{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "breastCancer.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "metadata": {
        "id": "_82qTr1dUFkj",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Kaggle"
      ]
    },
    {
      "metadata": {
        "id": "xnLjGVtMPAWP",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "!mkdir -p ~/.kaggle/"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "iV7PDpbTWakY",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "mv kaggle.json ~/.kaggle/"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "WdWMrkteEbU8",
        "colab_type": "code",
        "outputId": "cf85c4b4-f1b1-4780-b28a-ba2414e66037",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        }
      },
      "cell_type": "code",
      "source": [
        "!kaggle datasets download -d paultimothymooney/breast-histopathology-images"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Warning: Your Kaggle API key is readable by other users on this system! To fix this, you can run 'chmod 600 /root/.kaggle/kaggle.json'\n",
            "Downloading breast-histopathology-images.zip to /content\n",
            "100% 1.48G/1.49G [00:23<00:00, 101MB/s] \n",
            "100% 1.49G/1.49G [00:23<00:00, 67.3MB/s]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "QYc5Ba-0VE95",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "!mkdir images"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "EqwU2LbtVAoo",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "!unzip -q breast-histopathology-images.zip"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "4Kr1Rs0pUcQ5",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "!unzip -q IDC_regular_ps50_idx5.zip -d images"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "G5OMnAFBxWoT",
        "colab_type": "code",
        "outputId": "e8c5ef46-6587-49f3-f8f8-4675175102a8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import keras\n",
        "from keras import backend as K\n",
        "from keras.optimizers import Adam, SGD\n",
        "\n",
        "from keras.layers import Flatten, Dropout, BatchNormalization, Reshape, GlobalAveragePooling2D\n",
        "from keras.layers import Dense, Dropout, Input\n",
        "from keras.applications.mobilenet_v2 import MobileNetV2, preprocess_input\n",
        "\n",
        "from keras.models import Model\n",
        "from keras.callbacks import ReduceLROnPlateau, ModelCheckpoint, LearningRateScheduler\n",
        "\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "\n",
        "from sklearn.metrics import confusion_matrix\n",
        "import itertools\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "\n",
        "from keras.preprocessing import image\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.utils.class_weight import compute_class_weight\n",
        "\n",
        "import matplotlib.image as mpimg\n",
        "\n",
        "import shutil\n",
        "import os"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "metadata": {
        "id": "QKRHzcuXVxGl",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "images_folder = os.listdir(\"images\")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "IjHCfq7kWzWP",
        "colab_type": "code",
        "outputId": "9b3e1193-db6f-4d50-9913-f26f691d3568",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 392
        }
      },
      "cell_type": "code",
      "source": [
        "first_folder = images_folder[0]\n",
        "benign_images_first_folder = f'images/{first_folder}/0'\n",
        "benign_images = os.listdir(benign_images_first_folder)[0:3]\n",
        "malignant_images_first_folder = f'images/{first_folder}/1'\n",
        "malignant_images = os.listdir(malignant_images_first_folder)[0:3]\n",
        "\n",
        "def plot_images(image_index, folder_name, image_name, title):\n",
        "  sp = figure.add_subplot(2, 3, image_index + 1)\n",
        "  sp.axis('Off')\n",
        "  image_path = f'{folder_name}/{image_name}'\n",
        "  image_loaded = img=mpimg.imread(image_path)\n",
        "  sp.set_title(title, fontsize=16)\n",
        "  plt.imshow(image_loaded, interpolation=None)\n",
        "\n",
        "figure = plt.figure(figsize=(12, 6))\n",
        "\n",
        "for image_index, image_name in enumerate(benign_images):\n",
        "  plot_images(image_index, benign_images_first_folder, image_name, \"Benign\")\n",
        "\n",
        "for image_index, image_name in enumerate(malignant_images):\n",
        "  plot_images(image_index + 3, malignant_images_first_folder, image_name, \"Malignant\")"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAo4AAAF3CAYAAAAre8qsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 864x432 with 6 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "lC8HnFAZYpMj",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "!mkdir training && mkdir validation"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "w56vBtOqY4oH",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "!mkdir training/benign && mkdir training/malignant"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "7BgZMhLuY-0s",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "!mkdir validation/benign && mkdir validation/malignant"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "0rc3ePlDazMJ",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Move to training folder"
      ]
    },
    {
      "metadata": {
        "id": "th2Bp_Iue7l_",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "benign_images_training_folder_name = \"training/benign\"\n",
        "malignant_images_training_folder_name = \"training/malignant\""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "rzekONYnFhcX",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "for folder in images_folder:\n",
        "  benign_folder = os.listdir(f'images/{folder}/0')\n",
        "  malignant_folder = os.listdir(f'images/{folder}/1')\n",
        "  \n",
        "  for benign_image in benign_folder:\n",
        "    image_url = f'images/{folder}/0/{benign_image}'\n",
        "    shutil.move(image_url, benign_images_training_folder_name)\n",
        "    \n",
        "  for malign_image in malignant_folder:\n",
        "    image_url = f'images/{folder}/1/{malign_image}'\n",
        "    shutil.move(image_url, malignant_images_training_folder_name)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "1sZidsEEb-_p",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "benign_images = len(os.listdir(benign_images_training_folder_name))\n",
        "malignant_images = len(os.listdir(malignant_images_training_folder_name))\n",
        "total_images =  benign_images + malignant_images"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "6QE6JCrEcG-b",
        "colab_type": "code",
        "outputId": "206f0095-b5ec-42b5-986c-e5142d11c646",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "cell_type": "code",
      "source": [
        "print(f'Total images: {total_images}')\n",
        "print(f'Benign images: {benign_images}')\n",
        "print(f'Malignant images: {malignant_images}')"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Total images: 277524\n",
            "Benign images: 198738\n",
            "Malignant images: 78786\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "Olja0AE8bh7V",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Move to Validation folder"
      ]
    },
    {
      "metadata": {
        "id": "aLi-5IXJbg5i",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "validation_size = 0.20\n",
        "\n",
        "benign_validation_folder_size = int(benign_images * 0.20)\n",
        "malignant_validation_folder_size = int(malignant_images * 0.20)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "oqdsQ370dfXp",
        "colab_type": "code",
