0BPx/lNVxWJ6DsA/JZd/sHKnAMhCbzufZeRyq8Z2qa+9aGWrIZ5ACxiCZMm3TI8vdWKQgX7cYzKzlZe253RruM7IXklJ1+c1hpevNuwbVtIOqaiBWgSR/bxRWgX54Bexs3pZKqV1ovU9xuRNlfWIH76c1YJdk0dfK978yqMYgbXJeEl+jkqrOX+3aF8sndsBgaBzC/nawlCJ9EShKzMpxYNDm+WE6gulMFM1QQo0zx7QLkNPHyDMX5jwLn+TXMnOmZ187HaK4MmRPQWEb3jvwP4LIB/AOBLAD5vl30ewN+0378E4E+Stu8F8Dsvs98ABOysE6H/XC3cThuePX+G02kLhNLGMnF7uEQTwbt393hxdz8AlsTJuAyHKWClF04n3N7cDC66VftXtIhsAgSAIfGyzk90nVBWKaSEp+oWlVHbflX7rqmQ3vL7PPS8UG0XyffgM4YZfJXYrt8znuVlBP3e8/mMd999Fy9evMCLFy+m4rB5XLT0Ua1SF5lqqfYtAeQI9XA95PFke86BM8AyxY9jUrs9/y37Ot057qXtX7e9joT7NIC/YS9WAfw1EfmfiegrAH6SiP40gF8F8AN2/U9BXQJfhboF/tT7fVA2dPM/Zka3Umy5GhMA5WhAQMnO+VrrYLQI2/Hv/B7Yp/u+gywfbjudcOw71JZQN0KnbkHBVqBniURwDktUQFBo3VUZVxM7WjjwncPqOHQME4Sv4Z2TerPac+/Vrkm6tY8L5iB+zfV8vGn+0u/dEQyDHiNsqjXIccSmzs917cO1kmPfASvoFOp7GjMXipy+dV8AA3l0Jzkp94MHX9+fz9iPQ+1pHn7PeK/kK32s9soEJyK/DOBfuvL5/wPg37zyuQD4oQ/0jGS/5RSV3JgZXDq4DSnjqmgu/c3MocqoKjig7oxGAYj77lrDze1toJW7OVVrrYD4aTsMUJ8klf9kM957bxFFkjdpn466sqphNCBot3UQmQRa2gCTMJnDtd7LNsvXXotMWQkqVEQu9tw1afZSIoEI3D0xGFZfcqhprnU4gJLnyzUaQA9VOZ9dG+kBeDlkn8c8uW+AcAdk53cetx83tpuE9f2x+m8nbekR2pOONBG5Vk9kbFBgcKNSrDAPkdWw9BLopKqoJDWB1GbIqFtr4ww4IkUpCbb5wy5kDcQVOwqraont7g5f+0mlakiRbyIe8X8u+Xoy6IcLQaW1bu5h1zCADrqotn1Nwlz7/pobYP0uz+31PlTF13ILQ83M12XQxhmYMsSRy+e5jW4Xrb5Dd+dklFBRzuHeiXlcpBqzpjQxNBl1zOlIQnZNol85qYfIopaIwFaoNh9J9hjtSRMcZGRM+0S7arAa/CC1w0BkJ9cQymnT89b2JQ9O7KRL8aiSEpCzR53X0xYl1H2jQMQIFZNjlEWjJAZqZ5EoBPRSwjbs0ifi8GO04m8ygIQzc9GNzoZSCsZu9+iYYB5Ek02yElf2Nfrnq4TES+5XqN4zK2ZnuDPGLDWVCJAIYgskmJfnZU3DfZJi8z2YK7Sup4xDVsTXwcs3MGNP/eaghOz30xOVysVcdNFqZQSgEV3M2eu2p01wNOvPPik5aDWjks7hyEvI9ZE4aiGAqLXqYhGnDaN6GjFD2oHWFQljKpHSwpyKjzoYA4sWKRUgRm96+qc0Q+WkoRfB7TZPsxNqKUUPb8wqGY8Tb7Stgc+Z2eh/ZGUmxCT6hYqX7nepn5G3NaA5M7P52a4yjtN2MjFPQdVmwznBadxqw+3tLfZ9n57lkscJ0+NctSBvA3UKRFefp5pAa5Zk7MRKFMBHoNoY5SP8jUthK1/uBzqOrJPWGopnnOecyUdqT5rgCBppUiyNv9sCxCmmSbcmIo0KqaqOjbPaTLdvgyN7RLvfJ8hpI2bri+aREbT6MnNFM2DGLoNgZImXsoGo2wEh2m8Fo1ocZetGWDwIgkvRFP40FpAdz5Xez0saMHs9S50LpxE/8+w4hs2bVbwJuSMNf7omCbM66P3k2EKX0h5V4/3nNKBsGwLmkmHTBAhT/6sLx4OPvdSCczuXPBfvglH7ZN933N/f42SuG7e3Ob03kAOmjUnICG2LQlD39+BSIuChshY+eoz2tAmORiSAb4rz+Yx2NDCXKFUtQJRkAw/VL+v70kdtkFw9eVWBtpsTimyJmAlEWm6PawXagcOI5+h64k6GkedS26oKxsKX4ZAFhg9OgZH5rPAVQdVNZhu2sHJ9jH7CPlreKdvALuGKRY5w4YtNv9rLgRDn70Wz6zvmkKncxwQiQZmCn8oaa5JsvqhkJikn0N7Ba2QSjygefQ6DaPTTe8fd/X3siQzQ+Pjd71cKY6MaNp0XnwppL+NdeJmT12lPnuAylwWbatJ0Q+rCOuDRUQTGRQc44lEeAgK1ZkddpU1CKW6PDPVMGxUwsACajCgigAXbZoup92boovbdekflGpkO/u9oqk7FRtu0FJ7XtlT7U58FDFXLUTfCDEzoswcSu2YDOFHL8k6+oSYgCbPUyfPEqc9Qc3nEXw7XTe7PJKADGqn/bD/m9XZJsydtopQaUSHdmKnfX0pFrSrxD0sknVR0l2iuTouMfL3E8GoZZ0z4OrBJtlxV7HXbkyY4lxzn89k4pf59c6tOSLLgWF0kAHf32HCDbdNTcrwOCQBQYdTCekBiO6x0m2VnI4TieDan+DnJapdtQOIAN/QrUZAkccZu+imLJ3MOuH86AivhZQRMFaIieBamxooCFlm1mh3tlz6xLEHYn5mrNqcNyomIpv4SkfdOEHSzmUwydwGXGlLYblaVrDAKVxxtv2AW6xjcZLi/vwcIE9PS0oVWl8bGuG018hoLH1rtTEb0j5gWkpHK/ExnSIpId7XLi7uOxsqUNN7XaU+a4Bw0CYcoRkWrnIrjakioFu0EZrJKUOrYcn0AACAASURBVMOm2bYNINZjihNnNwwQwCA6tec4xpGlA1NBjQXw3CwKcCWIE27EdxTxMgtjzBFOlKSuiDvXKexLQZIuDyz8NYmxgh8TE0kMO6OUmWCvoZkgCimsJ86m6HsDTTVAW8dKBt/XUjQcD31KWM2l7PPz/bCTUliDk7PGQfNBm3mNfZy1VmzbCcd+nlN7DG0ejAWwOvlqNljp9iYa4nV/f57m9HXb0yY4WJhT4amwDWD2CRZ1BoC0BmkNrc3XDHtMC7iCLGPZdl4plp1tXE39RzN4ASx+I3JfmW4+9wHluooiAjkaDphrLqkt0S8N574+JG16IDaJb8jVmsg1K3NUTiawa0HLWY3MxOrv2BaUbvZbXTqqfc0Aq3RMgJCdBScd+7EDjDj7O94xqYgucW5vb9P7DMd5zjV0e3c/jjiwxMP/9CAQRjHCFVch7dkaX2uMGloBACIQrxTfNFi9i+C8H7P68xrtiROcNg/PyUhbF4G0Hgd2kG9+pI3XeyCbzAUE9+/08GkRzY7a4NQYSJ2rJ8BQz9RL4FW3DLlLJRcGIMCAB9na+1wQ8KYqjQbz6nt5IyIrDWHIZ/IdZSg9S4f8ewZNVk7t85RRyfzc0a+5K7o5602ir2io3zdUXS2iqpv2jL3tqFTBXAJZnRiTDNQzE6E6qtXvlgMIiAgv7u5CI3DG5XvET1C9mPOiSHYz94Tf7xXZFOFmMLUg0LVQ1Ku2J01wskgwV2mYlXO14MwmbDo0HQS26ZiA5qqRQMQiDGzfuaTxas36mUyL7W31VTkXBQH9TnO6CGOjttbizHEx1YtJFxcYRKEgDCDsaCZbhntiAMZM8n0+Bv8st5zvl+2l/Nmqbvo7XlWfSIGSsM8E8MiPdX0G0fZYq9YFd3d36TmmTtYKyBwjmQGg+EyMcGTUpHHGR0Q4rOxGrUV9p64hRAKvAjjn847Ch6rw6qALzSe0Dtt5zkCdMTxWe9IE56FSQFKrTLqBCGzqgqsvhQlAQfPTaUBhIzhKpmUVxmYJ4zpxwJE4OuyDjAju+459B25uNtP5LoOrxX1XGBy0sAItWx2cOzYscUSz5xN/8kk7gKKfbsBfs9kuVbyZu1+d5ge+y0Q5X3MpPX08AzDRTds7cMgKzujaeYhb9EoD3Al02aVwIfTDI3E4qqURFNDY244gHBnRIkOCqqp/dxxacq/oGQ4EV/NnKc3E6Kzg0pqw/DrtaROcNSp23pnIXHKOXf0bUq5wiVNpRERPanFHbccUET5FRoiYL0Yl69EEHinvi75KFC1w6oOZbRL/Rn04+nvrDZW0jLZgFBwVGdWH1YaZJRiGMjpJXb/mmuTKf18jyrl/XPSbP9OIDRrDWL6fpHW8tx6CSQJLz9GI0KGaAwfM+YxZFVWiOyLcDtDcuAMNsKrITnAwO2yjzZQAsf0gCV3WNWBmVdfLqGeSpfr0DuTv/Hg+OOCpExwl8MJScACbRMkVh9PBFExanQmI9JlilZX3QyXXNW7Vk3qhdtelPeSRHusmv6aysYYyLP13AIJKw5BXw3/A+v4+1wjpGmDi319r14jvGrHNNudlX0xFGdr80AeJn03VhgBChMqE7bRhP58B0HQCLYlAssSi2TYE1J4uTODTfDhnVj/ZmG5HNzBsqQlDpMERz59HoSNmCrW/D3/OGEcp2LCkTr1me9oEZ3PgPq9GBHh6BrTCVnPzJgEDXLQEOshsDMDCxAoaGtohk1sAwDh3wKVmmnhg2BRdBCzzvdcI0OMa3f/mC3ocHcy5nLnZoSQAD2d8JpBAGD/g9D0ktXLfedzX3iVLiHzddVvP12wgwgR3+hcwez0TYzBmM+UM7UxsqwTNf8f4dVADMbWy8RArBkXKQCOyiAicyjNEn6be51qjnnTaezdm8frtSRNcgB0YRjJaiwMbCBrSJYIpNd43ifUCX/yNVTUlVkDFfXz7vqt7oOhpnW46r5vL1dLcZ26XLoMcWG1+nrRxM0cXGHiCZSNPUumS6F6m7qzgyGpv5Xl+6LuHpNhqrzmoQ54mpX/FumkoniHJJHGqTZYehDEnORh69bHlXDW4p9Imp4YbAIbbjEzvCHbuo0DuZOMvQFJ+5kei1LlLqC6Cvh9RZrwZZxxOUz2QnWwiSyG7F5YSU6dNW5gAJkPLFN2iJoZ6MgqZCohuxnS3TVQg0sxwr9OGnogNtjFZVSoQUMglc78gAn1VCsbh3wF2llpztBKTP2gljGuS4KFr1+evn1+7b30/3/CaGoT4e5wdTkpcxKOQq1gcZh+HV3oUyXG0OPZrVX9r8qfleExHilvT8/wmm8y+U1twADK+Xll9zSj0tbP41kptr9qeNMEN9WMcDSXAJMlIvckoMgJ4J7UJPG2a5uihjGDm4JSufkYNjh6JqcRNQ35EplIHL1XJwrbzQyJGtPoKp/umcjSVmbRmJnniqToehMiO7H0YQFk36wpoPASyPGTjPUSY9tdFP/FccJRGBxz9dR+oMiMP32NmLXMu4xBOb16HJKvXgNWhKQWARrFoRFCKNQ3C6Ujx4EHgsewRlytT3RN/5nooyuu09yRbIvoiEf0WEf2D9NmniOjLRPRL9vOT9jkR0V8ioq8S0d8joj+U7vm8Xf9LRPT59zM439S5mlI+kMEjQwpxTL6m4LOGX9UN22nT4GS358xGck6pcZgtJKmihIztVFAqgwvQRaPXHXJeN1jmiOPn4udK0lpNzqHO5A2t71AVJaMsSWIep5/rZ9eIcOXkWZ196N+1Pq49Y/08978CVK0fEZ4Xlc1KBXNF4ToBJ+GABuM4Gu7u7nA+n7HvFo/JFvG/Vdzc3ODm9sYkq8ADFwpXs8t5hMilfeIopA/Rq7JlbcGZwWMd5vF+5ORfAfB9y2cfqJw56XkDPwrge6Cl9H7UifS9WmxKGFdMUsyPd5Le1eFpi+/O7djUltYf7gGZEcLelfPGWdiE4HSlFGybGs9x5HGfDfcIgHVCpB6FZ3PrgsjbOu/7RegUoKqlO4ZDxdM/xjUvIUAfj/8+SdwrXHolppd9/jLCvEaQmlSdE4h1EQQC4optu8F2ukE5nTTukitKPeG03er56QQI9DTUu7s7vHjxQrMIpCOXvg+GJYiSicHAuKKWLVBnPX5Kid3PltDPGTc3N1bmcJsY8kPo7au09yQ4EfnbAL62fPw5aBlz2M9/K33+46LtZwF8grQ25R8F8GUR+ZqI/DaAL+OSiK+0eRP5xo6cNuj5bZ5Q6Q7hbrlvvc0TVtjKraXJ62a3wbDMkJopZi+q/cJxkLHYcz8IQlN1eJTMC1vU0k/a4vAdkmCEhJGfz60X5TWJ+y5mbCGIVU28ii7iYSLO/9bQr9isqRqy+0Xj+TZb+re5VXiWhkYhgCWqCpNm0Zs2cnE2gxpm6F1wf3ePFy/ucH93NmZcJ7CllBJ23DrnGmAwn+U+pKzeexwH7u/urWrb67dXlZMftJz5Q59fNEqlzp+//fbg1rYokTxqvwv8wL6GCKi1CAEtFKOqWfhqAHC1PrrntPnm0Ixt7lpBaxAjQObIdXWReT4MQtWcQQie0xWHTjiowHaWnAzIP2ewh2MeDD8K/pqkeag9BKe/zM+W5v7is2vo3WXjQFP1Ekau90/QClzhT/N0GE876hrr6mMYhF0AGWBIHqOI+8d6ZPNnl4Jnj/j8OhNrRwsGoMxNn+QHwuxWws9NCCKC9Ia7F99cgosm8sHLmb9Hf1Hq/Nu+/dPi6BSgqlUxH1dO13DkshTNhi5dIKQoowfaxuKaz6735OQmVXO8yM9+Pix4eagbvpmIRiCyGOF76BGGIwrMjGfPniW70aeIUaVExIxz4d47Sq0oWzWflW30RUo5cLQSVf59Dbm6hormz681AuLZK8hyXRrOtqiqcASmdAALAGKPOvFqZ84UdG2GOu1QvWkoDHgQeH4XH9Ma63o+n6dMfD/FxxNUVWL6elEUB+67oCY7tJQCrhXHI/nhXhXr/E1TFUHvr5z5By5zHgMsjM3OX6u1hopFS7Vlb7oAXldk9CMiVsq8KeH4pDIDPJzOXK5xU91wbtN5qTeBp/G7GjWI0xHHNSCYzF+xqskuMSmpN/7s9f0cOFrLFeSxZjt1tVnzz0xQMc4Hrp+SZpfvLvujAUqQS7UhwfQ6iwohC02I98x2tv6NK3ZjmArJTvT7NGh6fleIhgY6+BLRJb7WNBhmXh8PJXuM9qoE90HLmf80gM8S0SdJwZLP2mcvbTpxpGembaPUNYhQFo7KxKhLla2hOpi65mpmBzROssAX2o+zKoVRN1NLbQOqz0jDwrJ6VmqNfCv3H4U0NJJcJU2TYVOutpaIoJIeROh221R2O951Vi2vIY/5+3zdCqjkjRXf2Rzn6yYbilY0MlYsPfOKPUmDOP1E0lhDyhE5M0PRsbvNJRfv6QQWKDWP8uTMmlmi69gjgqQUl3BDQnrp+izdHE390I4cJqK/Dj3v7Z8hol+Doo1/AR+gnLmIfI2I/hyAr9h1f1ZEViDmanMYXiMFUo4ZRtLkcTSUG7XnChFuTidAgF0OKzvXIYdCzAhneFos0li/1hqOPfv8JHxGrR+qFsoc/ZDDkqaaIYDpZbqBMpd1Dr9KCG8i3TSyS2KbQIY+bLxVGr/MVrt23UPX+3WrYz+3UopVI/MAAZgKLQEi1VKwn8mO/NLz8dzPCWyTY1n7WMYlQ831z/MarIHoHqrn1/p6Rj1RuITUZwUTdXW39+jjZar3B23vSXAi8scf+OoDlTMXkS8C+OIHGZxARqHUPp+zTCbltIIW0A7NSTttWxi9AWyocaBhYUkqRh5WJ80wwKq2aP5cVrEInvM1FmkUnrE4PEtyHREXo/Xeo8R5lm6ZeMlrZjKATpq9jFEiz4uoEl9Gul+q1w8T4Gr3XLMLrzWX4KFBECCJia3vxCnKBNJxvrtPqqD6zhxgGerhSI4dqm4insSIV0bnxJgRaicuMklJIBxxHkVWPUeki9dXeSzpBjz1SBNQLESG0rUKsqpbfnD9fuzYrIwaM6E3GaoXDxRtKpW9SJq8Of07Zq33GKqMjS1f73/3LmGIO3ddjXuVsJI47ALVE8URVxAxSZfUYyvRTcx+GPewJTHUq1Xije7nasIrYa1q7kqEsXHttlEcafSTI2hK4YiMCbuNdNOzASjiWPHy7Em6ixKLAkYpA5+1dils/vN8l2oRKAC6NAyGAEDmk1j9Occx7OHs1vlIxFISATenk9am4ALmUY8fadJuSLN+T7Uq3A7jcGIwu5hK5LdShpA7QDPx+ILl00wHEY4UmsmG5JG6kwl27ZMcFs/GPMYm1Ws4VKhMEBMBYEi49bqJwDFLnDy2VSIh9bOqqfPngLs5LhhKGgNk5MSFn5IKStnU0T0xtsVOXXx+GlZnWjQJCiNCsvRagZYVHPtHs+wtudRdQDQQUeccbkMWz+A3G9P9cKvK+jrtiROcSjAX74BLknHwhoim4xQuKK5yMQFw6aV1NdBnacSFUImRs5x8w6xSIUsr/Xs+o7oUDQcbAbzaaqnwRNMZbBi5cuOEU4uqMRWtDBMt5iI3MYJca/RnX9Q1Ilx/f0jK+c9VCmebas1suLiPCdz1MA/PGFi1iDxmNwHyeowxWEYFKeilarclqBqS6VW3Ikue9fNAHp2ooZJxKqzkzJATIip+gtGHaMN9s5vbCw7xttbQCbip24DTyY8IbkB3hJLsMAaTJN3NCCOaUlT1lOtZ0bmS1SqxatFTdEKq8bhPfxkO8VxGgdnOPvBDOOz9MjOJI7aubPr8DE5Bwx6SthKf33+htq7ze4U4p/dJ143vKRz9D9l8pgmimV0EokCQwaNA7TXGMBHDMhb1WV7WGyEa6ryvjQ+koILZyxYukS7WuJA7M2ItqqV/PRZw8i1BcG4st9Y0HtKICL2jbBWFix70JxIHufcu2PsBsXQYwXzkb60aQtT34YDOEu44mqqmZV6Y1poa3tItMVYPlfCxKvECwFzpy5uoTjoBBE5knm6iC26bb7k/nuMuBwwe7ldp9BlNdt3LbLL8d4wR11XTaV1wSSzzvekeGlH3XtV4ZXIkAqrVNI+RlaESyVRPFMv8zqpktsmdGSRp30VLLYZbB2GX6Rgk0HC/HtCKzza6R5NyT57gACgwQF6Tws54dvusd9RtQ88TSCm93u/rHiiV+pThJFW0SyD9iMldp5jSpsl+qxWSXu+ZwJljrn2ZN7xKYokD5XX4jgJeptwgSwcRiM2Ti5ZOuKji7M+6pl5eTjtd/akMzCSreBiXI4I+F054rsqNdJlV7SVAJb8YkslA3Yr1aV2RnxM3al5OZwcQaZlC6aCmrh9Ag5fBpnGY3ZZ9bSKC1htKAqp09BRS0s2Gx2hPnOCS2lQYJ9pwPo8NcxwNpTBuOJXydrAEgspVJRsaOoBSGdTGAjlXnLn0OCAiT/hD6iUwNnKusaGb8rKqlRCUg16ROnFNT+UXMKRdbuO+RAzDXDEbREI6PERU8cwHpNj67hOxwoOTzX7FGI85dSDolrHB6Hu/+iwfJ2igkDqH85Fkg0gceXR808bqqjkGA+s0onHynsouBwXOrJozBmPN4/zIoJQCQHrX0uKshqxH3SuSpEcBj7Ooh5TZtgo51JZj9vqHLlH6tJFjUZn8gFP4aZ+em6VRL3XafNmG9L6Gf2v0HdEcRaNbZMkkyFJDu1nsInrAjgiNbUg6kBn8GfkkhCP/ur3l4lADBLL0X6+T2OiDwpnmnEDPDOjtcsyzOpdINM3rAMrIqq31qFMDs7SI9Hy3ox1BRPEM62/vo/CUH3HsiccCJ7bxbppLqYeEDL9o0phesz1pglNYfwQLE4DTaTMitM0RnIkn1c2Rp8ociJQTwrUKTErAbhMYUXZB64LzsaNyQd3msChvTnT5dFZmglyJnKu1Qkzi5pqMPrYxDooTUp1Q/Nn559o8CTPhBXBpA8CKpKZ4RxoW4NSnjLSb3mVSA+fnXYaSxXdJQylcJhVQROLMNScOZ1xdNHCgHR4JolL0/v4cz/E0mi5tAqYy4RERjnbgOLSwkB/UMUH9hEmLKaVqOB2xng9IDmp9BGw46aKFYHhwtFor9nZolSuMBSdSG+GwtBoIQbipdyaFX51qRSPSk0oXoEAl56zuiIEjvmGd+1LejGWomq1ZXY6UI7YiboUrmiAyBrIvS9UqTwuxfgVR2Cj34/dOjQAGQ6sU59ClORJk9OVEOquWAAx99Uj99IhF7cxjWfvwzV8clfTP07xPKmbvCoBBme1WRj2a1WbWeWXUclImdiVqxg/m1DlQiUtEcR6cS75ay4hOIQIVMi1oCdd7zfakCY6YUCvB7V1gxLgFkbGeZqoHb4iVxRbApYMIdouP1ABVOwfsSDGTeaPIkFiAEXRrAYF7sRqyBRR0NEKoK9NmMNvTmyJx6p9TFHNGNyMESj8FQOPsOsyb/KrNBajKiOF28InL9uU1J27ub2IAMAlwxSe43pO/f9A+9bm2dCmv/1/sjAh9C89VnPvOtU1iPv1Z6V3y+221DgDKVUPC5JJwRqeupRHrysasPGXoMdqTJji3Z8IAFkHngtYEdVN7iFgPCxz2gB41VJDjDId6Q2YYMo00F51cQwbJQ8IGwbFFNThilm2cvPhum6jLAiYtexwzDCBqmqzFjiLFBa7KWYjasmmvbfJhQ2qKbQ4dk0TAIt1yRfX73vVdHS6HbVzpoqe9+p1EEOph/z0EHq0SNEsmZxyw93e1zv2rtzyKJ+VyGQOMwaQqZnSXme2eJPHFkctRVg/pOpjrQCH/Hkc/+4EqRECtm54WyzQxvNdpT5rgCFotuXfg7v4MYsa9+KF7I/RqgA8euTHAgkwEF+pd5pCiC8uEcGQH+mZEehwdoLHQ3jKSFmNP0k6PzOUwvHVsmIlNPIhWK0z5tddUNm8Z5h9cfXaZTIwhvT9EwBhxmwOPMPsFI5h7WhCZ1fh1jPn9fW2ime8U5HD8fFyw155BBBjreISGCpoZ4bbV+Kj3FlJLGdcIAnfGUKufRZWON8NMwMx+Sq3Z1BaDWfgjgFKCCASdAOeOh53/dUMWmtWblk4gYG/LBksthw0RAQWs8VMYh3eIUl1c46AMkyJazMlfltG/hdteqlLmXrAo9S4CpCOp1EMoIWkoEbmP6yE1TftfbCoxjr24KQBBb4pGssDr6sDEtr20DPCxSyTESqhtY8Nfk26T28DGXIjQW0Pbzyp5zF6qQGguLl2ndySAXaWmEeLHpGBY3Sx6pLVYhzVwO0vjISHH0VsadzsCEIiRStGrX1S6oNFHoNR55qS3t7doTZGrUhi325b8Xbple+9oCVLOi7eVakcC6yQLzcifckW1HyI9ppsdZchlKYyRIGmqKs/GfOb8ofKZ8NEa/d0OiRypQnrwhR2X1RSQUX+iwuLeV+7bx73ackQENN2smRGM+9ylMp8HzmV2bXh6EaVnpZW5eP6l3ZeIDpoHd7SOwsVsTT37u3i4m2sENJQ3wSgMZbNumf5+ja5DJrQ8npzPltVb/d3XEqY+VlPh52wBf3a/gmy/SnvSBKcSrZvNBT3B8nSD589u4lznUBn6qC25qninbVM1FHqCTUvqQ+j3gDJ3OzmTmEy1M9XU9HpitbE8siIvdHBQA2wcyInNauqYL2rYH7VAdtFwMYahnIZ6LjGdOi2XgEn+u1Qv2a4oLfn7EdCOQ6WNzA7l0+1pykVzqeLfK39K0m95pkvmTIyABolHwi/babZ6gaqvZrs56OSAClZCRlqH2B7DBTTZi8vnq8Qb/Y53IdKQswJnRBJFgJkTCPWa7UkTXBdokVaD9BVoALatTE7j8IP1