MRNet_EDA.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "MRNet-EDA.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "mount_file_id": "1SCrB_H_qbU0bZZF5MCA-JaKgEf3cAr9U",
      "authorship_tag": "ABX9TyNgRfdnYK++9RwJgOBOlgaG",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/ptnv-s/Knee-Abnormality-and-Common-Disorders-Detection/blob/main/MRNet_EDA.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Deep Learning Based Musculoskeletal Diagnosis, Abnormality and Common Disorders detection using Knee Magnetic Resonance Imaging"
      ],
      "metadata": {
        "id": "fosR27toc8kH"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "---\n",
        "\n",
        "Magnetic resonance (MR) imaging of the knee is the standard of care imaging modality to evaluate knee disorders, and more musculoskeletal MR examinations are performed on the knee than on any other region of the body.\n",
        "\n",
        "---\n",
        "\n",
        "We will be implementing deep learning baselines for detecting general abnormalities and specific diagnoses (anterior cruciate ligament [ACL] tears and meniscal tears) on knee MRI exams.\n",
        "\n",
        "---\n",
        "\n",
        "The most common indications for the knee MRI examinations in this study included acute and chronic pain, follow-up or preoperative evaluation, injury/trauma.\n",
        "\n",
        "---\n",
        "We have used the [MRNet dataset](https://stanfordmlgroup.github.io/competitions/mrnet/)  which consists of 1,370 knee MRI exams performed at Stanford University Medical Center. The dataset contains 1,104 (80.6%) abnormal exams, with 319 (23.3%) ACL tears and 508 (37.1%) meniscal tears. The labels were obtained through manual extraction from clinical reports.\n",
        "\n",
        "---\n",
        "\n",
        "Examinations were performed with GE scanners (GE Discovery, GE Healthcare, Waukesha, WI) with standard knee MRI coil and a routine non-contrast knee MRI protocol that included the following sequences: coronal T1 weighted, coronal T2 with fat saturation, sagittal proton density (PD) weighted, sagittal T2 with fat saturation, and axial PD weighted with fat saturation. A total of 775 (56.6%) examinations used a 3.0-T magnetic field the remaining used a 1.5-T magnetic field as reported [here](https://stanfordmlgroup.github.io/projects/mrnet/) .\n",
        "\n",
        "For more details refer to the paper related to this study [here](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002699).\n",
        "\n",
        "---\n",
        "\n",
        "Given a MRI scan our deep learning model will be predicting 3 outcomes for knee MRI exams (anterior cruciate ligament [ACL] tears, meniscal tears, and general abnormalities)."
      ],
      "metadata": {
        "id": "dT4KaPSJgPWT"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Importing Basic Libraries"
      ],
      "metadata": {
        "id": "gLgaQloxVQNB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import csv\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "from glob import glob"
      ],
      "metadata": {
        "id": "Eh8xmBGuU85P"
      },
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "id": "iqfOzZgQyj5d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "52c938bc-79ee-440a-a365-f325f49ea116"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "cd /content/drive/MyDrive/Medical/MRNet-v1.0"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NmbRdqMYxple",
        "outputId": "c678a7b4-f01a-4513-9317-d3375a6afb5b"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/drive/MyDrive/Medical/MRNet-v1.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pwd"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "sTy3NYRXxv2X",
        "outputId": "11260ebe-faa9-4481-a25e-f5f03088f3c4"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'/content/drive/MyDrive/Medical/MRNet-v1.0'"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Exploring Dataset and the File/Folder Structure\n",
        "This data comes as split dataset (training & validation in separate directories). \n",
        "Each data categories such as training & validation has sub-directories (pls see below)."
      ],
      "metadata": {
        "id": "KybYHW1hVYjV"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!ls"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "voC_7uukx0xM",
        "outputId": "f3c4eee3-a9f7-435b-8159-b26d02c6de20"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "train\t\t    train-acl.csv\tvalid\t\t    valid-acl.csv\n",
            "train-abnormal.csv  train-meniscus.csv\tvalid-abnormal.csv  valid-meniscus.csv\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Lets Explore the Training Set of the Dataset.\n",
        "We will try to find out the total number of the training examples and the examples in each category of conclusion of the knee MRI examinations made by the clinical expert.\n"
      ],
      "metadata": {
        "id": "t6nOdyzvV2h0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "with open(\"/content/drive/MyDrive/Medical/MRNet-v1.0/train-abnormal.csv\") as csv_file:\n",
        "    csv_reader = csv.reader(csv_file)\n",
        "    line_count = 0\n",
        "    trainAb_labels=[]\n",
        "    for row in csv_reader:\n",
        "           trainAb_labels.append(int(row[1]))\n",
        "trainAb_labels= np.array(trainAb_labels)\n",
        "\n",
        "with open(\"/content/drive/MyDrive/Medical/MRNet-v1.0/train-acl.csv\") as csv_file:\n",
        "    csv_reader = csv.reader(csv_file)\n",
        "    line_count = 0\n",
        "    trainACL_labels=[]\n",
        "    for row in csv_reader:\n",
        "           trainACL_labels.append(int(row[1]))\n",
        "trainACL_labels= np.array(trainACL_labels)\n",
        "\n",
        "\n",
        "\n",
        "with open(\"/content/drive/MyDrive/Medical/MRNet-v1.0/train-meniscus.csv\") as csv_file:\n",
        "    csv_reader = csv.reader(csv_file)\n",
        "    line_count = 0\n",
        "    trainMen_labels=[]\n",
        "    for row in csv_reader:\n",
        "           trainMen_labels.append(int(row[1]))\n",
        "trainMen_labels= np.array(trainMen_labels)\n",
        "\n",
        "print(\"Total No of Observations in the Training Set : \"+str(len(trainAb_labels)))\n",
        "print(\"No of Abnormal Investigation Cases : \"+str(len(trainAb_labels)-np.count_nonzero(trainAb_labels)))\n",
        "print(\"No of ACL tears Investigation Cases : \"+str(len(trainACL_labels)-np.count_nonzero(trainACL_labels)))\n",
        "print(\"No of Meniscal tears Investigation Cases : \"+str(len(trainMen_labels)-np.count_nonzero(trainMen_labels)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KWV7iHivzW9I",
        "outputId": "c17de840-0035-4237-f115-9c0b6df7a0b0"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Total No of Observations in the Training Set : 1130\n",
            "No of Abnormal Investigation Cases : 217\n",
            "No of ACL tears Investigation Cases : 922\n",
            "No of Meniscal tears Investigation Cases : 733\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### We now try to find the same above investigation for validation set."
      ],
      "metadata": {
        "id": "qGM241i_W4Y1"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "U4pEnRbJ8zbJ",
        "outputId": "a057f048-58a6-433d-f31d-0d5ea14aa9a1",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "with open(\"/content/drive/MyDrive/Medical/MRNet-v1.0/valid-abnormal.csv\") as csv_file:\n",
        "    csv_reader = csv.reader(csv_file)\n",
        "    line_count = 0\n",
        "    testAb_labels=[]\n",
        "    for row in csv_reader:\n",
        "           testAb_labels.append(int(row[1]))\n",
        "\n",
        "testAb_labels= np.array(testAb_labels)\n",
        "\n",
        "with open(\"/content/drive/MyDrive/Medical/MRNet-v1.0/valid-acl.csv\") as csv_file:\n",
        "    csv_reader = csv.reader(csv_file)\n",
        "    line_count = 0\n",
        "    testACL_labels=[]\n",
        "    for row in csv_reader:\n",
        "           testACL_labels.append(int(row[1]))\n",
        "\n",
        "testACL_labels= np.array(testACL_labels)\n",
        "\n",
        "\n",
        "with open(\"/content/drive/MyDrive/Medical/MRNet-v1.0/valid-meniscus.csv\") as csv_file:\n",
        "    csv_reader = csv.reader(csv_file)\n",
        "    line_count = 0\n",
        "    testMen_labels=[]\n",
        "    for row in csv_reader:\n",
        "           testMen_labels.append(int(row[1]))\n",
        "\n",
        "testMen_labels= np.array(testMen_labels)\n",
        "\n",
        "\n",
        "print(\"Total No of Observations in the Training Set : \"+str(len(testAb_labels)))\n",
        "print(\"No of Abnormal Investigation Cases : \"+str(len(testAb_labels)-np.count_nonzero(testAb_labels)))\n",
        "print(\"No of ACL tears Investigation Cases : \"+str(len(testACL_labels)-np.count_nonzero(testACL_labels)))\n",
        "print(\"No of Meniscal tears Investigation Cases : \"+str(len(testMen_labels)-np.count_nonzero(testMen_labels)))"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Total No of Observations in the Training Set : 120\n",
            "No of Abnormal Investigation Cases : 25\n",
            "No of ACL tears Investigation Cases : 66\n",
            "No of Meniscal tears Investigation Cases : 68\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_dir = 'train'\n",
        "valid_dir = 'valid'"
      ],
      "metadata": {
        "id": "hq-fzGHR4ZhZ"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "##### Now let us check for whether there is any disparity in the dataset.\n",
        "\n",
        "As we know this dataset has 3 views of Knee since datset has MRI scans namely - \n",
        "\n",
        "*   Axial\n",
        "*   Coronal\n",
        "*   Sagittal\n",
        "\n",
        "\n",
        " \n",
        " "
      ],
      "metadata": {
        "id": "9pA07_X1XFuK"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "planes = ['axial', 'coronal', 'sagittal']\n",
        "datasets = {'train': train_dir, 'valid': valid_dir}\n",
        "\n",
        "for dataset, path in datasets.items():\n",
        "    print(f'\\nNumber of exams in {dataset} set:')\n",
        "    for plane in planes:\n",
        "        print(plane, len(glob(f'{path}/{plane}/*.npy')))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GORw8xXm7xlT",
        "outputId": "22838b70-fde8-4bb7-c6f5-6c5eeb4bc070"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Number of exams in train set:\n",
            "axial 1130\n",
            "coronal 1130\n",
            "sagittal 1130\n",
            "\n",
            "Number of exams in valid set:\n",
            "axial 120\n",
            "coronal 120\n",
            "sagittal 120\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##### Lets us check the dimensions of each file/ scan of a particular view of a patient's Knee."
      ],
      "metadata": {
        "id": "8telFCchX4zU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_axial_exams = glob(f'{train_dir}/axial/*.npy')[:5]\n",
        "\n",
        "for exam in train_axial_exams:\n",
        "    series = np.load(exam)\n",
        "    print(series.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qsCbOqbc759e",
        "outputId": "e5ffa973-7f4f-4c3e-9801-918a62cce23a"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(32, 256, 256)\n",
            "(32, 256, 256)\n",
            "(39, 256, 256)\n",
            "(36, 256, 256)\n",
            "(39, 256, 256)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Helper Functon for plotting"
      ],
      "metadata": {
        "id": "sNzSstUtYDfz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def plot_first_ten(path):\n",
        "    series = np.load(path)[:10]\n",
        "\n",
        "    fig = plt.figure(figsize=(20, 6))\n",
        "    \n",
        "    for i, image in enumerate(series):\n",
        "        ax = fig.add_subplot(1, 10, i+1, xticks=[], yticks=[])\n",
        "        plt.imshow(image, cmap='gray')\n",
        "        plt.axis('off');"
      ],
      "metadata": {
        "id": "ICvovvgR8IXF"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Plotting results-\n",
        "\n",
        "(Commented Since Dataset requires special access - apply [here](https://stanfordmlgroup.github.io/competitions/mrnet/))"
      ],
      "metadata": {
        "id": "783rYXNHYHbr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#plot_first_ten(train_axial_exams[0])"
      ],
      "metadata": {
        "id": "jNI3rsOD8NGk"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Helper Function for Plotting the distribution "
      ],
      "metadata": {
        "id": "iL7OFqPvYj5E"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def get_num_images_per_exam(exams):\n",
        "    num_images_per_exam = []\n",
        "    \n",
        "    for exam in exams:\n",
        "        series = np.load(exam)\n",
        "        num_images_per_exam.append(series.shape[0])\n",
        "        \n",
        "    return np.asarray(num_images_per_exam)\n",
        "\n",
        "def plot_num_images_per_exam(dataset_path):\n",
        "    f, axes = plt.subplots(1, 3, figsize=(20, 6), sharex=True)\n",
        "    \n",
        "    for i, plane in enumerate(planes):\n",
        "        paths = glob(f'{dataset_path}/{plane}/*.npy')\n",
        "        \n",
        "        num_images = get_num_images_per_exam(paths)\n",
        "        print(f'{plane} - min: {num_images.min()}, max: {num_images.max()}')\n",
        "\n",
        "        sns.distplot(num_images, ax=axes[i])\n",
        "        axes[i].set_title(f'{plane} plane')"
      ],
      "metadata": {
        "id": "18zU9gqk8W6V"
      },
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Visualise distrubutions of number of images in a series\n",
        "\n",
        "Let's have a look at the distributions for **train**  and **valid** datasets."
