Notebook_2_Classification_of_WBC_VGG16_Model.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4",
      "authorship_tag": "ABX9TyNKV1yS4rTb8J5pXwTe4Hrg",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/shu-nya/Image-Classification-of-White-Blood-Cells/blob/main/Notebook_2_Classification_of_WBC_VGG16_Model.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Notebook 2: VGG16 model\n",
        "\n",
        "### to be run after notebook 1"
      ],
      "metadata": {
        "id": "Szuh4lMafAUW"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## This notebook contains the VGG16 model.\n",
        "\n",
        "Getting an accuracy score around 75% on the test split."
      ],
      "metadata": {
        "id": "DqDMOQ8NfO1G"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### SMOTE is performed before training the model to create a balanced training dataset."
      ],
      "metadata": {
        "id": "lWkiNu_Pk5yz"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## After training the model, predictions are done on the classify-images folder dataset and the output is stored in a dataframe."
      ],
      "metadata": {
        "id": "-zoe1Sa6fdQQ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Future Work to be done\n",
        "\n",
        "1. Fine-tuning the VGG16 model."
      ],
      "metadata": {
        "id": "C69xuKeQhj71"
      }
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "4AUYsAl6fwMP"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Install and import the necessary libraries"
      ],
      "metadata": {
        "id": "fzG3XTsyBFw0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install imblearn\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ddpnOBjBFJT8",
        "outputId": "df7133dc-cf9d-4eae-d68c-9b5d1ea7570c"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting imblearn\n",
            "  Downloading imblearn-0.0-py2.py3-none-any.whl (1.9 kB)\n",
            "Requirement already satisfied: imbalanced-learn in /usr/local/lib/python3.10/dist-packages (from imblearn) (0.10.1)\n",
            "Requirement already satisfied: numpy>=1.17.3 in /usr/local/lib/python3.10/dist-packages (from imbalanced-learn->imblearn) (1.22.4)\n",
            "Requirement already satisfied: scipy>=1.3.2 in /usr/local/lib/python3.10/dist-packages (from imbalanced-learn->imblearn) (1.10.1)\n",
            "Requirement already satisfied: scikit-learn>=1.0.2 in /usr/local/lib/python3.10/dist-packages (from imbalanced-learn->imblearn) (1.2.2)\n",
            "Requirement already satisfied: joblib>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from imbalanced-learn->imblearn) (1.2.0)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from imbalanced-learn->imblearn) (3.1.0)\n",
            "Installing collected packages: imblearn\n",
            "Successfully installed imblearn-0.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "KBmEj0RR1MRP"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import cv2\n",
        "import os\n",
        "import shutil\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "from imblearn.over_sampling import SMOTE\n",
        "from collections import Counter\n",
        "\n",
        "from keras.models import Sequential\n",
        "from keras import applications\n",
        "from keras import optimizers\n",
        "from keras.layers import Dense, Dropout, Activation, Flatten\n",
        "from keras.layers import Input, Conv2D, MaxPooling2D, Lambda\n",
        "from keras.layers import Dense\n",
        "from keras.models import Model\n",
        "from keras.wrappers.scikit_learn import KerasClassifier\n",
        "from keras.utils import to_categorical\n",
        "from keras.utils import np_utils\n",
        "\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.metrics import confusion_matrix\n",
        "\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n",
        "\n",
        "%matplotlib inline\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "2w21WL8pB0nm"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Mount google drive to load the dataset"
      ],
      "metadata": {
        "id": "C0bKnRCYB0MO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# mount the drive\n",
        "\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "M3TTF4TQ1PE1",
        "outputId": "da3b8ab0-17d4-4a5f-a4c8-97ad4d6858c7"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# path to the base folder\n",
        "\n",
        "BASE_DIR = '/content/drive/MyDrive/taskMedPrime'\n"
      ],
      "metadata": {
        "id": "eu15MYI01PBw"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# number of classes into which the classifications are to be made\n",
        "num_classes = 9\n",
        "\n",
        "# classes into which the classifications are to be made\n",
        "classes = ['bands', 'basophils', 'blasts', 'eosinophils', 'lymphocytes', 'metamyelocytes', 'monocytes', 'myelocytes', 'neutrophils']\n"
      ],
      "metadata": {
        "id": "DFUrlXARCIav"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "m-Pr4_SgCaop"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Create the variables X and y"
      ],
      "metadata": {
        "id": "0s7a_xUHCaWS"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# method to get the images as independent variable X and their labels as the dependent variable y\n",
        "\n",
        "def get_data(folder):\n",
        "    X = []\n",
        "    y = []\n",
        "\n",
        "    # if the image files in the folder wbc are not empty then read the files\n",
        "    # and convert the images as numpy arrays\n",
        "    # append the image numpy arrays to the independent variable X\n",
        "    # and append their labels to y\n",
        "\n",
        "    for wbc_type in os.listdir(folder):\n",
        "        if not wbc_type.startswith('.'):\n",
        "\n",
        "            for image_filename in os.listdir(folder + '/' + wbc_type):\n",
        "                img_file = cv2.imread(folder + '/' + wbc_type + '/' + image_filename)\n",
        "\n",
        "                if img_file is not None:\n",
        "                    img_arr = np.asarray(img_file)\n",
        "                    X.append(img_arr)\n",
        "                    y.append(wbc_type)\n",
        "\n",
        "    # create numpy arrays of X and y\n",
        "\n",
        "    X = np.asarray(X)\n",
        "    y = np.asarray(y)\n",
        "\n",
        "    return X,y\n"
      ],
      "metadata": {
        "id": "Zhh87iJO1O7l"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# call the get_data method using the path to the train and test folders as arguments\n",
        "# to get the variables X_train, y_train, X_test and y_test\n",
        "\n",
        "X_train, y_train = get_data(BASE_DIR + '/train')\n",
        "X_test, y_test = get_data(BASE_DIR + '/test')\n"
      ],
      "metadata": {
        "id": "D1YQH6cq1O4s"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# normalize X_train and X_test\n",
        "# it will make images contribute more evenly to the total loss\n",
        "\n",
        "X_train = X_train * 1./255.\n",
        "X_test = X_test * 1./255.\n"
      ],
      "metadata": {
        "id": "XBbEPUEv1O17"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# summarize class distribution in the train dataset after augmentation\n",
        "\n",
        "counter = Counter(y_train)\n",
        "print(counter)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7A8AL_K265eR",
        "outputId": "97ca5550-1233-4783-a99a-82e0b6b11c4a"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Counter({'neutrophils': 225, 'lymphocytes': 209, 'monocytes': 81, 'bands': 78, 'eosinophils': 77, 'metamyelocytes': 77, 'myelocytes': 64, 'basophils': 63, 'blasts': 60})\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# plot the number of images in each sub-folder of the train dataset\n",
        "\n",
        "y_train_unique_counts = np.unique(y_train, return_counts=True)\n",
        "\n",
        "plt.title('Number of Images in each WBC type')\n",
        "plt.barh(y_train_unique_counts[0], y_train_unique_counts[1])\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 452
        },
        "id": "JvuficAI5MJ4",
        "outputId": "2e417aa3-560a-48ba-9f12-40e140c486d9"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "6XqSdKRDaY1y"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## SMOTE to treat the imbalance in the dataset"
      ],
      "metadata": {
        "id": "QbcAkmGfGr56"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# get all the dimensions of X_train\n",
        "\n",
        "n_samples, height, width, n_channels = [X_train.shape[index] for index in range(4)]\n",
        "print(f\"Shape of X_train before reshape: {X_train.shape}\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sHrUMT2VHDGz",
        "outputId": "79225d4c-acc1-4a4e-dfee-72b78ee7a9a1"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Shape of X_train before reshape: (934, 256, 256, 3)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# reshape X because SMOTE accepts only (n_samples, n_channels*height*weight) type data\n",