HuXQgXR+mQezQqH4cXrKXJrA9awuHSzOCVUFG5vMfGwygJMLpDM1V8liLFACiOrRZp80SqCA3cugyGbIDnbgut3kn4ezFrqxDtHCONIF+90Zx/1ZQYuoLA289c5beOudtzXGNAjUovFdLYwTUBU3DVWVfPMmYrPP1FmtY+FSVDUVq3BtEo2x2Lzp/cXScdyWdiYgAvTWY379X/azZrssE5q7jYZNas8bxju6dPWZmso746Cv3p40wWVO7qkT27ahN+U+TRpKqSAI9mNG0bIqcRwHsJ30YPfeQ8opDD+e15uiZg6mePycO35jLLZCXoptBQgmewaMY284jiPqpoj4pqQJ2FCEzFE5PYDeSXA9YCT/zOrcQ1LQ++/NDiy0zPjz+Yzf/cY3IABubm9RyhxnCKZk6wGmBiyonY7aN/A4hdZMQ+ZU9j2Na3mX+D4xI1JRC18oJi052JsCI9nfmQnLVXXvM6PIbh/a1F7YokSWYc8EYdGK0Y8UvPyeEZl0vdT5f0JEv05Ef9f+fX/67j8iLXX+i0T0R9Pn32effZWIfmR9zksGoGeM1Rqq4f2+4/58xnGok3S3MCmH3/MCADAIPAEdUGcyGefXazTdP0P0wFisyd5hnv755/lnSNlF4up+NZdGkkLjxFTCiAsZ9t9KUGleL/7l5sHMPalcNzc3ePbsGW5vb/H8+XPc3NwAMiOKkz2ELOAHRN6lQdDSO3eAUskCDBQwn9paa0U9nabsgWne0vwMEItDujljhTB6TxWxkyqpFZ5NQwmfxHi3XFn7guDTvJZScNpOUX37ddv7kXB/BcB/CeDHl8//cxH5T/MHRPT7AfwggD8A4J8D8L8S0b9oX/9XAP4ItAjsV4joSyLyCy99MiGczJtN0H7suD9r7CQXs08WfX0ljtPpJjY3AD0njrzK8ohYIFZi8NNq3NUg4gQLPWmH5sI1mUDXZ3sLWy/Mf9vMziyIIXKAC2fmGxt2bLxLCafXUkiGwgWNjvE8EzcqbQiMkboTTENGINro00EP7ZqZQx0fquQ4k00BLCNWdaqEnZY3dxxakqVRIshiJfXE4lGZFiaSNJgWh2zquI6mqDW5CQE1E9hC/DJqmX8fa2Cq8CT1NLH1Mdr7KSL0t4noO99nf58D8BMicg/g/ySir0LPEgCAr4rILwMAEf2EXftygjP9urKWNmitRbGf7MCtpWC3YkLzBKoK6oU9xQKeXb05uiJn3cAKIs2T61wNQDDOKOqegJCFCo3oiGnxm42N1B5wIgtnOEwKCCIKRn1IQ42VDAsYsxmus0ukcmxa+BYPgElPch0MIG8ZUwLteoojwDKwQKbKkTGj5mlLaTMe7Zid53pBMA5HEYFLFDf/9N8j613EVFG1H/30W1/ffG9rHX1vEGk4WsNpO6HoQRKodcyVuwOOZOM5Y/X1KWWo0zEPaa5et71Oks8Pk56Q80UaB3M8SqlzIvo5Ivq5uxfvGnxscLptiq1WbBZkWkvBqVRU1gpeeXG9BdfNKJVxPtAcysU8jr91CH0kLkoUpYmS6wkwsex85DR9e6dRHz82cRqbBdECiKBsL7CTx7eqja6qrYmmpF/OhGmfIRONEYNAEdl8/rmrek6lAi9ZCHgZc+UX5ILVGIhH88x2kUu6rDqu/2Ks+W9CoMors8lrplFJBbVsmgOHQZx6QMc4z691wYu7M+7u7nB/f29FYTta260ywPDN5f3zGO1VCe4vA/gXAHwXgN8A8J89ymgAiMiPich3i8h339w+m4p7utFbqxbt3ErBjfnjwkZzRy0GYCJQn4+DHJO6sMQVZg6q/QkK5xCzsfi+6eK+whNq5tdn/xYhqXGEkLogQtmqRfNb3RQ/LJlmm9Sfp/MF2+SavuSqKmNUBwuIQ4B9bzjO+u98bri/O3D/Ykc7/F0WpNUOr9Rfx9xyUQCjlm0KU6NpbHMQwoV9mcAiV8u9NF2umkyE2Ad+H2we4+gpKzS1Va2Q7D5Xr+QFkBV/5chc8Cxy3zvqolBGRJhV4Wvxp6/SXgmlFJHf9N+J6L8B8D/Zny8raf7BS50TUEgj/HtvaBYi1VsDSHA66eTe3Z+xH6pSllLUYdwy4NEgUmdVBIx8VG56n0FsmaOKFgnNiMhn4wAAIABJREFUe0ZVNkKCB0zSlSFB0RciH9xfz60bf7tUicBaUKiJ/n3ewPkz1WRto/tBGgyge5Io4Wgd9y92oI9A4PvzPd599y4i+ddwLlnQ2QmUKp7RYLasAR60ji3dr1N5HXGdga1hqxEoGJFf70zWf3oCstvXBEK3gkPtUFdO2QpOJ0Ltgl4LmsVQenYHUxm2qGhAvBLa49hvwCsSHBF9h4zTc/5tAI5gfgnAXyOivwgFTT4D4H+HMvbPENHvgxLaDwL4E+/5HBBqYZwqY2+axT1qktSQYK6TeypHB/S43zY2oohW9wr1q3fzsYysb19we8eLDeN1NPLxtxBCa+6QTZnVompr702znXtO8zB7SjSS4kjZDwSLZEefxp7Vp2tEl3/3MCwis3VdKkOd4v3wsxcsoTaDDA8AGXlOnAi6pQC57bxKuPk+DuLwtgYNZGl32jZVl1nPZV+zvv3+/Fk+UthVfJdg7odVnyzQmcC9wwPRV2YytJkW2stjtFctdf6Hiei7oKT/KwD+PQAQkZ8nop+EgiEHgB8S0ZglIvph6HkCBcAXReTn3/vZwFs3myKVHaFqFNv0rbfJ2A9bBwQqBY39POcS6pVPaIegGwzoHDJDxNf8WNmGusi9kuEiCIAEduYAOFQ6J+pu6m8pRSW2aBSMY5iadiSQNpIws/p6zabI9tCsFtvGL4zTsxM4fd+OgqM/V23CSpavfV8lbrKDL7O67PbjMka3h/NcTmpoum74yShsM2eS6xo6Aee0nQnsEMS7u+3msZnxvMqRepTnzQOX991KHB6PY8O9aqnz//Yl1/95AH/+yuc/BT174H03T6HZ7+/RBXhxdw+AcLrZRlhWF/PDuDo14IgIFzJwYNggirjlMKBVhZxh4Us0LTghKBHEuIeLL75FrIdtpP+ISdVeKENA75Aw/lNfrh4mZ/MqhbNqlcebx09Q9XCrXkJAEUxmxulGQSkvsZD74OWdp3fkUewID8yRr0V2A6ySc3oXC05mGOMyW7cboXgJBD9EsQdDGQm7GdDK42lHC3+kn3nArMm9qkqPM+oUINPy+l6S4THaE480Ae7u7rFtG9ohOO/N1KCCUy04399bGkoqLc5kBCgR0lWIonCPSxftP9ef9IgSfXbmvpkjZwniqGU+Q2BsyHnjdgf8fdPK/LuImH9MXSAqNa/bOtcIalU710RT6SMliFhAZD4Xy6jPNWEmVToR+fr8HHicmYDPj48FiSHla3MisW/onNXh69OQpFtUTSNVi7vONTlBMhnKOIg6cg2TBhA1TVntvA4BPHLJ5xMDrJmApNdoT5vgDI7YasXRj1iMrTBuSsE3jFAKE2D+k73lmiXqNqhVq+f2phyTzHbzppt7+GNO2wY3lZlMtUsTPoh0gDIZQYvsgLzp9MsIzHVh11sL562WIdAoC4XrUxygzJvdx3PN5syMJL+j2nX5VFcaRYssgiSr1UNBxETM63PXn/m5vnEdZl8JN6SPRfZ7QvHR2kARJUX6MKOyQ1Sr1PVqW7PN5wQz1f90iXqs86ioKBdC3W5wgxNERjXn121PmuAAoHI1aFvwzvMTIIRPfewtnI8d+97gNUdc5xebGCc6vrnBtlWc7w/NddoKjgOgtmQNZ+lFhHo6RRYwgXDIOIwQSLZYv4xMUI47kh+zuidkcLS5K6IvAEQSaKFnCgAIm2YJapyILzayDDDk/v5+SGgigBmHXw+XdkpoQjMoEcRAM4p4rblqt55bN77XSJYAL3L/PmMy0mmaawB680B1Y/6dubjx5QEJqlqGumm2eV4vH9/qhF/eCNs2NJocVfS67ckTHOAvTDhRgYBwdz7w9Rd3sSG7CAqpilNLGYYuuQuJInM3NmkKx4oNJQLXtJhH/CEIIJlRzKGejPy54c+Z1bPZvtKFLNxxtK7j5UODirshfqaSNbEgYJsHEA3E9YpPDrbpO+ZNT6ZSZ0LL93v6jKQN5pkZWZJM0iQRe++axY3lmeMPLXe+2kEZ2PE0qX3fw7+2hn/FukFQUGA1po0e8/NEI4a6qZh9rEW27a63EaJHNIoQPXz9B2tPmuCICMK6EUotOPaOvTW8OJ/R7NimwyIzNio4Wcq9+lYE22mAK0CG1A2RM0d4NDFoHzrh+77bGc9l3mw+NjPYvfQcCGHQ+zW6aYfNNza9newiltVMasiPEKMcDibw9NjgASQQoSBIV/+ISAur0qgiFlclG2oGWgxFTO8G65cxiC0TQLYvXW3Mc5MRW2cS3hxZzJLOx7JKQV+3rAbuhlbbUbYQsNUtZZvbNBgZNu1xHBd9+fwOLcHHAUsA1nF/aCegfjObGq0W5MiEJgdaF41OYjYppVnAW6l4dnuD3nrUuljPgvPGsTB5wQnMFadNqzzt532cbCPjSKVsLyjhJoAg+aMAxAIiPUsDpl1lGtIqS6Fcg8W5MZvjjuC83H8TK7OSmEEiPg4xZejs3nF0QSUCSGHv+3fvNW2o+5nbCXRIwQLeZ24rEviQJMjZAh6pP88nLiI6Vvs0f35vzFDftwPJ9vQz63okAY++/Ohin8VsJuS/3W4TdPO9Ps4Z30+a4OAGL2ki4v3RsB8t0Ef0jlL0XK9PvP0cpRZ8I269PHjQN+KozJSjx8lOpEk2UdGy6ZIkZFYrvWQCMEskpDHoJhA79KMBHmrkaBlrSW890spAibSJ84aITW0byhFNFy8KjMwggF6vc9X2Ay9+510c+xHS7+7uHv/vb/8OPv6pj+ET3/bJsNnCzrmyLNcQy2tgSHYDuJo4z8voz8OzWhs5hut8ZhU0q6f5FKVCHBWnR/UvsQz+sQ8IY71yMIEzwMFsi/lKPwIoJTDQNY8o6SJ68IIITrWgsNU2AeHdF/e43/eI3KgJanaOBxEcTTOdfZKjpiRkRCsktQppgQeR+oERacfnMWfJJQlV7D3AgJCwFnnimzK7Lby/SyBiACQXn2eJw+zaLgDC0RrOZ60k3VrD+e4e9y/usB/Pp2dFv1ekWw6Ry+fbXRuLp8U40fm8ZLVudWEAlwBNvi7b0SGNfdw8Ez2gtS8Jo7y6l1B4iGlE/5ZvX7gEI3vd9q1BcL3H6ThczF/FhFOtARL802+8i9YFR2s4mhrxW92wbVX9cjIqK3udkFIYpRZTQQDH8wW2aTFv+nFcrv8by3DN0bpycRGBkNl96gqHiG6W1Ue03j+364jZIJJrqi3hdNrw1sfewvO3n1nKE+H53XNsz0549vYtcsHTsA2XfrJ/K6vZU5xq0hw0ZGxIpoxGOjgzSZQ0V1nyrGcBTs+CuhE2OzyxSTPwazjT/VARJVBzUfSZiWS0V8d+gnRBsYM1H6N9SxCcc9lTVbVDF4q1tkkTVZe62iYCAxlEjyGq9QYMwlkUXCEAlQsOCHq3KAYZeWzN4wMxNg7ghJcww7ww5Cospg2UiWhG9/R5XApYtHKz9+l1NbMUmF0O1gQPfrcSgDoVCVQYt2/dWAKABmLfvr2h3OrZ1n6sFOz/skidTEjr56s6ma/nUrQkRmvTvI33nRHgWPf0MwceZMkY79y7Hupi71+sPP1WtVqbgyc+d0DwV0giTt87bOqpo5RX1+AV2pMnuG7pK6ethHonvWA72dDFzgkw57QrggygkNauPGNIHw2GFRxeZtxDglh9OSbW1Cm9nJiSN8i8IZQYt9MIas72Y26UQsEIao+BR4aAStYFAXxQ0iGum+y9ZNMN1Q2hchGQEkRd1TI3H+Uxm4+RBhFOwdNEUREsj3H+XbM7JNu+STVctYD8PhMRJKmWJW08ZyoCa2fjGdHpmeqqujfRIlKVy/DXtZnQxU7Z3bhGAuu1A2BepT1pghMR3N2f8fzZLU6GErXegKqEcz7vWlCoqxoZC1oYt7Xi2c0JvXecza5jKoHtMRf0NtwCoU6wqo5ud+VFXu2OYWcksKXn4GGlXU9ihUdAyMiBE+OmeiBkn2w4LffQLghPiWot5JO1gcuNC0Go3wBmJ7VIUidHzKmSfrJ1cMW2SiDPwORnKQZgKrGQA8Vdsq2hc652Tu6FK0S9qrn5M9dKnAftx4GznSkoRceqx4f1IDq/95AeFdU+MhJOoVndnM9OJzsHTnCIEtF539Ga2m1cGLWq6lIL47TpsVbn48BhVbq0mpSm7Hu2wWoPeEa3uxJWB3Oe/AnWtg2naCNNGcbAkJZMdu64viAcO71wZBOZJjiILTuhVRTpj/BzmQTzAx0n8MPuW4OR3aEu3SqELY6s9Z1ziw2OoYhquouHjw3i47IckGII6Koy5qOn1jH4nsByn69DDr/Kmoakf0fr4Nsb7YO9Rqa+hVdn41Kwn+9H2bz0vNdtT5vgAAU/SsHtzYajHRMHzNy8cFG93WoYxik6pHXhu4VmqZ0G5DMCLuwO5IPZL0GDjGZ5FAzgh3ckRDLbeYkomCt6P8Z36dlePHZW7bTNS24HSlq+SYxrse1G30DnQXQRE+l9E60PgFdXGajdsvHS+LI6Tx6ADVeh3feVQ+J6PG5lTj4e1ySyapxjObOdOp5v7pusubQRVVSYI1PA7+29RziZA1YByC225eu2J01wTIS3biqe35xCZQPSpqmMm22LLAJfxC4jI8CL+cB1cNsJHo6VPMkABmTd+uC0+dRTYPYHxT9LEVrDl8J2BOOQI21sVeUkXadugeQEt+8e4q46J1rU1FNLrOdFiuWNf8XWovkZIQUH+7fZgdmYSw1Oew8hdf6vxzuxBSm4Oul/C4Zki3VNRLTGoV7azpcMgO0ZzSQaAFCtYJIIRWU73dRLbmiNy1GQVkSiVAPSZ4/RnjTBFWZ828fexk1h7F2PHT6oox0HpBNON1VVxMKaJXBvRJXUL8CjxIddlTcRmy1EZIdsuBriNt0D6kz+3I/G9c+yQ9WPZSrVNo/eARJzDPSuGepB0IMJSJsRsujb1Mj4yVD1lPWtFYafI1WoJ6lpmoGYXZclW/ZpRcEjQzl1NudSClnN1b7HvHW/F+O5XkuEErC0Zl+sQEpW7f1vt/F8/olIx+SSHkDdtuhfIHaehEQh7fP5iLWutQY4FAEOtpciVvQR2pMmOLcxele77GbbcHc+DACpsSkKF/S+B5dVjdFy3GSkZQDOfZWfrdyUhKaF9M+BWU0cwb00XZdtu1UqiXhYUUG3QF4S4Eg2ZGsNbT+MEBgdMzIXhAe6ZAZ+CAkk7MULiWD2WUjZa96GVRV1xznRBTEoISvD8KgUcVs5QB0Hozikj4nm6Mvt3mx3rUDItd8z6OLSU+cHE5EC47wGZzTAGlSuTHrvmgDrEUFNOhgay/sY7UkTHERLmJ8qsBU9iFGrORE6VCXQEy7FzpGTiJT3TSci6p8LgtGu3d7yazLh5QXPiZFulAsELLOvzNuFTeaEKLO9FM/rjG4xss3gc1fTcsrQuuFXKeDjMkfkrEIKQBbelIlNhQJnU8zszfG7quBrlM3ybKJBuzS0BWa34iyKJsoPIkLdPGDZ2+pYzyk2+d1Pp1MALFmND4TV371LgB86FZ5kzJHJvWoQHYAcXs2bUW5Ojybh3jPJh4h+LxH9b0T0C0T080T079vnnyKiLxPRL9nPT9rnRER/ibSk+d8joj+U+vq8Xf9LRPT593q2Syf9Hbg/DhxdPzufFepv0nHetaCrnwiz73ucte1q4tjkpjaaG8HLsgFaVTnqRy5qTDbY/bN8JJa934W6ma+PSWc9comLRquskka6TAt8jciu2T0k4zua7DIJ2wuutqZxBWqKwentrqhF+dC7UAo4ztI4j0+lWiJmu9/dA9u2xb8c2JxtZe/Pv1/tvZiLUrCdTgnwklQXhWw8NWzl6M/U0REGqGlAa/Dz67b3k1V3APgPReT3A/heAD9EWtL8RwD8jIh8BsDP2N8A8Meg1bo+A+AL0BqWIKJPQQsQfQ+0GvOP0igg+2Dbm+DcBV+/u8fXX9xhP1rUkidSh+1hGQTNj2ESy2oinipY+SETOQYxS49SC1bCAi5rcLhKt6bt5+Yb3vtbN0e3wrZarKbP2A1c67qMZF9/j2sc/whiT8Qf70PR90PEa52Gus5Xank8JNHXv8f7W1yrMz/Mc5Pr/Oe5fqiwU8xb75O5UGsNgCvGqaadqfPzO8Rza0HdqoX5jfFsp1NkS6wM9FXb+yki9BvQYq8Qka8T0T+EVk3+HLSaFwD8VQB/C8Cfsc9/XPStf5aIPkFE32HXfllEvmYv/GUA3wfgrz/4bKiN8+7dji5a08RdA1odmbEfGtbVLN5SCcjr+BuEUKyWZTdIuI7D+MYCI6Sdb3Yb52QrrD64dXP5NVkVBVKsJYDWDnXgWwupIlpl2KPd/Tufi1WajnkSFDtVZrQWKB2AAYxCGVJJUpDgZ48PsEQMeczO9YckWLYnMwOLsbrUMI0gE9W1YOSVaHOco8+jE1pmaKPMhUP5Er42Ik1xGrY0gbjA3ZlxEkG3sDBiPV2Phs/wMdoHsuGI6DsB/EEAfwfAp2XUpvwnAD5tv79WuXMi+gJUMuKdj30cQsC5HWit4zjs0AgeSNXhaqOGUugxuWR+H7fXWg80jVlLX5+PI6lLKhH2s0eeJElkmzYf3LjaVZkLt7SpXIXN1+KKzVeY0cSctjRiO2ejHpNknuYMi21FA0wSI2xiAlldQK/Zr9r1JdOIf3C75hIt9Z8rseX3ytI2q+mrmujzvGYerP9Wie8SMYNY4rYvX47lglnxYDBktm+nrhXXCgFHH9fw+1EG37u9716I6G0A/wOA/0BE/mn+TpydPEKTVOr8+VtvpQXgEO91q+ii532HgxNOHKbrV9aag47YYWygWhmnrWqdj0WaRdXjLEEWVW4qNuqElVS67DD3O9kzw3UQYwGJQooQqaugXNuUD8yuLOqhx0OC5o1CQKCfZATtXxBZWb8rkTciM4Gt8/HQZp7mgeekU59nYGgVLrHy7xlMyYTndp8T8Jbhf2cAGIhytsGJCJ001jK/i1CSsGxZJEnFfKz2vgiOiDYosf13IvI/2se/aaoi7Odv2ecPlTt/WRn0B1uoEomw9kMjw5s4KmdAg3NsVpVpK2zOTM0cL4VRC2vJcpgd1YY9sG0VtQwuvPp+fCwXYVOEKYYyt946jkNPfuFEPL4ZvA/fPHWxWV5mrKv7w4awSBv/3vsmUyEdvJiktV+ZvotnJ+ZybW1ye1k42TU1NNtkK5PL7Zq0u2Bsrma2huNouN8PHIf6R/VglNkmzO8Y5oIzL+szQDQA5ZGUyveDUhK08Os/FJG/mL76EgBHGj8P4G+mz/8kafteAL9jqudPA/gsEX2SFCz5rH32YOu948Xdjv0Yh2js+4F37+5wd95TPlMGCJy7klVoLqjMer6cxVuq2uGCYBCXiOBsJ6lMk5RshIyS2fxcSJ9MMO1o6GcrEnSM+ojeL5YoCq9nvxLvuomdi3u7UO28bqMTwJV7rLewfVfSVoT3upS5NqZ8jY8JUHW2pDlb+/O5zXPsv197xmpfA4jIkfvzWZOQ92NKByo80OesUXh/197fEW1nkI/R3o8N968B+HcA/H0i+rv22X8M4C8A+Eki+tMAfhXAD9h3PwXg+wF8FcC7AP4UAIjI14jozwH4il33Zx1Aeaj13nF/7NhOp4CSu2jpAyCSWABS+JY7WQUvQmVGLYReXUUEcmiECOwQ9THhx5FKW9tiZK6YJYO3SW1bNkfvXYN2b04GingQcg9/WW+zyjNQwbGpwmaE33Y5Bm+easREIOQjjCWiSJDv///Ie5dY3bblPOir8ZjzX3ufc1+5tmyHxL44iRBBwkKGhB5yRDARkhMJpNAAC6UBktMEQVo83UAKsoQEkYx0IdDAiSIhriIjZJTQSIMkNjGRExPF2BHEQXbs+/A9Z+/1zznGKBr1GDXn/6999z17Yy3pjKN91lr/Yz7GHDWq6quqr1zTAaHd6vHYJyF7qlTlvElEYIkEKjykT9l9RGQRmByVZ2vB5ihqxKipzJ+WeTf/ThiUkQmANL3soa3Z2XTvYyBjWjbSvupbl0i97XgblPKvAE/q0z905/MM4MeeONaXAXz5bS/OFlfSnbo1cfgLT8Yq44ofLMFxKTpMqJoELDmJsYrXBMfAhmPdWinlAJAAtyYNgANK9tSubwvAOtAkSuijSX2YLbhgpvqOHMxXXxDhWuL13mgNfc2LajUR2jaAeCxmdmHyanYTVlJfMsWwAu74eLfpWPH+z6ZuJCVqoWD0fMzzBnc4xp2Nz8xLK3Myy4G7aL5chCSpj47ezKq4ZQiz9K4p1MdEiHcdzzrTxB76pRYJByA+XK3bAiyui6rpNzlp5xTLS8TxoQ4eaBpKOC8Ym/xYniO0bllo7Pq0+61iYSnlJtE4pgy5LzgGIDxCIuw4Lkr//pjC4tfm2vFW2OL1W4Y/Hdejm8/Mkqg71FwSujnZDMDwxoyRHs9G1CT3Nhj7DM7fJZoCgLkpnLWLjbOfdd5kznQWRBPs8jCMbmaDhzSC0WPtuyDRAuBMK8aOS0PSuaLb4JbBexjPWuASER7WBdBSG7Gzh9vVOU3k0ftQGBxHJNyVwXRhloyK3qTt7LmcJvoU5ywTGwa02MNNKWGQadsprMBME7Njy7Woiae8JhIjvM9obMPNSb2veK23vsX04SQrX7+jtyCcLl1MV5Zawt6HE/1IuOA4HzdnuCOI5+s5CBGrH0zkC9+Aqnum3fn4JlBy/UcNG6/nwJym7+VafE7smc1nPFFYs1is9RmxZqgkmdN36hUcxrMWOCJBGgcPN/1IVqxnL+RMypdhD45RtPzi8XGTSvABMEkOHTO0Ohw3CzYKXHyYKZNrV4uF+b879Hj2YOMOPSw4jwhXa3eYgJQOC+CHYwEBl9GFcPblfGPwdTqTAKSEh9WEhuakNtFoWkVRL6tq0mB2ywwc5icGqs9mYJxDFxoyH1KOlG0uQkrdmT7vntDFezSz8axxz8nomlkrgqWbq4Fetp4mh6XEefetoZQihc9J1p9sRp8CDcc8/aC92+TJpNZS8LAW1FLwzVevJFlZtR6I8Hjd8fi4SfVua2JyDKBruYwl+s5Fq74ejv6Ymyh9NgCJO6U/bEw/6d4u7AtDA9Dix4lzfzBDaS5OkgP49w+7PqYQHjYO0/J2Tv00KzlSrgsGb/OYAC4vXqBoYnFrDTVXMaNVcw++j5jG+79nSoopBo0/3mrAc1JyFJz4ehx2DEsmv/eZlGSTJia3re1z01ydiK0fI5qkYyCR0F2UU1reu4xnLXCAMi8DSm0+TY4lC9zPujP14HMxs2sJS1kajGDGKFiH6JcACQxk3MDRVtN2NmFswVrpyXmcF09ciLVWbPvm5t/UetqbACo3AdgAPOx4WLiHigUFSJDgQAwgIE/bO6RWThcdhIOfSDhefKjszrDAUVCCTN/Myc3CZMzuofE+7vhnZ6E9Wx83QNLJIonXIm2p9RIYXuVuz2MmHswAe6kFORuPJWky8yQbeh/jPVmm//8MImCtEhLYtVyCoWXyEC7CV9er+2oEMYnASkGXsjfpszKU1qSJY6yHOyNQWeN1SU0Jo1s/+yq+QA7XfD8NaO6i0SwWG5cDail9AY7fjeeyv+MxzxpC/phC2PYdu8YXx5DGj0gJRsfnsUilRugafhLKwAAaJPKMjOjXRWE7+2GmycqyHLT53c3rjklp/8wMPfvkZ0H3nzBNWJBS9vBOPOe9Z5Qy6SYUzOpgRr/rePYa7tqaVAcze1PBouUefWhvAdDMHhlzwQolXIb16Wyjo4+GJUsb43HngWUNkgNBeMZAt51dP3eORcVF/9RISVi4BKWUxd71vqZPoqY00YFAyI99xzQ7aAYGDLZlVl8pNi1JCTV8J6VJ5mM+l1NUMCQpGnBzG8BdbX6eh7t/B7DEzx3u8Z4paTG56K/FY9v3zknifSidYdgO7zGEnY9l/zqOZMF9ezOw9bbj2Quc16olQq6itXIKHldKSOm4KLuyL0v+JbuJZP3AzeyynEVD5qyyPCUSJO9gnt4XpPPuHl+z3+fr0PQhsxQTwO2wCAfP2jS91GOg+jTOmm9qenbkdOYDTkESU1gMV/EH9XsE5EwC5gTel6g5AoTzpLaIpt9gRm9N6Ao38R+jsMUqe/sbOPZ0ixvcOccyMn611jC4o7Udy7JopfcU5HLKeLEyKTuOJ0MXXWslI1F+b5LyrAWOSNl/CbikOkMD4SEwrPONmIDG8NvV7CHZvgU0UGBFzCcxHaVXuvgzOasJZtW+ZOcZDrDYQopJzk/Vz9k1AvCFJ5nsJCX8w2JFU3OJuTxNphxDC8CNKXfWqDFIm5I0FsyZnADWTWP9vk+SrUEGtKne7HBzUFLjbmpUXMTnazLwhMcMNNvr5rvGTStS5cWNKwrlGZlkZve1+2h+HMIAleUgbDEs0Xq76eHt2SoE8WHt9/cwnrfAgVBLRk4kFd2Q/nCDGaPDu3amlLAsGYkIj9q+Cn1mCghhTYDYCdrSSjLkJ2RMhx2WWRNaEyExeWbEOb7mnw27OqAlJ6MLcqr+mSGVNRcAjL7DkUvK0h5ZfKl5jDkh39qUO4ARScCQxNmbXNhn7D7id+33s8YSnks1qVMy1Obm3NE8jO+x+qulVow2u+PYv3Vdb+btfD2H8psAfFjfthgyoA4MUqHNt9Xo58JV8xEP88n6Pz5vc+82nrXAWbzRHOBdNVcfc4cyNiyb0DEGttYkg4LIYz+UErhW8GDUXHBZFmytIdpvMXA9hfPWPIy7q5lqIIDV6pEecNDQg7YlVrDEKvM8kdb8j6xBcQoIpQ5bZKYJ7VrMbLzn1/kUEjm6GwUt/m7cHVHJ2XHlM8F/5alp33jOcHypFjgG7s/X4M9An2WsjTt/xubDrz0IpTwf6UEBYtSy+HO657MBSaznMIfMlsVEILrlfXmX8awFjgfj9XUD6WLbexf0rBl8nj0gyWOgqRm5te5oZkmiJXkwKjL21jXgO0SzDRGCkqV0jIRzAAAgAElEQVQ0JuWM7bodFxLFiT9mLMyHSK6ARGh7qEAPXI9mFqeZqjUTcmc9X5Q41xy4s9faSaNms78xBeEeymm/y2c0SI5gbtoJGa4h7bLOqKD9fu9v1ucTXzsnBsRrj8eJGs3ei+Gf+Hq8NxMyM4vtPRFQOKeJhJOmOWnny+kInEXSqXcZz1rgBjM+etxQS0FS/8pyGYmEuixnCTobIGKLB5AE8ZyTMHvpTjjGQG8dW2u6wCQLI6eCWguYj1nulnUVzZ0xBpZlubvDG7RvPp/7L7Zr6kI+5AOe0DrTcjfb6ll49HNRczjLlR3jiXFYoEQw/Ra/JSakbBKx7u6pcRYaG0lqphzxjXd1FrwobPu+H643fiY+D6s2iJvhvdicnQ+kff3oHMuTrJycM3Ip4DbBm/c1nrXA2YNhZiy1og8GuEvG9xigTihZbfgxpIMKGEupIWtEtEtjxqZ5gyAJZmf9bskZ61Kl084QP61WybbYd/n7ns9muXj9lAtJOWvZvkL/XVolx+oXZuVj1GNLpbdB8PJ/0zS20GLK1+F8J8G/B6bcG5L+dDRfHVQIWTesKXFnW/dthNDu1811PcTZ1zvnnO77rs8o3wgicOsz23vx83ZcIvKqcAuZmDDnnD3ITSAgZY/BDpqZRe+re87zDnxD8iWXknGpFdU6rgTHlyGTxiDsTZZJKcWBB0JCHx2P+46PX1+x98nxkRNhWSvWWpBT0u6gExhxPy78JAoL3ACVbJR3EzWLhZSkvdgYrG2iJgAjmswevN63tY86n9vef8qce8N7h3mNO//pdRuJJrGrffAGTJETHg9yOpaFKRizlRSrmR7HOWEcOFLiiUVDFq24sSwMMDkLnCOy86qP34N0zxWTkTx5QtJcp+Wx99ko5F3Gs9ZwiYC1ivYBoPC/oGQWX6r6UK5b83xL/34Sk3FvkN7gJIHhlKRAtZaM1JUGYW/Ye8e2d880sOwWYJqSkbeSQGI+2oMBeytjAIdF7TwiDM3QV3OlH+vxPPtEDiDHjRnwd0xY3xyiianj3sI8+z6uQQMAAd1Q/HPhudxoNP38Wbu6ENGMaaacZ6jjZDXY7/Y988NEPlXFEmG0duCVifR6503SgKlpcgKD9fucgEA2NGCks0OlGpAmJOMQNniX8awFDpCd1jgxakroJWFT/shaCy7rgm3ffUVsqjmWWrV8h9FNQElg8nUpImwJACf0NrCPjr01Kd/pjDGEaPYM9QPqkJdAnTZY+hs8udDl3wQ05PWuZpNtHkfTMAHoWkga5uPmc3cECE/7U/fGPbDCTpqgLNYnc+6gIfmk0U7nJTWXa61IWoEPjclF3+t8PfMFsQ5Glwrs1rpbCAfWZT2eCWPMPrGflksLPvrN8Rqku5Eg4pIDkJDHp6AAFRCzZUBsX0GPgBeXRTc7m+CBWgqAgcdNUM2aEz54uOC67fj48Xqw8Wupnp7Ux8Cmyc9eH5eOTMQRVYtmo8MzzMqLeeSunMIgJUPiw7HGABn7tgl3PWUR4KRkpOH+oxlpS3KExWrv+Xydrjtex40g3BFE/xzb8VRzgRRQCvVxE7vxcS+P1EzvXLJ/N6lWtHIkGHgT5tivhxNssnsXciAzIc/BcfvOYGmY4j7pKaXMPjd9wVi6wwGvUlPzW2xabzvehkToKarz/4CIfpWIfl7//ZHwnT9FQnX+d4joXwiv/7C+9ktE9O/dO18cDFlce5NwQCLx6WqZqTyP100mPyfkRKgpo5A0/rjUilqKVxz4Q6JZ3yb06GzW+wEdjP5YRL7sWFEAzz7OYeERIZeEUosEYtUPtePUKtcoJjOObMd3BOUMkz85fyetfE8Ib46jPllc9BYHdW1m/6kvFoU8fjduVqUUl8xoekZfnIJGipp8jAHCJBoqpWBdVyzLckhIdsHiLkBV0HTTpyZJ9zshlJ7GlZUpus7WYV7d/x7G22g4ozr/34noQwA/R8KaDAA/wcx/On6YhAb9jwP4/QC+B8D/QkS/T9/+LwD88xAS2L9ORF9h5r/9ppObX5E0ngaIVtobw+o3cxJh45Kkl1wirEsRX2nM8hJplG4PB7Lwe5NFgqDFlNcjarmUZjMP29oHS+pVSklQvXFsuwRYOMAWme6qpkkhwkVp8jOekcg3CdTZ9zkjf/HvM4r5puPPqyOtTD+biKb5kmtAP985jqjazeoPLc8zQu3WfpmIZj/xsCHs+66+GJzV7J4Z7BkklkdLdPNv8AgdYudxckkoJXlcN6U8U8V6O3RXfZfxLlTnT40fAfBTzHwF8CtE9EuQXgIA8EvM/MsAQEQ/pZ99o8DZ5Jrzu7UrXm9KC+CLxRAmoNYJxb9uHXuf1eJ9jAlPs9TRgWbN3RlwiV137DURyEAoipCVgCmcMWZkS7hrWGJwBwLYsfehjNE4+EPRbOXw973xlLDFhXvWbvfMP9dilnvJ06w8+m7kG5Ufmyz9S2wFy8G0DcjAkpkQoNkhcoCb63WgxUzAoT0YNBwDzDicxTXNR6ulHpp++DWG+U1JKC7AYwJymEJMRJIY8R7jcN+WJ0hE34dJdQ4Af5KkQ86XaTbmeGeqcyL6WSL62devX7kp0HrHx49XfPx4VYJPE0aJoyVNTWptYNs6Xj/ueNyMY9ISg4Xcs9TkvQiIUoDvSXuFF7sWANPEmR1ejmlGYwxQBzIda+x85wcLKLPvUuk9JgI4dEdWvFzIj+Li04Vs5zqbk4fFEV6PJt15Icd7i78ftJ2ahwek8nSeN12H6WmjXM85I9Os0MgK4UszjXK4r/O1z3zJeU0pQ3xiOoIuJsTWiefcYZUZouEoIKElH84ZwZzOQndeltkR9V3GWwsc3VKd/xkA3w/gByAa8D97HxfEker8xUsXhtYHHrc9cJeIiVarII6iKaTWrLMmDROpMErW/MNlxWWtUoIT0M+bhUhGg9cP3Bn2MP2zmD4f6R/3IHkvG1FgZfB03km1o9RdiS/xpkV91Jy378e/o6Y9nO8JYSNSEzJpwakJ/J1U+bumaLj38EmAROMI6JW0kcj8l/XfU/d9yKMMPJFWIEzKIGVpW7HdlV2XoZd9DLTOcKXFM+Z3nhvbaHMpd7sIfZLxVigl3aE6Z+ZfC+//VwD+ov75Jkrzb5vqHJi5hpJkS5pwrI6v+mdb79i2IW1lVaOVrCZFJiQqWFe53e26g0i0oZiZcWEyzBZqLdTEjUDhMBQFI0KJeXj9fnxLjjEAFNAQOm5jhcr6OemDwNNEfQMoFqHsewIUX3tKK52PZb+7ScgM5OxmXMrJww22USWkg9kYLgKDJYwDEo0iNIPSxWiM4ajkQSsakkgq9OFch+dzyrexmjZhH2MFP5JaIozr4x4qxue3Bc1kJ34VV2T2/LN19z7H26CUhDtU56R9BXT8MQC/oL9/BcAfJ6KViL4E6RP31yCMy7+XiL5ERAsEWPnKm84doVxjnipZSnGqxsG2vUmjxp3BJBPYNCugloxaM9aaUYoEsh+vV+xdQgBWfZCyIIjmC45xZNeK12L/eu8YbTYTEeqCKWjT/JOfSymoOR0Qz6I7fsqzeDKldIh7vcn0O2u98+dt3MTpTprI4XFNe7JskCjQEVUFHYEUN8Xsd9uYeNLvGfgVk5hhM56SB8RvrInDZjDh+3gvfu2wigv5d2basmOSrifmIfG2gECaCXvPDH8f412ozv9VIvoByJz9PQD/JgAw898ioj8PAUMagB9jlsIVIvqTkH4CGcCXmflvvfnU8qDXWtB7wxisCcYM1mJTyXfUHHfS0hp9+MsifkLv4j9dr7vstJQmDD3YGYZ7mxOcU8IID3YGTX1/hMXYDo4OptawnTclOjSGzKWg7bsALQkYez98TyrB3/ygv6UPdvrsU0J58G8Izt9BCNpMPxubkZxVsAFG9n3o/CYWtNeKhynMg970QePYtRgYQimh2Xz2gVxuNbqbkjWa47OQNfpmmWSrkAoBAVEiSDO/f9wo7sYXP8F4F6rzn37Dd34cwI/fef2n3/S98yACSspY11V8tBIKQ5kVqRQAxCq8GXC72yZPWL1sx5KdLxNhLdL2yqkG1JRjaA1dtUYXh3sAYH7CTDUjIhhxj5f8j6a7qByXdPtNSYox+7bJTbKZVLH0R1KKzujj/Xl6WtDuCdlt2IK8Hu4A2Oj1xs+6WWeCY8e08+mDS0SY2K/4qEwAJ/Jsk9haefTJvOa+nYZMmAWil41A54YJkYziDLQIjK/mPCbYlIvkz4pZS4DSnVvAOwqwsTSfffd3Gc860ySlhA9fXrDWjLYLR6DRk3fNmxTCG3IgouSMkqX+bdt3gIHrJrG2AWn04Uii2h6D4WYoR9hYTSARAi00BbRiW0tjaSJypFo2kznrGUxHej8wOzrXidBfDxAdczbP5uQZ1rfXzv7aWYvF96wsybSuETLJfSZHZ6VYV4+ngmh+zY3wRu143JX0dRHipMfMSZ6hlN9D/G2d74EJkHizewCjzd7tOslIJBXsYxD2vvkmdc6nFMqN7tcCTEDFOC2NMtDmyDZlA1liO+T3MZ63wJFQnfMYAunnLNqsS1U32AROBNEYt2TSBlofGANoVth50D5dEE8SJmY7RmtdslbKZKsC1C/Q2FQpBVlNWgC+OLjPReGnTARi2QiGZlTYGJhZJYJgHolNoyl476cfZxy5Qs5COIYkZ7e9Yd938FBzy8wlYm9weHl4QF6q+FRFzC6yxcYzI0etsaNJapuATFxonCKlTB4TJ8LAnCuD3rMSrzJmx1lb+JIZMv0wM/e2ffaBsPuNgxVBTvnYMTV+VrQaT83JssZMOL1ByHsYz1rgAMZ12z1tK6WC1hp2M3nkI2hdegXkZcG6VBGeLou+KwBhgVd76F2bNvQ20FlSh5iN9Ie1uFVo7Egd8JwzEgGlHCuE3Udg9t1RnQMA8vmsCF00z+L3LV5ooA0lBmP4znweZ43jpqjV3EFO3/Ydbe8icE0FjoVwB75QB/Z9F/Kc1vDw4gXyZQUgiQRQxLCzxg/19siyUO5YtAKuwFnKxBSc1zvpEAICSbBOJxhgjNE1hQwotapGlWPkUpCZkbd8rHLQ4SZknuli5s8fUc8plMY4nWHP6mlz/ZOOZy5w0xmPJShGvW3xuXa9+gRe1gWtNbzuXRKfx9zFasnqMCesy4Lr3gB09Ab3HyzXctEauV3pGiQKIf6emVjDzU0FOVi8GtEqYkpaeckRcFEIvJuPN4LQTY2qnzyCDDYvp79nKQzb19CaarWtOWmOaQzW65UNJgBCOo91DCwvHsQfhphwntEBrQAf0/89yB1JAgGSCKV2hZPLChuUad+zRo73Jx1I5zKNcVHLl4x5rlH727OJmjIKW8yVzUnWRvZskwTF+m6E+V3Gsxe4mEC6t47X16ubcilJ8NqSUl9cKi614KN9F0FT82b4bjYfqtWkxQ47Ywyt/haSod47EiWMZFv6bWWyPUgiAvchx3PhweGzB7RzdMnj1OaAU8PNB+yBa8CDR/dMzvmC/mTZaEYf4C7fz3TMU9RfQIM9ljaGEN5eFcwxkzClJEIT/FA/DgVJE4c2/Dk1IOGWPNeeYdy8zhC9lC7lA2o7mHG9vsZ2vTqaaceRRhwNu96Dx05PghyfnxxTrJylVt+c9rYh5fWtgKu3Hc9a4OKuNcbA6+uGbW9uHowhzTBePqx4WBe8XCsUNBRzcGjeY9AglBLavuO67eLAW//nIl9casFSCqQURCF9ryYgzeO7NUkApVHXYwLwBVpyxiDyHVbSu2ay8gQ4ZrpaXAxPjfh+zAO0ILCXoxCBU1IadRLQQoETMGOcGxOSgEV939FLRle/N2nb3kHkfB8zn5IPwiUXKPqZCX6v8drPfmnM7o/cLEZsxCx968YQrQwC1nV1vzwCInFuou8W59Weh62lslS/nt4btuuGdV2/rTX7rcazFzibjN6VkzI4vOYsF+UQ2dvArt0ta8kgreR20wLSmE8Cy1AEUs+DBMpSHjN4qBlmAVV47mWl7PGoaK4cYmi6+A05FXLamXHOmAtOgBSh++snregajI9a7d48+c5tyOnoYCYveSFmpDGQAOy9o113DNWuVBJKOcehNNlANwfm4pA/EYGTzB3D+qVrdoxaFQR4db2s71lRHk0+F6xwL1EQmVn9OgJrA0xm2UjX9eI9200TTzM/e31iNB1jPNVe8yRnIux9x7Vd0XvDUqu7GPHZvMt41gKHYOq1MTBYdquu5qVoM0EVXz9eASKstaCUjG3bne7Ad05Ngs0pgbNliOnDImngwWDhtewdOZW5e5MEXpElxhdNRnl4CbkQBpPv+hZXa2PWfHn6kppezbJdzH9MdGMC2WYdhZBolryc37PQgvmWKSUMJSzarhu++bWP8Bu//nU8Xne8eKj48HMPePGZD2bSNhB2+o7cG5irnycRMMxvFZwegMqFzgWRMIHRUKGjeD9HTXfeRCVGF3xKAB5+Z+v2Q1hXYU7btu3QJjqlBGRCJj4IVCyxsfPExHVGdyqOVUGj9xUOsPGsBW4MQdNal6rswdPB3tvAGNI87+WLVRpw6MMeg3HVKu7eO5oKbE5CmWfazuB7IkKp2ft7H/0voRlg9U9Smp8z00XACgKlDDLfSReKYAtHjhI7rtqO4C48lTkXWRwawpjmzy2oYL+TOngHzWTAS07avhignNGuV3zjq9/AL/yNX8bf/7+/io9fP+LlBwu+73s/j9/3+7+EdV3m/cgRZCEOBvcGcFWQxKUckq0vIAMQKCDC7yO+Hq4/QvzW3cde1xD3vD//jvJGGq+k+uLRNxO/TQtJS/GgtqwppThXRrADxZ5tUmF+7ee3Mu/fdjxrgWMIy7IFu8nSbVSwAHnP7Hcew1vobnvzHgMxAVa4NeS2BzMoy2KogZnXdj6yuA8yBqTNU6nldvETIRWp5h6DMcYGUz6Ws2faxgK8kuCu5s4Qvn7j7Yo+C970nNW0Mr+SSNOzvIkJA1k5J1k2gn/w//xD/J1f/Af4jd/4LTB3/MY/BD5+9QoffuYDfPCZD5Euq/pNMxdRq2AE4YWcQ3zkcRD0adaGoT6jMWSfUUpb4EZbNwVGKjpEwLQgeEgitflsrbWZgRIELrYePgvQvjcQyTHMPztfT7SKogZ8H+N5C5yhbRoPWWqRolAAuehiIoH7l5KxtYG+DzT14xjw+isAmlO5SR+CNMEYy9h/fb0Kwuf2fchKQT4sjkMglKZGgtKtQRNtrZyIecL7550+Z02rGto/3GJ4wXc7mpj6ewBK7B5TSiAMHNeH3Md12/Frv/Z1fO1rr8AsgUoewFd//Zv4zd/4On73dcfLz+iGxowaNha5foazQ6vmYY9VKXqploCbviySeg7Mm/lqieAWI5P5mAJjmkyegZq7FrwOfpgJ6pnBK87NtgsqfVnrHX9VNq9aqyeRt31HIzpQObzreNa8lAzWwLMIlqUeMTOWmvCwVHzwsODDlxfUmlRAJygB3MZQrtsG7l2y9PUhMrNnsLTW0duR9cl+p9MClGuUdbbvO/arNT7sMM4UU1EzEmXXlMScVfPHqBpEYG7La1xQMbTnuMTs4n2mpCYkkbJbMUYnMBMaA3tj7HvHGE2FA6r5Or75zddojbFtDYPJhSjGyhwcgrDLHa4NYUOw6yKCclr45nEWgsPzPr1nfq3XqqU0i1X1OqzQ1DbPQ2wU01fbti7aLU1ht3PadS+1elYND0bfmx/3U2FSSjxpeOVu1BA1JVxqwVIUodwFJHENoTuf5U7aAUvOeFgv2HvDngglL+4HJBLaBNk1j3vReTEYfO6gxuAJ8QOKkB1ROYTFi6z9x3BcJCaMlkAbz+3nPY2zWWfVELv6vSkR+t4x2szaP0/0DsJ1b9gbIy/HzWqipccY3sHcpaDVSONvyTaCpx/xWcCi2enaTc1mqFvRAgV69PtM4GJepVSTNGx7w7IsWOrsR2Gj946qPc4NgW69gcok8/1UpHYxZBEXFZpdE05NO5UstWSjWf9qcaozgFRiMwapDrisC2opIALSkO8PnpB/yglEQz8z08eiSWoPVga5NpXPmyBmbSMcvttmWEBSsCRheISd3xE5Fdq7m6qioPfg85SSVGwP4WphJHQGHh879seO6yNj7Eo6aPz6eh9I5CETB3j8lAHoCc/G5hdQQiXFdJMe0o7D4b5MEM6FnrXWuz60maWE5CCQaF0xb2udqG4UMoCQc3ErSTSlrAXo39EPNNM0nj9usp8KgQOgwekksaPePbisSgVgWSi2uC07QrSbHIMVOLEe1ntr079jbcTYjW58xtM8VhYCpvaQjArBYm2AunKA95Njng+qKzJmi47VZDPEbIyBPdTfiRd4RiTthuZrANwEJbsAFZVSM/bHju2xo2/SFZZdaoKmwiQ8nfQKcK0Vr8EWY+FZlGrXkKD8oUQYQTMSJvX7wRznGRIwdDia8XJf8swYUmFuJuG+yZzVE9XI6IZWKTsAA5QkLzblqTWropODGWPbXDsaDcOyLP4s4s93Hc9f4MbQOMvA68dHbwaRc0ZrXdBFpT2zNKmkCza2oh1j4KqJu6SOMJEI68KMHV0rDgyKhj8Quw5fEKQUenyvwYc+UA1T2AIrOWNHAE3GUARzUuTZMUzgcNIic9hCTqZGDtdgn0lE2LarhB0A9CEsZhaaZvmC0gJaAFssBFNRZ/PTzwMEU0uQ0IRpFUx5nulhBkiY0NpzObMoz3sRfZn0mSwqJG1vcm2DNS6nFBYMR2x7Z+0IxEhMLryUNLNTr/OqMbzL5YJSCrZ9R10WZKJD955PRXmOJOB2GB2CleQIaaqgZk21z3VvIlBjmhYxgDyYwdpo3ejaUkpe+LjtTT6TM4okEKGF1CvfDbuaXDxrp1KaeZ0x3lRLARSQQc54eHiYPhzsOzOeBMBjhwbJTH/udqeNpo7EkaDhiCkshISUGcxZNNxQ1YUMKcgHRIAhRLpamSEumrBbUYL8i2BO1G4ETVB2+ZfXTFuRCIIJWcx9PPcIsHuRY2lyNBEwTCOyg1pF/S6NhUzrYQwwZGMRX7pjaIMO6nLB8syFRa3UirIsSKXgomEno4IYull+KjQcQ7JBPOsAasYMRmagDQb2hs6Mx8dNOuOQEN7YJLXWhco8FlHaDtoattaweb+4I2RvwiwPVYAYhra60tQh0QS6CNNRAEwrIvgXB3+AZSHXYBdZUEEdwqMpidvwgAMt6oNlmilWACtgwRofTFIl7war/RSSn5yTm+TnMVSgD76YzqmRMLEJJE9E1obxnKSUpECXZrzN7ivOEaufBWhhKAN7MyCJUJfFQSsAKKX6MYzOAgDQ4RsYNwZqFYqLbUMfHXVdJ80iSc4p1MR3kC6lm/v5pONtSIQuRPTXiOj/IKE6/w/19S8R0V8loS3/cyTEQCAhD/pz+vpfJaLvC8e6S4H+1LAApJXO1FIlE59V4+0N3/joFT569YirthlurUsWCjMerxuu+4ZNbXQ7pglD6wNXRbAsG4RZKgn21ry49TaNi8BtCIlQbzPTvUnd2fV6RdP0MDObIrTMzPI9bg59l1JwuVywaN3XWZPZz7Omm+aXLPLR9XfLtsgKNiBUNIM91crHmJ184nldIADvQjM3pamJ4H/ZXkEeJwOkqtz4IikI2XkDseObdWJpcGZKttblOhDMXZLEg1kzdxwRPfVC4yFlPYsW3tpxLJyxa8WBuSfvS8O9TRzuCuCHmPmfhHBQ/jAR/UEA/ymE6vz3APgagD+hn/8TAL6mr/+Efg50pED/YQD/JRG90TBmANvWwAwsS0VROmp58NJa+LrvnvFAJJzwPJpMWGtqkkZWXlkerQ8BT3guRMtBtM/YiAvPgquiCRQV1EyIROm2OSPNpNpjtvoMtJrAGWgRH27c9c+vIQgNjwEM4Hrt2Pah4IL0Ycgla+qawu5qetosK63jwXQ1Ona/nqDl9YJOYQGplKZEoBzuwYVi+rPMx5Zek+j1WIFh12PCt7eGbds0BKT+n2qvRLq54HaDohAQZ2ioKWXnyozCbuvu41cfo42ZMva+UMpvKXAs4yP9s+o/BvBDAP6Cvv5nAfxR/f1H9G/o+3+I5KqdAp2ZfwVApEB/6uRgCKXBhy9eYKnFqdgs3WtdFixLQa0Zdcl4cXnwXMfD7gl4U3p7UF1prLM+kGq02wqM2PdNC62acVBKQa4FqWSUog1DklaT022a0VnoACHSWdfV4Wmday09CX7MCbQ4vMZhoZDc37YJu1nOhFoTLquksi2XisvDipcvLyhZEnUVzgM0faooI1pK0Ar7pKDRrACw0Xl4IkzSNDsicyKPsT43kQGfe1vApRTfdNzkzDP+FQPfDMZ127wAGQi1jrrhlVqRyxEsMyGMfeIIoWhX59T8tn3fQZRQgkBG9PRdxlsdhYgyCUXerwP4GQD/F4CvM7N53ZG23CnN9f1vAPgd+CRU569egQA8LBVLSbgsFQQ4jC5oo3QwrSmhkDrMajrYgjfnegzWUplxgLTFr1DAg9lNIWkSMs2OSJudi7SmJejuGpJlD9A2TWd+onDTxDqbjo5Q8nGnju/ffJ5IWjOB8PBywcPLirVm1JxQ14JV6eMeHh7wxe/8PD787MV9LQBYLxWf/fwHWNeqiGBybQAEcQu+UUraqhdTEM9ml28kmO5BLDQ1v9oELHJ2mj92WdTHCpyeZ21vgm7znEpGG+JS7NbRliZvpo0Y+0wAEBDTFw8PEo+17J/fTpSSpdb8B4jocwD+BwD/2Hs5+/1z/SSAnwSA7/qe38mf/eAFPvvBAzIR2t7QRz/sUktOWJeKx21H23bsXTJOJKg8EaYxpJ4ujVmSoWGam4RVAA4du7/CjKtSr5mPVWoB937Y+e07MxdwkgRFIIXHwD6GB8TdjCWrrLYC2NvE2fOiGWMg1wzK5IF9Cw+mJp/tnZHziu/6R74Lv+cf/yb+7i/+PWzXK5aa8b3f/x343d//u92XTd0AACAASURBVPBwWUU7nHg0fSHbZkAE6RNK6GDlhIHPi7Ghna81VmLEJIJlWQ5EPYdUKhJQqbO4AF5uA8LQHExA0F1Qx1oX0BBCImIt+tV0v0TS+dbAMI4FwOafAhKjC3+bhfQ+xreFUjLz14noLwP4ZwF8joiKarFIW25U53+fiAqAzwL4TbyZAv3uICJ88HBBTuTpOeZbAGL2vFirmw9bm3QFpml8dzWmqiQsv6SIm2WbNPX5kiFSIfvBiHeYWVm2TBBn+lUULLt2+45XersPJxQLnXlqa5aMGgR/lOjWrLzxN8LvJWcnyu3qMyWSNLOuHC8ffuGL+J1falgeXmD0HZ/97Irv/p4v4Atf/BwuLy5ahJucocw0TbYY1mkDMQ1n1Hru6yJqZPY5ikJnP7dtO2xS0YQbg7XYld0iMFTXCWLHAHUCkhIKKcOb+4YQc5HURy76nU3vwTKLhD9Fzt8B/3wqBfTb5cMR0XeoZgMRPUD6u/0igL8M4F/Wj/0ogP9Rf/+K/g19/y+xrIinKNDfMBh9dLx+vOLV4yMGDzXxMj58uODzHzxgLdn7r9VgzpVEWKuYoGMI18ghYx1m2h2Dxq13oUA/xYc8QbYbMtk1UzqwdoXUIhOR3ju2bXPU0sIbN36OXkPKxt9xS+Fgx77zjOY9ISZCix/32Q8v+OxnXuDhxYqUC1588Dl88bu+B7/r+74P3/uPfi+++J1fwIuHC1K1OjL9fjDxLOAfET9gzl8EHDycE14ToUyH40VrYNu2G39VUMUZmonf86wYILxubF/ik+ZMeHhxERLa6GIErdiDTxktEL+2w0b77uNtNNx3A/iziigmAH+emf8iEf1tAD9FRP8JgL8B6T8A/fnfkfSF+yoEmQS/gQL9qaH5ARL4blNoSkp4qBUv1sWrv0tWshkUJLpgXQTI6EOCnrYAiaQqXJ6tagZKgMZ1rK+3gx1qHhk7lO/sAbFjZmzbJr8P9mNH38Yh7jAOWkoftIAvyR14+/7597jwLZgs2l/oBTJ3MCXQQkiJkdOCy5LAo2PfGpYKrAvhww8zHh4kBueADzPAp3KXNP1LsxqIprmVcNutx0GnYJbH92OoJt5bBKsOm2TwWQF4DaQjjfYZBihlXC6rhwEMMKMsea7ui4ek+HgOWx3cO6iU91ZW8zZU538T0hPu/Pov4w7KyMyPAP6VJ451lwL9yUEy6U2rtg3okORYgYkfW9Omi5LSdakFL9YFKSc8XnfvmFJU4ErJBwFgWLaIgCOGXllbJd+RAU/N4tPC0XvTmBGBjX0AEw07mpqz6ti/p1qR+Vao7Pj3hM12YxG6fKyuYNE1KUmSLyEhffEFXj4Qen8AeEfNGr8DA0mZqSGJ3USElDNYsz0k91QQykOFA275UEg3KtcsAUgywCQKWrzX84ikQJb0VmuVa9H4K2OAOwAtKB5EQLKsE2inoiOD12GpqUbnqOGgFgdJUsH7GM8704QZHz8+CroY4liUE1Iu2LtoPjPgeDD21nFNO/qjpGtJl9N82AWNPIg1oNpVoy2leL5eNCsAXVB6/pgBYQvK/nXYomDf1YFpmsoizeh6LYeAuJs1U5ji8eO8HDQJpGNQ6RllHBcGAUpwJT7humpJ0t7QGgCWejiAQBmokDiamV6UBaFLOU/tpozVb4LKrR0zSOgSLBF6nDR31GrnzaW3hs3BLAZpMnhJyevxkBI6SyCbeGCM7JkjrU9/kFlagWU5gaCWVYhfzSxOKcnzs7lVMGUoedH7GM9c4IDrbo0MOTwU0niV0I0LugfNDJEvSmZKwQBhHwacyC5ofrxrOl0cjliab3byyUwQz/7HTNUiEKJvptUEQUD84WfhTzwcV2N08mxvd/2zcEOvt/cOSpLGlrVo0qggkom/QnUzTY79XlMRzQXNlyzasZSW6p1K5brlOOzmtHlndrdTs9FpgSaltuh9ce0WS3TsXmy+CUDbr9ivMvfrsgBKQtFbQzeNxIw+pHeEMP/qXCfIswa8X7uFfyTQmKdvac+dZ0jIhM3N50+4hs/jmQucNuTAcQccmsJlmg8IZlgSzWHUeVuXui9zjl2rjeEU5DMTfATTTxYndNInmez0PWLglEiC8ZZ1N2CLUnkVg9nUW1PHXruwmjbD9I3sOu/5Fva3+ZpzTloAJaQhZbiMwzFcs5ARJMmmU1JGygQq+QY0sflm1TYmdIRZuAngsDjNNCulYFkr2BBhRX5NmGMKnAEgzNlN2ZIXeZ8Ig2cMbyg/pTwQxsBAbzuAoa6HmIPF5joCVZhMy2Z6tn2/CXabyf4+xvMWOMhCt4fppsaQntpDiUyjujeTbO8dj9uOa2hp1dSUtGM3rfL23R/wTi+DBYQkvQbPWgh+k12PhyB4R29Cr5CTUO4NIvSgTZv6HDaippLMCiM/PRL0xFqxs9DEe7deeERm+gWBTQC6QviDtFSFfafPCv8jCxR+CPZz2HTY1vgxz5BOQmfCRmQm9hEUiRuA9G1gXDVulktBqYs/G9fsSXk+eYCGtJYW0JcVnSW0th2OPTQkEP1dQPyzFJ6jc9WwJMCDGUXLuMbj49su2zeOZy1w7mjd2eVNQ7gEBfOOEmHbd6XKk1Ql29eOWSAShyuloGg+pEHA0sJqePJy9KfsWgy59OMycG2yc9ci9VjRFLYCR4PIwUmy0/X7VZNvxYQhX0j2/lnAbJiwRhYrACgloK0JEKCZgX2AlwxuqslZQCHh2E9INR/664mfOxc+INTpOov+XxxHYTvS00XAxEatFTwI1+tV5jUnLMsFAND3hn3s09/VTbINazE2wSkO1ofPHaZmizHACSzN0Zmx9x0LCupSPfXsqbn/dsezFjhb/EVZnOLDknQrySgArF/c3FmHsmeRLlqj+wbm4q0p4YOHFUvOUmsHFUwS0lfL14xZI9HPsGs0zeVZ8yzxvEVNOlMh7oMxe/a95QsaGGDX6wiFjnvIZRymaZ0XX022Mcah/MdyNw+xpiE9FUAz99MF6GREODrKs9CX9HNius4SqIPGS2kmRNMx5CDHYbEEspX9aA5rLmggvL5ubjLL+YdQOpDwZpK/PrOLfOOpZcYREXocaF+FUopYN6UAyjcqtInCbXK5XD4dJiWlhId1RcoJr18/usCVUsCa31gVaQKaLwp5uN01lpX3mHCUlPCwLKhFiIjMVLIA+QDweN0OvP/ADBccAZzgXybJgohBc7XAFCWkA2pJlEFlonfZunDiIGt+jqeE7mze7vt+mCszRw9B4+CjOLwf7lVxFTBCwoABSeFabCQVUiNwAs+KCzPTo8ZelsV9Pz/naCBkLIvkqUq+aQJQUIeALSbIKUkHpD4GrvumQFSWjBgVPAs/GAelgWDLsuDx8RHb2LyurpbiwtB2YV+zYPdjfjxUf7/LeNYCl4jwcFm1O+lc3LITayAbLEFuIu89wMqgZZpm10lPJNQHtWRhbyJhtmpahAqNyXUOYQjYTivI3VoqrrsEuW0hesskPmpBcXzgVd3i4Efji6QPXPARGeJbDAUngFtf7d6IwmJCzzzT0myRxuTqe0HqCAjFn/Y8EDchogNOaQiyPaNjDwY9hurEoRqMVKibglpzc5utxQYLKzVISnDWZcVlXZFzQesNW7NyHVZhlUr76xhSm9hFM15ePKAsi5BOffSRmq5iPhs4E/1k26QfX7/+dJiUdpNtb2rTLyi1YqkVn3lxAbM4zcQa0OxSNCo93WSXs4B50gyMh8sq+ZNjeDCztYbXr68ArGBzpv60LmimZTUQzU6s9l3RBto80Z/L5GYc3KRNLinPigqTQewGdFiepmtMYs9njFrsqQBxnDPzT87+p2mz+Nl73z2/b1rXASwEPy1+1o4BuhFmwLrVdq2iNl+su/kezzt0w5SSnlmsKwnMFSkXIN0yLddFKNsHoBvOcHKklDNG66Cc0UNhcvxZa510GQGlfR/jWQscIClTjQcSJVxWoT17uS54WFdc9w3CstBx1d5xYKAbGSwkw8Sg/4d1wVqLgxfABBzqUtFa90k+m2zbtoFLQc67LmKWNC5RWmp+KqypO72CgQCA1sQkkUTZhD4aGIQWhMJAogPqR4oFnvy5c0D8fL33BDQipZYrSeG798xNfQNeWuFPRu4zmr/zuuHX5KYqLEymWhw2TwLJ9xNlOTNjbzu6V4DxJH0FwYll/RkkLNoqmW3O9GKsj3fvA/u2CY9JKRghnc+eeVLzEjj62G/a5L6d8awFbgzGRx+/kgzvKvl+VUGUbduVFAcC/7eGvYtmswYd5tMQCTnOqgWs/eR/EVkWCNzkI5Im8IkyFvXzeuvYkghOtxADwXnvD0JAk6desuQHcpaavZG6B2gtsZmI9OdRo1mQGkFjAEft89RiOOc2nr9HCtKYxDj0HgUZco/KWDLNU0xlHoXEhEsnUY+hWSdDzcgQt+s8DtotCqkBZWbO23wQZhOPbb+ij46lLqhFqwh2YZbe992FBszo+y5WCmTjS0RaKX9MkO5qDhuQFP31dx3PW+B44LdevcK6VLwsL9x8EbhezD0we0Y5iMB9gHn6KwBQc9I+4cm73tiQYPg0tQazBNVZGau0XKWmhJEz+mBfhYM7wIRUABChD0YGoy4VRMeOn8tSUVJCScCuDMitM6CFo65l7sDrJnTGlRm1l30GuB9COEPf8TvTDMSNYPr1GEux+WBvZVrR1GLQSnuGN4C0YPO5EiCW7sTNwoR0YGiPB+3zxpLU8LCujsRGukEAXkTsYJT6bfoh2VirVPBjsLsZthaO/um7j2ctcAC8GDIl8jxH+QdHo2SBDWAkX+jWSI+gNWFjSF8BRc9Y06iMQgCYgXbPqCBg0QaMpAtFjj+kAJPM9ExgEr+sDwaBUVMWM0lz9YTCwZDQ4fQOKWWAxK8RZG9yNt4gkZZVQQH1xFwM9xbFPWGz36O56bmmUfCMoi58Z8AQyeMxTSMBdzYNAKSmrE6XVoCEpGSaNBR9dKlvA3Qz1OaWDGRNWRP/9FjZQSQpbUP5agaMjTr5Zg31C+0KU0ooSp3RR9M0OW3eaaxk6v+9j/GsBY5AeLg8ICXCw7JgWao23GgH008Yt8QBtyaMZoqUTDc7lSFsfYi2TESHzHJmnnwdScxYW0R9DBhJuPGYCOytFHSkWekg9wdIBWQM4ULc9xYKTwHw5PLg3oUV86S97Fru1cNF0/Uwf0EgI1Dipl9S7RriY/4ZBMGxgHdcc2o3GvHqMGdW/8X1SZh8LpYZEi0NiwnavVASYfKYp24+Fn8jFlJbA6YsiG0x22u/qts5tZsdy8Ilnq4W5sTCR3rLbhp7IvZ7GM9b4Ijw8vKAh7Xg4SLU06Ox7FjM2HsTAexTY3iOHYtmiwiW76IKThg1nrLtQx2o0D6KlUF5mlbSPZXRBmsAm4U4NWXQAtSU8LBW2aU3ceZTYtds+y75nB4jUqLVmLN4bzd9yox8yqyMI5pnUdhqybqTjwNLF4CgEc6acfp1pNiF/Q1IJYd1RI33wQTpyxcsCiLCvm2e+wqCtkcXHxgsZiMpF7TcxyTJjUkIYwwkHjDSNDE7d0eBz6EQs5RSkmoAC+RnZYcGhLfGzuX39I7j2QvcuhSsS1Xfajgc/7htwic5xIwz03JZFoWscah9i3EjQPg4jmR4cNStlIylVjAPlFzQlUjWBCIDnlY0ANBglAR8cLkgJ0Evm5YNZQqwNcuCZo0LGipWasa+s8aE8syaCX7WDdp40kpy+bfI5ZPvMzzOBWZYpDNG1ubcwS0KeeP2OYHO35lmugXP67IATOhtF7pDFGyPV4nDiY0SUsqMImJqKDNNbf6sPpIx442GKA4NNXTW0iM1eeMGDL3/rtQcljdpx6AkWTD5iU3wk4xnL3BQ2B8hPtX6wON1U6q7+VkL6kpr4YTLuszmH6xV1SwwslEA+OLQxU9qBgr8PHkM3aQZA4/bJr2glWB1H10QuMboRN551TSJn5NnS6tlqahldlm1BSMbyrHe7QBw2NwkODR+DhOcP3svfmcMy4lIqqBZNgjyqnGtezATEXBBcyQyPCePy53OY2Ow9FtjMFImJBJSWDDjm9/8Ld+MCPBmIPH69Zd5Tn19jOEbaR8DsKJeBWQENJGNNxcRUup0uIdYcnXjyxIErf40mJQSi2kgmoRAnaVkh9QHi46tPZiqTF4JAGvcx7n72cJdRw0hpt1EN9veFMHUotOkZTAsMbgBBrTcRx0/sD5IBs/8RJhPGBYlNIBcpMLYiE1zylLnR7P/wRkQcX+Ep4Y+C1x8LQ573xKtuy7WBMnAMaIh948RzEo5gq30o7kYjn/WqgA8U7+PAacTI0kyyDWjLvWoFZkPz9M2xo6ZNO2ETGp92uW1rpT1mHV9ceSUQUnhfsuHVRAlbsBxLo03832Md6E6/2+I6FeI6Of13w/o60RE/zkJpfnfJKJ/KhzrR4no7+q/H/1W52aWMhvr4ba3rmakONmXpRyqr+V1RSSvG15fNzfPZIeyHtx3OloyS2nKYFyvVzzuG7a2C6GQ+mq2o57r25gZ+7bj8fHqGhUAWpvU600p2+3cvY8JJCgc3jkwP4cN4Qyc6AWffKsDSvFGjWdza9cj4Ec4hlkEdg2n48fj2iYSNcM9ZFSuRyjWex/Yrhu2xysIcELcHOJedq/2OpPUGBrTss2ba29Am08Ko/Ismj1G6/0u7mjNmJXj2lNf/+00KY3q/CMiqgD+ChH9T/rev8PMf+H0+X8Rwsj1ewH8AQB/BsAfIKIvAPj3AfwgZFP8OSL6CjN/7VtdgKBHgYhGX4+LwHysiWDNZFwhE5L2UDHLPxKxWkigtYY2uvsH4tMVuEkLSA7nINFyyv0xekLXnD5KRs82sPdp+hgyCRcGU47sqV2dj7GuuGAO4AfUtzLecpsKMkE9NrC3Y0VEMJNsQE25nDyXk8iZ0Oa6FC0dg982P+dncs+kBUn+6/bqit7F5OvUkBVJpFLU55oL3jgoAYCaort99vaDllSNPrzq3SwRUtOeiA6xyGgRHawA9Q3N1bD7GJrjWfgWHf4k421IhBjAParzp8aPAPhv9Xv/GxF9joi+G8A/B+BnmPmrAEBEPwPpMfDfv+n8ZkqOMXe9DIF3W49c/drhhYczbRlcb/a973gMN/vswXaW9rxWpuMPI/ovesxKsykEDXH0URh7EpNBUsmS78C2IzusHzUEN4fIL7BWxxO4sHEP/ADUl5MX7ZN3vxNfs4aSzB0paUoaoCCB3atoFCKooMGRSYS5sYyRoya7PW/KEg+L9zB6B+VJd16JsLcd2yb9Ioo9bwWTwI+SvofumSDLsnh1xGgDO3bt9TAr+WPVgNOmmxbvYXPVtWXXFze7e37pJxmfiOqcmf+qvvXjajb+BBGt+tpTlOafiOocmDcbuU3aAAbPws11XbGUgnWt0m+gFGf09fJ/3bFZEGdf6N7OKJgyphljbzDp3REWEyuqqT7jw2XF5bLishbUIq+ty3LoG2ewf84ZxFZVnrDUC1IqAEmDwWi2xd02zNPh91h+o7eqAmKfY/VRSUCLPK+JiAQAIt2IeMYaWbWpMyrQKdCum9q964rPLiFpyEWYnUfXjWZM4RFfqepmqM01c0a9rMiloK6LP8eUtRHKumBZl4k+kqTk5ZJRrelmiPkdQgn27E9mZLw/m/NYpf8u460Ejpk7M/8AhC35nyGifwLAn4JQnv/TAL4A4N99HxfEzD/JzD/IzD/48PKl7FzGd2H+Uu9gFtOg1Iy6FNSl4OWLC16sq/byzph0dIJyDjVNfTc2whldsMbWK9weQkIkgWuJ1nn/NMA1paQsaWMKrZi2IKk1NrTGGHYe81V2bW/VWUzClNNBY5z9OLvuGMQ+gyP+OUfqWRsqyrVEjv6UyZstGvxvPtmbfBbzmfR5heucr83PihBjzJzVpRZv6QtI9ooVkFreZEoGOGmFhqZulZz1uVRhbyPpFedNQXSDLLViuawupLHg1sbgo69smy6FZyB994bzjr7r+KRU5z/MzH9aX74S0X8N4N/Wv5+iNP9ViFkZX/9f3+Kc2FuT7G4OJDaaJWHgQU4JNc+eYr137WAKJZEV/6QUSf2RQHZGSoTLskweRktKBrQ+Cmj9Nsgqu6kIW2wukoBpOhWhNliWigEcdtmh12fJu03NniQq5JDQa+NNZlsMW0SBlMDxEeCR74rZ6kIcIH8zGe+ZseeNIL5n4CXRRDlNMntv6Hr/5pdZCASQvuuSrSPNLtMiPKE8GPu2SewZJBaDuwaaoKzXUrUo2a7HSqocjc1JWMggfr5VxqewwZ7BMGYWWvpvtVDfcnxSqvP/U/0ykMz6HwXwC/qVrwD410nGHwTwDWb+fwH8zwD+MBF9nog+D+AP62tvGPIEG2vsq89mHuT2zUylilW+e5cm8k7JFrJNSi1YzARcV7xYV7y4rFhCnp4sqvmgCErP1juu2qOstYbBw1O1BHAZE0kFlB5CKg0i6jWYgSQxMOOwZ2Zs12sAVeaCtu/deT6HLAr5CUDE7GRm3vp3RxSPffG6wOjIUWAhm5Ehm0+ZkX4uQJDfx+sEjuwYpPwz16s0zuwz99TGfpWSGtOQ9ox66wKYtO4a7rwxMoDLuuLh4QGlVqSccHm4eI/3EcAZuw+jp3dSoVNO6buMd6E6/0tE9B06nz8P4N/Sz/80gD8C6f/2CsC/AQDM/FUi+o8B/HX93H9kAMpTgxmCGJo5YBkD3QLIZiLKRQyNaJ/NrLOZYGbdpVZUTXFKiXDdJWh6vV7VJ6pILOGCAS3P2SfrFg8OPk3IhsgZRg1u/sPWOzoDAE/ey7CwinbwfHV9xKuPfwsPDx84Mnick9sg9tmktOthklb3osxCSpN9FlOjCSgyBY3nwaePxgElZnYLA0TaLMNsUgvVzQX86vVrXK/XWRgbQR5tPGLHlmfcFfyCa9yYosesNIM5gfs4CJrPiW4OUfPZ90zArCtrrsI/Y8XAMRh+b1P5pONdqM5/6InPM4Afe+K9LwP48rdzgTZBU9Uf4W0iJadJwe4mAQcAUrPyHBdiJACLyAWG9h8AAII46kspKLm4iZQAcM5AH0jhePYgYzaKMP/K+13bGhvKKvc0r3MWVcp9ffTNV/jqb34V3/kdFamUQ8bFeVFZSUu8t7ng9Dv+/aMmY2bJSTUzVIUtJvDaQWgIH6R9huUizH7UzY6VTFY+ITIvPz/65kd4/eq1AzLXXWjs7Hi11KOghHvJlA/NOOatkPrPrPmc4lNT8FGtLMgSkNEGUIonv9dF/MKs7Z4JCY/X12gtcGaW/N6EDXjmmSYADqUXYk9PkEM0lWi4qs0RAePKMPMrSd0apqmUCahFYlASiO2SfKzC8rkPFrx8+YDWB169flTkKwE0sC4Vezv2Csi5YFeek1qlG6st6u7NPeAbxFDNaDuvCY35nl/72tcBAC9efICyrsFsnObtIe4lv/nrpsNqqWhs7MbD70MCxMdF3HuX64GmvWlWifCxyHE9YEEzHGBxOyJJx2LdsHgwriym5Ne++jV8/OoV9n2f0Dys7EquI5IvRf8J4T6ZZ+sv23yzPGL01pGS8NRYyMHMQMt0EQmdm2PKGaWygFVqVqYs53l89dotIa/Gfw/jWQsc0bFOy3IHU2IAkodYXRsdU3hsfkyzRLDA4BYxlTROpj7CZV3xcFmRCNi2aVrABJyBRuPA+1GrPChixrpULwmRcAN77JDU1DKwxTqrGpekZZy8fv0aj4+PylZ8QalFfaYsdV5jaI87TYSGLVwAYNeKOYv2TxY8Bnn1+eDmdOjG7EVqGoIJNReNw5H2yRvY264hjslbmbNsLqNLlo8FpAFg2zteP75Ga91970hvbmiu9XmLpuIBtLDNq3dPrQMBpSStxNdqBxIqe1Y3gHLodqQzYDFRe3ZZ82/3fReEVBmnc5GW0tN6eT9r+nkLHGa7KH8Q6ABrHZoSyEhNFmCIZevia7U+0b6kmd/SfH1mHshCS2gaF7LPt9GxK0//YadN6WZBWOcaI5P1rBFLmyKJ1UEXaGtNOv0o4vbRxx97YNYWvgue+j22OLe9YXM67gQhUppgidSTyXmu2ujQSHHM/Ot9+AYjJTqEXHTxa8rZZVmRE5Bywbpe5HoeX6H3jvVyEVjeABk20x+z8Jek1gyYVsoYAnCYxpY6RatlDPB8mO/eJjcoQ6F8NR9Z+4wnkiUhzz95+Ad9oCv5kF1jU8T7gLaqBty7vJeLkMCaaV1qQf800OSBJGNgpvvo7sZHf2YfAyNkFUg6jlIw+KHIewWMLjDvtu/Y2w6AppYZQq0GhYkHsyN0Y7DTlvchxa5iWnUQMTDIY4bLuqrZq36e7uAlZ1T1MUqSVlyjD98wogkdS4LM70g5I7sPRyAqGJtUTpRSBaxRVM00Fw8GqwZm3XQ4mOhI6i+lrCl0Qh+RUQBo6QtJjVlKWZKsmW+6ghKrEFnbZMi9r+sqCdpaAX9Tn5bglQIW9B/cQBiSR2umppzF0/ximCOirtG6sC9aZtHYJuNy/L6Y9gNEcq6shcMvX7xALgXf+K1vvJcl/bwFDqzCBs0MEdjdaM4vKBgYE8pWQbAKcOA2CCsZ8STthYkmoKI/GXe0mGqtPqSOzmgYjMYtEUBcMWj4wzOtmlXTGVdGH8Mz8y1WiERAE+3qPk4KvP5QMygl1JS9FXJcMAvgC9129JqLzwERualnPlM060w4MgDOUoSbtPyIspU8VY8VsJqRMFArW8fU7P6Qmaur+qGyEb7WZ0FukupegzEAJAW0yqImrPhYYwjYAtYAvsZhLbfTzEaNLnoLMxfG6fRO/06rPlprAl7paxNxkjkopWC9XN55NQPPXeDUDOCh2qXPYPDoHSMrD36aHFJEBAzAskxssTlXBQ9A8xxjMPMM/xKRJZJo3Ezo9zrfUpxbalIJTrq8113wDfJmBjqNQ+aMm7gseZlLrZ5Ea0zPNSV0TD9WxKYYzAAAFINJREFUcjW1ayrz0fTWOj5HLVm1mJlQ6suZid1a8+YVCSK4yYTdUNoUqcmls+x2vaJpKCX21DbYXcBE0coGisTkYY+16cYFENJIoCILPQ1pNVKK+sQ8vPfAZb1M1HTM+CHCz5iozSzhC9uUATgNBqupu2+7cKKopVMS4brtaOE47zqetcB5gvFg96dsYiWHsggltjnD7nB3qGXg9v9