        "outputId": "8464580b-1e22-4619-a4ff-f48d0c0a57b7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "cell_type": "code",
      "source": [
        "print(f'Benign validation size: {benign_validation_folder_size}')\n",
        "print(f'Malignant validation size: {malignant_validation_folder_size}')"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Benign validation size: 39747\n",
            "Malignant validation size: 15757\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "weifQ46VgkZe",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "benign_images_validation_folder_name = \"validation/benign\"\n",
        "malignant_images_validation_folder_name = \"validation/malignant\""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "w4XQciDLdmpO",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "benign_images_training_folder = os.listdir(benign_images_training_folder_name)\n",
        "malignant_images_training_folder = os.listdir(malignant_images_training_folder_name)\n",
        "\n",
        "for image_name in benign_images_training_folder[:benign_validation_folder_size]:\n",
        "  image_url = f'{benign_images_training_folder_name}/{image_name}'\n",
        "  shutil.move(image_url, benign_images_validation_folder_name)\n",
        "  \n",
        "for image_name in malignant_images_training_folder[:malignant_validation_folder_size]:\n",
        "  image_url = f'{malignant_images_training_folder_name}/{image_name}'\n",
        "  shutil.move(image_url, malignant_images_validation_folder_name)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "qffWk7LSfo_X",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "training_folder_size = len(os.listdir(benign_images_training_folder_name)) + len(os.listdir(malignant_images_training_folder_name))\n",
        "validation_folder_size = len(os.listdir(benign_images_validation_folder_name)) + len(os.listdir(malignant_images_validation_folder_name))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "lEqYWskjgBnR",
        "colab_type": "code",
        "outputId": "0a2ff41b-57a0-46a6-835c-409fbb295bd3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "cell_type": "code",
      "source": [
        "print(f'Training folder size: {training_folder_size}')\n",
        "print(f'Validation folder size: {validation_folder_size}')"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Training folder size: 222020\n",
            "Validation folder size: 55504\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "aFu7w3UyhQsy",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Generators"
      ]
    },
    {
      "metadata": {
        "id": "LGSQbcPLhP4t",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "batch_size = 128\n",
        "img_size = 48\n",
        "input_img_size = (48, 48, 3)\n",
        "num_classes = 2"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "rV6ZP-o-hMJj",
        "colab_type": "code",
        "outputId": "3a029a27-1726-407d-935b-7e5dd87a4509",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "cell_type": "code",
      "source": [
        "train_generator = ImageDataGenerator(\n",
        "    preprocessing_function=preprocess_input,\n",
        "    rotation_range=20,\n",
        "    zoom_range=0.05,\n",
        "    width_shift_range=0.05,\n",
        "    height_shift_range=0.05,\n",
        "    shear_range=0.05,\n",
        "    horizontal_flip=True,\n",
        "    fill_mode=\"nearest\").flow_from_directory(\n",
        "    \"training\",\n",
        "    target_size=(img_size, img_size),\n",
        "    color_mode=\"rgb\",\n",
        "\t  shuffle=True,\n",
        "    batch_size=batch_size)\n",
        "\n",
        "val_generator = ImageDataGenerator(\n",
        "    preprocessing_function=preprocess_input).flow_from_directory(\n",
        "    \"validation\",\n",
        "    target_size=(img_size, img_size),\n",
        "    color_mode=\"rgb\",\n",
        "\t  shuffle=False,\n",
        "    batch_size=batch_size)"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 222020 images belonging to 2 classes.\n",
            "Found 55504 images belonging to 2 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "OsgcwsSoiVQO",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "train_steps = int(training_folder_size // batch_size)\n",
        "val_steps = int(validation_folder_size // batch_size) "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "ZhlHVPfeifqT",
        "colab_type": "code",
        "outputId": "5983c09c-fef5-4464-d3b5-b26dd64cbd6d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "train_steps, val_steps"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1734, 433)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "metadata": {
        "id": "CkMX9t7Gil6t",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Model"
      ]
    },
    {
      "metadata": {
        "id": "D86hdx7rihgP",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def create_model():\n",
        "  input_tensor = Input(shape=input_img_size)\n",
        "  \n",
        "  mobile_model = MobileNetV2(\n",
        "       weights=None,\n",
        "       input_tensor=input_tensor,\n",
        "       input_shape=input_img_size,\n",
        "       alpha=1.5,\n",
        "       include_top=False)\n",
        "  \n",
        "  for layer in mobile_model.layers:\n",
        "    layer.trainable = True\n",
        "  \n",
        "  mobile_model_output = mobile_model.output\n",
        "  classification_layer = Flatten()(mobile_model_output)\n",
        "  classification_layer = Dense(256, activation='relu')(classification_layer)\n",
        "  classification_layer = Dropout(0.5)(classification_layer)\n",
        "  predictions = Dense(activation=\"softmax\", units=num_classes)(classification_layer)\n",
        "\n",
        "  model = Model(inputs=input_tensor, outputs=predictions)\n",
        "  \n",
        "  return model"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "umAwfSoyiyO4",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "learning_rate = 3e-4\n",
        "epochs = 20"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "4S8u4nsWiztB",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def polynomial_decay(epoch):\n",
        "\tpower = 1.0\n",
        " \n",
        "\talpha = learning_rate * (1 - (epoch / float(epochs))) ** power\n",
        "\treturn alpha"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "MzsRYh6qkRHF",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "y_true = np.concatenate((val_generator.classes, train_generator.classes))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "QGTt43lYkGZE",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "cw = compute_class_weight('balanced', np.unique(y_true), y_true)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "KPmP00tplZOm",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "class_weights = {index: weight for index, weight in enumerate(cw)}"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "I9_4uMJLl-J0",
        "colab_type": "code",
        "outputId": "689b09b6-73b3-48c0-edd1-f3e9a47f0177",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "class_weights"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{0: 0.6982157413277783, 1: 1.761251999086132}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        }
      ]
    },
    {
      "metadata": {
        "id": "bOq-b2WKi1F3",
        "colab_type": "code",
        "outputId": "ec29af15-6546-47c0-de62-2d23b646a516",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 88
        }
      },
      "cell_type": "code",
      "source": [
        "weights_name = \"epoch={epoch:02d}|accuracy={val_acc:.4f}.h5\"\n",
        "\n",
        "checkpoint = ModelCheckpoint(weights_name, monitor=\"val_acc\", verbose=1, save_best_only=True,\n",
        "                                 save_weights_only=True, mode=\"max\", period=1)\n",