        "\n",
        "X_train_reshaped = X_train.reshape(n_samples, n_channels*height*width)\n",
        "print(f\"Shape of X_train after reshape: {X_train_reshaped.shape}\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OcM3HWYhHFxg",
        "outputId": "c24e5d6e-e7c7-4bc9-ab8f-65623ed0827f"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Shape of X_train after reshape: (934, 196608)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# initialize the SMOTE model\n",
        "\n",
        "smote = SMOTE(k_neighbors=5, random_state=62)\n"
      ],
      "metadata": {
        "id": "gM8-8ghXHyax"
      },
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# perform re-sampling on modified X given y of the train data\n",
        "\n",
        "X_smote, y_smote = smote.fit_resample(X_train_reshaped, y_train)\n"
      ],
      "metadata": {
        "id": "CRTnrJZQHyMo"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# number of images before and after applying SMOTE\n",
        "\n",
        "print(f'Before re-sampling, the number of images: {len(X_train):2}')\n",
        "print(f'After re-sampling, the number of images: {len(X_smote):2}')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XMwQvwWvI7-2",
        "outputId": "691eca9a-79ec-447c-c4fd-ae6612d098dc"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Before re-sampling, the number of images: 934\n",
            "After re-sampling, the number of images: 2025\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Observations:\n",
        "\n",
        "1. The number of samples in the train dataset has increased from 1158 to 2250 after applying SMOTE"
      ],
      "metadata": {
        "id": "iFS681VdY0Cc"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# summarize the new class distribution after applying SMOTE on the train data\n",
        "\n",
        "counter = Counter(y_smote)\n",
        "print(counter)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qd8GAHFaIwtm",
        "outputId": "ba1a0923-3857-4c92-d783-6b9c6176a65d"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Counter({'bands': 225, 'basophils': 225, 'blasts': 225, 'eosinophils': 225, 'lymphocytes': 225, 'metamyelocytes': 225, 'monocytes': 225, 'myelocytes': 225, 'neutrophils': 225})\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# view the resulting balanced data (modified distribution) after applying SMOTE\n",
        "\n",
        "y_smote_unique_counts = np.unique(y_smote, return_counts=True)\n",
        "print(y_smote_unique_counts)\n",
        "\n",
        "plt.title('Number of Images in each WBC type')\n",
        "plt.barh(y_smote_unique_counts[0], y_smote_unique_counts[1])\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 504
        },
        "id": "w-_2GUUilGgL",
        "outputId": "6df7aefa-e2e4-4c9d-d954-6a3b9992331f"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(array(['bands', 'basophils', 'blasts', 'eosinophils', 'lymphocytes',\n",
            "       'metamyelocytes', 'monocytes', 'myelocytes', 'neutrophils'],\n",
            "      dtype='<U14'), array([225, 225, 225, 225, 225, 225, 225, 225, 225]))\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAoEAAAGzCAYAAABQCc2DAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABdfUlEQVR4nO3deVxO+f8//sdV6Wq92lelopCUkKhmaNRIjLGHMci+JY3BaGaoMFO8MZax+0zZhrEMYze2LEnWbMXYUsg0lkohqvP7w6/znWtaZAZXdR732+263brOeZ3XeZ5zrvTwOsslEwRBABERERFJipqqCyAiIiKi948hkIiIiEiCGAKJiIiIJIghkIiIiEiCGAKJiIiIJIghkIiIiEiCGAKJiIiIJIghkIiIiEiCGAKJiIiIJIghkIjeWHx8PGQyGTZt2qTqUirlzz//RI8ePWBiYgKZTIa5c+equqQqTyaTITIyUtVl/CfV7XNK9L4xBBJVUXFxcZDJZNDS0sLdu3dLzff19UXjxo1VUFn188UXX2Dv3r0IDw/H6tWr0b59+3LbymQyhISEvMfqqCrasGEDZDIZtmzZUmpekyZNIJPJcOjQoVLz6tSpA29vb/G9vb09ZDKZ+NLS0oKTkxMmTJiAR48elbnuLVu2IDAwEKamptDU1IS1tTWCgoJw8ODBCmt++vQpIiMjER8f/2YbS5KloeoCiKhiBQUFiImJwYIFC1RdSrV18OBBdO7cGePHj1d1KdXGs2fPoKEh3T8RH3zwAQDg2LFj6Nq1qzg9NzcXly5dgoaGBhISEvDRRx+J8zIyMpCRkYHevXsr9eXu7o4vv/wSAPD8+XOcOXMGc+fOxeHDh3Hy5EmxnSAIGDRoEOLi4tC0aVOMGzcOlpaWyMzMxJYtW+Dn54eEhASlkPl3T58+RVRUFIBX/0kkeh3p/oYTVRPu7u5Yvnw5wsPDYW1trepy3qv8/Hzo6ur+536ysrJgaGj43wuSEC0tLVWXoFLW1tZwcHDAsWPHlKYnJiZCEAT07Nmz1LyS9yUBskTt2rXx+eefi++HDBkCPT09zJo1C9euXYOTkxMAYPbs2YiLi0NYWBjmzJkDmUwmLvPNN99g9erVkg7m9PbxdDBRFff111+jqKgIMTExFbZLS0uDTCZDXFxcqXn/vL4rMjISMpkMf/zxBz7//HMYGBjAzMwMkydPhiAIyMjIQOfOnaFQKGBpaYnZs2eXuc6ioiJ8/fXXsLS0hK6uLj799FNkZGSUapeUlIT27dvDwMAAOjo6aNOmDRISEpTalNSUkpKCzz77DEZGRqX+mP7TzZs30bNnTxgbG0NHRwetWrXCzp07xfklp9QFQcDChQvFU3JvouS6sg0bNiAqKgq1a9eGvr4+evTogZycHBQUFCAsLAzm5ubQ09PDwIEDUVBQoNRHbGws2rZtC3Nzc8jlcjRq1AiLFy8uta7i4mJERkbC2toaOjo6+Oijj5CSkgJ7e3sEBwcrtc3OzkZYWBhsbW0hl8vh6OiIGTNmoLi4WKnd+vXr0bx5c+jr60OhUMDV1RXz5s177XaX95m5fv06goODYWhoCAMDAwwcOBBPnz6t1L6szOfg9u3bGDVqFBo0aABtbW2YmJigZ8+eSEtLK9VfdnY2vvjiC9jb20Mul8PGxgb9+/fHgwcPlNoVFxfju+++g42NDbS0tODn54fr16+/tt4PPvgA586dw7Nnz8RpCQkJcHFxQWBgIE6cOKG0vxMSEiCTyeDj4/Pavi0tLQFADHXPnj1DdHQ0GjZsiFmzZpX5Oe3Xrx88PT3L7C8tLQ1mZmYAgKioKPGzHhkZidjYWMhkMpw7d67Uct9//z3U1dXFS05KLjM5c+YMvL29oa2tDQcHByxZsqTUsgUFBYiIiICjoyPkcjlsbW0xceLEUp9/qrr4XwqiKs7BwQH9+/fH8uXLMWnSpLc6GtirVy84OzsjJiYGO3fuxPTp02FsbIylS5eibdu2mDFjBtauXYvx48ejRYsWaN26tdLy3333HWQyGb766itkZWVh7ty58Pf3R3JyMrS1tQG8OhUbGBiI5s2bIyIiAmpqamIoOnr0aKk/aj179oSTkxO+//57CIJQbu1//vknvL298fTpU4SGhsLExAQrV67Ep59+ik2bNqFr165o3bo1Vq9ejX79+uHjjz9G//79//W+io6Ohra2NiZNmoTr169jwYIFqFWrFtTU1PD48WNERkbixIkTiIuLg4ODA6ZMmSIuu3jxYri4uODTTz+FhoYGtm/fjlGjRqG4uBijR48W24WHh2PmzJno1KkTAgICcP78eQQEBOD58+dKtTx9+hRt2rTB3bt3MXz4cNSpUwfHjx9HeHg4MjMzxRtf9u3bhz59+sDPzw8zZswAAKSmpiIhIQFjx479V/shKCgIDg4OiI6OxtmzZ7FixQqYm5uL/Zensp+DU6dO4fjx4+jduzdsbGyQlpaGxYsXw9fXFykpKdDR0QEA5OXl4cMPP0RqaioGDRqEZs2a4cGDB9i2bRvu3LkDU1NTcd0xMTFQU1PD+PHjkZOTg5kzZ6Jv375ISkqqsOYPPvgAq1evRlJSknh6teR0rLe3N3JycnDp0iW4ubmJ8xo2bAgTExOlfl6+fCkG0+fPn+PcuXOYM2cOWrduDQcHBwCvRhEfPXqEsLAwqKurV/Jo/D9mZmZYvHgxRo4cia5du6Jbt24AADc3Nzg4OGD06NFYu3YtmjZtqrTc2rVr4evri9q1a4vTHj9+jA4dOiAoKAh9+vTBhg0bMHLkSGhqamLQoEEAXgXrTz/9FMeOHcOwYcPg7OyMixcv4ocffsAff/yBrVu3vvE2kAoIRFQlxcbGCgCEU6dOCTdu3BA0NDSE0NBQcX6bNm0EFxcX8f2tW7cEAEJsbGypvgAIERER4vuIiAgBgDBs2DBxWmFhoWBjYyPIZDIhJiZGnP748WNBW1tbGDBggDjt0KFDAgChdu3aQm5urjh9w4YNAgBh3rx5giAIQnFxseDk5CQEBAQIxcXFYrunT58KDg4Owscff1yqpj59+lRq/4SFhQkAhKNHj4rTnjx5Ijg4OAj29vZCUVGR0vaPHj26Uv3+s23JtjZu3Fh48eKFOL1Pnz6CTCYTAgMDlZb38vIS7OzslKY9ffq01HoCAgKEunXriu/v378vaGhoCF26dFFqFxkZKQBQ2v/Tpk0TdHV1hT/++EOp7aRJkwR1dXUhPT1dEARBGDt2rKBQKITCwsJKbfvflfeZGTRokFK7rl27CiYmJhX29Safg7L2VWJiogBAWLVqlThtypQpAgDh119/LXN9gvD/jp2zs7NQUFAgzp83b54AQLh48WKFdV++fFkAIEybNk0QBEF4+fKloKurK6xcuVIQBEGwsLAQFi5cKAiCIOTm5grq6urC0KFDlfqws7MTAJR6+fj4CA8ePChV05YtWyqsqSJ//fVXqeNWok+fPoK1tbXS78XZs2dL/ZvRpk0bAYAwe/ZscVpBQYHg7u4umJubi78Dq1evFtTU1JR+/wRBEJYsWSIAEBISEv71dtD7w9PBRNVA3bp10a9fPyxbtgyZmZlvrd8hQ4aIP6urq8PDwwOCIGDw4MHidENDQzRo0AA3b94stXz//v2hr68vvu/RowesrKywa9cuAEBycjKuXbuGzz77DA8fPsSDBw/w4MED5Ofnw8/PD0eOHCl1+nLEiBGVqn3Xrl3w9PRUOmWsp6eHYcOGIS0tDSkpKZXbCZXUv39/1KpVS3zfsmVL8UL+v2vZsiUyMjJQWFgoTisZFQWAnJwcPHjwAG3atMHNmzeRk5MDADhw4AAKCwsxatQopf7GjBlTqpaNGzfiww8/hJGRkbhPHzx4AH9/fxQVFeHIkSMAXh27/Px87Nu377/vgP/fP4/Phx9+iIcPHyI3N7fcZd7kc/D3ffXy5Us8fPgQjo6OMDQ0xNmzZ8V5mzdvRpMmTZRu2ijxz1OpAwcOhKamplLNAMr8TP+ds7MzTExMxGv9zp8/j/z8fPHGDG9vb/F0dmJiIoqKisq8hKFly5bYt28f9u3bhx07duC7777D5cuX8emnn4qnmkv2399/n96m/v374969e0p3NK9duxba2tro3r27UlsNDQ0MHz5cfK+pqYnhw4cjKysLZ86cAfDqM+js7IyGDRsqfQbbtm0LAGXeOU1VD08HE1UT3377LVavXo2YmJhKXdNVGXXq1FF6b2BgAC0tLaVTaSXTHz58WGr5kgvaS8hkMjg6OorXb127dg0AMGDAgHJryMnJgZGRkfi+5PTY69y+fRstW7YsNd3Z2Vmc/zYfoVPWvgIAW1vbUtOLi4uRk5MjnhZMSEhAREQEEhMTS10/l5OTAwMDA9y+fRsA4OjoqDTf2NhYaf8Ar/brhQsXxGvA/ikrKwsAMGrUKGzYsAGBgYGoXbs22rVrh6CgoAofkfM6/9wPJbU9fvwYCoWizGXe5HNQcm1cbGws7t69q3RJQElgBoAbN26UCi//puaKyGQyeHt7iyE1ISEB5ubm4jHy9vbGjz/+CABiGCwrBJqamsLf319837FjRzRo0AA9evTAihUrMGbMGHHfPXnypFLb9KY+/vhjWFlZYe3atfDz80NxcTHWrVuHzp07lwqe1tbWpW7Iql+/PoBX1x62atUK165dQ2pq6ms/g1S1MQQSVRN169bF559/jmXLlmHSpEml5pd3w0NRUVG5fZZ17VF51yMJFVyfV56S0Z3//e9/cHd3L7ONnp6e0vu/jwRVJeXtl9ftrxs3bsDPzw8NGzbEnDlzYGtrC01NTezatQs//PBDqZHQyiguLsbHH3+MiRMnljm/5A+2ubk5kpOTsXfvXuzevRu7d+9GbGws+vfvj5UrV77xeoF/9/l4k8/BmDFjEBsbi7CwMHh5ecHAwAAymQy9e/f+V/vq39Zc4oMPPsD27dtx8eLFUo9n8fb2xoQJE3D37l0cO3YM1tbWqFu3bqVq8vPzAwAcOXIEY8aMQcOGDQEAFy9eRJcuXSrVx5tQV1fHZ599huXLl2PRokVISEjAvXv3lO5afhPFxcVwdXXFnDlzypz/z/8cUdXEEEhUjXz77bdYs2ZNmRfhl4xuZGdnK00vGWF6F0pGeEoIgoDr16+LF8rXq1cPAKBQKJRGQt4GOzs7XL16tdT0K1euiPOrgu3bt6OgoADbtm1TGpH65+myknqvX7+uNBr68OHDUiNW9erVQ15eXqX2qaamJjp16oROnTqhuLgYo0aNwtKlSzF58uRSo47vypt8DjZt2oQBAwYo3ZH+/PnzUp/revXq4dKlS2+91n/6+/MCExISEBYWJs5r3rw55HI54uPjkZSUhA4dOlS635LLBfLy8sT1GBkZYd26dfj666//1c0hr7vzvX///pg9eza2b9+O3bt3w8zMDAEBAaXa3bt3r9Tjmf744w8Arx5+Dbza/+fPn4efn98b33FPVQevCSSqRurVq4fPP/8cS5cuxf3795XmKRQKmJqaiteDlVi0aNE7q2fVqlVKp682bdqEzMxMBAYGAnj1R7JevXqYNWuW+Mfu7/76669/ve4OHTrg5MmTSExMFKfl5+dj2bJlsLe3R6NGjf51329TyR/zf57WjI2NVWrn5+cHDQ2NUo+OKTnd+HdBQUFITEzE3r17S83Lzs4WA8Y/T+GrqamJAf19PsbjTT4H6urqpUboFixYUGpEu3v37jh//nyZ3+jxb0aty+Ph4QEtLS2sXbsWd+/eVRoJlMvlaNasGRYuXIj8/PzXPtLo77Zv3w7g1bePAICOjg6++uorpKam4quvvipzG9asWaP0cOl/Krlz+p+BuYSbmxvc3NywYsUKbN68Gb179y7zuYOFhYVYunSp+P7FixdYunQpzMzM0Lx5cwCvPoN3797F8uXLSy3/7Nkz5Ofnl1snVR0cCSSqZkoeGnv16lW4uLgozRsyZAhiYmIwZMgQeHh44MiRI+L/4N8FY2NjfPDBBxg4cCD+/PNPzJ07F46Ojhg6dCiAV6FjxYoVCAwMhIuLCwYOHIjatWvj7t27OHToEBQKhfjH8E1NmjQJ69atQ2BgIEJDQ2FsbIyVK1fi1q1b2Lx5M9TUqsb/cdu1ayeOxg