122SXrtUh+xiX8XPyzNiIGsYg7rMZEz9zBnn6mJqQIQuj94beZ0spK5i1RVo0nUz8yCHaBYLGLacwQEr5+Ddlba4oF+MUAbagx/ByF/ezhhAiJeuemgi1LFKsGbJdYsxqKRW71p2NIazElrcaGdEMqb1um6oveMDZhsX/+mjISTrQ5kX8w5oIvU9wB4nw+KiU96WBlPDVr69k5Cxz11vD43aVTYCUVVmfr9P10QyjdG648gDoEvx7xrbtSL3d1NV90vGsBc403AHZ0l22FiF2LQb5EkADkhc4yAGVnAXZQzeGJ+v1nWUR6uTGdCinvA6O9QDcl7TFZylawGypdRgWB6ZjhbTcD1DK6sHZZVkOWqL3jqJakEiybeJunSgj1aSmrqWVqRms92P93VprSOG6PfZnQIaicTEuuCyLlxRFzeGbA4lwlVo8yyOa0Q509I7emp8r+qdDN4BEpNpGgY29YblUMDFyySDLr9RnKOewFmH6/HKSTJJgzqfro8/Py8uD34PE6hSdVCY3W2/mK/sjpJlw8T7G8xY4HxIdshSckgnrUrHU4nE5QJDATXde0AQsTIvknLFWcYbF1Bro2mduWXLY+efkmhlovoINSpIixkl8AuJxMKnks6wOvvpXibDvCfvesNSKmqcWsXQmMyFtR7Wm8QtlpQ2YGSK2AXnh5mBkZDedbBM5F2MaIhrTvmJmhf0dY2E29ybEYwh/yIsXL5xfxBd6AET21tzPs+NEgYv+pLgCDOxASwTKNJFQAMwyRy8eHrAu68FCSeF+qvqMRISvfu2rkoZXl5nYnLMHvUGLcNaEfus+BzZ3+X5f8E8ynr/AjWnmycTOejMCKUmMflTNz3Pir5XIZ5JEYgBCk74bB2XHZa1Y8jR3rI+AZIBgEtbYom0NCbFYMzlI4LmCIG2RNctyjDpvsfiP7v7eBITYOTPjEA2akQPwYdqHiJCJ0BOUKPVYfR0FwBa5CdhTca+o2VyLUMJSZusr04SAcPhvFjwv0lxlAGBtQJl1c6jaD8A07+E8SbJVGjdQJmQY4zLQ9g4ejBcPH2gvOe37rb62FyDnGSOs64JlEYEyCyTX4j6zb6x0DKzHOd23DZSWT4lJSVq/xYDlWffekRjYNkLPXTSPYlOjS9b9Prqy8op26V2YuGrJyCWh7bPlr8RtVaBpAguGnEm3TaBppoJlOuQkMHtKwluZczAzo8CT4mb6EBeNvVk8zmJqpWRQggaQSbNgZjC4jck+ZcO0icWo7DU7l5vJ+j1WiP9c9W7X8/j4ePBP7XcHh1JCasn5VgZUA0OE73K5oFRhWMulYFVA4jqkmUatVZobOpvzbNYxM19kYzIBk/4HEtS3EIMAILqRudmewEOzRB6vyEVcBinZKa6pLAlA2NZmABxQq2eIPxt9vL53p2J51/GsBY4AR98ENh6SuJytWFCyTQwPZobmTpqvBbCGXxMlLEtxH0dSqwqWKr3kLuuCnBOG9t9mZgwmqThgiCYd06dsjcEDuKxCax41ig1WezRpVr0Q4DAACW3sbQ/fA8ASAzN01I4l95LdN4vAjcfamLEoQHCm2bO+48Yk5mijCqSbYtpr76wZo9AREdiaWxgYpICPdaihMVtPres66wNDrVouRaoI1IeaG8m8vzE6KBekVanvwjVZ+phmoPm9ioDsGE36CJScwZYcTccqBLaMnTGwPV79eRTwoYmlVIkEMOgdxrMWOED550cwdcYAc8Jq6Uq6c4vZNw7xF1s0pWRJv0rJ6+JMA6y5ICex7xlAgwbPGdg0NsY08wrNP7JxDriKllD0MmdP7SpZ4HQVJVyv+6w60O8ZogY6mnIAPJ3Lzbugyc7+WFIzS6j3cCNcMWkgB/PK+Bi3bbuL3NqCPicaA9DQggAyKWfP9k9KhRczaEbrbsZFE9ziZebjMSQn8nq9+vOyec6aV3o2hZkZ7dpApMIYXS/9/XK5eKJD1mu9loyxb9LIU4Vybj4SNngf41kLHLNVdzOYp5kkmAhpYFbLRBKQxkydOmoa+X1X7WXJu9noBihAxInAIE/zApQNSmn0cipuUpWU/XoAuLk5rLrZoP+sapik/F8SiSetgy1iSyBGOjrpshCFfwNB0M6CH4XRFtO+t6NfEtBJq00zs9R+Wm8DO24U5jjsOLvSyoND0BhzEzqQ+PDA0GoBi6v6cfYGgiYdQNKz9v0K6gk5H6nqKGvvNnu+BL/fx9ev0NqO9XI50FvwGF6db3OxLBU5EfroaG3Hpulw0VqJ2v5dx/MWOI3DwYWNpANNTg5+OMqXSOq/1Ac6+x+PBO8dYN01QRrXynAmY0kbE3SsAJ7vB9KKYUpAZrRtRy7JzcW4w8oCakAn5Zpk6XQ6hgZUZ2DYfCpbmKBZtU00CX+IxJ88AyLxu6a9zqaifdZrwJT/MhbEppOQewV7WGhusvU+6QkhLXptYQsIlIVSISmKGkzQvXds+46SRONb/PF6vWJ0BqDhgmpAz9BjaWyyd+nfRqTgiDTaRCI8PFykcHS74tWrjwXd1R5wMFdg27CVGVIQUz9hWResq9Dy5DxRZgO8iD4FAudgg9f/KxwfcuGA6RuIf3VEKG1Sxa+xUpmhDTPgO/JIyYPBKTPIaBWCEGXMAGq5JGRNMQJw0BJyfvIHLfVmkyqvG9diuAcDFCxQ7eZhzgAYbZc0sMwslOGYu7Qs1o6qgIyZalnT4EzjGwRvgeuk/dLsfAIczTbKB+AlaFD7aVocKshSE8dIzGo+zzhiKcWRxMu6il9JhLJUPDw8IH/8Mcb4WNKsyoTip8ne0Xe1QMoMmeRF8j+TZugQES4PD3j96mOptriE8IFctG9UouHFquhtiG+ZtQstSSaKhWk+NXG46OTaTVtRqREC2evCAnVkXXJTUV4BQOpsk79ii9wYNnqfFciuBXi432ELUVMNvQAzXqP8PtAHAX3o7o3ZgDElNymzZpiQVZxZJxq9LkoJg2cJT1ENkzWWl0C4btfgy8nOnnIGKOGVsp+JNgwdflLSinJCbx0Pl4t2gs0eJ4v+ns1l3CgIwFKqdxsiZgwMLehlEHU1+2Rullpnt9GUsL54wPpw0WyTgbbvDrL0YU1GtE0xETIVMHfsfWCgoCixHbFklxAzciZ8+NnP+ObjJrWW3vhaaU1idHtzHz2HKoGlila9Xrewht5tPGuBI02VirEag8BrLQCyL0KjJrck4RjAtL+TazEcBLUp34jE9Ri9h+wWTF8EGgBP6uNk1QamTy1NaWoDhfe1CHWMIc3j7VpyQs3F6+dMqAczkkLpVAzZKxhsGf1qMkJZi2vFw3YRGgjthNN7x7Is0l9htzbJsqB7rm5uSSsoSB5nng3ma60YfRxSvKyMxTaMkrPWFU5qdnk+w3NQCQWkG0R8hu5PBgTz4eEBu/qCvQnt+s67zw+zmuVKdZ8gexOS5NbuWkKTSsFDlrBA17CLh36ima3z0bZNrAuyJPDp65YspFPttxs0IeE0+VkAv8rM/xIRfQnATwH4HQB+DsC/9v+1d7Yhcl1lHP89921mx21M4ksJKjbBgkQQjaW2KCIKVaPol4LpF4MWBFuwftKEgiCIED9YFZQ0GKGK2NYXMAhSqvaraau2MW1Ys1E/CI3xpa2myc7Mvffxw/OcO2fWbnbXbHZmlvuHYc+cuTv3PHfOc87zflR1IFaf8nvAO4B/Ah9X1b/4dxwG7gQq4LOqukoRoZH4EnQHi/ezH1vNQdcU81FlbPUNEyDWbfIkbXQeEcu3slwodwhXI5+WiBf5CVYz3/VUlVRGO2isu405mQm1QhibMIo7zH2ij0ScfmOsKIrCHMhZTppko4NKJKFIM4ZVSemVnrM8R1RZGgyo6oo0MdoHzmhZllKW5uNa7rczEatudJzUxb6iKNCqbg41nOv1bJVfWmpEWMuSHy06zQInAnXtuXdVszMGvS+ItuIMOxwMbFHCmAU1Q1jlpemGZTnm4wzpNkmSMBj0SeuUlASt7KAQAYv0V7WyGLHEk5jxK1hHh0tLVGVJrUoRBVgXeWExuon/gBuE9WiC9wBnovdHgPtU9U3A8xgj4X+f9/77/DpEZC9wAHgLVnH5287EKyLsCqP3USDw8vQLaAr3xCJl2AFNj7LKxnnIoA6+PfXQK2SM2ULYVAgqthSV1Ez8yXj6TLx6mvtBGwthbEoPOoIdm/y/Ju0YIzO8HfQx1+nSLYqGdlXlpcuX+c/Fi1zq9xl45EYdGMitkGExyLOMPPcyc7Fp38W7sQwGSZo4x5CeEuo+5l7wKEkTarG8wouXLvHS5cvNWXuj+pLjhpzwLMICqrWilY6Z6fM8J+8UhNqUzWRQtZQaPw21qiqGgz6Dfp+qNn0snEIUQvmSKPQtPNMsS+nOdej1unS7Xc8aL5vnYc+9YugLGthZcRuBNTGciLwe+DDwHX8vwPuAcK7AA1ipPLBS5w94+8fA+/36jwEPqmpfVf+MVfW6ebV7x6buzBXjblFYteUib87utjjEkU4nIvR6Pbrdrpl/s5y5PCfLrCiNiatWiEyJ9DL/gZog6TyniCZakdl90nT06NLEXMs2iWrPyi6tVF9dN39rsCOJ/ccflqMKYGmajh1q0RiCyoqyKm1Xqa1ScaU1FbXnclVNyb4wacIEb6yYdaxzjZ7PyCWhDa1J0HE1pB4l0c5iDFHkeXNmWuZiYVmWTflAnyPNzhL70EL9zf5gwLDfZ7C0xND12oAsz+nNz9Ob79lOGIw57uVuFtiggLsIPSyHTaJrXCs0NpQQMXWvN0enY1W6m9jVILWM5j5zvS7z869YbaquCWsVKb8OfB64zt+/CnhBVcNTisuWNyXNVbUUkRf9+tcBv4m+c8VS58CnAea3bRsT1cKEyDNzC+RpYtWLqaP/t0zqGhoxLE8tdrHwINlhFBGeJMmYnypm8DRNyYIPKrd2GsQnaJR/EU9qxM3m6v5DsIpSvuIm7mOqnAlF/IRVR+wji8citVLVpmsYM2aIBIf1uAURTMTOksQLq1amR+KTvVaKvEC1HivxnYwx4OgV/Fch7SccvBGMHrkbQUItx/A8lrsUml3Nv1PUi7hGCyTB6OO+y6LbYa4sjbmDuuDGJxPTa9+BFC1HzN24SjxyJoi3kiRNnRKtvZK2z4MwV0K0TOJGLUlNZ+90O2wEVmU4EfkIcEFVfysi792Qu14BqnoMOAbw2l27NP5BbJfx00lRBqVNJrD0DAS0tIiMTJJGtEnyjExCUmjpIUmWrxbXjB+Jg7ar2GGKdVOyHGiso1nIpvb/TbOMcmhxnCE+sPZIrhB6FMoP2Kpt1X8blwfBIFQ3k0mjPD8NBgn1Mglpagr9MqtsSLy1XUsaH2blJ4KaBVDNAU8o/6AsDQd0EiGto4wBr/OpAonrwvGikKYp3bk5inpU1aoIFtckgapkWJaojoIC8iwfOcnxk44Ut9K6ewVY6g/RuiLNcwoRai+2K/77WGmIxAMJgPAc4sVHxHZ978+KUVpSpXZAZFmWlhpVFG40spAxC/uysSRZwga54da0w70L+KiI7Ae6wDbgG8B2Ecl8lwvlzGFU6vyvIpIBr8SMJyuVQF8Rfz9//uLRI19ZWAc9s4JXA/+Y9CCuAbY6XW+86m9arrRf6YWdDfBzb/8IOODto8Bd3r4bOOrtA1ilZjBjydNAB9gN/AlIV7nfk+sZ36y8Wrpm67WRdF2NH+4LwIMi8mXg98Bx7z8OfF9EFoF/OdOhqs+IyMPAs1iM8N2qujHOjRYtZgTiHDyVEJEnVfWmSY9jo9HSNVvYSLo2SBW8Zjg26QFcI7R0zRY2jK6p3uFatNhqmPYdrkWLLYWW4Vq02ERMLcOJyAdFZEFEFkXk0KTHsxpE5LsickFETkd9O0XkURE56393eL+IyDedtlMisi/6n4N+/VkROTgJWmKIyBtE5DEReVZEnhGRe7x/ZmkTka6IPC4iTztNX/L+3SJy0sf+kIgU3t/x94v++Q3Rdx32/gUR+cCqN5+0j2MFv0cKnAP2AAXmv9s76XGtMub3APuA01HfV4FD3j4EHPH2fuAXWBDKLcBJ79+J+Sd3Aju8vWPCdO0C9nn7OuCPwN5Zps3HNu/tHDjpY32Ycd/yZ7x9F+O+5Ye8vZdx3/I5VvMtT3qirvBAbgUeid4fBg5PelxrGPcNyxhuAdgVTdwFb98P3LH8OuAO4P6of+y6aXgBP8POed8StAE94HfAO7Fokmz5HMTOor/V25lfJ8vnZXzdSq9pFSmbAGjHywY6zwCuV9XnvH0euN7bK9E31XS7KPV2bEeYadpEJBWRp4ALwKPY7rSmgHwgDshfF03TynBbDmpL4Mz6YERkHvgJ8DlV/Xf82SzSpqqVqr4Ni+m9GXjzZtx3Whlu3YHOU4q/icguAP97wftXom8q6RaRHGO2H6jqT717S9Cmqi8Aj2Ei5HYPuIeXD8jnagPyp5XhngBudKtRgSmqJyY8pv8HJ4BgjTuI6T+h/xNu0bsFeNHFs0eA20Rkh1v9bvO+iUEsV+c4cEZVvxZ9NLO0ichrRGS7t+cwnfQMxni3+2XLaQq03g782nf1E8ABt2LuBm4EHr/izSettF5Bmd2PWcTOAfdOejxrGO8PgeeAISbL34nJ+b8CzgK/BHb6tQJ8y2n7A3BT9D2fwrLhF4FPTgFd78bExVPAU/7aP8u0AW/FAu5PAaeBL3r/HmeYRSwbpuP9XX+/6J/vib7rXqd1AfjQavduQ7tatNhETKtI2aLFlkTLcC1abCJahmvRYhPRMlyLFpuIluFatNhEtAzXosUmomW4Fi02Ef8FBjx61vUOHjIAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fqw-N5SPKdIu",
        "colab_type": "code",
        "outputId": "32a8b6d0-dc95-4581-8353-5d13526dafae",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "im.shape"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(4032, 3024, 3)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 62
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AgRND7YPcWZX",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "data_dir = pathlib.Path(\".\")\n",
        "images = list(data_dir.glob('drive/My Drive/data/LFA_Iamges/Train/*.jpg'))\n",
        "\n",
        "for image_path in images:\n",
        "  image = load_img(image_path)\n",
        "  image = img_to_array(image)\n",
        "  image = np.expand_dims(image, axis=0)\n",
        "  \n",
        "\n",
        "  aug = ImageDataGenerator(\n",
        "    rotation_range=30,\n",
        "    zoom_range=0.15,\n",
        "    width_shift_range=0.2,\n",
        "    height_shift_range=0.2,\n",
        "    shear_range=0.15,\n",
        "    horizontal_flip=True,\n",
        "    fill_mode=\"nearest\")\n",
        "  total = 0\n",
        "  sum_total=2\n",
        "\n",
        "  imageGen = aug.flow(image, batch_size=1, save_to_dir=GENERATE_IMAGES_DIR,save_prefix=\"image_gen&\"+image_name, save_format=\".jpg\")\n",
        "\n",
        "  for image in imageGen:\n",
        "      # increment our counter\n",
        "      total += 1\n",
        "      # if we have reached the specified number of examples, break\n",
        "      # from the loop\n",
        "        \n",
        "      if total == sum_total:\n",
        "        break"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t1wKWxRUEKEG",
        "colab_type": "code",
        "outputId": "e5dc13fd-808b-4b6b-e30e-04e805c03eb1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 550
        }
      },
      "source": [
        "data_dir = pathlib.Path(\".\")\n",
        "\n",
        "images = list(data_dir.glob('drive/My Drive/data/LFA_Iamges/General/*.jpg'))\n",
        "test=[]\n",
        "i=0 \n",
        "for image_path in images:\n",
        "    image_name= path.basename(image_path).split('&')[1].split('_')[0]\n",
        "\n",
        "    image_name = str(image_name)\n",
        "    i+=1\n",
        "    for  k in range(len(df_labels)) : \n",
        "        if (df_labels.Result[k] == image_name):\n",
        "          test.append(\n",
        "              {\n",
        "                  'Labels': image_name,\n",
        "                  'Result': df_labels.Result[k]\n",
        "                  \n",
        "              }\n",
        "            )\n",
        "   #print(df_labels.Result[i])\n",
        "\n",
        "test=pd.DataFrame(test)\n",
        "test.head(39)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "error",
          "ename": "KeyError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-148-8276ab82e527>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     10\u001b[0m     \u001b[0mi\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     11\u001b[0m     \u001b[0;32mfor\u001b[0m  \u001b[0mk\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdf_labels\u001b[0m\u001b[0;34m)\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---> 12\u001b[0;31m         \u001b[0;32mif\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mdf_labels\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mResult\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mk\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mimage_name\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     13\u001b[0m           test.append(\n\u001b[1;32m     14\u001b[0m               {\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/pandas/core/series.py\u001b[0m in \u001b[0;36m__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m    869\u001b[0m         \u001b[0mkey\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcom\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply_if_callable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\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[0m\n\u001b[1;32m    870\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 871\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mindex\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_value\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkey\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    872\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    873\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mis_scalar\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresult\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/pandas/core/indexes/base.py\u001b[0m in \u001b[0;36mget_value\u001b[0;34m(self, series, key)\u001b[0m\n\u001b[1;32m   4402\u001b[0m         \u001b[0mk\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_convert_scalar_indexer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mk\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkind\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"getitem\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   4403\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 4404\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_value\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mk\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtz\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mseries\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"tz\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mNone\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   4405\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mKeyError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   4406\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m0\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mholds_integer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mis_boolean\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\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;32mpandas/_libs/index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_value\u001b[0;34m()\u001b[0m\n",