      ],
      "metadata": {
        "id": "PsLTp4F2ZiiQ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "plot_num_images_per_exam(train_dir)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 554
        },
        "id": "IvJ0t4EX93DQ",
        "outputId": "ac0369be-63e0-4d37-8840-cbc2edfef793"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "axial - min: 19, max: 61\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "coronal - min: 17, max: 58\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "sagittal - min: 17, max: 51\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABJUAAAF1CAYAAABRSNv7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzde5xcdZnv+8/T9/u900lf0p2QCyQhIgSCCIKCiKOCbkEQVPTgcdyOe2Z0HIc9e28G2XNmxOPIGY/MqKMyKIOgqGNGgyggIIKBBMg9IZ2kO53upO/pa/paz/6jqpmm6SRd3XXpqv6+X6+8UrXWb631VF6v1K/Ws36/52fujoiIiIiIiIiISDhS4h2AiIiIiIiIiIgkHiWVREREREREREQkbEoqiYiIiIiIiIhI2JRUEhERERERERGRsCmpJCIiIiIiIiIiYVNSSUREREREREREwqakksgkZnaZme2fYduPm9mzs7zOnWb2wGyOFRGR5GFmbmYrZnFcXejYtGjEJSIi4TOzW8zs1xE+p/oJmdeUVBKZxN1/5+6r4x2HiIiIiIgkFnf/N3e/euL91ISQmV1hZkfjE51IdCipJCIiInIKesIrIiIicmpKKklSMrPbzeygmfWZ2R4z+8Ckff9sZj+Z9P5uM3vCgl739OB05znD9SeGm37KzFrM7JiZfeE07X9sZsfNrMfMnjGztZP2/auZ3WtmvwzFscXMzpq0/2wz+42ZdZnZfjP7UDj/ViIiycrMaszsp2bWbmadZvaN0PYUM/ufZtZoZm1m9n0zKwztm/j+vs3MjgBPzrD9rWZ2xMw6zOx/TIrhIjN73sxOhPqCb5hZxgzjf8rM/t7MXjCzXjP7uZmVnKLtJ8xsb6ifOGRmfzxp3xVmdtTM/iIU/zEz+8Sk/Zlm9tVQ/K1m9k0zy57VP7qISIIws78ys+bQ9+Z+M7sytP2039tmdnWofY+Z/ZOZPW1mnwzte608hpk9Ezpku5n1m9mtwKNAZeh9v5lVqp+QRKekkiSrg8BlQCHwJeABM1sS2vcXwLmhL/3LgNuAW93dwzzPTLwdWAlcDfyVmV11inaPhtotAl4C/m3K/ptC1y8G6oH/B8DMcoHfAA+Gjr0J+CczWxNGjCIiScfMUoFfAI1AHVAFPBTa/fHQn7cDy4E84BtTTnE5cA7wrhm2vxRYDVwJ3GFm54S2jwOfA8qAt4T2fyaMj/Ix4P8ClgBjwNdP0a4NeC9QAHwCuMfMzp+0fzHBvqyKYL93r5kVh/Z9GVgFnAesCLW5I4wYRUQSipmtBj4LXOju+QS/6xtCu0/5vW1mZcAjwH8HSoH9wCXTXcPd3xZ6+SZ3z3P3+4F3Ay2h93nu3nK6682Q+gmJKyWVJCm5+4/dvcXdA+7+MHAAuCi0bxD4KPA14AHgv7n7tHObT3eeGfqSuw+4+07gPuDDp7jO99y9z92HgTuBN008BQ/5mbu/4O5jBBNO54W2vxdocPf73H3M3V8GfgLcEEaMIiLJ6CKgEvjL0PfwkLtPLK5wC/A1dz/k7v0Ebw5ustdPdbszdNzJGbb/krufdPftwHbgTQDuvs3d/xD6jm4AvkUwYTVTP3D3Xe4+APwv4EOhhNnruPsv3f2gBz0N/JrgQ5EJo8Bd7j7q7puBfmC1mRnwKeBz7t7l7n3A3xF8SCEikqzGgUxgjZmlu3uDux+EM35v/xGw291/Gvpd/nXg+FwCUT8hiU51AiQpmdnHgM8TfDoNwafKZRP73X2LmR0iOLrnR7M9zww0TXrdCJw7zTVSCY48ugEoBwKhXWVAT+j15M5qMBQHQC2w0cxOTNqfBvwgjBhFRJJRDdAY+tE/VSXB7+QJjQS/OysmbWsKs/2039NmtorgQ4wNQE7ouG1hfI6p/Ug60/RDZvZu4G8IPklOCV1r56QmnVP+LSZiLA+13Ra8bwieDnjDDYmISLJw93oz+3OCD3PXmtljwOfdveUM39uVTPpedne3ORbeVj8hiU4jlSTpmFkt8C8Eh7SWunsRsIvgl99Emz8h+HSiBfjibM8zAzWTXi8NXW+qm4HrgKsIDjmtmwhhBudvAp5296JJf/Lc/b+GEaOISDJqApba9IW2Wwgm5ScsJThloHXSNg+z/an8M7APWOnuBcBfM7d+ZBTomNzAzDIJjlL9KlAR6q82z/A6HcBJYO2kfqTQ3fPOdKCISCJz9wfd/VKC3+8O3B3adbrv7WNA9cQ5QqN4Xns/k8tOs039hCQ0JZUkGeUS/MJuh2BROmDdxM7Q04C/BT5CcBrcF83svHDPM0P/y8xyLFh4+xPAw9O0yQeGgU6CTwH+Lozz/wJYZWYfNbP00J8LJ9XyEBFZqF4g+OP/y2aWa2ZZZvbW0L4fAp8zs2Vmlkfwe/fhU4xqmk37yfKBXqDfzM4Gwk36f8TM1phZDnAX8Ii7j09pk0HwQUk7MBZ6Gn01M+DuAYIPUO4xs0UAZlZlZu8KM04RkYRhZqvN7B2hZMsQwaTJxGyB031v/5Jgbdb3hx5a/AnBWkSn0kqwFt/k96VTylyon5CEpqSSJB133wP8A/A8wS/uc4Hfw2tLQz8A3O3u2939AMGnAT8IdSozOk8YniZYWPsJ4Kvu/utp2nyf4FDVZmAP8IeZnjw0p/lqgnOaWwhOv7ibYKchIrJghX5Qv49gQdEjwFHgxtDu7xGcJvwMcJjgDcV/O83pwm0/2RcIjkjtI/ijfLqHC6fzA+BfCX6/ZwF/OrVBqC/4U4LTubtD19sUxjX+imBf9Qcz6wUeJ1h0XEQkWWUSLD7dQfD7dRHBenlwmu9td+8gWLLiKwQfCK8BthJ8QDydO4H7Qyu7fcjd9xF8UHEotK3ydNebIfUTElfm0y54JSJzYWZ1BG880mf4JFtEROR1zOwp4AF3/068YxERkTcysxSCDy1ucfffxuH6T6F+QuJMI5VEREREREREZsDM3mVmRaFZDhP1j2Y800Ak2SipJCIiIiIiIjIzbwEOEpw69z7g/e5+Mr4hicSPpr+JiIiIiIiIiEjYNFJJRERERERERETCpqSSiIiIiIiIiIiELS3eAURKWVmZ19XVxTsMEZF5adu2bR3uXh7vOOJJ/YSIyPTURwSpnxARmd7p+omkSSrV1dWxdevWeIchIjIvmVljvGOYjpldA/wjkAp8x92/PGV/JvB94AKgE7jR3RvMLB34DnA+wb7s++7+96e7lvoJEZHpzdc+ItbUT4iITO90/YSmv4mISFyYWSpwL/BuYA3wYTNbM6XZbUC3u68A7gHuDm2/Ach093MJJpz+2MzqYhG3iIiIiIgEKakkIiLxchFQ7+6H3H0EeAi4bkqb64D7Q68fAa40MwMcyDWzNCAbGAF6YxO2iIiIiIiAkkoiIhI/VUDTpPdHQ9umbePuY0APUEowwTQAHAOOAF91965oBywiIiIiIv9JSSUREUlEFwHjQCWwDPgLM1s+tZGZfcrMtprZ1vb29ljHKCIiIiKS1JRUEhGReGkGaia9rw5tm7ZNaKpbIcGC3TcDv3L3UXdvA34PbJh6AXf/trtvcPcN5eULfmEjEREREZGIUlJJRETi5UVgpZktM7MM4CZg05Q2m4BbQ6+vB550dyc45e0dAGaWC1wM7ItJ1CIiIiIiAiipJCIicRKqkfRZ4DFgL/Ajd99tZneZ2bWhZt8FSs2sHvg8cHto+71AnpntJpicus/dd8T2E4iIiIiILGxp8Q5AREQWLnffDGyesu2OSa+HgBumOa5/uu0iIiIiIhI7GqkkIiIiIiIiIiJhU1JJRERERERERETCpqSSiIiIiIiIiIiETUklEREREREREREJm5JKIiIiIiIiIiISNq3+JhJFD245Elb7mzcujVIkIrLQhft9NEHfSyIiMluz6XvU74gkFo1UEhERERERERGRsCmpJCIiIiIiIiIiYVNSSUREREREREREwqakkoiIiIiIiIiIhE1JJRERERERERERCZuSSiIiIiIiIiIiEjYllUREREREREREJGxKKomIiIiIiIiISNiUVBIRERERERERkbApqSQiIiIiIiIiImFTUklERERERERERMKmpJKIiIiIiIiIiIRNSSUREREREREREQmbkkoiIiIiIiIiIhI2JZVERERERERERCRsSiqJiIiIiIiIiEjY0uIdgCw8D245Elb7mzcujVIkIiIiIiIiIjJbGqkkIiIiIiIiIiJhU1JJRERERERERETCpqSSiIiIiIiIiIiETUklEREREREREREJm5JKIiIiIiIiIiISNiWVREREREREREQkbEoqiYiIiIiIiIhI2KKaVDKza8xsv5nVm9nt0+zPNLOHQ/u3mFndpH3rzex5M9ttZjvNLCuasYqIiIiIiIiIyMxFLalkZqnAvcC7gTXAh81szZRmtwHd7r4CuAe4O3RsGvAA8Gl3XwtcAYxGK1YREYmP2T58MLNbzOyVSX8CZnZerOMXEZHomkM/kW5m94ceTu81s/8e69hFRBaCaI5Uugiod/dD7j4CPARcN6XNdcD9odePAFeamQFXAzvcfTuAu3e6+3gUYxURkRiby8MHd/83dz/P3c8DPgocdvdXYhe9iIhE21z6CeAGINPdzwUuAP548qwIERGJjGgmlaqApknvj4a2TdvG3ceAHqAUWAW4mT1mZi+Z2RejGKeIiMTHXB4+TPbh0LEiIpJc5tJPOJAbmgGRDYwAvbEJW0Rk4ZivhbrTgEuBW0J/f8DMrpzayMw+ZWZbzWxre3t7rGMUEZG5mcvDh8luBH4YpRhFRCR+5tJPPAIMAMeAI8BX3b1r6gV0PyEiMjfRTCo1AzWT3leHtk3bJvQUoRDoJNhhPOPuHe4+CGwGzp96AXf/trtvcPcN5eXlUfgIIiIyn5nZRmDQ3XedYr9uFkREFqaLgHGgElgG/IWZLZ/aSPcTIiJzE82k0ovASjNbZmYZwE3ApiltNgG3hl5fDzzp7g48BpxrZjmhZNPlwJ4oxioiIrE3l4cPE27iNKOUdLMgIpLQ5tJP3Az8yt1H3b0N+D2wIeoRi4gsMFFLKoWGn36WYIJoL/Ajd99tZneZ2bWhZt8FSs2sHvg8cHvo2G7gawQTU68AL7n7L6MVq4iIxMVcHj5gZinAh1A9JRGRZDWXfuII8A4AM8sFLgb2xSRqEZEFJC2aJ3f3zQSnrk3edsek10MEV2aY7tgHgAeiGZ+IiMSPu4+Z2cTDh1TgexMPH4Ct7r6J4MOHH4QePnQRvKGY8Dagyd0PxTp2ERGJvjn2E/cC95nZbsCA+9x9R+w/hYhIcotqUklEROR05vjw4SmCT55FRCRJzbafcPf+6baLiEhkzdfV30REREREREREZB5TUklERERERERERMKmpJKIiIiIiIiIiIRNSSUREREREREREQmbkkoiIiIiIiIiIhI2JZVERERERERERCRsSiqJiIiIiIiIiEjYlFQSEREREREREZGwKakkIiIiIiIiIiJhU1JJRERERERERETCpqSSiIiIiIiIiIiETUklEREREREREREJm5JKIiIiIiIiIiISNiWVREREREREREQkbEoqiYiIiIiIiIhI2JRUEhERERERERGRsCmpJCIiIiIiIiIiYVNSSUREREREREREwpYW7wBEksnoeID/7/FX+c2eVlYsymPlonwqCrLiHZaIiIiIiIhIxGmkkkiEjI4HuOU7W7j3twcpysnguYOd/OtzDfSeHI13aCIiIiIiIiIRp6SSSIR86+mDvHC4i698cD0/+uO38OAnL+bk6Dg/fPEI7h7v8EREREREREQiSkklkQiob+vn60/U8571S/jQhTUArKks4Jq1i2nsHORw50CcIxQRERERERGJLCWVRCLgn35bT1qq8aVr175u+wW1xeRmpPK7VzviFJmIiIiIiIhIdCipJDJHLSdOsml7CzdeWENZXubr9qWnpnDxWaXsb+2jrW8oThGKiIiIiIiIRJ6SSiJz9L1nD+PAbZcum3b/RXUlGLCruSemcYmIiIiIiIhEk5JKInMwPDbOj7cd5Zp1i6kuzpm2TX5WOtXF2ew91hfj6ERERERERESiR0klkTl4cm8bPSdHueGC6tO2W1NZSPOJk5wYHIlRZCIiIiIiIiLRlRbvAEQSyYNbjrzu/Q+ebyA/K42j3SffsG+yc5bk89ju4+w93sdblpdGOUoRERERERGR6NNIJZFZ6h8eY39rH+fVFJFidtq2i/KzKM3NYP/x3hhFJyIiIiIiIhJdSiqJzNL+430EHNZXF82o/fLyPBo7Bwm4RzkyERERERERkehTUklklvYf76UgK43KwqwZta8rzWF4LMDxnqEoRyYiIiIiIiISfUoqiczCeMA50NbPqop87AxT3ybUleUC0Ng5EM3QRBKKmV1jZvvNrN7Mbp9mf6aZPRzav8XM6ibtW29mz5vZbjPbaWYzy/CKiIiIiEhEKKkkMgsNnQMMjwU4e3H+jI8pyk6nMDudhs7BKEYmkjjMLBW4F3g3sAb4sJmtmdLsNqDb3VcA9wB3h45NAx4APu3ua4ErgNEYhS4iIiIiIiipJDIr+4/3kZpinLUob8bHmBm1pTk0dA7gqqskAnARUO/uh9x9BHgIuG5Km+uA+0OvHwGutODwwKuBHe6+HcDdO919PEZxi4iIiIgISiqJzMr+1j6WleWSmZYa1nF1pbn0DY1xYlADKkSAKqBp0vujoW3TtnH3MaAHKAVWAW5mj5nZS2b2xRjEKyIiIiIikyipJBKmroER2vuGWV0x86lvE6qKsgFoPnEy0mGJLDRpwKXALaG/P2BmV05tZGafMrOtZra1vb091jGKiIiIiCQ1JZVEwrT/eC8Aq8OopzRhcWEWKQYtPUoqiQDNQM2k99WhbdO2CdVRKgQ6CY5qesbdO9x9ENgMnD/1Au7+bXff4O4bysvLo/ARREREREQWrqgmlWa7qo+Z1ZnZSTN7JfTnm9GMUyQc+1v7KM3NoCwvM+xj01NTWJSfRYtGKokAvAisNLNlZpYB3ARsmtJmE3Br6PX1wJMeLEr2GHCumeWEkk2XA3tiFLeIiIiIiBCcPhAVk1b1eSfBJ8ovmtkmd5/8o/+1VX3M7CaCq/rcGNp30N3Pi1Z8IrMxMhbgUPsAG5eVzPoclUVZ7G/tx90J1hsWWZjcfczMPkswQZQKfM/dd5vZXcBWd98EfBf4gZnVA10EE0+4e7eZfY1gYsqBze7+y7h8EBERERGRBSpqSSUmreoDYGYTq/pMTipdB9wZev0I8A3TXbbMYwfb+xkLOKsXF8z6HJVF2bx05AR9Q2MUZKdHMDqRxOPumwlOXZu87Y5Jr4eAG05x7APAA1ENUERERERETima09/msqoPwDIze9nMnjazy6IYp8iM7W/tIyMthbrSnFmfo7IwWKxbU+BEREREREQkkc3XQt3HgKXu/mbg88CDZvaGoSFa1Udiyd3Zf7yPFeV5pKXO/r/OksIsDGhWsW4RERERERFJYNFMKs16VR93H3b3TgB33wYcBFZNvYBW9ZFY2t/aR8/J0Vmt+jZZZnoqxbkZtPYORygyERERERERkdiLZlJp1qv6mFl5qNA3ZrYcWAkcimKsImf05L42AFZXzC2pBFCRn0lb79CczyMiIiIiIiISL1Er1D2XVX2AtwF3mdkoEAA+7e5d0YpVZCZ+u6+NysKsiBTXXlSQxf7WPsYCAdJS5ussVBEREREREZFTi+bqb7Ne1cfdfwL8JJqxiYTjxOAI2xq7uXxVZKZZVhRkEXDo6B9hcUFWRM4pIiIiIiIiEksaIiEyA88c6CDgsHrxG+rFz0pFQSaApsCJiIiIiIhIwlJSSWQGfruvjZLcDKqLsyNyvrK8TAxUrFtEREREREQSlpJKImcwHnCe2t/G5avKSTGLyDnTU1MozcukVSOVREREREREJEEpqSRyBq80naB7cJS3n70oouetKMikrU9JJREREREREUlMSiqJnMFv97WRmmJcvjIyRbonLMrPpLN/hLFAIKLnFREREREREYkFJZVEzuDJfW1csLSYwpz0iJ63LC8TB7oHRiN6XhEREREREZFYUFJJ5DSO9wyx51hvxKe+QTCpBNDRr2LdIiIiIiIikniUVBI5jaf2twHw9rMjO/UNoDQvA1BSSURERERERBJTWrwDEJnPntzXRmVhFqsr8iN+7pyMNHIyUunoH4n4uUVEJHwPbjkyq+Nu3rg0wpGIiIiIJAaNVBI5heGxcZ6t7+DtZy/CzKJyjbK8TDo1UklEREREREQSkJJKIqfwwuEuBkfGeUcU6ilNKMvL0PQ3ERERERERSUhKKomcwm/3tZOZlsIlZ5VF7RpleZn0Do0xPDYetWuIiIiIiIiIRIOSSiLTcHee2NfKW84qJTsjNWrXKQ2tANepukoiIiIiIiKSYJRUEpnGwfZ+GjsHufKciqhepyy0AlzngJJKIiIiIiIikliUVBKZxuN72wC4Mor1lABKc4MjlVRXSURERERERBKNkkoi03hibytrKwuoLMqO6nUy0lIozE6no09JJREREZGpzOwaM9tvZvVmdvs0+zPN7OHQ/i1mVjdp33oze97MdpvZTjPLimXsIiILgZJKIlN0DYywrbE76lPfJpTmagU4ERERkanMLBW4F3g3sAb4sJmtmdLsNqDb3VcA9wB3h45NAx4APu3ua4ErgNEYhS4ismAoqSQyxW/3tRFwuOqc6E59m1CWl6maSiIiIiJvdBFQ7+6H3H0EeAi4bkqb64D7Q68fAa40MwOuBna4+3YAd+90dy23KyISYUoqiUzxxL5WKgoyWVdZGJPrleVlMDgyzuDwWEyuJyLSc3KUJ/a28kpTN0OjuscSkXmrCmia9P5oaNu0bdx9DOgBSoFVgJvZY2b2kpl9MQbxiogsOGnxDkBkPhkeG+eZVzt435sqSUmxmFyzLC9UrFujlUQkBg519PPgliMMjgSTSYvyM/n05WeRlZ4a58hERCIqDbgUuBAYBJ4ws23u/sTkRmb2KeBTAEuXLo15kCIiiU4jlUQm2XKoi/7hsZhNfQMozdMKcCISG8Oj4zz8QhM5GWn82ZUr+cjGWjr6h3noxSME3OMdnojIVM1AzaT31aFt07YJ1VEqBDoJjmp6xt073H0Q2AycP/UC7v5td9/g7hvKy8uj8BFERJKbkkoikzy+t5Ws9BTeuqIsZtcszk0nxaBTSSURibJnDrTTNzzGDRdUU1GQxZrKAt5z7hJebe1nT0tvvMMTEZnqRWClmS0zswzgJmDTlDabgFtDr68HnnR3Bx4DzjWznFCy6XJgT4ziFhFZMJRUEgkJBJzHdh/n8lXlMZ0GkpaSQnFOBh39mv4mItHTNzTK7w50sL66kJqSnNe2b1xeSmluBk+92oZrtJKIzCOhGkmfJZgg2gv8yN13m9ldZnZtqNl3gVIzqwc+D9weOrYb+BrBxNQrwEvu/stYfwYRkWSnmkoiIS83ddPaO8y71y2J+bVL8zI0/U1EomrH0R7GAs7bV79+em+KGZevKuenLzdzoK2fVRX5cYpQROSN3H0zwalrk7fdMen1EHDDKY59AHggqgHKaY0HnN6hUXIz0shI03gGkWSkpJJIyKM7j5OearwjhvWUJpTmZdLYOYi7E1wFV0Qksl5pOkFlURYVBVlv2Hfe0iIe39vKHw51KqkkIiIRcbR7kHsef5WugRFyM9O49S21VBfnnPlAEUkoSirJtB7ccmTGbW/emPgrZbg7j+46zqUryijISo/59UtzMxgeC9A5MPLaanAiIpFS39ZP84mT/NG504/ETEtJYX11Ec8f7GRwZIycDP08EBGR2Ts5Ms7N/7KFwZEx3rt+Cb+v7+A7zx7mc1etojA79r+1RSR6NAZRBNjV3EvziZNxmfoGwaQSQGPnQFyuLxIvZnaNme03s3ozu32a/Zlm9nBo/xYzqwttrzOzk2b2SujPN2MdeyLZtL0FA9ZXF56yzfrqQsbd2a2C3SIiMkfff76BI12D3LKxlkvOKuO2S5czHnB+vft4vEMTkQhTUkkEeHTXMVJTjHeuqYjL9Utzg6OTGjsH43J9kXgws1TgXuDdwBrgw2a2Zkqz24Bud18B3APcPWnfQXc/L/Tn0zEJOkE9vb+NmpKc047ErCrKpjQ3g+1HT8QwMhERSTb9w2N88+mDvG1VOWeV5wFQkpvBpSvKeLnpBMd7huIcoYhEkpJKsuC5O7/adZyLl5dQHBoxFGtFuekY0KCkkiwsFwH17n7I3UeAh4DrprS5Drg/9PoR4EpT4bGw9AyOsqO5hxWL8k7bzsxYX13I4fYBBofHYhSdiIgkm807j9E9OMqfvmPF67ZftrKMtBTjxYauOEUmItGgpJIseK+29nOoY4Br4jT1DYL1TIpy0jX9TRaaKqBp0vujoW3TtgktLd0DlIb2LTOzl83saTO7bLoLmNmnzGyrmW1tb2+PbPQJ4vlDHbjDivLTJ5UAVi8uwIH69v7oByYiIklp0yst1JbmcEFt8eu252SksaaygJebuhkdD8QpOhGJNFXilAXv7l/tw4CB4bGwCpRHWmlupkYqiczcMWCpu3ea2QXAv5vZWnd/XUEgd/828G2ADRs2eBzijLtn6zvIzUilpuTMK+5UF2eTnZ7Kq639rK8uikF0IiKSTNr6hnjuYAefuWLFtCsaX1hXwo6jPew51sub1M+IJAWNVJIFzd3Z3nSCZWW5cVn1bbKS3AyOaKSSLCzNQM2k99WhbdO2MbM0oBDodPdhd+8EcPdtwEFgVdQjTkC/r+9k4/JSUlPOPGswxYwVi/I40NaH+4LMwYmIyBw8truVgMP73lQ57f5lZbnkZaaxR4tCiCQNjVSSBW1Xcy+dAyO8bWV5vEOhNC+DFxpG6RkcpTBHS63KgvAisNLMlhFMHt0E3DylzSbgVuB54HrgSXd3MysHutx93MyWAyuBQ7ELPTG09Q5xuGOAmy9aOuNjVlXksbO5h+O9QywpzI5idCIikmyePdBOVVE2qyqmn3KdYsbZi/PZ2dzDWCBAWsr8H+MQ7kyGmzfOvM8VSQbz/3+xSBRt2t5MqhlrqwriHQqloSLhjV0arSQLQ6hG0meBxzTkRMsAACAASURBVIC9wI/cfbeZ3WVm14aafRcoNbN64PPA7aHtbwN2mNkrBAt4f9rdVflzipebgiu5nV878ykGKxflA1DfprpKIiIyc2PjAZ472MmlK8qmnfo24ZwlBQyPBWjoUNkHkWSgkUqyYAUCzi92HGNlRR45GfH/r1CSmwlAY+egapnIguHum4HNU7bdMen1EHDDNMf9BPhJ1ANMcNubTpCWYqytLGT/8ZkliQqy0ynLy+RwxwCXzYNRnCIikhh2NvfQNzTGpSvLTtvurPI80lKMvcd7z7gyqYjMfxqpJAvWiw1dHOsZmjdFAksmRiqprpKIRMgrTSc4e0k+WempYR1XV5pDQ+cAAdVVEhGRGfp9fQcAl5xVetp2GWkp1JXlckgrjYokBSWVZMHatL2F7PRUzlkS/6lvEOxgKwq0ApyIREYg4Ow42sN5NeEnzpeV5TI0GqC1dygKkYmISDJ6oaGbsxfnU5qXeca2y8tyae0dpn94LAaRiUg0KakkC9LoeIDNO49x1ZoKMtLmz3+D2tJcjiipJCIRcLC9n/7hMc6rKQ772LqyXAAaOjRyUkREziwQcF450s2bl86szzmrPDjtTaOVRBJfVO+mzewaM9tvZvVmdvs0+zPN7OHQ/i1mVjdl/1Iz6zezL0QzTll4fnegne7BUa49xXKn8VJbEpxyIiIyVxNFus+rKQz72OKcDIqy0zVyUkREZuRQRz+9Q2O8eenMRsdWFmWTmZbCoXb97hVJdFFLKplZKnAv8G5gDfBhM1szpdltQLe7rwDuAe6esv9rwKPRilEWrke2HaUkN4PLV82vIrR1Zbm09Q0zOKKhwCIyN3taesnJSGV52eyKoNaV5dLQOYCrrpKIiJzBS0dCq43OcKRSaopRW5rDYT1MFUl40RypdBFQ7+6H3H0EeAi4bkqb64D7Q68fAa600PqTZvZ+4DCwO4oxygLUPTDC43vaeP95VfNq6htAbWkOEFwBTkRkLvYe6+XsxfmkpJx6WefTqSnOpm9ojOOqqyQiImfw8pFuCrLSWB6aPj0TS0tyae8b5uTIeBQjE5Foi+YddRXQNOn90dC2adu4+xjQA5SaWR7wV8CXohifLFCbtrcwMh7g+guq4x3KG9SWBDtiJZVEZC7cPZhUmsNCBNXFwST39tA0OhERkVN5pamH85YWh/Ugo6YkG4CjJ/S7VySRza9hGv/pTuAedz9t5TYz+5SZbTWzre3t7bGJTBLej7c1sbaygDWV82PVt8mWvjZSSUOBRWT2jvUM0Ts0xjmL82d9jiWFWaSa8UpTTwQjExGRZDM8Ns6B1j7WhfnbuqY4BwOaupRUEklk0UwqNQM1k95Xh7ZN28bM0oBCoBPYCHzFzBqAPwf+2sw+O/UC7v5td9/g7hvKy+dXbRyZn/Ye62VXc++8HKUEUJidTkluhorjisic7DveCzCnkUppqSksKcrSSCURETmtA639jAU87Ae2WemplOdn0tR1MkqRiUgspEXx3C8CK81sGcHk0U3AzVPabAJuBZ4Hrgee9GBF0MsmGpjZnUC/u38jirHKAvHItqOkpxrXnTd1Jub8sbQkhyNdGqkkIrO391gfAKvnMFIJoKoom53NPQQCPuvaTCIiktx2twRHtK6tDH+10ZqSHPa09OLuhErrikiCidpIpVCNpM8CjwF7gR+5+24zu8vMrg01+y7BGkr1wOeB26MVj8joeIB/f7mZK8+uoCQ3I97hnFJdaQ4NHRqpJCKzt/dYL9XF2RRkpc/pPDXFOfQPj3Go47Sz0UVEZAHb3dJLXmYatSU5YR+7tDiHk6PjdA6MRCEyEYmFaI5Uwt03A5unbLtj0ush4IYznOPOqAQnC86ju47TOTDCjRfWnLlxHNWW5vLz7S0Mj42TmZYa73BEJAHtO97H2YvnXjeuujhYRPWVph5WLJrbqCcREUlOu1t6OWfJ7FYbrQklopq6BinLy4x0aCISA/O1ULdIxN3/XAO1pTlcvmp+19+qLc3BHc0vF5FZGRkLcLhjgNWL8+Z8rrL8TPIy01RXSUREphUIBFcbXTPLGn6LCjLJSEvhiIp1iyQsJZVkQdjV3MO2xm4+enHtvK8LUluaC6C6SiIyK42dA4wHnBWL5p5USjFjfXUh248qqSQiIm/U0nOSwZFxVs2yhl+KGdXF2TR1K6kkkqiUVJIF4V+fayA7PZUbNszvqW8QrKkEqK6SiMxKfVuw/tGK8shMV1tfXcTeY70Mj41H5HwiIpI8DoT6nJVzmCK9tDiH4z1DjIwFIhWWiMTQjJJKZvZTM3uPmSkJJQmns3+YTdtb+C/nV1GYPbeitbFQkptBXmYajZ0aqSSJQ/3E/DGRVDprUW5EzndeTSGj4/7ainIiIrOhfiI51bdOJJVmPzq2piSHgMOxHpV+EElEM/1S/yfgZuCAmX3ZzFZHMSaRiHroxSZGxgJ8/JK6eIcyI2ZGbWkOjZpbLolF/cQ8Ud/eT1VRNjkZkVmL4001RQCqqyQic6V+IgkdaOujLC+T4jmsrFxVFFwU4mi3kkoiiWhGSSV3f9zdbwHOBxqAx83sOTP7hJnN/6EfsmCNjAV44A+NvHVFKSsrEmflorrSXBo7lVSSxKF+Yv6ob+vnrAjUU5qwuCCLRfmZSiqJyJyon0hOB9r65zRKCaAgO52CrDSaTyipJJKIZjz81MxKgY8DnwReBv6RYKfwm6hEJhIBP33pKMd6hvjkZcvjHUpYaktzaOoaZGxcc8slcaifiL9AwDnUPsBZ5ZGZ+gbB0ZPrq4tUrFtE5kz9RHJxd+pb+1lZMfcHGVVF2RqpJJKgZjQ23sx+BqwGfgC8z92PhXY9bGZboxWcyFyMjge496l61lcXcsWq8niHE5ba0hzGAs6xniFqSnLiHY7IGamfmB9aek5ycnQ8Iiu/TbauqoAn9rUyMDxGbmZkptWJyMKifiL5tPYO0zc8NueRSgBVxdnsPd7H0KgWhRBJNDP9Zfgv7r558gYzy3T3YXffEIW4RObspy8dpanrJH/z3rWYWbzDCUttaXCUQUPngJJKkijUT8wD/7nyW2STSudWFeIOe4/1sqGuJKLnFpEFQ/1EEnhwy5HXXh9oCy7g0Ng5+Lrts1FdHPy926IpcCIJZ6bT3/52mm3PRzIQWdj6h8foHhhhLBCZ6V4Dw2P8w69f5c1Li7jynEUROWcs1b2WVFJdJUkY6ifmgcMdwVUjl0c4qbSuqhCAXc09ET2viCwo6ieSTFvvMACLCrLmfK5KFesWSVinHalkZouBKiDbzN4MTAz3KAA0fELm5OTIOM8d6uDFw130Do0BkGrG6sX5XLKilOVls78p+tbTB2nrG+abH70g4UYpASzKzyQzLYXG0A2iyHylfmJ+aewcJC8zjbK82a/CM51F+ZmU5WWyq6U3oucVkeSnfiJ5tfUNkZORSm5G6pzPlZeZRlFOuop1iySgM01/exfBYnrVwNcmbe8D/jpKMckCUN/WzyPbmugdGmN1RT6XrcwjMy2F471DbD/aw57f9bKusoBrz6sK+9z7j/fxzacPce2bKjl/aXEUoo++lBSjtjSHxi6NVJJ5T/3EPNLYOcDSkpyIJ9PNjHVVBRqpJCKzoX4iSbX1DrMoPytifU51UbaSSiIJ6LRJJXe/H7jfzD7o7j+JUUyS5LY2dPGzl5spy8vkM1fUvjaHesK71i7mdwfaeWp/O41PHODsxfm8bYaFtkfHA3zhx9vJz0rjb963Jhrhx0xtaS6NnRqpJPOb+on5pbFzkLOX5Efl3OsqC/ndgQ6GRsfJSp/7U2kRWRjUTyQnd6etb5hzqwsjds6q4hx2tfTSPTBCcW5kR9yKSPScafrbR9z9AaDOzD4/db+7f22aw0RO6Tu/O8RPX25mxaI8btm4lMy0N96YpKem8I6zKzhnSQEPv9jEx773Ap+8dBl/ec3qadtPcHfu+Pkudjb38M+3nE9pXmY0P0rU1Zbk8Myr7QQCTkpK4k3hk4VB/cT8MTYeoKl7kHetWxyV86+rKmA84Ow73sd5NUVRuYaIJB/1E8mpf3iMk6PjLMqP3O/tqlBdpZ3NPTN+oCwi8XemQt25ob/zgPxp/ojMiLvztV/v529/uZe1lQV87OLa0yaIAJYUZvMnb1/Bx95Sy3eePcwH7n2O+tAqE9Od/57fvMoPX2jiT95+Fu8+d0k0PkZM1ZblMjwWoK1vON6hiJyO+ol54ljPEKPjTl1pdEqUqFi3iMyS+okkNPH7dFH+3It0T5icVBKRxHGm6W/fCv39pdiEI8koEHD+9y/3cN/vG7jhgmrWVxeROsORN+mpKdx13TretrKcL/5kB+/9/5/lz69axc0bl1KQlQ5AU9cgf7d5L4/uOs4NF1TzF+9cHc2PEzMTN4YNnQMsLoxchy0SSeon5o+G0HTZ2tLcM7ScnaqibIpy0tndoh/7IjJz6ieSU0d/MKlUHsGRStkZqZTmZrC96UTEziki0XemkUoAmNlXzKzAzNLN7Akzazezj0Q7OEl84wHn9p/u4L7fN/CJt9Zx9wfXzzihNNlVayr41Z9dxluWl/LlR/ex4W8f5z1f/x1X3/M0V3z1KZ7Y28Z/f/fZfOX69UkzVay2JHhjqLpKkgjUT8RfY2ewsH9dlJJKZsa6ykJ2NWsFOBEJn/qJ5NLZP0J6qpGfdaZ1n8JTVZytkUoiCWam3wJXu/sXzewDQAPwX4BngAeiFZgkvu6BEf7yke08vreNP71yJZ+7auWcVodYVJDFfZ+4iF3NPfz8lWb2t/aTmZbCNWsXc8vFtVQUJNdonsqiLNJS7LUbRZF5Tv1EnDV2DpCZlhLR+hZTra0q4L5nGxgZC5CRNqPnUiIiE9RPJJHO/mFKcjNIifBqo9XFOew42kNb31BEp9aF46Uj3TzwfCNbDnfRNTBCVnoKKxblc8lZpUl3vyESCTNNKk20ew/wY3fvifRyxZIYBofH2Hakm+7BEVLNqCzK5sK6YpaX5702AmlwZIyfvdzM1584QNfACF+6di23XlIXsRjWVRW+VtsjmaWlplBTkqOkkiQK9RNx1tA5SG1pTlRHa66rLGRkPMCrrX0L4ntYRCJK/UQS6RwYoSwKi+K8VlfpaA9XnhPbBM7gyBh3/Hw3j2w7Sn5mGpevLmdxQRbPHexke9MJtjZ0cenKMq5es3hWMy9EktVMk0q/MLN9wEngv5pZOTAUvbBkPnqpsZtN21sYGQ+QnZ7K6HiAsYDz421HyUhLoaY42Ak0dg4yFnDevLSI73zswoguNbrQ1JbmvFYnRWSem1U/YWbXAP8IpALfcfcvT9mfCXwfuADoBG5094ZJ+5cCe4A73f2rEfosCamxcyBq9ZQmTCSSdrf0KKkkIuHS/USSCLjTNTDC6sWRr7NeWZRFisGOoz1ceU5FxM9/Kh39w9z6vRfYe6yXz1xxFp99xwpyMoK3yg9uOcLg8BiP7TnO7w50cLxniI9cXEt6qkbsisAMk0rufruZfQXocfdxMxsArotuaDKfbG86wU9eOkpdWS7XvqmSioIsxgNOR/8wS0ty2N/aR1NXcETNNesWc9nKcjYuK5nTdDeB2pIctjZ04+76t5R5bTb9hJmlAvcC7wSOAi+a2SZ33zOp2W1At7uvMLObgLuBGyft/xrwaCQ/SyIKBJwjXYNcHuUlmGtLcsjPTGNXcy83XhjVS4lIktH9RPLoPTnKWMApzY38SKXMtFRWLMqLaV2lnpOjfOy7L3Coo5/v3nohbz970Rva5GSm8YE3V1NdlMO/v9LMg1uO8JGLazViSYSZj1QCOBuoM7PJx3w/wvHIPNQ9MMJPXjpKbWkut76l7rU6GqkpRkVBFh+8oDrOESav2tJc+ofH6BoYoTQKQ4xFIizcfuIioN7dDwGY2UMEbzAmJ5WuA+4MvX4E+IaZmbu7mb0fOAws+OF8bX3DDI0Goj5SKSXFWFNZwC6tACcis6P7iSTQOTACQGleRlTOf25VEU+/2h6Th6qBgPP5h1/h1dY+vnPrBq5Y/caE0mQXLishgPPzV1r4zZ7jXLNuSVTjE0kEM0oqmdkPgLOAV4Dx0GZHncCC8OiuY5jBjRfWqDBrjNWV5QDBWilKKsl8Nst+ogpomvT+KLDxVG3cfczMeoBSMxsC/orgKKcvnCauTwGfAli6dOlMP07CmZgmW1uaE/Vrrasq5N+2NDI2HiBNQ/9FZIZ0P5E8OvtDSaXc6CSV1lcX8pOXjnKsZ4jKUI2laPnWM4d4Yl8bd1239owJpQkbl5VyrGeIZw50sLw8j1UVkZ8GKJJIZjpSaQOwxt09msHI/NPYOcCull6uOqeCwuz0eIez4CwtCY46aOwc4ILa4jhHI3Jase4n7gTucff+0z3FdPdvA98G2LBhQ9L2YY2hpFJdlEcqAayrKmBoNMChjgH9kBaRcOh+Ikl0DgyTlmIUROneYH2oHuuOoz1RTSrtP97H136znz86dzEfvbg2rGPfc+4SDncM8LOXm/mzK1eSlZ4apShF5r+ZPmLcBSyOZiAyPz13sJPs9FQuXVEW71AWpJqSbMzQCnCSCGbTTzQDNZPeV4e2TdsmNF2ikGDB7o3AV8ysAfhz4K/N7LPhh50cGjoHSU81lhRGf6WcdZXBH/s7j2oKnIiERfcTSaKzf4SS3AxSojQ17ZwlBaSlGDubT0Tl/BCc9nb7T3eQn5XO/75uXdjT7NJTU7j+/Gp6T47yxN7WKEUpkhhmOlKpDNhjZi8AwxMb3f3aqEQl80L/8Bh7WnrZuLwkoaa9PbjlSLxDiJjMtFQqC7NfG4UgMo/Npp94EVhpZssIJo9uAm6e0mYTcCvwPHA98GToKfdlEw3M7E6g392/EYHPkZCOdA5SU5wTk+loy8vzyE5PZVdLj2rqiUg4dD+RJDoHhqM29Q0gKz2VVRX57Ijiw4ufvtzMy0dO8P9ev37WJSZqSnK4oLaY5w91snFZKWX5KlUhC9NMk0p3RjMImZ9eauxm3J2L6kriHcqCVluaQ4NGKsn8d2e4B4RqJH0WeAxIBb7n7rvN7C5gq7tvAr4L/MDM6oEugoknmaKhcyAm9ZQguEjDmsoCdjf3xuR688HgyBjHe4bAYFF+FnmZ4axzIiIhd8Y7AJm7gDtdAyOsXBTd6c9vqilk887jUSnWPTQ6zld+tY/zaor44PlzezjyzjUV7Gzu4bE9x7llY3hT6ESSxYx+Fbn702ZWC6x098fNLIfgDYAksZebuqktyWFRQfSnU8ip1Zbm8uiuYzFZAUNktmbbT7j7ZmDzlG13THo9BNxwhnPcOaugE9B0IzHdnfq2fgqy02M2UnNdZQGPbDtKIOCkJPFyyu19wzy+t5VdzT1MFIExYE1lAe88p0L9o0gYdD+RHPqGxhgd96it/Dbh3KoifvhCE01dJ1ka4YcmD/yhkba+Yb5x8/lz7sPys9J564oyntzXRsuJk1EvLC4yH81onLyZ/d8El3L+VmhTFfDv0QpK4q9rYITW3mHWVhXGO5QFb8WiPE4Mjr62fKvIfKR+In4GRsYZHgtEdSrCVGurChkYGX9t1blktOPoCb7x2wO82trHZSvL+PgldXzikjretqqcg+393PtUPS8f6Y53mCIJQ/1EcugcCM5cLM2N7lSviWLd249Gtq7S4MgY33z6IG9dUcpFyyIzG+OtZ5WRlZ7Cb/e3ReR8IolmpsUX/gR4K9AL4O4HgJmtuSgJae+x4LSGcxZrZZ94W7EoD4D6tv44RyJyWuon4qSrP/gDvySGSaWJYt27WpJzCtzWhi4eerGJJYXZfO6dq7hm3RJWVeSzsiKfd61dzOeuWkV1cQ4/3naULYc74x2uSKJQP5EEOvuDDzmj/SBjVUU+GWkp7GyObF2lB/7QSEf/CJ+7alXEzpmdkcrFy0rZ09JLlx4CywI006TSsLu/9j8ktAKPlgNNYvuO91KenznrwnUSOStDSaUDSirJ/KZ+Ik66BkcBKMmJXVJpZUUeGakp7Irwj/35YE9LLz97uZmVi/L45KXLKMh645LZ+VnpfOKtdayuyOfnr7Twix0tcYhUJOGon0gCnf0jpJpRmPPG78ZIykhL4ZwlBeyI4EilwZExvvX0IS5bWcaGCNeMvXh5KWbw/MGOiJ5XJBHMNKn0tJn9NZBtZu8Efgz8R/TCknjqHRrlcMeARinNE0sKs8jNSOWgkkoyv6mfiJMTg8F7tKIYJpXSU1M4e0l+0iWVOvuH+fG2JqqKs7llY+1pV9NLS0nh5o1LWVqSwxcf2cGB1r4YRiqSkNRPJIHOgWGKczNIiUGdz/VVhexq7iUQiEzu8eEXm+gcGOHPr1oZkfNNVpCdzvrqIrY2dtM3NBrx84vMZzNNKt0OtAM7gT8mWFT1f0YrKImvPxzsJOCwSkmlecHMWLEojwNtumGReU39RJx0D46Ql5lGRtpMu/TIWFdVGCxg7ckx0GBkLMAPXzxCihkfvmjpjP4901NT+PBFS8nJSOUz//YSQ6PjMYhUJGGpn0gCXQMjlEW5SPeE82qK6B8e49UI/AYeDzj3/b6BC2qLuaA2OitbX3JWKcNjAX689WhUzi8yX83oF6i7BwgW0vuMu1/v7v/iyfIrUt7ghcNdpKUYNcWxWZ5azuysRXmqqSTzmvqJ+OkeHKU4ytMQprOuspDeoTGOdp+M+bWj4dvPHKTlxBAfPL+K4jBGfRVmp/MPHzqPA239/OMTB6IYoUhiUz+R+Nydzv6RmC0MMVFI+4XDXXM+12/2tHKka5DbLl0253OdSnVxDrUlOdz33GHGIzS6SiQRnDapZEF3mlkHsB/Yb2btZnbH6Y6TxLblcBc1JTmkn2bYv8TWykX5tPYO06vhtDLPqJ+Iv+6BkZhOfZuwrqoAICmmwB1s7+frT9RzblUhayrDX/X08lXl3Lihhm89fZCdRxP/30MkktRPJI/2vmFGxgOUxKjmanVxNpWFWWyJQFLpe88epro4m6vXVEQgslO7ZEUZTV0nebZetZVk4ThT1uBzBFdpuNDdS9y9BNgIvNXMPhf16CTmeodG2d3SQ11pbrxDkUm0ApzMY+on4ijgzomTo2GNrImUVRX5pKVYxFfmiYe7/mMPmekpvHf9klmf43+89xxKcjP40n/sTpopgSIRon4iSTR0DgLRX/ltgplx0bISXjjcNafv1Z1He3ihoYuPX1J32lp5kXDO4nyKc9L50damqF5HZD450/+qjwIfdvfDExvc/RDwEeBj0QxM4mNbYzcBh2VlSirNJyuVVJL5S/1EHPUPjTEecIpzYz/9LSs9lZUV+exq6Y35tSPpqf1tPP1qO3925Uryp1npbaYKstL5wtWr2drYzS93HotghCIJT/1EkmjoHACgLIarQ1+0rJT2vmEOdwzM+hzfffYQeZlp3HhhTQQjm15aagrvf3MVv9ndSvfAyJkPEEkCaWfYn+7ubxi75+7tZhb7X7ASdS8c7iI91VhaonpK80lNSQ4ZaSlKKsl8pH4ijrpDK7/FY6QSwLrKAp7c14a7YzFYCSjSxgPO323eS11pDh97Sx2PbJtbcdUbNtRw//ON/P3mfVx1TgVZ6akRilQkoamfSBINHQOkWLCWXKxcvDxYV+n39R0sL88L+/jjPUP8YscxPvaWutceHPwf9u47TM6zPPv/956Z7b331e5qV2VVLMmyJBfcKwbLYBtsiOkhCSG8Lw784uQNDiFAIJAQCE5oDjiAMca44YKMe1OxbNVV26rtve9sn/v3x846QpZtlZ15ppyf49DB7DPP7Jwr5J2Z67nv67pne/OCZjzeTWeX8NOXm3h4dxsfOz9wPZxEQsU7rVR6u/LqO5ZejTFXG2MOG2PqjDG3n+D+OGPMr/33bzfGlPmPbzDG7Pb/2WOMed87PZcsjF3NA1QXpgV9ipC8PbfLUJGdpKKShKIzep2QMzNfVEp3oFE3zE2A6xubonN4wpHnP1OP7evgSNcoX7hq6YK87rldhi+9Zzltg+P85MWGBUgoEhH0OhEhjvZ5yUiMxe0K3kWE8uwkijMSeKH29HoU3b21CZ+1fPz8sgXN9XaqC1NZVZTGfZoCJ1Hind5BnWWMGT7BnxFg1ds90BjjBu4ErgGqgVuMMdXHnfZJYMBaWwl8B/im//h+YL21dg1wNfBDY8w7raqSMzTrs+xrHWJN8ak3KZXAq8xNpnYBRqqKLLDTfp2QMzfgnWve79hKpaK514v9beG3Bc7ns/zH07VU5Sbz7pWn30vpeOctzuaqFXn853P1dIdpsU1kgZ3R68TpXqQ+5v5SY8yoMeYLC/YTRammvjGykoP7emOM4cIlOWyt72N61ndKjx2bnOGe7c1ctSKfkiDvwrhpfTEHOoYjYpiFyDt526KStdZtrU09wZ8Ua+07XRbdANRZaxustVPAvcDm487ZDNztv30/cJkxxlhrvdbaGf/xeEAdL4OgoWeUsalZVhenOx1FTqAqN4XWgXHGp2adjiLyhjN8nZAzNDA2RXKcx7FpncsLUnCZ8JwA9/j+Dmq7R/mry6pwLfBV97+9ZjlTMz6+90ztgn5fkXB0Jq8TZ3iRet6/AU8szE8Tvay1HO3zkhXEfkrzLqzKYXRyhtePDpzS436zs4Wh8Wn+9MKKACV7a5vPKiLW7TrjbdUi4SCQ70KLgGPb3rf6j53wHH8RaQjIAjDGbDTG1AD7gD8/psgkAbK7ZRCAs0q0UikUVeUlY+3c6GsREYBB7zQZDm19A0iM9bA4Jznsiko+n+V7T9eyOCeJa1ct3CqleWXZSXzwnBLu3dFCs39akoicltO+SA1gjLkeaARqgpQ3YvWNTTE6ORO0yW/HOq8yixi34elD3Sf9mJlZH3e93Mj6RRmsK80IYLoTS0uM4bLluTy6t52ZU1xhJRJuQrZxjrV2u7V2BXAO8LfGmPjjzzHGfNoYs9MYs7Onpyf4EgCvMgAAIABJREFUISPM3tYhkuM8VGSfehM8CbxK/wQ4FZVEZN6Ad4p0h7a+zVtZlMa+MCsq/b6mkyNdo3zusqqA9Qb53GVVeNyG7zx1JCDfXyRKnPZFamNMMvA3wD++3RPo88TJafJPX8tKCv5KpdT4GM6vzObxfR1Ye3IbWH5f00lL/7gjq5TmXb+2iN7RKV6sO71+UCLhIpBFpTbg2LmNxf5jJzzH3zMpDeg79gRr7UFgFFh5/BNYa39krV1vrV2fk5OzgNGj097WQVYVpS34NgBZGGVZSbhdhtouFZVEBHzWMuidJtOBq8bHWlWURvfIJB1D447mOFnWWr7/TB0VOUm8Z3VhwJ4nLzWej55XxkO72zjUGX49p0QiwJeB71hr3/aNkz5PnJwm/6rLYPdUmvfuVQW0Doyf1EUMay0/fqGB8uwkLl+eF4R0J3bJ0lzSE2N4aNfxH4FFIksgi0qvAlXGmHJjTCxwM/DIcec8AnzUf/tG4BlrrfU/xgNgjFkELAOaApg16k3N+DjQMcxqbX0LWbEeF4uyEtWsW0QAGJmYYdZaxya/zVu3aG5bwa7mQUdznKyt9X0c6Bjmzy6sCPgEo7+4aDHJsR6+vUWrlURO05lcpN4I/Isxpgn4v8DfGWM+G+jAkepo3xhul3FsMMSV1Xl4XIbH9na847lb6/vY0zrEJy4oD+qkuuPFelxcu6qALTWdjE6qk4tEroAVlfzLTz8LbAEOAvdZa2uMMV8xxlznP+0u5pan1gG3AfMTHS4A9hhjdgMPAp+x1mrdYADVdo8wPWtZVaSiUihbmpfCEa1UEhFg0Ds3idupN/jzqgtSifW4TrmBqlPueqmRrKRYNq85fgfNwktPjOXTF1bw1MEuXguTvx+REHPaF6mtte+y1pZZa8uAfwe+bq39frCCR5qmPi/FGQmOFWnSE2O5ZFku97/WyuTMWw+tsdby7ScPU5AWz01nFwcx4Ym9b20RE9M+tuzvdDqKSMAEtKeStfZxa+0Sa+1ia+3X/MfusNY+4r89Ya29yVpbaa3dYK1t8B//ubV2hbV2jbV2nbX2oUDmFDjYMbf6ZXlBqsNJ5O1UF6TS2Dumqx0iwkCIFJViPS5WFaWxqyX0Vyo19Izy9KFuPrxpEfEx7qA85ycuKCc7OZZvbTl00r1ARGTOGV6klgXU1DvGoqwkRzPcumkRfWNTPL7vrVcrPX2wm9ebB/mrS6uC9nv+7Zy9KIOSzAQe2q0tcBK5PE4HkNBwoH2YhBg3ZQ6/WMjbqy6cK/od6hhmfVmmw2lExEn9Y9MAjm9/A1hXms7dW48yNeMj1hOyM0D46ctNxLpd3LppUdCeMynOw19eUsk//u4AL9X18q4q9WwRORXW2seBx487dscxtyeAm97he3w5IOGihLWWpr4x1pamO5rjgspsyrOT+MmLjWw+q+iNPrD3bG8GYHrWx3efriUnOY5Zn33juJOMMbxvTRHff7aOruEJ8lLfNHtKJOyF7js/CaqDHcMszU9xdN+xvLP5otKBDjV9FYl2g94pUuI8xLidfylfW5rxRm++UDXoneL+11q5bk0hOSnBnV70oY2lFKUn8K0th7VaSUTCzoB3mpGJGcdXKrlchr+6tJKa9mEeOEHz66cOdNE/NsV1awpD6jPN5rVF+Cz8bk+701FEAsL5d6LiOGstBzqG3yhYSOjKT40nMymWA+2h+8FNRIJjwDsVEquUANaVzjXrDuW+Sr/a0cL49CyfOL886M8d53Hzfy+vYm/rEL9XXw0RCTNNfWMAlGcnOpwErl9TxNrSdL7xxEGa/RPpAPa0DPJiXS8byjNZnJPsYMI3W5yTzFnFaTzwurbASWRSUUloH5pgaHxa/ZTCgDGG6oLUkF4NICLBMeCdJiPJ2X5K8/LT4ilMi+f15tAsKs36LL/YdpRNFZmOXUB5/7piKnOT+faTh5mZ9TmSQUTkdDT1zhWVnF6pBHOrlb5142pmfJZbfryNn287yqN727lvZwuLshJ5z+oCpyOe0PVrizjQMczhTk1xlsijopJw0L/qpVpFpbBQXZjKoc4RpvWhRCRq+axlyDvteJPuY60tzWBXc2g2637+SDdtg+PcuqnMsQxul+ELVy6hvmfshNs2RERCVVOfF5eB4owEp6MAUJmbws8/sZH4GBdfemg/r9T3sW5RBh8/rxyPKzQ/3r73rLkteWrYLZEoNP+rk6A61DlXVFqan+JwEjkZ1QWpTM34aOgZczqKiDhkZGKGWWtDZvsbwNrSdNoGx+kennA6ypv8clsz2clxXFGd52iOq1bks7o4je8+Vfu2I7FFRELJ0b4xCtMTiPM4P01t3qriNJ667SL+8PkL+dK11dywrjikB0VkJ8dxYVU2D+9qw+dTbz2JLKH7X54ETW33KEXpCSTHaRhgOPjfZt1DDicREacMjE0BhNRKpXWL/H2VQmwLXOuAl2cOd3PzOSWOf+AwxvDFq5bSNjjOL7c5P5VIRORkNPV5Q3JCtDGGqrwUEmJDp9j1dq5fW0T70ATbG/udjiKyoFRUEmq7RqnKC62GdvLWKrKTiPO41KxbJIoNeOeKSpkhVFRaUZhKrNsVclvg7t3RAsDNG0ocTjLngspszq3I4s5n6xibnHE6jojIO2rqHaMsBJp0h7srq/NJinXzkLZAS4RRUSnKzfos9T2jVOWqqBQuPG4Xy/JTqFFRSSRqDXinAUgLoe1vcR43K4pSQ2ql0vSsj3tfbeGSpbkUZ4TGByJjDF+8eil9Y1P890uNTscREXlbg94phsanQ3KlUrhJiHVz1cp8Ht/XwcS0tkBL5FBRKcq19HuZnPFRlad+SuGkunBuApy12pMtEo0GvFOkxHuIcYfWy/j6RRnsaR0KmTfLfzjQRe/oJH+yqdTpKH9kXWkGV1bn8YPn6+kKwR5UIiLzmvq8QGhMfosE719bzMjkDM8c6nY6isiCUROdKFfbPQoQ0iuV7tmuvhPHqy5I5Vc7WugYmqAwPTQmcYhI8Ax4p0Kqn9K8DeVZ/PjFRna3DLKpIsvpOPxi21GK0hO4aEmu01He5O/evZwrv/MCX3/8IN+9ea3TcURETuho39xgmLKs0FjtGe7OXZxFbkocD7zexrtXFTgdR2RBqKgU5Wq7RwCoDOGikrzZG82624dVVBKJQoPe6ZAZ7XysDWWZGAPbG/odLyrV94zySn0fX7xqKW6XcTTLiZRlJ/HnF1XwvWfquGVD6Sn/fZ3JBZcPbQytlVsiEroae8cwBkoyVVRaCG6XYfOaQn76chP9Y1NkJoXeBSKRU6WiUpSr6xqlIC2elPjQ6csh72xZfirGwIGOYS53eES2yJkwxlwNfBdwAz+x1n7juPvjgP8Bzgb6gA9aa5uMMRuAH82fBnzZWvtg8JI7x2ctg94pVhWlBeX5TrV4kZ8az8N72shJiXOkeDGf9/F9HbgMeFwmZFe8/sXFlfz29TbueHg/j33uXSG3nVFE5Gifl8K0BOJjwmPCWji4fm0RP36xkcf2tnPruWVOxxE5Y3r3EuVqu0e1SikMJcV5qMhOYm9raE1ZEjkVxhg3cCdwDVAN3GKMqT7utE8CA9baSuA7wDf9x/cD6621a4CrgR8aY6LiQsnw+DQ+S0hufwMoz06ipd/LjM/nWIbpWR+vHR2gujAtpC+aJMS6ueO91RzpGuXuV5qcjiMi8iZNfWMs0ta3BVVdkMqy/BR+taNF/VElIqioFMV8Pktd9yhVuWrSHY7Wlmawq3lQL0YSzjYAddbaBmvtFHAvsPm4czYDd/tv3w9cZowx1lqvtXZ+Hns8EDX/IcxPfssIoclvxyrPTmJ61tLaP+5Yhn1tQ4xPz7KxPNOxDCfryuo8Llmaw7/94Qgt/V6n44iI/JGjfV416V5gxhhuPXcRBzqGQ2piqsjpUlEpirUNjjM+PUtVnlYqhaO1pen0jU3RrA8hEr6KgJZjvm71HzvhOf4i0hCQBWCM2WiMqQH2AX9+TJHpDcaYTxtjdhpjdvb09ATgRwi+Qe8UENorlQxzPY2csqOxn+zkWCqyQ/+DkDGGr75vFS5j+Jvf7tWFAhEJGUPj0/SPTVGerZVKC+36NUWkxHn4n61HnY4icsZUVIpidWEw+U3e2rrSDAB2NWsLnEQna+12a+0K4Bzgb40x8Sc450fW2vXW2vU5OTnBDxkAA/6iUlqIrlRKjPVQmJ7wxmtMsHUMjdPc72VDeRbGhF6D7hMpSk/gb9+9jFfq+7j31ZZ3foCISBDMT37TSqWFlxTn4Yazi3l8Xwc9I5NOxxE5IyoqRTFNfgtvS/JSSIp1a9mshLM2oOSYr4v9x054jr9nUhpzDbvfYK09CIwCKwOWNIQMeKdJifeEdFPnytxkWga8jExMB/25tzf243EZ1pWmB/25z8Qt55RybkUWX3vsIO2Dzm0dFBGZ19Q3txq+TEWlgLj13EVMz1p+/WpoDpMQOVmh+45UAq62a5SclDjSQ3QLhbw9t8twVkm6VipJOHsVqDLGlBtjYoGbgUeOO+cR4KP+2zcCz1hrrf8xHgBjzCJgGdAUnNjOGvBOhezWt3mVucn4LGxr6A/q845OzrC7ZZDVxWkkxoZX33aXy/DNG1Yz67Pc/sA+bYMTEcc19c6tVCrN1Pa3QFick8wFldn8YlszUzPODbcQOVMqKkWx2u5RbX0Lc2tL0znYMcz41KzTUUROmb8H0meBLcBB4D5rbY0x5ivGmOv8p90FZBlj6oDbgNv9xy8A9hhjdgMPAp+x1vYG9ydwxsDYVMg26Z63KDORGLfhpdrg9rF6aFcbUzM+NpRnBfV5F0ppViJ/++5lvHCkh3t26Mq1iDirqXeMwrR4EmLdTkeJWJ96VzmdwxM88Hqr01FETpuKSlHK2vnJbyoqhbN1pRnM+Cz72oacjiJyWqy1j1trl1hrF1trv+Y/doe19hH/7Qlr7U3W2kpr7QZrbYP/+M+ttSustWusteustQ85+XMEy6zPMjQ+HfIrlTxuF+XZSTx/JHhFJWstv9zeTEFaPCUZCUF73oX2JxsXcUFlNl977OAb/UxERJzQ0DtGWRgMPAhnFy3J4aziNO58ro7pWa1WkvCkolKU6hiaYHRyhsq8FKejyBlY62/Wrb5KItFhZGIanw3dyW/HWpqfSlOfN2hT4F47OsDBjmE2hlGD7hNxuQz/cuNq3C7DF36zh1mftsGJiDMae8coV1EpoIwxfO6yKlr6x3l4d7vTcUROi4pKUapWk98iQmZSLGVZiexSUUkkKgx45xpfpyeF9vY3gGX5cxctnjnYHZTnu3vrUVLiPawpCa8G3SdSmJ7Al9+7glebBvjvlxqdjiMiUWhgbIqh8WkVlYLg0mW5rChM5c5n65jRaiUJQ+HVxVIWTG3X3OQ3FZVCyz3bT62Hxoc2lrKuNIMX63qx1ob11XkReWcD3ikgPFYqZSTGsiw/hacPdfGnF1YE9Lm6hyd4Yl8HHz2vjFhPZFwve/+6In5f08m3njzMRUtzWKKVxSISRA3+Jt0qKgWeMYa/urSKP//FazzwehsfOKfknR8kEkJUVIpSdd2jZCXFkpUc53QUOUNrS9N5YFcbbYPjFGdoOodIJJsvKqUnhP5KJYDLlufyg+cbGPJOkxbA5uK/2tHCjM/yJ5sWsbW+L2DPE0zGGP75/au48jsvcNt9u3nwM+cT446MgpmIhL4mFZVO2+lcJL5qRR7rStP5ly2HuXpVPqnx4fE6LwLa/ha1artHqdQqpYjwv32VBh1OIiKBNjg2TWq8B0+YFBcuX57HrM/y1MGugD3H9KyPX24/ykVLciLuw092chxff99K9rcN8/1n6pyOIyJRpLF3DLfLUJKpC5bBYIzhy9etoG9skv94utbpOCKnJDzelcqCstZS2zVCVZ6KSpFgWX4KyXEedjRGxtV5EXlrA94p0sNg69u8NSXpFKUn8Ni+joA9x5aaTrpHJvnoeYsC9hxOunplAe9bW8T3n61jX6smfYpIcDT2jlGSkaAVkkG0ujidD5xdwk9fbqKuOzhDLkQWgn5LRKGekUmGJ2aoylV/hkjgcbs4pywjYrZ8iMhbG/BOkRHAbWQLzRjDtasLeLG2hyF/k/GF9rOXmyjNTOSiJbkB+f6h4MvXrSAzKZb/99A+TYMTkaDQ5DdnfPHqpSTEuPnyIzVYq9/3Eh5UVIpCmvwWec5dnEV9zxjdwxNORxGRAJmZ9TE0Ph0WTbqPde2qAqZnLU8e6Fzw772zqZ+dRwf42HlluF2RO6ggLSGGv792OXtbh/jVjlPr1SEicqqstTT2jlGmolLQZSfH8cWrl/JSXS8PvN7mdByRk6KiUhSan/xWqe1vEePcimwAtjZotZJIpOoYmsBnISMpvIpKq4vTKMlM4OHd7Qv+vX/wfD0ZiTHcvCHyJ+Vcd1Yh51Zk8a0th+kdnXQ6johEsK7hScanZ6lQUckRf7JxEetK0/mnxw7o972EBU1/i0K13aOkJcSQs0CT3051woEsvOrCVFLjPWyt72PzmiKn44hIALQMeAHCbqWSMYYb15Xw708foaXfu2BNXw93jvDUwW4+f/kSEmMj/+2MMYZ/un4FV//7i3zjiUOs8w9pEBFZaI3+yW9aqeQMl8vwzRtW8+7vvcg/PXqA79681ulIIm9LK5WiUG33KFW5yRgTuVsFoo3bZdhYkaWVSiIRrLV/HIDMMFupBHDD2XPF7vtfa12w7/nD5+tJjHXzkXMjs0H3iVTmpvCpd1Vw/2utNPd7nY4jIhFqvqiknkrOqcpL4TMXV/Lw7naePdztdByRtxX5l/bkTeq6R7lqRZ7TMWSBnVuRxR8OdNE2OE5ReoLTcURkgbUMeDHM9dcJN8UZiVxQmc39r7XyucuqTqr/0dutgh3wTvHQ7jbOrcjiif0L36splP3VpZXc/1oLW2o6+dQF5bpAJCILrqlvjFiPi8I0vZ900mcuWcxj+zr4+wf38+TnLyQpTh/dJTTpX2aU6RudpH9sikpNfos451VmAbC1vo8bzy52OI2ILLSWfi9piTFh25D65nNK+ct7Xufpg11cuSL/jL7Xi7U9AJxfmb0Q0cJKUpyHv7ykkn/83QHqe8ao1NANEVlgDT1jlGUl4grT15tIEedx880bVnHjD7by7ScP8w/vXQGceuuRD20sDUQ8kTdo+1uUOdKlyW+RakluCplJsWyt1xY4kUjUMjAedv2UjnXVijyK0hP4yYuNZ/R9+semeLVxgLMXZZAexn8fZ+JDG0tJS4jhyQOdGjktIguusXdUW99CxNmLMrl10yJ+9koTu5oHnI4jckIqKkWZuu65yW9VmvwWcVwuw6aKTLbW9+pDhkgEaun3khnGRRSP28XHzy9jR1M/e1oGT/v7/OFAJy4XXLoserdxx3ncXLYsl9aBcQ52DDsdR0QiyKzP0tzvVZPuEPLFq5aSnxrP7b/dx9SMz+k4Im+iolKUqe0eJTnOQ35qvNNRJADOXZxN+9CEGriKRJiJ6Vm6RybJSAq/fkrHunlDKanxHr77dO1pPb6l38ue1iHOW5wdlr2lFtLa0gyyk2N58kAXPl1IEJEF0jYwzvSspUJFpZCREh/DV69fyeGuEX78YoPTcUTeRD2Vokxt1yiVmvwWsc5fPNdX6YXaXm7N0psBkUjROjA3+S2ct78BJMd5+IuLK/nm7w+xvaGPjRVZJ/3YWZ/l4d1tpMZ7uGhJTgBTnrpT7W9xrNPtdeF2GS5fnse9r7ZwoH2YlUVpp51BRGReY9/85Dftagglly3P44rqPH7wXD3/5/IqEmP1MV5CR0BXKhljrjbGHDbG1Bljbj/B/XHGmF/7799ujCnzH7/CGPOaMWaf/38vDWTOaFLbPap+ShGsPDuJ0sxEnjuk0aMikaRlYG71YWZSeBeVAD5+fhn5qfF8/fGDzPpOfoXN1oY+2ocmuHZ1IfEx7gAmDB8ri9LISIzhpbpep6OISIRo7Jnrv1qWnehwEjneX1+5hNGpGV6q1e98CS0BKyoZY9zAncA1QDVwizGm+rjTPgkMWGsrge8A3/Qf7wXea61dBXwU+HmgckaTQe8UvaOTLMnT5LdIZYzhkqU5vFzfy8T0rNNxRGSBtPq3tIb7SiWA+Bg3f3PNUva0DnHXSye3jL99cJwtNZ0sy09hZWFqgBOGD5cxnF+ZTXO/V9ueRWRBNPaOkRznISc5zukocpxl+am8Z3Uhr9T3MTo543QckTcEcqXSBqDOWttgrZ0C7gU2H3fOZuBu/+37gcuMMcZau8ta2+4/XgMkGGP0m+0MzU9+0/jhyHbxslwmpn1sb+x3OoqILJCWgXFiPS6S4yNjufv1a4q4sjqPb285woH2t280PTY5w692NJMU6+aGdcXavn2cs0sziI9xabWSiCyIxj4vZdmJ+l0bov7v5VVMz/p44UiP01FE3hDId6dFQMsxX7cCG9/qHGvtjDFmCMhibqXSvBuA1621kwHMGhUOd81NfluSr5VKkezciiziPC6ePdQdcn1HROT0tPR7Kc5IwBUhb/KNMXz9/au49nsv8rGf7uC3f3EeJZlv3moxMT3Lz15pYmh8mk+cX05SXGQU1Y51Jv2YAOJi3Gwoy+TF2l76x6YiYoukiDinsXeUNSUZTseQt7A4J5m1pRlsa+jjoiU5Efm6KOEnpKe/GWNWMLcl7s/e4v5PG2N2GmN29vSoWvtOartGSI7zUJimyW+RLD7GzXmLs3jusPoqiUSK1oFxSjIiq79FdnIc//