0fPhx5eXlYvnw5zM3NlW7ysbCwwNixYzF79mx8+umnaN++Pc6fP4/du3fD1NRUabRlwoQJ2LZtGz755BMEBwejefPmyM/Px8WLF7Fp0yakpaXB1NQUQ4YMwaNHj9C2bVvY2Njg9u3bWLBgAdzd3cVrJ9+HN/kcfPLJJ1i9ejUMDAzQqFEjJCYmYv/+/aUeuzJhwgRs2rQJPXv2xKBBg9C8eXM8evQI27Ztw5IlS8Rw9V9pamqiRYsWOHr0KORyuRiCSnh7e4ujluWFwLt372LNmjUAXgWq8+fPY+nSpTA1NVW68WfChAm4fPkyZs+ejUOHDqFHjx6wtLTE/fv3sXXrVpw8eRLHjx8vt1ZtbW00atQIv/zyC+rXrw9jY2M0btxY6drY/v37i9+cU96pYGtra8yYMQNpaWmoX78+fvnlFyQnJ2PZsmXizVH9+vXDhg0bMGLECBw6dAg+Pj4oKirClStXsGHDBuzduxceHh6v272kaqq5KZmIXufvj4j5pwEDBggAlB4RIwivHq8xePBgwcDAQNDX1xeCgoKErKysch/38ddff5XqV1dXt9T6/vk4mpJHb6xbt04IDw8XzM3NBW1tbaFjx47C7du3Sy1/7tw5oVu3boKJiYkgl8sFOzs7ISgoSDhw4MBra6rIjRs3hB49egiGhoaClpaW4OnpKezYsaNUO7yFR8Rs3LhRqV15x6es7di2bZvg5uYmaGlpCfb29sKMGTOEn376SQAg3Lp1S2xXWFgoTJ48WbC0tBS0tbWFtm3bCqmpqYKJiYkwYsQIpfU8efJECA8PFxwdHQVNTU3B1NRU8Pb2FmbNmiU+xmPTpk1Cu3btBHNzc0FTU1OoU6eOMHz4cCEzM7NS+6Eyn5mS/fD37ShPZT4Hjx8/FgYOHCiYmpoKenp6QkBAgHDlyhXBzs5O6TE5giAIDx8+FEJCQoTatWsLmpqago2NjTBgwADx0SvlHbuKHqdUlvDwcAGA4O3tXWrer7/+KgAQ9PX1y3wUzz8fEaOmpiaYm5sLffr0Ea5fv17m+kqOm7GxsaChoSFYWVkJvXr1EuLj419b6/Hjx4XmzZsLmpqaZT4uJjMzU1BXVxfq169f5vIlv+unT58WvLy8BC0tLcHOzk748ccfS7V98eKFMGPGDMHFxUWQy+WCkZGR0Lx5cyEqKkrIycl5ba2kejJBeIvj5kRE9FZlZ2fDyMgI06dPxzfffKPqcqiae/DgAaysrDBlyhRMnjy51HxfX188ePDgvVxvSapXNc6XEBGR0teTlSj59o+Sb6wg+i/i4uJQVFSEfv36qboUqgJ4TSARURXxyy+/IC4uDh06dICenh6OHTuGdevWoV27dpX6Plqi8hw8eBApKSn47rvv0KVLF/EuX5I2hkAioirCzc0NGhoamDlzJnJzc8WbRaZPn67q0qiamzp1Ko4fPw4fHx8sWLBA1eVQFcFrAomIiIgkiNcEEhEREUkQQyARERGRBPGaQCpTcXEx7t27B319fX4lEBERUTUhCAKePHkCa2vr1z40nyGQynTv3j1+ATgREVE1lZGRARsbmwrbMARSmfT19QG8+hApFAoVV0NERESVkZubC1tbW/HveEUYAqlMJaeAFQoFQyAREVE1U5lLuXhjCBEREZEEMQQSERERSRBDIBEREZEEMQQSERERSRBDIBEREZEEMQQSERERSRBDIBEREZEEMQQSERERSRBDIBEREZEEMQQSERERSRBDIBEREZEEMQQSERERSRBDIBEREZEEaai6AKraGkfshZpcR9VlEBER1ShpMR1VXQJHAomIiIikiCGQiIiISIIYAomIiIgkiCGQiIiISIIYAomIiIgkiCGQiIiISIIYAomIiIgkiCGQiIiISIIYAomIiIgkiCGQiIiISIIYAquouLg4GBoaVtgmODgYXbp0Ed/7+voiLCzsndZFRERENQO/O7iSfH194e7ujrlz56q6FNG8efMgCIKqyyAiIqJqiCHwLRIEAUVFRdDQeD+71cDA4L2sh4iIiGqeGnE62NfXF6GhoZg4cSKMjY1haWmJyMhIcX52djaGDBkCMzMzKBQKtG3bFufPnxfn//O0KgCEhYXB19dXnH/48GHMmzcPMpkMMpkMaWlpiI+Ph0wmw+7du9G8eXPI5XIcO3YMBQUFCA0Nhbm5ObS0tPDBBx/g1KlTYt8ly+3cuRNubm7Q0tJCq1atcOnSpVLbtnfvXjg7O0NPTw/t27dHZmZmhXX/3aJFi+Dk5AQtLS1YWFigR48eb7ZjiYiIqMaqESEQAFauXAldXV0kJSVh5syZmDp1Kvbt2wcA6NmzJ7KysrB7926cOXMGzZo1g5+fHx49elSpvufNmwcvLy8MHToUmZmZyMzMhK2trTh/0qRJiImJQWpqKtzc3DBx4kRs3rwZK1euxNmzZ+Ho6IiAgIBS65swYQJmz56NU6dOwczMDJ06dcLLly/F+U+fPsWsWbOwevVqHDlyBOnp6Rg/fnylaj59+jRCQ0MxdepUXL16FXv27EHr1q3LbV9QUIDc3FylFxEREdVcNeZ0sJubGyIiIgAATk5O+PHHH3HgwAFoa2vj5MmTyMrKglwuBwDMmjULW7duxaZNmzBs2LDX9m1gYABNTU3o6OjA0tKy1PypU6fi448/BgDk5+dj8eLFiIuLQ2BgIABg+fLl2LdvH/7v//4PEyZMEJeLiIgQl1u5ciVsbGywZcsWBAUFAQBevnyJJUuWoF69egCAkJAQTJ06tVL7Iz09Hbq6uvjkk0+gr68POzs7NG3atNz20dHRiIqKqlTfREREVP3VmJFANzc3pfdWVlbIysrC+fPnkZeXBxMTE+jp6YmvW7du4caNG29l3R4eHuLPN27cwMuXL+Hj4yNOq1WrFjw9PZGamqq0nJeXl/izsbExGjRooNRGR0dHDIB/36bK+Pjjj2FnZ4e6deuiX79+WLt2LZ4+fVpu+/DwcOTk5IivjIyMSq2HiIiIqqcaMxJYq1YtpfcymQzFxcXIy8uDlZUV4uPjSy1T8ggWNTW1UnfZ/v207Ovo6uq+cb2VUdY2VfZuYH19fZw9exbx8fH4/fffMWXKFERGRuLUqVNlPnpGLpeLI6VERERU89WYkcDyNGvWDPfv34eGhgYcHR2VXqampgAAMzMzpRsuACA5OVnpvaamJoqKil67vnr16kFTUxMJCQnitJcvX+LUqVNo1KiRUtsTJ06IPz9+/Bh//PEHnJ2d33QTy6WhoQF/f3/MnDkTFy5cQFpaGg4ePPjW+iciIqLqq8aMBJbH398fXl5e6NKlC2bOnIn69evj3r172LlzJ7p27QoPDw+0bdsW//vf/7Bq1Sp4eXlhzZo1uHTpktI1dPb29khKSkJaWhr09PRgbGxc5vp0dXUxcuRITJgwAcbGxqhTpw5mzpyJp0+fYvDgwUptp06dChMTE1hYWOCbb76BqalphXf7vokdO3bg5s2baN26NYyMjLBr1y4UFxejQYMGb6V/IiIiqt5q/EigTCbDrl270Lp1awwcOBD169dH7969cfv2bVhYWAAAAgICMHnyZEycOBEtWrTAkydP0L9/f6V+xo8fD3V1dTRq1AhmZmZIT08vd50xMTHo3r07+vXrh2bNmuH69evYu3cvjIyMSrUbO3Ysmjdvjvv372P79u3Q1NR8K9ttaGiIX3/9FW3btoWzszOWLFmCdevWwcXF5a30T0RERNWbTOBXTrx38fHx+Oijj/D48ePXfjWcquTm5sLAwAC2YRugJtdRdTlEREQ1SlpMx3fSb8nf75ycHCgUigrb1viRQCIiIiIqjSGQiIiISIJq/I0hVZGvr2+lH/VCRERE9C5wJJCIiIhIghgCiYiIiCSIIZCIiIhIghgCiYiIiCSIIZCIiIhIgnh3MFXoUlTAax82SURERNUPRwKJiIiIJIghkIiIiEiCGAKJiIiIJIghkIiIiEiCGAKJiIiIJIghkIiIiEiCGAKJiIiIJIjPCaQKNY7YCzW5jqrLICIiqlHSYjqqugSOBBIRERFJEUMgERERkQQxBBIRERFJEEMgERERkQQxBBIRERFJEEMgERERkQQxBBIRERFJEEMgERERkQQxBBIRERFJEEPgG4qMjIS7u7uqyyAiIiL6TxgCq7j4+HjIZDJkZ2eruhQiIiKqQRgCiYiIiCSoWodAX19fjBkzBmFhYTAyMoKFhQWWL1+O/Px8DBw4EPr6+nB0dMTu3bshCAIcHR0xa9YspT6Sk5Mhk8lw/fp1AEB2djaGDBkCMzMzKBQKtG3bFufPny+3huLiYkydOhU2NjaQy+Vwd3fHnj17lNrcuXMHffr0gbGxMXR1deHh4YGkpCSkpaVBTU0Np0+fVmo/d+5c2NnZ4ebNm/joo48AAEZGRpDJZAgODhbXGx0dDQcHB2hra6NJkybYtGmT2Mfjx4/Rt29fmJmZQVtbG05OToiNjf3X+5qIiIhqlmodAgFg5cqVMDU1xcmTJzFmzBiMHDkSPXv2hLe3N86ePYt27dqhX79+ePbsGQYNGlQqCMXGxqJ169ZwdHQEAPTs2RNZWVnYvXs3zpw5g2bNmsHPzw+PHj0qc/3z5s3D7NmzMWvWLFy4cAEBAQH49NNPce3aNQBAXl4e2rRpg7t372Lbtm04f/48Jk6ciOLiYtjb28Pf37/MmoKDg2FnZ4fNmzcDAK5evYrMzEzMmzcPABAdHY1Vq1ZhyZIluHz5Mr744gt8/vnnOHz4MABg8uTJSElJwe7du5GamorFixfD1NS03P1YUFCA3NxcpRcRERHVXDJBEARVF/Fv+fr6oqioCEePHgUAFBUVwcDAAN26dcOqVasAAPfv34eVlRUSExNRp04d1KlTB8ePH4enpydevnwJa2trzJo1CwMGDMCxY8fQsWNHZGVlQS6Xi+txdHTExIkTMWzYMERGRmLr1q1ITk4GANSuXRujR4/G119/Lbb39PREixYtsHDhQixbtgzjx49HWloajI2NS23Dhg0bMGLECGRmZkIul+Ps2bPw8PDAzZs3YW9vj/j4eHz00Ud4/PgxDA0NAbwKbMbGxti/fz+8vLzEvoYMGYKnT5/i559/xqeffgpTU1P89NNPldqXkZGRiIqKKjXdNmwD1OQ6leqDiIiIKictpuM76Tc3NxcGBgbIycmBQqGosG21Hwl0c3MTf1ZXV4eJiQlcXV3FaRYWFgCArKwsWFtbo2PHjmIw2r59OwoKCtCzZ08AwPnz55GXlwcTExPo6emJr1u3buHGjRul1p2bm4t79+7Bx8dHabqPjw9SU1MBvDrd3LRp0zIDIAB06dIF6urq2LJlCwAgLi4OH330Eezt7cvd5uvXr+Pp06f4+OOPlepctWqVWOfIkSOxfv16uLu7Y+LEiTh+/HiF+zE8PBw5OTniKyMjo8L2REREVL1pqLqA/6pWrVpK72UymdI0mUwG4NU1dMCr0bJ+/frhhx9+QGxsLHr16gUdnVcjXXl5ebCyskJ8fHyp9ZSMwr0pbW3tCudramqif//+iI2NRbdu3fDzzz+Lp3zLk5eXBwDYuXMnateurTSvZAQzMDAQt2/fxq5du7Bv3z74+flh9OjRpa6J/Ptyfx/9JCIiopqt2ofAN9WhQwfo6upi8eLF2LNnD44cOSLOa9asGe7fvw8NDY0KR+JKKBQKWFtbIyEhAW3atBGnJyQkwNPTE8CrkcoVK1bg0aNH5Y4GDhkyBI0bN8aiRYtQWFiIbt26ifM0NTUBvDrVXaJRo0aQy+VIT09XWu8/mZmZYcCAARgwYAA+/PBDTJgwodwQSERERNIiuRCorq6O4OBghIeHw8nJSemaOn9/f3h5eaFLly6YOXMm6tevj3v37mHnzp3o2rUrPDw8SvU3YcIEREREoF69enB3d0dsbCySk5Oxdu1aAECfPn3w/fffo0uXLoiOjoaVlRXOnTsHa2trcd3Ozs5o1aoVvvrqKwwaNEhp9NDOzg4ymQw7duxAhw4doK2tDX19fYwfPx5ffPEFiouL8cEHHyAnJwcJCQlQKBQYMGAApkyZgubNm8PFxQUFBQXYsWMHnJ2d3/HeJSIiouqi2l8T+G8MHjwYL168wMCBA5Wmy2Qy7Nq1C61bt8bAgQNRv3599O7dG7dv3xavLfyn0NBQjBs3Dl9++SVcXV2xZ88ebNu2DU5OTgBejeT9/vvvMDc3R4cOHeDq6oqYmBioq6uXWdOgQYOUpteuXRtRUVGYNGkSLCwsEBISAgCYNm0aJk+ejOjoaDg7O6N9+/bYuXMnHBwcxPWGh4fDzc0NrVu3hrq6OtavX/9W9h8RERFVf9X67uB/6+jRo/Dz80NGRka54e59mzZtGjZu3IgLFy6ouhQA/+/uIt4dTERE9PZVhbuDJXU6uKCgAH/99RciIyPRs2fPKhEA8/LykJaWhh9//BHTp09XdTlEREQkEZI6Hbxu3TrY2dkhOzsbM2fOVHU5AICQkBA0b94cvr6+pU4FExEREb0rkjwdTK/H08FERETvTlU4HSypkUAiIiIieoUhkIiIiEiCGAKJiIiIJIghkIiIiEiCGAKJiIiIJEhSzwmkN3cpKuC1dxcRERFR9cORQCIiIiIJYggkIiIikiCGQCIiIiIJYggkIiIikiCGQCIiIiIJYggkIiIikiCGQCIiIiIJ4nMCqUKNI/ZCTa6j6jKIiIhqlLSYjqougSOBRERERFLEEEhEREQkQQyBRERERBLEEEhEREQkQQyBRERERBLEEEhEREQkQQyBRERERBLEEEhEREQkQQyBRERERBLEEEhEREQkQQyB1UBwcDC6dOmi6jKIiIioBmEIJCIiIpKgGhUCfX19MWbMGISFhcHIyAgWFhZYvnw58vPzMXDgQOjr68PR0RG7d+8Wlzl8+DA8PT0hl8thZWWFSZMmobCwUKnP0NBQTJw4EcbGxrC0tERkZKTSetPT09G5c2fo6elBoVAgKCgIf/75p1Kb7du3o0WLFtDS0oKpqSm6du0KAJg6dSoaN25calvc3d0xefJkREZGYuXKlfjtt98gk8kgk8kQHx8PAMjIyEBQUBAMDQ1hbGyMzp07Iy0tTewjPj4enp6e0NXVhaGhIXx8fHD79u3/uJeJiIioJqhRIRAAVq5cCVNTU5w8eRJjxozByJEj0bNnT3h7e+Ps2bNo164d+vXrh6dPn+Lu3bvo0KEDWrRogfPnz2Px4sX4v//7P0yfPr1Un7q6ukhKSsLMmTMxdepU7Nu3DwBQXFyMzp0749GjRzh8+DD27duHmzdvolevXuLyO3fuRNeuXdGhQwecO3cOBw4cgKenJwBg0KBBSE1NxalTp8T2586dw4ULFzBw4ECMHz8eQUFBaN++PTIzM5GZmQlvb2+8fPkSAQEB0NfXx9GjR5GQkAA9PT20b98eL168QGFhIbp06YI2bdrgwoULSExMxLBhwyCTycrcbwUFBcjNzVV6ERERUc0lEwRBUHURb4uvry+Kiopw9OhRAEBRUREMDAzQrVs3rFq1CgBw//59WFlZITExEdu3b8fmzZuRmpoqhqNFixbhq6++Qk5ODtTU1Er1CQCenp5o27YtYmJisG/fPgQGBuLWrVuwtbUFAKSkpMDFxQUnT55EixYt4O3tjbp162LNmjVl1t2hQwfY29tj0aJFAIDQ0FBcvHgRhw4dAvDqmsDs7Gxs3bpVXGbNmjWYPn26Uu0vXryAoaEhtm7dCg8PD5iYmCA+Ph5t2rR57b6LjIxEVFRUqem2YRugJtd57fJERERUeWkxHd9Jv7m5uTAwMEBOTg4UCkWFbWvcSKCbm5v4s7q6OkxMTODq6ipOs7CwAABkZWUhNTUVXl5eSqNjPj4+yMvLw507d8rsEwCsrKyQlZUFAEhNTYWtra0YAAGgUaNGMDQ0RGpqKgAgOTkZfn5+5dY8dOhQrFu3Ds+fP8eLFy/w888/Y9CgQRVu5/nz53H9+nXo6+tDT08Penp6MDY2xvPnz3Hjxg0YGxsjODgYAQEB6NSpE+bNm4fMzMxy+wsPD0dOTo74ysjIqHD9REREVL1pqLqAt61WrVpK72UymdK0ksBXXFz8n/p8k+W1tbUrnN+pUyfI5XJs2bIFmpqaePnyJXr06FHhMnl5eWjevDnWrl1bap6ZmRkAIDY2FqGhodizZw9++eUXfPvtt9i3bx9atWpVahm5XA65XF7pbSIiIqLqrcaNBL4JZ2dnJCYm4u9nxBMSEqCvrw8bG5tK95GRkaE0cpaSkoLs7Gw0atQIwKuRxAMHDpTbh4aGBgYMGIDY2FjExsaid+/eSsFRU1MTRUVFSss0a9YM165dg7m5ORwdHZVeBgYGYrumTZsiPDwcx48fR+PGjfHzzz9XaruIiIioZpN0CBw1ahQyMjIwZswYXLlyBb/99hsiIiIwbtw4qKlVbtf4+/vD1dUVffv2xdmzZ3Hy5En0798fbdq0gYeHBwAgIiIC69atQ0REBFJTU3Hx4kXMmDFDqZ8hQ4bg4MGD2LNnT6lTwfb29rhw4QKuXr2KBw8e4OXLl+jbty9MTU3RuXNnHD16FLdu3UJ8fDxCQ0Nx584d3Lp1C+Hh4UhMTMTt27fx+++/49q1a3B2dn47O4+IiIiqNUmHwNq1a2PXrl04efIkmjRpghEjRmDw4MH49ttvK92HTCbDb7/9BiMjI7Ru3Rr+/v6oW7cufvnlF7GNr68vNm7ciG3btsHd3R1t27bFyZMnlfpxcnKCt7c3GjZsiJYtWyrNGzp0KBo0aAAPDw+YmZkhISEBOjo6OHLkCOrUqYNu3brB2dkZgwcPxvPnz6FQKKCjo4MrV66ge/fuqF+/PoYNG4bRo0dj+PDh/22nERERUY1Qo+4Ors4EQYCTkxNGjRqFcePGqboc8e4i3h1MRET09lWFu4Nr3I0h1dFff/2F9evX4/79+xg4cKCqyyEiIiIJYAisAszNzWFqaoply5bByMhI1eUQERGRBDAEVgE8I09ERETvm6RvDCEiIiKSKoZAIiIiIgliCCQiIiKSIIZAIiIiIgliCCQiIiKSIN4dTBW6FBXw2odNEhERUfXDkUAiIiIiCWIIJCIiIpIghkAiIiIiCWIIJCIiIpIghkAiIiIiCWIIJCIiIpIghkAiIiIiCeJzAqlCjSP2Qk2uo+oyiIiIapS0mI6qLoEjgURERERSxBBIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBD4huLi4mBoaKjqMoiIiIj+E5WHQF9fX4SFham6jCorLS0NMpkMycnJqi6FiIiIahCVh0AiIiIiev/eKAT6+vpizJgxCAsLg5GRESwsLLB8+XLk5+dj4MCB0NfXh6OjI3bv3i0uc+nSJQQGBkJPTw8WFhbo168fHjx4AAAIDg7G4cOHMW/ePMhkMshkMqSlpaGoqAiDBw+Gg4MDtLW10aBBA8ybN0+pluDgYHTp0gXff/89LCwsYGhoiKlTp6KwsBATJkyAsbExbGxsEBsbKy7Ttm1bhISEKPXz119/QVNTEwcOHAAAFBQUYPz48ahduzZ0dXXRsmVLxMfHV7hfFi9ejHr16kFTUxMNGjTA6tWrleZnZ2dj+PDhsLCwgJaWFho3bowdO3YgPz8fCoUCmzZtUmq/detW6Orq4smTJ3BwcAAANG3aFDKZDL6+vmK7FStWwNnZGVpaWmjYsCEWLVokznvx4gVCQkJgZWUFLS0t2NnZITo6usLtICIiIul445HAlStXwtTUFCdPnsSYMWMwcuRI9OzZE97e3jh79izatWuHfv364enTp8jOzkbbtm3RtGlTnD59Gnv27MGff/6JoKAgAMC8efPg5eWFoUOHIjMzE5mZmbC1tUVxcTFsbGywceNGpKSkYMqUKfj666+xYcMGpVoOHjyIe/fu4ciRI5gzZw4iIiLwySefwMjICElJSRgxYgSGDx+OO3fuAACGDBmCn3/+GQUFBWIfa9asQe3atdG2bVsAQEhICBITE7F+/XpcuHABPXv2RPv27XHt2rUy98eWLVswduxYfPnll7h06RKGDx+OgQMH4tChQwCA4uJiBAYGIiEhAWvWrEFKSgpiYmKgrq4OXV1d9O7dWymoAkBsbCx69OgBfX19nDx5EgCwf/9+ZGZm4tdffwUArF27FlOmTMF3332H1NRUfP/995g8eTJWrlwJAJg/fz62bduGDRs24OrVq1i7di3s7e3LPa4FBQXIzc1VehEREVHNJRMEQahsY19fXxQVFeHo0aMAgKKiIhgYGKBbt25YtWoVAOD+/fuwsrJCYmIi9u/fj6NHj2Lv3r1iH3fu3IGtrS2uXr2K+vXrw9fXF+7u7pg7d26F6w4JCcH9+/fFUbPg4GDEx8fj5s2bUFN7lWUbNmwIc3NzHDlyRKm+FStWoHfv3nj+/Dmsra2xZMkSMYg2adIE3bp1Q0REBNLT01G3bl2kp6fD2tpaXLe/vz88PT3x/fffIy4uDmFhYcjOzgYA+Pj4wMXFBcuWLRPbBwUFIT8/Hzt37sTvv/+OwMBApKamon79+qW26+TJk/D29kZGRgasrKyQlZWF2rVrY//+/WjTpg3S0tLg4OCAc+fOwd3dXVzO0dER06ZNQ58+fcRp06dPx65du3D8+HGEhobi8uXL2L9/P2QyWYX7FgAiIyMRFRVVarpt2AaoyXVeuzwRERFVXlpMx3fSb25uLgwMDJCTkwOFQlFh2zceCXRzcxN/VldXh4mJCVxdXcVpFhYWAICsrCycP38ehw4dgp6envhq2LAhAODGjRsVrmfhwoVo3rw5zMzMoKenh2XLliE9PV2pjYuLixgAS9b991pK6svKygIAaGlpoV+/fvjpp58AAGfPnsWlS5cQHBwMALh48SKKiopQv359pZoPHz5cbr2pqanw8fFRmubj44PU1FQAQHJyMmxsbMoMgADg6ekJFxcXcQRvzZo1sLOzQ+vWrcvdN/n5+bhx4wYGDx6sVOf06dPFOoODg5GcnIwGDRogNDQUv//+e7n9AUB4eDhycnLEV0ZGRoXtiYiIqHrTeNMFatWqpfReJpMpTSsZdSouLkZeXh46deqEGTNmlOrHysqq3HWsX78e48ePx+zZs+Hl5QV9fX3873//Q1JS0hvVUjKtuLhYfD9kyBC4u7vjzp07iI2NRdu2bWFnZwcAyMvLg7q6Os6cOQN1dXWlfvT09MqttyLa2tqvbTNkyBAsXLgQkyZNQmxsLAYOHFjh6F1eXh4AYPny5WjZsqXSvJK6mzVrhlu3bmH37t3Yv38/goKC4O/vX+r6wxJyuRxyubyym0VERETV3BuHwDfRrFkzbN68Gfb29tDQKHtVmpqaKCoqUpqWkJAAb29vjBo1Spz2upHDynJ1dYWHhweWL1+On3/+GT/++KM4r2nTpigqKkJWVhY+/PDDSvXn7OyMhIQEDBgwQKn+Ro0aAXg1cnrnzh388ccf5Y4Gfv7555g4cSLmz5+PlJQUpb40NTUBQGkfWVhYwNraGjdv3kTfvn3LrU2hUKBXr17o1asXevTogfbt2+PRo0cwNjau1LYRERFRzfVOQ+Do0aOxfPly9OnTBxMnToSxsTGuX7+O9evXY8WKFVBXV4e9vT2SkpKQlpYGPT09GBsbw8nJCatWrcLevXvh4OCA1atX49SpU+Kdsv/VkCFDEBISAl1dXXTt2lWcXr9+ffTt2xf9+/fH7Nmz0bRpU/z11184cOAA3Nzc0LFj6fP3EyZMQFBQEJo2bQp/f39s374dv/76K/bv3w8AaNOmDVq3bo3u3btjzpw5cHR0xJUrVyCTydC+fXsAgJGREbp164YJEyagXbt2sLGxEfs3NzeHtrY29uzZAxsbG2hpacHAwABRUVEIDQ2FgYEB2rdvj4KCApw+fRqPHz/GuHHjMGfOHFhZWaFp06ZQU1PDxo0bYWlpyQddExEREYB3/JxAa2trJCQkoKioCO3atYOrqyvCwsJgaGgoXss3fvx4qKuro1GjRjAzM0N6ejqGDx+Obt26oVevXmjZsiUePnyoNCr4X/Xp0wcaGhro06cPtLS0lObFxsaif//++PLLL9GgQQN06dIFp06dQp06dcrsq0uXLpg3bx5mzZoFFxcXLF26FLGxsUqPctm8eTNatGiBPn36oFGjRpg4cWKp0c/BgwfjxYsXGDRokNJ0DQ0NzJ8/H0uXLoW1tTU6d+4M4FWQXbFiBWJjY+Hq6oo2bdogLi5ODMr6+vqYOXMmPDw80KJFC6SlpWHXrl1K11ASERGRdL3R3cE1RVpaGurVq4dTp06hWbNmqi4HALB69Wp88cUXuHfvnngKWJVK7i7i3cFERERvX1W4O/idng6ual6+fImHDx/i22+/RatWrapEAHz69CkyMzMRExOD4cOHV4kASERERDWfpM4NJiQkwMrKCqdOncKSJUtUXQ4AYObMmWjYsCEsLS0RHh6u6nKIiIhIIiR5Ophej6eDiYiI3p2qcDpYUiOBRERERPQKQyARERGRBDEEEhEREUkQQyARERGRBDEEEhEREUkQQyARERGRBEnqYdH05i5FBbz2FnMiIiKqfjgSSERERCRBDIFEREREEsQQSERERCRBDIFEREREEsQQSERERCRBDIFEREREEsRHxFCFGkfshZpcR9VlEBER1ShpMR1VXQJHAomIiIikiCGQiIiISIIYAomIiIgkiCGQiIiISIIYAomIiIgkiCGQiIiISIIYAomIiIgkiCGQiIiISIIYAomIiIgkiCGQiIiISIKqbAj09fVFWFiYSmuIj4+HTCZDdna2SusgIiIietuqbAikV9LS0iCTyZCcnKzqUoiIiKgGYQgkIiIikqAqHwKnTp2Kxo0bl5ru7u6OyZMnAwCCg4PRpUsXfP/997CwsIChoSGmTp2KwsJCTJgwAcbGxrCxsUFsbKy4fMkI2/r16+Ht7Q0tLS00btwYhw8fLrWuM2fOwMPDAzo6OvD29sbVq1eV5i9evBj16tWDpqYmGjRogNWrVyvNz87OxvDhw2FhYSGuZ8eOHcjPz4dCocCmTZuU2m/duhW6urp48uQJHBwcAABNmzaFTCaDr6+v2G7FihVwdnaGlpYWGjZsiEWLFonzXrx4gZCQEFhZWUFLSwt2dnaIjo4udz8XFBQgNzdX6UVEREQ1V5UPgYMGDUJqaipOnTolTjt37hwuXLiAgQMHitMOHjyIe/fu4ciRI5gzZw4iIiLwySefwMjICElJSRgxYgSGDx+OO3fuKPU/YcIEfPnllzh37hy8vLzQqVMnPHz4UKnNN998g9mzZ+P06dPQ0NDAoEGDxHlbtmzB2LFj8eWXX+LSpUsYPnw4Bg4ciEOHDgEAiouLERgYiISEBKxZswYpKSmIiYmBuro6dHV10bt3b6VwCgCxsbHo0aMH9PX1cfLkSQDA/v37kZmZiV9//RUAsHbtWkyZMgXfffcdUlNT8f3332Py5MlYuXIlAGD+/PnYtm0bNmzYgKtXr2Lt2rWwt7cvdz9HR0fDwMBAfNna2lb2EBEREVE1JBMEQVB1EWXx9fWFu7s75s6diw4dOsDe3l4c6QoNDcXFixfFoBUcHIz4+HjcvHkTamqvcm3Dhg1hbm6OI0eOAACKiopgYGCAFStWoHfv3khLS4ODgwNiYmLw1VdfAQAKCwvh4OCAMWPGYOLEiYiPj8dHH32E/fv3w8/PDwCwa9cudOzYEc+ePYOWlhZ8fHzg4uKCZcuWibUHBQUhPz8fO3fuxO+//47AwECkpqaifv36pbbz5MmT8Pb2RkZGBqysrJCVlYXatWtj//79aNOmjVjnuXPn4O7uLi7n6OiIadOmoU+fPuK06dOnY9euXTh+/DhCQ0Nx+fJl7N+/HzKZ7LX7u6CgAAUFBeL73Nxc2NrawjZsA9TkOpU6ZkRERFQ5aTEd30m/ubm5MDAwQE5ODhQKRYVtq/xIIAAMHToU69atw/Pnz/HixQv8/PPPSqNxAODi4iIGQACwsLCAq6ur+F5dXR0mJibIyspSWs7Ly0v8WUNDAx4eHkhNTVVq4+bmJv5sZWUFAGI/qamp8PHxUWrv4+Mj9pGcnAwbG5syAyAAeHp6wsXFRRzBW7NmDezs7NC6dety90d+fj5u3LiBwYMHQ09PT3xNnz4dN27cAPAqGCcnJ6NBgwYIDQ3F77//Xm5/ACCXy6FQKJReREREVHNpqLqAyujUqRPkcjm2bNkCTU1NvHz5Ej169FBqU6tWLaX3MpmszGnFxcVvvP6/91MyqlbZfrS1tV/bZsiQIVi4cCEmTZqE2NhYDBw4sMLRu7y8PADA8uXL0bJlS6V56urqAIBmzZrh1q1b2L17N/bv34+goCD4+/uXuv6QiIiIpKlajARqaGhgwIABiI2NRWxsLHr37l2pcFUZJ06cEH8uLCzEmTNn4OzsXOnlnZ2dkZCQoDQtISEBjRo1AvBqFPHOnTv4448/yu3j888/x+3btzF//nykpKRgwIAB4jxNTU0Ar05nl7CwsIC1tTVu3rwJR0dHpVfJjSQAoFAo0KtXLyxfvhy//PILNm/ejEePHlV624iIiKjmqhYjgcCr0bKScPbP0PVfLFy4EE5OTnB2dsYPP/yAx48flzrVXJEJEyYgKCgITZs2hb+/P7Zv345ff/0V+/fvBwC0adMGrVu3Rvfu3TFnzhw4OjriypUrkMlkaN++PQDAyMgI3bp1w4QJE9CuXTvY2NiI/Zubm0NbWxt79uyBjY0NtLS0YGBggKioKISGhsLAwADt27dHQUEBTp8+jcePH2PcuHGYM2cOrKys0LRpU6ipqWHjxo2wtLSEoaHhW9t3REREVH1Vi5FAAHBycoK3tzcaNmxY6hTofxETE4OYmBg0adIEx44dw7Zt22Bqalrp5bt06YJ58+Zh1qxZcHFxwdKlSxEbG6v0KJfNmzejRYsW6NOnDxo1aoSJEycqjewBwODBg/HixYtSAVRDQwPz58/H0qVLYW1tjc6dOwN4FYpXrFiB2NhYuLq6ok2bNoiLixNHAvX19TFz5kx4eHigRYsWSEtLw65du5SumyQiIiLpqrJ3B/+TIAhwcnLCqFGjMG7cuP/cX3l33arK6tWr8cUXX+DevXviKWBVKrm7iHcHExERvX1V4e7ganE6+K+//sL69etx//59pWcD1gRPnz5FZmYmYmJiMHz48CoRAImIiKjmqxbnBs3NzTF16lQsW7YMRkZGqi7nrZo5cyYaNmwIS0tLhIeHq7ocIiIikohqczqY3i+eDiYiInp3qsLp4GoxEkhEREREbxdDIBEREZEEMQQSERERSRBDIBEREZEEMQQSERERSVC1eE4gqc6lqIDX3l1ERERE1Q9HAomIiIgkiCGQiIiISIIYAomIiIgkiCGQiIiISIIYAomIiIgkiCGQiIiISIL4iBiqUOOIvVCT66i6DCIioholLaajqkvgSCARERGRFDEEEhEREUkQQyARERGRBDEEEhEREUkQQyARERGRBDEEEhEREUkQQyARERGRBDEEEhEREUkQQyARERGRBDEEEhEREUkQQ2AlBAcHo0uXLlVyvfb29pg7d674XiaTYevWre+0LiIiIqr++N3BlTBv3jwIgqDqMsp06tQp6OrqqroMIiIiqmYYAivBwMBA1SWUy8zMTNUlEBERUTVUY04HFxcXIzo6Gg4ODtDW1kaTJk2wadMmcf7hw4fh6ekJuVwOKysrTJo0CYWFheL8TZs2wdXVFdra2jAxMYG/vz/y8/MBlD4t6+vri9DQUEycOBHGxsawtLREZGSkUj3p6eno3Lkz9PT0oFAoEBQUhD///FOcHxkZCXd3dyxduhS2trbQ0dFBUFAQcnJySm3brFmzYGVlBRMTE4wePRovX74U5/3zdPDfvXjxAiEhIbCysoKWlhbs7OwQHR39JruViIiIaqgaEwKjo6OxatUqLFmyBJcvX8YXX3yBzz//HIcPH8bdu3fRoUMHtGjRAufPn8fixYvxf//3f5g+fToAIDMzE3369MGgQYOQmpqK+Ph4dOvWrcJTwCtXroSuri6SkpIwc+ZMTJ06Ffv27QPwKpB27twZjx49wuHDh7Fv3z7cvHkTvXr1Uurj+vXr2LBhA7Zv3449e/bg3LlzGDVqlFKbQ4cO4caNGzh06BBWrlyJuLg4xMXFVWqfzJ8/H9u2bcOGDRtw9epVrF27Fvb29mW2LSgoQG5urtKLiIiIaq4acTq4oKAA33//Pfbv3w8vLy8AQN26dXHs2DEsXboUDg4OsLW1xY8//giZTIaGDRvi3r17+OqrrzBlyhRkZmaisLAQ3bp1g52dHQDA1dW1wnW6ubkhIiICAODk5IQff/wRBw4cwMcff4wDBw7g4sWLuHXrFmxtbQEAq1atgouLC06dOoUWLVoAAJ4/f45Vq1ahdu3aAIAFCxagY8eOmD17NiwtLQEARkZG+PHHH6Guro6GDRuiY8eOOHDgAIYOHfra/ZKeng4nJyd88MEHkMlk4raVJTo6GlFRUa/tk4iIiGqGGjESeP36dTx9+hQff/wx9PT0xNeqVatw48YNpKamwsvLCzKZTFzGx8cHeXl5uHPnDpo0aQI/Pz+4urqiZ8+eWL58OR4/flzhOt3c3JTeW1lZISsrCwCQmpoKW1tbMQACQKNGjWBoaIjU1FRxWp06dcQACABeXl4oLi7G1atXxWkuLi5QV1cvcz2vExwcjOTkZDRo0AChoaH4/fffy20bHh6OnJwc8ZWRkVGpdRAREVH1VCNCYF5eHgBg586dSE5OFl8pKSlK1wWWR11dHfv27cPu3bvRqFEjLFiwAA0aNMCtW7fKXaZWrVpK72UyGYqLi//bhrzl9TRr1gy3bt3CtGnT8OzZMwQFBaFHjx5ltpXL5VAoFEovIiIiqrlqRAhs1KgR5HI50tPT4ejoqPSytbWFs7MzEhMTla7xS0hIgL6+PmxsbAC8Clc+Pj6IiorCuXPnoKmpiS1btvyrepydnZGRkaE0mpaSkoLs7Gw0atRInJaeno579+6J70+cOAE1NTU0aNDgX623LAqFAr169cLy5cvxyy+/YPPmzXj06NFb65+IiIiqpxpxTaC+vj7Gjx+PL774AsXFxfjggw+Qk5ODhIQEKBQKjBo1CnPnzsWYMWMQEhKCq1evIiIiAuPGjYOamhqSkpJw4MABtGvXDubm5khKSsJff/0FZ2fnf1WPv78/XF1d0bdvX8ydOxeFhYUYNWoU2rRpAw8PD7GdlpYWBgwYgFmzZiE3NxehoaEICgoSrwf8r+bMmQMrKys0bdoUampq2LhxIywtLWFoaPhW+iciIqLqq0aEQACYNm0azMzMEB0djZs3b8LQ0BDNmjXD119/jdq1a2PXrl2YMGECmjRpAmNjYwwePBjffvstgFejZUeOHMHcuXORm5sLOzs7zJ49G4GBgf+qFplMht9++w1jxoxB69atoaamhvbt22PBggVK7RwdHdGtWzd06NABjx49wieffIJFixb9531RQl9fHzNnzsS1a9egrq6OFi1aYNeuXVBTqxEDwERERPQfyISq+lUYNVxkZCS2bt2K5ORkVZdSptzcXBgYGMA2bAPU5DqqLoeIiKhGSYvp+E76Lfn7nZOT89rr+zkkRERERCRBDIFEREREEsQQqCKRkZFV9lQwERER1XwMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMgUREREQSVGO+MYTejUtRAa992CQRERFVPxwJJCIiIpIghkAiIiIiCWIIJCIiIpIghkAiIiIiCWIIJCIiIpIghkAiIiIiCWIIJCIiIpIgPieQKtQ4Yi/U5DqqLoOIiKhGSYvpqOoSOBJIREREJEUMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMge+Ir68vwsLCyp1vb2+PuXPnvrd6iIiIiP6OIbAGiI+Ph0wmQ3Z2tqpLISIiomqCIZCIiIhIghgC36HCwkKEhITAwMAApqammDx5MgRBKLPtnDlz4OrqCl1dXdja2mLUqFHIy8sT59++fRudOnWCkZERdHV14eLigl27diEtLQ0fffQRAMDIyAgymQzBwcEAgE2bNsHV1RXa2towMTGBv78/8vPz3/l2ExERUdWnoeoCarKVK1di8ODBOHnyJE6fPo1hw4ahTp06GDp0aKm2ampqmD9/PhwcHHDz5k2MGjUKEydOxKJFiwAAo0ePxosXL3DkyBHo6uoiJSUFenp6sLW1xebNm9G9e3dcvXoVCoUC2trayMzMRJ8+fTBz5kx07doVT548wdGjR8sNoQUFBSgoKBDf5+bmvpudQkRERFUCQ+A7ZGtrix9++AEymQwNGjTAxYsX8cMPP5QZAv9+E4m9vT2mT5+OESNGiCEwPT0d3bt3h6urKwCgbt26YntjY2MAgLm5OQwNDQEAN27cQGFhIbp16wY7OzsAEJctS3