        "\n",
        "lr_decay = LearningRateScheduler(polynomial_decay)\n",
        "\n",
        "optimizer = SGD(lr=learning_rate, momentum=0.9)"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Colocations handled automatically by placer.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "d8SF8Y9Li7sJ",
        "colab_type": "code",
        "outputId": "27da9640-4830-44f4-86aa-e0ec07aaf193",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 88
        }
      },
      "cell_type": "code",
      "source": [
        "model = create_model()\n",
        "model.compile(loss='binary_crossentropy', optimizer=optimizer, metrics=['acc'])"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "wT-RGl9QjI9C",
        "colab_type": "code",
        "outputId": "f0788398-c1f2-4012-f6bd-8b0081b772b3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1448
        }
      },
      "cell_type": "code",
      "source": [
        "trained_model = model.fit_generator(train_generator,\n",
        "                        epochs=epochs,\n",
        "                        class_weight=class_weights,\n",
        "                        steps_per_epoch=train_steps,\n",
        "                        callbacks=[checkpoint, lr_decay],\n",
        "                        validation_data=val_generator,\n",
        "                        validation_steps=val_steps,\n",
        "                        verbose=1)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Use tf.cast instead.\n",
            "Epoch 1/20\n",
            "1734/1734 [==============================] - 1073s 619ms/step - loss: 0.4558 - acc: 0.8023 - val_loss: 0.3677 - val_acc: 0.8448\n",
            "\n",
            "Epoch 00001: val_acc improved from -inf to 0.84478, saving model to epoch=01|accuracy=0.8448.h5\n",
            "Epoch 2/20\n",
            "1734/1734 [==============================] - 1044s 602ms/step - loss: 0.4011 - acc: 0.8272 - val_loss: 0.4111 - val_acc: 0.8396\n",
            "\n",
            "Epoch 00002: val_acc did not improve from 0.84478\n",
            "Epoch 3/20\n",
            "1734/1734 [==============================] - 1045s 603ms/step - loss: 0.3841 - acc: 0.8336 - val_loss: 0.3658 - val_acc: 0.8501\n",
            "\n",
            "Epoch 00003: val_acc improved from 0.84478 to 0.85013, saving model to epoch=03|accuracy=0.8501.h5\n",
            "Epoch 4/20\n",
            "1734/1734 [==============================] - 1040s 600ms/step - loss: 0.3748 - acc: 0.8370 - val_loss: 0.4060 - val_acc: 0.8155\n",
            "\n",
            "Epoch 00004: val_acc did not improve from 0.85013\n",
            "Epoch 5/20\n",
            "1734/1734 [==============================] - 1049s 605ms/step - loss: 0.3676 - acc: 0.8411 - val_loss: 0.3786 - val_acc: 0.8337\n",
            "\n",
            "Epoch 00005: val_acc did not improve from 0.85013\n",
            "Epoch 6/20\n",
            "1734/1734 [==============================] - 1044s 602ms/step - loss: 0.3633 - acc: 0.8420 - val_loss: 0.3482 - val_acc: 0.8598\n",
            "\n",
            "Epoch 00006: val_acc improved from 0.85013 to 0.85978, saving model to epoch=06|accuracy=0.8598.h5\n",
            "Epoch 7/20\n",
            "1734/1734 [==============================] - 1048s 605ms/step - loss: 0.3596 - acc: 0.8437 - val_loss: 0.3772 - val_acc: 0.8391\n",
            "\n",
            "Epoch 00007: val_acc did not improve from 0.85978\n",
            "Epoch 8/20\n",
            "1734/1734 [==============================] - 1049s 605ms/step - loss: 0.3548 - acc: 0.8457 - val_loss: 0.3271 - val_acc: 0.8615\n",
            "\n",
            "Epoch 00008: val_acc improved from 0.85978 to 0.86146, saving model to epoch=08|accuracy=0.8615.h5\n",
            "Epoch 9/20\n",
            "1734/1734 [==============================] - 1053s 607ms/step - loss: 0.3512 - acc: 0.8464 - val_loss: 0.3410 - val_acc: 0.8553\n",
            "\n",
            "Epoch 00009: val_acc did not improve from 0.86146\n",
            "Epoch 10/20\n",
            "1734/1734 [==============================] - 1050s 605ms/step - loss: 0.3491 - acc: 0.8469 - val_loss: 0.3256 - val_acc: 0.8644\n",
            "\n",
            "Epoch 00010: val_acc improved from 0.86146 to 0.86440, saving model to epoch=10|accuracy=0.8644.h5\n",
            "Epoch 11/20\n",
            "1734/1734 [==============================] - 1052s 607ms/step - loss: 0.3451 - acc: 0.8487 - val_loss: 0.3150 - val_acc: 0.8650\n",
            "\n",
            "Epoch 00011: val_acc improved from 0.86440 to 0.86505, saving model to epoch=11|accuracy=0.8650.h5\n",
            "Epoch 12/20\n",
            "1734/1734 [==============================] - 1067s 615ms/step - loss: 0.3431 - acc: 0.8497 - val_loss: 0.3537 - val_acc: 0.8466\n",
            "\n",
            "Epoch 00012: val_acc did not improve from 0.86505\n",
            "Epoch 13/20\n",
            "1734/1734 [==============================] - 1069s 617ms/step - loss: 0.3408 - acc: 0.8514 - val_loss: 0.3410 - val_acc: 0.8544\n",
            "\n",
            "Epoch 00013: val_acc did not improve from 0.86505\n",
            "Epoch 14/20\n",
            "1734/1734 [==============================] - 1049s 605ms/step - loss: 0.3390 - acc: 0.8512 - val_loss: 0.3256 - val_acc: 0.8603\n",
            "\n",
            "Epoch 00014: val_acc did not improve from 0.86505\n",
            "Epoch 15/20\n",
            "1734/1734 [==============================] - 1058s 610ms/step - loss: 0.3366 - acc: 0.8522 - val_loss: 0.3063 - val_acc: 0.8687\n",
            "\n",
            "Epoch 00015: val_acc improved from 0.86505 to 0.86872, saving model to epoch=15|accuracy=0.8687.h5\n",
            "Epoch 16/20\n",
            "1734/1734 [==============================] - 1045s 603ms/step - loss: 0.3357 - acc: 0.8530 - val_loss: 0.3421 - val_acc: 0.8530\n",
            "\n",
            "Epoch 00016: val_acc did not improve from 0.86872\n",
            "Epoch 17/20\n",
            "1734/1734 [==============================] - 1043s 601ms/step - loss: 0.3349 - acc: 0.8533 - val_loss: 0.3329 - val_acc: 0.8576\n",
            "\n",
            "Epoch 00017: val_acc did not improve from 0.86872\n",
            "Epoch 18/20\n",
            "1734/1734 [==============================] - 1045s 603ms/step - loss: 0.3336 - acc: 0.8537 - val_loss: 0.3246 - val_acc: 0.8605\n",
            "\n",
            "Epoch 00018: val_acc did not improve from 0.86872\n",
            "Epoch 19/20\n",
            "1734/1734 [==============================] - 1045s 603ms/step - loss: 0.3327 - acc: 0.8545 - val_loss: 0.3354 - val_acc: 0.8548\n",
            "\n",
            "Epoch 00019: val_acc did not improve from 0.86872\n",
            "Epoch 20/20\n",
            "1734/1734 [==============================] - 1051s 606ms/step - loss: 0.3321 - acc: 0.8544 - val_loss: 0.3277 - val_acc: 0.8581\n",
            "\n",
            "Epoch 00020: val_acc did not improve from 0.86872\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "1D2eDrvHjLVv",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def plot_validation_training(metric, trained_model):\n",
        "  validation_metric = trained_model.history[f'val_{metric}']\n",
        "  training_metric = trained_model.history[metric]\n",
        "  epochs = range(len(training_metric))\n",
        "  plt.plot(epochs, training_metric, 'b', label=f'Training {metric}')\n",
        "  plt.plot(epochs, validation_metric, 'r', label=f'Validation {metric}')\n",
        "  plt.ylim(bottom=0)\n",
        "  plt.xlabel('Epochs ', fontsize=16)\n",
        "  plt.ylabel(metric, fontsize=16)\n",
        "  loc = 'upper right' if metric == \"loss\" else 'lower right'\n",
        "  plt.legend(loc=loc)\n",
        "  plt.title(f'Training and validation {metric}', fontsize = 20)\n",