OJjUzO+LjpB1t5pf6PV9q0D47zn8/V0TE0zi0bSjXe+m2cuzgbY3jT36GIyKmYnJmlbWCc8qzIer2JNO+qymbGZ9l5dMDpKCJAYItKbUDJMV8X+4+d8BxjjAdIA/r8XxcDDwIfsdbWn+gJrLU/staut9auz8nRiox3cqRrhKo8TX6LBpcsy6Wpz0tj75jTUURkAbQMeCnJTHA6xoJbmp/CPX+6kcRYNx/68Xauv/Nl/r/79/CBH2zl+8/WMTnj4xMXlLO8QH2U3k5aQgyri9PZeXRA/fRE5LQ19o7hs7BYrTJCWl5qPBU5SWxv6DulgRcigRLI9XKvAlXGmHLmikc3Ax867pxHmGvEvRW4EXjGWmuNMenAY8Dt1tqXA5gxqhzpGuXK6jynY8gCeqttE8Pjc837vr3lMOdXZr9x/HRHV4uIc0Ymphn0TkfcSqV5KwrTePRzF3D3K0d5Yn8Hzx/pISMxliur89hQlkmilvaflE0VWexuGWRv6xAbyjOdjiMiYaiuW/1Xw8W5FVn8cnszhzuHqS5MczqORLmAvVPz90j6LLAFcAP/ba2tMcZ8BdhprX0EuAv4uTGmDuhnrvAE8FmgErjDGHOH/9iV1lrt5zlNvaOT9I9NUaXJb1EhMymWnOQ4jnSN/FFRSUTCT0v/OAAlmYkMeqcdThMYibEe/uLixfzFxYvfOHamvYaiTUlGAnmpcew82q+ikoiclrruUYyBimwVlULdsvxU0hNi2NrQp6KSOC6gl/+stY8Djx937I5jbk8AN53gcV8FvhrIbNHmSOdck+6lKipFjaX5KWxt6GNyZpY4j9vpOCJymloG5kbFl2QkMugdcjiNhCpjDOsXZfLYvg46hsYpSIu87ZIiElj1PWMUpSeQEKv3jaHO7TJsLM9ky4EuuoYnyEtVz1xxTkg36paFc2R+8luerjxEi2X5Kcz6LEe6Rp2OIiJnoKXfX1SKwJ5KsrDWlqTjdhlebVLzVhE5dXXdo9r6FkbWl2XiNobX1LBbHKaiUpQ43DVKemIMOSlxTkeRIFmUlURirJuadq1sEAlnrQPjpMR5SEuIcTqKhLjEOA8rClPZ3TLA9KzP6TgiEkZmfZaGnlEqc1RUChdJcR6W5Kewp3UQn1XDbnGOikpRorZrhCW5KZr8FkXcLkN1QSqHO0eY0YcLkbDV0u+lODNRv7/lpJxTlsnEtE8XFETklLQNjDM549NKpTCztiSdkYkZ6ru1M0Gco6JSFLDWcrhrhCptfYs6KwrTmJzxvTHNQ0TCT8uAl5IMbX2Tk1OenURmUqy2wInIKanvmXuvuFhFpbCyLD+F+BgXu1oGnY4iUUxFpSjQNTzJyMQMS/PVpDvaLM5NIj7Gxf72YaejiMhp8Pkszf1eSjITnY4iYcJlDGcvyqCxd4wB75TTcUQkTMxfgNT2t/DicbtYVZROTfsQkzOzTseRKBXQ6W8SGg77m3RX5aqoFG08LhfL8lM52DHMrE97rSX0GGOuBr4LuIGfWGu/cdz9ccD/AGcDfcAHrbVNxpgrgG8AscAU8EVr7TNBDR8E3SOTTEz7KMtOcjqKhJGzitP5w4Eu9rYMctHSXKfjiEgYqOseJSsploykWKejcM/2ZqcjhJW1Jem82tTPgfZh1pZmOB1HopBWKkWBWk1+i2orC1MZn56loVdb4CS0GGPcwJ3ANUA1cIsxpvq40z4JDFhrK4HvAN/0H+8F3mutXQV8FPh5cFIHV1PfGADlWSoqycnLTIqlNDOR3a3aDiEiJ6euZ1Rb38LUoqxEMhJjtAVOHKOiUhQ40jVCdnIsWcma/BaNqvJSiHEbarQFTkLPBqDOWttgrZ0C7gU2H3fOZuBu/+37gcuMMcZau8ta2+4/XgMk+Fc1RZSm3rmi0qIsbX+TU7OmJJ2u4Uk6hsadjiIiIc5aS133qJp0hyljDGcVp1PfPcro5IzTcSQKaftbFDjcNUpqQoyWkkapGLeLpXkpHGif2wLndmmClISMIqDlmK9bgY1vdY61dsYYMwRkMbdSad4NwOvW2skAZnVEU5+XWLeLwnQ16pZTs6oojUf3trOnZZCCNP37EZG31jc2xdD4tPophbHVxek8d6SH/W1DbKrIcjqORBkVlSKcz2ep6xphdXG601HEQauK09nfPswr9b28qyrH6TgiC8YYs4K5LXG9McQlAAAgAElEQVRXvsX9nwY+DVBaWhrEZAvjaN8YJZkJKgbLKUuK81CVm8Ke1iGuXJGPy+jfkIi82T3bm99okdDc79VF6DCVlxpHbkoce1tVVJLgU1EpwjX3exmbmqUgLd7pKOKg+XGjD+5qU1FJQkkbUHLM18X+Yyc6p9UY4wHSmGvYjTGmGHgQ+Ii1tv5ET2Ct/RHwI4D169eHXbf6pj4vZeqnJKfprJJ0Du9s4Wifl3I1exeRt9A9PLfQNzcl4naRh4WFKOQZY1hVnMYzB7sZGp8mLSFmAZKJnBz1VIpw8310CrR1IqrFuF2sLExjy/5OvFPaay0h41WgyhhTboyJBW4GHjnunEeYa8QNcCPwjLXWGmPSgceA2621LwctcRBZaznaN8YiFZXkNFUXpBLrdrFbzVtF5G10DU8Q53GpEBHmVhelY4H9bUNOR5Eoo6JShKtpH8LjMuTpykPUW1OaztjULH840OV0FBFgrkcS8FlgC3AQuM9aW2OM+Yox5jr/aXcBWcaYOuA24Hb/8c8ClcAdxpjd/j8RNTu9Z2QS79Qs5dlq0i2nJ9bjYnlBCvvbhpj1hd1CPREJkq7hCfJS4zHaJhvWclLiKEyLZ68mf0qQqagU4Wrah6nMTcbj1v/V0a4sK4nCtHge2nX87iIR51hrH7fWLrHWLrbWfs1/7A5r7SP+2xPW2pustZXW2g3W2gb/8a9aa5OstWuO+dPt5M+y0Jr6vABaqSRnZFVRGuPTs2/0TBEROZa1lq7hSfJT1SojEqwqTqdlYJz+sSmno0gUUaUhwtW0D7OiMM3pGBICXMaweW0RL9T20jsacUOyRCJOU+8YgHoqyRmpyksh1uPSdggROaHhiRnGp2fJS9Wuhkiwumjuc98+/c6XIFJRKYJ1j0zQOzrJisJUp6NIiHjf2iJmfZZH97Q7HUVE3kFT3xgel6EwXVeP5fTFuF0sy0+hpn1YW+AkLBljrjbGHDbG1Bljbj/B/XHGmF/7799ujCnzH7/CGPOaMWaf/38vDXb2cNA1PAFAnob6RISMpFhKMhK0BU6CSkWlCDbfpFtFJZm3JC+F6oJUHtAWOJGQd7TPS2lmorYvyxlbWZiGd2qWRv/qN5FwYYxxA3cC1wDVwC3GmOrjTvskMGCtrQS+A3zTf7wXeK+1dhVzAx9+HpzU4WW+qJSfoqJSpFhdnE7H0AQ9I9qZIMGhd6oR7IC/qLRcRSU5xk3ri9nbOkRNu5bFioSyxt4xFmWpSbecuaX5KcS6tQVOwtIGoM5a22CtnQLuBTYfd85m4G7/7fuBy4wxxlq7y1o7vzS7BkgwxmiP13G6hidIifeQGOdxOooskJVFaRhgb5tWK0lwqKgUwWrahyjNTCQ1XuNB5X+9b20RcR4X9+5ocTqKiLwFay1H+8bUpFsWRIzbxdL8FGrah/BZbYGTsFIEHPuGpdV/7ITn+KeKDgFZx51zA/C6tfZNSzeMMZ82xuw0xuzs6elZsODhotM/+U0iR1pCDGXZSextGcLqd74EgYpKEexA+7C2vsmbpCfGcu2qAh7a1YZ3asbpOCJyAr2jU4xNzVKeraKSLIyVRWmMaQucRCFjzArmtsT92Ynut9b+yFq73lq7PicnJ7jhHDbrs3Rr8ltEWlWURs/oJJ3+7Y0igaSiUoQamZimqc+ropKc0C0bSxmZnOHRvR1ORxGRE2jqm/vgr+1vslCW5qUQ4zbaAifhpg0oOebrYv+xE55jjPEAaUCf/+ti4EHgI9ba+oCnDTNH+8aY8VlNfotAK4vScBnY16rf+RJ4KipFqIMdIwCsKExzOImEovWLMqjMTebeHc1ORxGRE2jyryYp0/Y3WSCxHhdL8zQFTsLOq0CVMabcGBML3Aw8ctw5jzDXiBvgRuAZa601xqQDjwG3W2tfDlriMHKka+7zgra/RZ7kOA+Lc5LZ26YtcBJ4KipFqPkmzNVaqSQnYIzh5nNKeL15kEOdw07HEZHjHO3z4nEZijMSnI4iEWRlURqjkzO82tTvdBSRk+LvkfRZYAtwELjPWltjjPmKMeY6/2l3AVnGmDrgNuB2//HPApXAHcaY3f4/uUH+EULaoc4RDJCryW8RaVVRGv1jU+zVaiUJMBWVIlRN+zDZybHkpmg5q5zYDeuKiXW7+NV2rVYSCTVNfWMUZyTgcetlWhbO0vy5LXBP7NPWZwkf1trHrbVLrLWLrbVf8x+7w1r7iP/2hLX2JmttpbV2g7W2wX/8q9baJGvtmmP+dDv5s4SaI10jZCbFEuvRa00kWlGYhtsYHt3b/s4ni5wBzY6MUDXtw1QXpmGMcTqKhKiMpFjes7qA377exheuWkqKpgSKhIzGXk1+k4UX53GzJC+FJ/Z38g/vXYHLpfcIItHsQPsw+WlapRSpEmLdVOUlc9/OVhZlJeE6yc+FH9pYGuBkEmlUVIpA41OzHOka4fLlWuErb+9j55fxwK42frOzlU9cUO50HBEBfD5LQ88YmyqOn4gdfu7RSsiQs7IojV+/2sLOowNsKM90Oo6IOGR+qM/ly/OcjiIBtLo4jUOdIzT3eSnTRFkJEK11jED72oaY9VnWlKQ7HUVC3OridM5elMHdW5vUuFUkRLQPjTM+PcvinGSno0gEWpaXQqzHxePaAicS1eaH+hSma6VSJFuen4rHZdiryZ8SQCoqRaA9LYMAnKWikpyEj51XxtE+L88dVpsBkVBQ3zM3+W1xjq4oysKLi3Fz8ZIcntjfgU8XE0Si1vxQn8I0DYSIZHExbpblp7yx6EAkEFRUikC7WwYpzkggO1lNuuWdXb0yn/zUeH76cpPTUUQEqO8eBaAyVyuVJDCuXV1A1/AkrzcPOB1FRBwyP9QnJV7dUCLdquJ0xiZnaOwdczqKRCgVlSLQ7pZBbX2TkxbjdnHruYt4qa6XI10jTscRiXp1PaOkJ8aQmRTrdBSJUJcuyyXW4+IxbYETiVoa6hM9lvq3Pe9rG3Q6ikQolabD1Fs1Px2ZmKZtcJyzitPUIFVO2i0bSvne07Xc9WIj37xxtdNxRKJaffcoi3OS9UZfAiYlPoaLluTwxL5OvnRttabAiUSZyZlZartGuHhpjtNRJAhiPS6qC1LZ3zbMe1f78Li1rkQWlv5FRZiWfi8AJZmJDieRcJKZFMtN64t5cFcb3cMTTscRiWr1PaNUqkm3BNh7VhfQOTzBa9oCJxJ1artGmfFZVhSmOh1FgmRtSTrj07PsU8NuCQAVlSLM0T4vbpehMF1N9+TUfOqCCmZ8Pv5bvZVEHDPonaJ3dIrFuWrSLYF12fI84jwuHturLXAi0Wa+SfeKwjSHk0iwLM5NJjs5lm0NfU5HkQikolKEOdrvpTg9gRgta5RTVJadxDUrC/jltqOMTEw7HUckKtX3zDXpXqyVShJgyXEeLlmay2P7OjQRSCTK7G8bJjnOwyLtbIgaLmPYVJFFy8A4rQNep+NIhFHlIYJMz/poGxhnUZZeIOT0/NlFFYxMzqgfl4hDDnfOFZWW5KU4nESiwbWrC+gZmeTVpn6no4hIENW0D7G8IEX91KLMutIMYj0uttZrtZIsLBWVIkjrwDiz1rIoS9sm5PSsLk7nvMVZ/PfLjUzOzDodRyTqHO4cJinWTZG2MEsQXLosl/gYbYETiSazPsvBjhFtfYtC8TFu1pWms7dtiNHJGafjSATR9LcI0tw3BkCplrLKWziZFUhL81J4pb6Ph3e384H1JUFIJSLzDneNsCRfV48lOJLiPFy2LI8n9nfwD++t1kQgkSjQ0DPK+PQs1WrSHZU2lWexraGfbQ19XL48z+k4EiH07iGCNPV5yUmOIylOtUI5fZW5yRSkxfPD5+vxqc+GSNBYazncOcJSbX2TILp2dQG9o1PsaNQWOJFosKtlEJibBibRJzc1nuqCVF6p72V8SrsSZGEEtKhkjLnaGHPYGFNnjLn9BPfHGWN+7b9/uzGmzH88yxjzrDFm1Bjz/UBmjBSzPktT3xjlOdr6JmfGGMOFVTnU94zx9KFup+OIRI3e0SkGvNPqpyRBdcnSXBJj3Ty6T1vgRKLB7pZBUuI8GggRxS5bnsvEtI+X63udjiIRImBFJWOMG7gTuAaoBm4xxlQfd9ongQFrbSXwHeCb/uMTwJeALwQqX6RpHxxncsZHRbaKSnLmVhalUZSewH89V4e1Wq0kEgyHO0cAWJavopIET0Ksm8uW5/H7/Z3MzPqcjiMiAba7eZCzStK1zTqKFaQlsKIwlZfrtFpJFkYgVyptAOqstQ3W2ingXmDzcedsBu72374fuMwYY6y1Y9bal5grLslJaOyd66dUrqKSLAC3y/BnF1XwevMg2xq0JUIkGA53zRWVlqioJEF27aoC+sem2NqgiUAikcw7NcPhrhHWaOtb1Lt0WS6TMz5eqtNqJTlzgSwqFQEtx3zd6j92wnOstTPAEJAVwEwRq6F3lJyUOFLiY5yOIhHiA+tLyE6O485n65yOIhIVjnSOkJUUS3ZynNNRJMpcvDSHpFg3j+7RFjiRSLavdYhZn2VtqYpK0a4gLYGVRWm8VNfD0Pi003EkzIV1o25jzKeNMTuNMTt7enqcjuOYuX5KXm19kwUVH+Pm0xeW81JdL7v9TR1FJHAOdAyzrECrlCT44mPcXFGdx+9rOpmc0VYIkUg1/35OK5UE4JoV+VgLT+zXBQU5M4EsKrUBx84jL/YfO+E5xhgPkAac9Npra+2PrLXrrbXrc3JyzjBu+God8DI146NCDfdkgX144yLSE2P4/jNarSQSSNOzPg53jrCiMM3pKBKlrl9bxND4NM8eit6LdCKRbnfLICWZCWRpRawAGUmxXLgkh72tQ2+0UhE5HYEsKr0KVBljyo0xscDNwCPHnfMI8FH/7RuBZ6y6Ap+y2u5RDLBYk99kgSXFefj4eeU8dbCLgx3DTscRiVh13aNMzfpYUZjqdBSJUhdUZpOdHMcDr7c6HUVEAmR3yyBrSzKcjiEh5MKqHNITYvjdnnZmffoYLqcnYEUlf4+kzwJbgIPAfdbaGmPMV4wx1/lPuwvIMsbUAbcBt88/3hjTBPwb8DFjTOsJJseJX23XCMUZCSTGepyOIhHoY+eVkRznUW8lCQhjzNXGmMPGmDpjzO0nuD/OGPNr//3bjTFl/uNZxphnjTGjxpjvBzv3QqtpnyvaqqgkTvG4XVy/ppBnD3fTPzbldBwRWWCdQxN0DE1o65v8kViPi3evKqBzeEJNu+W0BbQKYa19HHj8uGN3HHN7ArjpLR5bFshskWJ8apbWgXEuXprrdBSJMPdsb37j9rrSDB7b28GS3FqyU068ZPpDG0uDFU0ihDHGDdwJXMHcMIdXjTGPWGsPHHPaJ4EBa22lMeZm4JvAB5mbDvolYKX/T1iraR8iIcZNeba2MYtz3r+umJ+81Mije9v5yLllTscRkQW0u2UAgDVq0i3HWVGYysrCVJ462MUyTaCV0xDWjboF6ntGscCSPH0QkcA5vzILt8vw/BH12pAFtQGos9Y2WGungHuBzcedsxm423/7fuAyY4yx1o5Za19irrgU9mra55p0u13G6SgSxaoLU1mWn8JvXz++BaaIhLvXjg4Q63ZRXaAVsfLHjDFct6aIOI+L+19rZWbW53QkCTMqKoW5I10jxHlcFGckOh1FIlhKfAznlGWyq2WAAa+2RciCKQJajvm61X/shOf4t1UPAVlBSRckPp/lYPswK9WkW0LADeuK2dMySH3PqNNRRGQBbW/sZ01pOvExbqejSAhKjvOweU0RbYPj/PCFBqfjSJhRUSmM+azlcNcIS/J0dVsC711V2RgML9ZqtZKED2PMp40xO40xO3t6QvPf7tF+LyOTM+qnJCFh85pCXAYe1GolkYgxPDHN/rYhNlVE1DUZWWCritJYVZTGvz91hEOdGtAjJ09FpTDWMTjByMQMS7X3VYIgPTGWtaXp7GwaYGRi2uk4EhnagJJjvi72HzvhOcYYD5AG9J3sE1hrf2StXW+tXZ+Tk3OGcQNjvs/FWWqeKiEgNzWed1XlaAuESAR5rWkAn4VN5ZlOR5EQd91ZhaQlxPDX9+1hWq8BcpJUVApjhzqHMcCSPBWVJDguWpLDrM/yYq2mQ8iCeBWoMsaUG2NigZuBR4475xHgo/7bNwLPWGsjaubt7uZBEmPd+l0uIeOWDaV0Dk/wzKFup6OIyALY1tBHrNvF2tIMp6NIiEuK8/DV61dR0z7M95/R5Gc5OSoqhbFDnSOUZCaSHBfQIX4ib8hKjmNNSTrbGvoY1molOUP+HkmfBbYAB4H7rLU1xpivGGOu8592F5BljKkDbgNun3+8MaYJ+DfgY8aYVmNMdVB/gAWyu2WQ1cVp2sYsIePy5bnkp8bzi2OmgIpI+NrW0MeaknQSYtVPSd7Z1Svzef/aIr7/bB17WwedjiNhQEWlMDU0Pk3b4LjGPkrQXbosF5+1PH84NPvTSHix1j5urV1irV1srf2a/9gd1tpH/LcnrLU3WWsrrbUbrLUNxzy2zFqbaa1NttYWW2sPOPVznK6J6VkOdAyzpkRXjyV0eNwubt5QwgtHemju8zodR0TOwMjENPvbh9lYoa1vcvL+4boV5CTHcdt9e5iYnnU6joQ4FZXCVE37EAArNC1IgiwrOY51pRnsaOpnaFyrlUTOxIGOYaZnLWvUT0lCzM3nlOJ2GX6546jTUUTkDOw8OsCsz6pJt5yStIQY/uXG1dR1j/KvTx52Oo6EOBWVwlRN+zC5KXHkpMQ5HUWi0CXLcsHCs4fVb0PkTOxunltWvrZURSUJLflp8VyxPI/f7GxlckZXqUXC1baGPmLchnXqpySn6MIlOXx4Yyk/eamRHY39TseREKaiUhjqG52kqXdMq5TEMRmJsawvy+C1pgEGxqacjiMStnYe7acoPYG81Hino4i8yYc3ldI/NsUT+zqdjiIip2lbQz9nFaufkpyev3v3ckoyEvnCb/YwNjnjdBwJUSoqhaEtNV1YYEVhqtNRJIpdvDQXY7RaSeR0WWvZ0djPRo14lhB1/uJsyrOT+OkrTUTY0EWRqNA/NsXe1kHOr8x2OoqEqaQ4D9++6SxaBrx8/fGDTseREKWiUhh6ZE8b2cmxFKTpyrY4Jy0hhnPKM3m9eYC+0Umn44iEnfqeUXpHp9Q8VUKWy2X41LvK2dMyyNb6PqfjiMgpeu5wN9bCZctznY4iYWxDeSafuqCcX25v5vkjGtQjb6ZZ9GGmc2iC7Y39XLo0F2M0flqcdfGSHHY29fP0oW7+6rIqp+OIhJVtDXP9CTaWq3mqBNc925tP+tyZWUtKnIcvPbyfT15QwYc2lgYwmYgspGcOdZOdHMdKtcyQM/TXVy7l2cM93P7bvTz5+QtJiY9xOpKEEBWVwsyje9uxFs4qVlNXcV5KfAznVmTzQm0Pe1sHWa1/lyInbXtjP3mpcSzKSnQ6ishbinG7uKAqmyf2d9LS73U6joi8g/mi8azP8tTBLlYUpnHvqy0Op5JwFx/j5ls3ruaG/3qFrz9+iH9+/yqnI0kI0fa3MPPw7nZWFqWSralvEiIuXppDUqybrz56UD03RE7SXD+lPjaUZ2nVqYS8DWWZJMS4te1BJIwc7R9jYtrH0rwUp6NIhFhbmsGn3lXBr3Y083Jdr9NxJISoqBRGDnUOs69tiPevLXY6isgb4mPcXF6dx46mfrbUaEKQyMmo7R6la3iS8xZr65uEvrgYN5sqsjjQMUxt14jTcUTkJBzuGMFtDFW5yU5HkQhy2xVLKM9O4vYH9moanLxB29/CyG92thLjNly/tojf79eHdwkd6xdlcrBjmK8/fohLluUS59HYWpG385x/auLFS3McTiJycs5bnMVLdT1875k6/uOWtU7HEZF3cKhrhPLsJOJi9J5MTs079d27YnkeP36xgU/dvZP3nlWoXnuilUrhYnrWx0O72rh8eR6ZSbFOxxH5I26X4f9dW01zv5efvtzkdByRkPfc4R6W5qVQkJbgdBSRk5IU5+H8ymx+t6edva2DTscRkbfRNzpJz8gkS/O19U0WXll2Epsqstja0Edj75jTcSQEqKgUJv5woIu+sSluWq+tbxKaLlqSw+XLc/n3p47Q3KdmriJvZWxyhleb+rVKScLOhVU5ZCbF8o0nDqmHnkgI298+DMDyglSHk0ikunJFHhmJMTzweivjU7NOxxGHqagUJn6x7ShF6QlctCTX6Sgib+krm1ficbn4uwf36QOHyFt4pb6P6VnLRUtUVJLwEh/j5v9cVsUr9X3ahi8Swva2DlKSkaDdDRIwcR4371tbTN/YFN956ojTccRhKiqFgbruEV6p7+PDm0pxuzQlSEJXYXoCf3P1Ul6q6+W3r7c5HUckJG2p6SQl3sP6skyno4icsg9vLGV5QSpfefSAmrSKhKCekUk6hiZYXZzudBSJcJW5yZxTlslPXmxgV/OA03HEQWrUHQZ+sa2ZWLeLD6wvcTqKyFuab+pnjGFRViJfemg/faOTpMTHnPB8NfWTaDQ14+PJmk6uqM4j1qPrOhJ+PG4XX71+BTf811b+9ckj3PHeaqcjicgxdrcMYoCVRWlOR5EocM3KfFoHvHzx/r089rkLNKwnSukdbYgb9E7x61dbeM9ZBWQnxzkdR+QduYzhfWuLmJ718cDrbdoGJ3KMl+t6GZ6Y4dpVBU5HETltZy/K5CPnLuKnrzSyraHP6Tgi4jfrs7zePEBlbjJpCSe+qCeykOJj3Hz9/auo6x7lu0/VOh1HHKKiUoj7xbajjE/P8ukLK5yOInLSclPiuWZlPoe7Rni5Xh84ROY9tq+DlDgPF1RlOx1F5Izcfs0ySjMT+ev79jDonXI6jogwd+FiaHxa26slqC5ZmstNZxfzX8/Xs1Xv+6OSikohbGJ6lp+90sRFS3JYlq/pDRJeNlVkUV2Qyu/3d9DQO+p0HBHHjU/NsqWmkytW5Gl5uIS9xFgP3715Ld0jE9x23x58Pq1KFXHar3Y0kxDjZnl+itNRJMp8+boVlGcl8flf72ZgTBcaoo2KSiHsF9uO0js6xWcuXux0FJFTZozhxrOLyUyK457tzfTrBUai3GP7OhiZmOGD6o8nEWJNSTpfek81zxzq5ttPHnY6jkhUaxscZ0tNJ+eUZeJx6yOeBFdSnIfv3bKW/rEpvnj/HrW/iDL6jROivFMz/OD5es6vzGJjRZbTcUROS3yMm1s3LcJa+OnLjYxMTDsdScQxv9rRTEVOEhvKtS1BIsetmxZxy4ZS/vO5en6+tcnpOCJR63+2NmGMYVOFXmPEGSuL0rj9mmU8dbCb7z9T53QcCSIVlULU3a/MrVL6/OVLnI4ickZyUuL46LmLGJ6Y5r9VWJIodaRrhNeODnDLOaUYY5yOI7JgjDH80+YVXL48ly89XMMvth11OpJI1Bkan+aebc1cvSKf9MRYp+NIFPv4+WW8b20R//qHI/x+f6fTcSRIVFQKQT0jk9z5bB2XLctVoz2JCKVZSdy6qYz+sSl+/GKDtsJJ1Pnh8w3Ex7i44exip6OILDiP28WdH17H5ctz+fuH9vOdPxzR1geRIPrZy02MTM7wmUvUMkOcZYzhn9+/irNK0rntvt3UtA85HUmCQEWlEPSvTx5mYnqW/3ftcqejiCyYytxkPn5eOWOTs/znc3W8VNvrdCSRoGjp9/LQ7jZu2VBKZpKuIEtkivO4+c8Pn82NZxfz3adr+bOfv6ZmrSJBMOid4q6XGrh8eR4rCtOcjiNCfIybH996NmkJMXzkrh3UdY84HUkCTEWlELOjsZ97X23hY+eVUZGT7HQckQVVlp3EZy5eTHKchz+5azv/+LsabYeTiPfDF+pxGfj0hRVORxEJqFiPi2/duJq/v3Y5zx7u5urvvsCDu1o1GU4kgP79qVpGJ2f4wlVqmSGhIzc1nl9+aiMul+FDP95OY++Y05EkgDxOB5D/NTE9y9/8di/FGQl8/gq9MEhkykqO4zMXV9LYO8rPXmnid3s6+Owli7l5QynxMRqzLpGltmuEX+1o4ZYNJRSkJTgdRyTgjDF86l0VbKrI4m8f2Mfnf72Hn73cxIbyLMqyEk+rp9iHNpYGIKlI+DvSNcIvth3lg+eUsiw/1ek4In+kIieZez61kQ/+aBsf+OFW7vroelYXpzsdSwJAK5VCyD89eoDG3jG+8f7VJMWp3ieRK9bj4h83r+Shz5xPeXYiX/7dAc752lN88Td7eLG2h5lZn9MRRc6YtZYv/66GpFg3t12x1Ok4IkG1siiNh//yfL5901l0Dk/w/7d378Fxlecdx7/P3qSVtLrLkixZtmSEBTbGNsaXhFwgk8RtLpCUAB2aMKkLaZtM02koSfpHmgt0YKZJoZOUJCWpgSEhwbnAkHuISWkMvhtsbHyTZVuyLNm6X6zLSm//2IMtBAZL9mp1Vr/PzJlz9t1d+X3mHJ9nz3ve9z3//Vw9//XsIbYf6WBY53iRCxYfGeWfn3iR3GiYO9+nm9EyPdWWxnj8jlVEggFu+s7z/HJXc6qrJEmglotp4smdTTy26SifemcN19QWp7o6IlPiyjn5PPG3b2Pz4XZ+tOUYv9p9gie2NRINB1lcmcfSqgIWzs5lQVmM6uJswkG1g4t/rN/WyJ8OtvHlD12uuZRkRgoEjBuvquTPryjjCz/ZxQv1bazf3sjTu45zWVkuC2fnUVuao3O7yCR8c8NBXmzs4j//cilFORmpro7IOV1aGuPJz7ydTz26jb9/bDufWD2Xz6+pUyeKNKI9OQ1sbWjnrvUvcfW8Au58v+5my8yzorqQFdWF3DO8iGf3neSF+jZ2HOvke/9Xz/BIYi6OoBnFsQizYpmU5mZSlptBaW4mBdkRAuOGU2iohKTaoZO9fOnJl1lVU8jHV89LdXVEUiorEmJVTRErqwupP9XHzqOd7GnuZsexTiKhADMPr8AAAA75SURBVAtKYywojXFJaQ65meFUV1dk2tuwr5UHnjnAR5dV8KHF5amujshbKs7J4LG/Wcl9v36FdRsbeHbfSe6+YRHvqC2e1LBomV7UqJRiu5u6WPvwViryo3zn48t1t05mtMxwkDWLylizqAxIzDNWf7KP/S09/GxHEy3dAzR29LOr6ezjScNBoySWQanX2FSam0lH3xAF6hkiKdLWO8jtj2wlIxzg/puXEgzox5IIJOZbml+Sw/ySHG4YddSf7GX38W72NHefOa+X5WZSOyuH2tIYc4uy9LtIZJydxzr59GPbqSvL5e4bFumCXFLuB5uOnvdna2fFePz2Vdz1k5f4xPc3s7qmiDvfv4Cr5hYksYaSbGpUSqGNh07xqUe2kRsN8/Bfr9DwCJlRJpKAAN6/sOzM9mB8hNbuQVq6B2jtSawPnexlx7FOAB5+voGakmyWVRVw1dzEcklJDgFd3EuStfcN8cl1W2jqOM2ja1dSlpeZ6iqJTEvBgFFbGqO2NMb1S2ZzomuAA629HGjpYeOhNp47eIpw0KguzqZ7YJir5xWwqCKPjJAe6CAz1+bD7axdt4XC7AgPf/JqsiK6lBP/OXSyj7Vvr2ZzQzsbXmnlLx7cSGVBlFXVRSysyH3deV4jEKY/nYlSID4yynf+t56v/3YfNSU5PLp2hZ4KJDIBGaEgcwqzmFOY9Zry/qE4J7oGKMyJsP1IJ394pZX12xoBiGWGWFZVwLKqAhZX5lFdnE1lQZSQ7oLLRbK/pYfbH9lKc9cAD966jBXVhamukogvBMyYnR9ldn6Ud11awmB8hMOn+jjQ0suB1l7u/dUrAGSEAlw5J5+r5xWwfG4hl5XnUpqboZ4akvacczzy/BHu/sUeqgqzeHTtSmbl6qaF+FcoGOBt84u5qqqA7Uc7eKG+nfXbG/n5TuPS0hiLKvKoK4vpydA+kdRGJTNbAzwABIGHnHP3jns/A3gEuApoA252zjV4730RWAuMAP/gnPtNMus6FZxzbDzUxj2/2Mue5m4+sLicez96BTHNHyByUWRFQtSU5Jy5o+Gco6Gtn21HOth2pIPtRzq4/5n9uMQ0TYQCRlVhFqGgkR0JkRUJkpWRWGeGgmSEAkTCgbPboQAZoSC3vW2uLmIuknTIE32DcR567jDf2nCQ3GiIH96+St24RS5ARihIXVnumUekv29hKVsbOtjS0M7Whna+/cd6RkYPAZAXDbOgLEZdWYyqwizK86KU52cyOy9KcU5ENw7SQDrkiclyzrGloYN//+0+Nh9u57q6WXzjpivJz9LoBkkPGeEgq+cXs6qmiIa2fnY3dfHy8S72NHcTMKjIj9LU2c/qmmKWVOWTo8m9p6Wk7RUzCwLfAt4LNAJbzOwp59yeMR9bC3Q45y4xs1uA+4Cbzexy4BZgITAb+L2ZXeqcG0lWfZPFOcfR9n6e3XeS9dsa2dXURXleJg/euow1i8p0YSqSRGaJoRPVxdnceFUlAF2nh9nf0sPhU300nOrj8Kk+XmzspLV7kP6hEYbO41HXX/vFHrIjQWKZYYpjGZTlZlCeF01MIJ6XQVlulPK8TMryMnWH5U34OU/ER0bZ09zN0y8188TWY3T0D/OBxeV85cMLKdZTeCSNTXTo8sVQnJPxmvn2+gbjvNTYxf6WHl450ZOYd297Ez2D8dd9Nxy0xI2BcJBM7yZBxLtJEA4GiASNcChAJJgoiwQDhEMB3ntZ6ZkbDbHMELmZYWKZIZ3Tp5if88SFaOke4Nl9rfxw8zF2HuukMDvCvR+9gpuWz9FQfklLY3+zf2BxOcfa+9nX0kP9yT6+/cd6vrXhEGZQXZzNFRV5XFGRx6KKPGqKsymJqcdqqiWzqW8FcNA5Vw9gZo8D1wNjk8D1wJe97fXANy1xRFwPPO6cGwQOm9lB7+89n8T6TphzjtPDI/QOxOkZjCfWA3GOd52msb2f+lN9bG3o4ET3AAB1ZTHuvmERH1teqTkBRJLofC96KguyqCzI4h21JWfKhkdG6R8aYXB4hMH4KIPxUYbiZ7cH46PUzsqhdzDx/721Z4D6k31sPNRGz8DrL2gKssKU5mZ6jUxRSnMzyI+GyY2Gyc301tEQ2ZEQ4WCAUNAIBwOEvXUoYOmcKH2RJ4629XPwZA9NHadp7DzNnuPd7DjaSe9gnFDAuK5uFn/37vksrVLvJJGpkJ0RYvX8IlbPLzpT5pyj+3TiN1hz12mOdw6wYV8rg8OjDHjn84HhEQaGR+gdjDM0MspQfJShkVGG46O4cf/Gj7Yce8N/OxIKJM7dmSFi0cQ6cS4/2/D06vn91e1YZoiMUJBQwM6e5wOJ9avbaig4J1/kibfinCM+6hiKjzLsHXuD8VG6Tg/T1jdEW+8gzV0D7G/pYW9zN/tbegGYV5TF125YxI3LKolGdO0gM0PAjLlF2cwtygbgw0tms6WhnZeOdbGrqYtN9e08ufP4mc9Hw0GqCrOYW5RFRUGUouwIBdkRCrMiFGZHyPHOwZnhwGvW4WBa/8aeUslsVKoAxmbkRmDluT7jnIubWRdQ5JW/MO67Fcmr6uRsPNTGrQ9tesP3AgbleVGWzytgZU0Rq2sKuWRWbIprKCITFQ4GyIsGIHruYannmjCwbzDOie4BTnR5S/cAzV2nOdE1QHPXAC81dtHWNzThOoUCRsCMu29YxE1Xz5nw96cxX+SJ+3+/n5/uaAISvR7ml+TwkaUVLJ9XwDWXFFOknkkiSTWZ3lEBM95TV3pen331gn/Ya2Qaio9y3WWz6B8aoX8ocQOheyBO9+lhugeGE69PD58pO955+sz2YPyte7u+EbNEnc3bNgwMqgqz+P0/vWtSfzNN+CJPvJnrvv4sh0/1nRl6/2Yq8qNcWprDR5ZWcm1dCQtKY7rolRkvJyPEtQtmce2CWWfKWnsG2Nvcw5G2Po609XOkLTH64E8HT9E3NLHOiAHv/BswwwzWXlPNXWvqLnYYac3XgxLN7A7gDu9lr5ntm8DXi4FTF79WZx0GNpLosztFkh5TiqRjXIrJP14X160pqsjN/wY3T/7rcy9eTfwjGXniIOCrSTlea8b8P00Disk/khbXAcA+l4y//DozMkfA9LqeOELi2mHdxfqDk6P/p/4xI2Ka6t/dn/eWiyhd9tM580QyG5WagLG31Cu9sjf6TKOZhYA8EhPsnc93cc59F/juZCpnZludc8sn893pKh1jgvSMSzH5R7rGNU0oT0yxdIwJ0jMuxeQf6RrXNKE8McXSMSZIz7gUkz+kY0zjJfORGFuAWjOrNrMIiYnynhr3maeA27ztG4E/OOecV36LmWWYWTVQC2xOYl1FRGTqKU+IiMibUZ4QEZnmktZTyRvT/BkSowSCwPedcy+b2VeBrc65p4DvAY96E+e1k0gUeJ/7MYlJ+OLAp/3wpAYRETl/yhMiIvJmlCdERKa/pM6p5Jz7JfDLcWVfGrM9AHzsHN+9B7gnidWbVDfXaS4dY4L0jEsx+Ue6xjUtKE9MuXSMCdIzLsXkH+ka17SgPDHl0jEmSM+4FJM/pGNMr2HufB5FICIiIiIiIiIiMkYy51QSEREREREREZE0NSMalczs+2bWama7x5QVmtnvzOyAty5IZR0nyszmmNkGM9tjZi+b2We9ct/GZWaZZrbZzF70YvqKV15tZpvM7KCZ/cibqNFXzCxoZjvM7GnvdTrE1GBmu8xsp5lt9cp8e/wBmFm+ma03s1fMbK+ZrfZ7THJ+lCf8QXnCX5QnJJ0oT/iD8oS/KE+khxnRqASsA9aMK/sC8IxzrhZ4xnvtJ3Hgc865y4FVwKfN7HL8HdcgcJ1z7kpgCbDGzFYB9wH/4Zy7BOgA1qawjpP1WWDvmNfpEBPAtc65JWMek+nn4w/gAeDXzrk64EoS+8zvMcn5WYfyhB8oT/iP8oSki3UoT/iB8oT/KE/4nXNuRizAPGD3mNf7gHJvuxzYl+o6XmB8TwLvTZe4gCxgO7ASOAWEvPLVwG9SXb8JxlJJ4uRxHfA0YH6Pyat3A1A8rsy3xx+QBxzGm2suHWLSMuFjQHnCR4vyxPRflCe0pNuiPOGvRXli+i/KE+mxzJSeSm+k1DnX7G2fAEpTWZkLYWbzgKXAJnwel9etcyfQCvwOOAR0Oufi3kcagYpU1W+S7gfuAka910X4PyYAB/zWzLaZ2R1emZ+Pv2rgJPA/Xtfih8wsG3/HJBcmbfa98sS0pzzhD8oTMl7a7HvliWlPecIfZmSemMmNSme4RJOhLx+DZ2Y5wE+Af3TOdY99z49xOedGnHNLSLTGrwDqUlylC2JmHwRanXPbUl2XJLjGObcM+DMS3aXfOfZNHx5/IWAZ8KBzbinQx7iuqT6MSS4SP+975YnpTXnCV8ef8oSck5/3vfLE9KY84avjb0bmiZncqNRiZuUA3ro1xfWZMDMLk0gAjznnfuoV+z4uAOdcJ7CBRFfOfDMLeW9VAk0pq9jEvR34sJk1AI+T6LL6AP6OCQDnXJO3bgV+RiJp+/n4awQanXObvNfrSSQFP8ckF8b3+155wheUJ/xDeULG8/2+V57wBeUJ/5iReWImNyo9Bdzmbd9GYgyxb5iZAd8D9jrnvjHmLd/GZWYlZpbvbUdJjOneSyIZ3Oh9zFcxOee+6JyrdM7NA24B/uCcuxUfxwRgZtlmFnt1G3gfsBsfH3/OuRPAMTNb4BW9B9iDj2OSC+brfa884Q/KE/6JS3lC3oCv973yhD8oT/gnrpmaJyzR+yq9mdkPgXcDxUAL8K/Az4EfA1XAEeAm51x7quo4UWZ2DfAcsIuzY2v/hcQ4aF/GZWaLgYeBIIkGzx87575qZjUkWuULgR3AXznnBlNX08kxs3cDdzrnPuj3mLz6/8x7GQJ+4Jy7x8yK8OnxB2BmS4CHgAhQD3wS71jEpzHJ+VGe8EdcyhP+iUl5wj8xyflRnvBHXMoT/olJecI/Mb2VGdGoJCIiIiIiIiIiF9dMHv4mIiIiIiIiIiKTpEYlERERERERERGZMDUqiYiIiIiIiIjIhKlRSUREREREREREJkyNSiIiIiIiIiIiMmFqVBIRERERERERkQlTo5KIiIiIiIiIiEyYGpVERERERERERGTC/h+6rjd/965ghgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 1440x432 with 3 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_num_images_per_exam(valid_dir)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 554
        },
        "id": "2_L8Q37R9474",
        "outputId": "c997ccb9-c125-4271-fa6e-84430b43385a"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "axial - min: 20, max: 52\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "coronal - min: 17, max: 48\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "sagittal - min: 21, max: 45\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x432 with 3 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##### Helper Function for Visualization of Category of Distribution"
      ],
      "metadata": {
        "id": "2wBRHgwMZ0AM"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def label_race (row):\n",
        "  if row['meniscus'] == 0 and row['acl'] == 0 and row['abnormal'] == 0:\n",
        "    return 'Nothing Detected'\n",
        "  if row['meniscus'] == 0 and row['acl'] == 0 and row['abnormal'] == 1:\n",
        "    return 'Meniscus Only'\n",
        "  if row['meniscus'] == 1 and row['acl'] == 0 and row['abnormal'] == 1:\n",
        "    return 'Meniscus + Abnormal'\n",
        "  if row['meniscus'] == 0 and row['acl'] == 1 and row['abnormal'] == 1:\n",
        "    return 'Meniscus + Acl'\n",
        "  if row['meniscus'] == 1 and row['acl'] == 1 and row['abnormal'] == 1:\n",
        "    return 'Meniscus + Acl + Abnormal'\n",
        "  return None"
      ],
      "metadata": {
        "id": "X_e9yPjNTbdM"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Analyzing the positive negative cases in dataset and then checking for MIssing Values"
      ],
      "metadata": {
        "id": "bNREthzMcLDO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_abnormal_df = pd.read_csv('train-abnormal.csv',header=None,names=['case', 'abnormal'],dtype={'case': str, 'abnormal': np.int64})\n",
        "print(train_abnormal_df.groupby('abnormal').count())\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "n7K8EfeS97Kp",
        "outputId": "540b66c0-1f9a-4c2e-b10b-a6cbb6ed3d14"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "          case\n",
            "abnormal      \n",
            "0          217\n",
            "1          913\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_abnormal_df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "CYA_uENtaML9",
        "outputId": "866cfe64-b303-4512-c373-1705af14c10e"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-bb43c4e0-385a-4313-9bf6-07dbd47ff87d\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>case</th>\n",