R0NKKiov7T9hIREVH1wdPB71CrVq0gk8nE915eXrh27RqKiopKtd2/fz/8/PxQu3Zt6Ovro1+/fnj48CGePn0KAAgNDcX06dPh4+ODiIgIXLhwocJ1N2nSBH5+fnB1dUXPnj2xfPlyPH78uNz24eHhyMnJEV8ZGRn/cquJiIioOmAIrALS0tLwySefwM3NDZs3b8aZM2ewcOFCAMCLFy8AAEOGDMHNmzfRr18/XLx4ER4eHliwYEG5faqrq2Pfvn3YvXs3GjVqhAULFqBBgwa4detWme3lcjkUCoXSi4iIiGouhsB3KCkpSen9iRMn4OTkBHV1daXpZ86cQXFxMWbPno1WrVqhfv36uHfvXqn+bG1tMWLECPz666/48ssvsXz5cgCApqYmAJQaYZTJZPDx8UFUVBTOnTsHTU1NbNmy5W1uIhEREVVTvCbwHUpPT8e4ceMwfPhwnD17FgsWLMDs2bNLtXN0dMTLly+xYMECdOrUCQkJCViyZIlSm7CwMAQGBqJ+/fp4/PgxDh06BGdnZwCAnZ0dZDIZduzYgQ4dOkBbWxuXL1/GgQMH0K5dO5ibmyMpKQl//fWXuAwRERFJG0cC36H+/fvj2bNn8PT0xOjRozF27FgMGzasVLsmTZpgzpw5mDFjBho3boy1a9ciOjpaqU1RURFGjx4NZ2dntG/fHvXr1xdvGqlduzaioqIwadIkWFhYICQkBAqFAkeOHEGHDh1Qv359fPvtt5g9ezYCAwPfy7YTERFR1SYTyntmCElabm4uDAwMYBu2AWpyHVWXQ0REVKOkxXR8J/2W/P3Oycl57fX9HAkkIiIikiCGQCIiIiIJYggkIiIikiCGQCIiIiIJYggkIiIikiCGQCIiIiIJYggkIiIikiCGQCIiIiIJ4tfGUYUuRQW89mGTREREVP1wJJCIiIhIghgCiYiIiCSIIZCIiIhIghgCiYiIiCSIIZCIiIhIghgCiYiIiCSIIZCIiIhIgvicQKpQ44i9UJPrqLoMIiKiGiUtpqOqS+BIIBEREZEUMQQSERERSRBDIBEREZEEMQQSERERSRBDIBEREZEEMQQSERERSRBDIBEREZEEMQQSERERSRBDIBEREZEESTIE+vr6IiwsTNVlVKgyNcpkMmzduhUAkJaWBplMhuTk5HdeGxEREVV//Nq4aiwzMxNGRkaqLoOIiIiqIYbAaszS0lLVJRAREVE1JcnTwQBQWFiIkJAQGBgYwNTUFJMnT4YgCACA1atXw8PDA/r6+rC0tMRnn32GrKwscdnHjx+jb9++MDMzg7a2NpycnBAbGyvOv3jxItq2bQttbW2YmJhg2LBhyMvLE+cHBwejS5cuiIqKgpmZGRQKBUaMGIEXL14o1VhcXIyJEyfC2NgYlpaWiIyMVJr/99PB//S6GomIiEjaJBsCV65cCQ0NDZw8eRLz5s3DnDlzsGLFCgDAy5cvMW3aNJw/fx5bt25FWloagoODxWUnT56MlJQU7N69G6mpqVi8eDFMTU0BAPn5+QgICICRkRFOnTqFjRs3Yv/+/QgJCVFa/4EDB5Camor4+HisW7cOv/76K6KiokrVqKuri6SkJMycORNTp07Fvn37KrV9FdVYloKCAuTm5iq9iIiIqOaS7OlgW1tb/PDDD5DJZGjQoAEuXryIH374AUOHDsWgQYPEdnXr1sX8+fPRokUL5OXlQU9PD+np6WjatCk8PDwAAPb29mL7n3/+Gc+fP8eqVaugq6sLAPjxxx/RqVMnzJgxAxYWFgAATU1N/PTTT9DR0YGLiwumTp2KCRMmYNq0aVBTe5XN3dzcEBERAQBwcnLCjz/+iAMHDuDjjz9+7fZVVGNZoqOjS4VQIiIiqrkkOxLYqlUryGQy8b2XlxeuXbuGoqIinDlzBp06dUKdOnWgr6+PNm3aAHgVrABg5MiRWL9+Pdzd3TFx4kQcP35c7Cc1NRVNmjQRAyAA+Pj4oLi4GFevXhWnNWnSBDo6Okrrz8vLQ0ZGhjjNzc1NqWYrKyul09IVqajGsoSHhyMnJ0d8/b0OIiIiqnkkGwLL8/z5cwQEBEChUGDt2rU4deoUtmzZAgDiNXuBgYG4ffs2vvjiC9y7dw9+fn4YP378W6+lVq1aSu9lMhmKi4srteyb1iiXy6FQKJReREREVHNJNgQmJSUpvT9x4gScnJxw5coVPHz4EDExMfjwww/RsGHDMkffzMzMMGDAAKxZswZz587FsmXLAADOzs44f/488vPzxbYJCQlQU1NDgwYNxGnnz5/Hs2fPlNavp6cHW1vbt7aN5dVIREREJNkQmJ6ejnHjxuHq1atYt24dFixYgLFjx6JOnTrQ1NTEggULcPPmTWzbtg3Tpk1TWnbKlCn47bffcP36dVy+fBk7duyAs7MzAKBv377Q0tLCgAEDcOnSJRw6dAhjxoxBv379xOsBgVejioMHD0ZKSgp27dqFiIgIhISEiNcD/lcV1UhEREQk2RtD+vfvj2fPnsHT0xPq6uoYO3Yshg0bBplMhri4OHz99deYP38+mjVrhlmzZuHTTz8Vl9XU1ER4eDjS0tKgra2NDz/8EOvXrwcA6OjoYO/evRg7dixatGgBHR0ddO/eHXPmzFFav5+fH5ycnNC6dWsUFBSgT58+pR4B819UVCMRERGRTCh5OB69N8HBwcjOzi73GX9VQW5uLgwMDGAbtgFqcp3XL0BERESVlhbT8Z30W/L3Oycn57XX90v2dDARERGRlDEEEhEREUmQZK8JVKW4uDhVl0BEREQSx5FAIiIiIgliCCQiIiKSIIZAIiIiIgliCCQiIiKSIIZAIiIiIgni3cFUoUtRAa992CQRERFVPxwJJCIiIpIghkAiIiIiCWIIJCIiIpIghkAiIiIiCWIIJCIiIpIghkAiIiIiCWIIJCIiIpIgPieQKtQ4Yi/U5DqqLoOIiKhGSYvpqOoSOBJIREREJEUMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMge+Ar68vwsLC3vt6IyMj4e7u/t7XS0RERNUPQyARERGRBDEEEhEREUkQQ+A7UlhYiJCQEBgYGMDU1BSTJ0+GIAgAgNWrV8PDwwP6+vqwtLTEZ599hqysLHHZ+Ph4yGQyHDhwAB4eHtDR0YG3tzeuXr2qtI6YmBhYWFhAX18fgwcPxvPnz5Xmx8fHw9PTE7q6ujA0NISPjw9u37797jeeiIiIqjyGwHdk5cqV0NDQwMmTJzFv3jzMmTMHK1asAAC8fPkS06ZNw/nz57F161akpaUhODi4VB/ffPMNZs+ejdOnT0NDQwODBg0S523YsAGRkZH4/vvvcfr0aVhZWWHRokXi/MLCQnTp0gVt2rTBhQsXkJiYiGHDhkEmk5VZb0FBAXJzc5VeREREVHPJhJLhKXprfH19kZWVhcuXL4uha9KkSdi2bRtSUlJKtT99+jRatGiBJ0+eQE9PD/Hx8fjoo4+wf/9++Pn5AQB27dqFjh074tmzZ9DS0oK3tzeaNm2KhQsXiv20atUKz58/R3JyMh49egQTExPEx8ejTZs2r605MjISUVFRpabbhm2Amlzn3+4KIiIiKkNaTMd30m9ubi4MDAyQk5MDhUJRYVuOBL4jrVq1Uhp18/LywrVr11BUVIQzZ86gU6dOqFOnDvT19cWQlp6ertSHm5ub+LOVlRUAiKeNU1NT0bJlS6X2Xl5e4s/GxsYIDg5GQEAAOnXqhHnz5iEzM7PcesPDw5GTkyO+MjIy/uWWExERUXXAEPiePX/+HAEBAVAoFFi7di1OnTqFLVu2AABevHih1LZWrVrizyWBsri4uNLrio2NRWJiIry9vfHLL7+gfv36OHHiRJlt5XI5FAqF0ouIiIhqLobAdyQpKUnp/YkTJ+Dk5IQrV67g4cOHiImJwYcffoiGDRsq3RRSWc7OzmWu45+aNm2K8PBwHD9+HI0bN8bPP//8xusiIiKimoch8B1JT0/HuHHjcPXqVaxbtw4LFizA2LFjUadOHWhqamLBggW4efMmtm3bhmnTpr1x/2PHjsVPP/2E2NhY/PHHH4iIiMDly5fF+bdu3UJ4eDgSExNx+/Zt/P7777h27RqcnZ3f5mYSERFRNaWh6gJqqv79++PZs2fw9PSEuro6xo4dK96dGxcXh6+//hrz589Hs2bNMGvWLHz66adv1H+vXr1w48YNTJw4Ec+fP0f37t0xcuRI7N27FwCgo6ODK1euYOXKlXj48CGsrKwwevRoDB8+/F1sLhEREVUzvDuYylRydxHvDiYiInr7eHcwEREREakEQyARERGRBDEEEhEREUkQQyARERGRBDEEEhEREUkQQyARERGRBDEEEhEREUkQQyARERGRBPEbQ6hCl6ICXvuwSSIiIqp+OBJIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEaqi6AqiZBEAAAubm5Kq6EiIiIKqvk73bJ3/GKMARSmR4+fAgAsLW1VXElRERE9KaePHkCAwODCtswBFKZjI2NAQDp6emv/RCRauTm5sLW1hYZGRlQKBSqLofKwGNU9fEYVX08Rm9GEAQ8efIE1tbWr23LEEhlUlN7dbmogYEBf+mqOIVCwWNUxfEYVX08RlUfj1HlVXbwhjeGEBEREUkQQyARERGRBDEEUpnkcjkiIiIgl8tVXQqVg8eo6uMxqvp4jKo+HqN3RyZU5h5iIiIiIqpROBJIREREJEEMgUREREQSxBBIREREJEEMgUREREQSxBBIREREJEEMgVSmhQsXwt7eHlpaWmjZsiVOnjyp6pIkKzIyEjKZTOnVsGFDcf7z588xevRomJiYQE9PD927d8eff/6pwoprviNHjqBTp06wtraGTCbD1q1bleYLgoApU6bAysoK2tra8Pf3x7Vr15TaPHr0CH379oVCoYChoSEGDx6MvLy897gVNdvrjlFwcHCp36v27dsrteExeneio6PRokUL6Ovrw9zcHF26dMHVq1eV2lTm37b09HR07NgROjo6MDc3x4QJE1BYWPg+N6VaYwikUn755ReMGzcOEREROHv2LJo0aYKAgABkZWWpujTJcnFxQWZmpvg6duyYOO+LL77A9u3bsXHjRhw+fBj37t1Dt27dVFhtzZefn48mTZpg4cKFZc6fOXMm5s+fjyVLliApKQm6uroICAjA8+fPxTZ9+/bF5cuXsW/fPuzYsQNHjhzBsGHD3tcm1HivO0YA0L59e6Xfq3Xr1inN5zF6dw4fPozRo0fjxIkT2LdvH16+fIl27dohPz9fbPO6f9uKiorQsWNHvHjxAsePH8fKlSsRFxeHKVOmqGKTqieB6B88PT2F0aNHi++LiooEa2trITo6WoVVSVdERITQpEmTMudlZ2cLtWrVEjZu3ChOS01NFQAIiYmJ76lCaQMgbNmyRXxfXFwsWFpaCv/73//EadnZ2YJcLhfWrVsnCIIgpKSkCACEU6dOiW12794tyGQy4e7du++tdqn45zESBEEYMGCA0Llz53KX4TF6v7KysgQAwuHDhwVBqNy/bbt27RLU1NSE+/fvi20WL14sKBQKoaCg4P1uQDXFkUBS8uLFC5w5cwb+/v7iNDU1Nfj7+yMxMVGFlUnbtWvXYG1tjbp166Jv375IT08HAJw5cwYvX75UOl4NGzZEnTp1eLxU5NatW7h//77SMTEwMEDLli3FY5KYmAhDQ0N4eHiIbfz9/aGmpoakpKT3XrNUxcfHw9zcHA0aNMDIkSPx8OFDcR6P0fuVk5MDADA2NgZQuX/bEhMT4erqCgsLC7FNQEAAcnNzcfny5fdYffXFEEhKHjx4gKKiIqVfKgCwsLDA/fv3VVSVtLVs2RJxcXHYs2cPFi9ejFu3buHDDz/EkydPcP/+fWhqasLQ0FBpGR4v1SnZ7xX9Dt2/fx/m5uZK8zU0NGBsbMzj9p60b98eq1atwoEDBzBjxgwcPnwYgYGBKCoqAsBj9D4VFxcjLCwMPj4+aNy4MQBU6t+2+/fvl/l7VjKPXk9D1QUQUcUCAwPFn93c3NCyZUvY2dlhw4YN0NbWVmFlRNVX7969xZ9dXV3h5uaGevXqIT4+Hn5+fiqsTHpGjx6NS5cuKV3rTO8HRwJJiampKdTV1UvdgfXnn3/C0tJSRVXR3xkaGqJ+/fq4fv06LC0t8eLFC2RnZyu14fFSnZL9XtHvkKWlZakbrQoLC/Ho0SMeNxWpW7cuTE1Ncf36dQA8Ru9LSEgIduzYgUOHDsHGxkacXpl/2ywtLcv8PSuZR6/HEEhKNDU10bx5cxw4cECcVlxcjAMHDsDLy0uFlVGJvLw83LhxA1ZWVmjevDlq1aqldLyuXr2K9PR0Hi8VcXBwgKWlpdIxyc3NRVJSknhMvLy8kJ2djTNnzohtDh48iOLiYrRs2fK910zAnTt38PDhQ1hZWQHgMXrXBEFASEgItmzZgoMHD8LBwUFpfmX+bfPy8sLFixeVwvq+ffugUCjQqFGj97Mh1Z2q70yhqmf9+vWCXC4X4uLihJSUFGHYsGGCoaGh0h1Y9P58+eWXQnx8vHDr1i0hISFB8Pf3F0xNTYWsrCxBEARhxIgRQp06dYSDBw8Kp0+fFry8vAQvLy8VV12zPXnyRDh37pxw7tw5AYAwZ84c4dy5c8Lt27cFQRCEmJgYwdDQUPjtt9+ECxcuCJ07dxYcHByEZ8+eiX20b99eaNq0qZCUlCQcO3ZMcHJyEvr06aOqTapxKjpGT548EcaPHy8kJiYKt27dEvbv3y80a9ZMcHJyEp4/fy72wWP07owcOVIwMDAQ4uPjhczMTPH19OlTsc3r/m0rLCwUGjduLLRr105ITk4W9uzZI5iZmQnh4eGq2KRqiSGQyrRgwQKhTp06gqampuDp6SmcOHFC1SVJVq9evQQrKytBU1NTqF27ttCrVy/h+vXr4vxnz54Jo0aNEoyMjAQdHR2ha9euQmZmpgorrvkOHTokACj1GjBggCAIrx4TM3nyZMHCwkKQy+WCn5+fcPXqVaU+Hj58KPTp00fQ09MTFAqFMHDgQOHJkycq2JqaqaJj9PTpU6Fdu3aCmZmZUKtWLcHOzk4YOnRoqf/o8hi9O2UdGwBCbGys2KYy/7alpaUJgYGBgra2tmBqaip8+eWXwsuXL9/z1lRfMkEQhPc9+khEREREqsVrAomIiIgkiCGQiIiISIIYAomIiIgkiCGQiIiISIIYAomIiIgkiCGQiIiISIIYAomIiIgkiCGQiIiISIIYAomIiIgkiCGQiIiISIIYAomIiIgk6P8D+jfc/+bHQFIAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Observations:\n",
        "\n",
        "1. Each class in the train dataset has 250 samples after applying SMOTE\n",
        "2. This removes the imbalance in the train dataset"
      ],
      "metadata": {
        "id": "pHj3Ujf8ZPJ7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# assign the values of X_smote and y_smote to X_train and y_train respectively\n",
        "\n",
        "X_train, y_train = X_smote, y_smote\n"
      ],
      "metadata": {
        "id": "LFZAXhELlMXT"
      },
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# re-shape the dimension of X_train that was changed to apply SMOTE\n",
        "\n",
        "n_samples, height, width, n_channels = 225*9, 256, 256, 3\n",
        "\n",
        "X_train = X_train.reshape(n_samples, height, width, n_channels)\n"
      ],
      "metadata": {
        "id": "RdIBzGkSlMUS"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# get all the dimensions of X_train\n",
        "\n",
        "n_samples, height, width, n_channels = [X_train.shape[index] for index in range(4)]\n",
        "print(f\"Shape of X_train before reshape: {X_train.shape}\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mwEHXcS5Z1Lr",
        "outputId": "c396f251-45e6-457a-d6f6-5e8ff7778d12"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Shape of X_train before reshape: (2025, 256, 256, 3)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Observations:\n",
        "\n",
        "1. Shape of X_train has been restored for training the model"
      ],
      "metadata": {
        "id": "thi70TZxZxc8"
      }
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "9WmI-hcNafcq"
      },
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## One-hot encoding of the labels"
      ],
      "metadata": {
        "id": "ymywygQWFzbk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Convert the labels in y_train using one-hot encoding\n",
        "\n",
        "encoder = LabelEncoder()\n",
        "encoder.fit(y_train)\n",
        "encoded_y_train = encoder.transform(y_train)\n",
        "encoded_y_test = encoder.transform(y_test)\n",
        "\n",
        "y_train = np_utils.to_categorical(encoded_y_train)\n",
        "y_test = np_utils.to_categorical(encoded_y_test)\n"
      ],
      "metadata": {
        "id": "M-Hw6QM81OwW"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "7oil3Apr1zre"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Model: VGG16"
      ],
      "metadata": {
        "id": "PcDcxoZNITOv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# define the variables to be used in the model\n",
        "\n",
        "num_classes = 9\n",
        "\n",
        "epochs = 10\n",
        "batch_size = 32\n"
      ],
      "metadata": {
        "id": "GGhQlUJ81OrI"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# define the VGG16 model\n",
        "\n",
        "base_model = applications.VGG16(include_top=False, weights='imagenet')\n",
        "input = Input(shape=(256, 256, 3), name = 'image_input')\n",
        "vgg_output = base_model(input)\n",
        "\n",
        "top_model = Flatten()(vgg_output)\n",
        "top_model = Dense(64, activation='relu')(top_model)\n",
        "predictions = Dense(num_classes, activation='softmax', name='prediction_layer')(top_model)\n",
        "\n",
        "model = Model(inputs=input, outputs=predictions)\n",
        "\n",
        "# first: train only the top layers (which were randomly initialized)\n",
        "# i.e. freeze all convolutional InceptionV3 layers\n",
        "\n",
        "layers = base_model.layers[:-2]\n",
        "for layer in layers:\n",
        "    layer.trainable = False\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nf_h1opg1OoU",
        "outputId": "525c261b-13a0-4d26-95d5-33a0cc5daeb5"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/vgg16/vgg16_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
            "58889256/58889256 [==============================] - 0s 0us/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# compile the model (should be done after setting layers to non-trainable)\n",
        "\n",
        "model.compile(optimizer='rmsprop', loss='categorical_crossentropy', metrics=['accuracy'])\n"
      ],
      "metadata": {
        "id": "0cYDFX_D1Olg"
      },
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Note:\n",
        "\n",
        "Accuracy is used as a metric for model evaluation because it is more important that the model classifies each image correctly rather than penalizing the False-Positives (Precision) or the False-Negatives (Recall) or penalizing both (F1)."
      ],
      "metadata": {
        "id": "xNDKdYnB74wY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# model summary\n",
        "\n",
        "model.summary()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KcJXi6Ij1Of8",
        "outputId": "fb4a8919-9a2f-4992-b959-d9ff6189f630"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " image_input (InputLayer)    [(None, 256, 256, 3)]     0         \n",
            "                                                                 \n",
            " vgg16 (Functional)          (None, None, None, 512)   14714688  \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 32768)             0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 64)                2097216   \n",
            "                                                                 \n",
            " prediction_layer (Dense)    (None, 9)                 585       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 16,812,489\n",
            "Trainable params: 4,457,609\n",
            "Non-trainable params: 12,354,880\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# fit the model on train data\n",
        "\n",
        "model.fit(X_train, y_train, epochs=epochs, batch_size=batch_size, validation_split=0.2, shuffle=True, verbose=1)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zt0UdqYS1OdE",
        "outputId": "3abad552-3476-4551-9bb8-45f5897d486c"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "51/51 [==============================] - 39s 459ms/step - loss: 2.7645 - accuracy: 0.4432 - val_loss: 2.4323 - val_accuracy: 0.0494\n",
            "Epoch 2/10\n",
            "51/51 [==============================] - 11s 211ms/step - loss: 0.9276 - accuracy: 0.6994 - val_loss: 1.9017 - val_accuracy: 0.2420\n",
            "Epoch 3/10\n",
            "51/51 [==============================] - 10s 201ms/step - loss: 0.4159 - accuracy: 0.8691 - val_loss: 4.6440 - val_accuracy: 0.0864\n",
            "Epoch 4/10\n",
            "51/51 [==============================] - 10s 202ms/step - loss: 0.2133 - accuracy: 0.9302 - val_loss: 1.0255 - val_accuracy: 0.6272\n",
            "Epoch 5/10\n",
            "51/51 [==============================] - 11s 207ms/step - loss: 0.1087 - accuracy: 0.9710 - val_loss: 0.5444 - val_accuracy: 0.8222\n",
            "Epoch 6/10\n",
            "51/51 [==============================] - 11s 219ms/step - loss: 0.0683 - accuracy: 0.9778 - val_loss: 0.5444 - val_accuracy: 0.7951\n",
            "Epoch 7/10\n",
            "51/51 [==============================] - 11s 219ms/step - loss: 0.0814 - accuracy: 0.9809 - val_loss: 1.2309 - val_accuracy: 0.6568\n",
            "Epoch 8/10\n",
            "51/51 [==============================] - 11s 216ms/step - loss: 0.0234 - accuracy: 0.9914 - val_loss: 2.0224 - val_accuracy: 0.6346\n",
            "Epoch 9/10\n",
            "51/51 [==============================] - 11s 215ms/step - loss: 0.0434 - accuracy: 0.9883 - val_loss: 0.8137 - val_accuracy: 0.7802\n",
            "Epoch 10/10\n",
            "51/51 [==============================] - 10s 204ms/step - loss: 0.0297 - accuracy: 0.9938 - val_loss: 0.5884 - val_accuracy: 0.8148\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7f66876934c0>"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# save the weights of the model\n",
        "\n",
        "model.save_weights('vgg_top.h5')\n"
      ],
      "metadata": {
        "id": "hPLrgF6A1OaD"
      },
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# load the weights of the moodel\n",
        "\n",
        "model.load_weights('vgg_top.h5')\n"
      ],
      "metadata": {
        "id": "HXHOxjOG1OXn"
      },
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "IvriFN5CaiKS"
      },
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Make predictions on the test dataset that was created using 20% images of the wbc folder"
      ],
      "metadata": {
        "id": "Y_zmfE38OM5A"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# predictions on the test dataset\n",
        "\n",
        "y_pred = model.predict(X_test)\n",
        "\n",
        "print(y_pred)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_Aco1FCV1OUg",
        "outputId": "32d3018a-793e-4835-ae6d-c5407292b979"
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "7/7 [==============================] - 5s 859ms/step\n",
            "[[1.1997371e-03 5.0501665e-04 2.2212097e-04 ... 4.2700153e-04\n",
            "  1.4980124e-05 9.6989375e-01]\n",
            " [9.1864895e-06 5.5854028e-09 1.0568584e-05 ... 2.6217660e-07\n",
            "  1.8363458e-07 9.9997687e-01]\n",
            " [1.8853154e-06 7.1311776e-12 1.5850411e-08 ... 2.9189726e-10\n",
            "  2.9524664e-11 9.9999726e-01]\n",
            " ...\n",
            " [1.5493313e-07 3.8072159e-16 7.5505913e-14 ... 5.9242417e-11\n",
            "  5.3941239e-15 9.9999988e-01]\n",
            " [5.1076580e-07 1.0385508e-13 1.5703703e-08 ... 1.8461928e-11\n",
            "  2.1546596e-10 9.9999869e-01]\n",
            " [3.2014495e-06 1.9007351e-09 4.5967672e-06 ... 4.1000951e-07\n",
            "  6.7060242e-08 9.9243474e-01]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# rounding the elements of the y_pred array to the nearest integer using np.rint\n",
        "\n",
        "y_pred = np.rint(y_pred)\n",
        "\n",
        "print(y_pred)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Y3fLzNvT1OSJ",
        "outputId": "96fc32f6-98f6-4a15-de90-4ae5afe4dc01"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# print the accuracy obtained on the test data\n",
        "\n",
        "print(accuracy_score(y_test, y_pred))\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YuIDO5Ma1ONO",
        "outputId": "78063c1e-fd65-448c-a0f1-d98b304c49de"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.7149532710280374\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Observations:\n",
        "\n",
        "1. Getting accuracy score around 75%"
      ],
      "metadata": {
        "id": "txegCW7GekbG"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# the confusion matrix\n",
        "\n",
        "y_pred_unencoded = np.argmax(y_pred, axis=1)\n",
        "y_test_unencoded = np.argmax(y_test, axis=1)\n",
        "\n",
        "print(confusion_matrix(y_test_unencoded, y_pred_unencoded))\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ueA3Y1VO1OGM",