        "  plt.show()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "-6vbh3hQjQZv",
        "colab_type": "code",
        "outputId": "b87558dd-0ca0-44e4-c842-fdd5e1d8ff6b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 387
        }
      },
      "cell_type": "code",
      "source": [
        "plot_validation_training(\"loss\", trained_model)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfQAAAFyCAYAAAADPyhPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAQAwAIEOAIABCHQAAAzwf/zgT8v3TvEvAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 576x396 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "Q36sAG4-jRwI",
        "colab_type": "code",
        "outputId": "bb85c1a3-0689-44e0-ae7b-7df796ccc98b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 387
        }
      },
      "cell_type": "code",
      "source": [
        "plot_validation_training(\"acc\", trained_model)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfQAAAFyCAYAAAADPyhPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 576x396 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "o7J1YvjGjTs7",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def plot_confusion_matrix(cm, classes,\n",
        "                          normalize=False,\n",
        "                          title='Confusion matrix',\n",
        "                          cmap=plt.cm.Blues):\n",
        "    \"\"\"\n",
        "    This function prints and plots the confusion matrix.\n",
        "    Normalization can be applied by setting `normalize=True`.\n",
        "    \"\"\"\n",
        "    if normalize:\n",
        "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
        "        print(\"Normalized confusion matrix\")\n",
        "    else:\n",
        "        print('Confusion matrix, without normalization')\n",
        "\n",
        "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
        "    plt.title(title)\n",
        "    plt.colorbar()\n",
        "    tick_marks = np.arange(len(classes))\n",
        "    plt.xticks(tick_marks, classes, rotation=45)\n",
        "    plt.yticks(tick_marks, classes)\n",
        "\n",
        "    fmt = '.2f' if normalize else 'd'\n",
        "    thresh = cm.max() / 2.\n",
        "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
        "        plt.text(j, i, format(cm[i, j], fmt),\n",
        "                 horizontalalignment=\"center\",\n",
        "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
        "\n",
        "    plt.ylabel('True label')\n",
        "    plt.xlabel('Predicted label')\n",
        "    plt.tight_layout()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "kfaMMhrxs3-T",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model.load_weights('epoch=15_accuracy=0.8687.h5')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "3iy9IjpijVUw",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "val_generator.reset()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "h2zPTlBOjXHF",
        "colab_type": "code",
        "outputId": "f7a35465-d3c3-469f-ffd5-c9ce784d9971",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "classes_predicted = model.predict_generator(val_generator, steps=val_steps, verbose=1)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "433/433 [==============================] - 77s 177ms/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "LOGeMc1CjY3U",
        "colab_type": "code",
        "outputId": "8f49fb82-0b74-447d-a777-8d5ba45a402b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "len(classes_predicted)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "55424"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 43
        }
      ]
    },
    {
      "metadata": {
        "id": "JXkMfIxEjaz6",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "real_classes = np.argmax(classes_predicted, axis=1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "2QVsGDXgjcM-",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "val_labels = val_generator.classes"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "0svhad3D3SUs",
        "colab_type": "code",
        "outputId": "87720e1e-69af-42df-9cca-9571ce7a7ccc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "len(val_labels)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "55504"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 46
        }
      ]
    },
    {
      "metadata": {
        "id": "QoqePxngjds5",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "val_labels = val_labels[:55424]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "MMDBs85_jgQz",
        "colab_type": "code",
        "outputId": "e32fa14c-e5e5-430e-cd31-c7f0d56d6a1f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "len(val_labels), len(real_classes)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(55424, 55424)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 48
        }
      ]
    },
    {
      "metadata": {
        "id": "MLP3V48HjiL0",
        "colab_type": "code",
        "outputId": "4347503c-3a9f-429d-a231-85e1328df6d9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "val_generator.class_indices"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'benign': 0, 'malignant': 1}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 49
        }
      ]
    },
    {
      "metadata": {
        "id": "-I0hPHU9jjyf",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "classes_names = [\"Benign\", \"Malignant\"]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "fVooRo5OjuIb",
        "colab_type": "code",
        "outputId": "abbb52e8-96d0-42cf-fc99-dba51a553848",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 418
        }
      },
      "cell_type": "code",
      "source": [
        "cm = confusion_matrix(val_labels, real_classes, labels=range(num_classes))\n",
        "plot_confusion_matrix(cm, classes_names)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Confusion matrix, without normalization\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcQAAAGACAYAAAAzlhJlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3XdY1fX7x/HnOXAQVFBAcCSamgO/\nCjlyYA40crQst0laWFaaWaYhGWmpODLNHGk5SHKF/pLUnKFfB5KKqVRqmZYTQcQByP794eW54us4\nRDLP6+F1rjif8xn3+2jc536PzzHk5OTkICIiYuWMRR2AiIhIcaCEKCIighKiiIgIoIQoIiICKCGK\niIgASogiIiKAEqKUcDk5OSxevJgnn3ySzp0789hjjzFu3DiuXbv2r877zjvv0L59e3bu3PmPjz18\n+DABAQH/6vr324YNG7h+/fodX5s+fTrLly8v5IhEih+D1iFKSTZt2jR+/PFHZs+eTeXKlUlJSWHi\nxImcPHmSr7/+GoPBkK/zenp6smnTJmrUqHGfIy4aXbp0YcmSJVSpUqWoQxEptlQhSomVlJTE0qVL\nmTx5MpUrVwagbNmyBAcHM3jwYHJyckhLSyM4OJjOnTvTtWtXJk+eTFZWFgAdO3ZkxYoV9OzZk0cf\nfZTJkycD4O/vT3Z2NgEBAezYsYOOHTuyf/9+83VvPc/MzOS9996jc+fO+Pn5MWzYMK5fv050dDR+\nfn4A+br+//L392fBggX06dOHVq1a8fXXXzN37ly6dOlCt27dOH36NAB//PEH/fr1o2vXrvj5+bFu\n3ToAxowZw8mTJ/H392f//v0EBgYSEhLCU089xffff09gYCBz587l8OHDdOjQgeTkZAA+//xzhg8f\nfr//2kSKLSVEKbEOHTpElSpVqFOnTq7tZcqUoWPHjhiNRkJDQ7lw4QLr16/n//7v/9i/f785UQDs\n27ePlStXsnr1asLCwrhw4QJLly4FYOnSpbRv3/6u19+1axdnzpxh48aNbN68mYceeoiDBw/m2ic/\n17+Tffv28fXXXxMSEsK0adOoUqUKGzdu5KGHHmL16tUATJ06FV9fX77//nsmTZrEe++9R0ZGBiEh\nIeb2NG/eHICoqCjCw8Pp2rWr+RpeXl489thjzJ8/n7i4OJYtW8bYsWMt/j2IlBZKiFJiJSUl4erq\nes99tm/fTu/evbG1tcXe3p6nnnqK3bt3m19/6qmnsLGxoXLlyri6unL+/Pk8X9/FxYUTJ06wZcsW\nUlNTGTFiBG3bti2Q6/v6+mJra0u9evVITU2lc+fOANSrV4+LFy8CMHfuXPPYZbNmzUhLSyM+Pv6O\n52vdujVlypS5bftbb73Fxo0bGTNmDK+//jru7u55fj9ESjolRCmxnJ2diYuLu+c+iYmJVKhQwfy8\nQoUKXLp0yfy8fPny5p9tbGzM3Zl54eXlxdixY1m6dClt2rRh5MiRXL16tUCuX65cOfM+f39uNBrJ\nzs4GYOfOnTz//PN07tyZbt26kZOTY37tf/09pv+9TteuXTlw4ABPPfXUPdsvUtooIUqJ9fDDD3Pp\n0iV+/vnnXNszMjKYMWMGqampVKpUiaSkJPNrSUlJVKpU6R9d5+9JB+DKlSvmn7t06cLSpUuJjIwk\nNTWVhQsX5jr2flw/LzIyMhgxYgSvvfYamzZtIiIiIl8TiuLi4vjuu+944oknmD179n2PU6Q4U0KU\nEsvJyYnBgwfz7rvv8ueffwKQmppKcHAwv/zyCw4ODnTo0IHw8HCysrJISUlh7dq19xwXvBM3NzeO\nHj0K3Fy+kJaWBsDq1auZM2cOABUrVqR27dq3HXs/rp8XqamppKSk0KhRI+Dm2KXJZCIlJQUAW1vb\n26rXO5k4cSKDBw8mKCiI77//nl9//fW+xypSXCkhSon2xhtv0Lt3b1577TU6d+7Mc889h6urq7m6\n8ff3p0qVKjzxxBP06NGDDh065JpIkhevv/46S5Ys4cknn+TEiRM89NBDAHTq1Imff/6Zxx9/nK5d\nu/L777/z4osv5jr2flw/L259OOjevTvdu3enRo0aPPbYY7z66qukpKTQpUsX+vbty4YNG+56ju3b\nt3PmzBn69u1L+fLleeuttxg7duw/6kYWKcm0DlFERARViCIiIoASooiICKCEKCIiAighioiIAEqI\nIiIiANgWdQDFnUOTYUUdQomw/5sgmveaVNRhFHvHtn5c1CGUCFUq2HHhSnpRh1Ei1HC1L5Tr/Jvf\nhakHS8ZNHpQQ5b74z0PVijoEKUXsbNV5VewYSv/fSelvoYiISB6oQhQREcvy+WXbJYkSooiIWGYF\nXaZKiCIiYpkqRBEREVQhioiIAKoQRUREAFWIIiIigFVUiKU/5YuIiOSBKkQREbFMXaYiIiJYRZep\nEqKIiFimClFERARViCIiIoAqRBEREcAqEmLpb6GIiEgeqEIUERHLjBpDFBERsYouUyVEERGxTLNM\nRUREUIUoIiICqEIUEREBrKJCLP0tFBERyQNViCIiYpm6TEVERLCKLlMlRBERsUwVooiICKoQRURE\nAFWIIiIigCpEERERwCoSYulvoYiISB6oQhQREcs0higiIkKBdZmmpqYSGBjIpUuXSEtL4/XXX6dB\ngwaMHj2arKws3NzcmDZtGnZ2dkRERBAaGorRaKR379706tWLjIwMAgMDOXfuHDY2NoSEhODh4cHR\no0cZN24cAPXr12f8+PEWY1GXqYiIWGYw5P9xD5GRkTRq1IiwsDBmzpzJ5MmTmTVrFv3792fZsmXU\nrFmT8PBwUlJSmDNnDkuWLGHp0qWEhoaSlJTEunXrcHJyYvny5bz66qtMnz4dgIkTJxIUFMSKFSu4\nfv06O3bssNhEJUQREbHMYMz/4x66devGyy+/DMD58+epXLky0dHRdOrUCQBfX1+ioqI4dOgQjRs3\nxtHREXt7e5o2bUpMTAxRUVH4+fkB4OPjQ0xMDOnp6Zw9exYvL69c57BEXaYiImJZAY8h9u3blwsX\nLvD555/z4osvYmdnB4Crqyvx8fEkJCTg4uJi3t/FxeW27UajEYPBQEJCAk5OTuZ9b53DEiVEERGx\nyFDACXHFihX8+uuvjBo1ipycHPP2v//8d/9k+932/V/qMhURkSITGxvL+fPnAfD09CQrK4ty5cpx\n48YNAOLi4nB3d8fd3Z2EhATzcRcvXjRvv1X9ZWRkkJOTg5ubG0lJSeZ9b53DEiVEERGxyGAw5Ptx\nL/v372fRokUAJCQkkJKSgo+PD5s2bQJg8+bNtG3bFm9vb44cOcLVq1dJTk4mJiaG5s2b06ZNGzZu\n3AjcnKDTsmVLTCYTtWvXZv/+/bnOYYm6TEVExLIC6jHt27cv7733Hv379+fGjRsEBwfTqFEj3n33\nXVauXEm1atXo3r07JpOJkSNHEhAQgMFgYOjQoTg6OtKtWzf27NlDv379sLOzY/LkyQAEBQURHBxM\ndnY23t7e+Pj4WG5iTl47V62UQ5NhRR1CiZB6cLbeqzw4tvXjog6hRKjhas9fl24UdRglQg1X+0K5\nTvneS/J97PVVg+5bHAVJFaKIiFhU0JNqigMlRBERsUgJUUREBOtIiJplKiIigipEERHJi9JfICoh\nioiIZdbQZaqEKCIiFikhioiIoIQoIiICKCGKiIjcVPrzoRKiiIhYZg0VotYhioiIoApRRETywBoq\nRCVEERGxSAlRREQENKlGREQEVCGKiIgASogiIiKAdSRELbsQERFBFaKIiOSBNVSISogiImJZ6c+H\nSogiImKZKkSxGt07PUzgy12wt7PlUlIyb0xcwS8nzptfD3nrWZ597GEaPPEBABXKOzB/3PM0fKga\n6RmZuc6VenA2x05eMD8/d/EK3V79DIAGtauwYNwAXCuWI/FKMi9/EMbRPy4gpdfpv/6kQ4tG1Hyw\ntnmbd9PmzJi7kH17dxP0znBu3Ejlgeo1+HTeIipXrUZcXByvvPQyx4/+isFg4MMpM2jboRMAOTk5\nzJ89g2kTP2DFtxt5pFWbomqaVVFCFKvgUcWZWe/14dHnp/LX+csM7deB+eOep63/xwA0rvcAT3Xw\nynXMR28+w+kLl+n7zpc84F6R3zdNoJpbBc7FXwHg4ecm3HYdo9HAio8H88Hs71j7wyH6dnuEF5/1\n4d3pawq+kVKkqlStxg97D+Xadu3aVV4PGMD80BU0bd6SebM+Zu2aVbwydATDhw+nxoO1WfDVKo7+\nEsvzPZ5gx4+xlHd05L13hpOVlYVrJbciao11soaEqFmmQkZmFoPGhPLX+csARP54jLo1KwM3/yeY\nFdSH8XPX5Trmucea8EX4LgDOXkwC4IkOje95ndbetcnMymbtDzd/Ma7YsE/J0Ipt+X4djbya0LR5\nSwBeG/4OrwwdcfO1LVvo/fxAABo0bEQj7ybs3hkJQI++zzNl5lxMJlPRBG6tDP/iUUIoIQoXEq7y\nQ/RRAGxsjPg/3Yp1Ow4DMLhnG2J/O0f04ZPm/V0qlMO1Yjn+OBOf6zz1H6xs/nnRhBeIWf0eWxaO\noJV3LeBmpfnX+UQWjB/A4W+DWTPrVWpWcy3o5kkxcO3aNV7270XHVt680Ptpfjt+lF9iD+Pi6sor\nL/SmQ4vGDBvsT+KlBODmB7HsrCzz8eXKlePUHycAaPZIqyJpg5R+SohiNrRfB/7cGkKbJnUY++la\nKrs6Mqy/L+/PWptrv7L2JrKyssnMzM693aEMAAtX7+aT0K007TGRz1fsIHzmq1Qo70AFRwcebfoQ\nX3yzE+9nP+LQ0TMsmvBCobVPikb58uV5pkdvgid+zNY9B3m0fUdeHtCLq1eu8N/IrQSNC2HL7hjs\nypRh/HujAPDz82Ph/NlkZWXx689H2LNzB2lpaUXcEutmMBjy/Sgpik1CPHPmDE2aNMHf3x9/f3/6\n9OnD/v37/9E51qxZw5YtWwoowtJvzvLtVPd9l9nLIolc8jazgvoSsuB7kq6l5tovOTUdGxsjJlub\n3NtTbv7CGjZhObG/nQNg9ZaDnLuYRKuHa3H1+g0OHzvDvtg/ycnJYVbYD7TyrkVZe7vCaaAUCWcX\nVz6aMhOPGjUxGo28/PqbJMRfJOnyJdq08+XB2nUwmUy8+MpQdm7fBsCsWbO4eiWJTq0fZvYnU2jf\n0Q+nChWKuCXWzRoSYrGaVFOrVi2WLl0KwL59+5g3bx4LFy7M8/HPPfdcQYVWqtWvVZlq7hWJjD4G\nwKqNB/jk3d74tqxPC69ahLz9HDY2BlycynFyyyTqdwvmYuI1antU4tjJOPN5fv3jAuUc7KjmXpHf\n/rxo3m5rayQjI5u/zifiVN7BvD0rOzvXf6V0upJ0mStXrlCj5oPmbVnZWbRq047d/91u3mZjY4PR\nePNDlru7O/OXrDC/1rd7Fxp4NiqskOUOSlJiy69iUyH+r4SEBNzd3YmLi2Pw4MEMHDiQl156iXPn\nblYefn5+TJkyhT59+jB48GCys7P57LPPCAsLIyMjgxEjRtC7d29CQkJo164dAP7+/sybN4+BAwfy\n9NNPm89l7So5l2fhRy9Q1e3mJ/DW3rUx2Rqp03kstfyCqOUXxKPPT+NM3GVq+QWRnpHJms0xDOvv\nC9xcSgGwbvthqldxZnvoSGp7VAKgU6sGuFYsz77YU0RGH6OKmxOdWjUAIOC5NkT99Adp6Zl3iEpK\ni0MHD9Dv2S5cSrg55rz8q0VUe8CDx7s9RfSenRz9JRaAZV8t4tH2N/9NDRs2jC/nzQIgatd/iTt/\njkda+RRNAwRQhVjoTp48ib+/P2lpacTFxbFw4UI+/fRTXnrpJXx8fNixYwdz585lwoQJnD59mmee\neYZ3332X3r17c+zYMfN5du7cSVpaGqtWrSIyMpLQ0FDza+XLlyc0NJSPP/6YzZs3M2jQoCJoafGy\nO+YEU77cxPrPh2E0GEjLyOSFMYu5lnzjrscEz/6OL8YPIHbtB9xIzwDgYuI1LiZeY9THqwmfOQSj\nwUDStVR6v7XAfK6+b3/BZ2P7UcZky1/nE3n5g7BCaaMUnXa+j+H/4iv06NYRo9FI5arV+HzJcqp7\n1GTaZ/N5ZWAfDBio79mQkE/mADcTYu++zxO68HMqVHBm3uJl2NjcrB79Hm1GVmYmF86f481XX8Te\n3oFP5n7Jw00fKcpmln4lJ6/lmyEnJyenqIOAm2OIw4cPZ82am9PwT5w4wYgRI8jMzKRixYrY2tqS\nlZWFi4sLs2fPpnnz5uYxxuHDhzNgwACio6NxdnYmNfXmmNfLL79MdnY2Xl5exMbG4u/vT1BQEJ6e\nnoSFhZGUlMSwYcPuGdfPv5/jPw9VK9jGi4jkw1+XblDD1b5QrlX77Q35PvaPT7rdx0gKTrGqEP+u\nTp06lClThnPnzhEaGoq7u3uu1299Wrzl73k9JyfH/Pr/lut/Py4vnwWa95r0j2O3RqkHZ+PQ5N4f\nLgSObf24qEMoEWq42vPXpbv3UEjhK0ldn/lVbMcQk5KSiI+P5/HHH2fr1q0AREVF8d1331k8tkaN\nGsTG3hyX2LVrF1l/W88kIiJyJ8WqQrw1hgiQlpbG+++/T+PGjQkKCmL9+vUYDAZCQkIsnsfX15fV\nq1fTr18/WrRoQcWKFQs6dBGRUs0KCsTiM4Z4PyUlJREdHU3nzp2Ji4tj4MCBbNy4MV/nUjdg3qjL\nNG/UZZo36jLNu8IaQ6w7Kn+/QwF+m9blPkZScIpVhXi/lCtXju+//56FCxeSnZ3NmDFjijokEZES\nzRoqxFKZEE0mEzNnzizqMERESo2CnFQzdepUDhw4QGZmJkOGDOGHH37g559/Ng93BQQE0KFDByIi\nIggNDcVoNNK7d2969epFRkYGgYGBnDt3DhsbG0JCQvDw8ODo0aOMGzcOgPr16zN+/HiLcZTKhCgi\nIvdXQeXDvXv38ttvv7Fy5UouX77Ms88+S6tWrXj77bfx9fU175eSksKcOXMIDw/HZDLRs2dP/Pz8\niIyMxMnJienTp7Nr1y6mT5/OzJkzmThxIkFBQXh5eTFy5Eh27NhB+/bt7xlLsZ1lKiIixYfRaMj3\n414eeeQRPv30UwCcnJxITU2948qAQ4cO0bhxYxwdHbG3t6dp06bExMQQFRWFn58fAD4+PsTExJCe\nns7Zs2fx8rr5Pa6+vr5ERUVZbuM/fVNERETuFxsbG8qWLQtAeHg47dq1w8bGhrCwMF544QXeeust\nEhMTSUhIwMXFxXyci4sL8fHxubYbjUYMBgMJCQk4OTmZ93V1dSU+PvfX1d2JukxFRMSigp5Us3Xr\nVsLDw1m0aBGxsbFUrFgRT09PFixYwOzZs2nSpEmu/e+2QOJO2/O6mEIVooiIWFSQN/feuXMnn3/+\nOV988QWOjo60bt0aT09PADp27Mjx48dxd3cnISHBfMzFixdxd3fH3d3dXP1lZGSQk5ODm5sbSUlJ\n5n3j4uJuu9vZnSghioiIRQZD/h/3cu3aNaZOncr8+fPNs0rfeOMNTp8+DUB0dDR169bF29ubI0eO\ncPXqVZKTk4mJiaF58+a0adPGvM48MjKSli1bYjKZqF27tvl+15s3b6Zt27YW26guUxERsaigll1s\n2LCBy5cvM2LECPO25557jhEjRuDg4EDZsmUJCQnB3t6ekSNHEhAQgMFgYOjQoTg6OtKtWzf27NlD\nv379sLOzY/LkyQAEBQURHBxMdnY23t7e+PhY/vqwUnmnmvtJd1/JG92pJm90p5q80Z1q8q6w7lTj\n/cG2fB97aHyn+xhJwVGFKCIiFlnDnWo0higiIoIqRBERyQNr+D5EJUQREbHICvKhEqKIiFimClFE\nRARViCIiIoAqRBEREUAVooiICGAdFaLWIYqIiKAKUURE8sAKCkQlRBERscwaukyVEEVExCIryIdK\niCIiYpkqRBEREVQhioiIANZRIWrZhYiICKoQRUQkD6yhQlRCFBERi6wgHyohioiIZaoQRUREUIUo\nIiICqEIUEREBrKNC1LILERERVCGKiEgeGK2gRFRCFBERi6wgHyohioiIZZpUIyIiAhhLfz5UQhQR\nEctUIYqIiKAxRBEREQAMlP6MqHWIIiIiqEIUEZE80KQaERERNKlGREQE0KQaERERwMpv3RYeHn7P\nA3v27HnfgxERkeLJCvLh3RPigQMH7nmgEqKIiPWw6jHEkJAQ88/Z2dlcunQJNze3QglKRESsx9Sp\nUzlw4ACZmZkMGTKExo0bM3r0aLKysnBzc2PatGnY2dkRERFBaGgoRqOR3r1706tXLzIyMggMDOTc\nuXPY2NgQEhKCh4cHR48eZdy4cQDUr1+f8ePHW4zD4jrEqKgoHnvsMfz9/QGYNGkS27dv/1eNFxGR\nksVgyP/jXvbu3ctvv/3GypUr+fLLL5k0aRKzZs2if//+LFu2jJo1axIeHk5KSgpz5sxhyZIlLF26\nlNDQUJKSkli3bh1OTk4sX76cV199lenTpwMwceJEgoKCWLFiBdevX2fHjh0W22gxIc6YMYNVq1aZ\nq8NXX32VuXPn5uHtExGR0sJoMOT7cS+PPPIIn376KQBOTk6kpqYSHR1Np06dAPD19SUqKopDhw7R\nuHFjHB0dsbe3p2nTpsTExBAVFYWfnx8APj4+xMTEkJ6eztmzZ/Hy8sp1DotttLRD2bJlqVSpkvm5\ni4sLJpPJ4olFRKT0MPyLx73Y2NhQtmxZ4OZkznbt2pGamoqdnR0Arq6uxMfHk5CQgIuLi/k4FxeX\n27YbjUYMBgMJCQk4OTmZ9711DkssJkR7e3t+/PFHAK5cucKyZcsoU6aMxROLiEjpYTAY8v3Ii61b\ntxIeHk5wcHCu7Tk5OXfc/59sv9u+/8tiQvzggw9YuHAhR44cwc/Pj507d/Lhhx/m6eQiIlI6GA35\nf1iyc+dOPv/8c7744gscHR0pW7YsN27cACAuLg53d3fc3d1JSEgwH3Px4kXz9lvVX0ZGBjk5Obi5\nuZGUlGTe99Y5LLbR0g5Vq1Zl/vz5/PTTT/z444/MmzeP6tWrW26hiIiUGgVVIV67do2pU6cyf/58\nKlasCNwcC9y0aRMAmzdvpm3btnh7e3PkyBGuXr1KcnIyMTExNG/enDZt2rBx40YAIiMjadmyJSaT\nidq1a7N///5c57DE4p1q9u3bx+TJkzlx4gQGg4F69eoxevRomjVrZvHkIiJSOhTUMsQNGzZw+fJl\nRowYYd42efJkxo4dy8qVK6lWrRrdu3fHZDIxcuRIAgICMBgMDB06FEdHR7p168aePXvo168fdnZ2\nTJ48GYCgoCCCg4PJzs7G29sbHx8fy23MsdC5+tRTTxEUFETTpk3JycnhwIEDTJkyhYiIiH/5NpQM\nDk2GFXUIJULqwdl6r/Lg2NaPizqEEqGGqz1/XbpR1GGUCDVc7QvlOv5fH8r3sUuf976PkRQcixWi\nq6srrVu3Nj9v06YN1apVK9CgRESkeLHqO9WcPn0agMaNG7No0SJ8fHwwGo1ERUXRsGHDQgtQRESK\nnlV/H+LAgQMxGAzm6aphYWHm1wwGA8OHDy/46EREpFiw6grxhx9+uOtBMTExBRKMiIgUT6U/HeZh\nDPH69eusXbuWy5cvAzfXeaxevZpdu3YVeHAiIlI8WMP3IVpchzhixAiOHTvGmjVrSE5OJjIy0nwH\ncRERsQ4FdXPv4sRiQkxLS+PDDz/kgQce4N133+Wrr77i+++/L4zYRERECo3FLtOMjAxSUlLIzs7m\n8uXLODs7m2egioiIdbDqSTW3PPPMM6xatYpevXrRrVs3XFxcqFGjRmHEJiIixYQV5EPLCbFfv37m\nn1u3bs2lS5e0DlFExMpYw6SauybEW1/YeCdbtmzhzTffLJCARESk+LGCfHj3hGhjY1OYcRRbl/fN\nLuoQSgy9V5aFHzpT1CGUCANcq/PfUwmWdxQGuBbOtw9Z9RjisGG6UbOIiNxkcUlCKWANbRQREbHI\n4qQaERERa+gyzVOFePnyZY4cOQJAdnZ2gQYkIiLFj9GQ/0dJYTEhrlu3jj59+jBmzBgAPvroI775\n5psCD0xERIoPJURg8eLFrF27FmdnZwDeffddVq1aVeCBiYhI8WEwGPL9KCksjiE6Ojri4OBgfm5v\nb4/JZCrQoEREpHgpSZVefllMiM7Ozvzf//0faWlp/Pzzz2zYsAEXF5fCiE1ERIqJElTo5ZvFLtPx\n48dz5MgRkpOTGTt2LGlpaUyYMKEwYhMRkWLCaDDk+1FSWKwQnZycCA4OLoxYREREiozFhNi+ffs7\nDopu3769IOIREZFiyBru4mIxIS5btsz8c0ZGBlFRUaSlpRVoUCIiUryUoJ7PfLOYEB944IFczx98\n8EECAgIYNGhQQcUkIiLFTEkaC8wviwkxKioq1/MLFy7w119/FVhAIiJS/FhBPrScEOfOnWv+2WAw\nUL58ecaPH1+gQYmISPGidYhAYGAg//nPfwojFhERKaasocvU4sShKVOmFEYcIiIiRcpihVitWjX8\n/f3x9vbOdcu2N998s0ADExGR4sMKCkTLCbF69epUr169MGIREZFiyqrHECMiInj66acZNmxYYcYj\nIiLFkIHSnxHvOoYYHh5emHGIiEgxZg3fh2ixy1RERKQkJbb8umtCPHjwIB06dLhte05ODgaDQfcy\nFRGxIiXpi37z664JsWHDhnzyySeFGYuIiEiRuWtCtLOzu+0+piIiYp2susvUy8urMOMQEZFirKB7\nTI8fP87rr7/OoEGDGDBgAIGBgfz8889UrFgRgICAADp06EBERAShoaEYjUZ69+5Nr169yMjIIDAw\nkHPnzmFjY0NISAgeHh4cPXqUcePGAVC/fn2Ltx29a0IcNWrU/WupiIiUaAV567aUlBQ++ugjWrdu\nnWv722+/ja+vb6795syZQ3h4OCaTiZ49e+Ln50dkZCROTk5Mnz6dXbt2MX36dGbOnMnEiRMJCgrC\ny8uLkSNHsmPHDtq3b3/3NhZYC0VEpNQoyGUXdnZ2fPHFF7i7u99zv0OHDtG4cWMcHR2xt7enadOm\nxMTEEBUVhZ+fHwA+Pj7ExMSQnp7O2bNnzb2dvr6+t317021tzNtbISIi1sxgyP/DEltbW+zt7W/b\nHhYWxgsvvMBbb71FYmIiCQkF4S76AAAgAElEQVQJuLi4mF93cXEhPj4+13aj0YjBYCAhIQEnJyfz\nvq6ursTHx987jjy+FyIiYsWMhXynmmeeeYaKFSvi6enJggULmD17Nk2aNMm1T05Ozh2PvdP2u+37\nd6oQRUTEooKsEO+kdevWeHp6AtCxY0eOHz+Ou7s7CQkJ5n0uXryIu7s77u7u5uovIyODnJwc3Nzc\nSEpKMu8bFxdnsUtWCVFERIqdN954g9OnTwMQHR1N3bp18fb25siRI1y9epXk5GRiYmJo3rw5bdq0\nYePGjQBERkbSsmVLTCYTtWvXZv/+/QBs3ryZtm3b3vOa6jIVERGLCnIdYmxsLFOmTOHs2bPY2tqy\nadMmBgwYwIgRI3BwcKBs2bKEhIRgb2/PyJEjCQgIwGAwMHToUBwdHenWrRt79uyhX79+2NnZMXny\nZACCgoIIDg4mOzsbb29vfHx87hmHIScvHatW7EZmUUdQMtjb6r3Ki/BDZ4o6hBJhQLPqhB3Qe5UX\nA5oVztfzLdj7Z76PfaVVzfsYScFRhSgiIhZZwa1MlRBFRMSyglyYX1woIYqIiEVWkA+VEEVExDJr\nWJJgDW0UERGxSBWiiIhYZNVfECwiInJL6U+HSogiIpIHmmUqIiKCKkQRERFAyy5EREQA65hUo2UX\nIiIiqEIUEZE8sIbqSQlRREQssoYuUyVEERGxqPSnQyVEERHJA1WIIiIiaAxRREQEUIUoIiICWMcY\nojVUwSIiIhapQhQREYusoMdUCVFERCwzWkGnqRKiiIhYpApRREQEMKhCFBERUYUoIiICWMcYopZd\niIiIoApRRETyQF2mIiIiKCGKiIgAmmUqIiICgLH050NNqpE7W/ddBC2bPczDjT3p2P5Rfo6NBeDi\nxYs80cWP/zR46LZjFnw+jwZ1a9Ggbi2GvTaEjIwM/vrrL7wbNcj1cHEqy/p13xV2k6SQZWZmsGzG\nh/g39yAx7rx5++r50xndowOjnmvH7DGvkXztCgDzx73FsM7NGN2jAw0aNGB0jw6ciD1oPm7X+tUM\nbluf3RvW5LrOz/t2M/b5rox6rh2TX++f61py/xj+xZ+SQglRbnP27FlefmkgS5Yu46cjv9Knb3+G\nvT6ExMREHu/Unv80anzbMbt27WLWp5+wc8+PxP76G9euXyNqz25q1KjBodij5kfE+k08UL06HTs9\nVgQtk8I04+0AypQtl2tb1MZviY3eyYSvv2dK+Hays7KJWDTb/HrvYe8ydfV2jh49ytTV26nTqAkA\n3y2Zw49b11G1Zp1c57uRmsKcoKEMfn8q09b8lybtHmNxyJiCb5wVMhjy/ygplBDlNiaTidCw5Xg2\nbAiAT5tH+fWXnzEYDKwK/5Ynn3r6tmMWL15MwMtDcHNzw9bWltCly2jXvsNt+703ZjRjgt7HwcGh\noJshRaz74DfpMWRkrm3VatdjUOAk7OwdMBqNeDZrzYU/T1g8l2dzH976ZBH2/5Ngf9m3G/cHavBg\ng5sf0to/3Ycje/9LavL1+9cQsRpKiHIbd3d3Hu/cxfx806bveaRFS5ydnalXv/4djzl06BDJ16/T\nqUNbvP5Tn+CxQWRlZeXa5+fYWH46GEPf/s8XaPxSPNT1anbbtpr1GlKz3s0PWinXrxK9bR1N2j1u\nfj1q47cEv/AEDRs2JGLRZ+Tk5ADwUKMmd/yC2gt//YF79Zrm5/Zly1G+gjNxp0/d59aINXSZalKN\n3FPkD9uY/ekMvt/8wz33S0pKYs/uXXz73QbS0tLo6teRWrVq82LAYPM+M6ZPY9gbIzAa9TnM2s19\nbxgHtm+iVednePTJHgB4Nm1FdnYO7Z7qhW9VIy3b+uLiXpVHn+x51/Ok30jFZFcm1zY7e3vSbqQU\naPzWSJNq/oUzZ85Qv359fvrpp1zbe/ToQWBg4B2PWbNmDVOmTCE+Pp7g4OCCCg2Affv2cenSpQK9\nRkkXsfZbXg4YxOpv15m7T++mQoUK9O7TD0dHRypVqsSAFwaxdetm8+tpaWl8F/EtPXr1LuCopSR4\nfeJs5v1whDIOZfn8/eEAtHu6Dx2698VoY4OHhwe+z/bn4K5t9zxPGfuyZKSn5dqWfiMVe4dydzlC\n8ssaKsQC/aju4eHBunXrzM///PNPrl69avE4Nzc3Pvzww4IMjdWrVysh3sMP27byzttvsm7DZpo1\nb25x/5o1a3LlyhXzcxsbG2xsbMzP/7tjO/UbeOLm5lYg8UrJ8PO+3Zw5cQwAuzL2+D7bjyNR/wXg\n9O9HcyW3rKwsbGzv3YlV9cGHcnWPply/SvLVK1SuUev+B2/lrGFSTYF2mXp7e7Nnz56b/7BtbFi/\nfj1t2rThxo0bREREEBYWhtFopG7dunz00Ufm486cOcPw4cNZs2YN3377LQsXLqRKlSo4OzvTqlUr\nAA4cOEBiYiInT54kICCAXr163fGca9asuW3fatWqsXXrVn777Tc+++wzqlWrVpBvQ4mTkpLCK4Nf\nZFX4tzTw9MzTMX369OGDceMY9FIAdnZ2LF8WxsuvvGp+/fDhQzRokLdzSel1/Kd9/HZoH299sgiT\nXRkO/ncrHnUbALBoUiDNO3ThiRde5fLly+xaH87TLw275/kaNm/Nlx+9w7GffqT+wy3Y+PWXPPxo\nJ+wdyhZGc6xKQee148eP8/rrrzNo0CAGDBjA+fPnGT16NFlZWbi5uTFt2jTs7OyIiIggNDQUo9FI\n79696dWrFxkZGQQGBnLu3DlsbGwICQnBw8ODo0ePMm7cOADq16/P+PHj7xlDgSZEk8mEt7c30dHR\n+Pj4sG3bNoYNG8amTZtITU3lyy+/xMnJieeff55jx47ddnx2djaffPIJa9asoWzZsjz55JPmhHj8\n+HFWrFjBqVOnePvtt+nVq9ddz/m/+65duxZPT0/ef/99JcM7+C5iLQnx8bw4MPfkl1GjxzBtaggp\nKSnEXbiAd6MGVKv2AN9v3kafPn04dORnmj3cCAcHB5586hn8Bw4yH3v2zBkqV6lSyC2RonLlUjwT\nX+llfj5xSC9sbGwJnLecpIQ4gvo+DuTgUrkaAWOnAfDq+BksmjSG7d8uY2LZMvh06U7rzt0BmDrs\neRLOn+XShbNc+OsP1i6cRe9h79LctytDJ84hdMpY0lJTqOzxIK988ElRNLnUMxZgqZeSksJHH31E\n69atzdtmzZpF//796dq1K5988gnh4eF0796dOXPmEB4ejslkomfPnvj5+REZGYmTkxPTp09n165d\nTJ8+nZkzZzJx4kSCgoLw8vJi5MiR7Nixg/bt2981jgKfVNOlSxfWrVtHpUqVqFy5MmXL3vzkVqFC\nBV5//XUATpw4QVJS0m3HXr58mfLly1OpUiWAXG/Www8/jI2NDVWqVOHatWv3POed9s0rOxvrGEz+\nu4ED+jFwQL87vjb4pYF3PW7ShA+ZNOHOXd1zZ8+6L7GVdAOaVS/qEApJdYae+v2OrwzvuvTOhzSr\nzsjuu+740oCoHXe/VLNeTBzS6+6vl2JhB84U2rUK8tegnZ0dX3zxBV988YV5W3R0tLmi8/X1ZdGi\nRdSqVYvGjRvj6OgIQNOmTYmJiSEqKoru3W9+ePLx8SEoKIj09HTOnj2Ll5eX+RxRUVFFmxBbt27N\nhx9+iJubG507dwYgIyODDz/8kLVr1+Lm5saQIUPueGxOTk6uGYl/n3Zt+z9jC+np6Xc95//u+0+k\nZ1neR8DeFm5kFnUUxV/4ocL7BVaSDWhWvVB/2UvRsrW1ve33dGpqKnZ2dgC4uroSHx9PQkICLi4u\n5n1cXFxu2240GjEYDCQkJODk5GTe99Y57qXA57/b2dnxyCOPsHr1ajp27AhAcnIyNjY2uLm5cf78\neWJjY8nIyLjt2IoVK5KUlMSVK1e4ceMGP/74412vk9dz3mIwGG5bJyciIndh+BePf+nWetR/s/1u\n+/5doSwI69KlCw0bNjSXuRUrVqRNmzb06NGD2bNnM3jwYEJCQsjMzF1i2Nra8tprr/H8888zcuRI\nGjVqdNc1bM7Oznk65y0tWrRg+PDh/Pbbb/e3sSIipVBhL7soW7YsN27cACAuLg53d3fc3d1JSEgw\n73Px4kXz9lvVX0ZGBjk5Obi5ueUairt1jnspsC7T6tWrM3nyZAA6dOhAhw4dAGjZsiUtW7a8bf8X\nX3wx1/M1a27ewNfV1ZWwsDAqVqxIQEAANWrUoGnTpub9ypUrxw8/3Fw0fut6dzvn3/cdNmwYw4bd\newabiIjcVNjLJ3x8fNi0aRPPPPMMmzdvpm3btnh7ezN27FiuXr2KjY0NMTExBAUFcf36dTZu3Ejb\ntm2JjIykZcuWmEwmateuzf79+2nevDmbN2/G39//ntcs9nequXHjBgMHDsTBwQFPT89cyVBERApH\nQebD2NhYpkyZwtmzZ7G1tWXTpk18/PHHBAYGsnLlSqpVq0b37t0xmUyMHDmSgIAADAYDQ4cOxdHR\nkW7durFnzx769euHnZ2duTgKCgoiODiY7OxsvL298fHxuXcbc/LSsWrFNFEkbzSpJm80qSZvNKkm\n7wpr5vK+k1cs73QXj9SqcB8jKTjFvkIUEZGiV5JuwZZfusuyiIgIqhBFRCQPStI9SfNLCVFERCyy\ngnyohCgiInlgBRlRCVFERCyyhkk1SogiImKRxhBFRESwih5TLbsQEREBVYgiIpIXVlAiKiGKiIhF\nmlQjIiKCJtWIiIgAVtFjqoQoIiJ5YAUZUQlRREQs0hiiiIgI1jGGqHWIIiIiqEIUEZE8sIICUQlR\nRETywAoyohKiiIhYpEk1IiIiWMekGiVEERGxyAryoRKiiIjkgRVkRC27EBERQRWiiIjkgSbViIiI\noEk1IiIigFUMISohiohIHlhBRlRCFBERizSGKCIignWMIWrZhYiICKoQRUQkD6ygQFRCFBGRPLCC\njKiEKCIiFmlSjYiICNYxqUYJUURELLKCfKiEKCIilqlCFBERAQqqRoyOjubNN9+kbt26ANSrV4/B\ngwczevRosrKycHNzY9q0adjZ2REREUFoaChGo5HevXvTq1cvMjIyCAwM5Ny5c9jY2BASEoKHh0e+\nYlFCFBGRItWiRQtmzZplfj5mzBj69+9P165d+eSTTwgPD6d79+7MmTOH8PBwTCYTPXv2xM/Pj8jI\nSJycnJg+fTq7du1i+vTpzJw5M19xaGG+iIhYZDDk//FPRUdH06lTJwB8fX2Jiori0KFDNG7cGEdH\nR+zt7WnatCkxMTFERUXh5+cHgI+PDzExMfluoypEERGxqCCHEH///XdeffVVrly5wrBhw0hNTcXO\nzg4AV1dX4uPjSUhIwMXFxXyMi4vLbduNRiMGg4H09HTz8f+EEqKIiFhUUJNqHnzwQYYNG0bXrl05\nffo0L7zwAllZWebXc3Jy7njcP92eF+oyFRERiwz/4s+9VK5cmW7dumEwGKhRowaVKlXiypUr3Lhx\nA4C4uDjc3d1xd3cnISHBfNzFixfN2+Pj4wHIyMggJycnX9UhKCGKiEheGP7F4x4iIiJYuHAhAPHx\n8Vy6dInnnnuOTZs2AbB582batm2Lt7c3R44c4erVqyQnJxMTE0Pz5s1p06YNGzduBCAyMpKWLVvm\nu4nqMhUREYsKagyxY8eOvPPOO2zbto2MjAzGjRuHp6cn7777LitXrqRatWp0794dk8nEyJEjCQgI\nwGAwMHToUBwdHenWrRt79uyhX79+2NnZMXny5HzHYsj5Nx2uVuBGZlFHUDLY2+q9yovwQ2eKOoQS\nYUCz6oQd0HuVFwOaVS+U68Rdzcj3sZWdTPcxkoKjClFERCzSnWpERETQt12IiIjcVPrzoRKiiIhY\nZgX5UAlRREQs0xiiiIgI1jGGqIX5IiIiqEIUEZE8sIYuU1WIIiIiqEIUEZE8sIYKUQlRREQssoZJ\nNUqIIiJikSpEERERtDBfRETkJivIiJplKiIigipEERHJA02qERERQZNqREREAKsYQlRCFBGRPLCC\njKiEKCIiFmkMUUREBOsYQzTk5OTkFHUQIiIiRU3rEEVERFBCFBERAZQQRUREACVEERERQAlRREQE\nUEIUEREBlBBFREQAJUQRKcG0jFruJyVEESmRMjMzOXHiBADJyclFHI2UBrpTjRSYnJwcDNZwvycp\nEpmZmaxatYro6GhMJhMfffQRDg4ORR2WlGCqEKXAGAwGdu/ezdy5c4mKiiI9Pb2oQ5JSIicnB1tb\nW1q3bs2JEydITU1VMpR/TQlRCsyJEyeYM2cOFy9eZNu2baxevVpJUf61Wz0PJ0+e5NKlSyxatAhn\nZ2cmTZpEYmJiUYcnJZjNuHHjxhV1EFL6HDlyhLCwMHr06MGLL75Iamoqx44d48KFC9SrVw8bG5ui\nDlFKKIPBwJ49exg9ejTHjh3jzJkzBAUFsWHDBv744w/S09O5cuUKlStXLupQpYRRhSj3XUJCAtWq\nVePEiRNER0cD0LFjR5o0acKxY8f45ptvNDtQ8u3UqVOsXr2aOXPmsHDhQo4cOcLUqVOZNGkS165d\nY9q0aVy/fr2ow5QSSAlR7ovs7GwAjh8/zhNPPMHGjRuZP38+x48f56uvvsJkMtGxY0datGhBq1at\nNNlG8iUzM5Pt27dz8uRJzp07B8DChQs5duwY48eP5/333+err77Cx8eniCOVkkhdpvKvZGZmYjQa\nzRNoli1bRqtWrZg7dy4uLi688cYbfPnll8TFxfHII4/w0EMP4eLiUtRhSwny9zHD7OxsatWqhY2N\nDbGxsZQpU4aaNWvSrVs3Fi9eTOPGjalevXpRhywllG1RByAlV2JiInv37sXX1xdbW1tWrlxJu3bt\n6NmzJ/369aN79+4YjUZCQkIYOXIkXbp0wcPDo6jDlhLGYDCwfft2Fi5ciMlkomvXrjRr1gyTycS2\nbdvIysrCx8eHsLAw9TzIv6IKUfLtzJkzeHh4kJOTQ2pqKra2tsTGxlKrVi2qV69Oo0aNGD16NPXr\n1ycwMJAKFSoUdchSAp0+fZqPP/6YefPmERcXx1dffUWdOnVo2bIliYmJHDlyhEaNGmFvb6+EKP+K\nEqLkm6urK2XLlmXWrFmcPHmSihUrkpGRwYULF3jwwQfNv5yWLVtG1apVeeihh4o4YimJbty4QVZW\nFomJiWzbto0hQ4Ywf/58EhMTOXXqFG+88QZubm5KhvKv6U41km9Hjx7F1dWVU6dOERUVhbOzM9nZ\n2Vy7do3Dhw9z7tw5VqxYwdq1a3Fzc+Pxxx8v6pClBPnpp584e/YsLVq0oFy5cixevBgvLy/atm3L\n7NmzsbOz4+GHH6ZFixZFHaqUEpplKv/Irc9PBw4cYMSIEQwZMoT4+HgaNmxIYmIiNjY2tGjRglGj\nRjFw4ED27t3LqlWrqFevXhFHLiXJgQMHeO+994iOjmbAgAGcP3+eqlWrEhwczH//+19iY2Px8/NT\nMpT7ShWi/GP79+9n8+bNvPTSS1y8eJFFixbx+OOPU6ZMGfbv34+zszODBg3i999/Z8mSJQwZMoQ6\ndeoUddhSQpw9e5YZM2bQv39/mjZtytKlS1m2bBkLFixg586dbNq0iZdeeon27dsXdahSyighyj82\ne/Zs5s6dy7Zt26hatSq7d+8mPDwcX19fypYtS+3atalduzYA6enp2NnZFXHEUlJcvHiRH3/8kQ0b\nNlC9enVGjRqFyWQiLCyMBQsWEBYWRpUqVfRvSgqEJtWIRbfWgZ06dQpbW1seffRR0tPTmT59Os8+\n+yx16tTBwcGB7777jj59+lC9enXzMbpFm1hy69/KsWPHCA8Px93dnUaNGnHp0iV+//13GjduzMMP\nP4zBYMDBwYEaNWoUdchSSqlClDyJiopi5syZtG3bluPHjzNr1iwmTJjAvn37+PrrrylfvjyXL1/G\n2dm5qEOVEmjHjh0sW7aMa9eu4eTkxNNPPw3AsWPHKFeuHAEBAeYPV/paMSkomlQjd5SYmMi5c+e4\ndu0a169fZ+bMmUyfPp3KlSuTnJxMZmYmY8eOpXnz5vTs2ZOsrCycnJyKOmwpIeLi4vj2228BSE1N\nZfny5YwYMYJly5bRrFkzDh8+jI2NDXXr1iUpKcl8mzZAyVAKjCpEuc2ff/7JyJEjadiwIRcuXOCD\nDz5gw4YNODs7s2HDBiZNmgTc/PTevn17fvvtN+rWrVvEUUtJcf36dVJSUszfYeju7k5AQADt2rVj\n4MCBZGdn8/7775OQkMAzzzxDs2bN9M0VUihUIUouJ06cYOrUqfj7+/PBBx9Qp04drl27xo8//sia\nNWv4+OOPqVKlCnv37mXv3r0AWnAveXb9+nXmz59PZGQkHh4eBAYGsmTJEsaNG8f69etZuXIlRqOR\n5557jjJlyrB371598a8UGt3LVMxSU1MZOnQoTZo04ZlnniEjI4MtW7Zgb2/PkSNHSEpKIiIighs3\nbrB7925eeeUVQF1Y8s/Url2bX375hfXr1zN27FgmT55MVlYWISEhvPzyy/zyyy8cPHiQTz/9lE8/\n/ZSDBw9qiYUUCs0yFTOTyUSVKlXYuXMnZcqUYfny5TRs2JC3334bDw8Ptm/fzn/+8x/KlCnDE088\nQZs2bYo6ZClBsrOzKVOmDLVr1yY1NZWffvoJg8FAr169WLZsGXZ2dowbN46srCxatWoFwA8//EC/\nfv1wdHQs4ujFGqhClFwee+wx7OzsGDt2LJ6ensyfPx+Azp07c+jQIVq3bk3z5s2LOEopaTIzM7G1\nteWXX34hJSWFFi1akJOTw08//UROTg7vv/8+gYGBXL16lZdeeoktW7awfPlyxo4dS9WqVYs6fLES\nmlQjd7Rjxw7mzJnDyy+/jJ+fH4cPHyY4OJjJkyfToEGDog5PSoirV6+aZx9v3bqVxYsX4+TkhIeH\nB08//TSnTp3iyJEj1KtXj2bNmnH16lW8vLxIT08nIyODcuXKFXELxJqoQpQ7ujVms2DBAn799Vd+\n+eUXRo0apWQoeRYXF8fbb79Nx44defrpp9m0aROffvop0dHRTJgwAQcHB3r27MmNGzf466+/ePTR\nR3nwwQfJzs7Gzs5Od6ORQqeEKHfVvn17srKyCAoKYvLkyRozlH/EZDJx+fJlNm7cyOXLl/Hy8uLw\n4cMsX76c0NBQxo8fz7Fjx0hKSjKvcQUwGjX5XYqGukzFosTERFxcXHSHEMmz7OxsjEYjW7ZsITo6\nmkqVKnHhwgUaN25MWloa/fv3Z/369cTHx9O8eXMaNWpU1CGLqEIUy1xcXAAtrxDLEhMTKVeuHGXK\nlAGgUqVK/P7777Rp0waTycTq1as5fPgwRqOR9evXExQUhKenZxFHLXKTll2IyH2RnJxMt27d2L17\nN0lJSTRp0oSqVauSlpbG7t27zd+GcuHCBTIyMhg4cCBNmjQp6rBFzFQhish9Ua5cOT744ANWrlzJ\nkiVLOHv2LPXr16dq1aoYjUaqVatG+fLluXbtGt26ddOXRkuxo4QoIvfN448/TpUqVfj2228xGAzY\n29vz4Ycfkp2dzalTpxg1ahQeHh5aTiHFkqZzich91ahRI5566imSk5OpU6cOCxcupFOnTvz666/E\nxcUpGUqxpVmmIlIgoqOjWbt2Lb169aJJkyYkJSVRsWLFog5L5K7UZSoiBaJly5ZkZWWxfPlyateu\nrWQoxZ4qRBEpUKoMpaRQQhQREUGTakRERAAlRBEREUAJUUREBFBCFBERAZQQpRQ7c+YMjRo1wt/f\nH39/f/r27cvIkSO5evVqvs/5zTffEBgYCMBbb71FXFzcXfeNiYnh9OnTeT53ZmYm9evXv237Z599\nxowZM+55bMeOHfnzzz/zfK3AwEC++eabPO8vYg2UEKVUc3FxYenSpSxdupQVK1bg7u7OvHnz7su5\nZ8yYYf4OvztZs2bNP0qIIlK0tDBfrMojjzzCypUrgZtVVdeuXTl9+jSzZs1iw4YNhIWFkZOTg4uL\nCxMmTMDZ2Zmvv/6a5cuXU6VKFdzd3c3n6tixI4sXL8bDw4MJEyYQGxsLwIsvvoitrS0bN27k8OHD\njBkzhpo1azJ+/HhSU1NJSUnh7bffxsfHhz/++INRo0bh4OBAy5YtLca/bNky1q5di8lkokyZMsyY\nMQMnJyfgZvV65MgRLl26xPvvv0/Lli05d+7cHa8rIrdTQhSrkZWVxZYtW2jWrJl524MPPsioUaM4\nf/48n3/+OeHh4djZ2REaGsr8+fMZOnQos2bNYuPGjTg7O/Paa69RoUKFXOeNiIggISGBVatWcfXq\nVd555x3mzZuHp6cnr732Gq1bt+aVV17hpZdeolWrVsTHx9OnTx82b97MnDlz6NGjB/3792fz5s0W\n25CWlsbChQspX748wcHBREREMGDAAAAqVqxIaGgoUVFRTJkyhTVr1jBu3Lg7XldEbqeEKKVaYmIi\n/v7+wM1vcW/evDmDBg0yv37r+/gOHjxIfHw8AQEBAKSnp1O9enX+/PNPHnjgAZydnYGbtyM7evRo\nrmscPnzYXN05OTmxYMGC2+KIjo4mOTmZOXPmAGBra8ulS5c4fvw4r7zyCgCtWrWy2J6KFSvyyiuv\nYDQaOXv2LG5ububX2rRpY27T77//fs/risjtlBClVLs1hng3JpMJADs7O7y8vP6/nTt2OS+K4zj+\nvk9P8kzCYFAMMhgMMhrMdkqZTEhZlVuuxXBnkz9EDDaKSeJfkJSUZNAVfoN6evzu4/kLPq/tntP5\nnrPcPp17T4d+v//Sv16vMQzj+/l+v7tqGIbxa/tPHo+HXq9HIBB4aX88Hnx8PH/l3263P2vsdjts\n22YwGBAMBrFt27WO/2u+m1dE3HSoRgRIJpOsViv2+z0Aw+GQ8XhMJBJhs9lwOp14PB7MZjPX2FQq\nxWQyAeB8PlMoFHAcB8MwuF6vAKTTaYbDIfDctXa7XQBisRjL5RLg19o/HQ4H/H4/wWCQ4/HIdDrF\ncZzv/vl8DjxPt8bj8Y2PuRgAAADFSURBVD/nFRE37RBFgFAohGmaVCoVvr6+8Hq92LaNz+ejWq1S\nKpUIh8OEw2Eul8vL2Fwux2KxoFgscrvdKJfLeDweMpkMlmXRarUwTZN2u81gMMBxHGq1GgD1ep1m\ns8loNCKVSvH5+f6VTCQSRKNR8vk8kUiERqNBp9Mhm80Cz0u0K5UK2+0Wy7IA3s4rIm663FtERAR9\nMhUREQEUiCIiIoACUUREBFAgioiIAApEERERQIEoIiICKBBFREQABaKIiAgA/wCvtGJThLPjVgAA\nAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 576x396 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "mcnY6BLSjyEX",
        "colab_type": "code",
        "outputId": "b3e1bcd1-ec48-450f-fe64-91cf88e2d87b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "cell_type": "code",
      "source": [
        "sensitivity = cm[0, 0] / (cm[0, 0] + cm[0, 1])\n",
        "print(f'sensitivity: {sensitivity}')\n",
        "\n",
        "specificity = cm[1, 1] / (cm[1, 1] + cm[1, 0])\n",
        "print(f'specifity: {specificity}')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "sensitivity: 0.8568193825949129\n",
            "specifity: 0.8617720227084263\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "_chKt-onjzs7",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import classification_report"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "uUxmxH9Tj1aU",
        "colab_type": "code",
        "outputId": "498ebbe9-d3b2-451b-c736-d8379bb83710",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 170
        }
      },
      "cell_type": "code",
      "source": [
        "report = classification_report(val_labels, real_classes, target_names=classes_names)\n",
        "print(report)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "      Benign       0.94      0.86      0.90     39747\n",
            "   Malignant       0.70      0.86      0.77     15677\n",
            "\n",
            "   micro avg       0.86      0.86      0.86     55424\n",
            "   macro avg       0.82      0.86      0.84     55424\n",
            "weighted avg       0.87      0.86      0.86     55424\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "HONahXsYr2dM",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Gradient maps"
      ]
    },
    {
      "metadata": {
        "id": "AKZUVqtMr1xf",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 170
        },
        "outputId": "36780f60-8776-41c4-e464-19b75f4678a9"
      },
      "cell_type": "code",
      "source": [
        "!pip uninstall keras-vis"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Uninstalling keras-vis-0.4.1:\n",
            "  Would remove:\n",
            "    /usr/local/lib/python3.6/dist-packages/docs/*\n",
            "    /usr/local/lib/python3.6/dist-packages/keras_vis-0.4.1.dist-info/*\n",
            "    /usr/local/lib/python3.6/dist-packages/vis/*\n",
            "  Would not remove (might be manually added):\n",
            "    /usr/local/lib/python3.6/dist-packages/docs/autogen.py\n",
            "Proceed (y/n)? y\n",