              "      <th>abnormal</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0000</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0001</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0002</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0003</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0004</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-bb43c4e0-385a-4313-9bf6-07dbd47ff87d')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-bb43c4e0-385a-4313-9bf6-07dbd47ff87d button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-bb43c4e0-385a-4313-9bf6-07dbd47ff87d');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   case  abnormal\n",
              "0  0000         1\n",
              "1  0001         1\n",
              "2  0002         1\n",
              "3  0003         1\n",
              "4  0004         1"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print('\\nMissing values: ', train_abnormal_df.isnull().values.any())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1C1paB4jaJIU",
        "outputId": "0a58e647-45da-4450-b5c9-a6e420f0cb1b"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Missing values:  False\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_acl_df = pd.read_csv('train-acl.csv',header=None,names=['case', 'acl'],dtype={'case': str, 'acl': np.int64})\n",
        "train_acl_df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "b0-O7eOJ-EVX",
        "outputId": "edb8fd99-71b7-4f5e-efb9-d65ab77d7706"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-541005eb-4d15-454b-93ac-a33643291687\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>case</th>\n",
              "      <th>acl</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0001</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0002</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0003</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0004</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-541005eb-4d15-454b-93ac-a33643291687')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-541005eb-4d15-454b-93ac-a33643291687 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-541005eb-4d15-454b-93ac-a33643291687');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   case  acl\n",
              "0  0000    0\n",
              "1  0001    1\n",
              "2  0002    0\n",
              "3  0003    0\n",
              "4  0004    0"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(train_acl_df.groupby('acl').count())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cd-Clx4XaVTJ",
        "outputId": "43a35d5f-03b2-4542-e4d4-41811698574f"
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "     case\n",
            "acl      \n",
            "0     922\n",
            "1     208\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print('\\nMissing values: ', train_acl_df.isnull().values.any())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HFBcYzX1aUzj",
        "outputId": "adf38cbb-c524-4aae-e5d0-ecd5fff45124"
      },
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Missing values:  False\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_meniscus_df = pd.read_csv('train-meniscus.csv',header=None,names=['case', 'meniscus'],dtype={'case': str, 'meniscus': np.int64})\n",
        "print(train_meniscus_df.groupby('meniscus').count())\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5V_CGznC-IYH",
        "outputId": "a5fdbdd7-f370-496a-f13f-30f78de5839b"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "          case\n",
            "meniscus      \n",
            "0          733\n",
            "1          397\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_meniscus_df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "Ox6z4YGSa5lq",
        "outputId": "c2a4255f-f2ae-4187-c833-6fa841c598e4"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-aa29c72e-4ffb-4008-89f9-e7d8e435bfdc\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>case</th>\n",
              "      <th>meniscus</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0001</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0002</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0003</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0004</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-aa29c72e-4ffb-4008-89f9-e7d8e435bfdc')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-aa29c72e-4ffb-4008-89f9-e7d8e435bfdc button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-aa29c72e-4ffb-4008-89f9-e7d8e435bfdc');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   case  meniscus\n",
              "0  0000         0\n",
              "1  0001         1\n",
              "2  0002         0\n",
              "3  0003         1\n",
              "4  0004         0"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print('\\nMissing values: ', train_meniscus_df.isnull().values.any())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wYbJv1ZEa6yR",
        "outputId": "aa05d207-21bf-4290-dd94-cf29b29a3450"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Missing values:  False\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Lets Merge Dataset to see Distribution of How Many Belong to more than 1 category"
      ],
      "metadata": {
        "id": "ATUb5fpRa7iq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_df = pd.merge(train_abnormal_df, train_acl_df, on='case').merge(train_meniscus_df, on='case')\n",
        "train_df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "rSiIw14PIL_H",
        "outputId": "f2d7495e-e4be-4af9-f46b-70ad1c7c1034"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-3508aa30-faf2-429f-8273-998f3afe7d72\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>case</th>\n",
              "      <th>abnormal</th>\n",
              "      <th>acl</th>\n",
              "      <th>meniscus</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0000</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0001</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0002</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0003</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0004</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-3508aa30-faf2-429f-8273-998f3afe7d72')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-3508aa30-faf2-429f-8273-998f3afe7d72 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-3508aa30-faf2-429f-8273-998f3afe7d72');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   case  abnormal  acl  meniscus\n",
              "0  0000         1    0         0\n",
              "1  0001         1    1         1\n",
              "2  0002         1    0         0\n",
              "3  0003         1    0         1\n",
              "4  0004         1    0         0"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_df['diagnostic_label'] = train_df.apply (lambda row: label_race(row), axis=1)"
      ],
      "metadata": {
        "id": "CWdhYs6GABif"
      },
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Preprocessing and Merging the Validation Dataset"
      ],
      "metadata": {
        "id": "9BWYr-TZbTka"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "valid_abnormal_df = pd.read_csv('valid-abnormal.csv',header=None,names=['case', 'abnormal'],dtype={'case': str, 'abnormal': np.int64})\n",
        "valid_acl_df = pd.read_csv('valid-acl.csv',header=None,names=['case', 'acl'],dtype={'case': str, 'acl': np.int64})\n",
        "valid_meniscus_df = pd.read_csv('valid-meniscus.csv',header=None,names=['case', 'meniscus'],dtype={'case': str, 'meniscus': np.int64})\n",
        "valid_df = pd.merge(valid_abnormal_df, valid_acl_df, on='case').merge(valid_meniscus_df, on='case')\n",
        "valid_df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "QHZ727NC-RWk",
        "outputId": "05d1bf70-b9b2-4d87-815c-54ded15dfdc1"
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-1bd1334a-5a34-46db-bfbd-f4b47e6be3bf\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>case</th>\n",
              "      <th>abnormal</th>\n",
              "      <th>acl</th>\n",
              "      <th>meniscus</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1130</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1131</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1132</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1133</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1134</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1bd1334a-5a34-46db-bfbd-f4b47e6be3bf')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-1bd1334a-5a34-46db-bfbd-f4b47e6be3bf button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-1bd1334a-5a34-46db-bfbd-f4b47e6be3bf');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   case  abnormal  acl  meniscus\n",
              "0  1130         0    0         0\n",
              "1  1131         0    0         0\n",
              "2  1132         0    0         0\n",
              "3  1133         0    0         0\n",
              "4  1134         0    0         0"
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "valid_df['diagnostic_label'] = valid_df.apply (lambda row: label_race(row), axis=1)"
      ],
      "metadata": {
        "id": "_Bo7wBxpAvJW"
      },
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "valid_df.tail()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "U6znCWEUA2s-",
        "outputId": "aa237c6d-8795-480e-989e-f5b1da80679e"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-0062275f-3d17-4675-9211-e18a3319964a\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>case</th>\n",
              "      <th>abnormal</th>\n",
              "      <th>acl</th>\n",
              "      <th>meniscus</th>\n",
              "      <th>diagnostic_label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>115</th>\n",
              "      <td>1245</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>Meniscus + Acl + Abnormal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>116</th>\n",
              "      <td>1246</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>Meniscus + Acl + Abnormal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>117</th>\n",