        "outputId": "9a730cdb-a5b0-4e2b-9df2-678dc4f96d70"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[ 0  0  0  0  0  0  0  0 12]\n",
            " [ 0  3  0  0  0  0  0  1  0]\n",
            " [ 0  0  0  0  0  0  0  0  2]\n",
            " [ 2  0  0  3  2  0  0  0  3]\n",
            " [ 0  0  0  0 72  2  0  2  0]\n",
            " [ 3  0  0  0  0  2  1  0  4]\n",
            " [ 0  0  0  0  1  1  7  2  1]\n",
            " [ 0  0  0  0  4  0  0  0  0]\n",
            " [ 6  0  1  2  2  7  0  0 66]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# confusion matrix visualization using heatplot\n",
        "\n",
        "def plot_heatmap(y_true, y_pred, class_names, ax, title):\n",
        "\n",
        "    cm = confusion_matrix(y_true, y_pred)\n",
        "\n",
        "    sns.heatmap(\n",
        "        cm,\n",
        "        annot=True,\n",
        "        square=True,\n",
        "        xticklabels=class_names,\n",
        "        yticklabels=class_names,\n",
        "        fmt='d',\n",
        "        cmap=plt.cm.Blues,\n",
        "        cbar=False,\n",
        "    )\n",
        "\n",
        "    ax.set_title(title, fontsize=16)\n",
        "    ax.set_xticklabels(ax.get_xticklabels(), rotation=90, ha=\"right\")\n",
        "    ax.set_ylabel('True Label', fontsize=12)\n",
        "    ax.set_xlabel('Predicted Label', fontsize=12)\n",
        "\n",
        "# Get the names of the ten classes\n",
        "class_names = classes\n",
        "\n",
        "fig = plt.figure(figsize = (5, 5))\n",
        "ax = fig.gca()\n",
        "plot_heatmap(y_test_unencoded, y_pred_unencoded, class_names, ax, title=\"VGG16 Confusion Matrix\")\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 595
        },
        "id": "ATdmd7rYEmyS",
        "outputId": "594f3d61-7f2d-4f85-fc37-17022f43fe29"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 500x500 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "Gp8kZYoUyGVg"
      },
      "execution_count": 33,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Making prediction on the *classify-images* dataset"
      ],
      "metadata": {
        "id": "xYNFycXmK-vB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# define get_data_classify_images to load the images in the classify-images folder\n",
        "\n",
        "def get_data_classify_images(folder):\n",
        "\n",
        "    X = []\n",
        "    y = []\n",
        "\n",
        "    for image_filename in os.listdir(folder):\n",
        "        img_file = cv2.imread(folder + '/' + image_filename)\n",
        "\n",
        "        if img_file is not None:\n",
        "            img_arr = np.asarray(img_file)\n",
        "            X.append(img_arr)\n",
        "            y.append(image_filename)\n",
        "\n",
        "    X = np.asarray(X)\n",
        "    y = np.asarray(y)\n",
        "\n",
        "    return X, y\n"
      ],
      "metadata": {
        "id": "TXcLkHMsyHM6"
      },
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# call the get_data_classify_images method using the path of classify-images folder as the argument\n",
        "\n",
        "X_classify_images, classify_images_filenames = get_data_classify_images(BASE_DIR + '/classify-images')\n"
      ],
      "metadata": {
        "id": "w440nL2oyQ_m"
      },
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# load weights of the model\n",
        "\n",
        "model.load_weights('vgg_top.h5')\n"
      ],
      "metadata": {
        "id": "9oAnA9L41oJa"
      },
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# prediction on the classify-images data\n",
        "\n",
        "y_pred_classify_images = model.predict(X_classify_images)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iT2rXsN11tVU",
        "outputId": "498efc12-25e6-476b-d43a-07515e437a28"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2/2 [==============================] - 5s 5s/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# rounding the elements of the y_pred array to the nearest integer using np.rint\n",
        "\n",
        "y_pred_classify_images = np.rint(y_pred_classify_images)\n",
        "\n",
        "print('Prediction on the classify-images data')\n",
        "print()\n",
        "\n",
        "print(y_pred_classify_images)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UNe5m3w01tSr",
        "outputId": "32b7335c-bfe0-4b04-fd60-1d2ab9633278"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Prediction on the classify-images data\n",
            "\n",
            "[[0. 1. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 0. 0. 1. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 1. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 0. 0. 1. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 1. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 1. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [1. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [1. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [1. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [1. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [0. 1. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 1. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 1. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [1. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 1. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 1. 0.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 1. 0.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 0. 0. 0. 0. 1.]\n",
            " [0. 0. 0. 0. 1. 0. 0. 0. 0.]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "\n",
        "df = pd.DataFrame(data = y_pred_classify_images, index = classify_images_filenames, columns = classes)"
      ],
      "metadata": {
        "id": "QlS2FIOLKxhf"
      },
      "execution_count": 39,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "PfMZHjVSKxej",
        "outputId": "a3208ee6-ec59-43d0-a085-f45e07d57238"
      },
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        bands  basophils  blasts  eosinophils  lymphocytes  metamyelocytes  \\\n",
              "3.jpg     0.0        1.0     0.0          0.0          0.0             0.0   \n",
              "56.jpg    0.0        0.0     0.0          1.0          0.0             0.0   \n",
              "30.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "55.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "31.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "13.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "57.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "36.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "2.jpg     0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "42.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "26.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "19.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "35.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "48.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "9.jpg     0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "25.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "4.jpg     0.0        0.0     0.0          0.0          0.0             1.0   \n",
              "59.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "45.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "15.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "39.jpg    0.0        1.0     0.0          0.0          0.0             0.0   \n",
              "10.jpg    0.0        0.0     0.0          0.0          0.0             1.0   \n",
              "29.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "17.jpg    1.0        0.0     0.0          0.0          0.0             0.0   \n",
              "43.jpg    1.0        0.0     0.0          0.0          0.0             0.0   \n",
              "38.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "52.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "37.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "53.jpg    1.0        0.0     0.0          0.0          0.0             0.0   \n",
              "41.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "22.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "1.jpg     0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "58.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "50.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "8.jpg     0.0        0.0     0.0          1.0          0.0             0.0   \n",
              "33.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "47.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "12.jpg    1.0        0.0     0.0          0.0          0.0             0.0   \n",
              "34.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "40.jpg    0.0        1.0     0.0          0.0          0.0             0.0   \n",
              "6.jpg     0.0        1.0     0.0          0.0          0.0             0.0   \n",
              "28.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "21.jpg    0.0        1.0     0.0          0.0          0.0             0.0   \n",