            "  Successfully uninstalled keras-vis-0.4.1\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "uTbRFc5fr4sw",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "!pip install git+https://github.com/raghakot/keras-vis.git -U"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "BoGJ6T1Rr_Ob",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from vis.visualization import visualize_saliency, overlay\n",
        "from vis.utils import utils\n",
        "from keras import activations"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "nTe7r0qitLIb",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "benign_folder = os.listdir(\"training/benign\")\n",
        "malignant_folder = os.listdir(\"training/malignant\")\n",
        "\n",
        "benign_folder_name = \"training/benign\"\n",
        "malignant_folder_name = \"training/malignant\""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "SJXe1hljt04Y",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 105
        },
        "outputId": "ef1a54d3-1287-47b1-8d97-b056f121b755"
      },
      "cell_type": "code",
      "source": [
        "benign_image_path = f'{benign_folder_name}/{benign_folder[20]}'\n",
        "malignant_image_path = f'{malignant_folder_name}/{malignant_folder[58]}'\n",
        "\n",
        "benign_image = utils.load_img(benign_image_path, target_size=(48, 48))\n",
        "malignant_image = utils.load_img(malignant_image_path, target_size=(48, 48))"
      ],
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/skimage/transform/_warps.py:105: UserWarning: The default mode, 'constant', will be changed to 'reflect' in skimage 0.15.\n",
            "  warn(\"The default mode, 'constant', will be changed to 'reflect' in \"\n",
            "/usr/local/lib/python3.6/dist-packages/skimage/transform/_warps.py:110: UserWarning: Anti-aliasing will be enabled by default in skimage 0.15 to avoid aliasing artifacts when down-sampling images.\n",
            "  warn(\"Anti-aliasing will be enabled by default in skimage 0.15 to \"\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "metadata": {
        "id": "cCod8jnEuZq7",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "benign_image_model = image.load_img(benign_image_path, target_size=(48, 48))\n",
        "malignant_image_model = image.load_img(malignant_image_path, target_size=(48, 48))\n",
        "\n",
        "benign_image_model = image.img_to_array(benign_image_model)\n",
        "benign_image_model = preprocess_input(benign_image_model)\n",
        "\n",
        "malignant_image_model = image.img_to_array(malignant_image_model)\n",
        "malignant_image_model = preprocess_input(malignant_image_model)\n",
        "\n",
        "preprocessed_images = np.array([benign_image_model, malignant_image_model])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Esumi11iu0_I",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "predictions = model.predict(preprocessed_images)\n",
        "predicted_classes = np.argmax(predictions, axis=1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "-hh5VV2lu6KB",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "images_array = [benign_image, malignant_image]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "OukyS2Y4u_Is",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "outputId": "441ced24-40b4-4da9-e4b7-949825104eec"
      },
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots(1, 2, figsize=(8, 8))\n",
        "fig.suptitle('Normal Images')\n",
        "\n",
        "for index, img in enumerate(images_array):   \n",
        "  class_index = predicted_classes[index]\n",
        "  ax[index].set_title(f'class: {classes_names[index]}, pred: {classes_names[class_index]}')\n",
        "  ax[index].imshow(img)"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeMAAAGXCAYAAACAxDjvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 576x576 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "TrB5v9V2wSbJ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "outputId": "2f595e99-f07f-4c49-93c4-8f90bbbc6307"
      },
      "cell_type": "code",
      "source": [
        "layer_idx = utils.find_layer_idx(model, 'dense_2')\n",
        "\n",
        "fig, ax = plt.subplots(1, 2, figsize=(8, 8))\n",
        "fig.suptitle('Gradient maps')\n",
        "\n",
        "for index, img in enumerate(images_array):    \n",
        "    grads = visualize_saliency(model, layer_idx, filter_indices=index, seed_input=img)\n",
        "    ax[index].set_title(classes_names[index])\n",
        "    ax[index].imshow(grads, cmap='jet')"
      ],
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeMAAAGXCAYAAACAxDjvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 576x576 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "liVtgdUHwVya",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 541
        },
        "outputId": "c55511f9-a0bb-47a4-a3ae-bda12cb75068"
      },
      "cell_type": "code",
      "source": [
        "for modifier in ['guided', 'relu']:\n",
        "    plt.figure()\n",
        "    f, ax = plt.subplots(1, 2)\n",
        "    plt.suptitle(modifier)\n",
        "    \n",
        "    for index, img in enumerate(images_array):    \n",
        "        grads = visualize_saliency(model, layer_idx, filter_indices=index, \n",
        "                                   seed_input=img, backprop_modifier=modifier)   \n",
        "        ax[index].imshow(grads, cmap='jet')"
      ],
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD1CAYAAABA+A6aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\npEYY6JLUCANdkhphoEtSIwx0SWrE/wO6g5qDgrTbCQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD1CAYAAABA+A6aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nXZIaYaBLUiMMdElqhIEuSY34fzkTR6fO5HP3AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "biRT1DOrwb5f",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 803
        },
        "outputId": "f3b5f438-e572-4cda-ca4f-843ab006e936"
      },
      "cell_type": "code",
      "source": [
        "import matplotlib.cm as cm\n",
        "from vis.visualization import visualize_cam\n",
        "\n",
        "penultimate_layer = utils.find_layer_idx(model, 'Conv_1_bn')\n",
        "\n",
        "for modifier in [None, 'guided', 'relu']:\n",
        "    plt.figure()\n",
        "    f, ax = plt.subplots(1, 2)\n",
        "    plt.suptitle(\"vanilla\" if modifier is None else modifier)\n",
        "    for index, img in enumerate(images_array):    \n",
        "        grads = visualize_cam(model, layer_idx, filter_indices=index, \n",
        "                              seed_input=img, penultimate_layer_idx=penultimate_layer,\n",
        "                              backprop_modifier=modifier)        \n",
        "  \n",
        "        jet_heatmap = np.uint8(cm.jet(grads)[..., :3] * 255)\n",
        "        ax[index].imshow(overlay(jet_heatmap, img))"
      ],
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD1CAYAAABA+A6aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJInA03mCYS0AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD1CAYAAABA+A6aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n0A0Gg6ElsAXdYDAYWgJb0A0Gg6El+P8BnYrShVkneHcAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD1CAYAAABA+A6aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/n9skJi0lpcQRAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "BR3kX7Tq3tep",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Save model"
      ]
    },
    {
      "metadata": {
        "id": "hBAOuV1p3kpY",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model_json = model.to_json()\n",
        "with open(\"breast_cancer_model.json\", \"w\") as json_file:\n",
        "    json_file.write(model_json)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "B4Ed8a5goVgP",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}