            "\u001b[0;32mpandas/_libs/index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_value\u001b[0;34m()\u001b[0m\n",
            "\u001b[0;32mpandas/_libs/index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
            "\u001b[0;32mpandas/_libs/hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.Int64HashTable.get_item\u001b[0;34m()\u001b[0m\n",
            "\u001b[0;32mpandas/_libs/hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.Int64HashTable.get_item\u001b[0;34m()\u001b[0m\n",
            "\u001b[0;31mKeyError\u001b[0m: 0"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bbPk1vrKEZwb",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        " "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GmgwadgA-TqQ",
        "colab_type": "text"
      },
      "source": [
        "### 1. Helper functions\n",
        "\n",
        "* normalizing the pixel values between -1 and 1 \n",
        "* read and transform dcms to 3-channel inputs for e.g. ResNet50.\n",
        "\n",
        "\\* Source for windowing (although now partly removed from this kernel): https://www.kaggle.com/omission/eda-view-dicom-images-with-correct-windowing"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
        "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a",
        "id": "f0URQlqi-TqR",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def _normalize(img):\n",
        "    if img.max() == img.min():\n",
        "        return np.zeros(img.shape)-1\n",
        "    return 2 * (img - img.min())/(img.max() - img.min()) - 1\n",
        "\n",
        "\n",
        "  \n",
        "def _read(path, desired_size, augment_data=0, plot=False):\n",
        "    \"\"\"Will be used in DataGenerator\n",
        "    Loads image, crops and resizes. With optional image data augmentation.\n",
        "    We assume that the image has been centered. \n",
        "    Input:\n",
        "    ----------------------------------\n",
        "    desired_size     :  desired size for the image (tuple)\n",
        "    augment_data     :  nb of data augmented samples (int)\n",
        "    \"\"\"\n",
        "    new_width, new_height,_ = desired_size\n",
        "    print (path)\n",
        "    img = cv2.imread(path)\n",
        "    rows, cols,_ = img.shape\n",
        "   \n",
        "    if rows < cols:  \n",
        "      M = cv2.getRotationMatrix2D((cols/2,rows/2),90,1)\n",
        "      img = cv2.warpAffine(img,M,(cols,rows))\n",
        "    res = cv2.resize(img, dsize=desired_size[:2], interpolation=cv2.INTER_CUBIC)\n",
        "    samples = np.expand_dims(res, 0)\n",
        "    if augment_data>0:\n",
        "        # create image data augmentation generator\n",
        "        datagen = ImageDataGenerator(rotation_range=90)\n",
        "        # prepare iterator\n",
        "        it = datagen.flow(samples, batch_size=1)\n",
        "        for i in range(augment_data):\n",
        "          batch = it.next()    # generate batch of images\n",
        "          image = batch[0]\n",
        "          samples = np.vstack((samples, np.expand_dims(image, 0)))\n",
        "          if plot:\n",
        "            plt.subplot(330 + 1 + i)\n",
        "            plt.imshow(batch[0].astype('uint8'))\n",
        "          #img = np.stack((res,)*3, axis=-1)\n",
        "    if plot: plt.show()\n",
        "    print('samples size in read:', samples.shape)\n",
        "    return samples\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VO4vaCw7IfL3",
        "colab_type": "code",
        "outputId": "a2c8cf01-0279-4345-ace8-0bfa5988569e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 320
        }
      },
      "source": [
        "test = _read('drive/My Drive/data/LFA_Iamges/Train/0042.jpg',(512,512,3),9, plot=True)\n",
        "print(test.shape)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "drive/My Drive/data/LFA_Iamges/Train/0042.jpg\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 9 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "samples size in read: (10, 512, 512, 3)\n",
            "(10, 512, 512, 3)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tGSfWNhna91-",
        "colab_type": "code",
        "outputId": "17f180f0-81c2-4e40-f453-11532e2739db",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        }
      },
      "source": [
        "from numpy import expand_dims\n",
        "from keras.preprocessing.image import load_img\n",
        "from keras.preprocessing.image import img_to_array\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "import matplotlib.pyplot as plt\n",
        "import scipy as sc\n",
        "\n",
        "img = cv2.imread('drive/My Drive/data/LFA_Iamges/Train/0021.jpg')\n",
        "# convert to numpy array\n",
        "res = cv2.resize(img, dsize=(512, 512), interpolation=cv2.INTER_CUBIC)\n",
        "data = res\n",
        "# expand dimension to one sample\n",
        "\n",
        "samples = expand_dims(data, 0)\n",
        "# create image data augmentation generator\n",
        "datagen = ImageDataGenerator(rotation_range = 90, width_shift_range=[-100,100])\n",
        "# prepare iterator\n",
        "it = datagen.flow(samples, batch_size=1)\n",
        "for i in range(9):\n",
        "\t# define subplot\n",
        "\tplt.subplot(330 + 1 + i)\n",
        "\t# generate batch of images\n",
        "\tbatch = it.next()\n",
        "\t# convert to unsigned integers for viewing\n",
        "\timage = batch[0].astype('uint8')\n",
        "\t# plot raw pixel data\n",
        "\tplt.imshow(image)\n",
        "# show the figure\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 9 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RVD3MWoh-TqW",
        "colab_type": "text"
      },
      "source": [
        "### 2. Data generators\n",
        "\n",
        "Inherits from keras.utils.Sequence object and thus should be safe for multiprocessing.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GVupHBDB-TqX",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "\n",
        "\n",
        "class DataGenerator(keras.utils.Sequence):\n",
        "    def __init__(self, list_IDs, img_labels, batch_size=1, img_size=(512, 512,3), \n",
        "                 img_dir=TRAIN_IMAGES_DIR, n_classes = 6, train =True,\n",
        "                 n_augment = 9, shuffle=True,\n",
        "                 *args, **kwargs):\n",
        "\n",
        "        self.list_IDs = list_IDs\n",
        "        self.indices = np.arange(len(self.list_IDs))\n",
        "        self.img_labels = img_labels ### contains col1: names of images for loading + col2(!exits fr test) for labels\n",
        "        self.n_classes = n_classes   ### nb of classes\n",
        "        self.n_augment = n_augment   ### nb of additional data samples\n",
        "        self.batch_size = batch_size\n",
        "        self.img_size = img_size    ###  desired image size: (width, height, n_channels)\n",
        "        self.img_dir = img_dir\n",
        "        self.shuffle = shuffle\n",
        "        self.train = train\n",
        "        self.on_epoch_end()\n",
        "\n",
        "    def __len__(self):\n",
        "        return int(ceil(len(self.indices) / self.batch_size))\n",
        "\n",
        "    def __getitem__(self, index):\n",
        "        indices = self.indices[index*self.batch_size: (index+1)*self.batch_size]\n",
        "        list_IDs_temp = [self.list_IDs[k] for k in indices]\n",
        "        if self.train:\n",
        "            X, Y = self.__data_generation(list_IDs_temp)\n",
        "            return X, Y\n",
        "        else:\n",
        "            X = self.__data_generation(list_IDs_temp)\n",
        "            return X\n",
        "        \n",
        "    def on_epoch_end(self):\n",
        "        self.indices = np.arange(len(self.list_IDs))\n",
        "        if self.shuffle == True:\n",
        "            np.random.shuffle(self.indices)\n",
        "\n",
        "    def __data_generation(self, list_IDs_temp):\n",
        "        print(\"Self image size\",self.img_size )\n",
        "        X = np.empty((self.batch_size * (self.n_augment + 1), *self.img_size))\n",
        "        \n",
        "        if self.train: # training phase\n",
        "            Y = np.zeros((self.batch_size * (self.n_augment + 1), self.n_classes),\n",
        "                          dtype=np.float32)\n",
        "        \n",
        "            for i, ID in enumerate(list_IDs_temp):\n",
        "              test = _read(self.img_dir+self.img_labels['ID'].loc[ID] +\".jpg\",\n",
        "                              self.img_size,  augment_data= self.n_augment, plot=False)\n",
        "              print(\"Dim data gen\",  test.shape)\n",
        "              X[i:(i +self.n_augment + 1),] = test\n",
        "              ### Convert label  into one hot vector\n",
        "              Y[i:(i +self.n_augment + 1), int(self.img_labels['label'].loc[ID])] = 1\n",
        "\n",
        "            return X, Y\n",
        "        \n",
        "        else: # test phase\n",
        "            for i, ID in enumerate(list_IDs_temp):\n",
        "                X[i,] = _read(self.img_dir+self.img_labels['ID'].loc[ID] +\".jpg\",\n",
        "                              self.img_size, augment_data=0, plot=False)      \n",
        "            return X"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3dk9Plt--Tqe",
        "colab_type": "text"
      },
      "source": [
        "### 3a. loss function and metric"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Aj4BKXWH-Tqf",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from keras import backend as K\n",
        "\n",
        "def weighted_log_loss(y_true, y_pred):\n",
        "    \"\"\"\n",
        "    Can be used as the loss function in model.compile()\n",
        "    ---------------------------------------------------\n",
        "    \"\"\"\n",
        "    \n",
        "    class_weights = np.array([1., 2., 2., 2.])\n",
        "    \n",
        "    eps = K.epsilon()\n",
        "    \n",
        "    y_pred = K.clip(y_pred, eps, 1.0-eps)\n",
        "\n",
        "    out = -(         y_true  * K.log(      y_pred) * class_weights\n",
        "            + (1.0 - y_true) * K.log(1.0 - y_pred) * class_weights)\n",
        "    \n",
        "    return K.mean(out, axis=-1)\n",