              "32.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "18.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "16.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "46.jpg    1.0        0.0     0.0          0.0          0.0             0.0   \n",
              "23.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "54.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "20.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "5.jpg     0.0        0.0     0.0          1.0          0.0             0.0   \n",
              "24.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "7.jpg     0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "27.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "11.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "44.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "51.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "49.jpg    0.0        0.0     0.0          0.0          0.0             0.0   \n",
              "14.jpg    0.0        0.0     0.0          0.0          1.0             0.0   \n",
              "\n",
              "        monocytes  myelocytes  neutrophils  \n",
              "3.jpg         0.0         0.0          0.0  \n",
              "56.jpg        0.0         0.0          0.0  \n",
              "30.jpg        0.0         0.0          0.0  \n",
              "55.jpg        0.0         0.0          1.0  \n",
              "31.jpg        0.0         0.0          1.0  \n",
              "13.jpg        1.0         0.0          0.0  \n",
              "57.jpg        0.0         0.0          1.0  \n",
              "36.jpg        0.0         0.0          0.0  \n",
              "2.jpg         0.0         0.0          1.0  \n",
              "42.jpg        0.0         0.0          0.0  \n",
              "26.jpg        0.0         0.0          0.0  \n",
              "19.jpg        0.0         0.0          0.0  \n",
              "35.jpg        0.0         0.0          1.0  \n",
              "48.jpg        0.0         0.0          1.0  \n",
              "9.jpg         0.0         0.0          1.0  \n",
              "25.jpg        0.0         0.0          0.0  \n",
              "4.jpg         0.0         0.0          0.0  \n",
              "59.jpg        0.0         0.0          1.0  \n",
              "45.jpg        1.0         0.0          0.0  \n",
              "15.jpg        0.0         0.0          1.0  \n",
              "39.jpg        0.0         0.0          0.0  \n",
              "10.jpg        0.0         0.0          0.0  \n",
              "29.jpg        0.0         0.0          0.0  \n",
              "17.jpg        0.0         0.0          0.0  \n",
              "43.jpg        0.0         0.0          0.0  \n",
              "38.jpg        0.0         0.0          0.0  \n",
              "52.jpg        0.0         0.0          1.0  \n",
              "37.jpg        0.0         0.0          0.0  \n",
              "53.jpg        0.0         0.0          0.0  \n",
              "41.jpg        0.0         0.0          1.0  \n",
              "22.jpg        0.0         0.0          0.0  \n",
              "1.jpg         0.0         0.0          1.0  \n",
              "58.jpg        0.0         0.0          1.0  \n",
              "50.jpg        0.0         0.0          1.0  \n",
              "8.jpg         0.0         0.0          0.0  \n",
              "33.jpg        0.0         0.0          1.0  \n",
              "47.jpg        0.0         0.0          1.0  \n",
              "12.jpg        0.0         0.0          0.0  \n",
              "34.jpg        0.0         0.0          0.0  \n",
              "40.jpg        0.0         0.0          0.0  \n",
              "6.jpg         0.0         0.0          0.0  \n",
              "28.jpg        0.0         0.0          1.0  \n",
              "21.jpg        0.0         0.0          0.0  \n",
              "32.jpg        0.0         0.0          0.0  \n",
              "18.jpg        0.0         0.0          1.0  \n",
              "16.jpg        0.0         0.0          1.0  \n",
              "46.jpg        0.0         0.0          0.0  \n",
              "23.jpg        1.0         0.0          0.0  \n",
              "54.jpg        0.0         0.0          1.0  \n",
              "20.jpg        0.0         0.0          0.0  \n",
              "5.jpg         0.0         0.0          0.0  \n",
              "24.jpg        0.0         0.0          1.0  \n",
              "7.jpg         0.0         1.0          0.0  \n",
              "27.jpg        0.0         0.0          0.0  \n",
              "11.jpg        0.0         1.0          0.0  \n",
              "44.jpg        0.0         0.0          1.0  \n",
              "51.jpg        0.0         0.0          1.0  \n",
              "49.jpg        0.0         0.0          1.0  \n",
              "14.jpg        0.0         0.0          0.0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-4853d3ae-d45b-42c6-8253-340a425c9ee0\">\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>bands</th>\n",
              "      <th>basophils</th>\n",
              "      <th>blasts</th>\n",
              "      <th>eosinophils</th>\n",
              "      <th>lymphocytes</th>\n",
              "      <th>metamyelocytes</th>\n",
              "      <th>monocytes</th>\n",
              "      <th>myelocytes</th>\n",
              "      <th>neutrophils</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>3.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>56.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>30.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>55.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>57.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>36.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>42.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>26.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>48.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>59.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>45.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>39.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17.jpg</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>43.jpg</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>38.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>52.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>53.jpg</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>41.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>58.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>33.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>47.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12.jpg</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>34.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>40.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>28.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>46.jpg</th>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>54.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>44.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>51.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14.jpg</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4853d3ae-d45b-42c6-8253-340a425c9ee0')\"\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-4853d3ae-d45b-42c6-8253-340a425c9ee0 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-4853d3ae-d45b-42c6-8253-340a425c9ee0');\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",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "sm2XsIQ8VsT4"
      },
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Future work\n",
        "\n",
        "1. Fine-tuning the VGG16 model.\n",
        "\n",
        "2. Dealing with overfitting.\n",
        "\n",
        "3. Selecting the appropriate number of layers to unfreeze by careful experimentation.\n",
        "\n",
        "4. Other parameters to consider when training the network include:\n",
        "\n",
        "Optimizers: Here, rmsprop optimizer is used to update the weights during training. When training the network, experimentation with other optimizers and their learning rate can be done.\n",
        "\n",
        "\n",
        "Dropout: Dropout is a form of regularization to prevent overfitting of the network.\n",
        "\n",
        "\n",
        "Fully-Connected Layer: If a bootstrapped approach is taken for modelling, ensure that the Fully-Connected layer is structured appropriately for the classification task. The the number of input nodes should be correct for the outputted features. Number of densely-connected layers should be apprpriate.\n"
      ],
      "metadata": {
        "id": "IRuTBD5nG62O"
      }
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "eC93acBjVsKe"
      },
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### 1. A CNN model has also been built from scratch in another notebook. But the performance evaluation of the CNN model is poor.\n",
        "\n",
        "#### 2. Also, a Resnet152 model has been trained. But it is achieving around 40% accuracy on the test data. More work has to be done to improve it."
      ],
      "metadata": {
        "id": "qM3yPueJb-N6"
      }
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "moRc73YMjDpL"
      },
      "execution_count": 40,
      "outputs": []
    }
  ]
}