              "      <td>1247</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>Meniscus + Abnormal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>118</th>\n",
              "      <td>1248</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>Meniscus + Acl + Abnormal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>119</th>\n",
              "      <td>1249</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>Meniscus + Abnormal</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-0062275f-3d17-4675-9211-e18a3319964a')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-0062275f-3d17-4675-9211-e18a3319964a button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-0062275f-3d17-4675-9211-e18a3319964a');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "     case  abnormal  acl  meniscus           diagnostic_label\n",
              "115  1245         1    1         1  Meniscus + Acl + Abnormal\n",
              "116  1246         1    1         1  Meniscus + Acl + Abnormal\n",
              "117  1247         1    0         1        Meniscus + Abnormal\n",
              "118  1248         1    1         1  Meniscus + Acl + Abnormal\n",
              "119  1249         1    0         1        Meniscus + Abnormal"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Plotting Visualization for Distribution of the Dataset"
      ],
      "metadata": {
        "id": "-yw5QGDjbYlC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "pd.options.display.float_format = '{:,.1f}'.format\n",
        "train_co_occurency_df = train_df.groupby(['diagnostic_label']).count().reset_index()\n",
        "train_co_occurency_df['%_freq'] = (train_co_occurency_df.case / train_co_occurency_df.case.sum()) * 100\n",
        "train_co_occurency_df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "pLnB36hp_KaA",
        "outputId": "46097cac-a3c8-4767-ee9b-c0e056c3071d"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-2d3749a2-7c37-4bf7-95d2-2fc663b0ddcb\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>diagnostic_label</th>\n",
              "      <th>case</th>\n",
              "      <th>abnormal</th>\n",
              "      <th>acl</th>\n",
              "      <th>meniscus</th>\n",
              "      <th>%_freq</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Meniscus + Abnormal</td>\n",
              "      <td>272</td>\n",
              "      <td>272</td>\n",
              "      <td>272</td>\n",
              "      <td>272</td>\n",
              "      <td>24.1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Meniscus + Acl</td>\n",
              "      <td>83</td>\n",
              "      <td>83</td>\n",
              "      <td>83</td>\n",
              "      <td>83</td>\n",
              "      <td>7.3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Meniscus + Acl + Abnormal</td>\n",
              "      <td>125</td>\n",
              "      <td>125</td>\n",
              "      <td>125</td>\n",
              "      <td>125</td>\n",
              "      <td>11.1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Meniscus Only</td>\n",
              "      <td>433</td>\n",
              "      <td>433</td>\n",
              "      <td>433</td>\n",
              "      <td>433</td>\n",
              "      <td>38.3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Nothing Detected</td>\n",
              "      <td>217</td>\n",
              "      <td>217</td>\n",
              "      <td>217</td>\n",
              "      <td>217</td>\n",
              "      <td>19.2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2d3749a2-7c37-4bf7-95d2-2fc663b0ddcb')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-2d3749a2-7c37-4bf7-95d2-2fc663b0ddcb button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-2d3749a2-7c37-4bf7-95d2-2fc663b0ddcb');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "            diagnostic_label  case  abnormal  acl  meniscus  %_freq\n",
              "0        Meniscus + Abnormal   272       272  272       272    24.1\n",
              "1             Meniscus + Acl    83        83   83        83     7.3\n",
              "2  Meniscus + Acl + Abnormal   125       125  125       125    11.1\n",
              "3              Meniscus Only   433       433  433       433    38.3\n",
              "4           Nothing Detected   217       217  217       217    19.2"
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_co_occurency_df = train_co_occurency_df.drop(['abnormal', 'acl', 'meniscus'], axis = 1)"
      ],
      "metadata": {
        "id": "k0P0M1SKBkLe"
      },
      "execution_count": 32,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(15, 7))\n",
        "sns.set_theme(style=\"whitegrid\")\n",
        "ax = sns.barplot(x=\"diagnostic_label\", y=\"case\", data=train_co_occurency_df)\n",
        "for index, row in train_co_occurency_df.iterrows():\n",
        "    ax.text(row.name, row.case, round(row.case, 2),\n",
        "            color='black', ha='center')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 448
        },
        "id": "btlvC3Be-tU5",
        "outputId": "572515cf-669b-4812-8c82-e685786a55b3"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x504 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "plt.figure(figsize=(14, 7))\n",
        "sns.set_theme(style=\"ticks\")\n",
        "ax = sns.barplot(x=\"diagnostic_label\", y=\"%_freq\", data=train_co_occurency_df)\n",
        "for index, row in train_co_occurency_df.iterrows():\n",
        "    ax.text(row.name, row[\"%_freq\"], round(row[\"%_freq\"], 2),\n",
        "            color='black', ha='center')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 450
        },
        "id": "SIvuXzdBR7tM",
        "outputId": "15abf385-9ce1-4115-aef6-092f66324327"
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x504 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pd.options.display.float_format = '{:,.1f}'.format\n",
        "valid_co_occurency_df = valid_df.groupby(['diagnostic_label']).count().reset_index()\n",
        "valid_co_occurency_df['%_freq'] = (valid_co_occurency_df.case / valid_co_occurency_df.case.sum()) * 100\n",
        "valid_co_occurency_df.head()"
      ],
      "metadata": {
        "id": "71-riCRuPx10",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "8ccccc36-d351-4bfd-e926-20d06b4e4f22"
      },
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-f9eca216-3d86-4226-8384-ede605a5e70b\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>diagnostic_label</th>\n",
              "      <th>case</th>\n",
              "      <th>abnormal</th>\n",
              "      <th>acl</th>\n",
              "      <th>meniscus</th>\n",
              "      <th>%_freq</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Meniscus + Abnormal</td>\n",
              "      <td>21</td>\n",
              "      <td>21</td>\n",
              "      <td>21</td>\n",
              "      <td>21</td>\n",
              "      <td>17.5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Meniscus + Acl</td>\n",
              "      <td>23</td>\n",
              "      <td>23</td>\n",
              "      <td>23</td>\n",
              "      <td>23</td>\n",
              "      <td>19.2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Meniscus + Acl + Abnormal</td>\n",
              "      <td>31</td>\n",
              "      <td>31</td>\n",
              "      <td>31</td>\n",
              "      <td>31</td>\n",
              "      <td>25.8</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Meniscus Only</td>\n",
              "      <td>20</td>\n",
              "      <td>20</td>\n",
              "      <td>20</td>\n",
              "      <td>20</td>\n",
              "      <td>16.7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Nothing Detected</td>\n",
              "      <td>25</td>\n",
              "      <td>25</td>\n",
              "      <td>25</td>\n",
              "      <td>25</td>\n",
              "      <td>20.8</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f9eca216-3d86-4226-8384-ede605a5e70b')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-f9eca216-3d86-4226-8384-ede605a5e70b button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-f9eca216-3d86-4226-8384-ede605a5e70b');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "            diagnostic_label  case  abnormal  acl  meniscus  %_freq\n",
              "0        Meniscus + Abnormal    21        21   21        21    17.5\n",
              "1             Meniscus + Acl    23        23   23        23    19.2\n",
              "2  Meniscus + Acl + Abnormal    31        31   31        31    25.8\n",
              "3              Meniscus Only    20        20   20        20    16.7\n",
              "4           Nothing Detected    25        25   25        25    20.8"
            ]
          },
          "metadata": {},
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "valid_co_occurency_df = valid_co_occurency_df.drop(['abnormal', 'acl', 'meniscus'], axis = 1)"
      ],
      "metadata": {
        "id": "0w9TSuLPT7JJ"
      },
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "plt.figure(figsize=(14, 7))\n",
        "sns.set_theme(style=\"whitegrid\")\n",
        "ax = sns.barplot(x=\"diagnostic_label\", y=\"case\", data=valid_co_occurency_df)\n",
        "for index, row in valid_co_occurency_df.iterrows():\n",
        "    ax.text(row.name, row.case, round(row.case, 2),\n",
        "            color='black', ha='center',va= 'bottom')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 448
        },
        "outputId": "322cf79e-5468-4a03-c9b9-862251524e6c",
        "id": "nvgOa8e7T7JJ"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x504 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "plt.figure(figsize=(14, 7))\n",
        "sns.set_theme(style=\"ticks\")\n",
        "ax = sns.barplot(x=\"diagnostic_label\", y=\"%_freq\", data=valid_co_occurency_df)\n",
        "for index, row in valid_co_occurency_df.iterrows():\n",
        "    ax.text(row.name, row[\"%_freq\"], round(row[\"%_freq\"], 2),\n",
        "            color='black', ha='center')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 448
        },
        "outputId": "9a47d82f-b1a9-426c-e84a-1b4d1a00cb3c",
        "id": "anTAKigZT7JJ"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x504 with 1 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    }
  ]
}