        "\n",
        "\n",
        "def _normalized_weighted_average(arr, weights=None):\n",
        "    \"\"\"\n",
        "    A simple Keras implementation that mimics that of \n",
        "    numpy.average(), specifically for the this competition\n",
        "    \"\"\"\n",
        "    \n",
        "    if weights is not None:\n",
        "        scl = K.sum(weights)\n",
        "        weights = K.expand_dims(weights, axis=1)\n",
        "        return K.sum(K.dot(arr, weights), axis=1) / scl\n",
        "    return K.mean(arr, axis=1)\n",
        "\n",
        "\n",
        "def weighted_loss(y_true, y_pred):\n",
        "    \"\"\"\n",
        "    Will be used as the metric in model.compile()\n",
        "    ---------------------------------------------\n",
        "    \n",
        "    Similar to the custom loss function 'weighted_log_loss()' above\n",
        "    but with normalized weights, which should be very similar \n",
        "    to the official competition metric:\n",
        "        https://www.kaggle.com/kambarakun/lb-probe-weights-n-of-positives-scoring\n",
        "    and hence:\n",
        "        sklearn.metrics.log_loss with sample weights\n",
        "    \"\"\"\n",
        "    \n",
        "    class_weights = K.variable([1., 2., 2., 2.])\n",
        "    \n",
        "    eps = K.epsilon()\n",
        "    \n",
        "    y_pred = K.clip(y_pred, eps, 1.0-eps)\n",
        "\n",
        "    loss = -(        y_true  * K.log(      y_pred)\n",
        "            + (1.0 - y_true) * K.log(1.0 - y_pred))\n",
        "    \n",
        "    loss_samples = _normalized_weighted_average(loss, class_weights)\n",
        "    \n",
        "    return K.mean(loss_samples)\n",
        "\n",
        "\n",
        "def weighted_log_loss_metric(trues, preds):\n",
        "    \"\"\"\n",
        "    Will be used to calculate the log loss \n",
        "    of the validation set in PredictionCheckpoint()\n",
        "    ------------------------------------------\n",
        "    \"\"\"\n",
        "    class_weights = [1., 2., 2., 2.]\n",
        "    \n",
        "    epsilon = 1e-7\n",
        "    \n",
        "    preds = np.clip(preds, epsilon, 1-epsilon)\n",
        "    loss = trues * np.log(preds) + (1 - trues) * np.log(1 - preds)\n",
        "    loss_samples = np.average(loss, axis=1, weights=class_weights)\n",
        "\n",
        "    return - loss_samples.mean()\n",
        "\n",
        "def sparse_categorical_crossentropy(y_true, y_pred, from_logits=False, axis=-1):\n",
        "    return K.sparse_categorical_crossentropy(\n",
        "        y_true, y_pred, from_logits=from_logits, axis=axis)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6Iyq1X8R-Tqm",
        "colab_type": "text"
      },
      "source": [
        "### 3b. Model\n",
        "\n",
        "Model is divided into three parts: <br> \n",
        "\n",
        "* (REMOVED) The initial layer, which will transform/map input image of shape (\\_, \\_, 1) to another \"image\" of shape (\\_, \\_, 3).\n",
        "\n",
        "* The new input image is then passed through ResNet50 (which I named \"engine\"). ResNet50 could be replaced by any of the available architectures in keras_application.\n",
        "\n",
        "* Finally, the output from ResNet50 goes through average pooling followed by a dense output layer."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-KCkHGPm-Tqn",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "class PredictionCheckpoint(keras.callbacks.Callback):\n",
        "    \n",
        "    def __init__(self, test_df, valid_df, n_classes =5,\n",
        "                 test_images_dir=TEST_IMAGES_DIR, \n",
        "                 valid_images_dir=TRAIN_IMAGES_DIR, \n",
        "                 batch_size=32, input_size=(224, 224, 3)):\n",
        "        \n",
        "        self.test_df = test_df\n",
        "        self.valid_df = valid_df\n",
        "        self.test_images_dir = test_images_dir\n",
        "        self.valid_images_dir = valid_images_dir\n",
        "        self.batch_size = batch_size\n",
        "        self.input_size = input_size\n",
        "        self.n_classes = n_classes\n",
        "        \n",
        "        \n",
        "    def on_train_begin(self, logs={}):\n",
        "        self.test_predictions = []\n",
        "        self.valid_predictions = []\n",
        "        \n",
        "    def on_epoch_end(self,batch, logs={}):\n",
        "        self.test_predictions.append(\n",
        "            self.model.predict_generator(\n",
        "                DataGenerator(self.test_df.index, test_df,\n",
        "                              batch_size=self.batch_size, img_size=self.input_size, \n",
        "                              img_dir=self.test_images_dir, n_classes = self.n_classes,\n",
        "                              train =False, n_augment = 0, shuffle=True),\n",
        "                verbose=2)[:len(self.test_df)])\n",
        "        \n",
        "        self.valid_predictions.append(\n",
        "            self.model.predict_generator(\n",
        "                DataGenerator(self.valid_df.index, valid_df,\n",
        "                              batch_size=self.batch_size, img_size=self.input_size, \n",
        "                              img_dir=self.valid_images_dir, n_classes = self.n_classes,\n",
        "                              train =False, n_augment = 0, shuffle=True),\n",
        "                verbose=2)[:len(self.valid_df)])\n",
        "        valid_labels = np.zeros((self.valid_df.shape[0], self.n_classes))\n",
        "        valid_labels[np.arange(self.valid_df.shape[0]), self.valid_df['label']] = 1\n",
        "        print('valid_labels', valid_labels )\n",
        "        print('pred_labels', self.valid_predictions)\n",
        "        print(\"validation loss: %.4f\" %\n",
        "              weighted_log_loss_metric(valid_labels, \n",
        "                                   np.average(self.valid_predictions, axis=0)))\n",
        "        \n",
        "        # here you could save the predictions with np.save()\n",
        "\n",
        "\n",
        "class MyDeepModel:\n",
        "    \n",
        "    def __init__(self, engine, input_dims, batch_size=5, num_epochs=4,\n",
        "                 n_classes=4, learning_rate=1e-3, n_augment = 9,\n",
        "                 decay_rate=1.0, decay_steps=1, weights=WEIGHTS_PATH_NO_TOP, verbose=1):\n",
        "        \n",
        "        self.engine = engine\n",
        "        self.input_dims = input_dims\n",
        "        self.batch_size = batch_size\n",
        "        self.num_epochs = num_epochs\n",
        "        self.n_classes = n_classes\n",
        "        self.learning_rate = learning_rate\n",
        "        self.decay_rate = decay_rate\n",
        "        self.decay_steps = decay_steps\n",
        "        self.n_augment = n_augment\n",
        "        self.weights = weights\n",
        "        self.verbose = verbose\n",
        "        self._build()\n",
        "\n",
        "    def _build(self):\n",
        "        self.engine.trainable = True\n",
        "        \n",
        "      \n",
        "    \n",
        "        engine = self.engine(include_top=False,\n",
        "                             weights=self.weights, input_shape=(*self.input_dims[:2], 3),\n",
        "                             backend = keras.backend, layers = keras.layers, \n",
        "                             models = keras.models, utils = keras.utils,)\n",
        "        set_trainable = False\n",
        "        for layer in engine.layers:\n",
        "            if layer.name in ['res5c_branch2b', 'res5c_branch2c', 'activation_97']:\n",
        "              set_trainable = True\n",
        "            if set_trainable:\n",
        "              layer.trainable = True\n",
        "            else:\n",
        "              layer.trainable = False\n",
        "        x = keras.layers.GlobalAveragePooling2D(name='avg_pool')(engine.output)\n",
        "        out = keras.layers.Dense(self.n_classes, activation=\"sigmoid\", name='dense_output')(x)\n",
        "\n",
        "        self.model = keras.models.Model(inputs=engine.input, outputs=out)\n",
        "        # loss function has been changed needs to be investigated.\n",
        "        self.model.compile(loss='categorical_crossentropy',\n",
        "                           optimizer=keras.optimizers.Adam(0.0),\n",
        "                           metrics=[keras.metrics.Precision(), keras.metrics.Recall()])\n",
        "    #https://stackoverflow.com/questions/43076609/how-to-calculate-precision-and-recall-in-keras\n",
        "\n",
        "    def fit_and_predict(self, train_df, valid_df, test_df):\n",
        "        \n",
        "        # callbacks\n",
        "        pred_history = PredictionCheckpoint(test_df, valid_df,\n",
        "                                            n_classes=self.n_classes,\n",
        "                                            batch_size=self.batch_size,\n",
        "                                            input_size=self.input_dims)\n",
        "        checkpointer = keras.callbacks.ModelCheckpoint(filepath='%s-{epoch:02d}.hdf5' % self.engine.__name__, verbose=1, save_weights_only=True, save_best_only=False)\n",
        "        scheduler = keras.callbacks.LearningRateScheduler(lambda epoch: self.learning_rate * pow(self.decay_rate, floor(epoch / self.decay_steps)))\n",
        "        self.model.fit_generator(\n",
        "            DataGenerator(\n",
        "                list_IDs = train_df.index, \n",
        "                img_labels = train_df,\n",
        "                batch_size=self.batch_size,\n",
        "                img_size=self.input_dims,\n",
        "                img_dir=TRAIN_IMAGES_DIR,\n",
        "                n_classes = self.n_classes,\n",
        "                train=True,\n",
        "                n_augment = self.n_augment,\n",
        "                shuffle = True\n",
        "            ),\n",
        "            epochs=self.num_epochs,\n",
        "            verbose=self.verbose,\n",
        "            use_multiprocessing=True,\n",
        "            workers=4#,\n",
        "            #callbacks=[history]\n",
        "            #callbacks=[tensorboard_callback]\n",
        "        )\n",
        "        \n",
        "        return pred_history\n",
        "\n",
        "     \n",
        "\n",
        "\n",
        "    \n",
        "    def save(self, path):\n",
        "        self.model.save_weights(path)\n",
        "    \n",
        "    def load(self, path):\n",
        "        self.model.load_weights(path)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "D-siq1_GxoP1",
        "colab_type": "code",
        "outputId": "088c3800-9725-449c-d9bd-65dc67ae46bf",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "WEIGHTS_PATH_NO_TOP"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'resnet50_weights_tf_dim_ordering_tf_kernels_notop.h5'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Kx7Fww_f-Tqt",
        "colab_type": "text"
      },
      "source": [
        "### 4. Read csv files\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SWt5X9Fa-Tqu",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def read_testset(filename=\"drive/My Drive/data/classification_labels.txt\"):\n",
        "    ''' \n",
        "    Data in data folder\n",
        "    '''\n",
        "    wb = gc.open_by_url('https://docs.google.com/spreadsheets/d/1SoFbAYTInDf3eU7zBVJydmqUzgqFZ35-XL6Vk7pNAQk')\n",
        "    sheet = wb.sheet1\n",
        "    data = sheet.get_all_values()\n",
        "    df = pd.DataFrame(data)\n",
        "    cols = [1, 2,3, 4]\n",
        "    df.drop(df.columns[cols],axis=1,inplace=True)\n",
        "    df.columns = df.iloc[0]\n",
        "    df = df.iloc[1:]\n",
        "    df['Assay_no.']=df['Assay_no.'].astype(str).astype(int)\n",
        "    #df['Assay_no.'].dtypes\n",
        "    df['Assay_no.'] = df['Assay_no.'].apply(lambda x: '{0:0>4}'.format(x))\n",
        "    df['Assay_no.']= df['Assay_no.'].astype(str)\n",
        "    df[\"Result\"] = df[\"Result\"].astype('category')\n",
        "    df[\"Result\"] = df[\"Result\"].cat.codes\n",
        "    df[\"Result\"] = df[\"Result\"].astype('int')\n",
        "    #df = pd.read_csv(filename, sep=\" \", header=None)\n",
        "    df.rename(columns={'Assay_no.': 'ID', 'Result': 'label'})\n",
        "    df.columns = [\"ID\", \"label\"]\n",
        "    return df\n",
        "\n",
        "def read_trainset(filename=\"drive/My Drive/data/classification_labels.txt\"):\n",
        "    ''' \n",
        "    Data in data folder\n",
        "    '''\n",
        "    wb = gc.open_by_url('https://docs.google.com/spreadsheets/d/16isnCyge5lJ3LZ6viAv8U6zOTpQ3QU2B9ikxoosLTZA')\n",
        "    sheet = wb.sheet1\n",
        "    data = sheet.get_all_values()\n",
        "    df = pd.DataFrame(data)\n",
        "    cols = [1, 2,3, 4]\n",
        "    df.drop(df.columns[cols],axis=1,inplace=True)\n",
        "    df.columns = df.iloc[0]\n",
        "    df = df.iloc[1:]\n",
        "    df['Assay_no.']=df['Assay_no.'].astype(str).astype(int)\n",
        "    #df['Assay_no.'].dtypes\n",
        "    df['Assay_no.'] = df['Assay_no.'].apply(lambda x: '{0:0>4}'.format(x))\n",
        "    df['Assay_no.']= df['Assay_no.'].astype(str)\n",
        "    #df = pd.read_csv(filename, sep=\" \", header=None)\n",
        "    df[\"Result\"] = df[\"Result\"].astype('category')\n",
        "    df[\"Result\"] = df[\"Result\"].cat.codes\n",
        "    df[\"Result\"] = df[\"Result\"].astype('int')\n",
        "    df.rename(columns={'Assay_no.': 'ID', 'Result': 'label'})\n",
        "    df.columns = [\"ID\", \"label\"]\n",
        "    return df\n",
        "\n",
        "    \n",
        "test_df = read_testset()\n",
        "df = read_trainset()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_fVDXUo3-Tq6",
        "colab_type": "code",
        "outputId": "0f3366dc-ffa0-4bdb-dbc0-7efa64609d82",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "\n",
        "test_df[\"ID\"].dtype"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "dtype('O')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 29
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wL4LR8Gw-TrA",
        "colab_type": "code",
        "outputId": "809421e5-43c2-454a-ecef-a009a41c1ca6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 139
        }
      },
      "source": [
        "test_df.head(3)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>ID</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0049</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0050</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0051</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     ID  label\n",
              "1  0049      0\n",
              "2  0050      0\n",
              "3  0051      1"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1WEYOH5x-TrF",
        "colab_type": "text"
      },
      "source": [
        "### 5. Train model and predict\n",
        "\n",
        "*Using train, validation and test set* <br>\n",
        "\n",
        "Training for 6 epochs with Adam optimizer, with a learning rate of 0.0005 and decay rate of 0.8. The validation predictions are \\[exponentially weighted\\] averaged over all 6 epochs (same goes for the test set submission later). `fit_and_predict` returns validation and test predictions for all epochs.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zQpkQBx8-TrG",
        "colab_type": "code",
        "outputId": "306168b3-7eca-4a3c-8c85-ee3e390e2c1c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "# train set (80%) and validation set (20%)\n",
        "ss = ShuffleSplit(n_splits=10, test_size=0.1, random_state=42).split(df.index)\n",
        "\n",
        "# lets go for the first fold only\n",
        "train_idx, valid_idx = next(ss)\n",
        "\n",
        "# obtain model\n",
        "model = MyDeepModel(engine=ResNet50, input_dims=(224,224, 3),\n",
        "                    batch_size=10, learning_rate=5e-5, n_classes=2,n_augment=0,\n",
        "                    num_epochs=1, decay_rate=0.6, decay_steps=1,\n",
        "                    weights='imagenet', verbose=2)\n",
        "\n",
        "# obtain test + validation predictions (history.test_predictions, history.valid_predictions)\n",
        "history = model.fit_and_predict(df.iloc[train_idx], df.iloc[valid_idx], test_df)\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/1\n",
            "Self image size (224, 224, 3)\n",
            "Epoch 1/1\n",
            "Self image size (224, 224, 3)\n",
            "Self image size (224, 224, 3)\n",
            "Self image size (224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0039.jpg\n",
            "drive/My Drive/data/LFA_Iamges/Train/0001.jpg\n",
            "drive/My Drive/data/LFA_Iamges/Train/0008.jpg\n",
            "drive/My Drive/data/LFA_Iamges/Train/0043.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0005.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0022.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0026.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0003.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0037.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0032.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0040.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0019.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0048.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0030.jpg\n",
            "Self image size (224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0006.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0012.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0004.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0045.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0042.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0007.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0009.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0010.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0011.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0031.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0029.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0002.jpg\n",
            "drive/My Drive/data/LFA_Iamges/Train/0017.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0015.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0014.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0021.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0035.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0036.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0018.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0024.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0013.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0046.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0033.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0016.jpg\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0034.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0023.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0020.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0047.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0038.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "Self image size (224, 224, 3)\n",
            "Self image size (224, 224, 3)\n",
            "Self image size (224, 224, 3)\n",
            "Self image size (224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0002.jpg\n",
            "drive/My Drive/data/LFA_Iamges/Train/0007.jpg\n",
            "drive/My Drive/data/LFA_Iamges/Train/0033.jpg\n",
            "drive/My Drive/data/LFA_Iamges/Train/0040.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0011.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0036.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0004.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0018.jpg\n",
            "drive/My Drive/data/LFA_Iamges/Train/0021.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0034.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0023.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "Self image size (224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0030.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0029.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0022.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0017.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0048.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0019.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0039.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0042.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0020.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0008.jpg\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0026.jpg\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0016.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0005.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0006.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0043.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0003.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0037.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0009.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0038.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0047.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0001.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0015.jpg\n",
            " - 22s - loss: nan - precision_2: 0.0000e+00 - recall_2: 0.0000e+00\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0045.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0031.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0010.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0013.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0046.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0032.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "samples size in read: (3, 224, 224, 3)\n",
            "Dim data gen (3, 224, 224, 3)\n",
            "drive/My Drive/data/LFA_Iamges/Train/0035.jpg\n",
            "samples size in read: (3, 224, 224, 3)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uCRk9n8S0E7V",
        "colab_type": "text"
      },
      "source": [
        "Using the Tensor graph analysis"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XUscXfsW-TrO",
        "colab_type": "text"
      },
      "source": [
        "### 6. Submit test predictions"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ghTcolzR-TrQ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "test_df.iloc[:, :] = np.average(history.test_predictions, axis=0, weights=[2**i for i in range(len(history.test_predictions))])\n",
        "test_df = test_df.stack().reset_index()\n",
        "\n",
        "test_df.insert(loc=0, column='ID', value=test_df['Image'].astype(str) + \"_\" + test_df['Diagnosis'])\n",
        "\n",
        "test_df = test_df.drop([\"Image\", \"Diagnosis\"], axis=1)\n",
        "\n",
        "test_df.to_csv('submission.csv', index=False)"
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}